From a6e2f029ae34f41adb6ae3812c32c5d326e1abd2 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Wed, 29 Apr 2015 12:48:40 -0400 Subject: [PATCH 001/647] Make asm/word-at-a-time.h available on all architectures Added the x86 implementation of word-at-a-time to the generic version, which previously only supported big-endian. Omitted the x86-specific load_unaligned_zeropad(), which in any case is also not present for the existing BE-only implementation of a word-at-a-time, and is only used under CONFIG_DCACHE_WORD_ACCESS. Added as a "generic-y" to the Kbuilds of all architectures that didn't previously have it. Signed-off-by: Chris Metcalf --- arch/arc/include/asm/Kbuild | 1 + arch/avr32/include/asm/Kbuild | 1 + arch/blackfin/include/asm/Kbuild | 1 + arch/c6x/include/asm/Kbuild | 1 + arch/cris/include/asm/Kbuild | 1 + arch/frv/include/asm/Kbuild | 1 + arch/hexagon/include/asm/Kbuild | 1 + arch/ia64/include/asm/Kbuild | 1 + arch/m32r/include/asm/Kbuild | 1 + arch/metag/include/asm/Kbuild | 1 + arch/microblaze/include/asm/Kbuild | 1 + arch/mips/include/asm/Kbuild | 1 + arch/mn10300/include/asm/Kbuild | 1 + arch/nios2/include/asm/Kbuild | 1 + arch/powerpc/include/asm/Kbuild | 1 + arch/s390/include/asm/Kbuild | 1 + arch/score/include/asm/Kbuild | 1 + arch/tile/include/asm/Kbuild | 1 + arch/um/include/asm/Kbuild | 1 + arch/unicore32/include/asm/Kbuild | 1 + arch/xtensa/include/asm/Kbuild | 1 + include/asm-generic/word-at-a-time.h | 80 +++++++++++++++++++++++++--- 22 files changed, 93 insertions(+), 8 deletions(-) diff --git a/arch/arc/include/asm/Kbuild b/arch/arc/include/asm/Kbuild index 1a80cc91a03b..c8f07ce1c757 100644 --- a/arch/arc/include/asm/Kbuild +++ b/arch/arc/include/asm/Kbuild @@ -47,4 +47,5 @@ generic-y += types.h generic-y += ucontext.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/avr32/include/asm/Kbuild b/arch/avr32/include/asm/Kbuild index 1d66afdfac07..519810d0d5e1 100644 --- a/arch/avr32/include/asm/Kbuild +++ b/arch/avr32/include/asm/Kbuild @@ -19,4 +19,5 @@ generic-y += sections.h generic-y += topology.h generic-y += trace_clock.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/blackfin/include/asm/Kbuild b/arch/blackfin/include/asm/Kbuild index 07051a63415d..c80181e4454f 100644 --- a/arch/blackfin/include/asm/Kbuild +++ b/arch/blackfin/include/asm/Kbuild @@ -45,4 +45,5 @@ generic-y += types.h generic-y += ucontext.h generic-y += unaligned.h generic-y += user.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/c6x/include/asm/Kbuild b/arch/c6x/include/asm/Kbuild index 7aeb32272975..2f697020a8fd 100644 --- a/arch/c6x/include/asm/Kbuild +++ b/arch/c6x/include/asm/Kbuild @@ -58,4 +58,5 @@ generic-y += types.h generic-y += ucontext.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/cris/include/asm/Kbuild b/arch/cris/include/asm/Kbuild index d294f6aaff1d..09916c8073ac 100644 --- a/arch/cris/include/asm/Kbuild +++ b/arch/cris/include/asm/Kbuild @@ -25,4 +25,5 @@ generic-y += sections.h generic-y += topology.h generic-y += trace_clock.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/frv/include/asm/Kbuild b/arch/frv/include/asm/Kbuild index 30edce31e5c2..2c987dc05af4 100644 --- a/arch/frv/include/asm/Kbuild +++ b/arch/frv/include/asm/Kbuild @@ -6,3 +6,4 @@ generic-y += irq_work.h generic-y += mcs_spinlock.h generic-y += preempt.h generic-y += trace_clock.h +generic-y += word-at-a-time.h diff --git a/arch/hexagon/include/asm/Kbuild b/arch/hexagon/include/asm/Kbuild index 5ade4a163558..0988816dded0 100644 --- a/arch/hexagon/include/asm/Kbuild +++ b/arch/hexagon/include/asm/Kbuild @@ -57,4 +57,5 @@ generic-y += types.h generic-y += ucontext.h generic-y += unaligned.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/ia64/include/asm/Kbuild b/arch/ia64/include/asm/Kbuild index ccff13d33fa2..dc05773e1f11 100644 --- a/arch/ia64/include/asm/Kbuild +++ b/arch/ia64/include/asm/Kbuild @@ -7,3 +7,4 @@ generic-y += mcs_spinlock.h generic-y += preempt.h generic-y += trace_clock.h generic-y += vtime.h +generic-y += word-at-a-time.h diff --git a/arch/m32r/include/asm/Kbuild b/arch/m32r/include/asm/Kbuild index ba1cdc018731..219e54b33685 100644 --- a/arch/m32r/include/asm/Kbuild +++ b/arch/m32r/include/asm/Kbuild @@ -8,3 +8,4 @@ generic-y += module.h generic-y += preempt.h generic-y += sections.h generic-y += trace_clock.h +generic-y += word-at-a-time.h diff --git a/arch/metag/include/asm/Kbuild b/arch/metag/include/asm/Kbuild index 199320f3c345..611c0df2be39 100644 --- a/arch/metag/include/asm/Kbuild +++ b/arch/metag/include/asm/Kbuild @@ -53,4 +53,5 @@ generic-y += ucontext.h generic-y += unaligned.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/microblaze/include/asm/Kbuild b/arch/microblaze/include/asm/Kbuild index 9989ddb169ca..cefeabae24cc 100644 --- a/arch/microblaze/include/asm/Kbuild +++ b/arch/microblaze/include/asm/Kbuild @@ -9,3 +9,4 @@ generic-y += mcs_spinlock.h generic-y += preempt.h generic-y += syscalls.h generic-y += trace_clock.h +generic-y += word-at-a-time.h diff --git a/arch/mips/include/asm/Kbuild b/arch/mips/include/asm/Kbuild index 7fe5c61a3cb8..dee810fa973e 100644 --- a/arch/mips/include/asm/Kbuild +++ b/arch/mips/include/asm/Kbuild @@ -17,4 +17,5 @@ generic-y += serial.h generic-y += trace_clock.h generic-y += ucontext.h generic-y += user.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/mn10300/include/asm/Kbuild b/arch/mn10300/include/asm/Kbuild index de30b0c88796..27cbc0267b9c 100644 --- a/arch/mn10300/include/asm/Kbuild +++ b/arch/mn10300/include/asm/Kbuild @@ -8,3 +8,4 @@ generic-y += mcs_spinlock.h generic-y += preempt.h generic-y += sections.h generic-y += trace_clock.h +generic-y += word-at-a-time.h diff --git a/arch/nios2/include/asm/Kbuild b/arch/nios2/include/asm/Kbuild index 434639d510b3..e22478929719 100644 --- a/arch/nios2/include/asm/Kbuild +++ b/arch/nios2/include/asm/Kbuild @@ -60,4 +60,5 @@ generic-y += types.h generic-y += unaligned.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/powerpc/include/asm/Kbuild b/arch/powerpc/include/asm/Kbuild index 050712e1ce41..d9c0b44191b0 100644 --- a/arch/powerpc/include/asm/Kbuild +++ b/arch/powerpc/include/asm/Kbuild @@ -8,3 +8,4 @@ generic-y += preempt.h generic-y += rwsem.h generic-y += trace_clock.h generic-y += vtime.h +generic-y += word-at-a-time.h diff --git a/arch/s390/include/asm/Kbuild b/arch/s390/include/asm/Kbuild index dc5385ebb071..9e8089b44f46 100644 --- a/arch/s390/include/asm/Kbuild +++ b/arch/s390/include/asm/Kbuild @@ -5,3 +5,4 @@ generic-y += irq_work.h generic-y += mcs_spinlock.h generic-y += preempt.h generic-y += trace_clock.h +generic-y += word-at-a-time.h diff --git a/arch/score/include/asm/Kbuild b/arch/score/include/asm/Kbuild index 138fb3db45ba..ff19975beb33 100644 --- a/arch/score/include/asm/Kbuild +++ b/arch/score/include/asm/Kbuild @@ -12,3 +12,4 @@ generic-y += sections.h generic-y += trace_clock.h generic-y += xor.h generic-y += serial.h +generic-y += word-at-a-time.h diff --git a/arch/tile/include/asm/Kbuild b/arch/tile/include/asm/Kbuild index d53654488c2c..ea59330b4d90 100644 --- a/arch/tile/include/asm/Kbuild +++ b/arch/tile/include/asm/Kbuild @@ -38,4 +38,5 @@ generic-y += termbits.h generic-y += termios.h generic-y += trace_clock.h generic-y += types.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/um/include/asm/Kbuild b/arch/um/include/asm/Kbuild index 3d63ff6f583f..33c1d3e0caad 100644 --- a/arch/um/include/asm/Kbuild +++ b/arch/um/include/asm/Kbuild @@ -24,4 +24,5 @@ generic-y += preempt.h generic-y += switch_to.h generic-y += topology.h generic-y += trace_clock.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/unicore32/include/asm/Kbuild b/arch/unicore32/include/asm/Kbuild index d12b377b5a8b..932070cd754a 100644 --- a/arch/unicore32/include/asm/Kbuild +++ b/arch/unicore32/include/asm/Kbuild @@ -61,4 +61,5 @@ generic-y += ucontext.h generic-y += unaligned.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/xtensa/include/asm/Kbuild b/arch/xtensa/include/asm/Kbuild index 14d15bf1a95b..85acffd26217 100644 --- a/arch/xtensa/include/asm/Kbuild +++ b/arch/xtensa/include/asm/Kbuild @@ -28,4 +28,5 @@ generic-y += statfs.h generic-y += termios.h generic-y += topology.h generic-y += trace_clock.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/include/asm-generic/word-at-a-time.h b/include/asm-generic/word-at-a-time.h index 94f9ea8abcae..011dde083f23 100644 --- a/include/asm-generic/word-at-a-time.h +++ b/include/asm-generic/word-at-a-time.h @@ -1,15 +1,10 @@ #ifndef _ASM_WORD_AT_A_TIME_H #define _ASM_WORD_AT_A_TIME_H -/* - * This says "generic", but it's actually big-endian only. - * Little-endian can use more efficient versions of these - * interfaces, see for example - * arch/x86/include/asm/word-at-a-time.h - * for those. - */ - #include +#include + +#ifdef __BIG_ENDIAN struct word_at_a_time { const unsigned long high_bits, low_bits; @@ -53,4 +48,73 @@ static inline bool has_zero(unsigned long val, unsigned long *data, const struct #define zero_bytemask(mask) (~1ul << __fls(mask)) #endif +#else + +/* + * The optimal byte mask counting is probably going to be something + * that is architecture-specific. If you have a reliably fast + * bit count instruction, that might be better than the multiply + * and shift, for example. + */ +struct word_at_a_time { + const unsigned long one_bits, high_bits; +}; + +#define WORD_AT_A_TIME_CONSTANTS { REPEAT_BYTE(0x01), REPEAT_BYTE(0x80) } + +#ifdef CONFIG_64BIT + +/* + * Jan Achrenius on G+: microoptimized version of + * the simpler "(mask & ONEBYTES) * ONEBYTES >> 56" + * that works for the bytemasks without having to + * mask them first. + */ +static inline long count_masked_bytes(unsigned long mask) +{ + return mask*0x0001020304050608ul >> 56; +} + +#else /* 32-bit case */ + +/* Carl Chatfield / Jan Achrenius G+ version for 32-bit */ +static inline long count_masked_bytes(long mask) +{ + /* (000000 0000ff 00ffff ffffff) -> ( 1 1 2 3 ) */ + long a = (0x0ff0001+mask) >> 23; + /* Fix the 1 for 00 case */ + return a & mask; +} + +#endif + +/* Return nonzero if it has a zero */ +static inline unsigned long has_zero(unsigned long a, unsigned long *bits, const struct word_at_a_time *c) +{ + unsigned long mask = ((a - c->one_bits) & ~a) & c->high_bits; + *bits = mask; + return mask; +} + +static inline unsigned long prep_zero_mask(unsigned long a, unsigned long bits, const struct word_at_a_time *c) +{ + return bits; +} + +static inline unsigned long create_zero_mask(unsigned long bits) +{ + bits = (bits - 1) & ~bits; + return bits >> 7; +} + +/* The mask we created is directly usable as a bytemask */ +#define zero_bytemask(mask) (mask) + +static inline unsigned long find_zero(unsigned long mask) +{ + return count_masked_bytes(mask); +} + +#endif /* __BIG_ENDIAN */ + #endif /* _ASM_WORD_AT_A_TIME_H */ From 8f6429c7cb59f28433253575cc8e3262eed63592 Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Thu, 16 Jul 2015 19:40:12 +0100 Subject: [PATCH 002/647] iommu/iova: Avoid over-allocating when size-aligned Currently, allocating a size-aligned IOVA region quietly adjusts the actual allocation size in the process, returning a rounded-up power-of-two-sized allocation. This results in mismatched behaviour in the IOMMU driver if the original size was not a power of two, where the original size is mapped, but the rounded-up IOVA size is unmapped. Whilst some IOMMUs will happily unmap already-unmapped pages, others consider this an error, so fix it by computing the necessary alignment padding without altering the actual allocation size. Also clean up by making pad_size unsigned, since its callers always pass unsigned values and negative padding makes little sense here anyway. Signed-off-by: Robin Murphy Signed-off-by: David Woodhouse --- drivers/iommu/intel-iommu.c | 2 ++ drivers/iommu/iova.c | 23 ++++++----------------- 2 files changed, 8 insertions(+), 17 deletions(-) diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index a98a7b27aca1..92101597cede 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -3233,6 +3233,8 @@ static struct iova *intel_alloc_iova(struct device *dev, /* Restrict dma_mask to the width that the iommu can handle */ dma_mask = min_t(uint64_t, DOMAIN_MAX_ADDR(domain->gaw), dma_mask); + /* Ensure we reserve the whole size-aligned region */ + nrpages = __roundup_pow_of_two(nrpages); if (!dmar_forcedac && dma_mask > DMA_BIT_MASK(32)) { /* diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c index b7c3d923f3e1..29f2efcf668e 100644 --- a/drivers/iommu/iova.c +++ b/drivers/iommu/iova.c @@ -120,19 +120,14 @@ __cached_rbnode_delete_update(struct iova_domain *iovad, struct iova *free) } } -/* Computes the padding size required, to make the - * the start address naturally aligned on its size +/* + * Computes the padding size required, to make the start address + * naturally aligned on the power-of-two order of its size */ -static int -iova_get_pad_size(int size, unsigned int limit_pfn) +static unsigned int +iova_get_pad_size(unsigned int size, unsigned int limit_pfn) { - unsigned int pad_size = 0; - unsigned int order = ilog2(size); - - if (order) - pad_size = (limit_pfn + 1) % (1 << order); - - return pad_size; + return (limit_pfn + 1 - size) & (__roundup_pow_of_two(size) - 1); } static int __alloc_and_insert_iova_range(struct iova_domain *iovad, @@ -265,12 +260,6 @@ alloc_iova(struct iova_domain *iovad, unsigned long size, if (!new_iova) return NULL; - /* If size aligned is set then round the size to - * to next power of two. - */ - if (size_aligned) - size = __roundup_pow_of_two(size); - ret = __alloc_and_insert_iova_range(iovad, size, limit_pfn, new_iova, size_aligned); From ae1ff3d623905947158fd3394854c23026337810 Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Mon, 13 Jul 2015 14:31:28 +0300 Subject: [PATCH 003/647] iommu: iova: Move iova cache management to the iova library This is necessary to separate intel-iommu from the iova library. Signed-off-by: Sakari Ailus Signed-off-by: David Woodhouse --- drivers/iommu/intel-iommu.c | 6 +-- drivers/iommu/iova.c | 83 ++++++++++++++++++++++--------------- include/linux/iova.h | 4 +- 3 files changed, 54 insertions(+), 39 deletions(-) diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 92101597cede..2d5cf39e1053 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -3743,7 +3743,7 @@ static inline int iommu_devinfo_cache_init(void) static int __init iommu_init_mempool(void) { int ret; - ret = iommu_iova_cache_init(); + ret = iova_cache_get(); if (ret) return ret; @@ -3757,7 +3757,7 @@ static int __init iommu_init_mempool(void) kmem_cache_destroy(iommu_domain_cache); domain_error: - iommu_iova_cache_destroy(); + iova_cache_put(); return -ENOMEM; } @@ -3766,7 +3766,7 @@ static void __init iommu_exit_mempool(void) { kmem_cache_destroy(iommu_devinfo_cache); kmem_cache_destroy(iommu_domain_cache); - iommu_iova_cache_destroy(); + iova_cache_put(); } static void quirk_ioat_snb_local_iommu(struct pci_dev *pdev) diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c index 29f2efcf668e..ed95f7a0fad3 100644 --- a/drivers/iommu/iova.c +++ b/drivers/iommu/iova.c @@ -20,40 +20,6 @@ #include #include -static struct kmem_cache *iommu_iova_cache; - -int iommu_iova_cache_init(void) -{ - int ret = 0; - - iommu_iova_cache = kmem_cache_create("iommu_iova", - sizeof(struct iova), - 0, - SLAB_HWCACHE_ALIGN, - NULL); - if (!iommu_iova_cache) { - pr_err("Couldn't create iova cache\n"); - ret = -ENOMEM; - } - - return ret; -} - -void iommu_iova_cache_destroy(void) -{ - kmem_cache_destroy(iommu_iova_cache); -} - -struct iova *alloc_iova_mem(void) -{ - return kmem_cache_alloc(iommu_iova_cache, GFP_ATOMIC); -} - -void free_iova_mem(struct iova *iova) -{ - kmem_cache_free(iommu_iova_cache, iova); -} - void init_iova_domain(struct iova_domain *iovad, unsigned long granule, unsigned long start_pfn, unsigned long pfn_32bit) @@ -237,6 +203,55 @@ iova_insert_rbtree(struct rb_root *root, struct iova *iova) rb_insert_color(&iova->node, root); } +static struct kmem_cache *iova_cache; +static unsigned int iova_cache_users; +static DEFINE_MUTEX(iova_cache_mutex); + +struct iova *alloc_iova_mem(void) +{ + return kmem_cache_alloc(iova_cache, GFP_ATOMIC); +} +EXPORT_SYMBOL(alloc_iova_mem); + +void free_iova_mem(struct iova *iova) +{ + kmem_cache_free(iova_cache, iova); +} +EXPORT_SYMBOL(free_iova_mem); + +int iova_cache_get(void) +{ + mutex_lock(&iova_cache_mutex); + if (!iova_cache_users) { + iova_cache = kmem_cache_create( + "iommu_iova", sizeof(struct iova), 0, + SLAB_HWCACHE_ALIGN, NULL); + if (!iova_cache) { + mutex_unlock(&iova_cache_mutex); + printk(KERN_ERR "Couldn't create iova cache\n"); + return -ENOMEM; + } + } + + iova_cache_users++; + mutex_unlock(&iova_cache_mutex); + + return 0; +} + +void iova_cache_put(void) +{ + mutex_lock(&iova_cache_mutex); + if (WARN_ON(!iova_cache_users)) { + mutex_unlock(&iova_cache_mutex); + return; + } + iova_cache_users--; + if (!iova_cache_users) + kmem_cache_destroy(iova_cache); + mutex_unlock(&iova_cache_mutex); +} + /** * alloc_iova - allocates an iova * @iovad: - iova domain in question diff --git a/include/linux/iova.h b/include/linux/iova.h index 3920a19d8194..92f7177db2ce 100644 --- a/include/linux/iova.h +++ b/include/linux/iova.h @@ -68,8 +68,8 @@ static inline unsigned long iova_pfn(struct iova_domain *iovad, dma_addr_t iova) return iova >> iova_shift(iovad); } -int iommu_iova_cache_init(void); -void iommu_iova_cache_destroy(void); +int iova_cache_get(void); +void iova_cache_put(void); struct iova *alloc_iova_mem(void); void free_iova_mem(struct iova *iova); From 9b41760b03816b34f4c9eee2cbb8fda8439920fc Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Mon, 13 Jul 2015 14:31:29 +0300 Subject: [PATCH 004/647] iommu: iova: Export symbols Use EXPORT_SYMBOL_GPL() to export the iova library symbols. The symbols include: init_iova_domain(); iova_cache_get(); iova_cache_put(); iova_cache_init(); alloc_iova(); find_iova(); __free_iova(); free_iova(); put_iova_domain(); reserve_iova(); copy_reserved_iova(); Signed-off-by: Sakari Ailus Signed-off-by: David Woodhouse --- drivers/iommu/iova.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c index ed95f7a0fad3..cbd74c79b212 100644 --- a/drivers/iommu/iova.c +++ b/drivers/iommu/iova.c @@ -38,6 +38,7 @@ init_iova_domain(struct iova_domain *iovad, unsigned long granule, iovad->start_pfn = start_pfn; iovad->dma_32bit_pfn = pfn_32bit; } +EXPORT_SYMBOL_GPL(init_iova_domain); static struct rb_node * __get_cached_rbnode(struct iova_domain *iovad, unsigned long *limit_pfn) @@ -238,6 +239,7 @@ int iova_cache_get(void) return 0; } +EXPORT_SYMBOL_GPL(iova_cache_get); void iova_cache_put(void) { @@ -251,6 +253,7 @@ void iova_cache_put(void) kmem_cache_destroy(iova_cache); mutex_unlock(&iova_cache_mutex); } +EXPORT_SYMBOL_GPL(iova_cache_put); /** * alloc_iova - allocates an iova @@ -285,6 +288,7 @@ alloc_iova(struct iova_domain *iovad, unsigned long size, return new_iova; } +EXPORT_SYMBOL_GPL(alloc_iova); /** * find_iova - find's an iova for a given pfn @@ -325,6 +329,7 @@ struct iova *find_iova(struct iova_domain *iovad, unsigned long pfn) spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags); return NULL; } +EXPORT_SYMBOL_GPL(find_iova); /** * __free_iova - frees the given iova @@ -343,6 +348,7 @@ __free_iova(struct iova_domain *iovad, struct iova *iova) spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags); free_iova_mem(iova); } +EXPORT_SYMBOL_GPL(__free_iova); /** * free_iova - finds and frees the iova for a given pfn @@ -360,6 +366,7 @@ free_iova(struct iova_domain *iovad, unsigned long pfn) __free_iova(iovad, iova); } +EXPORT_SYMBOL_GPL(free_iova); /** * put_iova_domain - destroys the iova doamin @@ -382,6 +389,7 @@ void put_iova_domain(struct iova_domain *iovad) } spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags); } +EXPORT_SYMBOL_GPL(put_iova_domain); static int __is_range_overlap(struct rb_node *node, @@ -471,6 +479,7 @@ finish: spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags); return iova; } +EXPORT_SYMBOL_GPL(reserve_iova); /** * copy_reserved_iova - copies the reserved between domains @@ -497,6 +506,7 @@ copy_reserved_iova(struct iova_domain *from, struct iova_domain *to) } spin_unlock_irqrestore(&from->iova_rbtree_lock, flags); } +EXPORT_SYMBOL_GPL(copy_reserved_iova); struct iova * split_and_remove_iova(struct iova_domain *iovad, struct iova *iova, From 15bbdec3931e617231c12b0920e497e87ec8c2c6 Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Mon, 13 Jul 2015 14:31:30 +0300 Subject: [PATCH 005/647] iommu: Make the iova library a module The iova library has use outside the intel-iommu driver, thus make it a module. Signed-off-by: Sakari Ailus Signed-off-by: David Woodhouse --- drivers/iommu/Kconfig | 2 +- drivers/iommu/iova.c | 4 ++++ 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig index f1fb1d3ccc56..0a141cd44ddd 100644 --- a/drivers/iommu/Kconfig +++ b/drivers/iommu/Kconfig @@ -42,7 +42,7 @@ config IOMMU_IO_PGTABLE_LPAE_SELFTEST endmenu config IOMMU_IOVA - bool + tristate config OF_IOMMU def_bool y diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c index cbd74c79b212..fa0adef32bd6 100644 --- a/drivers/iommu/iova.c +++ b/drivers/iommu/iova.c @@ -18,6 +18,7 @@ */ #include +#include #include void @@ -548,3 +549,6 @@ error: free_iova_mem(prev); return NULL; } + +MODULE_AUTHOR("Anil S Keshavamurthy "); +MODULE_LICENSE("GPL"); From ca9f26a27949ba3b295e4f0841c0bec9ef440141 Mon Sep 17 00:00:00 2001 From: Leilk Liu Date: Mon, 31 Aug 2015 21:18:56 +0800 Subject: [PATCH 006/647] spi: mediatek: remove clk_disable_unprepare() This patch removes clk_disable_unprepare() in mtk_spi_remove(). clk_disable_prepare/unprepare must be balance, spi-clk is disabled in mtk_spi_probe, so not needs to disable again. Signed-off-by: Leilk Liu Signed-off-by: Mark Brown --- drivers/spi/spi-mt65xx.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/spi/spi-mt65xx.c b/drivers/spi/spi-mt65xx.c index 5f6315c47920..b609b1cbdcba 100644 --- a/drivers/spi/spi-mt65xx.c +++ b/drivers/spi/spi-mt65xx.c @@ -630,7 +630,6 @@ static int mtk_spi_remove(struct platform_device *pdev) pm_runtime_disable(&pdev->dev); mtk_spi_reset(mdata); - clk_disable_unprepare(mdata->spi_clk); spi_master_put(master); return 0; From adcbcfea15d62fab5ba40ac28f9d2a590cc5e5e8 Mon Sep 17 00:00:00 2001 From: Leilk Liu Date: Mon, 31 Aug 2015 21:18:57 +0800 Subject: [PATCH 007/647] spi: mediatek: fix spi clock usage error spi clock manages flow: CLK_TOP_SYSPLL3_D2 ---> CLK_TOP_SPI_SEL ---> CLK_PERI_SPI0 (source clock) (clock mux) (clock gate) spi driver should choose source clock by clock mux, then enable clock gate. Signed-off-by: Leilk Liu Signed-off-by: Mark Brown --- drivers/spi/spi-mt65xx.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) diff --git a/drivers/spi/spi-mt65xx.c b/drivers/spi/spi-mt65xx.c index b609b1cbdcba..6fbb5e5132b0 100644 --- a/drivers/spi/spi-mt65xx.c +++ b/drivers/spi/spi-mt65xx.c @@ -85,7 +85,7 @@ struct mtk_spi { void __iomem *base; u32 state; u32 pad_sel; - struct clk *spi_clk, *parent_clk; + struct clk *parent_clk, *sel_clk, *spi_clk; struct spi_transfer *cur_transfer; u32 xfer_len; struct scatterlist *tx_sgl, *rx_sgl; @@ -576,13 +576,6 @@ static int mtk_spi_probe(struct platform_device *pdev) goto err_put_master; } - mdata->spi_clk = devm_clk_get(&pdev->dev, "spi-clk"); - if (IS_ERR(mdata->spi_clk)) { - ret = PTR_ERR(mdata->spi_clk); - dev_err(&pdev->dev, "failed to get spi-clk: %d\n", ret); - goto err_put_master; - } - mdata->parent_clk = devm_clk_get(&pdev->dev, "parent-clk"); if (IS_ERR(mdata->parent_clk)) { ret = PTR_ERR(mdata->parent_clk); @@ -590,13 +583,27 @@ static int mtk_spi_probe(struct platform_device *pdev) goto err_put_master; } + mdata->sel_clk = devm_clk_get(&pdev->dev, "sel-clk"); + if (IS_ERR(mdata->sel_clk)) { + ret = PTR_ERR(mdata->spi_clk); + dev_err(&pdev->dev, "failed to get sel-clk: %d\n", ret); + goto err_put_master; + } + + mdata->spi_clk = devm_clk_get(&pdev->dev, "spi-clk"); + if (IS_ERR(mdata->spi_clk)) { + ret = PTR_ERR(mdata->parent_clk); + dev_err(&pdev->dev, "failed to get spi-clk: %d\n", ret); + goto err_put_master; + } + ret = clk_prepare_enable(mdata->spi_clk); if (ret < 0) { dev_err(&pdev->dev, "failed to enable spi_clk (%d)\n", ret); goto err_put_master; } - ret = clk_set_parent(mdata->spi_clk, mdata->parent_clk); + ret = clk_set_parent(mdata->sel_clk, mdata->parent_clk); if (ret < 0) { dev_err(&pdev->dev, "failed to clk_set_parent (%d)\n", ret); goto err_disable_clk; From 3d4fe182003bcde778e29e84c14c0c4bb70a452e Mon Sep 17 00:00:00 2001 From: Leilk Liu Date: Mon, 31 Aug 2015 21:18:58 +0800 Subject: [PATCH 008/647] spi: Mediatek: Document devicetree bindings update for spi bus This patch updates spi bindings, fixs clock usage description. Signed-off-by: Leilk Liu Signed-off-by: Mark Brown --- .../devicetree/bindings/spi/spi-mt65xx.txt | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/Documentation/devicetree/bindings/spi/spi-mt65xx.txt b/Documentation/devicetree/bindings/spi/spi-mt65xx.txt index dcefc438272f..6160ffbcb3d3 100644 --- a/Documentation/devicetree/bindings/spi/spi-mt65xx.txt +++ b/Documentation/devicetree/bindings/spi/spi-mt65xx.txt @@ -15,17 +15,18 @@ Required properties: - interrupts: Should contain spi interrupt - clocks: phandles to input clocks. - The first should be <&topckgen CLK_TOP_SPI_SEL>. - The second should be one of the following. + The first should be one of the following. It's PLL. - <&clk26m>: specify parent clock 26MHZ. - <&topckgen CLK_TOP_SYSPLL3_D2>: specify parent clock 109MHZ. It's the default one. - <&topckgen CLK_TOP_SYSPLL4_D2>: specify parent clock 78MHZ. - <&topckgen CLK_TOP_UNIVPLL2_D4>: specify parent clock 104MHZ. - <&topckgen CLK_TOP_UNIVPLL1_D8>: specify parent clock 78MHZ. + The second should be <&topckgen CLK_TOP_SPI_SEL>. It's clock mux. + The third is <&pericfg CLK_PERI_SPI0>. It's clock gate. -- clock-names: shall be "spi-clk" for the controller clock, and - "parent-clk" for the parent clock. +- clock-names: shall be "parent-clk" for the parent clock, "sel-clk" for the + muxes clock, and "spi-clk" for the clock gate. Optional properties: - mediatek,pad-select: specify which pins group(ck/mi/mo/cs) spi @@ -44,8 +45,11 @@ spi: spi@1100a000 { #size-cells = <0>; reg = <0 0x1100a000 0 0x1000>; interrupts = ; - clocks = <&topckgen CLK_TOP_SPI_SEL>, <&topckgen CLK_TOP_SYSPLL3_D2>; - clock-names = "spi-clk", "parent-clk"; + clocks = <&topckgen CLK_TOP_SYSPLL3_D2>, + <&topckgen CLK_TOP_SPI_SEL>, + <&pericfg CLK_PERI_SPI0>; + clock-names = "parent-clk", "sel-clk", "spi-clk"; + mediatek,pad-select = <0>; status = "disabled"; }; From 6bd3c6f75e0f9baddbf1196a7e3fceabb50c7e3c Mon Sep 17 00:00:00 2001 From: "Maciej S. Szmigiero" Date: Mon, 31 Aug 2015 17:07:12 +0200 Subject: [PATCH 009/647] ASoC: fsl-asoc-card: put ASRC OF node in case of unknown device In case of unknown DT compatible device the ASRC OF node possibly acquired earlier by of_parse_phandle() has to be put before returning from probe method. Signed-off-by: Maciej Szmigiero Signed-off-by: Mark Brown --- sound/soc/fsl/fsl-asoc-card.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/sound/soc/fsl/fsl-asoc-card.c b/sound/soc/fsl/fsl-asoc-card.c index 5aeb6ed4827e..96f55ae75c71 100644 --- a/sound/soc/fsl/fsl-asoc-card.c +++ b/sound/soc/fsl/fsl-asoc-card.c @@ -488,7 +488,8 @@ static int fsl_asoc_card_probe(struct platform_device *pdev) priv->dai_fmt |= SND_SOC_DAIFMT_CBM_CFM; } else { dev_err(&pdev->dev, "unknown Device Tree compatible\n"); - return -EINVAL; + ret = -EINVAL; + goto asrc_fail; } /* Common settings for corresponding Freescale CPU DAI driver */ From d76f41982f2fc88492efd96c7c3178044f32e125 Mon Sep 17 00:00:00 2001 From: Anatol Pomozov Date: Mon, 31 Aug 2015 08:24:23 -0700 Subject: [PATCH 010/647] ASoC: Document snd-soc-dummy-dai purpose Signed-off-by: Anatol Pomozov Signed-off-by: Mark Brown --- sound/soc/soc-utils.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/sound/soc/soc-utils.c b/sound/soc/soc-utils.c index 362c69ac1d6c..53dd085d3ee2 100644 --- a/sound/soc/soc-utils.c +++ b/sound/soc/soc-utils.c @@ -101,6 +101,15 @@ static struct snd_soc_codec_driver dummy_codec; SNDRV_PCM_FMTBIT_S32_LE | \ SNDRV_PCM_FMTBIT_U32_LE | \ SNDRV_PCM_FMTBIT_IEC958_SUBFRAME_LE) +/* + * The dummy CODEC is only meant to be used in situations where there is no + * actual hardware. + * + * If there is actual hardware even if it does not have a control bus + * the hardware will still have constraints like supported samplerates, etc. + * which should be modelled. And the data flow graph also should be modelled + * using DAPM. + */ static struct snd_soc_dai_driver dummy_dai = { .name = "snd-soc-dummy-dai", .playback = { From 51d2eeef1d958ef6834b24f548194f5acea0f499 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Tue, 1 Sep 2015 11:14:05 +0530 Subject: [PATCH 011/647] ASoC: wm0010: fix memory leak We were aborting if the kzalloc of img_swap fails but without freeing the already allocated out. Similarly we were aborting if spi_sync fails without releasing out and img_swap. Signed-off-by: Sudip Mukherjee Acked-by: Charles Keepax Signed-off-by: Mark Brown --- sound/soc/codecs/wm0010.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/sound/soc/codecs/wm0010.c b/sound/soc/codecs/wm0010.c index 6560a66b3f35..9d370a45abe8 100644 --- a/sound/soc/codecs/wm0010.c +++ b/sound/soc/codecs/wm0010.c @@ -672,8 +672,10 @@ static int wm0010_boot(struct snd_soc_codec *codec) } img_swap = kzalloc(len, GFP_KERNEL | GFP_DMA); - if (!img_swap) + if (!img_swap) { + kfree(out); goto abort; + } /* We need to re-order for 0010 */ byte_swap_64((u64 *)&pll_rec, img_swap, len); @@ -690,6 +692,8 @@ static int wm0010_boot(struct snd_soc_codec *codec) ret = spi_sync(spi, &m); if (ret != 0) { dev_err(codec->dev, "First PLL write failed: %d\n", ret); + kfree(img_swap); + kfree(out); goto abort; } @@ -697,6 +701,8 @@ static int wm0010_boot(struct snd_soc_codec *codec) ret = spi_sync(spi, &m); if (ret != 0) { dev_err(codec->dev, "Second PLL write failed: %d\n", ret); + kfree(img_swap); + kfree(out); goto abort; } From fe8291e82b5d0b79b743683e5c0fb7e273f7a99a Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Tue, 1 Sep 2015 13:59:24 -0700 Subject: [PATCH 012/647] ARM: OMAP2+: omap-device: fix race deferred probe of omap_hsmmc vs omap_device_late_init Kernel fails to boot 50% of times (form build to build) with RT-patchset applied due to the following race - on late boot stages deferred_probe_work_func->omap_hsmmc_probe races with omap_device_late_ini. The same issue has been reported now on linux-next (4.3) by Keerthy [1] late_initcall - deferred_probe_initcal() tries to re-probe all pending driver's probe. - later on, some driver is probing in this case It's cpsw.c (but could be any other drivers) cpsw_init - platform_driver_register - really_probe - driver_bound - driver_deferred_probe_trigger and boot proceed. So, at this moment we have deferred_probe_work_func scheduled. late_initcall_sync - omap_device_late_init - omap_device_idle CPU1 CPU2 - deferred_probe_work_func - really_probe - omap_hsmmc_probe - pm_runtime_get_sync late_initcall_sync - omap_device_late_init if (od->_driver_status != BUS_NOTIFY_BOUND_DRIVER) { if (od->_state == OMAP_DEVICE_STATE_ENABLED) { - omap_device_idle [ops - IP is disabled] - [fail] - pm_runtime_put_sync - omap_hsmmc_runtime_suspend [ooops!] == log == omap_hsmmc 480b4000.mmc: unable to get vmmc regulator -517 davinci_mdio 48485000.mdio: davinci mdio revision 1.6 davinci_mdio 48485000.mdio: detected phy mask fffffff3 libphy: 48485000.mdio: probed davinci_mdio 48485000.mdio: phy[2]: device 48485000.mdio:02, driver unknown davinci_mdio 48485000.mdio: phy[3]: device 48485000.mdio:03, driver unknown omap_hsmmc 480b4000.mmc: unable to get vmmc regulator -517 cpsw 48484000.ethernet: Detected MACID = b4:99:4c:c7:d2:48 cpsw 48484000.ethernet: cpsw: Detected MACID = b4:99:4c:c7:d2:49 hctosys: unable to open rtc device (rtc0) omap_hsmmc 480b4000.mmc: omap_device_late_idle: enabled but no driver. Idling ldousb: disabling Unhandled fault: imprecise external abort (0x1406) at 0x00000000 [00000000] *pgd=00000000 Internal error: : 1406 [#1] PREEMPT SMP ARM Modules linked in: CPU: 1 PID: 58 Comm: kworker/u4:1 Not tainted 4.1.2-rt1-00467-g6da3c0a-dirty #5 Hardware name: Generic DRA74X (Flattened Device Tree) Workqueue: deferwq deferred_probe_work_func task: ee6ddb00 ti: edd3c000 task.ti: edd3c000 PC is at omap_hsmmc_runtime_suspend+0x1c/0x12c LR is at _od_runtime_suspend+0xc/0x24 pc : [] lr : [] psr: a0000013 sp : edd3dda0 ip : ee6ddb00 fp : c07be540 r10: 00000000 r9 : c07be540 r8 : 00000008 r7 : 00000000 r6 : ee646c10 r5 : ee646c10 r4 : edd79380 r3 : fa0b4100 r2 : 00000000 r1 : 00000000 r0 : ee646c10 Flags: NzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment kernel Control: 10c5387d Table: 8000406a DAC: 00000015 Process kworker/u4:1 (pid: 58, stack limit = 0xedd3c218) Stack: (0xedd3dda0 to 0xedd3e000) dda0: ee646c70 ee646c10 c0029584 00000000 00000008 c0029590 ee646c70 ee646c10 ddc0: c0029584 c03adfb8 ee646c10 00000004 0000000c c03adff0 ee646c10 00000004 dde0: 0000000c c03ae4ec 00000000 edd3c000 ee646c10 00000004 ee646c70 00000004 de00: fa0b4000 c03aec20 ee6ddb00 ee646c10 00000004 ee646c70 ee646c10 fffffdfb de20: edd79380 00000000 fa0b4000 c03aee90 fffffdfb edd79000 ee646c00 c0474290 de40: 00000000 edda24c0 edd79380 edc81f00 00000000 00000200 00000001 c06dd488 de60: edda3960 ee646c10 ee646c10 c0824cc4 fffffdfb c0880c94 00000002 edc92600 de80: c0836378 c03a7f84 ee646c10 c0824cc4 00000000 c0880c80 c0880c94 c03a6568 dea0: 00000000 ee646c10 c03a66ac ee4f8000 00000000 00000001 edc92600 c03a4b40 dec0: ee404c94 edc83c4c ee646c10 ee646c10 ee646c44 c03a63c4 ee646c10 ee646c10 dee0: c0814448 c03a5aa8 ee646c10 c0814220 edd3c000 c03a5ec0 c0814250 ee6be400 df00: edd3c000 c004e5bc ee6ddb01 00000078 ee6ddb00 ee4f8000 ee6be418 edd3c000 df20: ee4f8028 00000088 c0836045 ee4f8000 ee6be400 c004e928 ee4f8028 00000000 df40: c004e8ec 00000000 ee6bf1c0 ee6be400 c004e8ec 00000000 00000000 00000000 df60: 00000000 c0053450 2e56fa97 00000000 afdffbd7 ee6be400 00000000 00000000 df80: edd3df80 edd3df80 00000000 00000000 edd3df90 edd3df90 edd3dfac ee6bf1c0 dfa0: c0053384 00000000 00000000 c000f668 00000000 00000000 00000000 00000000 dfc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 dfe0: 00000000 00000000 00000000 00000000 00000013 00000000 f1fc9d7e febfbdff [] (omap_hsmmc_runtime_suspend) from [] (_od_runtime_suspend+0xc/0x24) [] (_od_runtime_suspend) from [] (__rpm_callback+0x24/0x3c) [] (__rpm_callback) from [] (rpm_callback+0x20/0x80) [] (rpm_callback) from [] (rpm_suspend+0xe4/0x618) [] (rpm_suspend) from [] (__pm_runtime_idle+0x60/0x80) [] (__pm_runtime_idle) from [] (omap_hsmmc_probe+0x6bc/0xa7c) [] (omap_hsmmc_probe) from [] (platform_drv_probe+0x44/0xa4) [] (platform_drv_probe) from [] (driver_probe_device+0x170/0x2b4) [] (driver_probe_device) from [] (bus_for_each_drv+0x64/0x98) [] (bus_for_each_drv) from [] (device_attach+0x70/0x88) [] (device_attach) from [] (bus_probe_device+0x84/0xac) [] (bus_probe_device) from [] (deferred_probe_work_func+0x58/0x88) [] (deferred_probe_work_func) from [] (process_one_work+0x134/0x464) [] (process_one_work) from [] (worker_thread+0x3c/0x4fc) [] (worker_thread) from [] (kthread+0xcc/0xe4) [] (kthread) from [] (ret_from_fork+0x14/0x2c) Code: e594302c e593202c e584205c e594302c (e5932128) ---[ end trace 0000000000000002 ]--- The issue happens because omap_device_late_init() do not take into account that some drivers are present, but their probes were not finished successfully and where deferred instead. This is the valid case, and omap_device_late_init() should not idle such devices. To fix this issue, the value of omap_device->_driver_status field should be checked not only for BUS_NOTIFY_BOUND_DRIVER (driver is present and has been bound to device successfully), but also checked for BUS_NOTIFY_BIND_DRIVER (driver about to be bound) - which means driver is present and there was try to bind it to device. [1] http://www.spinics.net/lists/arm-kernel/msg441880.html Cc: Tero Kristo Cc: Keerthy Tested-by: Keerthy Signed-off-by: Grygorii Strashko Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/omap_device.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-omap2/omap_device.c b/arch/arm/mach-omap2/omap_device.c index 4cb8fd9f741f..72ebc4c16bae 100644 --- a/arch/arm/mach-omap2/omap_device.c +++ b/arch/arm/mach-omap2/omap_device.c @@ -901,7 +901,8 @@ static int __init omap_device_late_idle(struct device *dev, void *data) if (od->hwmods[i]->flags & HWMOD_INIT_NO_IDLE) return 0; - if (od->_driver_status != BUS_NOTIFY_BOUND_DRIVER) { + if (od->_driver_status != BUS_NOTIFY_BOUND_DRIVER && + od->_driver_status != BUS_NOTIFY_BIND_DRIVER) { if (od->_state == OMAP_DEVICE_STATE_ENABLED) { dev_warn(dev, "%s: enabled but no driver. Idling\n", __func__); From 874b300a6a0bfe97225ed1f8648fa92238e4d9e9 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Tue, 1 Sep 2015 13:59:25 -0700 Subject: [PATCH 013/647] ARM: OMAP2+: Fix booting if no timer parent clock is available When bringing up a new SoC we needlessly prevent booting at timer init if timer clock_set_parent fails. This can fail if the system is booting on bootloader configured PLL values until the clock framework driver for the PLL is implemented. Let's just WARN instead, this will provide helpful information for anybody bringing up a new SoC what needs to be fixed. This allows to boot dm814x that's still missing the PLL driver. Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/timer.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/arch/arm/mach-omap2/timer.c b/arch/arm/mach-omap2/timer.c index e4d8701f99f9..a55655127ef2 100644 --- a/arch/arm/mach-omap2/timer.c +++ b/arch/arm/mach-omap2/timer.c @@ -297,12 +297,8 @@ static int __init omap_dm_timer_init_one(struct omap_dm_timer *timer, if (IS_ERR(src)) return PTR_ERR(src); - r = clk_set_parent(timer->fclk, src); - if (r < 0) { - pr_warn("%s: %s cannot set source\n", __func__, oh->name); - clk_put(src); - return r; - } + WARN(clk_set_parent(timer->fclk, src) < 0, + "Cannot set timer parent clock, no PLL clock driver?"); clk_put(src); From 05f4f1abc274ac7e8e83424d0cef2d842db91815 Mon Sep 17 00:00:00 2001 From: Frans Klaver Date: Tue, 1 Sep 2015 13:59:25 -0700 Subject: [PATCH 014/647] ARM: OMAP3: vc: fix 'or' always true warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix the warning: arch/arm/mach-omap2/vc.c:302:47: warning: logical ‘or’ of collectively exhaustive tests is always true [-Wlogical-op] As we're toggling both CLKREQ and OFFMODE, we should also be checking OFFMODE. Signed-off-by: Frans Klaver Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/vc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-omap2/vc.c b/arch/arm/mach-omap2/vc.c index e5a35f6b83a7..d44d311704ba 100644 --- a/arch/arm/mach-omap2/vc.c +++ b/arch/arm/mach-omap2/vc.c @@ -300,7 +300,7 @@ static void __init omap3_vc_init_pmic_signaling(struct voltagedomain *voltdm) val = voltdm->read(OMAP3_PRM_POLCTRL_OFFSET); if (!(val & OMAP3430_PRM_POLCTRL_CLKREQ_POL) || - (val & OMAP3430_PRM_POLCTRL_CLKREQ_POL)) { + (val & OMAP3430_PRM_POLCTRL_OFFMODE_POL)) { val |= OMAP3430_PRM_POLCTRL_CLKREQ_POL; val &= ~OMAP3430_PRM_POLCTRL_OFFMODE_POL; pr_debug("PM: fixing sys_clkreq and sys_off_mode polarity to 0x%x\n", From 81032e34e184a8d63598d215a6d4a3352018ffea Mon Sep 17 00:00:00 2001 From: Vishal Mahaveer Date: Tue, 1 Sep 2015 13:59:25 -0700 Subject: [PATCH 015/647] ARM: DRA752: Add ID detect for ES2.0 ES2.0 is a minor variant of ES1.1. ES2.0 is an incremental revision with various fixes including the following: - reset logic fixes - few assymetric aging logic fixes - MMC clock rate fixes - Ethernet speed fixes - edma fixes for mcasp NOTE: even though we use a compatible of dra742 and dra752, the usage in the Linux kernel is more or less interchangable - we use dra752 more often in the linux kernel compared to dra742 and 4.2-rc6 Signed-off-by: Vishal Mahaveer Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/id.c | 8 ++++++-- arch/arm/mach-omap2/soc.h | 2 ++ 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-omap2/id.c b/arch/arm/mach-omap2/id.c index e3f713ffb06b..54a5ba54d2ff 100644 --- a/arch/arm/mach-omap2/id.c +++ b/arch/arm/mach-omap2/id.c @@ -653,8 +653,12 @@ void __init dra7xxx_check_revision(void) omap_revision = DRA752_REV_ES1_0; break; case 1: - default: omap_revision = DRA752_REV_ES1_1; + break; + case 2: + default: + omap_revision = DRA752_REV_ES2_0; + break; } break; @@ -674,7 +678,7 @@ void __init dra7xxx_check_revision(void) /* Unknown default to latest silicon rev as default*/ pr_warn("%s: unknown idcode=0x%08x (hawkeye=0x%08x,rev=0x%x)\n", __func__, idcode, hawkeye, rev); - omap_revision = DRA752_REV_ES1_1; + omap_revision = DRA752_REV_ES2_0; } sprintf(soc_name, "DRA%03x", omap_rev() >> 16); diff --git a/arch/arm/mach-omap2/soc.h b/arch/arm/mach-omap2/soc.h index f97654d11ea5..2d1d3845253c 100644 --- a/arch/arm/mach-omap2/soc.h +++ b/arch/arm/mach-omap2/soc.h @@ -469,6 +469,8 @@ IS_OMAP_TYPE(3430, 0x3430) #define DRA7XX_CLASS 0x07000000 #define DRA752_REV_ES1_0 (DRA7XX_CLASS | (0x52 << 16) | (0x10 << 8)) #define DRA752_REV_ES1_1 (DRA7XX_CLASS | (0x52 << 16) | (0x11 << 8)) +#define DRA752_REV_ES2_0 (DRA7XX_CLASS | (0x52 << 16) | (0x20 << 8)) +#define DRA722_REV_ES1_0 (DRA7XX_CLASS | (0x22 << 16) | (0x10 << 8)) #define DRA722_REV_ES1_0 (DRA7XX_CLASS | (0x22 << 16) | (0x10 << 8)) void omap2xxx_check_revision(void); From 21b430d23d233c67e6589ea5054d18392e15a28e Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Tue, 1 Sep 2015 13:59:25 -0700 Subject: [PATCH 016/647] ARM: omap2plus_defconfig: Enable MUSB DMA support With recent MUSB changes we can now build in support for multiple DMA implementations. So let's enable DMA by default to make life easier for distro use. Acked-by: Felipe Balbi Signed-off-by: Tony Lindgren --- arch/arm/configs/omap2plus_defconfig | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig index 50c84e1876fc..7b35e70d4bb3 100644 --- a/arch/arm/configs/omap2plus_defconfig +++ b/arch/arm/configs/omap2plus_defconfig @@ -350,6 +350,8 @@ CONFIG_USB_MUSB_HDRC=m CONFIG_USB_MUSB_OMAP2PLUS=m CONFIG_USB_MUSB_AM35X=m CONFIG_USB_MUSB_DSPS=m +CONFIG_USB_INVENTRA_DMA=y +CONFIG_USB_TI_CPPI41_DMA=y CONFIG_USB_DWC3=m CONFIG_USB_TEST=m CONFIG_AM335X_PHY_USB=y From 02bc933ebb59208f42c2e6305b2c17fd306f695d Mon Sep 17 00:00:00 2001 From: "Tan, Jui Nee" Date: Tue, 1 Sep 2015 10:22:51 +0800 Subject: [PATCH 017/647] spi: spi-pxa2xx: Check status register to determine if SSSR_TINT is disabled On Intel Baytrail, there is case when interrupt handler get called, no SPI message is captured. The RX FIFO is indeed empty when RX timeout pending interrupt (SSSR_TINT) happens. Use the BIOS version where both HSUART and SPI are on the same IRQ. Both drivers are using IRQF_SHARED when calling the request_irq function. When running two separate and independent SPI and HSUART application that generate data traffic on both components, user will see messages like below on the console: pxa2xx-spi pxa2xx-spi.0: bad message state in interrupt handler This commit will fix this by first checking Receiver Time-out Interrupt, if it is disabled, ignore the request and return without servicing. Signed-off-by: Tan, Jui Nee Acked-by: Jarkko Nikula Signed-off-by: Mark Brown Cc: stable@vger.kernel.org --- drivers/spi/spi-pxa2xx.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c index 7293d6d875c5..8e4b1a7c37ce 100644 --- a/drivers/spi/spi-pxa2xx.c +++ b/drivers/spi/spi-pxa2xx.c @@ -643,6 +643,10 @@ static irqreturn_t ssp_int(int irq, void *dev_id) if (!(sccr1_reg & SSCR1_TIE)) mask &= ~SSSR_TFS; + /* Ignore RX timeout interrupt if it is disabled */ + if (!(sccr1_reg & SSCR1_TINTE)) + mask &= ~SSSR_TINT; + if (!(status & mask)) return IRQ_NONE; From 9f8df6adf6c53abfa10e243cb289c34eb18ec581 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Wed, 2 Sep 2015 16:14:06 +0530 Subject: [PATCH 018/647] regulator: core: fix possible NULL dereference We were checking rdev->supply for NULL after dereferencing it. Lets check for rdev->supply along with _regulator_is_enabled() and call regulator_enable() only if rdev->supply is not NULL. Signed-off-by: Sudip Mukherjee Signed-off-by: Mark Brown --- drivers/regulator/core.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index de9f272a0faf..7150ff6ef46b 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -1422,11 +1422,10 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) return ret; /* Cascade always-on state to supply */ - if (_regulator_is_enabled(rdev)) { + if (_regulator_is_enabled(rdev) && rdev->supply) { ret = regulator_enable(rdev->supply); if (ret < 0) { - if (rdev->supply) - _regulator_put(rdev->supply); + _regulator_put(rdev->supply); return ret; } } From b9c93646fd5cb669d096fec5ad25a01f04cfde27 Mon Sep 17 00:00:00 2001 From: Kishon Vijay Abraham I Date: Thu, 3 Sep 2015 12:20:37 +0530 Subject: [PATCH 019/647] regulator: pbias: program pbias register offset in pbias driver Add separate compatible strings for every platform and populate the pbias register offset in the driver data. This helps avoid depending on the dt for pbias register offset. Also update the dt binding documentation for the new compatible strings. Suggested-by: Tony Lindgren Signed-off-by: Kishon Vijay Abraham I Acked-by: Tony Lindgren Signed-off-by: Mark Brown --- .../bindings/regulator/pbias-regulator.txt | 7 ++- drivers/regulator/pbias-regulator.c | 56 +++++++++++++++++-- 2 files changed, 56 insertions(+), 7 deletions(-) diff --git a/Documentation/devicetree/bindings/regulator/pbias-regulator.txt b/Documentation/devicetree/bindings/regulator/pbias-regulator.txt index 32aa26f1e434..acbcb452a69a 100644 --- a/Documentation/devicetree/bindings/regulator/pbias-regulator.txt +++ b/Documentation/devicetree/bindings/regulator/pbias-regulator.txt @@ -2,7 +2,12 @@ PBIAS internal regulator for SD card dual voltage i/o pads on OMAP SoCs. Required properties: - compatible: - - "ti,pbias-omap" for OMAP2, OMAP3, OMAP4, OMAP5, DRA7. + - should be "ti,pbias-dra7" for DRA7 + - should be "ti,pbias-omap2" for OMAP2 + - should be "ti,pbias-omap3" for OMAP3 + - should be "ti,pbias-omap4" for OMAP4 + - should be "ti,pbias-omap5" for OMAP5 + - "ti,pbias-omap" is deprecated - reg: pbias register offset from syscon base and size of pbias register. - syscon : phandle of the system control module - regulator-name : should be diff --git a/drivers/regulator/pbias-regulator.c b/drivers/regulator/pbias-regulator.c index bd2b75c0d1d1..c21cedbdf451 100644 --- a/drivers/regulator/pbias-regulator.c +++ b/drivers/regulator/pbias-regulator.c @@ -44,6 +44,10 @@ struct pbias_regulator_data { int voltage; }; +struct pbias_of_data { + unsigned int offset; +}; + static const unsigned int pbias_volt_table[] = { 1800000, 3000000 @@ -98,8 +102,35 @@ static struct of_regulator_match pbias_matches[] = { }; #define PBIAS_NUM_REGS ARRAY_SIZE(pbias_matches) +/* Offset from SCM general area (and syscon) base */ + +static const struct pbias_of_data pbias_of_data_omap2 = { + .offset = 0x230, +}; + +static const struct pbias_of_data pbias_of_data_omap3 = { + .offset = 0x2b0, +}; + +static const struct pbias_of_data pbias_of_data_omap4 = { + .offset = 0x60, +}; + +static const struct pbias_of_data pbias_of_data_omap5 = { + .offset = 0x60, +}; + +static const struct pbias_of_data pbias_of_data_dra7 = { + .offset = 0xe00, +}; + static const struct of_device_id pbias_of_match[] = { { .compatible = "ti,pbias-omap", }, + { .compatible = "ti,pbias-omap2", .data = &pbias_of_data_omap2, }, + { .compatible = "ti,pbias-omap3", .data = &pbias_of_data_omap3, }, + { .compatible = "ti,pbias-omap4", .data = &pbias_of_data_omap4, }, + { .compatible = "ti,pbias-omap5", .data = &pbias_of_data_omap5, }, + { .compatible = "ti,pbias-dra7", .data = &pbias_of_data_dra7, }, {}, }; MODULE_DEVICE_TABLE(of, pbias_of_match); @@ -114,6 +145,9 @@ static int pbias_regulator_probe(struct platform_device *pdev) const struct pbias_reg_info *info; int ret = 0; int count, idx, data_idx = 0; + const struct of_device_id *match; + const struct pbias_of_data *data; + unsigned int offset; count = of_regulator_match(&pdev->dev, np, pbias_matches, PBIAS_NUM_REGS); @@ -129,6 +163,20 @@ static int pbias_regulator_probe(struct platform_device *pdev) if (IS_ERR(syscon)) return PTR_ERR(syscon); + match = of_match_device(of_match_ptr(pbias_of_match), &pdev->dev); + if (match && match->data) { + data = match->data; + offset = data->offset; + } else { + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!res) + return -EINVAL; + + offset = res->start; + dev_WARN(&pdev->dev, + "using legacy dt data for pbias offset\n"); + } + cfg.regmap = syscon; cfg.dev = &pdev->dev; @@ -141,10 +189,6 @@ static int pbias_regulator_probe(struct platform_device *pdev) if (!info) return -ENODEV; - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!res) - return -EINVAL; - drvdata[data_idx].syscon = syscon; drvdata[data_idx].info = info; drvdata[data_idx].desc.name = info->name; @@ -154,9 +198,9 @@ static int pbias_regulator_probe(struct platform_device *pdev) drvdata[data_idx].desc.volt_table = pbias_volt_table; drvdata[data_idx].desc.n_voltages = 2; drvdata[data_idx].desc.enable_time = info->enable_time; - drvdata[data_idx].desc.vsel_reg = res->start; + drvdata[data_idx].desc.vsel_reg = offset; drvdata[data_idx].desc.vsel_mask = info->vmode; - drvdata[data_idx].desc.enable_reg = res->start; + drvdata[data_idx].desc.enable_reg = offset; drvdata[data_idx].desc.enable_mask = info->enable_mask; drvdata[data_idx].desc.enable_val = info->enable; From a394d635193b641f2c86ead5ada5b115d57c51f8 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sun, 6 Sep 2015 01:46:54 +0300 Subject: [PATCH 020/647] spi: Fix documentation of spi_alloc_master() Actually, spi_master_put() after spi_alloc_master() must _not_ be followed by kfree(). The memory is already freed with the call to spi_master_put() through spi_master_class, which registers a release function. Calling both spi_master_put() and kfree() results in often nasty (and delayed) crashes elsewhere in the kernel, often in the networking stack. This reverts commit eb4af0f5349235df2e4a5057a72fc8962d00308a. Link to patch and concerns: https://lkml.org/lkml/2012/9/3/269 or http://lkml.iu.edu/hypermail/linux/kernel/1209.0/00790.html Alexey Klimov: This revert becomes valid after 94c69f765f1b4a658d96905ec59928e3e3e07e6a when spi-imx.c has been fixed and there is no need to call kfree() so comment for spi_alloc_master() should be fixed. Signed-off-by: Guenter Roeck Signed-off-by: Alexey Klimov Signed-off-by: Mark Brown Cc: stable@vger.kernel.org --- drivers/spi/spi.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index cf8b91b23a76..9ce2f156d382 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -1437,8 +1437,7 @@ static struct class spi_master_class = { * * The caller is responsible for assigning the bus number and initializing * the master's methods before calling spi_register_master(); and (after errors - * adding the device) calling spi_master_put() and kfree() to prevent a memory - * leak. + * adding the device) calling spi_master_put() to prevent a memory leak. */ struct spi_master *spi_alloc_master(struct device *dev, unsigned size) { From 6583d2032d57df9f1c00c753ca58e1a822901bf0 Mon Sep 17 00:00:00 2001 From: Leilk Liu Date: Mon, 7 Sep 2015 19:37:57 +0800 Subject: [PATCH 021/647] spi: mediatek: fix spi cs polarity error Mediatek spi HW can't set cs inactive(keep cs high) directly. Instead, it supplies pause mode to do it indirectly. If driver unsets SPI_CMD_PAUSE_MODE in CMD_REG, it also needs to reset internal state machine to let cs inactive at once. Signed-off-by: Leilk Liu Signed-off-by: Mark Brown --- drivers/spi/spi-mt65xx.c | 27 +++++++-------------------- 1 file changed, 7 insertions(+), 20 deletions(-) diff --git a/drivers/spi/spi-mt65xx.c b/drivers/spi/spi-mt65xx.c index 5f6315c47920..b6073bbff8d9 100644 --- a/drivers/spi/spi-mt65xx.c +++ b/drivers/spi/spi-mt65xx.c @@ -173,22 +173,6 @@ static void mtk_spi_config(struct mtk_spi *mdata, writel(mdata->pad_sel, mdata->base + SPI_PAD_SEL_REG); } -static int mtk_spi_prepare_hardware(struct spi_master *master) -{ - struct spi_transfer *trans; - struct mtk_spi *mdata = spi_master_get_devdata(master); - struct spi_message *msg = master->cur_msg; - - trans = list_first_entry(&msg->transfers, struct spi_transfer, - transfer_list); - if (!trans->cs_change) { - mdata->state = MTK_SPI_IDLE; - mtk_spi_reset(mdata); - } - - return 0; -} - static int mtk_spi_prepare_message(struct spi_master *master, struct spi_message *msg) { @@ -228,11 +212,15 @@ static void mtk_spi_set_cs(struct spi_device *spi, bool enable) struct mtk_spi *mdata = spi_master_get_devdata(spi->master); reg_val = readl(mdata->base + SPI_CMD_REG); - if (!enable) + if (!enable) { reg_val |= SPI_CMD_PAUSE_EN; - else + writel(reg_val, mdata->base + SPI_CMD_REG); + } else { reg_val &= ~SPI_CMD_PAUSE_EN; - writel(reg_val, mdata->base + SPI_CMD_REG); + writel(reg_val, mdata->base + SPI_CMD_REG); + mdata->state = MTK_SPI_IDLE; + mtk_spi_reset(mdata); + } } static void mtk_spi_prepare_transfer(struct spi_master *master, @@ -509,7 +497,6 @@ static int mtk_spi_probe(struct platform_device *pdev) master->mode_bits = SPI_CPOL | SPI_CPHA; master->set_cs = mtk_spi_set_cs; - master->prepare_transfer_hardware = mtk_spi_prepare_hardware; master->prepare_message = mtk_spi_prepare_message; master->transfer_one = mtk_spi_transfer_one; master->can_dma = mtk_spi_can_dma; From 6af3e3adcac595a683bb55299a907d7d1ad61ab3 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 7 Sep 2015 10:41:12 +0100 Subject: [PATCH 022/647] Btrfs: don't initialize a space info as full to prevent ENOSPC Commit 2e6e518335f8 ("Btrfs: fix block group ->space_info null pointer dereference") accidently marked a space info as full when initializing it with a value of 0 total bytes. This introduces an ENOSPC problem when writing file data if we mount a filesystem that has no data block groups allocated, because the data space info is initialized with 0 total bytes, marked as full, and it never gets its total bytes incremented by a (positive) value to unmark it as full (because there are no data block groups loaded when the fs is mounted). For metadata and system spaces this issue can never happen since we always have at least one metadata block group and one system block group (even for an empty filesystem). So fix this by just not initializing a space info as full, reverting the offending part of the commit mentioned above. The following test case for fstests reproduces the issue: seq=`basename $0` seqres=$RESULT_DIR/$seq echo "QA output created by $seq" tmp=/tmp/$$ status=1 # failure is the default! trap "_cleanup; exit \$status" 0 1 2 3 15 _cleanup() { rm -f $tmp.* } # get standard environment, filters and checks . ./common/rc . ./common/filter # real QA test starts here _need_to_be_root _supported_fs btrfs _supported_os Linux _require_scratch rm -f $seqres.full _scratch_mkfs >>$seqres.full 2>&1 # Mount our filesystem without space caches enabled so that we do not # get any space used from the initial data block group that mkfs creates # (space caches used space from data block groups). _scratch_mount "-o nospace_cache" # Need an fs with at least 2Gb to make sure mkfs.btrfs does not create # an fs using mixed block groups (used both for data and metadata). We # really need to have dedicated block groups for data to reproduce the # issue and mkfs.btrfs defaults to mixed block groups only for small # filesystems (up to 1Gb). _require_fs_space $SCRATCH_MNT $((2 * 1024 * 1024)) # Run balance with the purpose of deleting the unused data block group # that mkfs created. We could also wait for the background kthread to # automatically delete the unused block group, but we do not have a way # to make it run and wait for it to complete, so just do a balance # instead of some unreliable sleep _run_btrfs_util_prog balance start -dusage=0 $SCRATCH_MNT # Now unmount the filesystem, mount it again (either with or with space # caches enabled, it does not matter to trigger the problem) and attempt # to create a file with some data - this used to fail with ENOSPC # because there were no data block groups when the filesystem was # mounted and the data space info object was marked as full when # initialized (because it had 0 total bytes), which prevented the file # write path from attempting to allocate a data block group and fail # immediately with ENOSPC. _scratch_remount echo "hello world" > $SCRATCH_MNT/foobar echo "Silence is golden" status=0 exit Signed-off-by: Filipe Manana --- fs/btrfs/extent-tree.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 5411f0ab5683..3686315bbd66 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3742,10 +3742,7 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags, found->bytes_reserved = 0; found->bytes_readonly = 0; found->bytes_may_use = 0; - if (total_bytes > 0) - found->full = 0; - else - found->full = 1; + found->full = 0; found->force_alloc = CHUNK_ALLOC_NO_FORCE; found->chunk_alloc = 0; found->flush = 0; From 721b51fcf91898299d96f4b72cb9434cda29dce6 Mon Sep 17 00:00:00 2001 From: John Lin Date: Wed, 9 Sep 2015 16:47:48 +0800 Subject: [PATCH 023/647] ASoC: rt5645: Add struct dmi_system_id "Google Ultima" for chrome platform Signed-off-by: John Lin Signed-off-by: Mark Brown --- sound/soc/codecs/rt5645.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index 610eacd34900..e620da164516 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -3236,6 +3236,13 @@ static struct dmi_system_id dmi_platform_intel_braswell[] = { DMI_MATCH(DMI_PRODUCT_NAME, "Celes"), }, }, + { + .ident = "Google Ultima", + .callback = strago_quirk_cb, + .matches = { + DMI_MATCH(DMI_PRODUCT_NAME, "Ultima"), + }, + }, { } }; From f1ec5ec7a94ba8138f9cbdc1e9e3b03aee29c4df Mon Sep 17 00:00:00 2001 From: Liam Girdwood Date: Wed, 9 Sep 2015 12:10:31 +0100 Subject: [PATCH 024/647] ASoC: intel: Fix SSP port configuration after RTD3 resume. Currently the SSP port settings are being clobbered as part of the DSP RTD3 restore logic. make sure we save the correct params and restore them at resume. The FW sadly does not save SSP settings as part of the PM context. Signed-off-by: Liam Girdwood Signed-off-by: Mark Brown --- sound/soc/intel/haswell/sst-haswell-ipc.c | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) diff --git a/sound/soc/intel/haswell/sst-haswell-ipc.c b/sound/soc/intel/haswell/sst-haswell-ipc.c index f95f271aab0c..5bc98550c81d 100644 --- a/sound/soc/intel/haswell/sst-haswell-ipc.c +++ b/sound/soc/intel/haswell/sst-haswell-ipc.c @@ -302,6 +302,10 @@ struct sst_hsw { struct sst_hsw_ipc_dx_reply dx; void *dx_context; dma_addr_t dx_context_paddr; + enum sst_hsw_device_id dx_dev; + enum sst_hsw_device_mclk dx_mclk; + enum sst_hsw_device_mode dx_mode; + u32 dx_clock_divider; /* boot */ wait_queue_head_t boot_wait; @@ -1400,10 +1404,10 @@ int sst_hsw_device_set_config(struct sst_hsw *hsw, trace_ipc_request("set device config", dev); - config.ssp_interface = dev; - config.clock_frequency = mclk; - config.mode = mode; - config.clock_divider = clock_divider; + hsw->dx_dev = config.ssp_interface = dev; + hsw->dx_mclk = config.clock_frequency = mclk; + hsw->dx_mode = config.mode = mode; + hsw->dx_clock_divider = config.clock_divider = clock_divider; if (mode == SST_HSW_DEVICE_TDM_CLOCK_MASTER) config.channels = 4; else @@ -1704,10 +1708,10 @@ int sst_hsw_dsp_runtime_resume(struct sst_hsw *hsw) return -EIO; } - /* Set ADSP SSP port settings */ - ret = sst_hsw_device_set_config(hsw, SST_HSW_DEVICE_SSP_0, - SST_HSW_DEVICE_MCLK_FREQ_24_MHZ, - SST_HSW_DEVICE_CLOCK_MASTER, 9); + /* Set ADSP SSP port settings - sadly the FW does not store SSP port + settings as part of the PM context. */ + ret = sst_hsw_device_set_config(hsw, hsw->dx_dev, hsw->dx_mclk, + hsw->dx_mode, hsw->dx_clock_divider); if (ret < 0) dev_err(dev, "error: SSP re-initialization failed\n"); From 2a3fffd45822070309bcf0b1e1dae624d633824a Mon Sep 17 00:00:00 2001 From: Martin Sperl Date: Thu, 10 Sep 2015 09:32:14 +0000 Subject: [PATCH 025/647] spi: bcm2835: BUG: fix wrong use of PAGE_MASK There is a bug in the alignment checking of transfers, that results in DMA not being used for un-aligned transfers that do not cross page-boundries, which is valid. This is due to a missconception of the meaning PAGE_MASK when implementing that check originally - (PAGE_SIZE - 1) should have been used instead. Also fixes a copy/paste error. Reported-by: Signed-off-by: Martin Sperl Signed-off-by: Mark Brown Cc: stable@vger.kernel.org --- drivers/spi/spi-bcm2835.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/spi/spi-bcm2835.c b/drivers/spi/spi-bcm2835.c index 59705ab23577..ed74786e68cb 100644 --- a/drivers/spi/spi-bcm2835.c +++ b/drivers/spi/spi-bcm2835.c @@ -386,14 +386,14 @@ static bool bcm2835_spi_can_dma(struct spi_master *master, /* otherwise we only allow transfers within the same page * to avoid wasting time on dma_mapping when it is not practical */ - if (((size_t)tfr->tx_buf & PAGE_MASK) + tfr->len > PAGE_SIZE) { + if (((size_t)tfr->tx_buf & (PAGE_SIZE - 1)) + tfr->len > PAGE_SIZE) { dev_warn_once(&spi->dev, "Unaligned spi tx-transfer bridging page\n"); return false; } - if (((size_t)tfr->rx_buf & PAGE_MASK) + tfr->len > PAGE_SIZE) { + if (((size_t)tfr->rx_buf & (PAGE_SIZE - 1)) + tfr->len > PAGE_SIZE) { dev_warn_once(&spi->dev, - "Unaligned spi tx-transfer bridging page\n"); + "Unaligned spi rx-transfer bridging page\n"); return false; } From d630526d0aa6acc0868dae892b1febda72029a3e Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Thu, 10 Sep 2015 10:19:52 +0200 Subject: [PATCH 026/647] spi: atmel: remove warning when !CONFIG_PM_SLEEP MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When CONFIG_PM is defined but not CONFIG_PM_SLEEP (this happens when CONFIG_SUSPEND is not defined), there is the following warning: drivers/spi/spi-atmel.c:1723:12: warning: ‘atmel_spi_suspend’ defined but not used [-Wunused-function] drivers/spi/spi-atmel.c:1741:12: warning: ‘atmel_spi_resume’ defined but not used [-Wunused-function] Enclose both atmel_spi_suspend and atmel_spi_resume in #ifdef CONFIG_PM_SLEEP/#endif to solve that. Signed-off-by: Alexandre Belloni Signed-off-by: Mark Brown --- drivers/spi/spi-atmel.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/spi/spi-atmel.c b/drivers/spi/spi-atmel.c index c9eca347787d..709fb6fa1c55 100644 --- a/drivers/spi/spi-atmel.c +++ b/drivers/spi/spi-atmel.c @@ -1721,6 +1721,7 @@ static int atmel_spi_runtime_resume(struct device *dev) return clk_prepare_enable(as->clk); } +#ifdef CONFIG_PM_SLEEP static int atmel_spi_suspend(struct device *dev) { struct spi_master *master = dev_get_drvdata(dev); @@ -1757,6 +1758,7 @@ static int atmel_spi_resume(struct device *dev) return ret; } +#endif static const struct dev_pm_ops atmel_spi_pm_ops = { SET_SYSTEM_SLEEP_PM_OPS(atmel_spi_suspend, atmel_spi_resume) From 3a0e27d84bb9abac5e39dc71706768a88c72cb71 Mon Sep 17 00:00:00 2001 From: Arnaud Pouliquen Date: Thu, 10 Sep 2015 09:45:55 +0200 Subject: [PATCH 027/647] ASoC: sti: check return of of_property_read Add check on of_property_read to return error when DT required property is not defined. Signed-off-by: Arnaud Pouliquen Signed-off-by: Mark Brown --- sound/soc/sti/uniperif_player.c | 14 ++++++++++---- sound/soc/sti/uniperif_reader.c | 6 +++++- 2 files changed, 15 insertions(+), 5 deletions(-) diff --git a/sound/soc/sti/uniperif_player.c b/sound/soc/sti/uniperif_player.c index f6eefe1b8f8f..843f037a317d 100644 --- a/sound/soc/sti/uniperif_player.c +++ b/sound/soc/sti/uniperif_player.c @@ -989,8 +989,8 @@ static int uni_player_parse_dt(struct platform_device *pdev, if (!info) return -ENOMEM; - of_property_read_u32(pnode, "version", &player->ver); - if (player->ver == SND_ST_UNIPERIF_VERSION_UNKNOWN) { + if (of_property_read_u32(pnode, "version", &player->ver) || + player->ver == SND_ST_UNIPERIF_VERSION_UNKNOWN) { dev_err(dev, "Unknown uniperipheral version "); return -EINVAL; } @@ -998,10 +998,16 @@ static int uni_player_parse_dt(struct platform_device *pdev, if (player->ver >= SND_ST_UNIPERIF_VERSION_UNI_PLR_TOP_1_0) info->underflow_enabled = 1; - of_property_read_u32(pnode, "uniperiph-id", &info->id); + if (of_property_read_u32(pnode, "uniperiph-id", &info->id)) { + dev_err(dev, "uniperipheral id not defined"); + return -EINVAL; + } /* Read the device mode property */ - of_property_read_string(pnode, "mode", &mode); + if (of_property_read_string(pnode, "mode", &mode)) { + dev_err(dev, "uniperipheral mode not defined"); + return -EINVAL; + } if (strcasecmp(mode, "hdmi") == 0) info->player_type = SND_ST_UNIPERIF_PLAYER_TYPE_HDMI; diff --git a/sound/soc/sti/uniperif_reader.c b/sound/soc/sti/uniperif_reader.c index c502626f339b..f791239a3087 100644 --- a/sound/soc/sti/uniperif_reader.c +++ b/sound/soc/sti/uniperif_reader.c @@ -316,7 +316,11 @@ static int uni_reader_parse_dt(struct platform_device *pdev, if (!info) return -ENOMEM; - of_property_read_u32(node, "version", &reader->ver); + if (of_property_read_u32(node, "version", &reader->ver) || + reader->ver == SND_ST_UNIPERIF_VERSION_UNKNOWN) { + dev_err(&pdev->dev, "Unknown uniperipheral version "); + return -EINVAL; + } /* Save the info structure */ reader->info = info; From 85e0a0f21a14bfd9145422a6a627c3df47101bd8 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 10 Sep 2015 10:36:06 +0100 Subject: [PATCH 028/647] Btrfs: remove unnecessary locking of cleaner_mutex to avoid deadlock After commmit e44163e17796 ("btrfs: explictly delete unused block groups in close_ctree and ro-remount"), added in the 4.3 merge window, we have calls to btrfs_delete_unused_bgs() while holding the cleaner_mutex. This can cause a deadlock with a concurrent block group relocation (when a filesystem balance or shrink operation is in progress for example) because btrfs_delete_unused_bgs() locks delete_unused_bgs_mutex and the relocation path locks first delete_unused_bgs_mutex and then it locks cleaner_mutex, resulting in a classic ABBA deadlock: CPU 0 CPU 1 lock fs_info->cleaner_mutex __btrfs_balance() || btrfs_shrink_device() lock fs_info->delete_unused_bgs_mutex btrfs_relocate_chunk() btrfs_relocate_block_group() lock fs_info->cleaner_mutex btrfs_delete_unused_bgs() lock fs_info->delete_unused_bgs_mutex Fix this by not taking the cleaner_mutex before calling btrfs_delete_unused_bgs() because it's no longer needed after commit 67c5e7d464bc ("Btrfs: fix race between balance and unused block group deletion"). The mutex fs_info->delete_unused_bgs_mutex, the spinlock fs_info->unused_bgs_lock and a block group's spinlock are enough to get correct serialization between tasks running relocation and unused block group deletion (as well as between multiple tasks concurrently calling btrfs_delete_unused_bgs()). This issue was discussed (in the mailing list) during the review of the patch titled "btrfs: explictly delete unused block groups in close_ctree and ro-remount" and it was agreed that acquiring the cleaner mutex had to be dropped after the patch titled "Btrfs: fix race between balance and unused block group deletion" got merged (both patches were submitted at about the same time, but one landed in kernel 4.2 and the other in the 4.3 merge window). Signed-off-by: Filipe Manana --- fs/btrfs/disk-io.c | 2 -- fs/btrfs/super.c | 2 -- 2 files changed, 4 deletions(-) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 0b658d0230f0..aa59871885da 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3762,9 +3762,7 @@ void close_ctree(struct btrfs_root *root) * block groups queued for removal, the deletion will be * skipped when we quit the cleaner thread. */ - mutex_lock(&root->fs_info->cleaner_mutex); btrfs_delete_unused_bgs(root->fs_info); - mutex_unlock(&root->fs_info->cleaner_mutex); ret = btrfs_commit_super(root); if (ret) diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index c389c13f0f38..5a186d7180d3 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -1658,9 +1658,7 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data) * groups on disk until we're mounted read-write again * unless we clean them up here. */ - mutex_lock(&root->fs_info->cleaner_mutex); btrfs_delete_unused_bgs(fs_info); - mutex_unlock(&root->fs_info->cleaner_mutex); btrfs_dev_replace_suspend_for_unmount(fs_info); btrfs_scrub_cancel(fs_info); From 135919a3a80565070b9645009e65f73e72c661c0 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Wed, 9 Sep 2015 13:35:05 -0400 Subject: [PATCH 029/647] intel_idle: Skylake Client Support - updated Addition of PC9 state, and minor tweaks to existing PC6 and PC8 states. Signed-off-by: Len Brown --- drivers/idle/intel_idle.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index 3a3738fe016b..cd4510a63375 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -620,7 +620,7 @@ static struct cpuidle_state skl_cstates[] = { .name = "C6-SKL", .desc = "MWAIT 0x20", .flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED, - .exit_latency = 75, + .exit_latency = 85, .target_residency = 200, .enter = &intel_idle, .enter_freeze = intel_idle_freeze, }, @@ -636,10 +636,18 @@ static struct cpuidle_state skl_cstates[] = { .name = "C8-SKL", .desc = "MWAIT 0x40", .flags = MWAIT2flg(0x40) | CPUIDLE_FLAG_TLB_FLUSHED, - .exit_latency = 174, + .exit_latency = 200, .target_residency = 800, .enter = &intel_idle, .enter_freeze = intel_idle_freeze, }, + { + .name = "C9-SKL", + .desc = "MWAIT 0x50", + .flags = MWAIT2flg(0x50) | CPUIDLE_FLAG_TLB_FLUSHED, + .exit_latency = 480, + .target_residency = 5000, + .enter = &intel_idle, + .enter_freeze = intel_idle_freeze, }, { .name = "C10-SKL", .desc = "MWAIT 0x60", From 30035e45753b708e7d47a98398500ca005e02b86 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Wed, 29 Apr 2015 12:52:04 -0400 Subject: [PATCH 030/647] string: provide strscpy() The strscpy() API is intended to be used instead of strlcpy(), and instead of most uses of strncpy(). - Unlike strlcpy(), it doesn't read from memory beyond (src + size). - Unlike strlcpy() or strncpy(), the API provides an easy way to check for destination buffer overflow: an -E2BIG error return value. - The provided implementation is robust in the face of the source buffer being asynchronously changed during the copy, unlike the current implementation of strlcpy(). - Unlike strncpy(), the destination buffer will be NUL-terminated if the string in the source buffer is too long. - Also unlike strncpy(), the destination buffer will not be updated beyond the NUL termination, avoiding strncpy's behavior of zeroing the entire tail end of the destination buffer. (A memset() after the strscpy() can be used if this behavior is desired.) - The implementation should be reasonably performant on all platforms since it uses the asm/word-at-a-time.h API rather than simple byte copy. Kernel-to-kernel string copy is not considered to be performance critical in any case. Signed-off-by: Chris Metcalf --- include/linux/string.h | 3 ++ lib/string.c | 88 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 91 insertions(+) diff --git a/include/linux/string.h b/include/linux/string.h index a8d90db9c4b0..9ef7795e65e4 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -25,6 +25,9 @@ extern char * strncpy(char *,const char *, __kernel_size_t); #ifndef __HAVE_ARCH_STRLCPY size_t strlcpy(char *, const char *, size_t); #endif +#ifndef __HAVE_ARCH_STRSCPY +ssize_t __must_check strscpy(char *, const char *, size_t); +#endif #ifndef __HAVE_ARCH_STRCAT extern char * strcat(char *, const char *); #endif diff --git a/lib/string.c b/lib/string.c index 13d1e84ddb80..8dbb7b1eab50 100644 --- a/lib/string.c +++ b/lib/string.c @@ -27,6 +27,10 @@ #include #include +#include +#include +#include + #ifndef __HAVE_ARCH_STRNCASECMP /** * strncasecmp - Case insensitive, length-limited string comparison @@ -146,6 +150,90 @@ size_t strlcpy(char *dest, const char *src, size_t size) EXPORT_SYMBOL(strlcpy); #endif +#ifndef __HAVE_ARCH_STRSCPY +/** + * strscpy - Copy a C-string into a sized buffer + * @dest: Where to copy the string to + * @src: Where to copy the string from + * @count: Size of destination buffer + * + * Copy the string, or as much of it as fits, into the dest buffer. + * The routine returns the number of characters copied (not including + * the trailing NUL) or -E2BIG if the destination buffer wasn't big enough. + * The behavior is undefined if the string buffers overlap. + * The destination buffer is always NUL terminated, unless it's zero-sized. + * + * Preferred to strlcpy() since the API doesn't require reading memory + * from the src string beyond the specified "count" bytes, and since + * the return value is easier to error-check than strlcpy()'s. + * In addition, the implementation is robust to the string changing out + * from underneath it, unlike the current strlcpy() implementation. + * + * Preferred to strncpy() since it always returns a valid string, and + * doesn't unnecessarily force the tail of the destination buffer to be + * zeroed. If the zeroing is desired, it's likely cleaner to use strscpy() + * with an overflow test, then just memset() the tail of the dest buffer. + */ +ssize_t strscpy(char *dest, const char *src, size_t count) +{ + const struct word_at_a_time constants = WORD_AT_A_TIME_CONSTANTS; + size_t max = count; + long res = 0; + + if (count == 0) + return -E2BIG; + +#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS + /* + * If src is unaligned, don't cross a page boundary, + * since we don't know if the next page is mapped. + */ + if ((long)src & (sizeof(long) - 1)) { + size_t limit = PAGE_SIZE - ((long)src & (PAGE_SIZE - 1)); + if (limit < max) + max = limit; + } +#else + /* If src or dest is unaligned, don't do word-at-a-time. */ + if (((long) dest | (long) src) & (sizeof(long) - 1)) + max = 0; +#endif + + while (max >= sizeof(unsigned long)) { + unsigned long c, data; + + c = *(unsigned long *)(src+res); + *(unsigned long *)(dest+res) = c; + if (has_zero(c, &data, &constants)) { + data = prep_zero_mask(c, data, &constants); + data = create_zero_mask(data); + return res + find_zero(data); + } + res += sizeof(unsigned long); + count -= sizeof(unsigned long); + max -= sizeof(unsigned long); + } + + while (count) { + char c; + + c = src[res]; + dest[res] = c; + if (!c) + return res; + res++; + count--; + } + + /* Hit buffer length without finding a NUL; force NUL-termination. */ + if (res) + dest[res-1] = '\0'; + + return -E2BIG; +} +EXPORT_SYMBOL(strscpy); +#endif + #ifndef __HAVE_ARCH_STRCAT /** * strcat - Append one %NUL-terminated string to another From 30059d494a72603d066baf55c748803df968aa08 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Wed, 29 Apr 2015 13:07:38 -0400 Subject: [PATCH 031/647] tile: use global strscpy() rather than private copy Now that strscpy() is a standard API, remove the local copy. Signed-off-by: Chris Metcalf --- arch/tile/gxio/mpipe.c | 33 ++++----------------------------- 1 file changed, 4 insertions(+), 29 deletions(-) diff --git a/arch/tile/gxio/mpipe.c b/arch/tile/gxio/mpipe.c index ee186e13dfe6..f102048d9c0e 100644 --- a/arch/tile/gxio/mpipe.c +++ b/arch/tile/gxio/mpipe.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -29,32 +30,6 @@ /* HACK: Avoid pointless "shadow" warnings. */ #define link link_shadow -/** - * strscpy - Copy a C-string into a sized buffer, but only if it fits - * @dest: Where to copy the string to - * @src: Where to copy the string from - * @size: size of destination buffer - * - * Use this routine to avoid copying too-long strings. - * The routine returns the total number of bytes copied - * (including the trailing NUL) or zero if the buffer wasn't - * big enough. To ensure that programmers pay attention - * to the return code, the destination has a single NUL - * written at the front (if size is non-zero) when the - * buffer is not big enough. - */ -static size_t strscpy(char *dest, const char *src, size_t size) -{ - size_t len = strnlen(src, size) + 1; - if (len > size) { - if (size) - dest[0] = '\0'; - return 0; - } - memcpy(dest, src, len); - return len; -} - int gxio_mpipe_init(gxio_mpipe_context_t *context, unsigned int mpipe_index) { char file[32]; @@ -540,7 +515,7 @@ int gxio_mpipe_link_instance(const char *link_name) if (!context) return GXIO_ERR_NO_DEVICE; - if (strscpy(name.name, link_name, sizeof(name.name)) == 0) + if (strscpy(name.name, link_name, sizeof(name.name)) < 0) return GXIO_ERR_NO_DEVICE; return gxio_mpipe_info_instance_aux(context, name); @@ -559,7 +534,7 @@ int gxio_mpipe_link_enumerate_mac(int idx, char *link_name, uint8_t *link_mac) rv = gxio_mpipe_info_enumerate_aux(context, idx, &name, &mac); if (rv >= 0) { - if (strscpy(link_name, name.name, sizeof(name.name)) == 0) + if (strscpy(link_name, name.name, sizeof(name.name)) < 0) return GXIO_ERR_INVAL_MEMORY_SIZE; memcpy(link_mac, mac.mac, sizeof(mac.mac)); } @@ -576,7 +551,7 @@ int gxio_mpipe_link_open(gxio_mpipe_link_t *link, _gxio_mpipe_link_name_t name; int rv; - if (strscpy(name.name, link_name, sizeof(name.name)) == 0) + if (strscpy(name.name, link_name, sizeof(name.name)) < 0) return GXIO_ERR_NO_DEVICE; rv = gxio_mpipe_link_open_aux(context, name, flags); From 75881df3fd7708f234c1e2573ade812eb5701708 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Thu, 10 Sep 2015 18:01:44 +0530 Subject: [PATCH 032/647] ASoC: dapm: fix memory leak Incase of an unknown event we were directly returning but we missed freeing params. Signed-off-by: Sudip Mukherjee Signed-off-by: Mark Brown --- sound/soc/soc-dapm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index f4bf21a5539b..ff8bda471b25 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -3501,7 +3501,7 @@ static int snd_soc_dai_link_event(struct snd_soc_dapm_widget *w, default: WARN(1, "Unknown event %d\n", event); - return -EINVAL; + ret = -EINVAL; } out: From f17b329b73a0393dc9d5fc5b4457189f92e5bbef Mon Sep 17 00:00:00 2001 From: Oder Chiou Date: Thu, 10 Sep 2015 13:40:16 +0800 Subject: [PATCH 033/647] ASoC: rt5645: Remove incorrect settings The patch removes the incorrect settings to avoid the pop sound in the first playback with headphone after boot. Signed-off-by: Oder Chiou Signed-off-by: Mark Brown --- sound/soc/codecs/rt5645.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index 4972bf3efa91..595ace9bcf58 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -2829,10 +2829,6 @@ static int rt5645_jack_detect(struct snd_soc_codec *codec, int jack_insert) snd_soc_dapm_sync(dapm); rt5645->jack_type = SND_JACK_HEADPHONE; } - - snd_soc_update_bits(codec, RT5645_CHARGE_PUMP, 0x0300, 0x0200); - snd_soc_write(codec, RT5645_DEPOP_M1, 0x001d); - snd_soc_write(codec, RT5645_DEPOP_M1, 0x0001); } else { /* jack out */ rt5645->jack_type = 0; @@ -2880,8 +2876,6 @@ int rt5645_set_jack_detect(struct snd_soc_codec *codec, rt5645->en_button_func = true; regmap_update_bits(rt5645->regmap, RT5645_GPIO_CTRL1, RT5645_GP1_PIN_IRQ, RT5645_GP1_PIN_IRQ); - regmap_update_bits(rt5645->regmap, RT5645_DEPOP_M1, - RT5645_HP_CB_MASK, RT5645_HP_CB_PU); regmap_update_bits(rt5645->regmap, RT5645_GEN_CTRL1, RT5645_DIG_GATE_CTRL, RT5645_DIG_GATE_CTRL); } From dd85ebf681ef0ee1fc985c353dd45e8b53b5dc1e Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Thu, 10 Sep 2015 16:48:13 +0530 Subject: [PATCH 034/647] spi: spidev: fix possible NULL dereference During the last close we are freeing spidev if spidev->spi is NULL, but just before checking if spidev->spi is NULL we are dereferencing it. Lets add a check there to avoid the NULL dereference. Fixes: 9169051617df ("spi: spidev: Don't mangle max_speed_hz in underlying spi device") Signed-off-by: Sudip Mukherjee Reviewed-by: Jarkko Nikula Tested-by: Jarkko Nikula Signed-off-by: Mark Brown --- drivers/spi/spidev.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c index dd616ff0ffc5..e58ca17eb247 100644 --- a/drivers/spi/spidev.c +++ b/drivers/spi/spidev.c @@ -651,7 +651,8 @@ static int spidev_release(struct inode *inode, struct file *filp) kfree(spidev->rx_buffer); spidev->rx_buffer = NULL; - spidev->speed_hz = spidev->spi->max_speed_hz; + if (spidev->spi) + spidev->speed_hz = spidev->spi->max_speed_hz; /* ... after we unbound from the underlying device? */ spin_lock_irq(&spidev->spi_lock); From 3758ff5f3dab57cd768d54279962a2f6bbc17188 Mon Sep 17 00:00:00 2001 From: Zidan Wang Date: Wed, 9 Sep 2015 19:29:10 +0800 Subject: [PATCH 035/647] ASoC: wm8960: correct the min gain value of some PGA The min gain is the corresponding gain value when the register value is 0 instead of 1, just correct it. Signed-off-by: Zidan Wang Acked-by: Charles Keepax Signed-off-by: Mark Brown --- sound/soc/codecs/wm8960.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/sound/soc/codecs/wm8960.c b/sound/soc/codecs/wm8960.c index 94c5c4681ce5..9c730490a02b 100644 --- a/sound/soc/codecs/wm8960.c +++ b/sound/soc/codecs/wm8960.c @@ -205,11 +205,11 @@ static int wm8960_put_deemph(struct snd_kcontrol *kcontrol, return wm8960_set_deemph(codec); } -static const DECLARE_TLV_DB_SCALE(adc_tlv, -9700, 50, 0); -static const DECLARE_TLV_DB_SCALE(dac_tlv, -12700, 50, 1); +static const DECLARE_TLV_DB_SCALE(adc_tlv, -9750, 50, 1); +static const DECLARE_TLV_DB_SCALE(dac_tlv, -12750, 50, 1); static const DECLARE_TLV_DB_SCALE(bypass_tlv, -2100, 300, 0); static const DECLARE_TLV_DB_SCALE(out_tlv, -12100, 100, 1); -static const DECLARE_TLV_DB_SCALE(boost_tlv, -1200, 300, 1); +static const DECLARE_TLV_DB_SCALE(boost_tlv, -1500, 300, 1); static const struct snd_kcontrol_new wm8960_snd_controls[] = { SOC_DOUBLE_R_TLV("Capture Volume", WM8960_LINVOL, WM8960_RINVOL, From 7e90f9b2b56669260e5f6f97974735d187f79b7d Mon Sep 17 00:00:00 2001 From: Zidan Wang Date: Wed, 9 Sep 2015 19:29:11 +0800 Subject: [PATCH 036/647] ASoC: wm8960: correct gain value for input PGA and add microphone PGA The input PGAs have a gain range from -17.25dB to +30dB in 0.75dB steps. The boost stage can provide additional gain. For line inputs, -12dB to +6dB gain is available on the boost mixer. For micphone inputs, it can provide up to +29dB additional gain from the microphone PGA. Signed-off-by: Zidan Wang Acked-by: Charles Keepax Signed-off-by: Mark Brown --- sound/soc/codecs/wm8960.c | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) diff --git a/sound/soc/codecs/wm8960.c b/sound/soc/codecs/wm8960.c index 9c730490a02b..70816f02ea8c 100644 --- a/sound/soc/codecs/wm8960.c +++ b/sound/soc/codecs/wm8960.c @@ -206,27 +206,37 @@ static int wm8960_put_deemph(struct snd_kcontrol *kcontrol, } static const DECLARE_TLV_DB_SCALE(adc_tlv, -9750, 50, 1); +static const DECLARE_TLV_DB_SCALE(inpga_tlv, -1725, 75, 0); static const DECLARE_TLV_DB_SCALE(dac_tlv, -12750, 50, 1); static const DECLARE_TLV_DB_SCALE(bypass_tlv, -2100, 300, 0); static const DECLARE_TLV_DB_SCALE(out_tlv, -12100, 100, 1); -static const DECLARE_TLV_DB_SCALE(boost_tlv, -1500, 300, 1); +static const DECLARE_TLV_DB_SCALE(lineinboost_tlv, -1500, 300, 1); +static const unsigned int micboost_tlv[] = { + TLV_DB_RANGE_HEAD(2), + 0, 1, TLV_DB_SCALE_ITEM(0, 1300, 0), + 2, 3, TLV_DB_SCALE_ITEM(2000, 900, 0), +}; static const struct snd_kcontrol_new wm8960_snd_controls[] = { SOC_DOUBLE_R_TLV("Capture Volume", WM8960_LINVOL, WM8960_RINVOL, - 0, 63, 0, adc_tlv), + 0, 63, 0, inpga_tlv), SOC_DOUBLE_R("Capture Volume ZC Switch", WM8960_LINVOL, WM8960_RINVOL, 6, 1, 0), SOC_DOUBLE_R("Capture Switch", WM8960_LINVOL, WM8960_RINVOL, 7, 1, 0), SOC_SINGLE_TLV("Right Input Boost Mixer RINPUT3 Volume", - WM8960_INBMIX1, 4, 7, 0, boost_tlv), + WM8960_INBMIX1, 4, 7, 0, lineinboost_tlv), SOC_SINGLE_TLV("Right Input Boost Mixer RINPUT2 Volume", - WM8960_INBMIX1, 1, 7, 0, boost_tlv), + WM8960_INBMIX1, 1, 7, 0, lineinboost_tlv), SOC_SINGLE_TLV("Left Input Boost Mixer LINPUT3 Volume", - WM8960_INBMIX2, 4, 7, 0, boost_tlv), + WM8960_INBMIX2, 4, 7, 0, lineinboost_tlv), SOC_SINGLE_TLV("Left Input Boost Mixer LINPUT2 Volume", - WM8960_INBMIX2, 1, 7, 0, boost_tlv), + WM8960_INBMIX2, 1, 7, 0, lineinboost_tlv), +SOC_SINGLE_TLV("Right Input Boost Mixer RINPUT1 Volume", + WM8960_RINPATH, 4, 4, 0, micboost_tlv), +SOC_SINGLE_TLV("Left Input Boost Mixer LINPUT1 Volume", + WM8960_LINPATH, 4, 4, 0, micboost_tlv), SOC_DOUBLE_R_TLV("Playback Volume", WM8960_LDAC, WM8960_RDAC, 0, 255, 0, dac_tlv), From 1853c949646005b5959c483becde86608f548f24 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 10 Sep 2015 20:05:46 +0200 Subject: [PATCH 037/647] netlink, mmap: transform mmap skb into full skb on taps Ken-ichirou reported that running netlink in mmap mode for receive in combination with nlmon will throw a NULL pointer dereference in __kfree_skb() on nlmon_xmit(), in my case I can also trigger an "unable to handle kernel paging request". The problem is the skb_clone() in __netlink_deliver_tap_skb() for skbs that are mmaped. I.e. the cloned skb doesn't have a destructor, whereas the mmap netlink skb has it pointed to netlink_skb_destructor(), set in the handler netlink_ring_setup_skb(). There, skb->head is being set to NULL, so that in such cases, __kfree_skb() doesn't perform a skb_release_data() via skb_release_all(), where skb->head is possibly being freed through kfree(head) into slab allocator, although netlink mmap skb->head points to the mmap buffer. Similarly, the same has to be done also for large netlink skbs where the data area is vmalloced. Therefore, as discussed, make a copy for these rather rare cases for now. This fixes the issue on my and Ken-ichirou's test-cases. Reference: http://thread.gmane.org/gmane.linux.network/371129 Fixes: bcbde0d449ed ("net: netlink: virtual tap device management") Reported-by: Ken-ichirou MATSUZAWA Signed-off-by: Daniel Borkmann Tested-by: Ken-ichirou MATSUZAWA Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 30 +++++++++++++++++++++++------- net/netlink/af_netlink.h | 9 +++++++++ 2 files changed, 32 insertions(+), 7 deletions(-) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 7f86d3b55060..4cad99d6c68b 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -125,6 +125,24 @@ static inline u32 netlink_group_mask(u32 group) return group ? 1 << (group - 1) : 0; } +static struct sk_buff *netlink_to_full_skb(const struct sk_buff *skb, + gfp_t gfp_mask) +{ + unsigned int len = skb_end_offset(skb); + struct sk_buff *new; + + new = alloc_skb(len, gfp_mask); + if (new == NULL) + return NULL; + + NETLINK_CB(new).portid = NETLINK_CB(skb).portid; + NETLINK_CB(new).dst_group = NETLINK_CB(skb).dst_group; + NETLINK_CB(new).creds = NETLINK_CB(skb).creds; + + memcpy(skb_put(new, len), skb->data, len); + return new; +} + int netlink_add_tap(struct netlink_tap *nt) { if (unlikely(nt->dev->type != ARPHRD_NETLINK)) @@ -206,7 +224,11 @@ static int __netlink_deliver_tap_skb(struct sk_buff *skb, int ret = -ENOMEM; dev_hold(dev); - nskb = skb_clone(skb, GFP_ATOMIC); + + if (netlink_skb_is_mmaped(skb) || is_vmalloc_addr(skb->head)) + nskb = netlink_to_full_skb(skb, GFP_ATOMIC); + else + nskb = skb_clone(skb, GFP_ATOMIC); if (nskb) { nskb->dev = dev; nskb->protocol = htons((u16) sk->sk_protocol); @@ -279,11 +301,6 @@ static void netlink_rcv_wake(struct sock *sk) } #ifdef CONFIG_NETLINK_MMAP -static bool netlink_skb_is_mmaped(const struct sk_buff *skb) -{ - return NETLINK_CB(skb).flags & NETLINK_SKB_MMAPED; -} - static bool netlink_rx_is_mmaped(struct sock *sk) { return nlk_sk(sk)->rx_ring.pg_vec != NULL; @@ -846,7 +863,6 @@ static void netlink_ring_set_copied(struct sock *sk, struct sk_buff *skb) } #else /* CONFIG_NETLINK_MMAP */ -#define netlink_skb_is_mmaped(skb) false #define netlink_rx_is_mmaped(sk) false #define netlink_tx_is_mmaped(sk) false #define netlink_mmap sock_no_mmap diff --git a/net/netlink/af_netlink.h b/net/netlink/af_netlink.h index 89008405d6b4..df9a06090db6 100644 --- a/net/netlink/af_netlink.h +++ b/net/netlink/af_netlink.h @@ -59,6 +59,15 @@ static inline struct netlink_sock *nlk_sk(struct sock *sk) return container_of(sk, struct netlink_sock, sk); } +static inline bool netlink_skb_is_mmaped(const struct sk_buff *skb) +{ +#ifdef CONFIG_NETLINK_MMAP + return NETLINK_CB(skb).flags & NETLINK_SKB_MMAPED; +#else + return false; +#endif /* CONFIG_NETLINK_MMAP */ +} + struct netlink_table { struct rhashtable hash; struct hlist_head mc_list; From 19539ce783dd27768d4f7f3b753152bf983db65b Mon Sep 17 00:00:00 2001 From: Tycho Andersen Date: Thu, 10 Sep 2015 18:25:07 -0600 Subject: [PATCH 038/647] ebpf: emit correct src_reg for conditional jumps Instead of always emitting BPF_REG_X, let's emit BPF_REG_X only when the source actually is BPF_X. This causes programs generated by the classic converter to not be importable via bpf(), as the eBPF verifier checks that the src_reg is correct or 0. While not a problem yet, this will be a problem when BPF_PROG_DUMP lands, and we can potentially dump and re-import programs generated by the converter. Signed-off-by: Tycho Andersen CC: Alexei Starovoitov CC: Daniel Borkmann Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- net/core/filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/filter.c b/net/core/filter.c index 13079f03902e..05a04ea87172 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -478,9 +478,9 @@ do_pass: bpf_src = BPF_X; } else { insn->dst_reg = BPF_REG_A; - insn->src_reg = BPF_REG_X; insn->imm = fp->k; bpf_src = BPF_SRC(fp->code); + insn->src_reg = bpf_src == BPF_X ? BPF_REG_X : 0; } /* Common case where 'jump_false' is next insn. */ From 8e2d61e0aed2b7c4ecb35844fe07e0b2b762dee4 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 10 Sep 2015 17:31:15 -0300 Subject: [PATCH 039/647] sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- net/sctp/protocol.c | 64 +++++++++++++++++++++++++++++---------------- 1 file changed, 41 insertions(+), 23 deletions(-) diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index b7143337e4fa..3d9ea9a48289 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1186,7 +1186,7 @@ static void sctp_v4_del_protocol(void) unregister_inetaddr_notifier(&sctp_inetaddr_notifier); } -static int __net_init sctp_net_init(struct net *net) +static int __net_init sctp_defaults_init(struct net *net) { int status; @@ -1279,12 +1279,6 @@ static int __net_init sctp_net_init(struct net *net) sctp_dbg_objcnt_init(net); - /* Initialize the control inode/socket for handling OOTB packets. */ - if ((status = sctp_ctl_sock_init(net))) { - pr_err("Failed to initialize the SCTP control sock\n"); - goto err_ctl_sock_init; - } - /* Initialize the local address list. */ INIT_LIST_HEAD(&net->sctp.local_addr_list); spin_lock_init(&net->sctp.local_addr_lock); @@ -1300,9 +1294,6 @@ static int __net_init sctp_net_init(struct net *net) return 0; -err_ctl_sock_init: - sctp_dbg_objcnt_exit(net); - sctp_proc_exit(net); err_init_proc: cleanup_sctp_mibs(net); err_init_mibs: @@ -1311,15 +1302,12 @@ err_sysctl_register: return status; } -static void __net_exit sctp_net_exit(struct net *net) +static void __net_exit sctp_defaults_exit(struct net *net) { /* Free the local address list */ sctp_free_addr_wq(net); sctp_free_local_addr_list(net); - /* Free the control endpoint. */ - inet_ctl_sock_destroy(net->sctp.ctl_sock); - sctp_dbg_objcnt_exit(net); sctp_proc_exit(net); @@ -1327,9 +1315,32 @@ static void __net_exit sctp_net_exit(struct net *net) sctp_sysctl_net_unregister(net); } -static struct pernet_operations sctp_net_ops = { - .init = sctp_net_init, - .exit = sctp_net_exit, +static struct pernet_operations sctp_defaults_ops = { + .init = sctp_defaults_init, + .exit = sctp_defaults_exit, +}; + +static int __net_init sctp_ctrlsock_init(struct net *net) +{ + int status; + + /* Initialize the control inode/socket for handling OOTB packets. */ + status = sctp_ctl_sock_init(net); + if (status) + pr_err("Failed to initialize the SCTP control sock\n"); + + return status; +} + +static void __net_init sctp_ctrlsock_exit(struct net *net) +{ + /* Free the control endpoint. */ + inet_ctl_sock_destroy(net->sctp.ctl_sock); +} + +static struct pernet_operations sctp_ctrlsock_ops = { + .init = sctp_ctrlsock_init, + .exit = sctp_ctrlsock_exit, }; /* Initialize the universe into something sensible. */ @@ -1462,8 +1473,11 @@ static __init int sctp_init(void) sctp_v4_pf_init(); sctp_v6_pf_init(); - status = sctp_v4_protosw_init(); + status = register_pernet_subsys(&sctp_defaults_ops); + if (status) + goto err_register_defaults; + status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; @@ -1471,9 +1485,9 @@ static __init int sctp_init(void) if (status) goto err_v6_protosw_init; - status = register_pernet_subsys(&sctp_net_ops); + status = register_pernet_subsys(&sctp_ctrlsock_ops); if (status) - goto err_register_pernet_subsys; + goto err_register_ctrlsock; status = sctp_v4_add_protocol(); if (status) @@ -1489,12 +1503,14 @@ out: err_v6_add_protocol: sctp_v4_del_protocol(); err_add_protocol: - unregister_pernet_subsys(&sctp_net_ops); -err_register_pernet_subsys: + unregister_pernet_subsys(&sctp_ctrlsock_ops); +err_register_ctrlsock: sctp_v6_protosw_exit(); err_v6_protosw_init: sctp_v4_protosw_exit(); err_protosw_init: + unregister_pernet_subsys(&sctp_defaults_ops); +err_register_defaults: sctp_v4_pf_exit(); sctp_v6_pf_exit(); sctp_sysctl_unregister(); @@ -1527,12 +1543,14 @@ static __exit void sctp_exit(void) sctp_v6_del_protocol(); sctp_v4_del_protocol(); - unregister_pernet_subsys(&sctp_net_ops); + unregister_pernet_subsys(&sctp_ctrlsock_ops); /* Free protosw registrations */ sctp_v6_protosw_exit(); sctp_v4_protosw_exit(); + unregister_pernet_subsys(&sctp_defaults_ops); + /* Unregister with socket layer. */ sctp_v6_pf_exit(); sctp_v4_pf_exit(); From a19a19de8310fb8ca2ca0621a9db1aab082943c5 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 11 Sep 2015 11:33:01 +0200 Subject: [PATCH 040/647] bnx2x: use ktime_get_seconds() for timestamp commit c48f350ff5e7 "bnx2x: Add MFW dump support" added the bnx2x_update_mfw_dump() function that reads the current time and stores it in a 32-bit field that gets passed into a buffer in a fixed format. This is potentially broken when the epoch overflows in 2038, and otherwise overflows in 2106. As we're trying to avoid uses of struct timeval for this reason, I noticed the addition of this function, and tried to rewrite it in a way that is more explicit about the overflow and that will keep working once we deprecate struct timeval. I assume that it is not possible to change the ABI any more, otherwise we should try to use a 64-bit field for the seconds right away. Signed-off-by: Arnd Bergmann Cc: Yuval Mintz Cc: Ariel Elior Acked-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index e3da2bddf143..89a174fa1300 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -3705,16 +3705,14 @@ out: void bnx2x_update_mfw_dump(struct bnx2x *bp) { - struct timeval epoc; u32 drv_ver; u32 valid_dump; if (!SHMEM2_HAS(bp, drv_info)) return; - /* Update Driver load time */ - do_gettimeofday(&epoc); - SHMEM2_WR(bp, drv_info.epoc, epoc.tv_sec); + /* Update Driver load time, possibly broken in y2038 */ + SHMEM2_WR(bp, drv_info.epoc, (u32)ktime_get_real_seconds()); drv_ver = bnx2x_update_mng_version_utility(DRV_MODULE_VERSION, true); SHMEM2_WR(bp, drv_info.drv_ver, drv_ver); From c2d4fbd2163e607915cc05798ce7fb7f31117cc1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Linus=20L=C3=BCssing?= Date: Fri, 11 Sep 2015 18:39:48 +0200 Subject: [PATCH 041/647] bridge: fix igmpv3 / mldv2 report parsing MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With the newly introduced helper functions the skb pulling is hidden in the checksumming function - and undone before returning to the caller. The IGMPv3 and MLDv2 report parsing functions in the bridge still assumed that the skb is pointing to the beginning of the IGMP/MLD message while it is now kept at the beginning of the IPv4/6 header, breaking the message parsing and creating packet loss. Fixing this by taking the offset between IP and IGMP/MLD header into account, too. Fixes: 9afd85c9e455 ("net: Export IGMP/MLD message validation code") Reported-by: Tobias Powalowski Tested-by: Tobias Powalowski Signed-off-by: Linus Lüssing Signed-off-by: David S. Miller --- net/bridge/br_multicast.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 66efdc21f548..480b3de1a0e3 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -1006,7 +1006,7 @@ static int br_ip4_multicast_igmp3_report(struct net_bridge *br, ih = igmpv3_report_hdr(skb); num = ntohs(ih->ngrec); - len = sizeof(*ih); + len = skb_transport_offset(skb) + sizeof(*ih); for (i = 0; i < num; i++) { len += sizeof(*grec); @@ -1067,7 +1067,7 @@ static int br_ip6_multicast_mld2_report(struct net_bridge *br, icmp6h = icmp6_hdr(skb); num = ntohs(icmp6h->icmp6_dataun.un_data16[1]); - len = sizeof(*icmp6h); + len = skb_transport_offset(skb) + sizeof(*icmp6h); for (i = 0; i < num; i++) { __be16 *nsrcs, _nsrcs; From 38c089d1d8d058f5dff018a811568aa8e8bc47fc Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Fri, 11 Sep 2015 15:01:16 -0700 Subject: [PATCH 042/647] openvswitch: Fix dependency on IPv6 defrag. When NF_CONNTRACK is built-in, NF_DEFRAG_IPV6 is a module, and OPENVSWITCH is built-in, the following build error would occur: net/built-in.o: In function `ovs_ct_execute': (.text+0x10f587): undefined reference to `nf_ct_frag6_gather' Fixes: 7f8a436eaa2c ("openvswitch: Add conntrack action") Reported-by: Jim Davis Signed-off-by: Joe Stringer Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/openvswitch/Kconfig b/net/openvswitch/Kconfig index 2a071f470d57..d143aa9f6654 100644 --- a/net/openvswitch/Kconfig +++ b/net/openvswitch/Kconfig @@ -5,7 +5,8 @@ config OPENVSWITCH tristate "Open vSwitch" depends on INET - depends on (!NF_CONNTRACK || NF_CONNTRACK) + depends on !NF_CONNTRACK || \ + (NF_CONNTRACK && (!NF_DEFRAG_IPV6 || NF_DEFRAG_IPV6)) select LIBCRC32C select MPLS select NET_MPLS_GSO From e8684c88774c0ddfeefdbed0aa469b25b9962f3e Mon Sep 17 00:00:00 2001 From: Alexey Khoroshilov Date: Sat, 12 Sep 2015 00:34:48 +0300 Subject: [PATCH 043/647] irda: ali-ircc: Fix deadlock in ali_ircc_sir_change_speed() ali_ircc_sir_change_speed() is always called with self->lock held, so acquiring the lock inside it leads to unavoidable deadlock. Call graph: ali_ircc_sir_change_speed() is called from ali_ircc_change_speed() ali_ircc_fir_hard_xmit() under spin_lock_irqsave(&self->lock, flags); ali_ircc_sir_hard_xmit() under spin_lock_irqsave(&self->lock, flags); ali_ircc_net_ioctl() under spin_lock_irqsave(&self->lock, flags); ali_ircc_dma_xmit_complete() ali_ircc_fir_interrupt() ali_ircc_interrupt() under spin_lock(&self->lock); ali_ircc_sir_write_wakeup() ali_ircc_sir_interrupt() ali_ircc_interrupt() under spin_lock(&self->lock); The patch removes spin_lock/unlock from ali_ircc_sir_change_speed(). Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov Signed-off-by: David S. Miller --- drivers/net/irda/ali-ircc.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/net/irda/ali-ircc.c b/drivers/net/irda/ali-ircc.c index 58ae11a14bb6..64bb44d5d867 100644 --- a/drivers/net/irda/ali-ircc.c +++ b/drivers/net/irda/ali-ircc.c @@ -1031,7 +1031,6 @@ static void ali_ircc_fir_change_speed(struct ali_ircc_cb *priv, __u32 baud) static void ali_ircc_sir_change_speed(struct ali_ircc_cb *priv, __u32 speed) { struct ali_ircc_cb *self = priv; - unsigned long flags; int iobase; int fcr; /* FIFO control reg */ int lcr; /* Line control reg */ @@ -1061,8 +1060,6 @@ static void ali_ircc_sir_change_speed(struct ali_ircc_cb *priv, __u32 speed) /* Update accounting for new speed */ self->io.speed = speed; - spin_lock_irqsave(&self->lock, flags); - divisor = 115200/speed; fcr = UART_FCR_ENABLE_FIFO; @@ -1089,9 +1086,6 @@ static void ali_ircc_sir_change_speed(struct ali_ircc_cb *priv, __u32 speed) /* without this, the connection will be broken after come back from FIR speed, but with this, the SIR connection is harder to established */ outb((UART_MCR_DTR | UART_MCR_RTS | UART_MCR_OUT2), iobase+UART_MCR); - - spin_unlock_irqrestore(&self->lock, flags); - } static void ali_ircc_change_dongle_speed(struct ali_ircc_cb *priv, int speed) From adf3442cc890f04af1e8e2c0b335c84e5092dca9 Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Sun, 23 Aug 2015 21:13:57 +0200 Subject: [PATCH 044/647] ARM: pxa: fix DFI bus lockups on startup After the conversion of pxa architecture to common clock framework, the NAND clock can be disabled on startup if no nand driver claims it. In this case, it happens that if the bootloader used the NAND and set the DFI arbitration bit, the next access to a static memory controller area, such as an ethernet card, the system bus will stall, and the core will be stalled forever. Fix this by clearing the DFI arbritration bit in pxa3xx startup. The bit will be enabled the pxa3xx-nand driver on need anyway. The only left requirement is that upon pxa3xx-nand removal, the bit should be cleared before the clock is disabled. Signed-off-by: Robert Jarzmik --- arch/arm/mach-pxa/include/mach/addr-map.h | 7 +++++++ arch/arm/mach-pxa/pxa3xx.c | 21 ++++++++++++++++++++- 2 files changed, 27 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-pxa/include/mach/addr-map.h b/arch/arm/mach-pxa/include/mach/addr-map.h index d28fe291233a..07b93fd24474 100644 --- a/arch/arm/mach-pxa/include/mach/addr-map.h +++ b/arch/arm/mach-pxa/include/mach/addr-map.h @@ -43,6 +43,13 @@ * 0xf6200000..0xf6201000 */ +/* + * DFI Bus for NAND, PXA3xx only + */ +#define NAND_PHYS 0x43100000 +#define NAND_VIRT IOMEM(0xf6300000) +#define NAND_SIZE 0x00100000 + /* * Internal Memory Controller (PXA27x and later) */ diff --git a/arch/arm/mach-pxa/pxa3xx.c b/arch/arm/mach-pxa/pxa3xx.c index ce0f8d6242e2..06005d3f2ba3 100644 --- a/arch/arm/mach-pxa/pxa3xx.c +++ b/arch/arm/mach-pxa/pxa3xx.c @@ -47,6 +47,13 @@ extern void __init pxa_dt_irq_init(int (*fn)(struct irq_data *, unsigned int)); #define ISRAM_START 0x5c000000 #define ISRAM_SIZE SZ_256K +/* + * NAND NFC: DFI bus arbitration subset + */ +#define NDCR (*(volatile u32 __iomem*)(NAND_VIRT + 0)) +#define NDCR_ND_ARB_EN (1 << 12) +#define NDCR_ND_ARB_CNTL (1 << 19) + static void __iomem *sram; static unsigned long wakeup_src; @@ -362,7 +369,12 @@ static struct map_desc pxa3xx_io_desc[] __initdata = { .pfn = __phys_to_pfn(PXA3XX_SMEMC_BASE), .length = SMEMC_SIZE, .type = MT_DEVICE - } + }, { + .virtual = (unsigned long)NAND_VIRT, + .pfn = __phys_to_pfn(NAND_PHYS), + .length = NAND_SIZE, + .type = MT_DEVICE + }, }; void __init pxa3xx_map_io(void) @@ -419,6 +431,13 @@ static int __init pxa3xx_init(void) */ ASCR &= ~(ASCR_RDH | ASCR_D1S | ASCR_D2S | ASCR_D3S); + /* + * Disable DFI bus arbitration, to prevent a system bus lock if + * somebody disables the NAND clock (unused clock) while this + * bit remains set. + */ + NDCR = (NDCR & ~NDCR_ND_ARB_EN) | NDCR_ND_ARB_CNTL; + if ((ret = pxa_init_dma(IRQ_DMA, 32))) return ret; From b692cb83b14d2f741f513221f5f78042c674c2a9 Mon Sep 17 00:00:00 2001 From: Jarkko Nikula Date: Mon, 7 Sep 2015 10:23:01 +0300 Subject: [PATCH 045/647] ARM: pxa: ssp: Fix build error by removing originally incorrect DT binding Commit 03fbf488cece ("spi: pxa2xx: Differentiate Intel LPSS types") caused build error here because it removed the type LPSS_SSP and I didn't notice the type was used here too. I believe commit a6e56c28a178 ("ARM: pxa: ssp: add DT bindings") added it accidentally by copying all enum pxa_ssp_type types from include/linux/pxa2xx_ssp.h even LPSS_SSP was for Intel LPSS SPI devices. Fix the build error by removing this incorrect binding. Fixes: 03fbf488cece ("spi: pxa2xx: Differentiate Intel LPSS types") Signed-off-by: Jarkko Nikula Reported-by: Axel Lin Cc: # 4.2 Signed-off-by: Robert Jarzmik --- arch/arm/plat-pxa/ssp.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/arm/plat-pxa/ssp.c b/arch/arm/plat-pxa/ssp.c index ad9529cc4203..daa1a65f2eb7 100644 --- a/arch/arm/plat-pxa/ssp.c +++ b/arch/arm/plat-pxa/ssp.c @@ -107,7 +107,6 @@ static const struct of_device_id pxa_ssp_of_ids[] = { { .compatible = "mvrl,pxa168-ssp", .data = (void *) PXA168_SSP }, { .compatible = "mrvl,pxa910-ssp", .data = (void *) PXA910_SSP }, { .compatible = "mrvl,ce4100-ssp", .data = (void *) CE4100_SSP }, - { .compatible = "mrvl,lpss-ssp", .data = (void *) LPSS_SSP }, { }, }; MODULE_DEVICE_TABLE(of, pxa_ssp_of_ids); From a797451b3a595759b4df48150e611237d4bd204a Mon Sep 17 00:00:00 2001 From: Douglas Anderson Date: Wed, 2 Sep 2015 14:25:48 -0700 Subject: [PATCH 046/647] ARM: dts: Add ddc i2c reference to veyron The ddc-i2c-bus property was missing from the veyron dtsi file since downstream the ddc-i2c-bus was still being specified in rk3288.dtsi and nobody noticed when the veyron dtsi was sent upstream. Add it. Signed-off-by: Douglas Anderson Tested-by: Brian Norris Signed-off-by: Heiko Stuebner --- arch/arm/boot/dts/rk3288-veyron.dtsi | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/boot/dts/rk3288-veyron.dtsi b/arch/arm/boot/dts/rk3288-veyron.dtsi index 2fa7a0dc83f7..275c78ccc0f3 100644 --- a/arch/arm/boot/dts/rk3288-veyron.dtsi +++ b/arch/arm/boot/dts/rk3288-veyron.dtsi @@ -158,6 +158,7 @@ }; &hdmi { + ddc-i2c-bus = <&i2c5>; status = "okay"; }; From 4de2ced795272db3ee5d668510b3a91a2bf1d1fb Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 17 Aug 2015 11:08:23 -0700 Subject: [PATCH 047/647] staging: most: fix HDM_USB dependencies and build errors Fix kconfig dependency warning and build errors. warning: (HDM_USB) selects AIM_NETWORK which has unmet direct dependencies (STAGING && MOST && NET) drivers/built-in.o: In function `aim_resume_tx_channel': networking.c:(.text+0xd6f7a2): undefined reference to `netif_tx_wake_queue' drivers/built-in.o: In function `aim_rx_data': networking.c:(.text+0xd6f8c5): undefined reference to `__netdev_alloc_skb' networking.c:(.text+0xd6f99a): undefined reference to `skb_put' networking.c:(.text+0xd6fa44): undefined reference to `eth_type_trans' networking.c:(.text+0xd6fa6f): undefined reference to `netif_rx' drivers/built-in.o: In function `most_nd_setup': networking.c:(.text+0xd6fad2): undefined reference to `ether_setup' drivers/built-in.o: In function `most_nd_set_mac_address': networking.c:(.text+0xd6fb0f): undefined reference to `eth_mac_addr' drivers/built-in.o: In function `most_nd_open': networking.c:(.text+0xd6fd37): undefined reference to `netif_tx_wake_queue' drivers/built-in.o: In function `aim_probe_channel': networking.c:(.text+0xd6febb): undefined reference to `alloc_netdev_mqs' networking.c:(.text+0xd6ff18): undefined reference to `register_netdev' networking.c:(.text+0xd6ff4a): undefined reference to `free_netdev' drivers/built-in.o: In function `most_net_rm_netdev_safe.isra.0': networking.c:(.text+0xd6ffcf): undefined reference to `unregister_netdev' networking.c:(.text+0xd6ffdf): undefined reference to `free_netdev' drivers/built-in.o: In function `most_nd_start_xmit': networking.c:(.text+0xd70390): undefined reference to `kfree_skb' drivers/built-in.o: In function `most_deliver_netinfo': (.text+0xd70499): undefined reference to `netif_tx_wake_queue' Signed-off-by: Randy Dunlap Cc: Christian Gromm Cc: Michael Fabry Cc: Christian Gromm Signed-off-by: Greg Kroah-Hartman --- drivers/staging/most/hdm-usb/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/most/hdm-usb/Kconfig b/drivers/staging/most/hdm-usb/Kconfig index a482c3fdf34b..ec1546312ee6 100644 --- a/drivers/staging/most/hdm-usb/Kconfig +++ b/drivers/staging/most/hdm-usb/Kconfig @@ -4,7 +4,7 @@ config HDM_USB tristate "USB HDM" - depends on USB + depends on USB && NET select AIM_NETWORK ---help--- Say Y here if you want to connect via USB to network tranceiver. From b786a8ad6a34b45ff6e40cf9c21f9c13c38e2f70 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 1 Sep 2015 22:05:58 +0200 Subject: [PATCH 048/647] Staging: most: MOST and MOSTCORE should depend on HAS_DMA If NO_DMA=y: ERROR: "dma_free_coherent" [drivers/staging/most/mostcore/mostcore.ko] undefined! ERROR: "dma_alloc_coherent" [drivers/staging/most/mostcore/mostcore.ko] undefined! As all MOST sub drivers use DMA functionality, add a dependency on HAS_DMA to MOSTCORE, and to MOST, which selects MOSTCORE. Signed-off-by: Geert Uytterhoeven Acked-by: Christian Gromm Signed-off-by: Greg Kroah-Hartman --- drivers/staging/most/Kconfig | 1 + drivers/staging/most/mostcore/Kconfig | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/staging/most/Kconfig b/drivers/staging/most/Kconfig index d50de03de7b9..0b9b9b539f70 100644 --- a/drivers/staging/most/Kconfig +++ b/drivers/staging/most/Kconfig @@ -1,5 +1,6 @@ menuconfig MOST tristate "MOST driver" + depends on HAS_DMA select MOSTCORE default n ---help--- diff --git a/drivers/staging/most/mostcore/Kconfig b/drivers/staging/most/mostcore/Kconfig index 38abf1b21b66..47172546d728 100644 --- a/drivers/staging/most/mostcore/Kconfig +++ b/drivers/staging/most/mostcore/Kconfig @@ -4,6 +4,7 @@ config MOSTCORE tristate "MOST Core" + depends on HAS_DMA ---help--- Say Y here if you want to enable MOST support. From a6ee5ac15ddf2425a066339f1ab030c97ec74b25 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Fri, 21 Aug 2015 14:02:57 -0700 Subject: [PATCH 049/647] staging/android: Update ION TODO per LPC discussion We discussed a bit with the folks on the Cc: list below what to do with ION. Two big take-aways: - High-performance drivers (like gpus) always want to play tricks with coherency and will lie to the dma api (radeon, nouveau, i915 gpu drivers all do so in upstream). What needs to be done here is fill gaps in dma-buf so that we can do this without breaking the dma-api expections of other clients like v4l. The consesus is that hw won't stop needing these tricks anytime soon. - Placement constraints for shared buffers won't be solved any other way than through something platform-specific like ion with platform-specific knowledge in userspace in something like gralloc. For general-purpose devices where this assumption would be painful for userspace (like servers) the consensus is that such devices will have proper MMUs where placement constraint handling is fairly irrelevant. Hence it is reasonable to destage ion as-is without changing the overall design to enable these use-cases and just fixing up a these few fairly minor things. Since there won't relly be an open-source userspace for ion (and hence drm maintainers won't take it) the proposal is to eventually move it to drivers/android/ion.[hc]. Laura would be ok with being maintainer once this is all done and ion is destaged. Note that Thiago is working on exposing the cpu cache flushing for cpu access from userspace through mmaps so this is alread in progress. Also adding him to the Cc: list. v2: Add ION_IOC_IMPORT to the list of ioctl that probably should go. Cc: Laura Abbott Cc: sumit.semwal@linaro.org Cc: laurent.pinchart@ideasonboard.com Cc: ghackmann@google.com Cc: robdclark@gmail.com Cc: david.brown@arm.com Cc: romlem@google.com Cc: Tiago Vignatti Signed-off-by: Daniel Vetter Signed-off-by: Greg Kroah-Hartman --- drivers/staging/android/TODO | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/drivers/staging/android/TODO b/drivers/staging/android/TODO index 20288fc53946..8f3ac37bfe12 100644 --- a/drivers/staging/android/TODO +++ b/drivers/staging/android/TODO @@ -5,5 +5,25 @@ TODO: - add proper arch dependencies as needed - audit userspace interfaces to make sure they are sane + +ion/ + - Remove ION_IOC_SYNC: Flushing for devices should be purely a kernel internal + interface on top of dma-buf. flush_for_device needs to be added to dma-buf + first. + - Remove ION_IOC_CUSTOM: Atm used for cache flushing for cpu access in some + vendor trees. Should be replaced with an ioctl on the dma-buf to expose the + begin/end_cpu_access hooks to userspace. + - Clarify the tricks ion plays with explicitly managing coherency behind the + dma api's back (this is absolutely needed for high-perf gpu drivers): Add an + explicit coherency management mode to flush_for_device to be used by drivers + which want to manage caches themselves and which indicates whether cpu caches + need flushing. + - With those removed there's probably no use for ION_IOC_IMPORT anymore either + since ion would just be the central allocator for shared buffers. + - Add dt-binding to expose cma regions as ion heaps, with the rule that any + such cma regions must already be used by some device for dma. I.e. ion only + exposes existing cma regions and doesn't reserve unecessarily memory when + booting a system which doesn't use ion. + Please send patches to Greg Kroah-Hartman and Cc: Arve Hjønnevåg and Riley Andrews From d98229f0299d5d5a741873c5900159844b2b2de5 Mon Sep 17 00:00:00 2001 From: Andreas Dilger Date: Mon, 24 Aug 2015 11:11:49 -0400 Subject: [PATCH 050/647] staging/lustre: change Lustre URLs and mailing list Now that the lustre.org domain has been liberated we can again use that for the main website URL and mailing list. Also update the URL for userspace tools downloads and Git repo. Signed-off-by: Andreas Dilger Signed-off-by: Oleg Drokin Signed-off-by: Greg Kroah-Hartman --- MAINTAINERS | 4 ++-- drivers/staging/lustre/README.txt | 16 ++++++---------- 2 files changed, 8 insertions(+), 12 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 7ba7ab749c85..c4c7804fb8e3 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -9904,8 +9904,8 @@ F: drivers/staging/media/lirc/ STAGING - LUSTRE PARALLEL FILESYSTEM M: Oleg Drokin M: Andreas Dilger -L: HPDD-discuss@lists.01.org (moderated for non-subscribers) -W: http://lustre.opensfs.org/ +L: lustre-devel@lists.lustre.org (moderated for non-subscribers) +W: http://wiki.lustre.org/ S: Maintained F: drivers/staging/lustre diff --git a/drivers/staging/lustre/README.txt b/drivers/staging/lustre/README.txt index cf0ca50ff83b..0676243eea9e 100644 --- a/drivers/staging/lustre/README.txt +++ b/drivers/staging/lustre/README.txt @@ -14,10 +14,8 @@ Unlike shared disk storage cluster filesystems (e.g. OCFS2, GFS, GPFS), Lustre has independent Metadata and Data servers that clients can access in parallel to maximize performance. -In order to use Lustre client you will need to download lustre client -tools from -https://downloads.hpdd.intel.com/public/lustre/latest-feature-release/ -the package name is lustre-client. +In order to use Lustre client you will need to download the "lustre-client" +package that contains the userspace tools from http://lustre.org/download/ You will need to install and configure your Lustre servers separately. @@ -76,12 +74,10 @@ Mount Options More Information ================ -You can get more information at -OpenSFS website: http://lustre.opensfs.org/about/ -Intel HPDD wiki: https://wiki.hpdd.intel.com +You can get more information at the Lustre website: http://wiki.lustre.org/ -Out of tree Lustre client and server code is available at: -http://git.whamcloud.com/fs/lustre-release.git +Source for the userspace tools and out-of-tree client and server code +is available at: http://git.hpdd.intel.com/fs/lustre-release.git Latest binary packages: -http://lustre.opensfs.org/download-lustre/ +http://lustre.org/download/ From cabb5b2a2866829177df2f1cee70f0dd42ee6ae0 Mon Sep 17 00:00:00 2001 From: Stefan Wahren Date: Tue, 25 Aug 2015 21:04:11 +0000 Subject: [PATCH 051/647] staging: fbtft: fix 9-bit SPI support detection MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since the result of the setup function isn't adequate to check 9-bit SPI support, we better check bits_per_word_mask. Btw this change avoids a NULL pointer dereference with master drivers without a separate setup function. Signed-off-by: Stefan Wahren Acked-by: Noralf Trønnes Signed-off-by: Greg Kroah-Hartman --- drivers/staging/fbtft/fbtft-core.c | 10 +++------- drivers/staging/fbtft/flexfb.c | 11 ++++------- 2 files changed, 7 insertions(+), 14 deletions(-) diff --git a/drivers/staging/fbtft/fbtft-core.c b/drivers/staging/fbtft/fbtft-core.c index 23392eb6799e..7f5fa3d1cab0 100644 --- a/drivers/staging/fbtft/fbtft-core.c +++ b/drivers/staging/fbtft/fbtft-core.c @@ -1436,15 +1436,11 @@ int fbtft_probe_common(struct fbtft_display *display, /* 9-bit SPI setup */ if (par->spi && display->buswidth == 9) { - par->spi->bits_per_word = 9; - ret = par->spi->master->setup(par->spi); - if (ret) { + if (par->spi->master->bits_per_word_mask & SPI_BPW_MASK(9)) { + par->spi->bits_per_word = 9; + } else { dev_warn(&par->spi->dev, "9-bit SPI not available, emulating using 8-bit.\n"); - par->spi->bits_per_word = 8; - ret = par->spi->master->setup(par->spi); - if (ret) - goto out_release; /* allocate buffer with room for dc bits */ par->extra = devm_kzalloc(par->info->device, par->txbuf.len + (par->txbuf.len / 8) + 8, diff --git a/drivers/staging/fbtft/flexfb.c b/drivers/staging/fbtft/flexfb.c index c763efc5de7d..3f380a0086c3 100644 --- a/drivers/staging/fbtft/flexfb.c +++ b/drivers/staging/fbtft/flexfb.c @@ -463,15 +463,12 @@ static int flexfb_probe_common(struct spi_device *sdev, } par->fbtftops.write_register = fbtft_write_reg8_bus9; par->fbtftops.write_vmem = fbtft_write_vmem16_bus9; - sdev->bits_per_word = 9; - ret = sdev->master->setup(sdev); - if (ret) { + if (par->spi->master->bits_per_word_mask + & SPI_BPW_MASK(9)) { + par->spi->bits_per_word = 9; + } else { dev_warn(dev, "9-bit SPI not available, emulating using 8-bit.\n"); - sdev->bits_per_word = 8; - ret = sdev->master->setup(sdev); - if (ret) - goto out_release; /* allocate buffer with room for dc bits */ par->extra = devm_kzalloc(par->info->device, par->txbuf.len + (par->txbuf.len / 8) + 8, From dd3afa57c515801f37071734f56be9035fdeaa8c Mon Sep 17 00:00:00 2001 From: Stefan Wahren Date: Tue, 25 Aug 2015 21:04:12 +0000 Subject: [PATCH 052/647] staging: fbtft: replace master->setup() with spi_setup() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calling the setup of the SPI master directly causes a NULL pointer dereference with master drivers without a separate setup function. This problem is reproduceable on ARM MXS platform. So fix this issue by using spi_setup() instead. Signed-off-by: Stefan Wahren Acked-by: Noralf Trønnes Signed-off-by: Greg Kroah-Hartman --- drivers/staging/fbtft/fb_uc1611.c | 2 +- drivers/staging/fbtft/fb_watterott.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/staging/fbtft/fb_uc1611.c b/drivers/staging/fbtft/fb_uc1611.c index 32f3a9d921d6..5cafa50d1fac 100644 --- a/drivers/staging/fbtft/fb_uc1611.c +++ b/drivers/staging/fbtft/fb_uc1611.c @@ -76,7 +76,7 @@ static int init_display(struct fbtft_par *par) /* Set CS active high */ par->spi->mode |= SPI_CS_HIGH; - ret = par->spi->master->setup(par->spi); + ret = spi_setup(par->spi); if (ret) { dev_err(par->info->device, "Could not set SPI_CS_HIGH\n"); return ret; diff --git a/drivers/staging/fbtft/fb_watterott.c b/drivers/staging/fbtft/fb_watterott.c index 88fb2c0132d5..8eae6ef25846 100644 --- a/drivers/staging/fbtft/fb_watterott.c +++ b/drivers/staging/fbtft/fb_watterott.c @@ -169,7 +169,7 @@ static int init_display(struct fbtft_par *par) /* enable SPI interface by having CS and MOSI low during reset */ save_mode = par->spi->mode; par->spi->mode |= SPI_CS_HIGH; - ret = par->spi->master->setup(par->spi); /* set CS inactive low */ + ret = spi_setup(par->spi); /* set CS inactive low */ if (ret) { dev_err(par->info->device, "Could not set SPI_CS_HIGH\n"); return ret; @@ -180,7 +180,7 @@ static int init_display(struct fbtft_par *par) par->fbtftops.reset(par); mdelay(1000); par->spi->mode = save_mode; - ret = par->spi->master->setup(par->spi); + ret = spi_setup(par->spi); if (ret) { dev_err(par->info->device, "Could not restore SPI mode\n"); return ret; From 5b12100a53c11ef760e78c2899be8f1bf33812ca Mon Sep 17 00:00:00 2001 From: David Kershner Date: Fri, 4 Sep 2015 12:01:29 -0400 Subject: [PATCH 053/647] staging: unisys: unregister netdev when create debugfs fails Noticed we were not unregistering the netdevice if we failed to create the debugfs entries. This patch fixes that problem. Signed-off-by: David Kershner Signed-off-by: Benjamin Romer Signed-off-by: Greg Kroah-Hartman --- drivers/staging/unisys/visornic/visornic_main.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/staging/unisys/visornic/visornic_main.c b/drivers/staging/unisys/visornic/visornic_main.c index 8c9da7ea7845..bacf1af4e8d4 100644 --- a/drivers/staging/unisys/visornic/visornic_main.c +++ b/drivers/staging/unisys/visornic/visornic_main.c @@ -1924,13 +1924,16 @@ static int visornic_probe(struct visor_device *dev) "%s debugfs_create_dir %s failed\n", __func__, netdev->name); err = -ENOMEM; - goto cleanup_xmit_cmdrsp; + goto cleanup_register_netdev; } dev_info(&dev->device, "%s success netdev=%s\n", __func__, netdev->name); return 0; +cleanup_register_netdev: + unregister_netdev(netdev); + cleanup_napi_add: del_timer_sync(&devdata->irq_poll_timer); netif_napi_del(&devdata->napi); From f6b6a8ec9a3bcba8cc9a9c302b514cf3848e7d35 Mon Sep 17 00:00:00 2001 From: David Kershner Date: Fri, 4 Sep 2015 12:01:30 -0400 Subject: [PATCH 054/647] staging: unisys: visornic: Fix receive bytes statistics The receive byte statistics was wrong in /proc/net/dev. Move the collection of statistics after the proper amount of bytes has been calculated and make sure you add it to rx_bytes instead of just replacing it. Signed-off-by: David Kershner Signed-off-by: Benjamin Romer Signed-off-by: Greg Kroah-Hartman --- drivers/staging/unisys/visornic/visornic_main.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/staging/unisys/visornic/visornic_main.c b/drivers/staging/unisys/visornic/visornic_main.c index bacf1af4e8d4..7b08a8998a6d 100644 --- a/drivers/staging/unisys/visornic/visornic_main.c +++ b/drivers/staging/unisys/visornic/visornic_main.c @@ -1189,16 +1189,16 @@ visornic_rx(struct uiscmdrsp *cmdrsp) spin_lock_irqsave(&devdata->priv_lock, flags); atomic_dec(&devdata->num_rcvbuf_in_iovm); - /* update rcv stats - call it with priv_lock held */ - devdata->net_stats.rx_packets++; - devdata->net_stats.rx_bytes = skb->len; - /* set length to how much was ACTUALLY received - * NOTE: rcv_done_len includes actual length of data rcvd * including ethhdr */ skb->len = cmdrsp->net.rcv.rcv_done_len; + /* update rcv stats - call it with priv_lock held */ + devdata->net_stats.rx_packets++; + devdata->net_stats.rx_bytes += skb->len; + /* test enabled while holding lock */ if (!(devdata->enabled && devdata->enab_dis_acked)) { /* don't process it unless we're in enable mode and until From 418627ddc6d1ed50a3dfd2c8f938741b0cbcc1ed Mon Sep 17 00:00:00 2001 From: David Kershner Date: Fri, 4 Sep 2015 12:01:31 -0400 Subject: [PATCH 055/647] staging: unisys: visorbus: Unregister driver on error If there is an error in registering driver attributes, unregister the driver as well. Signed-off-by: David Kershner Signed-off-by: Benjamin Romer Signed-off-by: Greg Kroah-Hartman --- drivers/staging/unisys/visorbus/visorbus_main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/staging/unisys/visorbus/visorbus_main.c b/drivers/staging/unisys/visorbus/visorbus_main.c index 2309f5f2b238..7905ea920a4a 100644 --- a/drivers/staging/unisys/visorbus/visorbus_main.c +++ b/drivers/staging/unisys/visorbus/visorbus_main.c @@ -885,6 +885,8 @@ int visorbus_register_visor_driver(struct visor_driver *drv) if (rc < 0) return rc; rc = register_driver_attributes(drv); + if (rc < 0) + driver_unregister(&drv->driver); return rc; } EXPORT_SYMBOL_GPL(visorbus_register_visor_driver); From 6155a3cf11a6d4ed102ef5a30f655d4d2a38076b Mon Sep 17 00:00:00 2001 From: Benjamin Romer Date: Fri, 4 Sep 2015 12:01:32 -0400 Subject: [PATCH 056/647] staging: unisys: stop device registration before visorbus registration In cases where visorbus is compiled directly into the kernel, if visorbus registration fails for any reason, it is still possible for other drivers to call visorbus_register_visor_driver(), which could cause an oops. Prevent this by saving the result of the call to create_bus() in a static variable, and return an error code when the bus hasn't been registered successfully. Signed-off-by: Benjamin Romer Signed-off-by: Greg Kroah-Hartman --- drivers/staging/unisys/visorbus/visorbus_main.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/staging/unisys/visorbus/visorbus_main.c b/drivers/staging/unisys/visorbus/visorbus_main.c index 7905ea920a4a..a272b48bab28 100644 --- a/drivers/staging/unisys/visorbus/visorbus_main.c +++ b/drivers/staging/unisys/visorbus/visorbus_main.c @@ -37,6 +37,8 @@ static int visorbus_debugref; #define POLLJIFFIES_TESTWORK 100 #define POLLJIFFIES_NORMALCHANNEL 10 +static int busreg_rc = -ENODEV; /* stores the result from bus registration */ + static int visorbus_uevent(struct device *xdev, struct kobj_uevent_env *env); static int visorbus_match(struct device *xdev, struct device_driver *xdrv); static void fix_vbus_dev_info(struct visor_device *visordev); @@ -863,6 +865,9 @@ int visorbus_register_visor_driver(struct visor_driver *drv) { int rc = 0; + if (busreg_rc < 0) + return -ENODEV; /*can't register on a nonexistent bus*/ + drv->driver.name = drv->name; drv->driver.bus = &visorbus_type; drv->driver.probe = visordriver_probe_device; @@ -1262,10 +1267,8 @@ remove_bus_instance(struct visor_device *dev) static int create_bus_type(void) { - int rc = 0; - - rc = bus_register(&visorbus_type); - return rc; + busreg_rc = bus_register(&visorbus_type); + return busreg_rc; } /** Remove the one-and-only one instance of the visor bus type (visorbus_type). From 8b5081c876bde9b1df5e96ee6990d5acd20c128b Mon Sep 17 00:00:00 2001 From: Benjamin Romer Date: Fri, 4 Sep 2015 12:01:33 -0400 Subject: [PATCH 057/647] staging: unisys: visornic: handle error return from device registration There is no code to handle an error return in visornic, when it tries to register with visorbus. This patch handles an error return from visorbus_register_visor_driver() by dropping out of initialization. Signed-off-by: Benjamin Romer Signed-off-by: Greg Kroah-Hartman --- drivers/staging/unisys/visornic/visornic_main.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/staging/unisys/visornic/visornic_main.c b/drivers/staging/unisys/visornic/visornic_main.c index 7b08a8998a6d..9d3c1e282062 100644 --- a/drivers/staging/unisys/visornic/visornic_main.c +++ b/drivers/staging/unisys/visornic/visornic_main.c @@ -2131,8 +2131,9 @@ static int visornic_init(void) if (!dev_num_pool) goto cleanup_workqueue; - visorbus_register_visor_driver(&visornic_driver); - return 0; + err = visorbus_register_visor_driver(&visornic_driver); + if (!err) + return 0; cleanup_workqueue: if (visornic_timeout_reset_workqueue) { From 216076705d6ac291d42e0f8dd85e6a0da98c0fa3 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 8 Sep 2015 08:56:13 -0400 Subject: [PATCH 058/647] dm thin: disable discard support for thin devices if pool's is disabled If the pool is configured with 'ignore_discard' its discard support is disabled. The pool's thin devices should also have queue_limits that reflect discards are disabled. Fixes: 34fbcf62 ("dm thin: range discard support") Signed-off-by: Mike Snitzer Cc: stable@vger.kernel.org # 4.1+ --- drivers/md/dm-thin.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c index 6578b7bc1fbb..6fcbfb063366 100644 --- a/drivers/md/dm-thin.c +++ b/drivers/md/dm-thin.c @@ -4249,6 +4249,10 @@ static void thin_io_hints(struct dm_target *ti, struct queue_limits *limits) { struct thin_c *tc = ti->private; struct pool *pool = tc->pool; + struct queue_limits *pool_limits = dm_get_queue_limits(pool->pool_md); + + if (!pool_limits->discard_granularity) + return; /* pool's discard support is disabled */ limits->discard_granularity = pool->sectors_per_block << SECTOR_SHIFT; limits->max_discard_sectors = 2048 * 1024 * 16; /* 16G */ From cd33dc9ac2977ebe30cecbf39d2992190fbac5b4 Mon Sep 17 00:00:00 2001 From: Punit Agrawal Date: Tue, 8 Sep 2015 14:51:12 +0100 Subject: [PATCH 059/647] thermal: Fix thermal_zone_of_sensor_register to match documentation thermal_zone_of_sensor_register is documented as returning a pointer to either a valid thermal_zone_device on success, or a corresponding ERR_PTR() value. In contrast, the function returns NULL when THERMAL_OF is configured off. Fix this. Signed-off-by: Punit Agrawal Acked-by: Guenter Roeck Cc: Eduardo Valentin Cc: Zhang Rui Signed-off-by: Eduardo Valentin --- include/linux/thermal.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/thermal.h b/include/linux/thermal.h index 17292fee8686..0c5518e13584 100644 --- a/include/linux/thermal.h +++ b/include/linux/thermal.h @@ -360,7 +360,7 @@ static inline struct thermal_zone_device * thermal_zone_of_sensor_register(struct device *dev, int id, void *data, const struct thermal_zone_of_device_ops *ops) { - return NULL; + return ERR_PTR(-ENODEV); } static inline From f9e2b05218e64fc6c614b90a650863981701f974 Mon Sep 17 00:00:00 2001 From: Eduardo Valentin Date: Wed, 9 Sep 2015 20:39:44 -0700 Subject: [PATCH 060/647] thermal: hisi: allow compile test Adding COMPILE_TEST flag to hisi driver to facilitate maintenance. Cc: Zhang Rui Cc: linux-pm@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Eduardo Valentin --- drivers/thermal/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index 039004400987..ef48f23823c8 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -163,7 +163,7 @@ config THERMAL_EMULATION config HISI_THERMAL tristate "Hisilicon thermal driver" - depends on ARCH_HISI && CPU_THERMAL && OF + depends on (ARCH_HISI && CPU_THERMAL && OF) || COMPILE_TEST help Enable this to plug hisilicon's thermal sensor driver into the Linux thermal framework. cpufreq is used as the cooling device to throttle From aa2937b73a45e49704b24db8d9163d83002332d9 Mon Sep 17 00:00:00 2001 From: Eduardo Valentin Date: Wed, 9 Sep 2015 20:42:01 -0700 Subject: [PATCH 061/647] thermal: spear: allow compile test Adding COMPILE_TEST flag to spear driver to facilitate maintenance. Cc: Zhang Rui Cc: linux-pm@vger.kernel.org Cc: linux-kernel@vger.kernel.org Acked-by: Viresh Kumar Signed-off-by: Eduardo Valentin --- drivers/thermal/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index ef48f23823c8..e2cfaa44ee1b 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -182,7 +182,7 @@ config IMX_THERMAL config SPEAR_THERMAL bool "SPEAr thermal sensor driver" - depends on PLAT_SPEAR + depends on PLAT_SPEAR || COMPILE_TEST depends on OF help Enable this to plug the SPEAr thermal sensor driver into the Linux From 444f9b00073668f028db0c14f609cf08abc9a2b3 Mon Sep 17 00:00:00 2001 From: Eduardo Valentin Date: Wed, 9 Sep 2015 20:43:57 -0700 Subject: [PATCH 062/647] thermal: rockchip: allow compile test Adding COMPILE_TEST flag to rockchip driver to facilitate maintenance. Cc: Zhang Rui Cc: Heiko Stuebner Cc: linux-pm@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org Cc: linux-rockchip@lists.infradead.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Eduardo Valentin --- drivers/thermal/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index e2cfaa44ee1b..f114dd6919f2 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -190,7 +190,7 @@ config SPEAR_THERMAL config ROCKCHIP_THERMAL tristate "Rockchip thermal driver" - depends on ARCH_ROCKCHIP + depends on ARCH_ROCKCHIP || COMPILE_TEST depends on RESET_CONTROLLER help Rockchip thermal driver provides support for Temperature sensor From 9c8aa959e13f5cb082ad244d2c9b806a67d9a8b6 Mon Sep 17 00:00:00 2001 From: Eduardo Valentin Date: Wed, 9 Sep 2015 20:44:46 -0700 Subject: [PATCH 063/647] thermal: kirkwood: allow compile test Adding COMPILE_TEST flag to kirkwood driver to facilitate maintenance. Cc: Zhang Rui Cc: linux-pm@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Eduardo Valentin --- drivers/thermal/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index f114dd6919f2..1b74f4eae16d 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -208,7 +208,7 @@ config RCAR_THERMAL config KIRKWOOD_THERMAL tristate "Temperature sensor on Marvell Kirkwood SoCs" - depends on MACH_KIRKWOOD + depends on MACH_KIRKWOOD || COMPILE_TEST depends on OF help Support for the Kirkwood thermal sensor driver into the Linux thermal From 07fffd5ccab70bc00fd69b9ce2297a5e8d05b838 Mon Sep 17 00:00:00 2001 From: Eduardo Valentin Date: Wed, 9 Sep 2015 20:45:25 -0700 Subject: [PATCH 064/647] thermal: dove: allow compile test Adding COMPILE_TEST flag to dove driver to facilitate maintenance. Cc: Zhang Rui Cc: linux-pm@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Eduardo Valentin --- drivers/thermal/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index 1b74f4eae16d..09fe5485ba7e 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -216,7 +216,7 @@ config KIRKWOOD_THERMAL config DOVE_THERMAL tristate "Temperature sensor on Marvell Dove SoCs" - depends on ARCH_DOVE || MACH_DOVE + depends on ARCH_DOVE || MACH_DOVE || COMPILE_TEST depends on OF help Support for the Dove thermal sensor driver in the Linux thermal From 1b1582672121e1c1138e68b64128d94e7f61d10c Mon Sep 17 00:00:00 2001 From: Eduardo Valentin Date: Wed, 9 Sep 2015 20:48:00 -0700 Subject: [PATCH 065/647] thermal: armada: allow compile test Adding COMPILE_TEST flag to armada driver to facilitate maintenance. Cc: Zhang Rui Cc: linux-pm@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Eduardo Valentin --- drivers/thermal/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index 09fe5485ba7e..f153f5871382 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -234,7 +234,7 @@ config DB8500_THERMAL config ARMADA_THERMAL tristate "Armada 370/XP thermal management" - depends on ARCH_MVEBU + depends on ARCH_MVEBU || COMPILE_TEST depends on OF help Enable this option if you want to have support for thermal management From 2bf427353eb606481d15d2843c76cdffa29ed418 Mon Sep 17 00:00:00 2001 From: Eduardo Valentin Date: Wed, 9 Sep 2015 20:53:35 -0700 Subject: [PATCH 066/647] thermal: exynos: allow compile test Adding COMPILE_TEST flag to exynos driver to facilitate maintenance. Cc: Lukasz Majewski Cc: Zhang Rui Cc: linux-pm@vger.kernel.org Cc: linux-samsung-soc@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Eduardo Valentin --- drivers/thermal/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index f153f5871382..15f7cde46628 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -353,7 +353,7 @@ source "drivers/thermal/ti-soc-thermal/Kconfig" endmenu menu "Samsung thermal drivers" -depends on ARCH_EXYNOS +depends on ARCH_EXYNOS || COMPILE_TEST source "drivers/thermal/samsung/Kconfig" endmenu From cb7fb4d342023d03070fcc1ed73fcd43682973d7 Mon Sep 17 00:00:00 2001 From: Eduardo Valentin Date: Wed, 9 Sep 2015 20:56:26 -0700 Subject: [PATCH 067/647] thermal: qcom_spmi: allow compile test Adding COMPILE_TEST flag to qcom_spmi driver to facilitate maintenance. Cc: Zhang Rui Cc: linux-pm@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Eduardo Valentin --- drivers/thermal/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index 15f7cde46628..7d95587b2757 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -364,7 +364,7 @@ endmenu config QCOM_SPMI_TEMP_ALARM tristate "Qualcomm SPMI PMIC Temperature Alarm" - depends on OF && SPMI && IIO + depends on OF && (SPMI || COMPILE_TEST) && IIO select REGMAP_SPMI help This enables a thermal sysfs driver for Qualcomm plug-and-play (QPNP) From 41ae3c02743e8f1f21ceebca614342d26b4eb080 Mon Sep 17 00:00:00 2001 From: Eduardo Valentin Date: Wed, 9 Sep 2015 20:58:36 -0700 Subject: [PATCH 068/647] thermal: ti-soc: allow compile test Adding COMPILE_TEST flag to ti-soc driver to facilitate maintenance. Cc: Zhang Rui Cc: linux-pm@vger.kernel.org Cc: linux-omap@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Eduardo Valentin --- drivers/thermal/ti-soc-thermal/Kconfig | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/thermal/ti-soc-thermal/Kconfig b/drivers/thermal/ti-soc-thermal/Kconfig index bd4c7beba679..7a24018df7cd 100644 --- a/drivers/thermal/ti-soc-thermal/Kconfig +++ b/drivers/thermal/ti-soc-thermal/Kconfig @@ -1,7 +1,7 @@ config TI_SOC_THERMAL tristate "Texas Instruments SoCs temperature sensor driver" depends on THERMAL - depends on ARCH_HAS_BANDGAP + depends on ARCH_HAS_BANDGAP || COMPILE_TEST help If you say yes here you get support for the Texas Instruments OMAP4460+ on die bandgap temperature sensor support. The register @@ -24,7 +24,7 @@ config TI_THERMAL config OMAP4_THERMAL bool "Texas Instruments OMAP4 thermal support" depends on TI_SOC_THERMAL - depends on ARCH_OMAP4 + depends on ARCH_OMAP4 || COMPILE_TEST help If you say yes here you get thermal support for the Texas Instruments OMAP4 SoC family. The current chip supported are: @@ -38,7 +38,7 @@ config OMAP4_THERMAL config OMAP5_THERMAL bool "Texas Instruments OMAP5 thermal support" depends on TI_SOC_THERMAL - depends on SOC_OMAP5 + depends on SOC_OMAP5 || COMPILE_TEST help If you say yes here you get thermal support for the Texas Instruments OMAP5 SoC family. The current chip supported are: @@ -50,7 +50,7 @@ config OMAP5_THERMAL config DRA752_THERMAL bool "Texas Instruments DRA752 thermal support" depends on TI_SOC_THERMAL - depends on SOC_DRA7XX + depends on SOC_DRA7XX || COMPILE_TEST help If you say yes here you get thermal support for the Texas Instruments DRA752 SoC family. The current chip supported are: From ec2feb475f4880246166a47e066e7a601a6103af Mon Sep 17 00:00:00 2001 From: Eduardo Valentin Date: Wed, 9 Sep 2015 21:13:41 -0700 Subject: [PATCH 069/647] thermal: ti-soc: Kconfig fix to avoid menu showing wrongly Move the dependencies to menu, so we avoid showing it wrongly. Cc: Zhang Rui Cc: linux-pm@vger.kernel.org Cc: linux-omap@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Eduardo Valentin --- drivers/thermal/Kconfig | 1 + drivers/thermal/ti-soc-thermal/Kconfig | 2 -- 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index 7d95587b2757..5aabc4bc0d75 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -349,6 +349,7 @@ config INTEL_PCH_THERMAL programmable trip points and other information. menu "Texas Instruments thermal drivers" +depends on ARCH_HAS_BANDGAP || COMPILE_TEST source "drivers/thermal/ti-soc-thermal/Kconfig" endmenu diff --git a/drivers/thermal/ti-soc-thermal/Kconfig b/drivers/thermal/ti-soc-thermal/Kconfig index 7a24018df7cd..cb6686ff09ae 100644 --- a/drivers/thermal/ti-soc-thermal/Kconfig +++ b/drivers/thermal/ti-soc-thermal/Kconfig @@ -1,7 +1,5 @@ config TI_SOC_THERMAL tristate "Texas Instruments SoCs temperature sensor driver" - depends on THERMAL - depends on ARCH_HAS_BANDGAP || COMPILE_TEST help If you say yes here you get support for the Texas Instruments OMAP4460+ on die bandgap temperature sensor support. The register From 64e05d8bcc837bc66cb2c6e71796c8883aa45939 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Thu, 27 Aug 2015 07:32:11 +0530 Subject: [PATCH 070/647] thermal: cpu_cooling: Add MAINTAINERS entry None of the patches are reaching Viresh or Daniel directly as get_maintainers doesn't report us as maintainers. Looks like file header or history of commits isn't able to do that properly. Add a separate entry for cpu_cooling driver in MAINTAINERS. Acked-by: Eduardo Valentin Acked-by: Amit Daniel Kachhap Acked-by: Amit Daniel Kachhap Signed-off-by: Viresh Kumar Signed-off-by: Eduardo Valentin --- MAINTAINERS | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 7ba7ab749c85..6cc919abf246 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -10338,6 +10338,16 @@ F: include/uapi/linux/thermal.h F: include/linux/cpu_cooling.h F: Documentation/devicetree/bindings/thermal/ +THERMAL/CPU_COOLING +M: Amit Daniel Kachhap +M: Viresh Kumar +M: Javi Merino +L: linux-pm@vger.kernel.org +S: Supported +F: Documentation/thermal/cpu-cooling-api.txt +F: drivers/thermal/cpu_cooling.c +F: include/linux/cpu_cooling.h + THINGM BLINK(1) USB RGB LED DRIVER M: Vivien Didelot S: Maintained From 0847e26a84faa0fcb4572d7e42d0a44cab69a83d Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Thu, 3 Sep 2015 13:10:14 +0200 Subject: [PATCH 071/647] thermal: db8500_cpufreq_cooling: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: Eduardo Valentin --- drivers/thermal/db8500_cpufreq_cooling.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/thermal/db8500_cpufreq_cooling.c b/drivers/thermal/db8500_cpufreq_cooling.c index 607b62c7e611..e58bd0b658b5 100644 --- a/drivers/thermal/db8500_cpufreq_cooling.c +++ b/drivers/thermal/db8500_cpufreq_cooling.c @@ -72,6 +72,7 @@ static const struct of_device_id db8500_cpufreq_cooling_match[] = { { .compatible = "stericsson,db8500-cpufreq-cooling" }, {}, }; +MODULE_DEVICE_TABLE(of, db8500_cpufreq_cooling_match); #endif static struct platform_driver db8500_cpufreq_cooling_driver = { From 459ac37506d195713b5e82271a2ac44a777e47df Mon Sep 17 00:00:00 2001 From: Javi Merino Date: Mon, 17 Aug 2015 19:21:42 +0100 Subject: [PATCH 072/647] thermal: cpu_cooling: don't call kcalloc() under rcu_read_lock build_dyn_power_table() allocates the power table while holding rcu_read_lock. kcalloc using GFP_KERNEL may sleep, so it can't be called in an RCU read-side path. Move the rcu protection to the part of the function that really needs it: the part that handles the dev_pm_opp pointer received from dev_pm_opp_find_freq_ceil(). In the unlikely case that there is an OPP added to the cpu while this function is running, return -EAGAIN. Fixes: c36cf0717631 ("thermal: cpu_cooling: implement the power cooling device API") Cc: Zhang Rui Cc: Eduardo Valentin Signed-off-by: Javi Merino Signed-off-by: Eduardo Valentin --- drivers/thermal/cpu_cooling.c | 47 +++++++++++++++++------------------ 1 file changed, 23 insertions(+), 24 deletions(-) diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c index 620dcd405ff6..b32755679395 100644 --- a/drivers/thermal/cpu_cooling.c +++ b/drivers/thermal/cpu_cooling.c @@ -262,7 +262,9 @@ static int cpufreq_thermal_notifier(struct notifier_block *nb, * efficiently. Power is stored in mW, frequency in KHz. The * resulting table is in ascending order. * - * Return: 0 on success, -E* on error. + * Return: 0 on success, -EINVAL if there are no OPPs for any CPUs, + * -ENOMEM if we run out of memory or -EAGAIN if an OPP was + * added/enabled while the function was executing. */ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, u32 capacitance) @@ -270,11 +272,9 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, struct power_table *power_table; struct dev_pm_opp *opp; struct device *dev = NULL; - int num_opps = 0, cpu, i, ret = 0; + int num_opps = 0, cpu, i; unsigned long freq; - rcu_read_lock(); - for_each_cpu(cpu, &cpufreq_device->allowed_cpus) { dev = get_cpu_device(cpu); if (!dev) { @@ -284,24 +284,20 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, } num_opps = dev_pm_opp_get_opp_count(dev); - if (num_opps > 0) { + if (num_opps > 0) break; - } else if (num_opps < 0) { - ret = num_opps; - goto unlock; - } + else if (num_opps < 0) + return num_opps; } - if (num_opps == 0) { - ret = -EINVAL; - goto unlock; - } + if (num_opps == 0) + return -EINVAL; power_table = kcalloc(num_opps, sizeof(*power_table), GFP_KERNEL); - if (!power_table) { - ret = -ENOMEM; - goto unlock; - } + if (!power_table) + return -ENOMEM; + + rcu_read_lock(); for (freq = 0, i = 0; opp = dev_pm_opp_find_freq_ceil(dev, &freq), !IS_ERR(opp); @@ -309,6 +305,11 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, u32 freq_mhz, voltage_mv; u64 power; + if (i >= num_opps) { + rcu_read_unlock(); + return -EAGAIN; + } + freq_mhz = freq / 1000000; voltage_mv = dev_pm_opp_get_voltage(opp) / 1000; @@ -326,18 +327,16 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, power_table[i].power = power; } - if (i == 0) { - ret = PTR_ERR(opp); - goto unlock; - } + rcu_read_unlock(); + + if (i != num_opps) + return PTR_ERR(opp); cpufreq_device->cpu_dev = dev; cpufreq_device->dyn_power_table = power_table; cpufreq_device->dyn_power_table_entries = i; -unlock: - rcu_read_unlock(); - return ret; + return 0; } static u32 cpu_freq_to_power(struct cpufreq_cooling_device *cpufreq_device, From eba4f88d5af84e0fcaa5d6eb4fe35a75c47203cb Mon Sep 17 00:00:00 2001 From: Javi Merino Date: Mon, 17 Aug 2015 19:21:43 +0100 Subject: [PATCH 073/647] thermal: cpu_cooling: free power table on error or when unregistering The power table is not being freed on error from cpufreq_cooling register or when unregistering. Free it. Fixes: c36cf0717631 ("thermal: cpu_cooling: implement the power cooling device API") Cc: Zhang Rui Cc: Eduardo Valentin Signed-off-by: Javi Merino Signed-off-by: Eduardo Valentin --- drivers/thermal/cpu_cooling.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c index b32755679395..42c6f71bdcc1 100644 --- a/drivers/thermal/cpu_cooling.c +++ b/drivers/thermal/cpu_cooling.c @@ -272,7 +272,7 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, struct power_table *power_table; struct dev_pm_opp *opp; struct device *dev = NULL; - int num_opps = 0, cpu, i; + int num_opps = 0, cpu, i, ret = 0; unsigned long freq; for_each_cpu(cpu, &cpufreq_device->allowed_cpus) { @@ -307,7 +307,8 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, if (i >= num_opps) { rcu_read_unlock(); - return -EAGAIN; + ret = -EAGAIN; + goto free_power_table; } freq_mhz = freq / 1000000; @@ -329,14 +330,21 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, rcu_read_unlock(); - if (i != num_opps) - return PTR_ERR(opp); + if (i != num_opps) { + ret = PTR_ERR(opp); + goto free_power_table; + } cpufreq_device->cpu_dev = dev; cpufreq_device->dyn_power_table = power_table; cpufreq_device->dyn_power_table_entries = i; return 0; + +free_power_table: + kfree(power_table); + + return ret; } static u32 cpu_freq_to_power(struct cpufreq_cooling_device *cpufreq_device, @@ -846,7 +854,7 @@ __cpufreq_cooling_register(struct device_node *np, ret = get_idr(&cpufreq_idr, &cpufreq_dev->id); if (ret) { cool_dev = ERR_PTR(ret); - goto free_table; + goto free_power_table; } snprintf(dev_name, sizeof(dev_name), "thermal-cpufreq-%d", @@ -888,6 +896,8 @@ __cpufreq_cooling_register(struct device_node *np, remove_idr: release_idr(&cpufreq_idr, cpufreq_dev->id); +free_power_table: + kfree(cpufreq_dev->dyn_power_table); free_table: kfree(cpufreq_dev->freq_table); free_time_in_idle_timestamp: @@ -1038,6 +1048,7 @@ void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev) thermal_cooling_device_unregister(cpufreq_dev->cool_dev); release_idr(&cpufreq_idr, cpufreq_dev->id); + kfree(cpufreq_dev->dyn_power_table); kfree(cpufreq_dev->time_in_idle_timestamp); kfree(cpufreq_dev->time_in_idle); kfree(cpufreq_dev->freq_table); From 19ab6bc5674a30fdb6a2436b068d19a3c17dc73e Mon Sep 17 00:00:00 2001 From: "Liu.Zhao" Date: Mon, 24 Aug 2015 08:36:12 -0700 Subject: [PATCH 074/647] USB: option: add ZTE PIDs This is intended to add ZTE device PIDs on kernel. Signed-off-by: Liu.Zhao Cc: stable [johan: sort the new entries ] Signed-off-by: Johan Hovold --- drivers/usb/serial/option.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index 6d1941a2396a..6956c4f62216 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -278,6 +278,10 @@ static void option_instat_callback(struct urb *urb); #define ZTE_PRODUCT_MF622 0x0001 #define ZTE_PRODUCT_MF628 0x0015 #define ZTE_PRODUCT_MF626 0x0031 +#define ZTE_PRODUCT_ZM8620_X 0x0396 +#define ZTE_PRODUCT_ME3620_MBIM 0x0426 +#define ZTE_PRODUCT_ME3620_X 0x1432 +#define ZTE_PRODUCT_ME3620_L 0x1433 #define ZTE_PRODUCT_AC2726 0xfff1 #define ZTE_PRODUCT_MG880 0xfffd #define ZTE_PRODUCT_CDMA_TECH 0xfffe @@ -544,6 +548,18 @@ static const struct option_blacklist_info zte_mc2716_z_blacklist = { .sendsetup = BIT(1) | BIT(2) | BIT(3), }; +static const struct option_blacklist_info zte_me3620_mbim_blacklist = { + .reserved = BIT(2) | BIT(3) | BIT(4), +}; + +static const struct option_blacklist_info zte_me3620_xl_blacklist = { + .reserved = BIT(3) | BIT(4) | BIT(5), +}; + +static const struct option_blacklist_info zte_zm8620_x_blacklist = { + .reserved = BIT(3) | BIT(4) | BIT(5), +}; + static const struct option_blacklist_info huawei_cdc12_blacklist = { .reserved = BIT(1) | BIT(2), }; @@ -1591,6 +1607,14 @@ static const struct usb_device_id option_ids[] = { .driver_info = (kernel_ulong_t)&zte_ad3812_z_blacklist }, { USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, ZTE_PRODUCT_MC2716, 0xff, 0xff, 0xff), .driver_info = (kernel_ulong_t)&zte_mc2716_z_blacklist }, + { USB_DEVICE(ZTE_VENDOR_ID, ZTE_PRODUCT_ME3620_L), + .driver_info = (kernel_ulong_t)&zte_me3620_xl_blacklist }, + { USB_DEVICE(ZTE_VENDOR_ID, ZTE_PRODUCT_ME3620_MBIM), + .driver_info = (kernel_ulong_t)&zte_me3620_mbim_blacklist }, + { USB_DEVICE(ZTE_VENDOR_ID, ZTE_PRODUCT_ME3620_X), + .driver_info = (kernel_ulong_t)&zte_me3620_xl_blacklist }, + { USB_DEVICE(ZTE_VENDOR_ID, ZTE_PRODUCT_ZM8620_X), + .driver_info = (kernel_ulong_t)&zte_zm8620_x_blacklist }, { USB_VENDOR_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff, 0x02, 0x01) }, { USB_VENDOR_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff, 0x02, 0x05) }, { USB_VENDOR_AND_INTERFACE_INFO(ZTE_VENDOR_ID, 0xff, 0x86, 0x10) }, From 54c12bc374408faddbff75dbf1a6167c19af39c4 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Mon, 14 Sep 2015 01:13:11 -0700 Subject: [PATCH 075/647] drm/vmwgfx: Fix up user_dmabuf refcounting If user space calls unreference on a user_dmabuf it will typically kill the struct ttm_base_object member which is responsible for the user-space visibility. However the dmabuf part may still be alive and refcounted. In some situations, like for shared guest-backed surface referencing/opening, the driver may try to reference the struct ttm_base_object member again, causing an immediate kernel warning and a later kernel NULL pointer dereference. Fix this by always maintaining a reference on the struct ttm_base_object member, in situations where it might subsequently be referenced. Cc: Signed-off-by: Thomas Hellstrom Reviewed-by: Brian Paul Reviewed-by: Sinclair Yeh --- drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 6 +++-- drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 6 +++-- drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c | 2 +- drivers/gpu/drm/vmwgfx/vmwgfx_resource.c | 29 +++++++++++++++++------- drivers/gpu/drm/vmwgfx/vmwgfx_shader.c | 2 +- drivers/gpu/drm/vmwgfx/vmwgfx_surface.c | 11 ++++++--- 6 files changed, 39 insertions(+), 17 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 8f40692cf48a..b60b41f95e74 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -631,7 +631,8 @@ extern int vmw_user_dmabuf_alloc(struct vmw_private *dev_priv, uint32_t size, bool shareable, uint32_t *handle, - struct vmw_dma_buffer **p_dma_buf); + struct vmw_dma_buffer **p_dma_buf, + struct ttm_base_object **p_base); extern int vmw_user_dmabuf_reference(struct ttm_object_file *tfile, struct vmw_dma_buffer *dma_buf, uint32_t *handle); @@ -645,7 +646,8 @@ extern uint32_t vmw_dmabuf_validate_node(struct ttm_buffer_object *bo, uint32_t cur_validate_node); extern void vmw_dmabuf_validate_clear(struct ttm_buffer_object *bo); extern int vmw_user_dmabuf_lookup(struct ttm_object_file *tfile, - uint32_t id, struct vmw_dma_buffer **out); + uint32_t id, struct vmw_dma_buffer **out, + struct ttm_base_object **base); extern int vmw_stream_claim_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); extern int vmw_stream_unref_ioctl(struct drm_device *dev, void *data, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index b56565457c96..5da5de0cb522 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -1236,7 +1236,8 @@ static int vmw_translate_mob_ptr(struct vmw_private *dev_priv, struct vmw_relocation *reloc; int ret; - ret = vmw_user_dmabuf_lookup(sw_context->fp->tfile, handle, &vmw_bo); + ret = vmw_user_dmabuf_lookup(sw_context->fp->tfile, handle, &vmw_bo, + NULL); if (unlikely(ret != 0)) { DRM_ERROR("Could not find or use MOB buffer.\n"); ret = -EINVAL; @@ -1296,7 +1297,8 @@ static int vmw_translate_guest_ptr(struct vmw_private *dev_priv, struct vmw_relocation *reloc; int ret; - ret = vmw_user_dmabuf_lookup(sw_context->fp->tfile, handle, &vmw_bo); + ret = vmw_user_dmabuf_lookup(sw_context->fp->tfile, handle, &vmw_bo, + NULL); if (unlikely(ret != 0)) { DRM_ERROR("Could not find or use GMR region.\n"); ret = -EINVAL; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c index 76069f093ccf..222c9c2123a1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c @@ -484,7 +484,7 @@ int vmw_overlay_ioctl(struct drm_device *dev, void *data, goto out_unlock; } - ret = vmw_user_dmabuf_lookup(tfile, arg->handle, &buf); + ret = vmw_user_dmabuf_lookup(tfile, arg->handle, &buf, NULL); if (ret) goto out_unlock; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c index c1912f852b42..e57667ca7557 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c @@ -354,7 +354,7 @@ int vmw_user_lookup_handle(struct vmw_private *dev_priv, } *out_surf = NULL; - ret = vmw_user_dmabuf_lookup(tfile, handle, out_buf); + ret = vmw_user_dmabuf_lookup(tfile, handle, out_buf, NULL); return ret; } @@ -481,7 +481,8 @@ int vmw_user_dmabuf_alloc(struct vmw_private *dev_priv, uint32_t size, bool shareable, uint32_t *handle, - struct vmw_dma_buffer **p_dma_buf) + struct vmw_dma_buffer **p_dma_buf, + struct ttm_base_object **p_base) { struct vmw_user_dma_buffer *user_bo; struct ttm_buffer_object *tmp; @@ -515,6 +516,10 @@ int vmw_user_dmabuf_alloc(struct vmw_private *dev_priv, } *p_dma_buf = &user_bo->dma; + if (p_base) { + *p_base = &user_bo->prime.base; + kref_get(&(*p_base)->refcount); + } *handle = user_bo->prime.base.hash.key; out_no_base_object: @@ -631,6 +636,7 @@ int vmw_user_dmabuf_synccpu_ioctl(struct drm_device *dev, void *data, struct vmw_dma_buffer *dma_buf; struct vmw_user_dma_buffer *user_bo; struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; + struct ttm_base_object *buffer_base; int ret; if ((arg->flags & (drm_vmw_synccpu_read | drm_vmw_synccpu_write)) == 0 @@ -643,7 +649,8 @@ int vmw_user_dmabuf_synccpu_ioctl(struct drm_device *dev, void *data, switch (arg->op) { case drm_vmw_synccpu_grab: - ret = vmw_user_dmabuf_lookup(tfile, arg->handle, &dma_buf); + ret = vmw_user_dmabuf_lookup(tfile, arg->handle, &dma_buf, + &buffer_base); if (unlikely(ret != 0)) return ret; @@ -651,6 +658,7 @@ int vmw_user_dmabuf_synccpu_ioctl(struct drm_device *dev, void *data, dma); ret = vmw_user_dmabuf_synccpu_grab(user_bo, tfile, arg->flags); vmw_dmabuf_unreference(&dma_buf); + ttm_base_object_unref(&buffer_base); if (unlikely(ret != 0 && ret != -ERESTARTSYS && ret != -EBUSY)) { DRM_ERROR("Failed synccpu grab on handle 0x%08x.\n", @@ -692,7 +700,8 @@ int vmw_dmabuf_alloc_ioctl(struct drm_device *dev, void *data, return ret; ret = vmw_user_dmabuf_alloc(dev_priv, vmw_fpriv(file_priv)->tfile, - req->size, false, &handle, &dma_buf); + req->size, false, &handle, &dma_buf, + NULL); if (unlikely(ret != 0)) goto out_no_dmabuf; @@ -721,7 +730,8 @@ int vmw_dmabuf_unref_ioctl(struct drm_device *dev, void *data, } int vmw_user_dmabuf_lookup(struct ttm_object_file *tfile, - uint32_t handle, struct vmw_dma_buffer **out) + uint32_t handle, struct vmw_dma_buffer **out, + struct ttm_base_object **p_base) { struct vmw_user_dma_buffer *vmw_user_bo; struct ttm_base_object *base; @@ -743,7 +753,10 @@ int vmw_user_dmabuf_lookup(struct ttm_object_file *tfile, vmw_user_bo = container_of(base, struct vmw_user_dma_buffer, prime.base); (void)ttm_bo_reference(&vmw_user_bo->dma.base); - ttm_base_object_unref(&base); + if (p_base) + *p_base = base; + else + ttm_base_object_unref(&base); *out = &vmw_user_bo->dma; return 0; @@ -1004,7 +1017,7 @@ int vmw_dumb_create(struct drm_file *file_priv, ret = vmw_user_dmabuf_alloc(dev_priv, vmw_fpriv(file_priv)->tfile, args->size, false, &args->handle, - &dma_buf); + &dma_buf, NULL); if (unlikely(ret != 0)) goto out_no_dmabuf; @@ -1032,7 +1045,7 @@ int vmw_dumb_map_offset(struct drm_file *file_priv, struct vmw_dma_buffer *out_buf; int ret; - ret = vmw_user_dmabuf_lookup(tfile, handle, &out_buf); + ret = vmw_user_dmabuf_lookup(tfile, handle, &out_buf, NULL); if (ret != 0) return -EINVAL; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c index bba1ee395478..fd47547b0234 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c @@ -855,7 +855,7 @@ static int vmw_shader_define(struct drm_device *dev, struct drm_file *file_priv, if (buffer_handle != SVGA3D_INVALID_ID) { ret = vmw_user_dmabuf_lookup(tfile, buffer_handle, - &buffer); + &buffer, NULL); if (unlikely(ret != 0)) { DRM_ERROR("Could not find buffer for shader " "creation.\n"); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c index 3361769842f4..64b50409fa07 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c @@ -46,6 +46,7 @@ struct vmw_user_surface { struct vmw_surface srf; uint32_t size; struct drm_master *master; + struct ttm_base_object *backup_base; }; /** @@ -656,6 +657,7 @@ static void vmw_user_surface_base_release(struct ttm_base_object **p_base) struct vmw_resource *res = &user_srf->srf.res; *p_base = NULL; + ttm_base_object_unref(&user_srf->backup_base); vmw_resource_unreference(&res); } @@ -851,7 +853,8 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, res->backup_size, true, &backup_handle, - &res->backup); + &res->backup, + &user_srf->backup_base); if (unlikely(ret != 0)) { vmw_resource_unreference(&res); goto out_unlock; @@ -1321,7 +1324,8 @@ int vmw_gb_surface_define_ioctl(struct drm_device *dev, void *data, if (req->buffer_handle != SVGA3D_INVALID_ID) { ret = vmw_user_dmabuf_lookup(tfile, req->buffer_handle, - &res->backup); + &res->backup, + &user_srf->backup_base); if (ret == 0 && res->backup->base.num_pages * PAGE_SIZE < res->backup_size) { DRM_ERROR("Surface backup buffer is too small.\n"); @@ -1335,7 +1339,8 @@ int vmw_gb_surface_define_ioctl(struct drm_device *dev, void *data, req->drm_surface_flags & drm_vmw_surface_flag_shareable, &backup_handle, - &res->backup); + &res->backup, + &user_srf->backup_base); if (unlikely(ret != 0)) { vmw_resource_unreference(&res); From 2e586a7e017a502410ba1c1a7411179e1d3fbb2b Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Mon, 14 Sep 2015 01:17:43 -0700 Subject: [PATCH 076/647] drm/vmwgfx: Map the fifo as cached On the guest kernel side, previously the FIFO has been mapped write- combined. This has worked since VMs up to now has not honored the mapping type and mapped the FIFO cached anyway. Since the FIFO is accessed cached by the CPU on the virtual device side, this leads to inconsistent mappings once the guest starts to honor the mapping types. So ask for cached mappings when we map the FIFO. We do this by using ioremap_cache() instead of ioremap_wc(), and remove the MTRR setup. On the TTM side, MOBs, GMRs and VRAM buffers are already requesting cached mappings for kernel- and user-space. Cc: Signed-off-by: Thomas Hellstrom Reviewed-by: Sinclair Yeh --- drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 10 ++-------- drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 1 - 2 files changed, 2 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index e13b20bd9908..2c7a25c71af2 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -752,12 +752,8 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) ttm_lock_set_kill(&dev_priv->fbdev_master.lock, false, SIGTERM); dev_priv->active_master = &dev_priv->fbdev_master; - - dev_priv->mmio_mtrr = arch_phys_wc_add(dev_priv->mmio_start, - dev_priv->mmio_size); - - dev_priv->mmio_virt = ioremap_wc(dev_priv->mmio_start, - dev_priv->mmio_size); + dev_priv->mmio_virt = ioremap_cache(dev_priv->mmio_start, + dev_priv->mmio_size); if (unlikely(dev_priv->mmio_virt == NULL)) { ret = -ENOMEM; @@ -913,7 +909,6 @@ out_no_device: out_err4: iounmap(dev_priv->mmio_virt); out_err3: - arch_phys_wc_del(dev_priv->mmio_mtrr); vmw_ttm_global_release(dev_priv); out_err0: for (i = vmw_res_context; i < vmw_res_max; ++i) @@ -964,7 +959,6 @@ static int vmw_driver_unload(struct drm_device *dev) ttm_object_device_release(&dev_priv->tdev); iounmap(dev_priv->mmio_virt); - arch_phys_wc_del(dev_priv->mmio_mtrr); if (dev_priv->ctx.staged_bindings) vmw_binding_state_free(dev_priv->ctx.staged_bindings); vmw_ttm_global_release(dev_priv); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index b60b41f95e74..a20f482848ee 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -376,7 +376,6 @@ struct vmw_private { uint32_t initial_width; uint32_t initial_height; u32 __iomem *mmio_virt; - int mmio_mtrr; uint32_t capabilities; uint32_t max_gmr_ids; uint32_t max_gmr_pages; From 205ee117d4dc4a11ac3bd9638bb9b2e839f4de9a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 9 Sep 2015 02:57:21 +0200 Subject: [PATCH 077/647] netfilter: nf_log: don't zap all loggers on unregister like nf_log_unset, nf_log_unregister must not reset the list of loggers. Otherwise, a call to nf_log_unregister() will render loggers of other nf protocols unusable: iptables -A INPUT -j LOG modprobe nf_log_arp ; rmmod nf_log_arp iptables -A INPUT -j LOG iptables: No chain/target/match by that name Fixes: 30e0c6a6be ("netfilter: nf_log: prepare net namespace support for loggers") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_log.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index 675d12c69e32..a5ebd7d9c472 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -107,11 +107,15 @@ EXPORT_SYMBOL(nf_log_register); void nf_log_unregister(struct nf_logger *logger) { + const struct nf_logger *log; int i; mutex_lock(&nf_log_mutex); - for (i = 0; i < NFPROTO_NUMPROTO; i++) - RCU_INIT_POINTER(loggers[i][logger->type], NULL); + for (i = 0; i < NFPROTO_NUMPROTO; i++) { + log = nft_log_dereference(loggers[i][logger->type]); + if (log == logger) + RCU_INIT_POINTER(loggers[i][logger->type], NULL); + } mutex_unlock(&nf_log_mutex); } EXPORT_SYMBOL(nf_log_unregister); From c973c3bcec3752455c4d7545edd42935cd7942d9 Mon Sep 17 00:00:00 2001 From: Javi Merino Date: Mon, 14 Sep 2015 14:23:50 +0100 Subject: [PATCH 078/647] thermal: Add a function to get the minimum power The thermal core already has a function to get the maximum power of a cooling device: power_actor_get_max_power(). Add a function to get the minimum power of a cooling device. Cc: Zhang Rui Cc: Eduardo Valentin Reviewed-by: Daniel Kurtz Signed-off-by: Javi Merino Signed-off-by: Eduardo Valentin --- drivers/thermal/thermal_core.c | 28 ++++++++++++++++++++++++++++ include/linux/thermal.h | 6 ++++++ 2 files changed, 34 insertions(+) diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c index 5e5fc7015c7f..d9e525cc9c1c 100644 --- a/drivers/thermal/thermal_core.c +++ b/drivers/thermal/thermal_core.c @@ -1012,6 +1012,34 @@ int power_actor_get_max_power(struct thermal_cooling_device *cdev, return cdev->ops->state2power(cdev, tz, 0, max_power); } +/** + * power_actor_get_min_power() - get the mainimum power that a cdev can consume + * @cdev: pointer to &thermal_cooling_device + * @tz: a valid thermal zone device pointer + * @min_power: pointer in which to store the minimum power + * + * Calculate the minimum power consumption in milliwatts that the + * cooling device can currently consume and store it in @min_power. + * + * Return: 0 on success, -EINVAL if @cdev doesn't support the + * power_actor API or -E* on other error. + */ +int power_actor_get_min_power(struct thermal_cooling_device *cdev, + struct thermal_zone_device *tz, u32 *min_power) +{ + unsigned long max_state; + int ret; + + if (!cdev_is_power_actor(cdev)) + return -EINVAL; + + ret = cdev->ops->get_max_state(cdev, &max_state); + if (ret) + return ret; + + return cdev->ops->state2power(cdev, tz, max_state, min_power); +} + /** * power_actor_set_power() - limit the maximum power that a cooling device can consume * @cdev: pointer to &thermal_cooling_device diff --git a/include/linux/thermal.h b/include/linux/thermal.h index 0c5518e13584..157d366e761b 100644 --- a/include/linux/thermal.h +++ b/include/linux/thermal.h @@ -380,6 +380,8 @@ static inline bool cdev_is_power_actor(struct thermal_cooling_device *cdev) int power_actor_get_max_power(struct thermal_cooling_device *, struct thermal_zone_device *tz, u32 *max_power); +int power_actor_get_min_power(struct thermal_cooling_device *, + struct thermal_zone_device *tz, u32 *min_power); int power_actor_set_power(struct thermal_cooling_device *, struct thermal_instance *, u32); struct thermal_zone_device *thermal_zone_device_register(const char *, int, int, @@ -415,6 +417,10 @@ static inline bool cdev_is_power_actor(struct thermal_cooling_device *cdev) static inline int power_actor_get_max_power(struct thermal_cooling_device *cdev, struct thermal_zone_device *tz, u32 *max_power) { return 0; } +static inline int power_actor_get_min_power(struct thermal_cooling_device *cdev, + struct thermal_zone_device *tz, + u32 *min_power) +{ return -ENODEV; } static inline int power_actor_set_power(struct thermal_cooling_device *cdev, struct thermal_instance *tz, u32 power) { return 0; } From e055bb0f9a6e5c09bedf41c2a5b881edbd7f2ed0 Mon Sep 17 00:00:00 2001 From: Javi Merino Date: Mon, 14 Sep 2015 14:23:51 +0100 Subject: [PATCH 079/647] thermal: power_allocator: relax the requirement of a sustainable_power in tzp The power allocator governor currently requires that a sustainable power is passed as part of the thermal zone's thermal zone parameters. If that parameter is not provided, it doesn't register with the thermal zone. While this parameter is strongly recommended for optimal performance, it doesn't need to be mandatory. Relax the requirement and allow the governor to bind to thermal zones that don't provide it by estimating it from the cooling devices' power model. Cc: Zhang Rui Cc: Eduardo Valentin Reviewed-by: Daniel Kurtz Signed-off-by: Javi Merino Signed-off-by: Eduardo Valentin --- drivers/thermal/power_allocator.c | 125 ++++++++++++++++++++++++------ 1 file changed, 100 insertions(+), 25 deletions(-) diff --git a/drivers/thermal/power_allocator.c b/drivers/thermal/power_allocator.c index 9c8a7aad0252..51473473f154 100644 --- a/drivers/thermal/power_allocator.c +++ b/drivers/thermal/power_allocator.c @@ -72,6 +72,88 @@ struct power_allocator_params { int trip_max_desired_temperature; }; +/** + * estimate_sustainable_power() - Estimate the sustainable power of a thermal zone + * @tz: thermal zone we are operating in + * + * For thermal zones that don't provide a sustainable_power in their + * thermal_zone_params, estimate one. Calculate it using the minimum + * power of all the cooling devices as that gives a valid value that + * can give some degree of functionality. For optimal performance of + * this governor, provide a sustainable_power in the thermal zone's + * thermal_zone_params. + */ +static u32 estimate_sustainable_power(struct thermal_zone_device *tz) +{ + u32 sustainable_power = 0; + struct thermal_instance *instance; + struct power_allocator_params *params = tz->governor_data; + + list_for_each_entry(instance, &tz->thermal_instances, tz_node) { + struct thermal_cooling_device *cdev = instance->cdev; + u32 min_power; + + if (instance->trip != params->trip_max_desired_temperature) + continue; + + if (power_actor_get_min_power(cdev, tz, &min_power)) + continue; + + sustainable_power += min_power; + } + + return sustainable_power; +} + +/** + * estimate_pid_constants() - Estimate the constants for the PID controller + * @tz: thermal zone for which to estimate the constants + * @sustainable_power: sustainable power for the thermal zone + * @trip_switch_on: trip point number for the switch on temperature + * @control_temp: target temperature for the power allocator governor + * @force: whether to force the update of the constants + * + * This function is used to update the estimation of the PID + * controller constants in struct thermal_zone_parameters. + * Sustainable power is provided in case it was estimated. The + * estimated sustainable_power should not be stored in the + * thermal_zone_parameters so it has to be passed explicitly to this + * function. + * + * If @force is not set, the values in the thermal zone's parameters + * are preserved if they are not zero. If @force is set, the values + * in thermal zone's parameters are overwritten. + */ +static void estimate_pid_constants(struct thermal_zone_device *tz, + u32 sustainable_power, int trip_switch_on, + int control_temp, bool force) +{ + int ret; + int switch_on_temp; + u32 temperature_threshold; + + ret = tz->ops->get_trip_temp(tz, trip_switch_on, &switch_on_temp); + if (ret) + switch_on_temp = 0; + + temperature_threshold = control_temp - switch_on_temp; + + if (!tz->tzp->k_po || force) + tz->tzp->k_po = int_to_frac(sustainable_power) / + temperature_threshold; + + if (!tz->tzp->k_pu || force) + tz->tzp->k_pu = int_to_frac(2 * sustainable_power) / + temperature_threshold; + + if (!tz->tzp->k_i || force) + tz->tzp->k_i = int_to_frac(10) / 1000; + /* + * The default for k_d and integral_cutoff is 0, so we can + * leave them as they are. + */ +} + /** * pid_controller() - PID controller * @tz: thermal zone we are operating in @@ -98,10 +180,20 @@ static u32 pid_controller(struct thermal_zone_device *tz, { s64 p, i, d, power_range; s32 err, max_power_frac; + u32 sustainable_power; struct power_allocator_params *params = tz->governor_data; max_power_frac = int_to_frac(max_allocatable_power); + if (tz->tzp->sustainable_power) { + sustainable_power = tz->tzp->sustainable_power; + } else { + sustainable_power = estimate_sustainable_power(tz); + estimate_pid_constants(tz, sustainable_power, + params->trip_switch_on, control_temp, + true); + } + err = control_temp - current_temp; err = int_to_frac(err); @@ -139,7 +231,7 @@ static u32 pid_controller(struct thermal_zone_device *tz, power_range = p + i + d; /* feed-forward the known sustainable dissipatable power */ - power_range = tz->tzp->sustainable_power + frac_to_int(power_range); + power_range = sustainable_power + frac_to_int(power_range); power_range = clamp(power_range, (s64)0, (s64)max_allocatable_power); @@ -416,19 +508,18 @@ static int power_allocator_bind(struct thermal_zone_device *tz) { int ret; struct power_allocator_params *params; - int switch_on_temp, control_temp; - u32 temperature_threshold; + int control_temp; - if (!tz->tzp || !tz->tzp->sustainable_power) { - dev_err(&tz->device, - "power_allocator: missing sustainable_power\n"); + if (!tz->tzp) return -EINVAL; - } params = kzalloc(sizeof(*params), GFP_KERNEL); if (!params) return -ENOMEM; + if (!tz->tzp->sustainable_power) + dev_warn(&tz->device, "power_allocator: sustainable_power will be estimated\n"); + ret = get_governor_trips(tz, params); if (ret) { dev_err(&tz->device, @@ -437,29 +528,13 @@ static int power_allocator_bind(struct thermal_zone_device *tz) goto free; } - ret = tz->ops->get_trip_temp(tz, params->trip_switch_on, - &switch_on_temp); - if (ret) - goto free; - ret = tz->ops->get_trip_temp(tz, params->trip_max_desired_temperature, &control_temp); if (ret) goto free; - temperature_threshold = control_temp - switch_on_temp; - - tz->tzp->k_po = tz->tzp->k_po ?: - int_to_frac(tz->tzp->sustainable_power) / temperature_threshold; - tz->tzp->k_pu = tz->tzp->k_pu ?: - int_to_frac(2 * tz->tzp->sustainable_power) / - temperature_threshold; - tz->tzp->k_i = tz->tzp->k_i ?: int_to_frac(10) / 1000; - /* - * The default for k_d and integral_cutoff is 0, so we can - * leave them as they are. - */ - + estimate_pid_constants(tz, tz->tzp->sustainable_power, + params->trip_switch_on, control_temp, false); reset_pid_controller(params); tz->governor_data = params; From 8b7b390f805f09ff252351468a79ebabde1ab55a Mon Sep 17 00:00:00 2001 From: Javi Merino Date: Mon, 14 Sep 2015 14:23:52 +0100 Subject: [PATCH 080/647] thermal: power_allocator: relax the requirement of two passive trip points The power allocator governor currently requires that the thermal zone has at least two passive trip points. If there aren't, the governor refuses to bind to the thermal zone. This commit relaxes that requirement. Now the governor will bind to all thermal zones regardless of how many trip points they have. Cc: Zhang Rui Cc: Eduardo Valentin Reviewed-by: Daniel Kurtz Signed-off-by: Javi Merino Signed-off-by: Eduardo Valentin --- Documentation/thermal/power_allocator.txt | 2 +- drivers/thermal/power_allocator.c | 101 ++++++++++++---------- 2 files changed, 58 insertions(+), 45 deletions(-) diff --git a/Documentation/thermal/power_allocator.txt b/Documentation/thermal/power_allocator.txt index c3797b529991..a1ce2235f121 100644 --- a/Documentation/thermal/power_allocator.txt +++ b/Documentation/thermal/power_allocator.txt @@ -4,7 +4,7 @@ Power allocator governor tunables Trip points ----------- -The governor requires the following two passive trip points: +The governor works optimally with the following two passive trip points: 1. "switch on" trip point: temperature above which the governor control loop starts operating. This is the first passive trip diff --git a/drivers/thermal/power_allocator.c b/drivers/thermal/power_allocator.c index 51473473f154..ea57759eb095 100644 --- a/drivers/thermal/power_allocator.c +++ b/drivers/thermal/power_allocator.c @@ -24,6 +24,8 @@ #include "thermal_core.h" +#define INVALID_TRIP -1 + #define FRAC_BITS 10 #define int_to_frac(x) ((x) << FRAC_BITS) #define frac_to_int(x) ((x) >> FRAC_BITS) @@ -61,6 +63,8 @@ static inline s64 div_frac(s64 x, s64 y) * Used to calculate the derivative term. * @trip_switch_on: first passive trip point of the thermal zone. The * governor switches on when this trip point is crossed. + * If the thermal zone only has one passive trip point, + * @trip_switch_on should be INVALID_TRIP. * @trip_max_desired_temperature: last passive trip point of the thermal * zone. The temperature we are * controlling for. @@ -432,43 +436,66 @@ unlock: return ret; } -static int get_governor_trips(struct thermal_zone_device *tz, - struct power_allocator_params *params) +/** + * get_governor_trips() - get the number of the two trip points that are key for this governor + * @tz: thermal zone to operate on + * @params: pointer to private data for this governor + * + * The power allocator governor works optimally with two trips points: + * a "switch on" trip point and a "maximum desired temperature". These + * are defined as the first and last passive trip points. + * + * If there is only one trip point, then that's considered to be the + * "maximum desired temperature" trip point and the governor is always + * on. If there are no passive or active trip points, then the + * governor won't do anything. In fact, its throttle function + * won't be called at all. + */ +static void get_governor_trips(struct thermal_zone_device *tz, + struct power_allocator_params *params) { - int i, ret, last_passive; + int i, last_active, last_passive; bool found_first_passive; found_first_passive = false; - last_passive = -1; - ret = -EINVAL; + last_active = INVALID_TRIP; + last_passive = INVALID_TRIP; for (i = 0; i < tz->trips; i++) { enum thermal_trip_type type; + int ret; ret = tz->ops->get_trip_type(tz, i, &type); - if (ret) - return ret; + if (ret) { + dev_warn(&tz->device, + "Failed to get trip point %d type: %d\n", i, + ret); + continue; + } - if (!found_first_passive) { - if (type == THERMAL_TRIP_PASSIVE) { + if (type == THERMAL_TRIP_PASSIVE) { + if (!found_first_passive) { params->trip_switch_on = i; found_first_passive = true; + } else { + last_passive = i; } - } else if (type == THERMAL_TRIP_PASSIVE) { - last_passive = i; + } else if (type == THERMAL_TRIP_ACTIVE) { + last_active = i; } else { break; } } - if (last_passive != -1) { + if (last_passive != INVALID_TRIP) { params->trip_max_desired_temperature = last_passive; - ret = 0; + } else if (found_first_passive) { + params->trip_max_desired_temperature = params->trip_switch_on; + params->trip_switch_on = INVALID_TRIP; } else { - ret = -EINVAL; + params->trip_switch_on = INVALID_TRIP; + params->trip_max_desired_temperature = last_active; } - - return ret; } static void reset_pid_controller(struct power_allocator_params *params) @@ -497,11 +524,10 @@ static void allow_maximum_power(struct thermal_zone_device *tz) * power_allocator_bind() - bind the power_allocator governor to a thermal zone * @tz: thermal zone to bind it to * - * Check that the thermal zone is valid for this governor, that is, it - * has two thermal trips. If so, initialize the PID controller - * parameters and bind it to the thermal zone. + * Initialize the PID controller parameters and bind it to the thermal + * zone. * - * Return: 0 on success, -EINVAL if the trips were invalid or -ENOMEM + * Return: 0 on success, -EINVAL if the thermal zone doesn't have tzp or -ENOMEM * if we ran out of memory. */ static int power_allocator_bind(struct thermal_zone_device *tz) @@ -520,30 +546,23 @@ static int power_allocator_bind(struct thermal_zone_device *tz) if (!tz->tzp->sustainable_power) dev_warn(&tz->device, "power_allocator: sustainable_power will be estimated\n"); - ret = get_governor_trips(tz, params); - if (ret) { - dev_err(&tz->device, - "thermal zone %s has wrong trip setup for power allocator\n", - tz->type); - goto free; + get_governor_trips(tz, params); + + if (tz->trips > 0) { + ret = tz->ops->get_trip_temp(tz, + params->trip_max_desired_temperature, + &control_temp); + if (!ret) + estimate_pid_constants(tz, tz->tzp->sustainable_power, + params->trip_switch_on, + control_temp, false); } - ret = tz->ops->get_trip_temp(tz, params->trip_max_desired_temperature, - &control_temp); - if (ret) - goto free; - - estimate_pid_constants(tz, tz->tzp->sustainable_power, - params->trip_switch_on, control_temp, false); reset_pid_controller(params); tz->governor_data = params; return 0; - -free: - kfree(params); - return ret; } static void power_allocator_unbind(struct thermal_zone_device *tz) @@ -574,13 +593,7 @@ static int power_allocator_throttle(struct thermal_zone_device *tz, int trip) ret = tz->ops->get_trip_temp(tz, params->trip_switch_on, &switch_on_temp); - if (ret) { - dev_warn(&tz->device, - "Failed to get switch on temperature: %d\n", ret); - return ret; - } - - if (current_temp < switch_on_temp) { + if (!ret && (current_temp < switch_on_temp)) { tz->passive = 0; reset_pid_controller(params); allow_maximum_power(tz); From f5cbb182586ea36c6ad0e90e3534d18a5e9af094 Mon Sep 17 00:00:00 2001 From: Javi Merino Date: Mon, 14 Sep 2015 14:23:53 +0100 Subject: [PATCH 081/647] thermal: power_allocator: don't require tzp to be present for the thermal zone Thermal zones created using thermal_zone_device_create() may not have tzp. As the governor gets its parameters from there, allocate it while the governor is bound to the thermal zone so that it can operate in it. In this case, tzp is freed when the thermal zone switches to another governor. Cc: Zhang Rui Cc: Eduardo Valentin Reviewed-by: Daniel Kurtz Signed-off-by: Javi Merino Signed-off-by: Eduardo Valentin --- drivers/thermal/power_allocator.c | 32 ++++++++++++++++++++++++++----- 1 file changed, 27 insertions(+), 5 deletions(-) diff --git a/drivers/thermal/power_allocator.c b/drivers/thermal/power_allocator.c index ea57759eb095..718363f5be40 100644 --- a/drivers/thermal/power_allocator.c +++ b/drivers/thermal/power_allocator.c @@ -58,6 +58,8 @@ static inline s64 div_frac(s64 x, s64 y) /** * struct power_allocator_params - parameters for the power allocator governor + * @allocated_tzp: whether we have allocated tzp for this thermal zone and + * it needs to be freed on unbind * @err_integral: accumulated error in the PID controller. * @prev_err: error in the previous iteration of the PID controller. * Used to calculate the derivative term. @@ -70,6 +72,7 @@ static inline s64 div_frac(s64 x, s64 y) * controlling for. */ struct power_allocator_params { + bool allocated_tzp; s64 err_integral; s32 prev_err; int trip_switch_on; @@ -527,8 +530,7 @@ static void allow_maximum_power(struct thermal_zone_device *tz) * Initialize the PID controller parameters and bind it to the thermal * zone. * - * Return: 0 on success, -EINVAL if the thermal zone doesn't have tzp or -ENOMEM - * if we ran out of memory. + * Return: 0 on success, or -ENOMEM if we ran out of memory. */ static int power_allocator_bind(struct thermal_zone_device *tz) { @@ -536,13 +538,20 @@ static int power_allocator_bind(struct thermal_zone_device *tz) struct power_allocator_params *params; int control_temp; - if (!tz->tzp) - return -EINVAL; - params = kzalloc(sizeof(*params), GFP_KERNEL); if (!params) return -ENOMEM; + if (!tz->tzp) { + tz->tzp = kzalloc(sizeof(*tz->tzp), GFP_KERNEL); + if (!tz->tzp) { + ret = -ENOMEM; + goto free_params; + } + + params->allocated_tzp = true; + } + if (!tz->tzp->sustainable_power) dev_warn(&tz->device, "power_allocator: sustainable_power will be estimated\n"); @@ -563,11 +572,24 @@ static int power_allocator_bind(struct thermal_zone_device *tz) tz->governor_data = params; return 0; + +free_params: + kfree(params); + + return ret; } static void power_allocator_unbind(struct thermal_zone_device *tz) { + struct power_allocator_params *params = tz->governor_data; + dev_dbg(&tz->device, "Unbinding from thermal zone %d\n", tz->id); + + if (params->allocated_tzp) { + kfree(tz->tzp); + tz->tzp = NULL; + } + kfree(tz->governor_data); tz->governor_data = NULL; } From e4c1b1ba7f663fd77204201eef83baf58e2e5777 Mon Sep 17 00:00:00 2001 From: Robert Baldyga Date: Fri, 21 Aug 2015 11:01:29 +0200 Subject: [PATCH 082/647] usb: gadget: fix possible regression introduced with ep->claimed This patch fixes possible regression introduced by patch reworking endpoint claiming mechanism. It restores setring ep->driver_data to NULL in usb_ep_autoconfig_reset(), which was removed by patch commit cc476b42a39d. Fixes: cc476b42a39d ("usb: gadget: encapsulate endpoint claiming mechanism") Reported-by: Felipe Balbi Signed-off-by: Robert Baldyga Signed-off-by: Felipe Balbi --- drivers/usb/gadget/epautoconf.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/gadget/epautoconf.c b/drivers/usb/gadget/epautoconf.c index 978435a51038..6399c106a3a5 100644 --- a/drivers/usb/gadget/epautoconf.c +++ b/drivers/usb/gadget/epautoconf.c @@ -186,6 +186,7 @@ void usb_ep_autoconfig_reset (struct usb_gadget *gadget) list_for_each_entry (ep, &gadget->ep_list, ep_list) { ep->claimed = false; + ep->driver_data = NULL; } gadget->in_epnum = 0; gadget->out_epnum = 0; From e2ae0692bf6f71c8b841889b655d0dc08413e4e3 Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Tue, 25 Aug 2015 12:07:45 -0500 Subject: [PATCH 083/647] usb: dwc3: omap: enable irqs lately If we enable IRQs before requesting our extcon device, we might fall into a situation where and IRQ fires before we're ready to handle it. Signed-off-by: Felipe Balbi --- drivers/usb/dwc3/dwc3-omap.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/usb/dwc3/dwc3-omap.c b/drivers/usb/dwc3/dwc3-omap.c index a5a1b7c45743..22e9606d8e08 100644 --- a/drivers/usb/dwc3/dwc3-omap.c +++ b/drivers/usb/dwc3/dwc3-omap.c @@ -514,8 +514,6 @@ static int dwc3_omap_probe(struct platform_device *pdev) goto err1; } - dwc3_omap_enable_irqs(omap); - ret = dwc3_omap_extcon_register(omap); if (ret < 0) goto err2; @@ -526,6 +524,8 @@ static int dwc3_omap_probe(struct platform_device *pdev) goto err3; } + dwc3_omap_enable_irqs(omap); + return 0; err3: From 40af177efc9385af15c49a40976f71e58e6af418 Mon Sep 17 00:00:00 2001 From: Bin Liu Date: Mon, 14 Sep 2015 09:12:34 -0500 Subject: [PATCH 084/647] usb: musb: ensure in peripheral mode when checking session The change ensures otg is not in a A- state when checking for VBUS in peripheral mode. musb_start() where VBUS checking is in can be called in many situations. One example is in babble recovery routine, in which otg is transitioning from A-HOST to A-WAIT-BCON, but VBUS discharge takes time, so musb->is_active could be set to 1 due to this improper checking, then it causes musb_bus_suspend() failed which leads to warning log message flooding. Signed-off-by: Bin Liu Signed-off-by: Felipe Balbi --- drivers/usb/musb/musb_core.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/musb/musb_core.c b/drivers/usb/musb/musb_core.c index 514a6cdaeff6..d105c6dcbbec 100644 --- a/drivers/usb/musb/musb_core.c +++ b/drivers/usb/musb/musb_core.c @@ -1051,6 +1051,7 @@ void musb_start(struct musb *musb) * (c) peripheral initiates, using SRP */ if (musb->port_mode != MUSB_PORT_MODE_HOST && + musb->xceiv->otg->state != OTG_STATE_A_WAIT_BCON && (devctl & MUSB_DEVCTL_VBUS) == MUSB_DEVCTL_VBUS) { musb->is_active = 1; } else { From 88ccdbd5fc59ce8c69261a20746ca2533be7e358 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 10 Sep 2015 10:37:39 -0700 Subject: [PATCH 085/647] usb: phy: fix phy-qcom-8x16-usb build Fix build errors that happen when USB_QCOM_8X16_PHY=y and EXTCON=m: drivers/built-in.o: In function `phy_8x16_init': phy-qcom-8x16-usb.c:(.text+0x86ef4): undefined reference to `extcon_get_cable_state' drivers/built-in.o: In function `phy_8x16_probe': phy-qcom-8x16-usb.c:(.text+0x870bf): undefined reference to `extcon_get_edev_by_phandle' phy-qcom-8x16-usb.c:(.text+0x87133): undefined reference to `extcon_register_interest' phy-qcom-8x16-usb.c:(.text+0x87151): undefined reference to `extcon_unregister_interest' drivers/built-in.o: In function `phy_8x16_remove': phy-qcom-8x16-usb.c:(.text+0x872ec): undefined reference to `extcon_unregister_interest' Signed-off-by: Randy Dunlap Cc: Felipe Balbi Signed-off-by: Felipe Balbi --- drivers/usb/phy/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/phy/Kconfig b/drivers/usb/phy/Kconfig index 7d3beee2a587..173132416170 100644 --- a/drivers/usb/phy/Kconfig +++ b/drivers/usb/phy/Kconfig @@ -155,7 +155,7 @@ config USB_MSM_OTG config USB_QCOM_8X16_PHY tristate "Qualcomm APQ8016/MSM8916 on-chip USB PHY controller support" depends on ARCH_QCOM || COMPILE_TEST - depends on RESET_CONTROLLER + depends on RESET_CONTROLLER && EXTCON select USB_PHY select USB_ULPI_VIEWPORT help From 6527cc27761a124de8c08e6488ce17fdcc74dbba Mon Sep 17 00:00:00 2001 From: Alexey Khoroshilov Date: Sun, 6 Sep 2015 01:11:51 +0300 Subject: [PATCH 086/647] usb: gadget: amd5536udc: fix error handling in udc_pci_probe() If a failure happens early in udc_pci_probe(), error handling code just kfree(dev) and returns. The patch adds proper resource deallocations in udc_pci_probe() itself, since udc_pci_remove() is not suitabe to be called so early in initialization process. By the way, iounmap(dev->regs) is replaced by iounmap(dev->virt_addr) in udc_pci_remove() for clarity. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov Signed-off-by: Felipe Balbi --- drivers/usb/gadget/udc/amd5536udc.c | 43 ++++++++++++++--------------- 1 file changed, 20 insertions(+), 23 deletions(-) diff --git a/drivers/usb/gadget/udc/amd5536udc.c b/drivers/usb/gadget/udc/amd5536udc.c index fdacddb18c00..175ca93fe5e2 100644 --- a/drivers/usb/gadget/udc/amd5536udc.c +++ b/drivers/usb/gadget/udc/amd5536udc.c @@ -3138,8 +3138,8 @@ static void udc_pci_remove(struct pci_dev *pdev) writel(AMD_BIT(UDC_DEVCFG_SOFTRESET), &dev->regs->cfg); if (dev->irq_registered) free_irq(pdev->irq, dev); - if (dev->regs) - iounmap(dev->regs); + if (dev->virt_addr) + iounmap(dev->virt_addr); if (dev->mem_region) release_mem_region(pci_resource_start(pdev, 0), pci_resource_len(pdev, 0)); @@ -3226,17 +3226,13 @@ static int udc_pci_probe( /* init */ dev = kzalloc(sizeof(struct udc), GFP_KERNEL); - if (!dev) { - retval = -ENOMEM; - goto finished; - } + if (!dev) + return -ENOMEM; /* pci setup */ if (pci_enable_device(pdev) < 0) { - kfree(dev); - dev = NULL; retval = -ENODEV; - goto finished; + goto err_pcidev; } dev->active = 1; @@ -3246,28 +3242,22 @@ static int udc_pci_probe( if (!request_mem_region(resource, len, name)) { dev_dbg(&pdev->dev, "pci device used already\n"); - kfree(dev); - dev = NULL; retval = -EBUSY; - goto finished; + goto err_memreg; } dev->mem_region = 1; dev->virt_addr = ioremap_nocache(resource, len); if (dev->virt_addr == NULL) { dev_dbg(&pdev->dev, "start address cannot be mapped\n"); - kfree(dev); - dev = NULL; retval = -EFAULT; - goto finished; + goto err_ioremap; } if (!pdev->irq) { dev_err(&pdev->dev, "irq not set\n"); - kfree(dev); - dev = NULL; retval = -ENODEV; - goto finished; + goto err_irq; } spin_lock_init(&dev->lock); @@ -3283,10 +3273,8 @@ static int udc_pci_probe( if (request_irq(pdev->irq, udc_irq, IRQF_SHARED, name, dev) != 0) { dev_dbg(&pdev->dev, "request_irq(%d) fail\n", pdev->irq); - kfree(dev); - dev = NULL; retval = -EBUSY; - goto finished; + goto err_irq; } dev->irq_registered = 1; @@ -3314,8 +3302,17 @@ static int udc_pci_probe( return 0; finished: - if (dev) - udc_pci_remove(pdev); + udc_pci_remove(pdev); + return retval; + +err_irq: + iounmap(dev->virt_addr); +err_ioremap: + release_mem_region(resource, len); +err_memreg: + pci_disable_device(pdev); +err_pcidev: + kfree(dev); return retval; } From 6fc6f4b87cb32fcd0bf5e5bce434301958a3b624 Mon Sep 17 00:00:00 2001 From: Pascal Huerst Date: Thu, 3 Sep 2015 10:50:58 +0200 Subject: [PATCH 087/647] usb: musb: Disable interrupts on suspend, enable them on resume In certain situations, an interrupt triggers on resume, before musb_start() has been called. This has been observed to cause enumeration issues after suspend/resume cycles with AM335x. Signed-off-by: Pascal Huerst Signed-off-by: Felipe Balbi --- drivers/usb/musb/musb_core.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/usb/musb/musb_core.c b/drivers/usb/musb/musb_core.c index d105c6dcbbec..4a518ff12310 100644 --- a/drivers/usb/musb/musb_core.c +++ b/drivers/usb/musb/musb_core.c @@ -2449,6 +2449,9 @@ static int musb_suspend(struct device *dev) struct musb *musb = dev_to_musb(dev); unsigned long flags; + musb_platform_disable(musb); + musb_generic_disable(musb); + spin_lock_irqsave(&musb->lock, flags); if (is_peripheral_active(musb)) { @@ -2502,6 +2505,9 @@ static int musb_resume(struct device *dev) pm_runtime_disable(dev); pm_runtime_set_active(dev); pm_runtime_enable(dev); + + musb_start(musb); + return 0; } From 523d5daf86e5e8b540bdc91e090197b5665f0519 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Thu, 27 Aug 2015 17:45:37 +0200 Subject: [PATCH 088/647] usb: musb: ux500: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: Felipe Balbi --- drivers/usb/musb/ux500.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/usb/musb/ux500.c b/drivers/usb/musb/ux500.c index 39168fe9b406..b2685e75a683 100644 --- a/drivers/usb/musb/ux500.c +++ b/drivers/usb/musb/ux500.c @@ -379,6 +379,8 @@ static const struct of_device_id ux500_match[] = { {} }; +MODULE_DEVICE_TABLE(of, ux500_match); + static struct platform_driver ux500_driver = { .probe = ux500_probe, .remove = ux500_remove, From 762982db33b23029e98c844611e2e8beeb75bc0d Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Thu, 13 Aug 2015 13:28:42 +0300 Subject: [PATCH 089/647] usb: phy: phy-generic: Fix reset behaviour on legacy boot The gpio-desc migration done in v4.0 caused a regression with legacy boots due to reversed reset logic. e.g. omap3-beagle USB host breaks on legacy boot. Request the reset GPIO with GPIOF_ACTIVE_LOW flag so that it matches the driver logic and pin behaviour. Fixes: e9f2cefb0cdc ("usb: phy: generic: migrate to gpio_desc") Cc: # 4.0+ Tested-by: Fabio Estevam Signed-off-by: Roger Quadros Signed-off-by: Felipe Balbi --- drivers/usb/phy/phy-generic.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/usb/phy/phy-generic.c b/drivers/usb/phy/phy-generic.c index ec6ecd03269c..5320cb8642cb 100644 --- a/drivers/usb/phy/phy-generic.c +++ b/drivers/usb/phy/phy-generic.c @@ -232,7 +232,8 @@ int usb_phy_gen_create_phy(struct device *dev, struct usb_phy_generic *nop, clk_rate = pdata->clk_rate; needs_vcc = pdata->needs_vcc; if (gpio_is_valid(pdata->gpio_reset)) { - err = devm_gpio_request_one(dev, pdata->gpio_reset, 0, + err = devm_gpio_request_one(dev, pdata->gpio_reset, + GPIOF_ACTIVE_LOW, dev_name(dev)); if (!err) nop->gpiod_reset = From 586b286b110e94eb31840ac5afc0c24e0881fe34 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Wed, 9 Sep 2015 21:34:51 -0400 Subject: [PATCH 090/647] dm crypt: constrain crypt device's max_segment_size to PAGE_SIZE Setting the dm-crypt device's max_segment_size to PAGE_SIZE is an unfortunate constraint that is required to avoid the potential for exceeding dm-crypt's underlying device's max_segments limits -- due to crypt_alloc_buffer() possibly allocating pages for the encryption bio that are not as physically contiguous as the original bio. It is interesting to note that this problem was already fixed back in 2007 via commit 91e106259 ("dm crypt: use bio_add_page"). But Linux 4.0 commit cf2f1abfb ("dm crypt: don't allocate pages for a partial request") regressed dm-crypt back to _not_ using bio_add_page(). But given dm-crypt's cpu parallelization changes all depend on commit cf2f1abfb's abandoning of the more complex io fragments processing that dm-crypt previously had we cannot easily go back to using bio_add_page(). So all said the cleanest way to resolve this issue is to fix dm-crypt to properly constrain the original bios entering dm-crypt so the encryption bios that dm-crypt generates from the original bios are always compatible with the underlying device's max_segments queue limits. It should be noted that technically Linux 4.3 does _not_ need this fix because of the block core's new late bio-splitting capability. But, it is reasoned, there is little to be gained by having the block core split the encrypted bio that is composed of PAGE_SIZE segments. That said, in the future we may revert this change. Fixes: cf2f1abfb ("dm crypt: don't allocate pages for a partial request") Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=104421 Suggested-by: Jeff Moyer Signed-off-by: Mike Snitzer Cc: stable@vger.kernel.org # 4.0+ --- drivers/md/dm-crypt.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index d60c88df5234..4b3b6f8aff0c 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -968,7 +968,8 @@ static void crypt_free_buffer_pages(struct crypt_config *cc, struct bio *clone); /* * Generate a new unfragmented bio with the given size - * This should never violate the device limitations + * This should never violate the device limitations (but only because + * max_segment_size is being constrained to PAGE_SIZE). * * This function may be called concurrently. If we allocate from the mempool * concurrently, there is a possibility of deadlock. For example, if we have @@ -2045,9 +2046,20 @@ static int crypt_iterate_devices(struct dm_target *ti, return fn(ti, cc->dev, cc->start, ti->len, data); } +static void crypt_io_hints(struct dm_target *ti, struct queue_limits *limits) +{ + /* + * Unfortunate constraint that is required to avoid the potential + * for exceeding underlying device's max_segments limits -- due to + * crypt_alloc_buffer() possibly allocating pages for the encryption + * bio that are not as physically contiguous as the original bio. + */ + limits->max_segment_size = PAGE_SIZE; +} + static struct target_type crypt_target = { .name = "crypt", - .version = {1, 14, 0}, + .version = {1, 14, 1}, .module = THIS_MODULE, .ctr = crypt_ctr, .dtr = crypt_dtr, @@ -2058,6 +2070,7 @@ static struct target_type crypt_target = { .resume = crypt_resume, .message = crypt_message, .iterate_devices = crypt_iterate_devices, + .io_hints = crypt_io_hints, }; static int __init dm_crypt_init(void) From 63cdbc06b357dcb3a7104a421ee4a4550d7fadfd Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 14 Sep 2015 17:06:27 +0200 Subject: [PATCH 091/647] netfilter: bridge: fix routing of bridge frames with call-iptables=1 We can't re-use the physoutdev storage area. 1. When using NFQUEUE in PREROUTING, we attempt to bump a bogus refcnt since nf_bridge->physoutdev is garbage (ipv4/ipv6 address) 2. for same reason, we crash in physdev match in FORWARD or later if skb is routed instead of bridged. This increases nf_bridge_info to 40 bytes, but we have no other choice. Fixes: 72b1e5e4cac7 ("netfilter: bridge: reduce nf_bridge_info to 32 bytes again") Reported-by: Sander Eikelenboom Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/skbuff.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 2738d355cdf9..9987af080fa0 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -179,6 +179,9 @@ struct nf_bridge_info { u8 bridged_dnat:1; __u16 frag_max_size; struct net_device *physindev; + + /* always valid & non-NULL from FORWARD on, for physdev match */ + struct net_device *physoutdev; union { /* prerouting: detect dnat in orig/reply direction */ __be32 ipv4_daddr; @@ -189,9 +192,6 @@ struct nf_bridge_info { * skb is out in neigh layer. */ char neigh_header[8]; - - /* always valid & non-NULL from FORWARD on, for physdev match */ - struct net_device *physoutdev; }; }; #endif From ba378ca9c04a5fc1b2cf0f0274a9d02eb3d1bad9 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 14 Sep 2015 18:04:09 +0200 Subject: [PATCH 092/647] netfilter: nft_compat: skip family comparison in case of NFPROTO_UNSPEC Fix lookup of existing match/target structures in the corresponding list by skipping the family check if NFPROTO_UNSPEC is used. This is resulting in the allocation and insertion of one match/target structure for each use of them. So this not only bloats memory consumption but also severely affects the time to reload the ruleset from the iptables-compat utility. After this patch, iptables-compat-restore and iptables-compat take almost the same time to reload large rulesets. Fixes: 0ca743a55991 ("netfilter: nf_tables: add compatibility layer for x_tables") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_compat.c | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c index 66def315eb56..9c8fab00164b 100644 --- a/net/netfilter/nft_compat.c +++ b/net/netfilter/nft_compat.c @@ -619,6 +619,13 @@ struct nft_xt { static struct nft_expr_type nft_match_type; +static bool nft_match_cmp(const struct xt_match *match, + const char *name, u32 rev, u32 family) +{ + return strcmp(match->name, name) == 0 && match->revision == rev && + (match->family == NFPROTO_UNSPEC || match->family == family); +} + static const struct nft_expr_ops * nft_match_select_ops(const struct nft_ctx *ctx, const struct nlattr * const tb[]) @@ -626,7 +633,7 @@ nft_match_select_ops(const struct nft_ctx *ctx, struct nft_xt *nft_match; struct xt_match *match; char *mt_name; - __u32 rev, family; + u32 rev, family; if (tb[NFTA_MATCH_NAME] == NULL || tb[NFTA_MATCH_REV] == NULL || @@ -641,8 +648,7 @@ nft_match_select_ops(const struct nft_ctx *ctx, list_for_each_entry(nft_match, &nft_match_list, head) { struct xt_match *match = nft_match->ops.data; - if (strcmp(match->name, mt_name) == 0 && - match->revision == rev && match->family == family) { + if (nft_match_cmp(match, mt_name, rev, family)) { if (!try_module_get(match->me)) return ERR_PTR(-ENOENT); @@ -693,6 +699,13 @@ static LIST_HEAD(nft_target_list); static struct nft_expr_type nft_target_type; +static bool nft_target_cmp(const struct xt_target *tg, + const char *name, u32 rev, u32 family) +{ + return strcmp(tg->name, name) == 0 && tg->revision == rev && + (tg->family == NFPROTO_UNSPEC || tg->family == family); +} + static const struct nft_expr_ops * nft_target_select_ops(const struct nft_ctx *ctx, const struct nlattr * const tb[]) @@ -700,7 +713,7 @@ nft_target_select_ops(const struct nft_ctx *ctx, struct nft_xt *nft_target; struct xt_target *target; char *tg_name; - __u32 rev, family; + u32 rev, family; if (tb[NFTA_TARGET_NAME] == NULL || tb[NFTA_TARGET_REV] == NULL || @@ -715,8 +728,7 @@ nft_target_select_ops(const struct nft_ctx *ctx, list_for_each_entry(nft_target, &nft_target_list, head) { struct xt_target *target = nft_target->ops.data; - if (strcmp(target->name, tg_name) == 0 && - target->revision == rev && target->family == family) { + if (nft_target_cmp(target, tg_name, rev, family)) { if (!try_module_get(target->me)) return ERR_PTR(-ENOENT); From 723831927e8813b5b336d383174f686ad708bf10 Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Mon, 14 Sep 2015 16:06:48 +0300 Subject: [PATCH 093/647] ASoC: davinci-mcasp: Revise the FIFO threshold calculation The FIFO threshold for McASP should be <=[tx/rx]numevt so the initial value for the refining should meet this requirement as well. Signed-off-by: Peter Ujfalusi Signed-off-by: Mark Brown --- sound/soc/davinci/davinci-mcasp.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/sound/soc/davinci/davinci-mcasp.c b/sound/soc/davinci/davinci-mcasp.c index add6bb99661d..1260b315a96c 100644 --- a/sound/soc/davinci/davinci-mcasp.c +++ b/sound/soc/davinci/davinci-mcasp.c @@ -663,7 +663,7 @@ static int mcasp_common_hw_param(struct davinci_mcasp *mcasp, int stream, u8 rx_ser = 0; u8 slots = mcasp->tdm_slots; u8 max_active_serializers = (channels + slots - 1) / slots; - int active_serializers, numevt, n; + int active_serializers, numevt; u32 reg; /* Default configuration */ if (mcasp->version < MCASP_VERSION_3) @@ -745,9 +745,8 @@ static int mcasp_common_hw_param(struct davinci_mcasp *mcasp, int stream, * The number of words for numevt need to be in steps of active * serializers. */ - n = numevt % active_serializers; - if (n) - numevt += (active_serializers - n); + numevt = (numevt / active_serializers) * active_serializers; + while (period_words % numevt && numevt > 0) numevt -= active_serializers; if (numevt <= 0) From e4fba9b5be12d577d2e2c19fdca6b0744c3f271e Mon Sep 17 00:00:00 2001 From: Koro Chen Date: Mon, 14 Sep 2015 14:51:17 +0800 Subject: [PATCH 094/647] ASoC: mediatek: Increase periods_min in capture In capture, there is chance that hw_ptr reported at IRQ is a little smaller than period_size due to internal AFE buffer. In the case of ping-pong buffer: |xxxxxxxxxxxxxxxxxxxxxxxxxxxx--|-----------------------------| hw_ptr < period_size This available buffer will not be read since its size is smaller than avail_min (which is period_size by default), and read thread continues to sleep. If the next hw_ptr is just a little larger than buffer_size, overrun occurs. One more period can hold the possible unread buffer. Signed-off-by: Koro Chen Signed-off-by: Mark Brown --- sound/soc/mediatek/mtk-afe-pcm.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/sound/soc/mediatek/mtk-afe-pcm.c b/sound/soc/mediatek/mtk-afe-pcm.c index d190fe017559..f5baf3c38863 100644 --- a/sound/soc/mediatek/mtk-afe-pcm.c +++ b/sound/soc/mediatek/mtk-afe-pcm.c @@ -549,6 +549,23 @@ static int mtk_afe_dais_startup(struct snd_pcm_substream *substream, memif->substream = substream; snd_soc_set_runtime_hwparams(substream, &mtk_afe_hardware); + + /* + * Capture cannot use ping-pong buffer since hw_ptr at IRQ may be + * smaller than period_size due to AFE's internal buffer. + * This easily leads to overrun when avail_min is period_size. + * One more period can hold the possible unread buffer. + */ + if (substream->stream == SNDRV_PCM_STREAM_CAPTURE) { + ret = snd_pcm_hw_constraint_minmax(runtime, + SNDRV_PCM_HW_PARAM_PERIODS, + 3, + mtk_afe_hardware.periods_max); + if (ret < 0) { + dev_err(afe->dev, "hw_constraint_minmax failed\n"); + return ret; + } + } ret = snd_pcm_hw_constraint_integer(runtime, SNDRV_PCM_HW_PARAM_PERIODS); if (ret < 0) From 42617869bf095c650e67aad4001cab4224e7fa98 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Tue, 1 Sep 2015 12:31:24 +0800 Subject: [PATCH 095/647] ASoC: SPEAr: Make SND_SPEAR_SOC select SND_SOC_GENERIC_DMAENGINE_PCM devm_snd_dmaengine_pcm_register() is guarded by CONFIG_SND_SOC_GENERIC_DMAENGINE_PCM. Signed-off-by: Axel Lin Signed-off-by: Mark Brown --- sound/soc/spear/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/spear/Kconfig b/sound/soc/spear/Kconfig index 0a53053495f3..4fb91412ebec 100644 --- a/sound/soc/spear/Kconfig +++ b/sound/soc/spear/Kconfig @@ -1,6 +1,6 @@ config SND_SPEAR_SOC tristate - select SND_DMAENGINE_PCM + select SND_SOC_GENERIC_DMAENGINE_PCM config SND_SPEAR_SPDIF_OUT tristate From 921e54680aefe52f28d9ce9485edb1bfef4b92a8 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Tue, 1 Sep 2015 14:50:15 +0800 Subject: [PATCH 096/647] ASoC: au1x: psc-i2s: Fix unused variable 'ret' warning Fix below build warning: sound/soc/au1x/psc-i2s.c: In function 'au1xpsc_i2s_drvprobe': sound/soc/au1x/psc-i2s.c:299:6: warning: unused variable 'ret' [-Wunused-variable] Reported-by: kbuild test robot Signed-off-by: Axel Lin Acked-by: Manuel Lauss Signed-off-by: Mark Brown --- sound/soc/au1x/psc-i2s.c | 1 - 1 file changed, 1 deletion(-) diff --git a/sound/soc/au1x/psc-i2s.c b/sound/soc/au1x/psc-i2s.c index 38e853add96e..0bf9d62b91a0 100644 --- a/sound/soc/au1x/psc-i2s.c +++ b/sound/soc/au1x/psc-i2s.c @@ -296,7 +296,6 @@ static int au1xpsc_i2s_drvprobe(struct platform_device *pdev) { struct resource *iores, *dmares; unsigned long sel; - int ret; struct au1xpsc_audio_data *wd; wd = devm_kzalloc(&pdev->dev, sizeof(struct au1xpsc_audio_data), From 295c3405a8bbd69ee8c8eb6580f30b0b8739b33a Mon Sep 17 00:00:00 2001 From: Jyri Sarha Date: Wed, 9 Sep 2015 21:27:42 +0300 Subject: [PATCH 097/647] ASoC: davinci-mcasp: Set .symmetric_rates = 1 in snd_soc_dai_driver The TX and RX direction share the same bit clock and frame sync, so the samplerate must be the same to both directions. Signed-off-by: Jyri Sarha Acked-by: Peter Ujfalusi Signed-off-by: Mark Brown --- sound/soc/davinci/davinci-mcasp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/davinci/davinci-mcasp.c b/sound/soc/davinci/davinci-mcasp.c index 1260b315a96c..2c5a2c5444ab 100644 --- a/sound/soc/davinci/davinci-mcasp.c +++ b/sound/soc/davinci/davinci-mcasp.c @@ -1298,6 +1298,7 @@ static struct snd_soc_dai_driver davinci_mcasp_dai[] = { .ops = &davinci_mcasp_dai_ops, .symmetric_samplebits = 1, + .symmetric_rates = 1, }, { .name = "davinci-mcasp.1", From 3a2fa775bd1d0579113666c1a2e37654a34018a0 Mon Sep 17 00:00:00 2001 From: Carl Frederik Werner Date: Wed, 2 Sep 2015 10:07:57 +0900 Subject: [PATCH 098/647] ARM: dts: omap3-beagle: make i2c3, ddc and tfp410 gpio work again Let's fix pinmux address of gpio 170 used by tfp410 powerdown-gpio. According to the OMAP35x Technical Reference Manual CONTROL_PADCONF_I2C3_SDA[15:0] 0x480021C4 mode0: i2c3_sda CONTROL_PADCONF_I2C3_SDA[31:16] 0x480021C4 mode4: gpio_170 the pinmux address of gpio 170 must be 0x480021C6. The former wrong address broke i2c3 (used by hdmi ddc), resulting in kernel message: omap_i2c 48060000.i2c: controller timed out Fixes: 8cecf52befd7 ("ARM: omap3-beagle.dts: add display information") Cc: stable@vger.kernel.org # v3.15+ Signed-off-by: Carl Frederik Werner Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/omap3-beagle.dts | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/omap3-beagle.dts b/arch/arm/boot/dts/omap3-beagle.dts index a5474113cd50..67659a0ed13e 100644 --- a/arch/arm/boot/dts/omap3-beagle.dts +++ b/arch/arm/boot/dts/omap3-beagle.dts @@ -202,7 +202,7 @@ tfp410_pins: pinmux_tfp410_pins { pinctrl-single,pins = < - 0x194 (PIN_OUTPUT | MUX_MODE4) /* hdq_sio.gpio_170 */ + 0x196 (PIN_OUTPUT | MUX_MODE4) /* hdq_sio.gpio_170 */ >; }; From 87ee15ec26000c6edc10e63c415bee1239e042c8 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Mon, 14 Sep 2015 07:07:28 -0700 Subject: [PATCH 099/647] ARM: dts: Fix dm814x control base to properly initialize Ethernet PHY Looks like I made a typo on the control base, all the 81xx SoCs have it at 0x48140000 base. We've just gotten away with the typo as the Ethernet phy was configured by the bootloader on my test system and we're not yet using the pinctrl. In addition to fixing the contol base, we need to also use the right Ethernet phy flags to initialize it. And we are still missing the PLL driver for dm814x and only relying on the divider and mux clocks. Fixes: f3d953ea3721 ("ARM: dts: Add minimal dm814x support") Cc: Matthijs van Duin Cc: Nicolas Chauvet Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/dm8148-evm.dts | 4 ++-- arch/arm/boot/dts/dm8148-t410.dts | 4 ++-- arch/arm/boot/dts/dm814x.dtsi | 8 ++++---- 3 files changed, 8 insertions(+), 8 deletions(-) diff --git a/arch/arm/boot/dts/dm8148-evm.dts b/arch/arm/boot/dts/dm8148-evm.dts index 92bacd3c8fab..109fd4711647 100644 --- a/arch/arm/boot/dts/dm8148-evm.dts +++ b/arch/arm/boot/dts/dm8148-evm.dts @@ -19,10 +19,10 @@ &cpsw_emac0 { phy_id = <&davinci_mdio>, <0>; - phy-mode = "mii"; + phy-mode = "rgmii"; }; &cpsw_emac1 { phy_id = <&davinci_mdio>, <1>; - phy-mode = "mii"; + phy-mode = "rgmii"; }; diff --git a/arch/arm/boot/dts/dm8148-t410.dts b/arch/arm/boot/dts/dm8148-t410.dts index 8c4bbc7573df..aecd7dfa2eda 100644 --- a/arch/arm/boot/dts/dm8148-t410.dts +++ b/arch/arm/boot/dts/dm8148-t410.dts @@ -19,10 +19,10 @@ &cpsw_emac0 { phy_id = <&davinci_mdio>, <0>; - phy-mode = "mii"; + phy-mode = "rgmii"; }; &cpsw_emac1 { phy_id = <&davinci_mdio>, <1>; - phy-mode = "mii"; + phy-mode = "rgmii"; }; diff --git a/arch/arm/boot/dts/dm814x.dtsi b/arch/arm/boot/dts/dm814x.dtsi index 972c9c9e885b..7988b42e5764 100644 --- a/arch/arm/boot/dts/dm814x.dtsi +++ b/arch/arm/boot/dts/dm814x.dtsi @@ -181,9 +181,9 @@ ti,hwmods = "timer3"; }; - control: control@160000 { + control: control@140000 { compatible = "ti,dm814-scm", "simple-bus"; - reg = <0x160000 0x16d000>; + reg = <0x140000 0x16d000>; #address-cells = <1>; #size-cells = <1>; ranges = <0 0x160000 0x16d000>; @@ -321,9 +321,9 @@ mac-address = [ 00 00 00 00 00 00 ]; }; - phy_sel: cpsw-phy-sel@0x48160650 { + phy_sel: cpsw-phy-sel@48140650 { compatible = "ti,am3352-cpsw-phy-sel"; - reg= <0x48160650 0x4>; + reg= <0x48140650 0x4>; reg-names = "gmii-sel"; }; }; From 423fe5f3c35ff0f374d5452eb2db93a28bf6b9e2 Mon Sep 17 00:00:00 2001 From: Kishon Vijay Abraham I Date: Wed, 29 Jul 2015 16:39:42 +0530 Subject: [PATCH 100/647] ARM: dts: am57xx-beagle-x15: Fix regulator populated in MMC1 dt node For beagle x15, both the vdd and io lines are connected to the same regulator (ldo1_reg). However vmmc_aux is populated to vdd_3v3. Remove it. Signed-off-by: Kishon Vijay Abraham I Acked-by: Nishanth Menon [tony@atomide.com: updated to apply] Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/am57xx-beagle-x15.dts | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/arm/boot/dts/am57xx-beagle-x15.dts b/arch/arm/boot/dts/am57xx-beagle-x15.dts index 3a05b94f59ed..1ad3474efcf7 100644 --- a/arch/arm/boot/dts/am57xx-beagle-x15.dts +++ b/arch/arm/boot/dts/am57xx-beagle-x15.dts @@ -579,7 +579,6 @@ pinctrl-0 = <&mmc1_pins_default>; vmmc-supply = <&ldo1_reg>; - vmmc_aux-supply = <&vdd_3v3>; bus-width = <4>; cd-gpios = <&gpio6 27 0>; /* gpio 219 */ }; From 259c0c04acde19d6286def1661ed22d0fb3b05b9 Mon Sep 17 00:00:00 2001 From: Teresa Remmet Date: Thu, 3 Sep 2015 14:00:06 +0200 Subject: [PATCH 101/647] ARM: dts: am335x-phycore-som: Fix mpu voltage Fix the mpu voltage as it is set too low for the silicon revision 2.1. Signed-off-by: Teresa Remmet Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/am335x-phycore-som.dtsi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/boot/dts/am335x-phycore-som.dtsi b/arch/arm/boot/dts/am335x-phycore-som.dtsi index 4d28fc3aac69..5dd084f3c81c 100644 --- a/arch/arm/boot/dts/am335x-phycore-som.dtsi +++ b/arch/arm/boot/dts/am335x-phycore-som.dtsi @@ -252,10 +252,10 @@ }; vdd1_reg: regulator@2 { - /* VDD_MPU voltage limits 0.95V - 1.26V with +/-4% tolerance */ + /* VDD_MPU voltage limits 0.95V - 1.325V with +/-4% tolerance */ regulator-name = "vdd_mpu"; regulator-min-microvolt = <912500>; - regulator-max-microvolt = <1312500>; + regulator-max-microvolt = <1378000>; regulator-boot-on; regulator-always-on; }; From c22c7f3e667bece46efe97780ab2df3af700aea0 Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Thu, 3 Sep 2015 14:24:00 -0500 Subject: [PATCH 102/647] ARM: dts: am57xx-beagle-x15: Add wakeup irq for mcp79410 With the support in the generic PM framework for wakeirq and capability added to the rtc-ds1307 driver to support this, we can now define the optional wakeup irq to allow the RTC to wakeup the system from low power modes as part of suspend. Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/am57xx-beagle-x15.dts | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/am57xx-beagle-x15.dts b/arch/arm/boot/dts/am57xx-beagle-x15.dts index 1ad3474efcf7..a4274abf1437 100644 --- a/arch/arm/boot/dts/am57xx-beagle-x15.dts +++ b/arch/arm/boot/dts/am57xx-beagle-x15.dts @@ -517,7 +517,8 @@ mcp_rtc: rtc@6f { compatible = "microchip,mcp7941x"; reg = <0x6f>; - interrupts = ; /* IRQ_SYS_1N */ + interrupts-extended = <&crossbar_mpu GIC_SPI 2 IRQ_TYPE_EDGE_RISING>, + <&dra7_pmx_core 0x424>; pinctrl-names = "default"; pinctrl-0 = <&mcp79410_pins_default>; From d34cf0d56658ff040c707313b4a71e86a767cbc2 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Mon, 7 Sep 2015 18:24:18 +0200 Subject: [PATCH 103/647] ARM: dts: omap3-igep: Move eth IRQ pinmux to IGEPv2 common dtsi Only the IGEPv2 boards have a LAN9221i chip connected to the GPMC so the pinmux configuration for the GPIO connected to the IRQ line of the LAN chip should not be defined in the IGEP common dtsi but in the one common to the IGEPv2 boards. While there, use the OMAP3_CORE1_IOPAD() macro for the padconf reg. Suggested-by: Ladislav Michl Signed-off-by: Javier Martinez Canillas Acked-by: Enric Balletbo i Serra Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/omap3-igep.dtsi | 6 ------ arch/arm/boot/dts/omap3-igep0020-common.dtsi | 6 ++++++ 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/arm/boot/dts/omap3-igep.dtsi b/arch/arm/boot/dts/omap3-igep.dtsi index d5e5cd449b16..2230e1c03320 100644 --- a/arch/arm/boot/dts/omap3-igep.dtsi +++ b/arch/arm/boot/dts/omap3-igep.dtsi @@ -78,12 +78,6 @@ >; }; - smsc9221_pins: pinmux_smsc9221_pins { - pinctrl-single,pins = < - 0x1a2 (PIN_INPUT | MUX_MODE4) /* mcspi1_cs2.gpio_176 */ - >; - }; - i2c1_pins: pinmux_i2c1_pins { pinctrl-single,pins = < 0x18a (PIN_INPUT | MUX_MODE0) /* i2c1_scl.i2c1_scl */ diff --git a/arch/arm/boot/dts/omap3-igep0020-common.dtsi b/arch/arm/boot/dts/omap3-igep0020-common.dtsi index e458c2185e3c..5ad688c57a00 100644 --- a/arch/arm/boot/dts/omap3-igep0020-common.dtsi +++ b/arch/arm/boot/dts/omap3-igep0020-common.dtsi @@ -156,6 +156,12 @@ OMAP3_CORE1_IOPAD(0x217a, PIN_INPUT | MUX_MODE0) /* uart2_rx.uart2_rx */ >; }; + + smsc9221_pins: pinmux_smsc9221_pins { + pinctrl-single,pins = < + OMAP3_CORE1_IOPAD(0x21d2, PIN_INPUT | MUX_MODE4) /* mcspi1_cs2.gpio_176 */ + >; + }; }; &omap3_pmx_core2 { From 6d785c963a660b0279c43a2b1cd9b1c89ce5514b Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Wed, 9 Sep 2015 16:18:11 -0500 Subject: [PATCH 104/647] ARM: OMAP4+: PM: erratum is used by OMAP5 and DRA7 as well OMAP5 and DRA7 reuse the same pm44xx_erratum variable so, enable the same, else PM features such as Suspend to ram is broken in a SoC only build configuration. Reported-by: Carlos Hernandez Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/pm.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-omap2/pm.h b/arch/arm/mach-omap2/pm.h index 425bfcd67db6..b668719b9b25 100644 --- a/arch/arm/mach-omap2/pm.h +++ b/arch/arm/mach-omap2/pm.h @@ -103,7 +103,8 @@ static inline void enable_omap3630_toggle_l2_on_restore(void) { } #define PM_OMAP4_ROM_SMP_BOOT_ERRATUM_GICD (1 << 0) #define PM_OMAP4_CPU_OSWR_DISABLE (1 << 1) -#if defined(CONFIG_PM) && defined(CONFIG_ARCH_OMAP4) +#if defined(CONFIG_PM) && (defined(CONFIG_ARCH_OMAP4) ||\ + defined(CONFIG_SOC_OMAP5) || defined(CONFIG_SOC_DRA7XX)) extern u16 pm44xx_errata; #define IS_PM44XX_ERRATUM(id) (pm44xx_errata & (id)) #else From a3b7470951ab49e5592d20c2bcfe5ee675221591 Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Wed, 9 Sep 2015 16:18:12 -0500 Subject: [PATCH 105/647] ARM: OMAP2+: board-generic: Remove stale of_irq macros When commit c4082d499fa2 ("ARM: omap2+: board-generic: clean up the irq data from board file") cleaned up the direct usage of gic_of_init and omap_intc_of_init, it failed to clean up the macros properly. Since these macros are no longer used, lets just remove them. Fixes: c4082d499fa2 ("ARM: omap2+: board-generic: clean up the irq data from board file") Reported-by: Carlos Hernandez Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/board-generic.c | 7 ------- 1 file changed, 7 deletions(-) diff --git a/arch/arm/mach-omap2/board-generic.c b/arch/arm/mach-omap2/board-generic.c index 24c9afc9e8a7..6133eaac685d 100644 --- a/arch/arm/mach-omap2/board-generic.c +++ b/arch/arm/mach-omap2/board-generic.c @@ -20,13 +20,6 @@ #include "common.h" -#if !(defined(CONFIG_ARCH_OMAP2) || defined(CONFIG_ARCH_OMAP3)) -#define intc_of_init NULL -#endif -#ifndef CONFIG_ARCH_OMAP4 -#define gic_of_init NULL -#endif - static const struct of_device_id omap_dt_match_table[] __initconst = { { .compatible = "simple-bus", }, { .compatible = "ti,omap-infra", }, From d2e104c63a81e64574a6af9806f2a5d5a48248d9 Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Wed, 9 Sep 2015 16:18:13 -0500 Subject: [PATCH 106/647] ARM: DRA7: Select missing options for SoC only build DRA7 does use OPP, uses OMAP interconnect and also does require SCU. These are missing in the SoC only build of DRA7 breaking various PM features in DRA7 only build. Reported-by: Carlos Hernandez Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/Kconfig | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/arm/mach-omap2/Kconfig b/arch/arm/mach-omap2/Kconfig index 07d2e100caab..403022a38607 100644 --- a/arch/arm/mach-omap2/Kconfig +++ b/arch/arm/mach-omap2/Kconfig @@ -70,10 +70,13 @@ config SOC_DRA7XX select ARCH_OMAP2PLUS select ARM_CPU_SUSPEND if PM select ARM_GIC + select HAVE_ARM_SCU if SMP select HAVE_ARM_ARCH_TIMER select IRQ_CROSSBAR select ARM_ERRATA_798181 if SMP + select OMAP_INTERCONNECT select OMAP_INTERCONNECT_BARRIER + select PM_OPP if PM config ARCH_OMAP2PLUS bool From d8f8004ec7048fb2faa6967ab51259b5603562fb Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Wed, 9 Sep 2015 16:18:14 -0500 Subject: [PATCH 107/647] ARM: OMAP5: Cleanup options for SoC only build OMAP5 SoC has Cortex-A15 which does not use TWD timer. It uses ARCH_TIMER instead, clean up unwanted configuration and enable OMAP_INTERCONNECT and OPP which is necessary for expected functionality on the SoC. Reported-by: Carlos Hernandez Reported-by: Felipe Balbi Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-omap2/Kconfig b/arch/arm/mach-omap2/Kconfig index 403022a38607..b3a0dff67e3f 100644 --- a/arch/arm/mach-omap2/Kconfig +++ b/arch/arm/mach-omap2/Kconfig @@ -44,10 +44,11 @@ config SOC_OMAP5 select ARM_CPU_SUSPEND if PM select ARM_GIC select HAVE_ARM_SCU if SMP - select HAVE_ARM_TWD if SMP select HAVE_ARM_ARCH_TIMER select ARM_ERRATA_798181 if SMP + select OMAP_INTERCONNECT select OMAP_INTERCONNECT_BARRIER + select PM_OPP if PM config SOC_AM33XX bool "TI AM33XX" From 737f146fe0dd1353f5e2c6153fedd81f48b5cd65 Mon Sep 17 00:00:00 2001 From: Kishon Vijay Abraham I Date: Fri, 4 Sep 2015 17:30:25 +0530 Subject: [PATCH 108/647] ARM: dts: Use ti,pbias compatible string for pbias Use platform specific compatible strings instead of the common "ti,pbias-omap" compatible string. Signed-off-by: Kishon Vijay Abraham I Acked-by: Tony Lindgren Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/dra7.dtsi | 2 +- arch/arm/boot/dts/omap2430.dtsi | 2 +- arch/arm/boot/dts/omap3.dtsi | 2 +- arch/arm/boot/dts/omap4.dtsi | 2 +- arch/arm/boot/dts/omap5.dtsi | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/arm/boot/dts/dra7.dtsi b/arch/arm/boot/dts/dra7.dtsi index 5d65db9ebc2b..d1bf0b7f0ee9 100644 --- a/arch/arm/boot/dts/dra7.dtsi +++ b/arch/arm/boot/dts/dra7.dtsi @@ -122,7 +122,7 @@ #size-cells = <1>; pbias_regulator: pbias_regulator { - compatible = "ti,pbias-omap"; + compatible = "ti,pbias-dra7", "ti,pbias-omap"; reg = <0xe00 0x4>; syscon = <&scm_conf>; pbias_mmc_reg: pbias_mmc_omap5 { diff --git a/arch/arm/boot/dts/omap2430.dtsi b/arch/arm/boot/dts/omap2430.dtsi index 2390f387c271..3961a6f170e1 100644 --- a/arch/arm/boot/dts/omap2430.dtsi +++ b/arch/arm/boot/dts/omap2430.dtsi @@ -63,7 +63,7 @@ }; pbias_regulator: pbias_regulator { - compatible = "ti,pbias-omap"; + compatible = "ti,pbias-omap2", "ti,pbias-omap"; reg = <0x230 0x4>; syscon = <&scm_conf>; pbias_mmc_reg: pbias_mmc_omap2430 { diff --git a/arch/arm/boot/dts/omap3.dtsi b/arch/arm/boot/dts/omap3.dtsi index 69a40cfc1f29..9af9ae190299 100644 --- a/arch/arm/boot/dts/omap3.dtsi +++ b/arch/arm/boot/dts/omap3.dtsi @@ -203,7 +203,7 @@ }; pbias_regulator: pbias_regulator { - compatible = "ti,pbias-omap"; + compatible = "ti,pbias-omap3", "ti,pbias-omap"; reg = <0x2b0 0x4>; syscon = <&scm_conf>; pbias_mmc_reg: pbias_mmc_omap2430 { diff --git a/arch/arm/boot/dts/omap4.dtsi b/arch/arm/boot/dts/omap4.dtsi index abc4473e6f8a..5aad7f37df8b 100644 --- a/arch/arm/boot/dts/omap4.dtsi +++ b/arch/arm/boot/dts/omap4.dtsi @@ -198,7 +198,7 @@ #size-cells = <1>; pbias_regulator: pbias_regulator { - compatible = "ti,pbias-omap"; + compatible = "ti,pbias-omap4", "ti,pbias-omap"; reg = <0x60 0x4>; syscon = <&omap4_padconf_global>; pbias_mmc_reg: pbias_mmc_omap4 { diff --git a/arch/arm/boot/dts/omap5.dtsi b/arch/arm/boot/dts/omap5.dtsi index 4205a8ac9ddb..8d5f9d26110e 100644 --- a/arch/arm/boot/dts/omap5.dtsi +++ b/arch/arm/boot/dts/omap5.dtsi @@ -187,7 +187,7 @@ #size-cells = <1>; pbias_regulator: pbias_regulator { - compatible = "ti,pbias-omap"; + compatible = "ti,pbias-omap5", "ti,pbias-omap"; reg = <0x60 0x4>; syscon = <&omap5_padconf_global>; pbias_mmc_reg: pbias_mmc_omap5 { From 3ab74b89c4d50e6d29d3ace2d400a8a3f1e6de3f Mon Sep 17 00:00:00 2001 From: Kishon Vijay Abraham I Date: Wed, 5 Aug 2015 19:13:49 +0530 Subject: [PATCH 109/647] ARM: omap2plus_defconfig: make PCF857x built-in One of the lines from PCF857x is connected to the vdd line of MMC1 in DRA74x and DRA72x EVMs and is modelled as a regulator. If PCF857x is not made as built-in, the regulator_get in omap_hsmmc fails making it difficult to use MMC1 as rootfs. Make PCF857x built-in. Signed-off-by: Kishon Vijay Abraham I Signed-off-by: Sekhar Nori Signed-off-by: Tony Lindgren --- arch/arm/configs/omap2plus_defconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig index 50c84e1876fc..1860f517421d 100644 --- a/arch/arm/configs/omap2plus_defconfig +++ b/arch/arm/configs/omap2plus_defconfig @@ -240,7 +240,7 @@ CONFIG_SSI_PROTOCOL=m CONFIG_PINCTRL_SINGLE=y CONFIG_DEBUG_GPIO=y CONFIG_GPIO_SYSFS=y -CONFIG_GPIO_PCF857X=m +CONFIG_GPIO_PCF857X=y CONFIG_GPIO_TWL4030=y CONFIG_GPIO_PALMAS=y CONFIG_W1=m From c263a5b8120523b736ed4641657075f2024e75b7 Mon Sep 17 00:00:00 2001 From: Vishal Mahaveer Date: Tue, 25 Aug 2015 13:57:49 -0500 Subject: [PATCH 110/647] ARM: dts: DRA7: fix a typo in ethernet Register address in name of the node is wrong Signed-off-by: Vishal Mahaveer Acked-by: Mugunthan V N Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/dra7.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/dra7.dtsi b/arch/arm/boot/dts/dra7.dtsi index d1bf0b7f0ee9..fed95a4547ee 100644 --- a/arch/arm/boot/dts/dra7.dtsi +++ b/arch/arm/boot/dts/dra7.dtsi @@ -1417,7 +1417,7 @@ ti,irqs-safe-map = <0>; }; - mac: ethernet@4a100000 { + mac: ethernet@48484000 { compatible = "ti,dra7-cpsw","ti,cpsw"; ti,hwmods = "gmac"; clocks = <&dpll_gmac_ck>, <&gmac_gmii_ref_clk_div>; From 60fdcb8863d9b4a8b6c6b367886fadb50d4c0b07 Mon Sep 17 00:00:00 2001 From: Nicolas Chauvet Date: Thu, 3 Sep 2015 17:34:40 +0200 Subject: [PATCH 111/647] ARM: dts: Fixup model name for HP t410 dts This fix the model name for the device. Whole string taken from the HP support center web page Signed-off-by: Nicolas Chauvet Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/dm8148-t410.dts | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/dm8148-t410.dts b/arch/arm/boot/dts/dm8148-t410.dts index aecd7dfa2eda..79838dd8dee7 100644 --- a/arch/arm/boot/dts/dm8148-t410.dts +++ b/arch/arm/boot/dts/dm8148-t410.dts @@ -8,7 +8,7 @@ #include "dm814x.dtsi" / { - model = "DM8148 EVM"; + model = "HP t410 Smart Zero Client"; compatible = "hp,t410", "ti,dm8148"; memory { From 005efedf2c7d0a270ffbe28d8997b03844f3e3e7 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 14 Sep 2015 09:09:31 +0100 Subject: [PATCH 112/647] Btrfs: fix read corruption of compressed and shared extents If a file has a range pointing to a compressed extent, followed by another range that points to the same compressed extent and a read operation attempts to read both ranges (either completely or part of them), the pages that correspond to the second range are incorrectly filled with zeroes. Consider the following example: File layout [0 - 8K] [8K - 24K] | | | | points to extent X, points to extent X, offset 4K, length of 8K offset 0, length 16K [extent X, compressed length = 4K uncompressed length = 16K] If a readpages() call spans the 2 ranges, a single bio to read the extent is submitted - extent_io.c:submit_extent_page() would only create a new bio to cover the second range pointing to the extent if the extent it points to had a different logical address than the extent associated with the first range. This has a consequence of the compressed read end io handler (compression.c:end_compressed_bio_read()) finish once the extent is decompressed into the pages covering the first range, leaving the remaining pages (belonging to the second range) filled with zeroes (done by compression.c:btrfs_clear_biovec_end()). So fix this by submitting the current bio whenever we find a range pointing to a compressed extent that was preceded by a range with a different extent map. This is the simplest solution for this corner case. Making the end io callback populate both ranges (or more, if we have multiple pointing to the same extent) is a much more complex solution since each bio is tightly coupled with a single extent map and the extent maps associated to the ranges pointing to the shared extent can have different offsets and lengths. The following test case for fstests triggers the issue: seq=`basename $0` seqres=$RESULT_DIR/$seq echo "QA output created by $seq" tmp=/tmp/$$ status=1 # failure is the default! trap "_cleanup; exit \$status" 0 1 2 3 15 _cleanup() { rm -f $tmp.* } # get standard environment, filters and checks . ./common/rc . ./common/filter # real QA test starts here _need_to_be_root _supported_fs btrfs _supported_os Linux _require_scratch _require_cloner rm -f $seqres.full test_clone_and_read_compressed_extent() { local mount_opts=$1 _scratch_mkfs >>$seqres.full 2>&1 _scratch_mount $mount_opts # Create a test file with a single extent that is compressed (the # data we write into it is highly compressible no matter which # compression algorithm is used, zlib or lzo). $XFS_IO_PROG -f -c "pwrite -S 0xaa 0K 4K" \ -c "pwrite -S 0xbb 4K 8K" \ -c "pwrite -S 0xcc 12K 4K" \ $SCRATCH_MNT/foo | _filter_xfs_io # Now clone our extent into an adjacent offset. $CLONER_PROG -s $((4 * 1024)) -d $((16 * 1024)) -l $((8 * 1024)) \ $SCRATCH_MNT/foo $SCRATCH_MNT/foo # Same as before but for this file we clone the extent into a lower # file offset. $XFS_IO_PROG -f -c "pwrite -S 0xaa 8K 4K" \ -c "pwrite -S 0xbb 12K 8K" \ -c "pwrite -S 0xcc 20K 4K" \ $SCRATCH_MNT/bar | _filter_xfs_io $CLONER_PROG -s $((12 * 1024)) -d 0 -l $((8 * 1024)) \ $SCRATCH_MNT/bar $SCRATCH_MNT/bar echo "File digests before unmounting filesystem:" md5sum $SCRATCH_MNT/foo | _filter_scratch md5sum $SCRATCH_MNT/bar | _filter_scratch # Evicting the inode or clearing the page cache before reading # again the file would also trigger the bug - reads were returning # all bytes in the range corresponding to the second reference to # the extent with a value of 0, but the correct data was persisted # (it was a bug exclusively in the read path). The issue happened # only if the same readpages() call targeted pages belonging to the # first and second ranges that point to the same compressed extent. _scratch_remount echo "File digests after mounting filesystem again:" # Must match the same digests we got before. md5sum $SCRATCH_MNT/foo | _filter_scratch md5sum $SCRATCH_MNT/bar | _filter_scratch } echo -e "\nTesting with zlib compression..." test_clone_and_read_compressed_extent "-o compress=zlib" _scratch_unmount echo -e "\nTesting with lzo compression..." test_clone_and_read_compressed_extent "-o compress=lzo" status=0 exit Cc: stable@vger.kernel.org Signed-off-by: Filipe Manana Reviewed-by: Qu Wenruo Reviewed-by: Liu Bo --- fs/btrfs/extent_io.c | 65 ++++++++++++++++++++++++++++++++++++++------ 1 file changed, 57 insertions(+), 8 deletions(-) diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index fa19f2f68c1b..11aa8f743b90 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2805,7 +2805,8 @@ static int submit_extent_page(int rw, struct extent_io_tree *tree, bio_end_io_t end_io_func, int mirror_num, unsigned long prev_bio_flags, - unsigned long bio_flags) + unsigned long bio_flags, + bool force_bio_submit) { int ret = 0; struct bio *bio; @@ -2823,6 +2824,7 @@ static int submit_extent_page(int rw, struct extent_io_tree *tree, contig = bio_end_sector(bio) == sector; if (prev_bio_flags != bio_flags || !contig || + force_bio_submit || merge_bio(rw, tree, page, offset, page_size, bio, bio_flags) || bio_add_page(bio, page, page_size, offset) < page_size) { ret = submit_one_bio(rw, bio, mirror_num, @@ -2922,7 +2924,8 @@ static int __do_readpage(struct extent_io_tree *tree, get_extent_t *get_extent, struct extent_map **em_cached, struct bio **bio, int mirror_num, - unsigned long *bio_flags, int rw) + unsigned long *bio_flags, int rw, + u64 *prev_em_start) { struct inode *inode = page->mapping->host; u64 start = page_offset(page); @@ -2970,6 +2973,7 @@ static int __do_readpage(struct extent_io_tree *tree, } while (cur <= end) { unsigned long pnr = (last_byte >> PAGE_CACHE_SHIFT) + 1; + bool force_bio_submit = false; if (cur >= last_byte) { char *userpage; @@ -3020,6 +3024,49 @@ static int __do_readpage(struct extent_io_tree *tree, block_start = em->block_start; if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) block_start = EXTENT_MAP_HOLE; + + /* + * If we have a file range that points to a compressed extent + * and it's followed by a consecutive file range that points to + * to the same compressed extent (possibly with a different + * offset and/or length, so it either points to the whole extent + * or only part of it), we must make sure we do not submit a + * single bio to populate the pages for the 2 ranges because + * this makes the compressed extent read zero out the pages + * belonging to the 2nd range. Imagine the following scenario: + * + * File layout + * [0 - 8K] [8K - 24K] + * | | + * | | + * points to extent X, points to extent X, + * offset 4K, length of 8K offset 0, length 16K + * + * [extent X, compressed length = 4K uncompressed length = 16K] + * + * If the bio to read the compressed extent covers both ranges, + * it will decompress extent X into the pages belonging to the + * first range and then it will stop, zeroing out the remaining + * pages that belong to the other range that points to extent X. + * So here we make sure we submit 2 bios, one for the first + * range and another one for the third range. Both will target + * the same physical extent from disk, but we can't currently + * make the compressed bio endio callback populate the pages + * for both ranges because each compressed bio is tightly + * coupled with a single extent map, and each range can have + * an extent map with a different offset value relative to the + * uncompressed data of our extent and different lengths. This + * is a corner case so we prioritize correctness over + * non-optimal behavior (submitting 2 bios for the same extent). + */ + if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags) && + prev_em_start && *prev_em_start != (u64)-1 && + *prev_em_start != em->orig_start) + force_bio_submit = true; + + if (prev_em_start) + *prev_em_start = em->orig_start; + free_extent_map(em); em = NULL; @@ -3069,7 +3116,8 @@ static int __do_readpage(struct extent_io_tree *tree, bdev, bio, pnr, end_bio_extent_readpage, mirror_num, *bio_flags, - this_bio_flag); + this_bio_flag, + force_bio_submit); if (!ret) { nr++; *bio_flags = this_bio_flag; @@ -3101,6 +3149,7 @@ static inline void __do_contiguous_readpages(struct extent_io_tree *tree, struct inode *inode; struct btrfs_ordered_extent *ordered; int index; + u64 prev_em_start = (u64)-1; inode = pages[0]->mapping->host; while (1) { @@ -3116,7 +3165,7 @@ static inline void __do_contiguous_readpages(struct extent_io_tree *tree, for (index = 0; index < nr_pages; index++) { __do_readpage(tree, pages[index], get_extent, em_cached, bio, - mirror_num, bio_flags, rw); + mirror_num, bio_flags, rw, &prev_em_start); page_cache_release(pages[index]); } } @@ -3184,7 +3233,7 @@ static int __extent_read_full_page(struct extent_io_tree *tree, } ret = __do_readpage(tree, page, get_extent, NULL, bio, mirror_num, - bio_flags, rw); + bio_flags, rw, NULL); return ret; } @@ -3210,7 +3259,7 @@ int extent_read_full_page_nolock(struct extent_io_tree *tree, struct page *page, int ret; ret = __do_readpage(tree, page, get_extent, NULL, &bio, mirror_num, - &bio_flags, READ); + &bio_flags, READ, NULL); if (bio) ret = submit_one_bio(READ, bio, mirror_num, bio_flags); return ret; @@ -3463,7 +3512,7 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode, sector, iosize, pg_offset, bdev, &epd->bio, max_nr, end_bio_extent_writepage, - 0, 0, 0); + 0, 0, 0, false); if (ret) SetPageError(page); } @@ -3765,7 +3814,7 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb, ret = submit_extent_page(rw, tree, wbc, p, offset >> 9, PAGE_CACHE_SIZE, 0, bdev, &epd->bio, -1, end_bio_extent_buffer_writepage, - 0, epd->bio_flags, bio_flags); + 0, epd->bio_flags, bio_flags, false); epd->bio_flags = bio_flags; if (ret) { set_btree_ioerr(p); From a30e577c96f59b1e1678ea5462432b09bf7d5cbc Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Fri, 11 Sep 2015 21:44:17 -0400 Subject: [PATCH 113/647] btrfs: skip waiting on ordered range for special files In btrfs_evict_inode, we properly truncate the page cache for evicted inodes but then we call btrfs_wait_ordered_range for every inode as well. It's the right thing to do for regular files but results in incorrect behavior for device inodes for block devices. filemap_fdatawrite_range gets called with inode->i_mapping which gets resolved to the block device inode before getting passed to wbc_attach_fdatawrite_inode and ultimately to inode_to_bdi. What happens next depends on whether there's an open file handle associated with the inode. If there is, we write to the block device, which is unexpected behavior. If there isn't, we through normally and inode->i_data is used. We can also end up racing against open/close which can result in crashes when i_mapping points to a block device inode that has been closed. Since there can't be any page cache associated with special file inodes, it's safe to skip the btrfs_wait_ordered_range call entirely and avoid the problem. Cc: Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=100911 Tested-by: Christoph Biedl Signed-off-by: Jeff Mahoney Reviewed-by: Filipe Manana --- fs/btrfs/inode.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 37dd8d0f1fb3..3d983dea57af 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -5080,7 +5080,8 @@ void btrfs_evict_inode(struct inode *inode) goto no_delete; } /* do we really want it for ->i_nlink > 0 and zero btrfs_root_refs? */ - btrfs_wait_ordered_range(inode, 0, (u64)-1); + if (!special_file(inode->i_mode)) + btrfs_wait_ordered_range(inode, 0, (u64)-1); btrfs_free_io_failure_record(inode, 0, (u64)-1); From e30f3963f2c3c03119d71f1804bdb94ccd33580b Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Mon, 14 Sep 2015 01:24:41 -0700 Subject: [PATCH 114/647] drm/ttm: Fix memory space allocation v2 In the event that TTM doesn't find a compatible memory type for the driver's first placement choice (placement without eviction), TTM returns -EINVAL without trying the driver's second choice. This causes problems on vmwgfx when VRAM is disabled before first modeset and during VT switches when fbdev is not enabled. Fix this by also trying the driver's second choice before returning -EINVAL. v2: Also check that man->use_type is true for the driver's second choice. Fixes a bug where disallowed memory types could be used. Signed-off-by: Thomas Hellstrom Reviewed-by: Brian Paul Reviewed-by: Sinclair Yeh --- drivers/gpu/drm/ttm/ttm_bo.c | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index 8d9b7de25613..745e996d2dbc 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -882,6 +882,8 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, if (ret) return ret; man = &bdev->man[mem_type]; + if (!man->has_type || !man->use_type) + continue; type_ok = ttm_bo_mt_compatible(man, mem_type, place, &cur_flags); @@ -889,6 +891,7 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, if (!type_ok) continue; + type_found = true; cur_flags = ttm_bo_select_caching(man, bo->mem.placement, cur_flags); /* @@ -901,12 +904,10 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, if (mem_type == TTM_PL_SYSTEM) break; - if (man->has_type && man->use_type) { - type_found = true; - ret = (*man->func->get_node)(man, bo, place, mem); - if (unlikely(ret)) - return ret; - } + ret = (*man->func->get_node)(man, bo, place, mem); + if (unlikely(ret)) + return ret; + if (mem->mm_node) break; } @@ -917,9 +918,6 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, return 0; } - if (!type_found) - return -EINVAL; - for (i = 0; i < placement->num_busy_placement; ++i) { const struct ttm_place *place = &placement->busy_placement[i]; @@ -927,11 +925,12 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, if (ret) return ret; man = &bdev->man[mem_type]; - if (!man->has_type) + if (!man->has_type || !man->use_type) continue; if (!ttm_bo_mt_compatible(man, mem_type, place, &cur_flags)) continue; + type_found = true; cur_flags = ttm_bo_select_caching(man, bo->mem.placement, cur_flags); /* @@ -957,8 +956,13 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, if (ret == -ERESTARTSYS) has_erestartsys = true; } - ret = (has_erestartsys) ? -ERESTARTSYS : -ENOMEM; - return ret; + + if (!type_found) { + printk(KERN_ERR TTM_PFX "No compatible memory type found.\n"); + return -EINVAL; + } + + return (has_erestartsys) ? -ERESTARTSYS : -ENOMEM; } EXPORT_SYMBOL(ttm_bo_mem_space); From 56ffa1d154c7e12af16273f0cdc42690dd05caf5 Mon Sep 17 00:00:00 2001 From: Peter Chen Date: Mon, 24 Aug 2015 14:10:07 +0800 Subject: [PATCH 115/647] usb: chipidea: udc: using the correct stall implementation According to spec, there are functional and protocol stalls. For functional stall, it is for bulk and interrupt endpoints, below are cases for it: - Host sends SET_FEATURE request for Set-Halt, the udc driver needs to set stall, and return true unconditionally. - The gadget driver may call usb_ep_set_halt to stall certain endpoints, if there is a transfer in pending, the udc driver should not set stall, and return -EAGAIN accordingly. These two kinds of stall need to be cleared by host using CLEAR_FEATURE request (Clear-Halt). For protocol stall, it is for control endpoint, this stall will be set if the control request has failed. This stall will be cleared by next setup request (hardware will do it). It fixed usbtest (drivers/usb/misc/usbtest.c) Test 13 "set/clear halt" test failure, meanwhile, this change has been verified by USB2 CV Compliance Test and MSC Tests. Cc: #3.10+ Cc: Alan Stern Cc: Felipe Balbi Signed-off-by: Peter Chen --- drivers/usb/chipidea/udc.c | 84 ++++++++++++++++++++------------------ 1 file changed, 44 insertions(+), 40 deletions(-) diff --git a/drivers/usb/chipidea/udc.c b/drivers/usb/chipidea/udc.c index a637da25dda0..8223fe73ea85 100644 --- a/drivers/usb/chipidea/udc.c +++ b/drivers/usb/chipidea/udc.c @@ -656,6 +656,44 @@ __acquires(hwep->lock) return 0; } +static int _ep_set_halt(struct usb_ep *ep, int value, bool check_transfer) +{ + struct ci_hw_ep *hwep = container_of(ep, struct ci_hw_ep, ep); + int direction, retval = 0; + unsigned long flags; + + if (ep == NULL || hwep->ep.desc == NULL) + return -EINVAL; + + if (usb_endpoint_xfer_isoc(hwep->ep.desc)) + return -EOPNOTSUPP; + + spin_lock_irqsave(hwep->lock, flags); + + if (value && hwep->dir == TX && check_transfer && + !list_empty(&hwep->qh.queue) && + !usb_endpoint_xfer_control(hwep->ep.desc)) { + spin_unlock_irqrestore(hwep->lock, flags); + return -EAGAIN; + } + + direction = hwep->dir; + do { + retval |= hw_ep_set_halt(hwep->ci, hwep->num, hwep->dir, value); + + if (!value) + hwep->wedge = 0; + + if (hwep->type == USB_ENDPOINT_XFER_CONTROL) + hwep->dir = (hwep->dir == TX) ? RX : TX; + + } while (hwep->dir != direction); + + spin_unlock_irqrestore(hwep->lock, flags); + return retval; +} + + /** * _gadget_stop_activity: stops all USB activity, flushes & disables all endpts * @gadget: gadget @@ -1051,7 +1089,7 @@ __acquires(ci->lock) num += ci->hw_ep_max / 2; spin_unlock(&ci->lock); - err = usb_ep_set_halt(&ci->ci_hw_ep[num].ep); + err = _ep_set_halt(&ci->ci_hw_ep[num].ep, 1, false); spin_lock(&ci->lock); if (!err) isr_setup_status_phase(ci); @@ -1117,8 +1155,8 @@ delegate: if (err < 0) { spin_unlock(&ci->lock); - if (usb_ep_set_halt(&hwep->ep)) - dev_err(ci->dev, "error: ep_set_halt\n"); + if (_ep_set_halt(&hwep->ep, 1, false)) + dev_err(ci->dev, "error: _ep_set_halt\n"); spin_lock(&ci->lock); } } @@ -1149,9 +1187,9 @@ __acquires(ci->lock) err = isr_setup_status_phase(ci); if (err < 0) { spin_unlock(&ci->lock); - if (usb_ep_set_halt(&hwep->ep)) + if (_ep_set_halt(&hwep->ep, 1, false)) dev_err(ci->dev, - "error: ep_set_halt\n"); + "error: _ep_set_halt\n"); spin_lock(&ci->lock); } } @@ -1397,41 +1435,7 @@ static int ep_dequeue(struct usb_ep *ep, struct usb_request *req) */ static int ep_set_halt(struct usb_ep *ep, int value) { - struct ci_hw_ep *hwep = container_of(ep, struct ci_hw_ep, ep); - int direction, retval = 0; - unsigned long flags; - - if (ep == NULL || hwep->ep.desc == NULL) - return -EINVAL; - - if (usb_endpoint_xfer_isoc(hwep->ep.desc)) - return -EOPNOTSUPP; - - spin_lock_irqsave(hwep->lock, flags); - -#ifndef STALL_IN - /* g_file_storage MS compliant but g_zero fails chapter 9 compliance */ - if (value && hwep->type == USB_ENDPOINT_XFER_BULK && hwep->dir == TX && - !list_empty(&hwep->qh.queue)) { - spin_unlock_irqrestore(hwep->lock, flags); - return -EAGAIN; - } -#endif - - direction = hwep->dir; - do { - retval |= hw_ep_set_halt(hwep->ci, hwep->num, hwep->dir, value); - - if (!value) - hwep->wedge = 0; - - if (hwep->type == USB_ENDPOINT_XFER_CONTROL) - hwep->dir = (hwep->dir == TX) ? RX : TX; - - } while (hwep->dir != direction); - - spin_unlock_irqrestore(hwep->lock, flags); - return retval; + return _ep_set_halt(ep, value, true); } /** From 84bc70f94d81f1d3107dbcdafc1b193169e82131 Mon Sep 17 00:00:00 2001 From: Nathan Sullivan Date: Mon, 31 Aug 2015 09:49:51 -0500 Subject: [PATCH 116/647] usb: chipidea: add xilinx zynq platform data Due to having hardware tx buffers less than 512 bytes in size, streaming must be enabled on the Zynq for the udc to work at all. Add platform data specific to the Zynq udc, which does not set the CI_HDRC_DISABLE_STREAMING flag. Based on a patch by the same name from the Xilinx vendor tree. Signed-off-by: Nathan Sullivan Signed-off-by: Peter Chen --- drivers/usb/chipidea/ci_hdrc_usb2.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) diff --git a/drivers/usb/chipidea/ci_hdrc_usb2.c b/drivers/usb/chipidea/ci_hdrc_usb2.c index 9eae1a16cef9..4456d2cf80ff 100644 --- a/drivers/usb/chipidea/ci_hdrc_usb2.c +++ b/drivers/usb/chipidea/ci_hdrc_usb2.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -30,18 +31,36 @@ static const struct ci_hdrc_platform_data ci_default_pdata = { .flags = CI_HDRC_DISABLE_STREAMING, }; +static struct ci_hdrc_platform_data ci_zynq_pdata = { + .capoffset = DEF_CAPOFFSET, +}; + +static const struct of_device_id ci_hdrc_usb2_of_match[] = { + { .compatible = "chipidea,usb2"}, + { .compatible = "xlnx,zynq-usb-2.20a", .data = &ci_zynq_pdata}, + { } +}; +MODULE_DEVICE_TABLE(of, ci_hdrc_usb2_of_match); + static int ci_hdrc_usb2_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct ci_hdrc_usb2_priv *priv; struct ci_hdrc_platform_data *ci_pdata = dev_get_platdata(dev); int ret; + const struct of_device_id *match; if (!ci_pdata) { ci_pdata = devm_kmalloc(dev, sizeof(*ci_pdata), GFP_KERNEL); *ci_pdata = ci_default_pdata; /* struct copy */ } + match = of_match_device(ci_hdrc_usb2_of_match, &pdev->dev); + if (match && match->data) { + /* struct copy */ + *ci_pdata = *(struct ci_hdrc_platform_data *)match->data; + } + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); if (!priv) return -ENOMEM; @@ -96,12 +115,6 @@ static int ci_hdrc_usb2_remove(struct platform_device *pdev) return 0; } -static const struct of_device_id ci_hdrc_usb2_of_match[] = { - { .compatible = "chipidea,usb2" }, - { } -}; -MODULE_DEVICE_TABLE(of, ci_hdrc_usb2_of_match); - static struct platform_driver ci_hdrc_usb2_driver = { .probe = ci_hdrc_usb2_probe, .remove = ci_hdrc_usb2_remove, From c1ceb5fff01c0357de0386f87a620a4636ca68d1 Mon Sep 17 00:00:00 2001 From: Nathan Sullivan Date: Mon, 31 Aug 2015 09:49:52 -0500 Subject: [PATCH 117/647] Documentation: bindings: add doc for zynq USB Document the binding for the zynq specific chipidea UDC binding. Signed-off-by: Nathan Sullivan Signed-off-by: Peter Chen --- Documentation/devicetree/bindings/usb/ci-hdrc-usb2.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/usb/ci-hdrc-usb2.txt b/Documentation/devicetree/bindings/usb/ci-hdrc-usb2.txt index d71ef07bca5d..a057b75ba4b5 100644 --- a/Documentation/devicetree/bindings/usb/ci-hdrc-usb2.txt +++ b/Documentation/devicetree/bindings/usb/ci-hdrc-usb2.txt @@ -6,6 +6,7 @@ Required properties: "lsi,zevio-usb" "qcom,ci-hdrc" "chipidea,usb2" + "xlnx,zynq-usb-2.20a" - reg: base address and length of the registers - interrupts: interrupt for the USB controller From 237865f195f6b10e4724ce49eeb3972641da882a Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Tue, 15 Sep 2015 13:18:04 -0500 Subject: [PATCH 118/647] PCI: Revert "PCI: Call pci_read_bridge_bases() from core instead of arch code" Revert dff22d2054b5 ("PCI: Call pci_read_bridge_bases() from core instead of arch code"). Reading PCI bridge windows is not arch-specific in itself, but there is PCI core code that doesn't work correctly if we read them too early. For example, Hannes found this case on an ARM Freescale i.mx6 board: pci_bus 0000:00: root bus resource [mem 0x01000000-0x01efffff] pci 0000:00:00.0: PCI bridge to [bus 01-ff] pci 0000:00:00.0: BAR 8: no space for [mem size 0x01000000] (mem window) pci 0000:01:00.0: BAR 2: failed to assign [mem size 0x00200000] pci 0000:01:00.0: BAR 1: failed to assign [mem size 0x00004000] pci 0000:01:00.0: BAR 0: failed to assign [mem size 0x00000100] The 00:00.0 mem window needs to be at least 3MB: the 01:00.0 device needs 0x204100 of space, and mem windows are megabyte-aligned. Bus sizing can increase a bridge window size, but never *decrease* it (see d65245c3297a ("PCI: don't shrink bridge resources")). Prior to dff22d2054b5, ARM didn't read bridge windows at all, so the "original size" was zero, and we assigned a 3MB window. After dff22d2054b5, we read the bridge windows before sizing the bus. The firmware programmed a 16MB window (size 0x01000000) in 00:00.0, and since we never decrease the size, we kept 16MB even though we only needed 3MB. But 16MB doesn't fit in the host bridge aperture, so we failed to assign space for the window and the downstream devices. I think this is a defect in the PCI core: we shouldn't rely on the firmware to assign sensible windows. Ray reported a similar problem, also on ARM, with Broadcom iProc. Issues like this are too hard to fix right now, so revert dff22d2054b5. Reported-by: Hannes Reported-by: Ray Jui Link: http://lkml.kernel.org/r/CAAa04yFQEUJm7Jj1qMT57-LG7ZGtnhNDBe=PpSRa70Mj+XhW-A@mail.gmail.com Link: http://lkml.kernel.org/r/55F75BB8.4070405@broadcom.com Signed-off-by: Bjorn Helgaas Acked-by: Yinghai Lu Acked-by: Lorenzo Pieralisi --- arch/alpha/kernel/pci.c | 7 ++++++- arch/frv/mb93090-mb00/pci-vdk.c | 2 ++ arch/ia64/pci/pci.c | 5 +++-- arch/microblaze/pci/pci-common.c | 9 ++++++++- arch/mips/pci/pci.c | 6 ++++++ arch/mn10300/unit-asb2305/pci.c | 1 + arch/powerpc/kernel/pci-common.c | 8 +++++++- arch/x86/pci/common.c | 1 + arch/xtensa/kernel/pci.c | 4 ++++ drivers/parisc/dino.c | 3 +++ drivers/parisc/lba_pci.c | 1 + drivers/pci/probe.c | 6 ------ 12 files changed, 42 insertions(+), 11 deletions(-) diff --git a/arch/alpha/kernel/pci.c b/arch/alpha/kernel/pci.c index cded02c890aa..5f387ee5b5c5 100644 --- a/arch/alpha/kernel/pci.c +++ b/arch/alpha/kernel/pci.c @@ -242,7 +242,12 @@ pci_restore_srm_config(void) void pcibios_fixup_bus(struct pci_bus *bus) { - struct pci_dev *dev; + struct pci_dev *dev = bus->self; + + if (pci_has_flag(PCI_PROBE_ONLY) && dev && + (dev->class >> 8) == PCI_CLASS_BRIDGE_PCI) { + pci_read_bridge_bases(bus); + } list_for_each_entry(dev, &bus->devices, bus_list) { pdev_save_srm_config(dev); diff --git a/arch/frv/mb93090-mb00/pci-vdk.c b/arch/frv/mb93090-mb00/pci-vdk.c index f9c86c475bbd..f211839e2cae 100644 --- a/arch/frv/mb93090-mb00/pci-vdk.c +++ b/arch/frv/mb93090-mb00/pci-vdk.c @@ -294,6 +294,8 @@ void pcibios_fixup_bus(struct pci_bus *bus) printk("### PCIBIOS_FIXUP_BUS(%d)\n",bus->number); #endif + pci_read_bridge_bases(bus); + if (bus->number == 0) { struct pci_dev *dev; list_for_each_entry(dev, &bus->devices, bus_list) { diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c index d89b6013c941..7cc3be9fa7c6 100644 --- a/arch/ia64/pci/pci.c +++ b/arch/ia64/pci/pci.c @@ -533,9 +533,10 @@ void pcibios_fixup_bus(struct pci_bus *b) { struct pci_dev *dev; - if (b->self) + if (b->self) { + pci_read_bridge_bases(b); pcibios_fixup_bridge_resources(b->self); - + } list_for_each_entry(dev, &b->devices, bus_list) pcibios_fixup_device_resources(dev); platform_pci_fixup_bus(b); diff --git a/arch/microblaze/pci/pci-common.c b/arch/microblaze/pci/pci-common.c index 6b8b75266801..ae838ed5fcf2 100644 --- a/arch/microblaze/pci/pci-common.c +++ b/arch/microblaze/pci/pci-common.c @@ -863,7 +863,14 @@ void pcibios_setup_bus_devices(struct pci_bus *bus) void pcibios_fixup_bus(struct pci_bus *bus) { - /* Fixup the bus */ + /* When called from the generic PCI probe, read PCI<->PCI bridge + * bases. This is -not- called when generating the PCI tree from + * the OF device-tree. + */ + if (bus->self != NULL) + pci_read_bridge_bases(bus); + + /* Now fixup the bus bus */ pcibios_setup_bus_self(bus); /* Now fixup devices on that bus */ diff --git a/arch/mips/pci/pci.c b/arch/mips/pci/pci.c index c6996cf67a5c..b8a0bf5766f2 100644 --- a/arch/mips/pci/pci.c +++ b/arch/mips/pci/pci.c @@ -311,6 +311,12 @@ int pcibios_enable_device(struct pci_dev *dev, int mask) void pcibios_fixup_bus(struct pci_bus *bus) { + struct pci_dev *dev = bus->self; + + if (pci_has_flag(PCI_PROBE_ONLY) && dev && + (dev->class >> 8) == PCI_CLASS_BRIDGE_PCI) { + pci_read_bridge_bases(bus); + } } EXPORT_SYMBOL(PCIBIOS_MIN_IO); diff --git a/arch/mn10300/unit-asb2305/pci.c b/arch/mn10300/unit-asb2305/pci.c index deaa893efba5..3dfe2d31c67b 100644 --- a/arch/mn10300/unit-asb2305/pci.c +++ b/arch/mn10300/unit-asb2305/pci.c @@ -324,6 +324,7 @@ void pcibios_fixup_bus(struct pci_bus *bus) struct pci_dev *dev; if (bus->self) { + pci_read_bridge_bases(bus); pcibios_fixup_bridge_resources(bus->self); } diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c index a1d0632d97c6..7587b2ae5f77 100644 --- a/arch/powerpc/kernel/pci-common.c +++ b/arch/powerpc/kernel/pci-common.c @@ -1032,7 +1032,13 @@ void pcibios_set_master(struct pci_dev *dev) void pcibios_fixup_bus(struct pci_bus *bus) { - /* Fixup the bus */ + /* When called from the generic PCI probe, read PCI<->PCI bridge + * bases. This is -not- called when generating the PCI tree from + * the OF device-tree. + */ + pci_read_bridge_bases(bus); + + /* Now fixup the bus bus */ pcibios_setup_bus_self(bus); /* Now fixup devices on that bus */ diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c index 09d3afc0a181..dc78a4a9a466 100644 --- a/arch/x86/pci/common.c +++ b/arch/x86/pci/common.c @@ -166,6 +166,7 @@ void pcibios_fixup_bus(struct pci_bus *b) { struct pci_dev *dev; + pci_read_bridge_bases(b); list_for_each_entry(dev, &b->devices, bus_list) pcibios_fixup_device_resources(dev); } diff --git a/arch/xtensa/kernel/pci.c b/arch/xtensa/kernel/pci.c index d27b4dcf221f..b848cc3dc913 100644 --- a/arch/xtensa/kernel/pci.c +++ b/arch/xtensa/kernel/pci.c @@ -210,6 +210,10 @@ subsys_initcall(pcibios_init); void pcibios_fixup_bus(struct pci_bus *bus) { + if (bus->parent) { + /* This is a subordinate bridge */ + pci_read_bridge_bases(bus); + } } void pcibios_set_master(struct pci_dev *dev) diff --git a/drivers/parisc/dino.c b/drivers/parisc/dino.c index baec33c4e698..a0580afe1713 100644 --- a/drivers/parisc/dino.c +++ b/drivers/parisc/dino.c @@ -560,6 +560,9 @@ dino_fixup_bus(struct pci_bus *bus) } else if (bus->parent) { int i; + pci_read_bridge_bases(bus); + + for(i = PCI_BRIDGE_RESOURCES; i < PCI_NUM_RESOURCES; i++) { if((bus->self->resource[i].flags & (IORESOURCE_IO | IORESOURCE_MEM)) == 0) diff --git a/drivers/parisc/lba_pci.c b/drivers/parisc/lba_pci.c index 7b9e89ba0465..a32c1f6c252c 100644 --- a/drivers/parisc/lba_pci.c +++ b/drivers/parisc/lba_pci.c @@ -693,6 +693,7 @@ lba_fixup_bus(struct pci_bus *bus) if (bus->parent) { int i; /* PCI-PCI Bridge */ + pci_read_bridge_bases(bus); for (i = PCI_BRIDGE_RESOURCES; i < PCI_NUM_RESOURCES; i++) pci_claim_bridge_resource(bus->self, i); } else { diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 0b2be174d981..c8cc0e62a7b9 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -855,9 +855,6 @@ int pci_scan_bridge(struct pci_bus *bus, struct pci_dev *dev, int max, int pass) child->bridge_ctl = bctl; } - /* Read and initialize bridge resources */ - pci_read_bridge_bases(child); - cmax = pci_scan_child_bus(child); if (cmax > subordinate) dev_warn(&dev->dev, "bridge has subordinate %02x but max busn %02x\n", @@ -918,9 +915,6 @@ int pci_scan_bridge(struct pci_bus *bus, struct pci_dev *dev, int max, int pass) if (!is_cardbus) { child->bridge_ctl = bctl; - - /* Read and initialize bridge resources */ - pci_read_bridge_bases(child); max = pci_scan_child_bus(child); } else { /* From a3c119d392d7d7c68865fe76f5732ca9b8164d68 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 15 Sep 2015 14:30:05 -0700 Subject: [PATCH 119/647] ipv6: Refactor common ip6gre_tunnel_init codes It is a prep work to fix the dst_entry refcnt bugs in ip6_tunnel. This patch refactors some common init codes used by both ip6gre_tunnel_init and ip6gre_tap_init. Signed-off-by: Martin KaFai Lau Signed-off-by: David S. Miller --- net/ipv6/ip6_gre.c | 37 ++++++++++++++++++++++++------------- 1 file changed, 24 insertions(+), 13 deletions(-) diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 4038c694ec03..af60d46129c1 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -1245,7 +1245,7 @@ static void ip6gre_tunnel_setup(struct net_device *dev) netif_keep_dst(dev); } -static int ip6gre_tunnel_init(struct net_device *dev) +static int ip6gre_tunnel_init_common(struct net_device *dev) { struct ip6_tnl *tunnel; @@ -1255,16 +1255,30 @@ static int ip6gre_tunnel_init(struct net_device *dev) tunnel->net = dev_net(dev); strcpy(tunnel->parms.name, dev->name); + dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); + if (!dev->tstats) + return -ENOMEM; + + return 0; +} + +static int ip6gre_tunnel_init(struct net_device *dev) +{ + struct ip6_tnl *tunnel; + int ret; + + ret = ip6gre_tunnel_init_common(dev); + if (ret) + return ret; + + tunnel = netdev_priv(dev); + memcpy(dev->dev_addr, &tunnel->parms.laddr, sizeof(struct in6_addr)); memcpy(dev->broadcast, &tunnel->parms.raddr, sizeof(struct in6_addr)); if (ipv6_addr_any(&tunnel->parms.raddr)) dev->header_ops = &ip6gre_header_ops; - dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); - if (!dev->tstats) - return -ENOMEM; - return 0; } @@ -1460,19 +1474,16 @@ static void ip6gre_netlink_parms(struct nlattr *data[], static int ip6gre_tap_init(struct net_device *dev) { struct ip6_tnl *tunnel; + int ret; + + ret = ip6gre_tunnel_init_common(dev); + if (ret) + return ret; tunnel = netdev_priv(dev); - tunnel->dev = dev; - tunnel->net = dev_net(dev); - strcpy(tunnel->parms.name, dev->name); - ip6gre_tnl_link_config(tunnel, 1); - dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); - if (!dev->tstats) - return -ENOMEM; - return 0; } From f230d1e891ba1da5953460516960894154f265db Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 15 Sep 2015 14:30:06 -0700 Subject: [PATCH 120/647] ipv6: Rename the dst_cache helper functions in ip6_tunnel It is a prep work to fix the dst_entry refcnt bugs in ip6_tunnel. This patch rename: 1. ip6_tnl_dst_check() to ip6_tnl_dst_get() to better reflect that it will take a dst refcnt in the next patch. 2. ip6_tnl_dst_store() to ip6_tnl_dst_set() to have a more conventional name matching with ip6_tnl_dst_get(). Signed-off-by: Martin KaFai Lau Signed-off-by: David S. Miller --- include/net/ip6_tunnel.h | 4 ++-- net/ipv6/ip6_gre.c | 4 ++-- net/ipv6/ip6_tunnel.c | 12 ++++++------ 3 files changed, 10 insertions(+), 10 deletions(-) diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h index b8529aa1dae7..979b081a47e8 100644 --- a/include/net/ip6_tunnel.h +++ b/include/net/ip6_tunnel.h @@ -60,9 +60,9 @@ struct ipv6_tlv_tnl_enc_lim { __u8 encap_limit; /* tunnel encapsulation limit */ } __packed; -struct dst_entry *ip6_tnl_dst_check(struct ip6_tnl *t); +struct dst_entry *ip6_tnl_dst_get(struct ip6_tnl *t); void ip6_tnl_dst_reset(struct ip6_tnl *t); -void ip6_tnl_dst_store(struct ip6_tnl *t, struct dst_entry *dst); +void ip6_tnl_dst_set(struct ip6_tnl *t, struct dst_entry *dst); int ip6_tnl_rcv_ctl(struct ip6_tnl *t, const struct in6_addr *laddr, const struct in6_addr *raddr); int ip6_tnl_xmit_ctl(struct ip6_tnl *t, const struct in6_addr *laddr, diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index af60d46129c1..24f5dd8f76a8 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -634,7 +634,7 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb, } if (!fl6->flowi6_mark) - dst = ip6_tnl_dst_check(tunnel); + dst = ip6_tnl_dst_get(tunnel); if (!dst) { ndst = ip6_route_output(net, NULL, fl6); @@ -763,7 +763,7 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb, ip6tunnel_xmit(NULL, skb, dev); if (ndst) - ip6_tnl_dst_store(tunnel, ndst); + ip6_tnl_dst_set(tunnel, ndst); return 0; tx_err_link_failure: stats->tx_carrier_errors++; diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index b0ab420612bc..599b0b419fbc 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -126,7 +126,7 @@ static struct net_device_stats *ip6_get_stats(struct net_device *dev) * Locking : hash tables are protected by RCU and RTNL */ -struct dst_entry *ip6_tnl_dst_check(struct ip6_tnl *t) +struct dst_entry *ip6_tnl_dst_get(struct ip6_tnl *t) { struct dst_entry *dst = t->dst_cache; @@ -139,7 +139,7 @@ struct dst_entry *ip6_tnl_dst_check(struct ip6_tnl *t) return dst; } -EXPORT_SYMBOL_GPL(ip6_tnl_dst_check); +EXPORT_SYMBOL_GPL(ip6_tnl_dst_get); void ip6_tnl_dst_reset(struct ip6_tnl *t) { @@ -148,14 +148,14 @@ void ip6_tnl_dst_reset(struct ip6_tnl *t) } EXPORT_SYMBOL_GPL(ip6_tnl_dst_reset); -void ip6_tnl_dst_store(struct ip6_tnl *t, struct dst_entry *dst) +void ip6_tnl_dst_set(struct ip6_tnl *t, struct dst_entry *dst) { struct rt6_info *rt = (struct rt6_info *) dst; t->dst_cookie = rt6_get_cookie(rt); dst_release(t->dst_cache); t->dst_cache = dst; } -EXPORT_SYMBOL_GPL(ip6_tnl_dst_store); +EXPORT_SYMBOL_GPL(ip6_tnl_dst_set); /** * ip6_tnl_lookup - fetch tunnel matching the end-point addresses @@ -1010,7 +1010,7 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, memcpy(&fl6->daddr, addr6, sizeof(fl6->daddr)); neigh_release(neigh); } else if (!fl6->flowi6_mark) - dst = ip6_tnl_dst_check(t); + dst = ip6_tnl_dst_get(t); if (!ip6_tnl_xmit_ctl(t, &fl6->saddr, &fl6->daddr)) goto tx_err_link_failure; @@ -1102,7 +1102,7 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, ipv6h->daddr = fl6->daddr; ip6tunnel_xmit(NULL, skb, dev); if (ndst) - ip6_tnl_dst_store(t, ndst); + ip6_tnl_dst_set(t, ndst); return 0; tx_err_link_failure: stats->tx_carrier_errors++; From cdf3464e6c6bd764277cbbe992cd12da735b92fb Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 15 Sep 2015 14:30:07 -0700 Subject: [PATCH 121/647] ipv6: Fix dst_entry refcnt bugs in ip6_tunnel Problems in the current dst_entry cache in the ip6_tunnel: 1. ip6_tnl_dst_set is racy. There is no lock to protect it: - One major problem is that the dst refcnt gets messed up. F.e. the same dst_cache can be released multiple times and then triggering the infamous dst refcnt < 0 warning message. - Another issue is the inconsistency between dst_cache and dst_cookie. It can be reproduced by adding and removing the ip6gre tunnel while running a super_netperf TCP_CRR test. 2. ip6_tnl_dst_get does not take the dst refcnt before returning the dst. This patch: 1. Create a percpu dst_entry cache in ip6_tnl 2. Use a spinlock to protect the dst_cache operations 3. ip6_tnl_dst_get always takes the dst refcnt before returning Signed-off-by: Martin KaFai Lau Signed-off-by: David S. Miller --- include/net/ip6_tunnel.h | 11 +++- net/ipv6/ip6_gre.c | 38 +++++++------ net/ipv6/ip6_tunnel.c | 120 +++++++++++++++++++++++++++++---------- 3 files changed, 122 insertions(+), 47 deletions(-) diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h index 979b081a47e8..60b4f402f78c 100644 --- a/include/net/ip6_tunnel.h +++ b/include/net/ip6_tunnel.h @@ -32,6 +32,12 @@ struct __ip6_tnl_parm { __be32 o_key; }; +struct ip6_tnl_dst { + spinlock_t lock; + struct dst_entry *dst; + u32 cookie; +}; + /* IPv6 tunnel */ struct ip6_tnl { struct ip6_tnl __rcu *next; /* next tunnel in list */ @@ -39,8 +45,7 @@ struct ip6_tnl { struct net *net; /* netns for packet i/o */ struct __ip6_tnl_parm parms; /* tunnel configuration parameters */ struct flowi fl; /* flowi template for xmit */ - struct dst_entry *dst_cache; /* cached dst */ - u32 dst_cookie; + struct ip6_tnl_dst __percpu *dst_cache; /* cached dst */ int err_count; unsigned long err_time; @@ -61,6 +66,8 @@ struct ipv6_tlv_tnl_enc_lim { } __packed; struct dst_entry *ip6_tnl_dst_get(struct ip6_tnl *t); +int ip6_tnl_dst_init(struct ip6_tnl *t); +void ip6_tnl_dst_destroy(struct ip6_tnl *t); void ip6_tnl_dst_reset(struct ip6_tnl *t); void ip6_tnl_dst_set(struct ip6_tnl *t, struct dst_entry *dst); int ip6_tnl_rcv_ctl(struct ip6_tnl *t, const struct in6_addr *laddr, diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 24f5dd8f76a8..646512488c28 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -637,17 +637,17 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb, dst = ip6_tnl_dst_get(tunnel); if (!dst) { - ndst = ip6_route_output(net, NULL, fl6); + dst = ip6_route_output(net, NULL, fl6); - if (ndst->error) + if (dst->error) goto tx_err_link_failure; - ndst = xfrm_lookup(net, ndst, flowi6_to_flowi(fl6), NULL, 0); - if (IS_ERR(ndst)) { - err = PTR_ERR(ndst); - ndst = NULL; + dst = xfrm_lookup(net, dst, flowi6_to_flowi(fl6), NULL, 0); + if (IS_ERR(dst)) { + err = PTR_ERR(dst); + dst = NULL; goto tx_err_link_failure; } - dst = ndst; + ndst = dst; } tdev = dst->dev; @@ -702,12 +702,9 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb, skb = new_skb; } - if (fl6->flowi6_mark) { - skb_dst_set(skb, dst); - ndst = NULL; - } else { - skb_dst_set_noref(skb, dst); - } + if (!fl6->flowi6_mark && ndst) + ip6_tnl_dst_set(tunnel, ndst); + skb_dst_set(skb, dst); proto = NEXTHDR_GRE; if (encap_limit >= 0) { @@ -762,14 +759,12 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb, skb_set_inner_protocol(skb, protocol); ip6tunnel_xmit(NULL, skb, dev); - if (ndst) - ip6_tnl_dst_set(tunnel, ndst); return 0; tx_err_link_failure: stats->tx_carrier_errors++; dst_link_failure(skb); tx_err_dst_release: - dst_release(ndst); + dst_release(dst); return err; } @@ -1223,6 +1218,9 @@ static const struct net_device_ops ip6gre_netdev_ops = { static void ip6gre_dev_free(struct net_device *dev) { + struct ip6_tnl *t = netdev_priv(dev); + + ip6_tnl_dst_destroy(t); free_percpu(dev->tstats); free_netdev(dev); } @@ -1248,6 +1246,7 @@ static void ip6gre_tunnel_setup(struct net_device *dev) static int ip6gre_tunnel_init_common(struct net_device *dev) { struct ip6_tnl *tunnel; + int ret; tunnel = netdev_priv(dev); @@ -1259,6 +1258,13 @@ static int ip6gre_tunnel_init_common(struct net_device *dev) if (!dev->tstats) return -ENOMEM; + ret = ip6_tnl_dst_init(tunnel); + if (ret) { + free_percpu(dev->tstats); + dev->tstats = NULL; + return ret; + } + return 0; } diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 599b0b419fbc..851cf6d1eb45 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -126,37 +126,90 @@ static struct net_device_stats *ip6_get_stats(struct net_device *dev) * Locking : hash tables are protected by RCU and RTNL */ +static void __ip6_tnl_per_cpu_dst_set(struct ip6_tnl_dst *idst, + struct dst_entry *dst) +{ + dst_release(idst->dst); + if (dst) { + dst_hold(dst); + idst->cookie = rt6_get_cookie((struct rt6_info *)dst); + } else { + idst->cookie = 0; + } + idst->dst = dst; +} + +static void ip6_tnl_per_cpu_dst_set(struct ip6_tnl_dst *idst, + struct dst_entry *dst) +{ + + spin_lock_bh(&idst->lock); + __ip6_tnl_per_cpu_dst_set(idst, dst); + spin_unlock_bh(&idst->lock); +} + struct dst_entry *ip6_tnl_dst_get(struct ip6_tnl *t) { - struct dst_entry *dst = t->dst_cache; + struct ip6_tnl_dst *idst; + struct dst_entry *dst; - if (dst && dst->obsolete && - !dst->ops->check(dst, t->dst_cookie)) { - t->dst_cache = NULL; - dst_release(dst); - return NULL; + idst = raw_cpu_ptr(t->dst_cache); + spin_lock_bh(&idst->lock); + dst = idst->dst; + if (dst) { + if (!dst->obsolete || dst->ops->check(dst, idst->cookie)) { + dst_hold(idst->dst); + } else { + __ip6_tnl_per_cpu_dst_set(idst, NULL); + dst = NULL; + } } - + spin_unlock_bh(&idst->lock); return dst; } EXPORT_SYMBOL_GPL(ip6_tnl_dst_get); void ip6_tnl_dst_reset(struct ip6_tnl *t) { - dst_release(t->dst_cache); - t->dst_cache = NULL; + int i; + + for_each_possible_cpu(i) + ip6_tnl_per_cpu_dst_set(raw_cpu_ptr(t->dst_cache), NULL); } EXPORT_SYMBOL_GPL(ip6_tnl_dst_reset); void ip6_tnl_dst_set(struct ip6_tnl *t, struct dst_entry *dst) { - struct rt6_info *rt = (struct rt6_info *) dst; - t->dst_cookie = rt6_get_cookie(rt); - dst_release(t->dst_cache); - t->dst_cache = dst; + ip6_tnl_per_cpu_dst_set(raw_cpu_ptr(t->dst_cache), dst); + } EXPORT_SYMBOL_GPL(ip6_tnl_dst_set); +void ip6_tnl_dst_destroy(struct ip6_tnl *t) +{ + if (!t->dst_cache) + return; + + ip6_tnl_dst_reset(t); + free_percpu(t->dst_cache); +} +EXPORT_SYMBOL_GPL(ip6_tnl_dst_destroy); + +int ip6_tnl_dst_init(struct ip6_tnl *t) +{ + int i; + + t->dst_cache = alloc_percpu(struct ip6_tnl_dst); + if (!t->dst_cache) + return -ENOMEM; + + for_each_possible_cpu(i) + spin_lock_init(&per_cpu_ptr(t->dst_cache, i)->lock); + + return 0; +} +EXPORT_SYMBOL_GPL(ip6_tnl_dst_init); + /** * ip6_tnl_lookup - fetch tunnel matching the end-point addresses * @remote: the address of the tunnel exit-point @@ -271,6 +324,9 @@ ip6_tnl_unlink(struct ip6_tnl_net *ip6n, struct ip6_tnl *t) static void ip6_dev_free(struct net_device *dev) { + struct ip6_tnl *t = netdev_priv(dev); + + ip6_tnl_dst_destroy(t); free_percpu(dev->tstats); free_netdev(dev); } @@ -1016,17 +1072,17 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, goto tx_err_link_failure; if (!dst) { - ndst = ip6_route_output(net, NULL, fl6); + dst = ip6_route_output(net, NULL, fl6); - if (ndst->error) + if (dst->error) goto tx_err_link_failure; - ndst = xfrm_lookup(net, ndst, flowi6_to_flowi(fl6), NULL, 0); - if (IS_ERR(ndst)) { - err = PTR_ERR(ndst); - ndst = NULL; + dst = xfrm_lookup(net, dst, flowi6_to_flowi(fl6), NULL, 0); + if (IS_ERR(dst)) { + err = PTR_ERR(dst); + dst = NULL; goto tx_err_link_failure; } - dst = ndst; + ndst = dst; } tdev = dst->dev; @@ -1072,12 +1128,11 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, consume_skb(skb); skb = new_skb; } - if (fl6->flowi6_mark) { - skb_dst_set(skb, dst); - ndst = NULL; - } else { - skb_dst_set_noref(skb, dst); - } + + if (!fl6->flowi6_mark && ndst) + ip6_tnl_dst_set(t, ndst); + skb_dst_set(skb, dst); + skb->transport_header = skb->network_header; proto = fl6->flowi6_proto; @@ -1101,14 +1156,12 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, ipv6h->saddr = fl6->saddr; ipv6h->daddr = fl6->daddr; ip6tunnel_xmit(NULL, skb, dev); - if (ndst) - ip6_tnl_dst_set(t, ndst); return 0; tx_err_link_failure: stats->tx_carrier_errors++; dst_link_failure(skb); tx_err_dst_release: - dst_release(ndst); + dst_release(dst); return err; } @@ -1573,12 +1626,21 @@ static inline int ip6_tnl_dev_init_gen(struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); + int ret; t->dev = dev; t->net = dev_net(dev); dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); if (!dev->tstats) return -ENOMEM; + + ret = ip6_tnl_dst_init(t); + if (ret) { + free_percpu(dev->tstats); + dev->tstats = NULL; + return ret; + } + return 0; } From 8e3d5be7368107f0c27a1f8126d79b01a47e9567 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 15 Sep 2015 14:30:08 -0700 Subject: [PATCH 122/647] ipv6: Avoid double dst_free It is a prep work to get dst freeing from fib tree undergo a rcu grace period. The following is a common paradigm: if (ip6_del_rt(rt)) dst_free(rt) which means, if rt cannot be deleted from the fib tree, dst_free(rt) now. 1. We don't know the ip6_del_rt(rt) failure is because it was not managed by fib tree (e.g. DST_NOCACHE) or it had already been removed from the fib tree. 2. If rt had been managed by the fib tree, ip6_del_rt(rt) failure means dst_free(rt) has been called already. A second dst_free(rt) is not always obviously safe. The rt may have been destroyed already. 3. If rt is a DST_NOCACHE, dst_free(rt) should not be called. 4. It is a stopper to make dst freeing from fib tree undergo a rcu grace period. This patch is to use a DST_NOCACHE flag to indicate a rt is not managed by the fib tree. Signed-off-by: Martin KaFai Lau Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 7 +++---- net/ipv6/ip6_fib.c | 11 +++++++++-- net/ipv6/route.c | 7 ++++--- 3 files changed, 16 insertions(+), 9 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 030fefdc9aed..900113376d4e 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -5127,13 +5127,12 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp) rt = addrconf_get_prefix_route(&ifp->peer_addr, 128, ifp->idev->dev, 0, 0); - if (rt && ip6_del_rt(rt)) - dst_free(&rt->dst); + if (rt) + ip6_del_rt(rt); } dst_hold(&ifp->rt->dst); - if (ip6_del_rt(ifp->rt)) - dst_free(&ifp->rt->dst); + ip6_del_rt(ifp->rt); rt_genid_bump_ipv6(net); break; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 418d9823692b..e68350bf838b 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -933,6 +933,10 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, int replace_required = 0; int sernum = fib6_new_sernum(info->nl_net); + if (WARN_ON_ONCE((rt->dst.flags & DST_NOCACHE) && + !atomic_read(&rt->dst.__refcnt))) + return -EINVAL; + if (info->nlh) { if (!(info->nlh->nlmsg_flags & NLM_F_CREATE)) allow_create = 0; @@ -1025,6 +1029,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, fib6_start_gc(info->nl_net, rt); if (!(rt->rt6i_flags & RTF_CACHE)) fib6_prune_clones(info->nl_net, pn); + rt->dst.flags &= ~DST_NOCACHE; } out: @@ -1049,7 +1054,8 @@ out: atomic_inc(&pn->leaf->rt6i_ref); } #endif - dst_free(&rt->dst); + if (!(rt->dst.flags & DST_NOCACHE)) + dst_free(&rt->dst); } return err; @@ -1060,7 +1066,8 @@ out: st_failure: if (fn && !(fn->fn_flags & (RTN_RTINFO|RTN_ROOT))) fib6_repair_tree(info->nl_net, fn); - dst_free(&rt->dst); + if (!(rt->dst.flags & DST_NOCACHE)) + dst_free(&rt->dst); return err; #endif } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 53617d715188..3d3c1b294725 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1322,8 +1322,7 @@ static void ip6_link_failure(struct sk_buff *skb) if (rt) { if (rt->rt6i_flags & RTF_CACHE) { dst_hold(&rt->dst); - if (ip6_del_rt(rt)) - dst_free(&rt->dst); + ip6_del_rt(rt); } else if (rt->rt6i_node && (rt->rt6i_flags & RTF_DEFAULT)) { rt->rt6i_node->fn_sernum = -1; } @@ -2028,7 +2027,8 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info) struct fib6_table *table; struct net *net = dev_net(rt->dst.dev); - if (rt == net->ipv6.ip6_null_entry) { + if (rt == net->ipv6.ip6_null_entry || + rt->dst.flags & DST_NOCACHE) { err = -ENOENT; goto out; } @@ -2515,6 +2515,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, rt->rt6i_dst.addr = *addr; rt->rt6i_dst.plen = 128; rt->rt6i_table = fib6_get_table(net, RT6_TABLE_LOCAL); + rt->dst.flags |= DST_NOCACHE; atomic_set(&rt->dst.__refcnt, 1); From 70da5b5c532f0ec8aa76b4f46158da5f010f34b3 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 15 Sep 2015 14:30:09 -0700 Subject: [PATCH 123/647] ipv6: Replace spinlock with seqlock and rcu in ip6_tunnel This patch uses a seqlock to ensure consistency between idst->dst and idst->cookie. It also makes dst freeing from fib tree to undergo a rcu grace period. Signed-off-by: Martin KaFai Lau Signed-off-by: David S. Miller --- include/net/ip6_tunnel.h | 4 ++-- net/ipv6/ip6_fib.c | 9 +++++-- net/ipv6/ip6_tunnel.c | 51 +++++++++++++++++++++------------------- 3 files changed, 36 insertions(+), 28 deletions(-) diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h index 60b4f402f78c..65c2a9397b3c 100644 --- a/include/net/ip6_tunnel.h +++ b/include/net/ip6_tunnel.h @@ -33,8 +33,8 @@ struct __ip6_tnl_parm { }; struct ip6_tnl_dst { - spinlock_t lock; - struct dst_entry *dst; + seqlock_t lock; + struct dst_entry __rcu *dst; u32 cookie; }; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index e68350bf838b..8a9ec01f4d01 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -155,6 +155,11 @@ static void node_free(struct fib6_node *fn) kmem_cache_free(fib6_node_kmem, fn); } +static void rt6_rcu_free(struct rt6_info *rt) +{ + call_rcu(&rt->dst.rcu_head, dst_rcu_free); +} + static void rt6_free_pcpu(struct rt6_info *non_pcpu_rt) { int cpu; @@ -169,7 +174,7 @@ static void rt6_free_pcpu(struct rt6_info *non_pcpu_rt) ppcpu_rt = per_cpu_ptr(non_pcpu_rt->rt6i_pcpu, cpu); pcpu_rt = *ppcpu_rt; if (pcpu_rt) { - dst_free(&pcpu_rt->dst); + rt6_rcu_free(pcpu_rt); *ppcpu_rt = NULL; } } @@ -181,7 +186,7 @@ static void rt6_release(struct rt6_info *rt) { if (atomic_dec_and_test(&rt->rt6i_ref)) { rt6_free_pcpu(rt); - dst_free(&rt->dst); + rt6_rcu_free(rt); } } diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 851cf6d1eb45..983f0d20f96d 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -126,45 +126,48 @@ static struct net_device_stats *ip6_get_stats(struct net_device *dev) * Locking : hash tables are protected by RCU and RTNL */ -static void __ip6_tnl_per_cpu_dst_set(struct ip6_tnl_dst *idst, - struct dst_entry *dst) +static void ip6_tnl_per_cpu_dst_set(struct ip6_tnl_dst *idst, + struct dst_entry *dst) { - dst_release(idst->dst); + write_seqlock_bh(&idst->lock); + dst_release(rcu_dereference_protected( + idst->dst, + lockdep_is_held(&idst->lock.lock))); if (dst) { dst_hold(dst); idst->cookie = rt6_get_cookie((struct rt6_info *)dst); } else { idst->cookie = 0; } - idst->dst = dst; -} - -static void ip6_tnl_per_cpu_dst_set(struct ip6_tnl_dst *idst, - struct dst_entry *dst) -{ - - spin_lock_bh(&idst->lock); - __ip6_tnl_per_cpu_dst_set(idst, dst); - spin_unlock_bh(&idst->lock); + rcu_assign_pointer(idst->dst, dst); + write_sequnlock_bh(&idst->lock); } struct dst_entry *ip6_tnl_dst_get(struct ip6_tnl *t) { struct ip6_tnl_dst *idst; struct dst_entry *dst; + unsigned int seq; + u32 cookie; idst = raw_cpu_ptr(t->dst_cache); - spin_lock_bh(&idst->lock); - dst = idst->dst; - if (dst) { - if (!dst->obsolete || dst->ops->check(dst, idst->cookie)) { - dst_hold(idst->dst); - } else { - __ip6_tnl_per_cpu_dst_set(idst, NULL); - dst = NULL; - } + + rcu_read_lock(); + do { + seq = read_seqbegin(&idst->lock); + dst = rcu_dereference(idst->dst); + cookie = idst->cookie; + } while (read_seqretry(&idst->lock, seq)); + + if (dst && !atomic_inc_not_zero(&dst->__refcnt)) + dst = NULL; + rcu_read_unlock(); + + if (dst && dst->obsolete && !dst->ops->check(dst, cookie)) { + ip6_tnl_per_cpu_dst_set(idst, NULL); + dst_release(dst); + dst = NULL; } - spin_unlock_bh(&idst->lock); return dst; } EXPORT_SYMBOL_GPL(ip6_tnl_dst_get); @@ -204,7 +207,7 @@ int ip6_tnl_dst_init(struct ip6_tnl *t) return -ENOMEM; for_each_possible_cpu(i) - spin_lock_init(&per_cpu_ptr(t->dst_cache, i)->lock); + seqlock_init(&per_cpu_ptr(t->dst_cache, i)->lock); return 0; } From daf158d0d544cec80b7b30deff8cfc59a6e17610 Mon Sep 17 00:00:00 2001 From: Simon Guinot Date: Tue, 15 Sep 2015 22:41:21 +0200 Subject: [PATCH 124/647] net: mvneta: fix DMA buffer unmapping in mvneta_rx() This patch fixes a regression introduced by the commit a84e32894191 ("net: mvneta: fix refilling for Rx DMA buffers"). Due to this commit the newly allocated Rx buffers are DMA-unmapped in place of those passed to the networking stack. Obviously, this causes data corruptions. This patch fixes the issue by ensuring that the right Rx buffers are DMA-unmapped. Reported-by: Oren Laskin Signed-off-by: Simon Guinot Fixes: a84e32894191 ("net: mvneta: fix refilling for Rx DMA buffers") Cc: # v3.8+ Tested-by: Oren Laskin Signed-off-by: David S. Miller --- drivers/net/ethernet/marvell/mvneta.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index fe2299ac4f5c..09ec32e33076 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -1479,6 +1479,7 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo, struct mvneta_rx_desc *rx_desc = mvneta_rxq_next_desc_get(rxq); struct sk_buff *skb; unsigned char *data; + dma_addr_t phys_addr; u32 rx_status; int rx_bytes, err; @@ -1486,6 +1487,7 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo, rx_status = rx_desc->status; rx_bytes = rx_desc->data_size - (ETH_FCS_LEN + MVNETA_MH_SIZE); data = (unsigned char *)rx_desc->buf_cookie; + phys_addr = rx_desc->buf_phys_addr; if (!mvneta_rxq_desc_is_first_last(rx_status) || (rx_status & MVNETA_RXD_ERR_SUMMARY)) { @@ -1534,7 +1536,7 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo, if (!skb) goto err_drop_frame; - dma_unmap_single(dev->dev.parent, rx_desc->buf_phys_addr, + dma_unmap_single(dev->dev.parent, phys_addr, MVNETA_RX_BUF_SIZE(pp->pkt_size), DMA_FROM_DEVICE); rcvd_pkts++; From d64f69b0373a7d0bcec8b5da7712977518a8f42b Mon Sep 17 00:00:00 2001 From: Roopa Prabhu Date: Tue, 15 Sep 2015 14:44:29 -0700 Subject: [PATCH 125/647] rtnetlink: catch -EOPNOTSUPP errors from ndo_bridge_getlink problem reported: kernel 4.1.3 ------------ # bridge vlan port vlan ids eth0 1 PVID Egress Untagged 90 91 92 93 94 95 96 97 98 99 100 vmbr0 1 PVID Egress Untagged 94 kernel 4.2 ----------- # bridge vlan port vlan ids ndo_bridge_getlink can return -EOPNOTSUPP when an interfaces ndo_bridge_getlink op is set to switchdev_port_bridge_getlink and CONFIG_SWITCHDEV is not defined. This today can happen to bond, rocker and team devices. This patch adds -EOPNOTSUPP checks after calls to ndo_bridge_getlink. Fixes: 85fdb956726ff2a ("switchdev: cut over to new switchdev_port_bridge_getlink") Reported-by: Alexandre DERUMIER Signed-off-by: Roopa Prabhu Signed-off-by: David S. Miller --- net/core/rtnetlink.c | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index a466821d1441..0ec48403ed68 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -3047,6 +3047,7 @@ static int rtnl_bridge_getlink(struct sk_buff *skb, struct netlink_callback *cb) u32 portid = NETLINK_CB(cb->skb).portid; u32 seq = cb->nlh->nlmsg_seq; u32 filter_mask = 0; + int err; if (nlmsg_len(cb->nlh) > sizeof(struct ifinfomsg)) { struct nlattr *extfilt; @@ -3067,20 +3068,25 @@ static int rtnl_bridge_getlink(struct sk_buff *skb, struct netlink_callback *cb) struct net_device *br_dev = netdev_master_upper_dev_get(dev); if (br_dev && br_dev->netdev_ops->ndo_bridge_getlink) { - if (idx >= cb->args[0] && - br_dev->netdev_ops->ndo_bridge_getlink( - skb, portid, seq, dev, filter_mask, - NLM_F_MULTI) < 0) - break; + if (idx >= cb->args[0]) { + err = br_dev->netdev_ops->ndo_bridge_getlink( + skb, portid, seq, dev, + filter_mask, NLM_F_MULTI); + if (err < 0 && err != -EOPNOTSUPP) + break; + } idx++; } if (ops->ndo_bridge_getlink) { - if (idx >= cb->args[0] && - ops->ndo_bridge_getlink(skb, portid, seq, dev, - filter_mask, - NLM_F_MULTI) < 0) - break; + if (idx >= cb->args[0]) { + err = ops->ndo_bridge_getlink(skb, portid, + seq, dev, + filter_mask, + NLM_F_MULTI); + if (err < 0 && err != -EOPNOTSUPP) + break; + } idx++; } } From 892aa01df2ad67237f213c8f9d9b491e908aa910 Mon Sep 17 00:00:00 2001 From: Sjoerd Simons Date: Fri, 11 Sep 2015 22:25:48 +0200 Subject: [PATCH 126/647] net: stmmac: Use msleep rather then udelay for reset delay The reset delays used for stmmac are in the order of 10ms to 1 second, which is far too long for udelay usage, so switch to using msleep. Practically this fixes the PHY not being reliably detected in some cases as udelay wouldn't actually delay for long enough to let the phy reliably be reset. Signed-off-by: Sjoerd Simons Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c index b735fa22ac95..ebf6abc4853f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c @@ -161,11 +161,16 @@ int stmmac_mdio_reset(struct mii_bus *bus) if (!gpio_request(reset_gpio, "mdio-reset")) { gpio_direction_output(reset_gpio, active_low ? 1 : 0); - udelay(data->delays[0]); + if (data->delays[0]) + msleep(DIV_ROUND_UP(data->delays[0], 1000)); + gpio_set_value(reset_gpio, active_low ? 0 : 1); - udelay(data->delays[1]); + if (data->delays[1]) + msleep(DIV_ROUND_UP(data->delays[1], 1000)); + gpio_set_value(reset_gpio, active_low ? 1 : 0); - udelay(data->delays[2]); + if (data->delays[2]) + msleep(DIV_ROUND_UP(data->delays[2], 1000)); } } #endif From a4c15cd957cbd728f685645de7a150df5912591a Mon Sep 17 00:00:00 2001 From: Jenny Derzhavetz Date: Sun, 6 Sep 2015 14:52:20 +0300 Subject: [PATCH 127/647] iser-target: remove command with state ISTATE_REMOVE As documented in iscsit_sequence_cmd: /* * Existing callers for iscsit_sequence_cmd() will silently * ignore commands with CMDSN_LOWER_THAN_EXP, so force this * return for CMDSN_MAXCMDSN_OVERRUN as well.. */ We need to silently finish a command when it's in ISTATE_REMOVE. This fixes an teardown hang we were seeing where a mis-behaved initiator (triggered by allocation error injections) sent us a cmdsn which was lower than expected. Signed-off-by: Jenny Derzhavetz Signed-off-by: Sagi Grimberg Cc: # v3.10+ Signed-off-by: Nicholas Bellinger --- drivers/infiniband/ulp/isert/ib_isert.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index 403bd29443b8..443ef33fbf30 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -2999,9 +2999,16 @@ isert_get_dataout(struct iscsi_conn *conn, struct iscsi_cmd *cmd, bool recovery) static int isert_immediate_queue(struct iscsi_conn *conn, struct iscsi_cmd *cmd, int state) { - int ret; + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + int ret = 0; switch (state) { + case ISTATE_REMOVE: + spin_lock_bh(&conn->cmd_lock); + list_del_init(&cmd->i_conn_node); + spin_unlock_bh(&conn->cmd_lock); + isert_put_cmd(isert_cmd, true); + break; case ISTATE_SEND_NOPIN_WANT_RESPONSE: ret = isert_put_nopin(cmd, conn, false); break; From 3e03c4b01da3e6a5f3081eb0aa252490fe83e352 Mon Sep 17 00:00:00 2001 From: Jenny Derzhavetz Date: Sun, 6 Sep 2015 14:52:21 +0300 Subject: [PATCH 128/647] iser-target: Put the reference on commands waiting for unsol data The iscsi target core teardown sequence calls wait_conn for all active commands to finish gracefully by: - move the queue-pair to error state - drain all the completions - wait for the core to finish handling all session commands However, when tearing down a session while there are sequenced commands that are still waiting for unsolicited data outs, we can block forever as these are missing an extra reference put. We basically need the equivalent of iscsit_free_queue_reqs_for_conn() which is called after wait_conn has returned. Address this by an explicit walk on conn_cmd_list and put the extra reference. Signed-off-by: Jenny Derzhavetz Signed-off-by: Sagi Grimberg Cc: # v3.10+ Signed-off-by: Nicholas Bellinger --- drivers/infiniband/ulp/isert/ib_isert.c | 38 ++++++++++++++++++++++++- 1 file changed, 37 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index 443ef33fbf30..c8749a9e1ccd 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -3352,6 +3352,41 @@ isert_wait4flush(struct isert_conn *isert_conn) wait_for_completion(&isert_conn->wait_comp_err); } +/** + * isert_put_unsol_pending_cmds() - Drop commands waiting for + * unsolicitate dataout + * @conn: iscsi connection + * + * We might still have commands that are waiting for unsolicited + * dataouts messages. We must put the extra reference on those + * before blocking on the target_wait_for_session_cmds + */ +static void +isert_put_unsol_pending_cmds(struct iscsi_conn *conn) +{ + struct iscsi_cmd *cmd, *tmp; + static LIST_HEAD(drop_cmd_list); + + spin_lock_bh(&conn->cmd_lock); + list_for_each_entry_safe(cmd, tmp, &conn->conn_cmd_list, i_conn_node) { + if ((cmd->cmd_flags & ICF_NON_IMMEDIATE_UNSOLICITED_DATA) && + (cmd->write_data_done < conn->sess->sess_ops->FirstBurstLength) && + (cmd->write_data_done < cmd->se_cmd.data_length)) + list_move_tail(&cmd->i_conn_node, &drop_cmd_list); + } + spin_unlock_bh(&conn->cmd_lock); + + list_for_each_entry_safe(cmd, tmp, &drop_cmd_list, i_conn_node) { + list_del_init(&cmd->i_conn_node); + if (cmd->i_state != ISTATE_REMOVE) { + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + + isert_info("conn %p dropping cmd %p\n", conn, cmd); + isert_put_cmd(isert_cmd, true); + } + } +} + static void isert_wait_conn(struct iscsi_conn *conn) { struct isert_conn *isert_conn = conn->context; @@ -3370,8 +3405,9 @@ static void isert_wait_conn(struct iscsi_conn *conn) isert_conn_terminate(isert_conn); mutex_unlock(&isert_conn->mutex); - isert_wait4cmds(conn); isert_wait4flush(isert_conn); + isert_put_unsol_pending_cmds(conn); + isert_wait4cmds(conn); isert_wait4logout(isert_conn); queue_work(isert_release_wq, &isert_conn->release_work); From f27dfa1f0eb91494577a395e6b8a9aac2832e9cf Mon Sep 17 00:00:00 2001 From: Jenny Derzhavetz Date: Sun, 6 Sep 2015 14:52:22 +0300 Subject: [PATCH 129/647] iser-target: Remove unused variables Signed-off-by: Jenny Derzhavetz Signed-off-by: Sagi Grimberg Signed-off-by: Nicholas Bellinger --- drivers/infiniband/ulp/isert/ib_isert.c | 1 - drivers/infiniband/ulp/isert/ib_isert.h | 5 ----- 2 files changed, 6 deletions(-) diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index c8749a9e1ccd..97475b2eecb7 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -3116,7 +3116,6 @@ isert_setup_np(struct iscsi_np *np, sema_init(&isert_np->np_sem, 0); mutex_init(&isert_np->np_accept_mutex); INIT_LIST_HEAD(&isert_np->np_accept_list); - init_completion(&isert_np->np_login_comp); isert_np->np = np; /* diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h index 6a04ba3c0f72..0480ab89ea40 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.h +++ b/drivers/infiniband/ulp/isert/ib_isert.h @@ -113,7 +113,6 @@ enum { }; struct isert_rdma_wr { - struct list_head wr_list; struct isert_cmd *isert_cmd; enum iser_ib_op_code iser_ib_op; struct ib_sge *ib_sge; @@ -134,9 +133,6 @@ struct isert_cmd { uint64_t write_va; u64 pdu_buf_dma; u32 pdu_buf_len; - u32 read_va_off; - u32 write_va_off; - u32 rdma_wr_num; struct isert_conn *conn; struct iscsi_cmd *iscsi_cmd; struct iser_tx_desc tx_desc; @@ -226,5 +222,4 @@ struct isert_np { struct rdma_cm_id *np_cm_id; struct mutex np_accept_mutex; struct list_head np_accept_list; - struct completion np_login_comp; }; From ed8cb0a4378c0e7035db047987fe0e7309020ab5 Mon Sep 17 00:00:00 2001 From: Jenny Derzhavetz Date: Sun, 6 Sep 2015 14:52:23 +0300 Subject: [PATCH 130/647] iser-target: Remove np_ prefix from isert_np members These are always referenced from np-> so no need for the prefix. Signed-off-by: Jenny Derzhavetz Signed-off-by: Sagi Grimberg Signed-off-by: Nicholas Bellinger --- drivers/infiniband/ulp/isert/ib_isert.c | 58 ++++++++++++------------- drivers/infiniband/ulp/isert/ib_isert.h | 8 ++-- 2 files changed, 33 insertions(+), 33 deletions(-) diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index 97475b2eecb7..b163ec0ba2eb 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -778,12 +778,12 @@ isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event) if (ret) goto out_conn_dev; - mutex_lock(&isert_np->np_accept_mutex); - list_add_tail(&isert_conn->accept_node, &isert_np->np_accept_list); - mutex_unlock(&isert_np->np_accept_mutex); + mutex_lock(&isert_np->mutex); + list_add_tail(&isert_conn->accept_node, &isert_np->accept_list); + mutex_unlock(&isert_np->mutex); isert_info("np %p: Allow accept_np to continue\n", np); - up(&isert_np->np_sem); + up(&isert_np->sem); return 0; out_conn_dev: @@ -903,14 +903,14 @@ isert_np_cma_handler(struct isert_np *isert_np, switch (event) { case RDMA_CM_EVENT_DEVICE_REMOVAL: - isert_np->np_cm_id = NULL; + isert_np->cm_id = NULL; break; case RDMA_CM_EVENT_ADDR_CHANGE: - isert_np->np_cm_id = isert_setup_id(isert_np); - if (IS_ERR(isert_np->np_cm_id)) { + isert_np->cm_id = isert_setup_id(isert_np); + if (IS_ERR(isert_np->cm_id)) { isert_err("isert np %p setup id failed: %ld\n", - isert_np, PTR_ERR(isert_np->np_cm_id)); - isert_np->np_cm_id = NULL; + isert_np, PTR_ERR(isert_np->cm_id)); + isert_np->cm_id = NULL; } break; default: @@ -929,7 +929,7 @@ isert_disconnected_handler(struct rdma_cm_id *cma_id, struct isert_conn *isert_conn; bool terminating = false; - if (isert_np->np_cm_id == cma_id) + if (isert_np->cm_id == cma_id) return isert_np_cma_handler(cma_id->context, event); isert_conn = cma_id->qp->qp_context; @@ -945,13 +945,13 @@ isert_disconnected_handler(struct rdma_cm_id *cma_id, if (terminating) goto out; - mutex_lock(&isert_np->np_accept_mutex); + mutex_lock(&isert_np->mutex); if (!list_empty(&isert_conn->accept_node)) { list_del_init(&isert_conn->accept_node); isert_put_conn(isert_conn); queue_work(isert_release_wq, &isert_conn->release_work); } - mutex_unlock(&isert_np->np_accept_mutex); + mutex_unlock(&isert_np->mutex); out: return 0; @@ -3113,9 +3113,9 @@ isert_setup_np(struct iscsi_np *np, isert_err("Unable to allocate struct isert_np\n"); return -ENOMEM; } - sema_init(&isert_np->np_sem, 0); - mutex_init(&isert_np->np_accept_mutex); - INIT_LIST_HEAD(&isert_np->np_accept_list); + sema_init(&isert_np->sem, 0); + mutex_init(&isert_np->mutex); + INIT_LIST_HEAD(&isert_np->accept_list); isert_np->np = np; /* @@ -3131,7 +3131,7 @@ isert_setup_np(struct iscsi_np *np, goto out; } - isert_np->np_cm_id = isert_lid; + isert_np->cm_id = isert_lid; np->np_context = isert_np; return 0; @@ -3220,7 +3220,7 @@ isert_accept_np(struct iscsi_np *np, struct iscsi_conn *conn) int ret; accept_wait: - ret = down_interruptible(&isert_np->np_sem); + ret = down_interruptible(&isert_np->sem); if (ret) return -ENODEV; @@ -3237,15 +3237,15 @@ accept_wait: } spin_unlock_bh(&np->np_thread_lock); - mutex_lock(&isert_np->np_accept_mutex); - if (list_empty(&isert_np->np_accept_list)) { - mutex_unlock(&isert_np->np_accept_mutex); + mutex_lock(&isert_np->mutex); + if (list_empty(&isert_np->accept_list)) { + mutex_unlock(&isert_np->mutex); goto accept_wait; } - isert_conn = list_first_entry(&isert_np->np_accept_list, + isert_conn = list_first_entry(&isert_np->accept_list, struct isert_conn, accept_node); list_del_init(&isert_conn->accept_node); - mutex_unlock(&isert_np->np_accept_mutex); + mutex_unlock(&isert_np->mutex); conn->context = isert_conn; isert_conn->conn = conn; @@ -3263,28 +3263,28 @@ isert_free_np(struct iscsi_np *np) struct isert_np *isert_np = np->np_context; struct isert_conn *isert_conn, *n; - if (isert_np->np_cm_id) - rdma_destroy_id(isert_np->np_cm_id); + if (isert_np->cm_id) + rdma_destroy_id(isert_np->cm_id); /* * FIXME: At this point we don't have a good way to insure * that at this point we don't have hanging connections that * completed RDMA establishment but didn't start iscsi login * process. So work-around this by cleaning up what ever piled - * up in np_accept_list. + * up in accept_list. */ - mutex_lock(&isert_np->np_accept_mutex); - if (!list_empty(&isert_np->np_accept_list)) { + mutex_lock(&isert_np->mutex); + if (!list_empty(&isert_np->accept_list)) { isert_info("Still have isert connections, cleaning up...\n"); list_for_each_entry_safe(isert_conn, n, - &isert_np->np_accept_list, + &isert_np->accept_list, accept_node) { isert_info("cleaning isert_conn %p state (%d)\n", isert_conn, isert_conn->state); isert_connect_release(isert_conn); } } - mutex_unlock(&isert_np->np_accept_mutex); + mutex_unlock(&isert_np->mutex); np->np_context = NULL; kfree(isert_np); diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h index 0480ab89ea40..6e85da3dd835 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.h +++ b/drivers/infiniband/ulp/isert/ib_isert.h @@ -218,8 +218,8 @@ struct isert_device { struct isert_np { struct iscsi_np *np; - struct semaphore np_sem; - struct rdma_cm_id *np_cm_id; - struct mutex np_accept_mutex; - struct list_head np_accept_list; + struct semaphore sem; + struct rdma_cm_id *cm_id; + struct mutex mutex; + struct list_head accept_list; }; From bd3792205aaeb79b994338af2e5499fa503d79c7 Mon Sep 17 00:00:00 2001 From: Jenny Derzhavetz Date: Sun, 6 Sep 2015 14:52:24 +0300 Subject: [PATCH 131/647] iser-target: Fix pending connections handling in target stack shutdown sequnce Instead of handing a connection to the iscsi stack for processing right after accepting (rdma_accept) we only hand the connection to the iscsi core after we reached to a connected state (ESTABLISHED CM event). This will prevent two error scenrios: 1. race between rdma connection teardown and iscsi login sequence reported by Nic in: (ce9a9fc20a78a "iser-target: Fix REJECT CM event use-after-free OOPs") 2. target stack shutdown sequence race with constant login attempts by multiple initiators. We address this by maintaining two queues at the isert_np level: - accepted: connections that were accepted but have not reached connected state (might get rejected, unreachable or error). - pending: connections in connected state, but have yet to handed to the iscsi core for login processing. iser connections are promoted to the pending queue only from the accepted queue. This way the iscsi core now will only handle functional iser connections and once we shutdown the target stack, we look for any stales that got left behind so we can safely release them. Signed-off-by: Jenny Derzhavetz Signed-off-by: Sagi Grimberg Cc: # v3.10+ Signed-off-by: Nicholas Bellinger --- drivers/infiniband/ulp/isert/ib_isert.c | 66 ++++++++++++++----------- drivers/infiniband/ulp/isert/ib_isert.h | 5 +- 2 files changed, 40 insertions(+), 31 deletions(-) diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index b163ec0ba2eb..f3f498f6ecc5 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -634,7 +634,7 @@ static void isert_init_conn(struct isert_conn *isert_conn) { isert_conn->state = ISER_CONN_INIT; - INIT_LIST_HEAD(&isert_conn->accept_node); + INIT_LIST_HEAD(&isert_conn->node); init_completion(&isert_conn->login_comp); init_completion(&isert_conn->login_req_comp); init_completion(&isert_conn->wait); @@ -762,28 +762,15 @@ isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event) ret = isert_rdma_post_recvl(isert_conn); if (ret) goto out_conn_dev; - /* - * Obtain the second reference now before isert_rdma_accept() to - * ensure that any initiator generated REJECT CM event that occurs - * asynchronously won't drop the last reference until the error path - * in iscsi_target_login_sess_out() does it's ->iscsit_free_conn() -> - * isert_free_conn() -> isert_put_conn() -> kref_put(). - */ - if (!kref_get_unless_zero(&isert_conn->kref)) { - isert_warn("conn %p connect_release is running\n", isert_conn); - goto out_conn_dev; - } ret = isert_rdma_accept(isert_conn); if (ret) goto out_conn_dev; mutex_lock(&isert_np->mutex); - list_add_tail(&isert_conn->accept_node, &isert_np->accept_list); + list_add_tail(&isert_conn->node, &isert_np->accepted); mutex_unlock(&isert_np->mutex); - isert_info("np %p: Allow accept_np to continue\n", np); - up(&isert_np->sem); return 0; out_conn_dev: @@ -831,13 +818,21 @@ static void isert_connected_handler(struct rdma_cm_id *cma_id) { struct isert_conn *isert_conn = cma_id->qp->qp_context; + struct isert_np *isert_np = cma_id->context; isert_info("conn %p\n", isert_conn); mutex_lock(&isert_conn->mutex); - if (isert_conn->state != ISER_CONN_FULL_FEATURE) - isert_conn->state = ISER_CONN_UP; + isert_conn->state = ISER_CONN_UP; + kref_get(&isert_conn->kref); mutex_unlock(&isert_conn->mutex); + + mutex_lock(&isert_np->mutex); + list_move_tail(&isert_conn->node, &isert_np->pending); + mutex_unlock(&isert_np->mutex); + + isert_info("np %p: Allow accept_np to continue\n", isert_np); + up(&isert_np->sem); } static void @@ -946,8 +941,8 @@ isert_disconnected_handler(struct rdma_cm_id *cma_id, goto out; mutex_lock(&isert_np->mutex); - if (!list_empty(&isert_conn->accept_node)) { - list_del_init(&isert_conn->accept_node); + if (!list_empty(&isert_conn->node)) { + list_del_init(&isert_conn->node); isert_put_conn(isert_conn); queue_work(isert_release_wq, &isert_conn->release_work); } @@ -962,6 +957,7 @@ isert_connect_error(struct rdma_cm_id *cma_id) { struct isert_conn *isert_conn = cma_id->qp->qp_context; + list_del_init(&isert_conn->node); isert_conn->cm_id = NULL; isert_put_conn(isert_conn); @@ -3115,7 +3111,8 @@ isert_setup_np(struct iscsi_np *np, } sema_init(&isert_np->sem, 0); mutex_init(&isert_np->mutex); - INIT_LIST_HEAD(&isert_np->accept_list); + INIT_LIST_HEAD(&isert_np->accepted); + INIT_LIST_HEAD(&isert_np->pending); isert_np->np = np; /* @@ -3238,13 +3235,13 @@ accept_wait: spin_unlock_bh(&np->np_thread_lock); mutex_lock(&isert_np->mutex); - if (list_empty(&isert_np->accept_list)) { + if (list_empty(&isert_np->pending)) { mutex_unlock(&isert_np->mutex); goto accept_wait; } - isert_conn = list_first_entry(&isert_np->accept_list, - struct isert_conn, accept_node); - list_del_init(&isert_conn->accept_node); + isert_conn = list_first_entry(&isert_np->pending, + struct isert_conn, node); + list_del_init(&isert_conn->node); mutex_unlock(&isert_np->mutex); conn->context = isert_conn; @@ -3271,14 +3268,25 @@ isert_free_np(struct iscsi_np *np) * that at this point we don't have hanging connections that * completed RDMA establishment but didn't start iscsi login * process. So work-around this by cleaning up what ever piled - * up in accept_list. + * up in accepted and pending lists. */ mutex_lock(&isert_np->mutex); - if (!list_empty(&isert_np->accept_list)) { - isert_info("Still have isert connections, cleaning up...\n"); + if (!list_empty(&isert_np->pending)) { + isert_info("Still have isert pending connections\n"); list_for_each_entry_safe(isert_conn, n, - &isert_np->accept_list, - accept_node) { + &isert_np->pending, + node) { + isert_info("cleaning isert_conn %p state (%d)\n", + isert_conn, isert_conn->state); + isert_connect_release(isert_conn); + } + } + + if (!list_empty(&isert_np->accepted)) { + isert_info("Still have isert accepted connections\n"); + list_for_each_entry_safe(isert_conn, n, + &isert_np->accepted, + node) { isert_info("cleaning isert_conn %p state (%d)\n", isert_conn, isert_conn->state); isert_connect_release(isert_conn); diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h index 6e85da3dd835..b81dfe07ce62 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.h +++ b/drivers/infiniband/ulp/isert/ib_isert.h @@ -159,7 +159,7 @@ struct isert_conn { struct iser_rx_desc *rx_descs; struct ib_recv_wr rx_wr[ISERT_MIN_POSTED_RX]; struct iscsi_conn *conn; - struct list_head accept_node; + struct list_head node; struct completion login_comp; struct completion login_req_comp; struct iser_tx_desc login_tx_desc; @@ -221,5 +221,6 @@ struct isert_np { struct semaphore sem; struct rdma_cm_id *cm_id; struct mutex mutex; - struct list_head accept_list; + struct list_head accepted; + struct list_head pending; }; From 4366b19ca5eb15e63d6640565ade135cf06be91a Mon Sep 17 00:00:00 2001 From: Jenny Derzhavetz Date: Sun, 6 Sep 2015 14:52:25 +0300 Subject: [PATCH 132/647] iser-target: Change the recv buffers posting logic iser target batches post recv operations to avoid the overhead of acquiring the recv queue lock and posting a HW doorbell for each command. We change it to be per command in order to support zcopy immediate data for IOs that fits in the 8K transfer boundary (in the next patch). (Fix minor patch fuzz due to ib_mr removal - nab) Signed-off-by: Jenny Derzhavetz Signed-off-by: Sagi Grimberg Signed-off-by: Nicholas Bellinger --- drivers/infiniband/ulp/isert/ib_isert.c | 111 ++++++++++++++---------- drivers/infiniband/ulp/isert/ib_isert.h | 4 +- 2 files changed, 67 insertions(+), 48 deletions(-) diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index f3f498f6ecc5..233b8c7a78df 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -238,8 +238,6 @@ isert_alloc_rx_descriptors(struct isert_conn *isert_conn) rx_sg->lkey = device->pd->local_dma_lkey; } - isert_conn->rx_desc_head = 0; - return 0; dma_map_fail: @@ -1002,35 +1000,51 @@ isert_cma_handler(struct rdma_cm_id *cma_id, struct rdma_cm_event *event) } static int -isert_post_recv(struct isert_conn *isert_conn, u32 count) +isert_post_recvm(struct isert_conn *isert_conn, u32 count) { struct ib_recv_wr *rx_wr, *rx_wr_failed; int i, ret; - unsigned int rx_head = isert_conn->rx_desc_head; struct iser_rx_desc *rx_desc; for (rx_wr = isert_conn->rx_wr, i = 0; i < count; i++, rx_wr++) { - rx_desc = &isert_conn->rx_descs[rx_head]; - rx_wr->wr_id = (uintptr_t)rx_desc; - rx_wr->sg_list = &rx_desc->rx_sg; - rx_wr->num_sge = 1; - rx_wr->next = rx_wr + 1; - rx_head = (rx_head + 1) & (ISERT_QP_MAX_RECV_DTOS - 1); + rx_desc = &isert_conn->rx_descs[i]; + rx_wr->wr_id = (uintptr_t)rx_desc; + rx_wr->sg_list = &rx_desc->rx_sg; + rx_wr->num_sge = 1; + rx_wr->next = rx_wr + 1; } - rx_wr--; rx_wr->next = NULL; /* mark end of work requests list */ isert_conn->post_recv_buf_count += count; ret = ib_post_recv(isert_conn->qp, isert_conn->rx_wr, - &rx_wr_failed); + &rx_wr_failed); if (ret) { isert_err("ib_post_recv() failed with ret: %d\n", ret); isert_conn->post_recv_buf_count -= count; - } else { - isert_dbg("Posted %d RX buffers\n", count); - isert_conn->rx_desc_head = rx_head; } + + return ret; +} + +static int +isert_post_recv(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc) +{ + struct ib_recv_wr *rx_wr_failed, rx_wr; + int ret; + + rx_wr.wr_id = (uintptr_t)rx_desc; + rx_wr.sg_list = &rx_desc->rx_sg; + rx_wr.num_sge = 1; + rx_wr.next = NULL; + + isert_conn->post_recv_buf_count++; + ret = ib_post_recv(isert_conn->qp, &rx_wr, &rx_wr_failed); + if (ret) { + isert_err("ib_post_recv() failed with ret: %d\n", ret); + isert_conn->post_recv_buf_count--; + } + return ret; } @@ -1201,7 +1215,8 @@ isert_put_login_tx(struct iscsi_conn *conn, struct iscsi_login *login, if (ret) return ret; - ret = isert_post_recv(isert_conn, ISERT_MIN_POSTED_RX); + ret = isert_post_recvm(isert_conn, + ISERT_QP_MAX_RECV_DTOS); if (ret) return ret; @@ -1274,7 +1289,7 @@ isert_rx_login_req(struct isert_conn *isert_conn) } static struct iscsi_cmd -*isert_allocate_cmd(struct iscsi_conn *conn) +*isert_allocate_cmd(struct iscsi_conn *conn, struct iser_rx_desc *rx_desc) { struct isert_conn *isert_conn = conn->context; struct isert_cmd *isert_cmd; @@ -1288,6 +1303,7 @@ static struct iscsi_cmd isert_cmd = iscsit_priv_cmd(cmd); isert_cmd->conn = isert_conn; isert_cmd->iscsi_cmd = cmd; + isert_cmd->rx_desc = rx_desc; return cmd; } @@ -1403,6 +1419,15 @@ isert_handle_iscsi_dataout(struct isert_conn *isert_conn, if (rc < 0) return rc; + /* + * multiple data-outs on the same command can arrive - + * so post the buffer before hand + */ + rc = isert_post_recv(isert_conn, rx_desc); + if (rc) { + isert_err("ib_post_recv failed with %d\n", rc); + return rc; + } return 0; } @@ -1475,7 +1500,7 @@ isert_rx_opcode(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc, switch (opcode) { case ISCSI_OP_SCSI_CMD: - cmd = isert_allocate_cmd(conn); + cmd = isert_allocate_cmd(conn, rx_desc); if (!cmd) break; @@ -1489,7 +1514,7 @@ isert_rx_opcode(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc, rx_desc, (unsigned char *)hdr); break; case ISCSI_OP_NOOP_OUT: - cmd = isert_allocate_cmd(conn); + cmd = isert_allocate_cmd(conn, rx_desc); if (!cmd) break; @@ -1502,7 +1527,7 @@ isert_rx_opcode(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc, (unsigned char *)hdr); break; case ISCSI_OP_SCSI_TMFUNC: - cmd = isert_allocate_cmd(conn); + cmd = isert_allocate_cmd(conn, rx_desc); if (!cmd) break; @@ -1510,22 +1535,20 @@ isert_rx_opcode(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc, (unsigned char *)hdr); break; case ISCSI_OP_LOGOUT: - cmd = isert_allocate_cmd(conn); + cmd = isert_allocate_cmd(conn, rx_desc); if (!cmd) break; ret = iscsit_handle_logout_cmd(conn, cmd, (unsigned char *)hdr); break; case ISCSI_OP_TEXT: - if (be32_to_cpu(hdr->ttt) != 0xFFFFFFFF) { + if (be32_to_cpu(hdr->ttt) != 0xFFFFFFFF) cmd = iscsit_find_cmd_from_itt(conn, hdr->itt); - if (!cmd) - break; - } else { - cmd = isert_allocate_cmd(conn); - if (!cmd) - break; - } + else + cmd = isert_allocate_cmd(conn, rx_desc); + + if (!cmd) + break; isert_cmd = iscsit_priv_cmd(cmd); ret = isert_handle_text_cmd(isert_conn, isert_cmd, cmd, @@ -1585,7 +1608,7 @@ isert_rcv_completion(struct iser_rx_desc *desc, struct ib_device *ib_dev = isert_conn->cm_id->device; struct iscsi_hdr *hdr; u64 rx_dma; - int rx_buflen, outstanding; + int rx_buflen; if ((char *)desc == isert_conn->login_req_buf) { rx_dma = isert_conn->login_req_dma; @@ -1625,22 +1648,6 @@ isert_rcv_completion(struct iser_rx_desc *desc, DMA_FROM_DEVICE); isert_conn->post_recv_buf_count--; - isert_dbg("Decremented post_recv_buf_count: %d\n", - isert_conn->post_recv_buf_count); - - if ((char *)desc == isert_conn->login_req_buf) - return; - - outstanding = isert_conn->post_recv_buf_count; - if (outstanding + ISERT_MIN_POSTED_RX <= ISERT_QP_MAX_RECV_DTOS) { - int err, count = min(ISERT_QP_MAX_RECV_DTOS - outstanding, - ISERT_MIN_POSTED_RX); - err = isert_post_recv(isert_conn, count); - if (err) { - isert_err("isert_post_recv() count: %d failed, %d\n", - count, err); - } - } } static int @@ -2152,6 +2159,12 @@ isert_post_response(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd) struct ib_send_wr *wr_failed; int ret; + ret = isert_post_recv(isert_conn, isert_cmd->rx_desc); + if (ret) { + isert_err("ib_post_recv failed with %d\n", ret); + return ret; + } + ret = ib_post_send(isert_conn->qp, &isert_cmd->tx_desc.send_wr, &wr_failed); if (ret) { @@ -2946,6 +2959,12 @@ isert_put_datain(struct iscsi_conn *conn, struct iscsi_cmd *cmd) &isert_cmd->tx_desc.send_wr); isert_cmd->rdma_wr.s_send_wr.next = &isert_cmd->tx_desc.send_wr; wr->send_wr_num += 1; + + rc = isert_post_recv(isert_conn, isert_cmd->rx_desc); + if (rc) { + isert_err("ib_post_recv failed with %d\n", rc); + return rc; + } } rc = ib_post_send(isert_conn->qp, wr->send_wr, &wr_failed); diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h index b81dfe07ce62..e586ee1691e9 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.h +++ b/drivers/infiniband/ulp/isert/ib_isert.h @@ -136,6 +136,7 @@ struct isert_cmd { struct isert_conn *conn; struct iscsi_cmd *iscsi_cmd; struct iser_tx_desc tx_desc; + struct iser_rx_desc *rx_desc; struct isert_rdma_wr rdma_wr; struct work_struct comp_work; }; @@ -155,9 +156,8 @@ struct isert_conn { u64 login_req_dma; int login_req_len; u64 login_rsp_dma; - unsigned int rx_desc_head; struct iser_rx_desc *rx_descs; - struct ib_recv_wr rx_wr[ISERT_MIN_POSTED_RX]; + struct ib_recv_wr rx_wr[ISERT_QP_MAX_RECV_DTOS]; struct iscsi_conn *conn; struct list_head node; struct completion login_comp; From 9fd60088ffed7573c2d409ddc63a2150a5edd5d8 Mon Sep 17 00:00:00 2001 From: Jenny Derzhavetz Date: Sun, 6 Sep 2015 14:52:26 +0300 Subject: [PATCH 133/647] iser-target: Skip data copy if all the command data comes as immediate Given that supporting zcopy immediate data for all IOs requires iser driver to use its own buffer allocations, we settle with avoiding data copy for IOs with data length of up to 8K (which is more latency sensitive anyway). This trims IO write latency by up to 3us and increase IOPs by up to 40% by saving CPU time doing sg_copy_from_buffer (8K IO size is the obvious winner here). Signed-off-by: Jenny Derzhavetz Signed-off-by: Sagi Grimberg Signed-off-by: Nicholas Bellinger --- drivers/infiniband/ulp/isert/ib_isert.c | 26 +++++++++++++++++-------- drivers/infiniband/ulp/isert/ib_isert.h | 1 + 2 files changed, 19 insertions(+), 8 deletions(-) diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index 233b8c7a78df..aa59037d7504 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -1315,9 +1315,9 @@ isert_handle_scsi_cmd(struct isert_conn *isert_conn, { struct iscsi_conn *conn = isert_conn->conn; struct iscsi_scsi_req *hdr = (struct iscsi_scsi_req *)buf; - struct scatterlist *sg; int imm_data, imm_data_len, unsol_data, sg_nents, rc; bool dump_payload = false; + unsigned int data_len; rc = iscsit_setup_scsi_cmd(conn, cmd, buf); if (rc < 0) @@ -1326,7 +1326,10 @@ isert_handle_scsi_cmd(struct isert_conn *isert_conn, imm_data = cmd->immediate_data; imm_data_len = cmd->first_burst_len; unsol_data = cmd->unsolicited_data; + data_len = cmd->se_cmd.data_length; + if (imm_data && imm_data_len == data_len) + cmd->se_cmd.se_cmd_flags |= SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC; rc = iscsit_process_scsi_cmd(conn, cmd, hdr); if (rc < 0) { return 0; @@ -1338,13 +1341,20 @@ isert_handle_scsi_cmd(struct isert_conn *isert_conn, if (!imm_data) return 0; - sg = &cmd->se_cmd.t_data_sg[0]; - sg_nents = max(1UL, DIV_ROUND_UP(imm_data_len, PAGE_SIZE)); - - isert_dbg("Copying Immediate SG: %p sg_nents: %u from %p imm_data_len: %d\n", - sg, sg_nents, &rx_desc->data[0], imm_data_len); - - sg_copy_from_buffer(sg, sg_nents, &rx_desc->data[0], imm_data_len); + if (imm_data_len != data_len) { + sg_nents = max(1UL, DIV_ROUND_UP(imm_data_len, PAGE_SIZE)); + sg_copy_from_buffer(cmd->se_cmd.t_data_sg, sg_nents, + &rx_desc->data[0], imm_data_len); + isert_dbg("Copy Immediate sg_nents: %u imm_data_len: %d\n", + sg_nents, imm_data_len); + } else { + sg_init_table(&isert_cmd->sg, 1); + cmd->se_cmd.t_data_sg = &isert_cmd->sg; + cmd->se_cmd.t_data_nents = 1; + sg_set_buf(&isert_cmd->sg, &rx_desc->data[0], imm_data_len); + isert_dbg("Transfer Immediate imm_data_len: %d\n", + imm_data_len); + } cmd->write_data_done += imm_data_len; diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h index e586ee1691e9..c5b99bcecbcf 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.h +++ b/drivers/infiniband/ulp/isert/ib_isert.h @@ -139,6 +139,7 @@ struct isert_cmd { struct iser_rx_desc *rx_desc; struct isert_rdma_wr rdma_wr; struct work_struct comp_work; + struct scatterlist sg; }; struct isert_device; From 982b527004826b40de1e821b123c70f05b41496c Mon Sep 17 00:00:00 2001 From: Jesse Gross Date: Fri, 11 Sep 2015 18:38:28 -0700 Subject: [PATCH 134/647] openvswitch: Fix mask generation for nested attributes. Masks were added to OVS flows in a way that was backwards compatible with userspace programs that did not generate masks. As a result, it is possible that we may receive flows that do not have a mask and we need to synthesize one. Generating a mask requires iterating over attributes and descending into nested attributes. For each level we need to know the size to generate the correct mask. We do this with a linked table of attribute types. Although the logic to handle these nested attributes was there in concept, there are a number of bugs in practice. Examples include incomplete links between tables, variable length attributes being treated as nested and missing sanity checks. Signed-off-by: Jesse Gross Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/flow_netlink.c | 82 ++++++++++++++++++++++++---------- 1 file changed, 59 insertions(+), 23 deletions(-) diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index c92d6a262bc5..5c030a4d7338 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -57,6 +57,7 @@ struct ovs_len_tbl { }; #define OVS_ATTR_NESTED -1 +#define OVS_ATTR_VARIABLE -2 static void update_range(struct sw_flow_match *match, size_t offset, size_t size, bool is_mask) @@ -304,6 +305,10 @@ size_t ovs_key_attr_size(void) + nla_total_size(28); /* OVS_KEY_ATTR_ND */ } +static const struct ovs_len_tbl ovs_vxlan_ext_key_lens[OVS_VXLAN_EXT_MAX + 1] = { + [OVS_VXLAN_EXT_GBP] = { .len = sizeof(u32) }, +}; + static const struct ovs_len_tbl ovs_tunnel_key_lens[OVS_TUNNEL_KEY_ATTR_MAX + 1] = { [OVS_TUNNEL_KEY_ATTR_ID] = { .len = sizeof(u64) }, [OVS_TUNNEL_KEY_ATTR_IPV4_SRC] = { .len = sizeof(u32) }, @@ -315,8 +320,9 @@ static const struct ovs_len_tbl ovs_tunnel_key_lens[OVS_TUNNEL_KEY_ATTR_MAX + 1] [OVS_TUNNEL_KEY_ATTR_TP_SRC] = { .len = sizeof(u16) }, [OVS_TUNNEL_KEY_ATTR_TP_DST] = { .len = sizeof(u16) }, [OVS_TUNNEL_KEY_ATTR_OAM] = { .len = 0 }, - [OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS] = { .len = OVS_ATTR_NESTED }, - [OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS] = { .len = OVS_ATTR_NESTED }, + [OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS] = { .len = OVS_ATTR_VARIABLE }, + [OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS] = { .len = OVS_ATTR_NESTED, + .next = ovs_vxlan_ext_key_lens }, }; /* The size of the argument for each %OVS_KEY_ATTR_* Netlink attribute. */ @@ -349,6 +355,13 @@ static const struct ovs_len_tbl ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = { [OVS_KEY_ATTR_CT_LABEL] = { .len = sizeof(struct ovs_key_ct_label) }, }; +static bool check_attr_len(unsigned int attr_len, unsigned int expected_len) +{ + return expected_len == attr_len || + expected_len == OVS_ATTR_NESTED || + expected_len == OVS_ATTR_VARIABLE; +} + static bool is_all_zero(const u8 *fp, size_t size) { int i; @@ -388,7 +401,7 @@ static int __parse_flow_nlattrs(const struct nlattr *attr, } expected_len = ovs_key_lens[type].len; - if (nla_len(nla) != expected_len && expected_len != OVS_ATTR_NESTED) { + if (!check_attr_len(nla_len(nla), expected_len)) { OVS_NLERR(log, "Key %d has unexpected len %d expected %d", type, nla_len(nla), expected_len); return -EINVAL; @@ -473,29 +486,50 @@ static int genev_tun_opt_from_nlattr(const struct nlattr *a, return 0; } -static const struct nla_policy vxlan_opt_policy[OVS_VXLAN_EXT_MAX + 1] = { - [OVS_VXLAN_EXT_GBP] = { .type = NLA_U32 }, -}; - -static int vxlan_tun_opt_from_nlattr(const struct nlattr *a, +static int vxlan_tun_opt_from_nlattr(const struct nlattr *attr, struct sw_flow_match *match, bool is_mask, bool log) { - struct nlattr *tb[OVS_VXLAN_EXT_MAX+1]; + struct nlattr *a; + int rem; unsigned long opt_key_offset; struct vxlan_metadata opts; - int err; BUILD_BUG_ON(sizeof(opts) > sizeof(match->key->tun_opts)); - err = nla_parse_nested(tb, OVS_VXLAN_EXT_MAX, a, vxlan_opt_policy); - if (err < 0) - return err; - memset(&opts, 0, sizeof(opts)); + nla_for_each_nested(a, attr, rem) { + int type = nla_type(a); - if (tb[OVS_VXLAN_EXT_GBP]) - opts.gbp = nla_get_u32(tb[OVS_VXLAN_EXT_GBP]); + if (type > OVS_VXLAN_EXT_MAX) { + OVS_NLERR(log, "VXLAN extension %d out of range max %d", + type, OVS_VXLAN_EXT_MAX); + return -EINVAL; + } + + if (!check_attr_len(nla_len(a), + ovs_vxlan_ext_key_lens[type].len)) { + OVS_NLERR(log, "VXLAN extension %d has unexpected len %d expected %d", + type, nla_len(a), + ovs_vxlan_ext_key_lens[type].len); + return -EINVAL; + } + + switch (type) { + case OVS_VXLAN_EXT_GBP: + opts.gbp = nla_get_u32(a); + break; + default: + OVS_NLERR(log, "Unknown VXLAN extension attribute %d", + type); + return -EINVAL; + } + } + if (rem) { + OVS_NLERR(log, "VXLAN extension message has %d unknown bytes.", + rem); + return -EINVAL; + } if (!is_mask) SW_FLOW_KEY_PUT(match, tun_opts_len, sizeof(opts), false); @@ -528,8 +562,8 @@ static int ipv4_tun_from_nlattr(const struct nlattr *attr, return -EINVAL; } - if (ovs_tunnel_key_lens[type].len != nla_len(a) && - ovs_tunnel_key_lens[type].len != OVS_ATTR_NESTED) { + if (!check_attr_len(nla_len(a), + ovs_tunnel_key_lens[type].len)) { OVS_NLERR(log, "Tunnel attr %d has unexpected len %d expected %d", type, nla_len(a), ovs_tunnel_key_lens[type].len); return -EINVAL; @@ -1052,10 +1086,13 @@ static void nlattr_set(struct nlattr *attr, u8 val, /* The nlattr stream should already have been validated */ nla_for_each_nested(nla, attr, rem) { - if (tbl && tbl[nla_type(nla)].len == OVS_ATTR_NESTED) - nlattr_set(nla, val, tbl[nla_type(nla)].next); - else + if (tbl[nla_type(nla)].len == OVS_ATTR_NESTED) { + if (tbl[nla_type(nla)].next) + tbl = tbl[nla_type(nla)].next; + nlattr_set(nla, val, tbl); + } else { memset(nla_data(nla), val, nla_len(nla)); + } } } @@ -1922,8 +1959,7 @@ static int validate_set(const struct nlattr *a, key_len /= 2; if (key_type > OVS_KEY_ATTR_MAX || - (ovs_key_lens[key_type].len != key_len && - ovs_key_lens[key_type].len != OVS_ATTR_NESTED)) + !check_attr_len(key_len, ovs_key_lens[key_type].len)) return -EINVAL; if (masked && !validate_masked(nla_data(ovs_key), key_len)) From 58d29e3ce903dcafacee9e355225d64922325cf0 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 13 Sep 2015 14:15:03 +0200 Subject: [PATCH 135/647] atm: he: drop null test before destroy functions Remove unneeded NULL test. The semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // @@ expression x; @@ -if (x != NULL) \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x); // Signed-off-by: Julia Lawall Signed-off-by: David S. Miller --- drivers/atm/he.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/atm/he.c b/drivers/atm/he.c index a8da3a50e374..0f5cb37636bc 100644 --- a/drivers/atm/he.c +++ b/drivers/atm/he.c @@ -1578,9 +1578,7 @@ he_stop(struct he_dev *he_dev) kfree(he_dev->rbpl_virt); kfree(he_dev->rbpl_table); - - if (he_dev->rbpl_pool) - dma_pool_destroy(he_dev->rbpl_pool); + dma_pool_destroy(he_dev->rbpl_pool); if (he_dev->rbrq_base) dma_free_coherent(&he_dev->pci_dev->dev, CONFIG_RBRQ_SIZE * sizeof(struct he_rbrq), @@ -1594,8 +1592,7 @@ he_stop(struct he_dev *he_dev) dma_free_coherent(&he_dev->pci_dev->dev, CONFIG_TBRQ_SIZE * sizeof(struct he_tbrq), he_dev->tpdrq_base, he_dev->tpdrq_phys); - if (he_dev->tpd_pool) - dma_pool_destroy(he_dev->tpd_pool); + dma_pool_destroy(he_dev->tpd_pool); if (he_dev->pci_dev) { pci_read_config_word(he_dev->pci_dev, PCI_COMMAND, &command); From adf78edac09f9640cd9676571586c4be46fb527c Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 13 Sep 2015 14:15:18 +0200 Subject: [PATCH 136/647] net: core: drop null test before destroy functions Remove unneeded NULL test. The semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // @@ expression x; @@ -if (x != NULL) { \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x); x = NULL; -} // Signed-off-by: Julia Lawall Signed-off-by: David S. Miller --- net/core/sock.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/net/core/sock.c b/net/core/sock.c index ca2984afe16e..3307c02244d3 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2740,10 +2740,8 @@ static void req_prot_cleanup(struct request_sock_ops *rsk_prot) return; kfree(rsk_prot->slab_name); rsk_prot->slab_name = NULL; - if (rsk_prot->slab) { - kmem_cache_destroy(rsk_prot->slab); - rsk_prot->slab = NULL; - } + kmem_cache_destroy(rsk_prot->slab); + rsk_prot->slab = NULL; } static int req_prot_init(const struct proto *prot) @@ -2828,10 +2826,8 @@ void proto_unregister(struct proto *prot) list_del(&prot->node); mutex_unlock(&proto_list_mutex); - if (prot->slab != NULL) { - kmem_cache_destroy(prot->slab); - prot->slab = NULL; - } + kmem_cache_destroy(prot->slab); + prot->slab = NULL; req_prot_cleanup(prot->rsk_prot); From 20471ed4d403a5f4de6aa0c10cd1e446f7f2b3c7 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 13 Sep 2015 14:15:27 +0200 Subject: [PATCH 137/647] dccp: drop null test before destroy functions Remove unneeded NULL test. The semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // @@ expression x; @@ -if (x != NULL) \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x); @@ expression x; @@ -if (x != NULL) { \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x); x = NULL; -} // Signed-off-by: Julia Lawall Signed-off-by: David S. Miller --- net/dccp/ackvec.c | 12 ++++-------- net/dccp/ccid.c | 3 +-- 2 files changed, 5 insertions(+), 10 deletions(-) diff --git a/net/dccp/ackvec.c b/net/dccp/ackvec.c index bd9e718c2a20..3de0d0362d7f 100644 --- a/net/dccp/ackvec.c +++ b/net/dccp/ackvec.c @@ -398,12 +398,8 @@ out_err: void dccp_ackvec_exit(void) { - if (dccp_ackvec_slab != NULL) { - kmem_cache_destroy(dccp_ackvec_slab); - dccp_ackvec_slab = NULL; - } - if (dccp_ackvec_record_slab != NULL) { - kmem_cache_destroy(dccp_ackvec_record_slab); - dccp_ackvec_record_slab = NULL; - } + kmem_cache_destroy(dccp_ackvec_slab); + dccp_ackvec_slab = NULL; + kmem_cache_destroy(dccp_ackvec_record_slab); + dccp_ackvec_record_slab = NULL; } diff --git a/net/dccp/ccid.c b/net/dccp/ccid.c index 83498975165f..90f77d08cc37 100644 --- a/net/dccp/ccid.c +++ b/net/dccp/ccid.c @@ -95,8 +95,7 @@ static struct kmem_cache *ccid_kmem_cache_create(int obj_size, char *slab_name_f static void ccid_kmem_cache_destroy(struct kmem_cache *slab) { - if (slab != NULL) - kmem_cache_destroy(slab); + kmem_cache_destroy(slab); } static int __init ccid_activate(struct ccid_operations *ccid_ops) From 8315b77d72c5f0b18ceb513303d845e73166133c Mon Sep 17 00:00:00 2001 From: Li Jun Date: Wed, 16 Sep 2015 14:46:32 +0800 Subject: [PATCH 138/647] usb: chipidea: imx: fix a typo for imx6sx Use imx6sx instead of imx6sl's platform flags for imx6sx. Fixes: e14db48dfcf3 ("usb: chipidea: imx: add runtime power management support") Cc: # v4.1+ Signed-off-by: Li Jun Signed-off-by: Peter Chen --- drivers/usb/chipidea/ci_hdrc_imx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/chipidea/ci_hdrc_imx.c b/drivers/usb/chipidea/ci_hdrc_imx.c index 867e9f3f3859..dcc50c878159 100644 --- a/drivers/usb/chipidea/ci_hdrc_imx.c +++ b/drivers/usb/chipidea/ci_hdrc_imx.c @@ -61,7 +61,7 @@ static const struct of_device_id ci_hdrc_imx_dt_ids[] = { { .compatible = "fsl,imx27-usb", .data = &imx27_usb_data}, { .compatible = "fsl,imx6q-usb", .data = &imx6q_usb_data}, { .compatible = "fsl,imx6sl-usb", .data = &imx6sl_usb_data}, - { .compatible = "fsl,imx6sx-usb", .data = &imx6sl_usb_data}, + { .compatible = "fsl,imx6sx-usb", .data = &imx6sx_usb_data}, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, ci_hdrc_imx_dt_ids); From 385877c01361401113c101ef5a80a9f0998e072b Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Tue, 1 Sep 2015 09:52:21 +0800 Subject: [PATCH 139/647] ARM: pxa: balloon3: Fix build error irq_data_get_chip() function does not exist, call irq_desc_get_chip() instead. Fixes: 9ec97561aa6d ("ARM/pxa: Prepare balloon3_irq_handler for irq argument removal") Signed-off-by: Axel Lin Signed-off-by: Robert Jarzmik --- arch/arm/mach-pxa/balloon3.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-pxa/balloon3.c b/arch/arm/mach-pxa/balloon3.c index 70366b35d299..c1cda989c705 100644 --- a/arch/arm/mach-pxa/balloon3.c +++ b/arch/arm/mach-pxa/balloon3.c @@ -502,7 +502,7 @@ static void balloon3_irq_handler(unsigned int __irq, struct irq_desc *desc) balloon3_irq_enabled; do { struct irq_data *d = irq_desc_get_irq_data(desc); - struct irq_chip *chip = irq_data_get_chip(d); + struct irq_chip *chip = irq_desc_get_chip(desc); unsigned int irq; /* clear useless edge notification */ From 0243ed44ad4a25dbd2e92ad97e5e12a1a6c72d6c Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Tue, 15 Sep 2015 04:59:21 -0700 Subject: [PATCH 140/647] spi: fix kernel-doc warnings in spi.h Fix the following 'make htmldocs' warnings: .//include/linux/spi/spi.h:71: warning: No description found for parameter 'lock' .//include/linux/spi/spi.h:71: warning: Excess struct/union/enum/typedef member 'clock' description in 'spi_statistics' Signed-off-by: Geliang Tang Signed-off-by: Mark Brown --- include/linux/spi/spi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index 269e8afd3e2a..6b00f18f5e6b 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -34,7 +34,7 @@ extern struct bus_type spi_bus_type; /** * struct spi_statistics - statistics for spi transfers - * @clock: lock protecting this structure + * @lock: lock protecting this structure * * @messages: number of spi-messages handled * @transfers: number of spi_transfers handled From e26d15f735f570a1178c3bba4d85e6f58e098fdd Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Tue, 15 Sep 2015 14:46:45 +0200 Subject: [PATCH 141/647] spi: mediatek: fix wrong error return value on probe Commit adcbcfea15d62 ("spi: mediatek: fix spi clock usage error") added a new sel_clk but introduced bugs in the error paths since the wrong struct clk pointers are passed to PTR_ERR(). Fixes: adcbcfea15d62 ("spi: mediatek: fix spi clock usage error") Signed-off-by: Javier Martinez Canillas Signed-off-by: Mark Brown --- drivers/spi/spi-mt65xx.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/spi/spi-mt65xx.c b/drivers/spi/spi-mt65xx.c index 6fbb5e5132b0..e9839a4e0175 100644 --- a/drivers/spi/spi-mt65xx.c +++ b/drivers/spi/spi-mt65xx.c @@ -585,14 +585,14 @@ static int mtk_spi_probe(struct platform_device *pdev) mdata->sel_clk = devm_clk_get(&pdev->dev, "sel-clk"); if (IS_ERR(mdata->sel_clk)) { - ret = PTR_ERR(mdata->spi_clk); + ret = PTR_ERR(mdata->sel_clk); dev_err(&pdev->dev, "failed to get sel-clk: %d\n", ret); goto err_put_master; } mdata->spi_clk = devm_clk_get(&pdev->dev, "spi-clk"); if (IS_ERR(mdata->spi_clk)) { - ret = PTR_ERR(mdata->parent_clk); + ret = PTR_ERR(mdata->spi_clk); dev_err(&pdev->dev, "failed to get spi-clk: %d\n", ret); goto err_put_master; } From 3c8f7710c1c44fb650bc29b6ef78ed8b60cfaa28 Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Tue, 15 Sep 2015 20:51:31 +0200 Subject: [PATCH 142/647] ASoC: fix broken pxa SoC support The previous fix of pxa library support, which was introduced to fix the library dependency, broke the previous SoC behavior, where a machine code binding pxa2xx-ac97 with a coded relied on : - sound/soc/pxa/pxa2xx-ac97.c - sound/soc/codecs/XXX.c For example, the mioa701_wm9713.c machine code is currently broken. The "select ARM" statement wrongly selects the soc/arm/pxa2xx-ac97 for compilation, as per an unfortunate fate SND_PXA2XX_AC97 is both declared in sound/arm/Kconfig and sound/soc/pxa/Kconfig. Fix this by ensuring that SND_PXA2XX_SOC correctly triggers the correct pxa2xx-ac97 compilation. Fixes: 846172dfe33c ("ASoC: fix SND_PXA2XX_LIB Kconfig warning") Signed-off-by: Robert Jarzmik Signed-off-by: Mark Brown --- sound/arm/Kconfig | 15 ++++++++------- sound/soc/pxa/Kconfig | 2 -- 2 files changed, 8 insertions(+), 9 deletions(-) diff --git a/sound/arm/Kconfig b/sound/arm/Kconfig index 885683a3b0bd..e0406211716b 100644 --- a/sound/arm/Kconfig +++ b/sound/arm/Kconfig @@ -9,6 +9,14 @@ menuconfig SND_ARM Drivers that are implemented on ASoC can be found in "ALSA for SoC audio support" section. +config SND_PXA2XX_LIB + tristate + select SND_AC97_CODEC if SND_PXA2XX_LIB_AC97 + select SND_DMAENGINE_PCM + +config SND_PXA2XX_LIB_AC97 + bool + if SND_ARM config SND_ARMAACI @@ -21,13 +29,6 @@ config SND_PXA2XX_PCM tristate select SND_PCM -config SND_PXA2XX_LIB - tristate - select SND_AC97_CODEC if SND_PXA2XX_LIB_AC97 - -config SND_PXA2XX_LIB_AC97 - bool - config SND_PXA2XX_AC97 tristate "AC97 driver for the Intel PXA2xx chip" depends on ARCH_PXA diff --git a/sound/soc/pxa/Kconfig b/sound/soc/pxa/Kconfig index 39cea80846c3..f2bf8661dd21 100644 --- a/sound/soc/pxa/Kconfig +++ b/sound/soc/pxa/Kconfig @@ -1,7 +1,6 @@ config SND_PXA2XX_SOC tristate "SoC Audio for the Intel PXA2xx chip" depends on ARCH_PXA - select SND_ARM select SND_PXA2XX_LIB help Say Y or M if you want to add support for codecs attached to @@ -25,7 +24,6 @@ config SND_PXA2XX_AC97 config SND_PXA2XX_SOC_AC97 tristate select AC97_BUS - select SND_ARM select SND_PXA2XX_LIB_AC97 select SND_SOC_AC97_BUS From 0f1d08dd6aaf7e9dd1039dce910c1f8ff58ee8c1 Mon Sep 17 00:00:00 2001 From: "Andrew F. Davis" Date: Tue, 15 Sep 2015 15:34:22 -0500 Subject: [PATCH 143/647] regulator: tps65218: Fix missing zero typo Add missing zero to value. This will be needed when range checking is implemented. Signed-off-by: Andrew F. Davis Acked-by: Dan Murphy Signed-off-by: Mark Brown --- drivers/regulator/tps65218-regulator.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/regulator/tps65218-regulator.c b/drivers/regulator/tps65218-regulator.c index 7f97223f95c5..a02c1b961039 100644 --- a/drivers/regulator/tps65218-regulator.c +++ b/drivers/regulator/tps65218-regulator.c @@ -73,7 +73,7 @@ static const struct regulator_linear_range dcdc4_ranges[] = { }; static struct tps_info tps65218_pmic_regs[] = { - TPS65218_INFO(DCDC1, "DCDC1", 850000, 167500), + TPS65218_INFO(DCDC1, "DCDC1", 850000, 1675000), TPS65218_INFO(DCDC2, "DCDC2", 850000, 1675000), TPS65218_INFO(DCDC3, "DCDC3", 900000, 3400000), TPS65218_INFO(DCDC4, "DCDC4", 1175000, 3400000), From 02d8dabc50f94353075f2f62b1047c1306e8bf92 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Thu, 3 Sep 2015 15:23:40 +0200 Subject: [PATCH 144/647] perf stat: Fix per-pkg event reporting bug Per-pkg events need to be captured once per processor socket. The code in check_per_pkg() ensures only one value per processor package is used. However there is a problem with this function in case the first CPU of the package does not measure anything for the per-pkg event, but other CPUs do. Consider the following: $ create cgroup FOO; echo $$ >FOO/tasks; taskset -c 1 noploop & $ perf stat -a -I 1000 -e intel_cqm/llc_occupancy/ -G FOO sleep 100 1.00000 Bytes intel_cqm/llc_occupancy/ FOO The reason for this is that CPU0 in the cgroup has nothing running on it. Yet check_per_plg() will mark socket0 as processed and no other event value will be considered for the socket. This patch fixes the problem by having check_per_pkg() only consider events which actually ran. Signed-off-by: Stephane Eranian Cc: Adrian Hunter Cc: Andi Kleen Cc: David Ahern Cc: Jiri Olsa Cc: Kan Liang Cc: Namhyung Kim Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1441286620-10117-1-git-send-email-eranian@google.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/stat.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index 415c359de465..2d065d065b67 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -196,7 +196,8 @@ static void zero_per_pkg(struct perf_evsel *counter) memset(counter->per_pkg_mask, 0, MAX_NR_CPUS); } -static int check_per_pkg(struct perf_evsel *counter, int cpu, bool *skip) +static int check_per_pkg(struct perf_evsel *counter, + struct perf_counts_values *vals, int cpu, bool *skip) { unsigned long *mask = counter->per_pkg_mask; struct cpu_map *cpus = perf_evsel__cpus(counter); @@ -218,6 +219,17 @@ static int check_per_pkg(struct perf_evsel *counter, int cpu, bool *skip) counter->per_pkg_mask = mask; } + /* + * we do not consider an event that has not run as a good + * instance to mark a package as used (skip=1). Otherwise + * we may run into a situation where the first CPU in a package + * is not running anything, yet the second is, and this function + * would mark the package as used after the first CPU and would + * not read the values from the second CPU. + */ + if (!(vals->run && vals->ena)) + return 0; + s = cpu_map__get_socket(cpus, cpu); if (s < 0) return -1; @@ -235,7 +247,7 @@ process_counter_values(struct perf_stat_config *config, struct perf_evsel *evsel static struct perf_counts_values zero; bool skip = false; - if (check_per_pkg(evsel, cpu, &skip)) { + if (check_per_pkg(evsel, count, cpu, &skip)) { pr_err("failed to read per-pkg counter\n"); return -1; } From 5005296e7408c055a02fc8be9f62218cb4b52b13 Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Thu, 3 Sep 2015 14:23:45 -0500 Subject: [PATCH 145/647] ARM: dts: am57xx-beagle-x15: Update Phy supplies Originally, all the SoC PHY rails were supplied by LDO3. However, as a result of characterization, it was determined that this posed a risk in extreme load conditions. Hence the PHY rails are split between two different LDOs. Update the related node as a result LDO3/VDDA_1V8_PHYA supplies vdda_usb1, vdda_usb2, vdda_sata, vdda_usb3 LDO4/VDDA_1V8_PHYB supplies vdda_pcie1, vdda_pcie0, vdda_hdmi, vdda_pcie NOTE: We break compatibility with pre-production boards with this change since, the PMIC LDO4 is disabled at OTP level. The new configuration is the plan of record and all pre-production boards are supposed to be replaced with the latest boards matching the mentioned configuration. Some very few 10 something boards have been created and stopped production till the latest modifications were done (PMIC USB interrupt, LDO4 etc) - and all of those boards are now getting scrapped.. If there are any (as per tracking information, there should not be any), TI should be contacted to have them replaced. Signed-off-by: Nishanth Menon [tony@atomide.com: updated commit about these being TI internal protos] Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/am57xx-beagle-x15.dts | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/arch/arm/boot/dts/am57xx-beagle-x15.dts b/arch/arm/boot/dts/am57xx-beagle-x15.dts index a4274abf1437..53884bc65abd 100644 --- a/arch/arm/boot/dts/am57xx-beagle-x15.dts +++ b/arch/arm/boot/dts/am57xx-beagle-x15.dts @@ -432,7 +432,7 @@ }; ldo3_reg: ldo3 { - /* VDDA_1V8_PHY */ + /* VDDA_1V8_PHYA */ regulator-name = "ldo3"; regulator-min-microvolt = <1800000>; regulator-max-microvolt = <1800000>; @@ -440,6 +440,15 @@ regulator-boot-on; }; + ldo4_reg: ldo4 { + /* VDDA_1V8_PHYB */ + regulator-name = "ldo4"; + regulator-min-microvolt = <1800000>; + regulator-max-microvolt = <1800000>; + regulator-always-on; + regulator-boot-on; + }; + ldo9_reg: ldo9 { /* VDD_RTC */ regulator-name = "ldo9"; @@ -681,7 +690,7 @@ &hdmi { status = "ok"; - vdda-supply = <&ldo3_reg>; + vdda-supply = <&ldo4_reg>; pinctrl-names = "default"; pinctrl-0 = <&hdmi_pins>; From 08224a7d410699abce42674ee15a7cf37d54f340 Mon Sep 17 00:00:00 2001 From: Dave Gerlach Date: Tue, 15 Sep 2015 14:47:34 -0500 Subject: [PATCH 146/647] ARM: OMAP2+: AM43XX: Enable autoidle for clks in am43xx_init_late Add omap2_clk_enable_autoidle_all to am43xx_init_late otherwise the call to omap2_clk_disable_autoidle_all in am43xx_init_early may cause some clocks to always stay active and prevent low power mode transitions. Signed-off-by: Dave Gerlach Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/io.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/mach-omap2/io.c b/arch/arm/mach-omap2/io.c index 980c9372e6fd..3eaeaca5da05 100644 --- a/arch/arm/mach-omap2/io.c +++ b/arch/arm/mach-omap2/io.c @@ -676,6 +676,7 @@ void __init am43xx_init_early(void) void __init am43xx_init_late(void) { omap_common_late_init(); + omap2_clk_enable_autoidle_all(); } #endif From 1dbdad75074d16c3e3005180f81a01cdc04a7872 Mon Sep 17 00:00:00 2001 From: Grazvydas Ignotas Date: Wed, 16 Sep 2015 01:34:31 +0300 Subject: [PATCH 147/647] ARM: dts: omap5-uevm.dts: fix i2c5 pinctrl offsets The i2c5 pinctrl offsets are wrong. If the bootloader doesn't set the pins up, communication with tca6424a doesn't work (controller timeouts) and it is not possible to enable HDMI. Fixes: 9be495c42609 ("ARM: dts: omap5-evm: Add I2c pinctrl data") Signed-off-by: Grazvydas Ignotas Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/omap5-uevm.dts | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/boot/dts/omap5-uevm.dts b/arch/arm/boot/dts/omap5-uevm.dts index 3cc8f357d5b8..3cb030f9d2c4 100644 --- a/arch/arm/boot/dts/omap5-uevm.dts +++ b/arch/arm/boot/dts/omap5-uevm.dts @@ -174,8 +174,8 @@ i2c5_pins: pinmux_i2c5_pins { pinctrl-single,pins = < - 0x184 (PIN_INPUT | MUX_MODE0) /* i2c5_scl */ - 0x186 (PIN_INPUT | MUX_MODE0) /* i2c5_sda */ + 0x186 (PIN_INPUT | MUX_MODE0) /* i2c5_scl */ + 0x188 (PIN_INPUT | MUX_MODE0) /* i2c5_sda */ >; }; From f4746176f95a0d06f07ea83191375d374611e170 Mon Sep 17 00:00:00 2001 From: Grazvydas Ignotas Date: Wed, 16 Sep 2015 01:36:20 +0300 Subject: [PATCH 148/647] ARM: omap2plus_defconfig: enable GPIO_PCA953X This enables tca6424a GPIO expander driver that in turn enables TPD12S015 HDMI ESD protection and level shifter on OMAP5 uevm. In other words, it makes HDMI work on OMAP5 uevm. Signed-off-by: Grazvydas Ignotas Signed-off-by: Tony Lindgren --- arch/arm/configs/omap2plus_defconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig index ffc8fa2565f9..3f15a5cae167 100644 --- a/arch/arm/configs/omap2plus_defconfig +++ b/arch/arm/configs/omap2plus_defconfig @@ -240,6 +240,7 @@ CONFIG_SSI_PROTOCOL=m CONFIG_PINCTRL_SINGLE=y CONFIG_DEBUG_GPIO=y CONFIG_GPIO_SYSFS=y +CONFIG_GPIO_PCA953X=m CONFIG_GPIO_PCF857X=y CONFIG_GPIO_TWL4030=y CONFIG_GPIO_PALMAS=y From 84ad1bab0eebd2e2b2dd3c3db7b88cade22a6de1 Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Mon, 27 Jul 2015 16:10:17 +0300 Subject: [PATCH 149/647] ARM: dts: am57xx-beagle-x15: use palmas-usb for USB2 The VBUS line of USB2 is connected to VBUS detect logic on the PMIC. Use the palmas-usb driver to report VBUS events to the USB driver. As the palmas-usb driver supports GPIO based ID reporting provide the GPIO for ID pin as well. Signed-off-by: Roger Quadros Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/am57xx-beagle-x15.dts | 29 ++++++++++++++----------- 1 file changed, 16 insertions(+), 13 deletions(-) diff --git a/arch/arm/boot/dts/am57xx-beagle-x15.dts b/arch/arm/boot/dts/am57xx-beagle-x15.dts index 53884bc65abd..568adf5efde0 100644 --- a/arch/arm/boot/dts/am57xx-beagle-x15.dts +++ b/arch/arm/boot/dts/am57xx-beagle-x15.dts @@ -98,13 +98,6 @@ pinctrl-0 = <&extcon_usb1_pins>; }; - extcon_usb2: extcon_usb2 { - compatible = "linux,extcon-usb-gpio"; - id-gpio = <&gpio7 24 GPIO_ACTIVE_HIGH>; - pinctrl-names = "default"; - pinctrl-0 = <&extcon_usb2_pins>; - }; - hdmi0: connector { compatible = "hdmi-connector"; label = "hdmi"; @@ -326,12 +319,6 @@ >; }; - extcon_usb2_pins: extcon_usb2_pins { - pinctrl-single,pins = < - 0x3e8 (PIN_INPUT_PULLUP | MUX_MODE14) /* uart1_ctsn.gpio7_24 */ - >; - }; - tpd12s015_pins: pinmux_tpd12s015_pins { pinctrl-single,pins = < 0x3b0 (PIN_OUTPUT | MUX_MODE14) /* gpio7_10 CT_CP_HPD */ @@ -504,6 +491,14 @@ gpio-controller; #gpio-cells = <2>; }; + + extcon_usb2: tps659038_usb { + compatible = "ti,palmas-usb-vid"; + ti,enable-vbus-detection; + ti,enable-id-detection; + id-gpios = <&gpio7 24 GPIO_ACTIVE_HIGH>; + }; + }; tmp102: tmp102@48 { @@ -632,6 +627,14 @@ }; &usb2 { + /* + * Stand alone usage is peripheral only. + * However, with some resistor modifications + * this port can be used via expansion connectors + * as "host" or "dual-role". If so, provide + * the necessary dr_mode override in the expansion + * board's DT. + */ dr_mode = "peripheral"; }; From b07bb761cf6a333b2108cdbe4ffee66407189bb1 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Wed, 16 Sep 2015 08:38:08 -0700 Subject: [PATCH 150/647] drm/vmwgfx: Only build on X86 ioremap_cache() is currently not available on some architectures. Reported-by: kbuild test robot Signed-off-by: Thomas Hellstrom --- drivers/gpu/drm/vmwgfx/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vmwgfx/Kconfig b/drivers/gpu/drm/vmwgfx/Kconfig index 67720f70fe29..b49445df8a7e 100644 --- a/drivers/gpu/drm/vmwgfx/Kconfig +++ b/drivers/gpu/drm/vmwgfx/Kconfig @@ -1,6 +1,6 @@ config DRM_VMWGFX tristate "DRM driver for VMware Virtual GPU" - depends on DRM && PCI + depends on DRM && PCI && X86 select FB_DEFERRED_IO select FB_CFB_FILLRECT select FB_CFB_COPYAREA From d8949aad3eab5d396f4fefcd581773bf07b9a79e Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Fri, 4 Sep 2015 12:22:46 +0300 Subject: [PATCH 151/647] Bluetooth: Delay check for conn->smp in smp_conn_security() There are several actions that smp_conn_security() might make that do not require a valid SMP context (conn->smp pointer). One of these actions is to encrypt the link with an existing LTK. If the SMP context wasn't initialized properly we should still allow the independent actions to be done, i.e. the check for the context should only be done at the last possible moment. Reported-by: Chuck Ebbert Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann Cc: stable@vger.kernel.org # 4.0+ --- net/bluetooth/smp.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/net/bluetooth/smp.c b/net/bluetooth/smp.c index ad82324f710f..0510a577a7b5 100644 --- a/net/bluetooth/smp.c +++ b/net/bluetooth/smp.c @@ -2311,12 +2311,6 @@ int smp_conn_security(struct hci_conn *hcon, __u8 sec_level) if (!conn) return 1; - chan = conn->smp; - if (!chan) { - BT_ERR("SMP security requested but not available"); - return 1; - } - if (!hci_dev_test_flag(hcon->hdev, HCI_LE_ENABLED)) return 1; @@ -2330,6 +2324,12 @@ int smp_conn_security(struct hci_conn *hcon, __u8 sec_level) if (smp_ltk_encrypt(conn, hcon->pending_sec_level)) return 0; + chan = conn->smp; + if (!chan) { + BT_ERR("SMP security requested but not available"); + return 1; + } + l2cap_chan_lock(chan); /* If SMP is already in progress ignore this request */ From ad5001cc7cdf9aaee5eb213fdee657e4a3c94776 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 17 Sep 2015 13:37:00 +0200 Subject: [PATCH 152/647] netfilter: nf_log: wait for rcu grace after logger unregistration The nf_log_unregister() function needs to call synchronize_rcu() to make sure that the objects are not dereferenced anymore on module removal. Fixes: 5962815a6a56 ("netfilter: nf_log: use an array of loggers instead of list") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_log.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index a5ebd7d9c472..a5d41dfa9f05 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -117,6 +117,7 @@ void nf_log_unregister(struct nf_logger *logger) RCU_INIT_POINTER(loggers[i][logger->type], NULL); } mutex_unlock(&nf_log_mutex); + synchronize_rcu(); } EXPORT_SYMBOL(nf_log_unregister); From 2110d70c5e58326a10e93cfefdc0b3686e2ada12 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Sat, 8 Aug 2015 10:46:02 +0200 Subject: [PATCH 153/647] cpu/cacheinfo: Fix teardown path MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Philip Müller reported a hang when booting 32-bit 4.1 kernel on an AMD box. A fragment of the splat was enough to pinpoint the issue: task: f58e0000 ti: f58e8000 task.ti: f58e800 EIP: 0060:[] EFLAGS: 00010206 CPU: 0 EIP is at free_cache_attributes+0x83/0xd0 EAX: 00000001 EBX: f589d46c ECX: 00000090 EDX: 360c2000 ESI: 00000000 EDI: c1724a80 EBP: f58e9ec0 ESP: f58e9ea0 DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 CR0: 8005003b CR2: 000000ac CR3: 01731000 CR4: 000006d0 cache_shared_cpu_map_setup() did check sibling CPUs cacheinfo descriptor while the respective teardown path cache_shared_cpu_map_remove() didn't. Fix that. >From tglx's version: to be on the safe side, move the cacheinfo descriptor check to free_cache_attributes(), thus cleaning up the hotplug path a little and making this even more robust. Reported-and-tested-by: Philip Müller Reviewed-by: Thomas Gleixner Acked-by: Sudeep Holla Cc: # 4.1 Cc: Andre Przywara Cc: Guenter Roeck Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: linux-kernel@vger.kernel.org Cc: manjaro-dev@manjaro.org Cc: Philip Müller Link: https://lkml.kernel.org/r/55B47BB8.6080202@manjaro.org Signed-off-by: Borislav Petkov Signed-off-by: Greg Kroah-Hartman --- drivers/base/cacheinfo.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/base/cacheinfo.c b/drivers/base/cacheinfo.c index 764280a91776..e9fd32e91668 100644 --- a/drivers/base/cacheinfo.c +++ b/drivers/base/cacheinfo.c @@ -148,7 +148,11 @@ static void cache_shared_cpu_map_remove(unsigned int cpu) if (sibling == cpu) /* skip itself */ continue; + sib_cpu_ci = get_cpu_cacheinfo(sibling); + if (!sib_cpu_ci->info_list) + continue; + sib_leaf = sib_cpu_ci->info_list + index; cpumask_clear_cpu(cpu, &sib_leaf->shared_cpu_map); cpumask_clear_cpu(sibling, &this_leaf->shared_cpu_map); @@ -159,6 +163,9 @@ static void cache_shared_cpu_map_remove(unsigned int cpu) static void free_cache_attributes(unsigned int cpu) { + if (!per_cpu_cacheinfo(cpu)) + return; + cache_shared_cpu_map_remove(cpu); kfree(per_cpu_cacheinfo(cpu)); @@ -514,8 +521,7 @@ static int cacheinfo_cpu_callback(struct notifier_block *nfb, break; case CPU_DEAD: cache_remove_dev(cpu); - if (per_cpu_cacheinfo(cpu)) - free_cache_attributes(cpu); + free_cache_attributes(cpu); break; } return notifier_from_errno(rc); From f5ab93fa5e79bacd83f0866ad5c3950eee279b66 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Wed, 16 Sep 2015 22:16:38 +0530 Subject: [PATCH 154/647] staging: unisys: remove reference of visorutil commit 53490b545cb0 ("staging: unisys: move periodic_work.c into the visorbus directory") has removed the visorutil directory but missed removing the reference in the Makefile. Fixes: 53490b545cb0 ("staging: unisys: move periodic_work.c into the visorbus directory") Signed-off-by: Sudip Mukherjee Signed-off-by: Greg Kroah-Hartman --- drivers/staging/unisys/visorbus/Makefile | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/staging/unisys/visorbus/Makefile b/drivers/staging/unisys/visorbus/Makefile index fa27ee5f336c..fc790e7592fc 100644 --- a/drivers/staging/unisys/visorbus/Makefile +++ b/drivers/staging/unisys/visorbus/Makefile @@ -10,4 +10,3 @@ visorbus-y += visorchipset.o visorbus-y += periodic_work.o ccflags-y += -Idrivers/staging/unisys/include -ccflags-y += -Idrivers/staging/unisys/visorutil From 981473c74c0930190aa4651efebb9bb1f784416e Mon Sep 17 00:00:00 2001 From: Christian Gromm Date: Wed, 16 Sep 2015 17:33:35 +0200 Subject: [PATCH 155/647] staging: most: Add dependency to HAS_IOMEM This patch prevents the module hdm_dim2 from breaking the build in case HAS_IOMEM is not configured. Reported-by: Signed-off-by: Christian Gromm Signed-off-by: Greg Kroah-Hartman --- drivers/staging/most/hdm-dim2/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/staging/most/hdm-dim2/Kconfig b/drivers/staging/most/hdm-dim2/Kconfig index 1d4ad1d67758..fc548769479b 100644 --- a/drivers/staging/most/hdm-dim2/Kconfig +++ b/drivers/staging/most/hdm-dim2/Kconfig @@ -5,6 +5,7 @@ config HDM_DIM2 tristate "DIM2 HDM" depends on AIM_NETWORK + depends on HAS_IOMEM ---help--- Say Y here if you want to connect via MediaLB to network transceiver. From 989e503c929d82f34a3cae51e98cbc2745e7472c Mon Sep 17 00:00:00 2001 From: Mark Hounschell Date: Tue, 15 Sep 2015 08:56:10 -0400 Subject: [PATCH 156/647] staging: dgap: Remove myself from the MAINTAINERS file Removes myself from the MAINTAINERS file for the dgap driver. There appears to be no way to get the firmware files required by the dgap driver into the linux-firmware tree. The dgap driver is useless wihtout this firmware. This product is considered an obsolete product by Digi. They will not respond to an inquiry concerning it or its firmware. Signed-off-by: Mark Hounschell Cc: Greg Kroah-Hartman Cc: Ben Hutchings Signed-off-by: Greg Kroah-Hartman --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index c4c7804fb8e3..72fc3936b423 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3394,7 +3394,6 @@ F: drivers/staging/dgnc/ DIGI EPCA PCI PRODUCTS M: Lidza Louina -M: Mark Hounschell M: Daeseok Youn L: driverdev-devel@linuxdriverproject.org S: Maintained From 6fa92e2bcf6390e64895b12761e851c452d87bd8 Mon Sep 17 00:00:00 2001 From: Shawn Lin Date: Wed, 9 Sep 2015 15:41:52 +0800 Subject: [PATCH 157/647] staging: ion: fix corruption of ion_import_dma_buf we found this issue but still exit in lastest kernel. Simply keep ion_handle_create under mutex_lock to avoid this race. WARNING: CPU: 2 PID: 2648 at drivers/staging/android/ion/ion.c:512 ion_handle_add+0xb4/0xc0() ion_handle_add: buffer already found. Modules linked in: iwlmvm iwlwifi mac80211 cfg80211 compat CPU: 2 PID: 2648 Comm: TimedEventQueue Tainted: G W 3.14.0 #7 00000000 00000000 9a3efd2c 80faf273 9a3efd6c 9a3efd5c 80935dc9 811d7fd3 9a3efd88 00000a58 812208a0 00000200 80e128d4 80e128d4 8d4ae00c a8cd8600 a8cd8094 9a3efd74 80935e0e 00000009 9a3efd6c 811d7fd3 9a3efd88 9a3efd9c Call Trace: [<80faf273>] dump_stack+0x48/0x69 [<80935dc9>] warn_slowpath_common+0x79/0x90 [<80e128d4>] ? ion_handle_add+0xb4/0xc0 [<80e128d4>] ? ion_handle_add+0xb4/0xc0 [<80935e0e>] warn_slowpath_fmt+0x2e/0x30 [<80e128d4>] ion_handle_add+0xb4/0xc0 [<80e144cc>] ion_import_dma_buf+0x8c/0x110 [<80c517c4>] reg_init+0x364/0x7d0 [<80993363>] ? futex_wait+0x123/0x210 [<80992e0e>] ? get_futex_key+0x16e/0x1e0 [<8099308f>] ? futex_wake+0x5f/0x120 [<80c51e19>] vpu_service_ioctl+0x1e9/0x500 [<80994aec>] ? do_futex+0xec/0x8e0 [<80971080>] ? prepare_to_wait_event+0xc0/0xc0 [<80c51c30>] ? reg_init+0x7d0/0x7d0 [<80a22562>] do_vfs_ioctl+0x2d2/0x4c0 [<80b198ad>] ? inode_has_perm.isra.41+0x2d/0x40 [<80b199cf>] ? file_has_perm+0x7f/0x90 [<80b1a5f7>] ? selinux_file_ioctl+0x47/0xf0 [<80a227a8>] SyS_ioctl+0x58/0x80 [<80fb45e8>] syscall_call+0x7/0x7 [<80fb0000>] ? mmc_do_calc_max_discard+0xab/0xe4 Fixes: 83271f626 ("ion: hold reference to handle...") Signed-off-by: Shawn Lin Reviewed-by: Laura Abbott Cc: stable # 3.14+ Signed-off-by: Greg Kroah-Hartman --- drivers/staging/android/ion/ion.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/staging/android/ion/ion.c b/drivers/staging/android/ion/ion.c index 217aa537c4eb..6e8d8392ca38 100644 --- a/drivers/staging/android/ion/ion.c +++ b/drivers/staging/android/ion/ion.c @@ -1179,13 +1179,13 @@ struct ion_handle *ion_import_dma_buf(struct ion_client *client, int fd) mutex_unlock(&client->lock); goto end; } - mutex_unlock(&client->lock); handle = ion_handle_create(client, buffer); - if (IS_ERR(handle)) + if (IS_ERR(handle)) { + mutex_unlock(&client->lock); goto end; + } - mutex_lock(&client->lock); ret = ion_handle_add(client, handle); mutex_unlock(&client->lock); if (ret) { From 179f36dde3cec0f9f05a757b68f6a58e4edbcc95 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 17 Sep 2015 11:30:20 -0300 Subject: [PATCH 158/647] Revert "perf symbols: Fix mismatched declarations for elf_getphdrnum" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit f785f2357673d520a0b7b468973cdd197f336494. We have a test to check if elf_getphdrnum() is present, so, if it fails, we'll get: [acme@rhel5 linux]$ cat /tmp/build/perf/feature/test-libelf-getphdrnum.make.output cc1: warnings being treated as errors test-libelf-getphdrnum.c: In function ‘main’: test-libelf-getphdrnum.c:7: warning: implicit declaration of function ‘elf_getphdrnum’ [acme@rhel5 linux]$ And this block will not be compiled: #ifndef HAVE_ELF_GETPHDRNUM_SUPPORT static int elf_getphdrnum(Elf *elf, size_t *dst) ... #endif So, if elf_getphdrnum() is being defined somewhere, there is a problem with the test that is not detecting that function, go fix it. Reported-by: Vinson Lee Cc: Adrian Hunter Cc: Borislav Petkov Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Namhyung Kim Cc: "Naveen N. Rao" Cc: Peter Zijlstra Cc: Srikar Dronamraju Cc: Stephane Eranian Cc: Victor Kamensky Cc: Wang Nan Link: http://lkml.kernel.org/n/tip-qn459fal6acvcvm50i8zxx9k@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/symbol-elf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index 53bb5f59ec58..f78ea3dc4c08 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -38,7 +38,7 @@ static inline char *bfd_demangle(void __maybe_unused *v, #endif #ifndef HAVE_ELF_GETPHDRNUM_SUPPORT -int elf_getphdrnum(Elf *elf, size_t *dst) +static int elf_getphdrnum(Elf *elf, size_t *dst) { GElf_Ehdr gehdr; GElf_Ehdr *ehdr; From f8ac8606fd3cd72183de8eec2b151ff05040c70f Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 17 Sep 2015 12:20:28 -0300 Subject: [PATCH 159/647] tools build: Add test for presence of numa_num_possible_cpus() in libnuma The existing numa test checks only if numa.h and numa_available() are present, but that can be satisfied with an old libnuma that is not enough for the 'perf bench numa' entry, so add a test to check for that: [acme@rhel5 linux]$ make NO_AUXTRACE=1 NO_LIBPERL=1 -C tools/perf O=/tmp/build/perf install-bin make: Entering directory `/home/acme/git/linux/tools/perf' BUILD: Doing 'make -j2' parallel build Auto-detecting system features: ... libelf: [ on ] ... libnuma: [ on ] ... numa_num_possible_cpus: [ OFF ] ... libperl: [ on ] config/Makefile:577: Old numa library found, disables 'perf bench numa mem' benchmark, please install numactl-devel/libnuma-devel/libnuma-dev >= 2.0.8 INSTALL binaries This fixes the build on old systems such as RHEL/CentOS 5.11. Cc: Adrian Hunter Cc: Borislav Petkov Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Namhyung Kim Cc: "Naveen N. Rao" Cc: Peter Zijlstra Cc: Srikar Dronamraju Cc: Stephane Eranian Cc: Victor Kamensky Cc: Vinson Lee Cc: Wang Nan Link: http://lkml.kernel.org/n/tip-zqriqkezppi2de2iyjin1tnc@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/build/Makefile.feature | 2 ++ tools/build/feature/Makefile | 4 ++++ tools/build/feature/test-all.c | 5 +++++ tools/build/feature/test-numa_num_possible_cpus.c | 6 ++++++ tools/perf/config/Makefile | 11 ++++++++--- 5 files changed, 25 insertions(+), 3 deletions(-) create mode 100644 tools/build/feature/test-numa_num_possible_cpus.c diff --git a/tools/build/Makefile.feature b/tools/build/Makefile.feature index 2975632d51e2..970242098a7c 100644 --- a/tools/build/Makefile.feature +++ b/tools/build/Makefile.feature @@ -41,6 +41,7 @@ FEATURE_TESTS ?= \ libelf-getphdrnum \ libelf-mmap \ libnuma \ + numa_num_possible_cpus \ libperl \ libpython \ libpython-version \ @@ -61,6 +62,7 @@ FEATURE_DISPLAY ?= \ libbfd \ libelf \ libnuma \ + numa_num_possible_cpus \ libperl \ libpython \ libslang \ diff --git a/tools/build/feature/Makefile b/tools/build/feature/Makefile index 74ca42093d70..e13a42bd0274 100644 --- a/tools/build/feature/Makefile +++ b/tools/build/feature/Makefile @@ -19,6 +19,7 @@ FILES= \ test-libelf-getphdrnum.bin \ test-libelf-mmap.bin \ test-libnuma.bin \ + test-numa_num_possible_cpus.bin \ test-libperl.bin \ test-libpython.bin \ test-libpython-version.bin \ @@ -87,6 +88,9 @@ test-libelf-getphdrnum.bin: test-libnuma.bin: $(BUILD) -lnuma +test-numa_num_possible_cpus.bin: + $(BUILD) -lnuma + test-libunwind.bin: $(BUILD) -lelf diff --git a/tools/build/feature/test-all.c b/tools/build/feature/test-all.c index 84689a67814a..7a8cdbad3e1b 100644 --- a/tools/build/feature/test-all.c +++ b/tools/build/feature/test-all.c @@ -77,6 +77,10 @@ # include "test-libnuma.c" #undef main +#define main main_test_numa_num_possible_cpus +# include "test-numa_num_possible_cpus.c" +#undef main + #define main main_test_timerfd # include "test-timerfd.c" #undef main @@ -136,6 +140,7 @@ int main(int argc, char *argv[]) main_test_libbfd(); main_test_backtrace(); main_test_libnuma(); + main_test_numa_num_possible_cpus(); main_test_timerfd(); main_test_stackprotector_all(); main_test_libdw_dwarf_unwind(); diff --git a/tools/build/feature/test-numa_num_possible_cpus.c b/tools/build/feature/test-numa_num_possible_cpus.c new file mode 100644 index 000000000000..2606e94b0659 --- /dev/null +++ b/tools/build/feature/test-numa_num_possible_cpus.c @@ -0,0 +1,6 @@ +#include + +int main(void) +{ + return numa_num_possible_cpus(); +} diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile index 827557fc7511..053e65b04dc3 100644 --- a/tools/perf/config/Makefile +++ b/tools/perf/config/Makefile @@ -573,9 +573,14 @@ ifndef NO_LIBNUMA msg := $(warning No numa.h found, disables 'perf bench numa mem' benchmark, please install numactl-devel/libnuma-devel/libnuma-dev); NO_LIBNUMA := 1 else - CFLAGS += -DHAVE_LIBNUMA_SUPPORT - EXTLIBS += -lnuma - $(call detected,CONFIG_NUMA) + ifeq ($(feature-numa_num_possible_cpus), 0) + msg := $(warning Old numa library found, disables 'perf bench numa mem' benchmark, please install numactl-devel/libnuma-devel/libnuma-dev >= 2.0.8); + NO_LIBNUMA := 1 + else + CFLAGS += -DHAVE_LIBNUMA_SUPPORT + EXTLIBS += -lnuma + $(call detected,CONFIG_NUMA) + endif endif endif From b0063dbfb031a7c728ed0d9533257e8329292cf1 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 17 Sep 2015 12:54:30 -0300 Subject: [PATCH 160/647] tools build: Add test for presence of __get_cpuid() gcc builtin The auxtrace code needed by Intel PT uses the __get_cpuid() gcc builtin, that is not present in old systems, breaking the build. Add a test to check for that builtin and disable AUXTRACE in those systems. [acme@rhel5 linux]$ make NO_LIBPERL=1 -C tools/perf O=/tmp/build/perf install-bin make: Entering directory `/home/acme/git/linux/tools/perf' BUILD: Doing 'make -j2' parallel build Auto-detecting system features: ... lzma: [ on ] ... get_cpuid: [ OFF ] config/Makefile:630: Your gcc lacks the __get_cpuid() builtin, disables support for auxtrace/Intel PT, please install a newer gcc MKDIR /tmp/build/perf/util/ This fixes the build on old systems such as RHEL/CentOS 5.11. Cc: Adrian Hunter Cc: Borislav Petkov Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Namhyung Kim Cc: "Naveen N. Rao" Cc: Peter Zijlstra Cc: Srikar Dronamraju Cc: Stephane Eranian Cc: Victor Kamensky Cc: Vinson Lee Cc: Wang Nan Link: http://lkml.kernel.org/n/tip-d4puslul0jltoodzpx9r4sje@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/build/Makefile.feature | 6 ++++-- tools/build/feature/Makefile | 6 +++++- tools/build/feature/test-all.c | 5 +++++ tools/build/feature/test-get_cpuid.c | 7 +++++++ tools/perf/config/Makefile | 9 +++++++-- 5 files changed, 28 insertions(+), 5 deletions(-) create mode 100644 tools/build/feature/test-get_cpuid.c diff --git a/tools/build/Makefile.feature b/tools/build/Makefile.feature index 970242098a7c..c8fe6d177119 100644 --- a/tools/build/Makefile.feature +++ b/tools/build/Makefile.feature @@ -52,7 +52,8 @@ FEATURE_TESTS ?= \ timerfd \ libdw-dwarf-unwind \ zlib \ - lzma + lzma \ + get_cpuid FEATURE_DISPLAY ?= \ dwarf \ @@ -69,7 +70,8 @@ FEATURE_DISPLAY ?= \ libunwind \ libdw-dwarf-unwind \ zlib \ - lzma + lzma \ + get_cpuid # Set FEATURE_CHECK_(C|LD)FLAGS-all for all FEATURE_TESTS features. # If in the future we need per-feature checks/flags for features not diff --git a/tools/build/feature/Makefile b/tools/build/feature/Makefile index e13a42bd0274..e43a2971bf56 100644 --- a/tools/build/feature/Makefile +++ b/tools/build/feature/Makefile @@ -35,7 +35,8 @@ FILES= \ test-compile-x32.bin \ test-zlib.bin \ test-lzma.bin \ - test-bpf.bin + test-bpf.bin \ + test-get_cpuid.bin CC := $(CROSS_COMPILE)gcc -MD PKG_CONFIG := $(CROSS_COMPILE)pkg-config @@ -166,6 +167,9 @@ test-zlib.bin: test-lzma.bin: $(BUILD) -llzma +test-get_cpuid.bin: + $(BUILD) + test-bpf.bin: $(BUILD) diff --git a/tools/build/feature/test-all.c b/tools/build/feature/test-all.c index 7a8cdbad3e1b..33cf6f20bd4e 100644 --- a/tools/build/feature/test-all.c +++ b/tools/build/feature/test-all.c @@ -121,6 +121,10 @@ # include "test-lzma.c" #undef main +#define main main_test_get_cpuid +# include "test-get_cpuid.c" +#undef main + int main(int argc, char *argv[]) { main_test_libpython(); @@ -148,6 +152,7 @@ int main(int argc, char *argv[]) main_test_zlib(); main_test_pthread_attr_setaffinity_np(); main_test_lzma(); + main_test_get_cpuid(); return 0; } diff --git a/tools/build/feature/test-get_cpuid.c b/tools/build/feature/test-get_cpuid.c new file mode 100644 index 000000000000..d7a2c407130d --- /dev/null +++ b/tools/build/feature/test-get_cpuid.c @@ -0,0 +1,7 @@ +#include + +int main(void) +{ + unsigned int eax = 0, ebx = 0, ecx = 0, edx = 0; + return __get_cpuid(0x15, &eax, &ebx, &ecx, &edx); +} diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile index 053e65b04dc3..38a08539f4bf 100644 --- a/tools/perf/config/Makefile +++ b/tools/perf/config/Makefile @@ -626,8 +626,13 @@ ifdef LIBBABELTRACE endif ifndef NO_AUXTRACE - $(call detected,CONFIG_AUXTRACE) - CFLAGS += -DHAVE_AUXTRACE_SUPPORT + ifeq ($(feature-get_cpuid), 0) + msg := $(warning Your gcc lacks the __get_cpuid() builtin, disables support for auxtrace/Intel PT, please install a newer gcc); + NO_AUXTRACE := 1 + else + $(call detected,CONFIG_AUXTRACE) + CFLAGS += -DHAVE_AUXTRACE_SUPPORT + endif endif # Among the variables below, these: From 23c3f310e897837aeb8ffe8700b803cb58e7b35d Mon Sep 17 00:00:00 2001 From: Charles Keepax Date: Thu, 17 Sep 2015 14:50:20 +0100 Subject: [PATCH 161/647] regulator: core: Correct return value check in regulator_resolve_supply The ret pointer passed to regulator_dev_lookup is only filled with a valid error code if regulator_dev_lookup returned NULL. Currently regulator_resolve_supply checks this ret value before it checks if a regulator was returned, this can result in valid regulator lookups being ignored. Fixes: 6261b06de565 ("regulator: Defer lookup of supply to regulator_get") Signed-off-by: Charles Keepax Signed-off-by: Mark Brown Cc: stable@vger.kernel.org --- drivers/regulator/core.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index 7150ff6ef46b..d0fbaea9bf54 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -1394,15 +1394,15 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) return 0; r = regulator_dev_lookup(dev, rdev->supply_name, &ret); - if (ret == -ENODEV) { - /* - * No supply was specified for this regulator and - * there will never be one. - */ - return 0; - } - if (!r) { + if (ret == -ENODEV) { + /* + * No supply was specified for this regulator and + * there will never be one. + */ + return 0; + } + if (have_full_constraints()) { r = dummy_regulator_rdev; } else { From bf6445631c6f00882b25516a174d5073ce0c6f81 Mon Sep 17 00:00:00 2001 From: Peter Senna Tschudin Date: Thu, 17 Sep 2015 12:08:53 +0200 Subject: [PATCH 162/647] perf tools: Bool functions shouldn't return -1 Returning a negative value for a boolean function seem to have the undesired effect of returning true. Replace -1 by false in a bool-returning function. The diff of the .s file before and after the change (for x86_64): 3907c3907 < movl $1, %ebx --- > xorl %ebx, %ebx while if -1 is replaced by true, the diff is empty. This issue was found by the following Coccinelle semantic patch: @@ identifier f; constant C; typedef bool; @@ bool f (...){ <+... * return -C; ...+> } Signed-off-by: Peter Senna Tschudin Cc: Jiri Olsa Cc: Kan Liang Cc: Matt Fleming Cc: Milos Vyletel Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Steven Rostedt Link: http://lkml.kernel.org/r/1442484533-19742-1-git-send-email-peter.senna@gmail.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/util.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c index 7acafb3c5592..c2cd9bf2348b 100644 --- a/tools/perf/util/util.c +++ b/tools/perf/util/util.c @@ -709,7 +709,7 @@ bool find_process(const char *name) dir = opendir(procfs__mountpoint()); if (!dir) - return -1; + return false; /* Walk through the directory. */ while (ret && (d = readdir(dir)) != NULL) { From d7ba2a024cf19261b34825f89d30ceef9a5ff864 Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Wed, 29 Jul 2015 18:45:33 +0900 Subject: [PATCH 163/647] of: add vendor prefix for Socionext Inc. Signed-off-by: Masahiro Yamada Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/vendor-prefixes.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt index ac5f0c34ae00..82d2ac97af74 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.txt +++ b/Documentation/devicetree/bindings/vendor-prefixes.txt @@ -203,6 +203,7 @@ sitronix Sitronix Technology Corporation skyworks Skyworks Solutions, Inc. smsc Standard Microsystems Corporation snps Synopsys, Inc. +socionext Socionext Inc. solidrun SolidRun solomon Solomon Systech Limited sony Sony Corporation From dc4dae00d82fedcd7a632f786f9f76c1f7f929a5 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Mon, 7 Sep 2015 10:49:03 +0100 Subject: [PATCH 164/647] Docs: dt: add #msi-cells to GICv3 ITS binding The GICv3 ITS uses sideband master identification data (known as a DeviceID) to identify which master wrote to a doorbell, and this data is used to determine how to react in response to the write. Commit 1e6db000482fa65a ("irqchip/gicv3-its: Add platform MSI support") added support per this binding, but failed to update the documentation. This patch fixes the documentation. Signed-off-by: Mark Rutland Acked-by: Marc Zyngier Cc: Rob Herring Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/arm/gic-v3.txt | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/Documentation/devicetree/bindings/arm/gic-v3.txt b/Documentation/devicetree/bindings/arm/gic-v3.txt index ddfade40ac59..7803e77d85cb 100644 --- a/Documentation/devicetree/bindings/arm/gic-v3.txt +++ b/Documentation/devicetree/bindings/arm/gic-v3.txt @@ -57,6 +57,8 @@ used to route Message Signalled Interrupts (MSI) to the CPUs. These nodes must have the following properties: - compatible : Should at least contain "arm,gic-v3-its". - msi-controller : Boolean property. Identifies the node as an MSI controller +- #msi-cells: Must be <1>. The single msi-cell is the DeviceID of the device + which will generate the MSI. - reg: Specifies the base physical address and size of the ITS registers. @@ -83,6 +85,7 @@ Examples: gic-its@2c200000 { compatible = "arm,gic-v3-its"; msi-controller; + #msi-cells = <1>; reg = <0x0 0x2c200000 0 0x200000>; }; }; @@ -107,12 +110,14 @@ Examples: gic-its@2c200000 { compatible = "arm,gic-v3-its"; msi-controller; + #msi-cells = <1>; reg = <0x0 0x2c200000 0 0x200000>; }; gic-its@2c400000 { compatible = "arm,gic-v3-its"; msi-controller; + #msi-cells = <1>; reg = <0x0 0x2c400000 0 0x200000>; }; }; From eb168b70dea54578a45119fcdb3b48ad5d75fed9 Mon Sep 17 00:00:00 2001 From: Punit Agrawal Date: Tue, 8 Sep 2015 12:20:48 +0100 Subject: [PATCH 165/647] of: thermal: Fix inconsitency between cooling-*-state and cooling-*-level The device trees in the kernel as well as the binding description in Documentation/devicetree/bindings/cpufreq/cpufreq-dt.txt use the cooling-{min,max}-level property. Fix the inconsistency with the binding description in Documentation/devicetree/bindings/thermal/thermal.txt by changing cooling-*-state properties to cooling-*-level. Signed-off-by: Punit Agrawal Cc: Eduardo Valentin Cc: Rob Herring Cc: Mark Rutland Cc: Ian Campbell Cc: Kumar Gala Signed-off-by: Rob Herring --- .../devicetree/bindings/thermal/thermal.txt | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/Documentation/devicetree/bindings/thermal/thermal.txt b/Documentation/devicetree/bindings/thermal/thermal.txt index 8a49362dea6e..8320186738ff 100644 --- a/Documentation/devicetree/bindings/thermal/thermal.txt +++ b/Documentation/devicetree/bindings/thermal/thermal.txt @@ -55,16 +55,16 @@ of heat dissipation). For example a fan's cooling states correspond to the different fan speeds possible. Cooling states are referred to by single unsigned integers, where larger numbers mean greater heat dissipation. The precise set of cooling states associated with a device -(as referred to be the cooling-min-state and cooling-max-state +(as referred to by the cooling-min-level and cooling-max-level properties) should be defined in a particular device's binding. For more examples of cooling devices, refer to the example sections below. Required properties: -- cooling-min-state: An integer indicating the smallest +- cooling-min-level: An integer indicating the smallest Type: unsigned cooling state accepted. Typically 0. Size: one cell -- cooling-max-state: An integer indicating the largest +- cooling-max-level: An integer indicating the largest Type: unsigned cooling state accepted. Size: one cell @@ -225,8 +225,8 @@ cpus { 396000 950000 198000 850000 >; - cooling-min-state = <0>; - cooling-max-state = <3>; + cooling-min-level = <0>; + cooling-max-level = <3>; #cooling-cells = <2>; /* min followed by max */ }; ... @@ -240,8 +240,8 @@ cpus { */ fan0: fan@0x48 { ... - cooling-min-state = <0>; - cooling-max-state = <9>; + cooling-min-level = <0>; + cooling-max-level = <9>; #cooling-cells = <2>; /* min followed by max */ }; }; From 9fa04fbeb78eab6f3817e444644348c46c7e2370 Mon Sep 17 00:00:00 2001 From: Punit Agrawal Date: Tue, 8 Sep 2015 12:20:49 +0100 Subject: [PATCH 166/647] of: thermal: Mark cooling-*-level properties optional The cooling-{min,max}-level properties are marked as optional in Documentation/devicetree/bindings/cpufreq/cpufreq-dt.txt and the usage in various device tree matches this, i.e., some cooling device in the device trees provide these properties while others do not. Make the bindings in Documentation/devicetree/bindings/thermal/thermal.txt consistent with the cpufreq-dt bindings by marking the cooling-*-level properties as optional. Signed-off-by: Punit Agrawal Cc: Eduardo Valentin Cc: Rob Herring Cc: Mark Rutland Cc: Ian Campbell Cc: Kumar Gala Signed-off-by: Rob Herring --- .../devicetree/bindings/thermal/thermal.txt | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/Documentation/devicetree/bindings/thermal/thermal.txt b/Documentation/devicetree/bindings/thermal/thermal.txt index 8320186738ff..41b817f7b670 100644 --- a/Documentation/devicetree/bindings/thermal/thermal.txt +++ b/Documentation/devicetree/bindings/thermal/thermal.txt @@ -60,14 +60,6 @@ properties) should be defined in a particular device's binding. For more examples of cooling devices, refer to the example sections below. Required properties: -- cooling-min-level: An integer indicating the smallest - Type: unsigned cooling state accepted. Typically 0. - Size: one cell - -- cooling-max-level: An integer indicating the largest - Type: unsigned cooling state accepted. - Size: one cell - - #cooling-cells: Used to provide cooling device specific information Type: unsigned while referring to it. Must be at least 2, in order Size: one cell to specify minimum and maximum cooling state used @@ -77,6 +69,15 @@ Required properties: See Cooling device maps section below for more details on how consumers refer to cooling devices. +Optional properties: +- cooling-min-level: An integer indicating the smallest + Type: unsigned cooling state accepted. Typically 0. + Size: one cell + +- cooling-max-level: An integer indicating the largest + Type: unsigned cooling state accepted. + Size: one cell + * Trip points The trip node is a node to describe a point in the temperature domain From 31b47ae3f18b31d86f553198e624b3b38f6397a2 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Sat, 12 Sep 2015 12:52:49 +0200 Subject: [PATCH 167/647] devicetree: bindings: Extend the bma180 bindings with bma250 info The bma180 / bma250 accelerometers share a driver (at least under Linux), so it makes sense to also have their bindings info in a single .txt. This commit extends the bma180 bindings with bma250 bindings, specifically it specifies how the 2 seperate interrupts the bma250 has must be listed in devicetree. The existing bma180 driver is already fully compatible with the specified bindings. Signed-off-by: Hans de Goede Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/iio/accel/bma180.txt | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/Documentation/devicetree/bindings/iio/accel/bma180.txt b/Documentation/devicetree/bindings/iio/accel/bma180.txt index c5933573e0f6..4a3679d54457 100644 --- a/Documentation/devicetree/bindings/iio/accel/bma180.txt +++ b/Documentation/devicetree/bindings/iio/accel/bma180.txt @@ -1,10 +1,11 @@ -* Bosch BMA180 triaxial acceleration sensor +* Bosch BMA180 / BMA250 triaxial acceleration sensor http://omapworld.com/BMA180_111_1002839.pdf +http://ae-bst.resource.bosch.com/media/products/dokumente/bma250/bst-bma250-ds002-05.pdf Required properties: - - compatible : should be "bosch,bma180" + - compatible : should be "bosch,bma180" or "bosch,bma250" - reg : the I2C address of the sensor Optional properties: @@ -13,6 +14,9 @@ Optional properties: - interrupts : interrupt mapping for GPIO IRQ, it should by configured with flags IRQ_TYPE_LEVEL_HIGH | IRQ_TYPE_EDGE_RISING + For the bma250 the first interrupt listed must be the one + connected to the INT1 pin, the second (optional) interrupt + listed must be the one connected to the INT2 pin. Example: From c8aa33a7114704f3f5a359893a8e1de9ea2940a5 Mon Sep 17 00:00:00 2001 From: David Daney Date: Wed, 9 Sep 2015 15:57:59 -0700 Subject: [PATCH 168/647] of_pci_irq: Silence bogus "of_irq_parse_pci() failed ..." messages. It is perfectly legitimate for a PCI device to have an PCI_INTERRUPT_PIN value of zero. This happens if the device doesn't use interrupts, or on PCIe devices, where only MSI/MSI-X are supported. Silence the annoying "of_irq_parse_pci() failed with rc=-19" error messages by moving the printing code into of_irq_parse_pci(), and only emitting the message for cases where PCI_INTERRUPT_PIN == 0 is not the cause for an early exit. Signed-off-by: David Daney Signed-off-by: Rob Herring --- drivers/of/of_pci_irq.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) diff --git a/drivers/of/of_pci_irq.c b/drivers/of/of_pci_irq.c index 1710d9dc7fc2..2306313c0029 100644 --- a/drivers/of/of_pci_irq.c +++ b/drivers/of/of_pci_irq.c @@ -38,8 +38,8 @@ int of_irq_parse_pci(const struct pci_dev *pdev, struct of_phandle_args *out_irq */ rc = pci_read_config_byte(pdev, PCI_INTERRUPT_PIN, &pin); if (rc != 0) - return rc; - /* No pin, exit */ + goto err; + /* No pin, exit with no error message. */ if (pin == 0) return -ENODEV; @@ -53,8 +53,10 @@ int of_irq_parse_pci(const struct pci_dev *pdev, struct of_phandle_args *out_irq ppnode = pci_bus_to_OF_node(pdev->bus); /* No node for host bridge ? give up */ - if (ppnode == NULL) - return -EINVAL; + if (ppnode == NULL) { + rc = -EINVAL; + goto err; + } } else { /* We found a P2P bridge, check if it has a node */ ppnode = pci_device_to_OF_node(ppdev); @@ -86,7 +88,13 @@ int of_irq_parse_pci(const struct pci_dev *pdev, struct of_phandle_args *out_irq out_irq->args[0] = pin; laddr[0] = cpu_to_be32((pdev->bus->number << 16) | (pdev->devfn << 8)); laddr[1] = laddr[2] = cpu_to_be32(0); - return of_irq_parse_raw(laddr, out_irq); + rc = of_irq_parse_raw(laddr, out_irq); + if (rc) + goto err; + return 0; +err: + dev_err(&pdev->dev, "of_irq_parse_pci() failed with rc=%d\n", rc); + return rc; } EXPORT_SYMBOL_GPL(of_irq_parse_pci); @@ -105,10 +113,8 @@ int of_irq_parse_and_map_pci(const struct pci_dev *dev, u8 slot, u8 pin) int ret; ret = of_irq_parse_pci(dev, &oirq); - if (ret) { - dev_err(&dev->dev, "of_irq_parse_pci() failed with rc=%d\n", ret); + if (ret) return 0; /* Proper return code 0 == NO_IRQ */ - } return irq_create_of_mapping(&oirq); } From a41cbe86df3afbc82311a1640e20858c0cd7e065 Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Mon, 14 Sep 2015 19:54:36 -0400 Subject: [PATCH 169/647] Failing to send a CLOSE if file is opened WRONLY and server reboots on a 4.x mount MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A test case is as the description says: open(foobar, O_WRONLY); sleep() --> reboot the server close(foobar) The bug is because in nfs4state.c in nfs4_reclaim_open_state() a few line before going to restart, there is clear_bit(NFS4CLNT_RECLAIM_NOGRACE, &state->flags). NFS4CLNT_RECLAIM_NOGRACE is a flag for the client states not open owner states. Value of NFS4CLNT_RECLAIM_NOGRACE is 4 which is the value of NFS_O_WRONLY_STATE in nfs4_state->flags. So clearing it wipes out state and when we go to close it, “call_close” doesn’t get set as state flag is not set and CLOSE doesn’t go on the wire. Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- fs/nfs/nfs4state.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index da73bc443238..5db324635e92 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1481,7 +1481,7 @@ restart: spin_unlock(&state->state_lock); } nfs4_put_open_state(state); - clear_bit(NFS4CLNT_RECLAIM_NOGRACE, + clear_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags); spin_lock(&sp->so_lock); goto restart; From 048883e0b934d9a5103d40e209cb14b7f33d2933 Mon Sep 17 00:00:00 2001 From: Peng Tao Date: Fri, 11 Sep 2015 11:14:06 +0800 Subject: [PATCH 170/647] nfs: fix pg_test page count calculation We really want sizeof(struct page *) instead. Otherwise we limit maximum IO size to 64 pages rather than 512 pages on a 64bit system. Fixes 2e11f829(nfs: cap request size to fit a kmalloced page array). Cc: Christoph Hellwig Signed-off-by: Peng Tao Fixes: 2e11f8296d22 ("nfs: cap request size to fit a kmalloced page array") Signed-off-by: Trond Myklebust --- fs/nfs/pagelist.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 7c5718ba625e..fe3ddd20ff89 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -508,7 +508,7 @@ size_t nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, * for it without upsetting the slab allocator. */ if (((mirror->pg_count + req->wb_bytes) >> PAGE_SHIFT) * - sizeof(struct page) > PAGE_SIZE) + sizeof(struct page *) > PAGE_SIZE) return 0; return min(mirror->pg_bsize - mirror->pg_count, (size_t)req->wb_bytes); From 03c78827db35be20ffa71cb5ccd2cedb192f95d6 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 17 Sep 2015 10:42:27 -0400 Subject: [PATCH 171/647] SUNRPC: Fix races between socket connection and destroy code When we're destroying the socket transport, we need to ensure that we cancel any existing delayed connection attempts, and order them w.r.t. the call to xs_close(). Reported-by:"Suzuki K. Poulose" Acked-by: Jeff Layton Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 7be90bc1a7c2..d2dfbd043bea 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -881,8 +881,11 @@ static void xs_xprt_free(struct rpc_xprt *xprt) */ static void xs_destroy(struct rpc_xprt *xprt) { + struct sock_xprt *transport = container_of(xprt, + struct sock_xprt, xprt); dprintk("RPC: xs_destroy xprt %p\n", xprt); + cancel_delayed_work_sync(&transport->connect_worker); xs_close(xprt); xs_xprt_free(xprt); module_put(THIS_MODULE); From 306a5549355966e480e0dcacdc6b9321d153e0c0 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 16 Sep 2015 17:21:27 -0400 Subject: [PATCH 172/647] nfs: fix v4.2 SEEK on files over 2 gigs We're incorrectly assigning a loff_t return to an int. If SEEK_HOLE or SEEK_DATA returns an offset over 2^31 then the application will see a weird lseek() result (usually -EIO). Cc: stable@vger.kernel.org Fixes: bdcc2cd14e4e "NFSv4.2: handle NFS-specific llseek errors" Signed-off-by: J. Bruce Fields Reviewed-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/nfs42proc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index d731bbf974aa..0f020e4d8421 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -175,10 +175,12 @@ loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence) { struct nfs_server *server = NFS_SERVER(file_inode(filep)); struct nfs4_exception exception = { }; - int err; + loff_t err; do { err = _nfs42_proc_llseek(filep, offset, whence); + if (err >= 0) + break; if (err == -ENOTSUPP) return -EOPNOTSUPP; err = nfs4_handle_exception(server, err, &exception); From 17a9618e984234fda167ea5e07eae6f4f2ea2186 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 13 Sep 2015 14:15:07 +0200 Subject: [PATCH 173/647] SUNRPC: drop null test before destroy functions Remove unneeded NULL test. The semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // @@ expression x; @@ -if (x != NULL) \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x); // Signed-off-by: Julia Lawall Signed-off-by: Trond Myklebust --- net/sunrpc/sched.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index b140c092d226..425ca2f37dd9 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -1092,14 +1092,10 @@ void rpc_destroy_mempool(void) { rpciod_stop(); - if (rpc_buffer_mempool) - mempool_destroy(rpc_buffer_mempool); - if (rpc_task_mempool) - mempool_destroy(rpc_task_mempool); - if (rpc_task_slabp) - kmem_cache_destroy(rpc_task_slabp); - if (rpc_buffer_slabp) - kmem_cache_destroy(rpc_buffer_slabp); + mempool_destroy(rpc_buffer_mempool); + mempool_destroy(rpc_task_mempool); + kmem_cache_destroy(rpc_task_slabp); + kmem_cache_destroy(rpc_buffer_slabp); rpc_destroy_wait_queue(&delay_queue); } From 74c600e36455af85c593702670f38709ca1951ce Mon Sep 17 00:00:00 2001 From: Martyn Welch Date: Thu, 17 Sep 2015 20:38:47 +0100 Subject: [PATCH 174/647] MAINTAINERS: Update email address for Martyn Welch I have recently left GE and the email address listed for me in the maintainers file is no longer valid. Updating email address. Signed-off-by: Martyn Welch Signed-off-by: Greg Kroah-Hartman --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 72fc3936b423..4cb7ee314887 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -11186,7 +11186,7 @@ F: drivers/vlynq/vlynq.c F: include/linux/vlynq.h VME SUBSYSTEM -M: Martyn Welch +M: Martyn Welch M: Manohar Vanga M: Greg Kroah-Hartman L: devel@driverdev.osuosl.org From 37a1d3611c126fd9782ce5235791f898f053e763 Mon Sep 17 00:00:00 2001 From: Roopa Prabhu Date: Sun, 13 Sep 2015 10:18:33 -0700 Subject: [PATCH 175/647] ipv6: include NLM_F_REPLACE in route replace notifications This patch adds NLM_F_REPLACE flag to ipv6 route replace notifications. This makes nlm_flags in ipv6 replace notifications consistent with ipv4. Signed-off-by: Roopa Prabhu Acked-by: Nicolas Dichtel Reviewed-by: Michal Kubecek Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 3 ++- net/ipv6/ip6_fib.c | 6 +++--- net/ipv6/route.c | 5 +++-- 3 files changed, 8 insertions(+), 6 deletions(-) diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 063d30474cf6..aaf9700fc9e5 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -275,7 +275,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info, struct mx6_config *mxc); int fib6_del(struct rt6_info *rt, struct nl_info *info); -void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info); +void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info, + unsigned int flags); void fib6_run_gc(unsigned long expires, struct net *net, bool force); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 8a9ec01f4d01..7d2e0023c72d 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -851,7 +851,7 @@ add: *ins = rt; rt->rt6i_node = fn; atomic_inc(&rt->rt6i_ref); - inet6_rt_notify(RTM_NEWROUTE, rt, info); + inet6_rt_notify(RTM_NEWROUTE, rt, info, 0); info->nl_net->ipv6.rt6_stats->fib_rt_entries++; if (!(fn->fn_flags & RTN_RTINFO)) { @@ -877,7 +877,7 @@ add: rt->rt6i_node = fn; rt->dst.rt6_next = iter->dst.rt6_next; atomic_inc(&rt->rt6i_ref); - inet6_rt_notify(RTM_NEWROUTE, rt, info); + inet6_rt_notify(RTM_NEWROUTE, rt, info, NLM_F_REPLACE); if (!(fn->fn_flags & RTN_RTINFO)) { info->nl_net->ipv6.rt6_stats->fib_route_nodes++; fn->fn_flags |= RTN_RTINFO; @@ -1422,7 +1422,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, fib6_purge_rt(rt, fn, net); - inet6_rt_notify(RTM_DELROUTE, rt, info); + inet6_rt_notify(RTM_DELROUTE, rt, info, 0); rt6_release(rt); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 3d3c1b294725..d5fa50297f80 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -3304,7 +3304,8 @@ errout: return err; } -void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) +void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info, + unsigned int nlm_flags) { struct sk_buff *skb; struct net *net = info->nl_net; @@ -3319,7 +3320,7 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) goto errout; err = rt6_fill_node(net, skb, rt, NULL, NULL, 0, - event, info->portid, seq, 0, 0, 0); + event, info->portid, seq, 0, 0, nlm_flags); if (err < 0) { /* -EMSGSIZE implies BUG in rt6_nlmsg_size() */ WARN_ON(err == -EMSGSIZE); From 0fdea1e8a2853f79d39b8555cc9de16a7e0ab26f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 16 Sep 2015 23:43:17 -0400 Subject: [PATCH 176/647] SUNRPC: Ensure that we wait for connections to complete before retrying Commit 718ba5b87343, moved the responsibility for unlocking the socket to xs_tcp_setup_socket, meaning that the socket will be unlocked before we know that it has finished trying to connect. The following patch is based on an initial patch by Russell King to ensure that we delay clearing the XPRT_CONNECTING flag until we either know that we failed to initiate a connection attempt, or the connection attempt itself failed. Fixes: 718ba5b87343 ("SUNRPC: Add helpers to prevent socket create from racing") Reported-by: Russell King Reported-by: Russell King Tested-by: Russell King Tested-by: Benjamin Coddington Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprtsock.h | 3 +++ net/sunrpc/xprtsock.c | 11 ++++++++--- 2 files changed, 11 insertions(+), 3 deletions(-) diff --git a/include/linux/sunrpc/xprtsock.h b/include/linux/sunrpc/xprtsock.h index 7591788e9fbf..357e44c1a46b 100644 --- a/include/linux/sunrpc/xprtsock.h +++ b/include/linux/sunrpc/xprtsock.h @@ -42,6 +42,7 @@ struct sock_xprt { /* * Connection of transports */ + unsigned long sock_state; struct delayed_work connect_worker; struct sockaddr_storage srcaddr; unsigned short srcport; @@ -76,6 +77,8 @@ struct sock_xprt { */ #define TCP_RPC_REPLY (1UL << 6) +#define XPRT_SOCK_CONNECTING 1U + #endif /* __KERNEL__ */ #endif /* _LINUX_SUNRPC_XPRTSOCK_H */ diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index d2dfbd043bea..c35038511686 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1438,6 +1438,7 @@ out: static void xs_tcp_state_change(struct sock *sk) { struct rpc_xprt *xprt; + struct sock_xprt *transport; read_lock_bh(&sk->sk_callback_lock); if (!(xprt = xprt_from_sock(sk))) @@ -1449,13 +1450,12 @@ static void xs_tcp_state_change(struct sock *sk) sock_flag(sk, SOCK_ZAPPED), sk->sk_shutdown); + transport = container_of(xprt, struct sock_xprt, xprt); trace_rpc_socket_state_change(xprt, sk->sk_socket); switch (sk->sk_state) { case TCP_ESTABLISHED: spin_lock(&xprt->transport_lock); if (!xprt_test_and_set_connected(xprt)) { - struct sock_xprt *transport = container_of(xprt, - struct sock_xprt, xprt); /* Reset TCP record info */ transport->tcp_offset = 0; @@ -1464,6 +1464,8 @@ static void xs_tcp_state_change(struct sock *sk) transport->tcp_flags = TCP_RCV_COPY_FRAGHDR | TCP_RCV_COPY_XID; xprt->connect_cookie++; + clear_bit(XPRT_SOCK_CONNECTING, &transport->sock_state); + xprt_clear_connecting(xprt); xprt_wake_pending_tasks(xprt, -EAGAIN); } @@ -1499,6 +1501,9 @@ static void xs_tcp_state_change(struct sock *sk) smp_mb__after_atomic(); break; case TCP_CLOSE: + if (test_and_clear_bit(XPRT_SOCK_CONNECTING, + &transport->sock_state)) + xprt_clear_connecting(xprt); xs_sock_mark_closed(xprt); } out: @@ -2182,6 +2187,7 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) /* Tell the socket layer to start connecting... */ xprt->stat.connect_count++; xprt->stat.connect_start = jiffies; + set_bit(XPRT_SOCK_CONNECTING, &transport->sock_state); ret = kernel_connect(sock, xs_addr(xprt), xprt->addrlen, O_NONBLOCK); switch (ret) { case 0: @@ -2243,7 +2249,6 @@ static void xs_tcp_setup_socket(struct work_struct *work) case -EINPROGRESS: case -EALREADY: xprt_unlock_connect(xprt, transport); - xprt_clear_connecting(xprt); return; case -EINVAL: /* Happens, for instance, if the user specified a link From 3ec0c97959abff33a42db9081c22132bcff5b4f2 Mon Sep 17 00:00:00 2001 From: Kinglong Mee Date: Mon, 14 Sep 2015 20:12:21 +0800 Subject: [PATCH 177/647] nfs/filelayout: Fix NULL reference caused by double freeing of fh_array If filelayout_decode_layout fail, _filelayout_free_lseg will causes a double freeing of fh_array. [ 1179.279800] BUG: unable to handle kernel NULL pointer dereference at (null) [ 1179.280198] IP: [] filelayout_free_fh_array.isra.11+0x1d/0x70 [nfs_layout_nfsv41_files] [ 1179.281010] PGD 0 [ 1179.281443] Oops: 0000 [#1] [ 1179.281831] Modules linked in: nfs_layout_nfsv41_files(OE) nfsv4(OE) nfs(OE) fscache(E) xfs libcrc32c coretemp nfsd crct10dif_pclmul ppdev crc32_pclmul crc32c_intel auth_rpcgss ghash_clmulni_intel nfs_acl lockd vmw_balloon grace sunrpc parport_pc vmw_vmci parport shpchp i2c_piix4 vmwgfx drm_kms_helper ttm drm serio_raw mptspi scsi_transport_spi mptscsih e1000 mptbase ata_generic pata_acpi [last unloaded: fscache] [ 1179.283891] CPU: 0 PID: 13336 Comm: cat Tainted: G OE 4.3.0-rc1-pnfs+ #244 [ 1179.284323] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 05/20/2014 [ 1179.285206] task: ffff8800501d48c0 ti: ffff88003e3c4000 task.ti: ffff88003e3c4000 [ 1179.285668] RIP: 0010:[] [] filelayout_free_fh_array.isra.11+0x1d/0x70 [nfs_layout_nfsv41_files] [ 1179.286612] RSP: 0018:ffff88003e3c77f8 EFLAGS: 00010202 [ 1179.287092] RAX: 0000000000000000 RBX: ffff88001fe78900 RCX: 0000000000000000 [ 1179.287731] RDX: ffffea0000f40760 RSI: ffff88001fe789c8 RDI: ffff88001fe789c0 [ 1179.288383] RBP: ffff88003e3c7810 R08: ffffea0000f40760 R09: 0000000000000000 [ 1179.289170] R10: 0000000000000000 R11: 0000000000000001 R12: ffff88001fe789c8 [ 1179.289959] R13: ffff88001fe789c0 R14: ffff88004ec05a80 R15: ffff88004f935b88 [ 1179.290791] FS: 00007f4e66bb5700(0000) GS:ffffffff81c29000(0000) knlGS:0000000000000000 [ 1179.291580] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1179.292209] CR2: 0000000000000000 CR3: 00000000203f8000 CR4: 00000000001406f0 [ 1179.292731] Stack: [ 1179.293195] ffff88001fe78900 00000000000000d0 ffff88001fe78178 ffff88003e3c7868 [ 1179.293676] ffffffffa0272737 0000000000000001 0000000000000001 ffff88001fe78800 [ 1179.294151] 00000000614fffce ffffffff81727671 ffff88001fe78100 ffff88001fe78100 [ 1179.294623] Call Trace: [ 1179.295092] [] filelayout_alloc_lseg+0xa7/0x2d0 [nfs_layout_nfsv41_files] [ 1179.295625] [] ? out_of_line_wait_on_bit+0x81/0xb0 [ 1179.296133] [] pnfs_layout_process+0xae/0x320 [nfsv4] [ 1179.296632] [] nfs4_proc_layoutget+0x2b1/0x360 [nfsv4] [ 1179.297134] [] pnfs_update_layout+0x853/0xb30 [nfsv4] [ 1179.297632] [] ? nfs_get_lock_context+0x74/0x170 [nfs] [ 1179.298158] [] filelayout_pg_init_read+0x37/0x50 [nfs_layout_nfsv41_files] [ 1179.298834] [] __nfs_pageio_add_request+0x119/0x460 [nfs] [ 1179.299385] [] ? nfs_create_request.part.9+0x37/0x2e0 [nfs] [ 1179.299872] [] nfs_pageio_add_request+0xa3/0x1b0 [nfs] [ 1179.300362] [] readpage_async_filler+0x85/0x260 [nfs] [ 1179.300907] [] read_cache_pages+0x91/0xd0 [ 1179.301391] [] ? nfs_read_completion+0x220/0x220 [nfs] [ 1179.301867] [] nfs_readpages+0x128/0x200 [nfs] [ 1179.302330] [] __do_page_cache_readahead+0x203/0x280 [ 1179.302784] [] ? __do_page_cache_readahead+0xd8/0x280 [ 1179.303413] [] ondemand_readahead+0x1a6/0x2f0 [ 1179.303855] [] page_cache_sync_readahead+0x31/0x50 [ 1179.304286] [] generic_file_read_iter+0x4a6/0x5c0 [ 1179.304711] [] ? __nfs_revalidate_mapping+0x1f6/0x240 [nfs] [ 1179.305132] [] nfs_file_read+0x52/0xa0 [nfs] [ 1179.305540] [] __vfs_read+0xcc/0x100 [ 1179.305936] [] vfs_read+0x85/0x130 [ 1179.306326] [] SyS_read+0x58/0xd0 [ 1179.306708] [] entry_SYSCALL_64_fastpath+0x12/0x76 [ 1179.307094] Code: c4 66 66 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 41 55 41 54 53 8b 07 49 89 f4 85 c0 74 47 48 8b 06 49 89 fd <48> 8b 38 48 85 ff 74 22 31 db eb 0c 48 63 d3 48 8b 3c d0 48 85 [ 1179.308357] RIP [] filelayout_free_fh_array.isra.11+0x1d/0x70 [nfs_layout_nfsv41_files] [ 1179.309177] RSP [ 1179.309582] CR2: 0000000000000000 Signed-off-by: Kinglong Mee Signed-off-by: Trond Myklebust --- fs/nfs/filelayout/filelayout.c | 31 ++++++++++++------------------- 1 file changed, 12 insertions(+), 19 deletions(-) diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c index b34f2e228601..02ec07973bc4 100644 --- a/fs/nfs/filelayout/filelayout.c +++ b/fs/nfs/filelayout/filelayout.c @@ -629,23 +629,18 @@ out_put: goto out; } -static void filelayout_free_fh_array(struct nfs4_filelayout_segment *fl) +static void _filelayout_free_lseg(struct nfs4_filelayout_segment *fl) { int i; - for (i = 0; i < fl->num_fh; i++) { - if (!fl->fh_array[i]) - break; - kfree(fl->fh_array[i]); + if (fl->fh_array) { + for (i = 0; i < fl->num_fh; i++) { + if (!fl->fh_array[i]) + break; + kfree(fl->fh_array[i]); + } + kfree(fl->fh_array); } - kfree(fl->fh_array); - fl->fh_array = NULL; -} - -static void -_filelayout_free_lseg(struct nfs4_filelayout_segment *fl) -{ - filelayout_free_fh_array(fl); kfree(fl); } @@ -716,21 +711,21 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo, /* Do we want to use a mempool here? */ fl->fh_array[i] = kmalloc(sizeof(struct nfs_fh), gfp_flags); if (!fl->fh_array[i]) - goto out_err_free; + goto out_err; p = xdr_inline_decode(&stream, 4); if (unlikely(!p)) - goto out_err_free; + goto out_err; fl->fh_array[i]->size = be32_to_cpup(p++); if (sizeof(struct nfs_fh) < fl->fh_array[i]->size) { printk(KERN_ERR "NFS: Too big fh %d received %d\n", i, fl->fh_array[i]->size); - goto out_err_free; + goto out_err; } p = xdr_inline_decode(&stream, fl->fh_array[i]->size); if (unlikely(!p)) - goto out_err_free; + goto out_err; memcpy(fl->fh_array[i]->data, p, fl->fh_array[i]->size); dprintk("DEBUG: %s: fh len %d\n", __func__, fl->fh_array[i]->size); @@ -739,8 +734,6 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo, __free_page(scratch); return 0; -out_err_free: - filelayout_free_fh_array(fl); out_err: __free_page(scratch); return -EIO; From cc5706056baa3002b844ff240a1cc2199a978795 Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Mon, 14 Sep 2015 11:14:50 -0700 Subject: [PATCH 178/647] openvswitch: Fix IPv6 exthdr handling with ct helpers. Static code analysis reveals the following bug: net/openvswitch/conntrack.c:281 ovs_ct_helper() warn: unsigned 'protoff' is never less than zero. This signedness bug breaks error handling for IPv6 extension headers when using conntrack helpers. Fix the error by using a local signed variable. Fixes: cae3a2627520: "openvswitch: Allow attaching helpers to ct action" Reported-by: Dan Carpenter Signed-off-by: Joe Stringer Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/conntrack.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index e8e524ad8a01..002a755fa07e 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -275,13 +275,15 @@ static int ovs_ct_helper(struct sk_buff *skb, u16 proto) case NFPROTO_IPV6: { u8 nexthdr = ipv6_hdr(skb)->nexthdr; __be16 frag_off; + int ofs; - protoff = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), - &nexthdr, &frag_off); - if (protoff < 0 || (frag_off & htons(~0x7)) != 0) { + ofs = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &nexthdr, + &frag_off); + if (ofs < 0 || (frag_off & htons(~0x7)) != 0) { pr_debug("proto header not found\n"); return NF_ACCEPT; } + protoff = ofs; break; } default: From 562d897d15a6e2bab3cc9b4c172286b612834fe8 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Tue, 15 Sep 2015 10:50:14 -0600 Subject: [PATCH 179/647] net: Add documentation for VRF device Signed-off-by: David Ahern Signed-off-by: David S. Miller --- Documentation/networking/vrf.txt | 96 ++++++++++++++++++++++++++++++++ MAINTAINERS | 1 + 2 files changed, 97 insertions(+) create mode 100644 Documentation/networking/vrf.txt diff --git a/Documentation/networking/vrf.txt b/Documentation/networking/vrf.txt new file mode 100644 index 000000000000..031ef4a63485 --- /dev/null +++ b/Documentation/networking/vrf.txt @@ -0,0 +1,96 @@ +Virtual Routing and Forwarding (VRF) +==================================== +The VRF device combined with ip rules provides the ability to create virtual +routing and forwarding domains (aka VRFs, VRF-lite to be specific) in the +Linux network stack. One use case is the multi-tenancy problem where each +tenant has their own unique routing tables and in the very least need +different default gateways. + +Processes can be "VRF aware" by binding a socket to the VRF device. Packets +through the socket then use the routing table associated with the VRF +device. An important feature of the VRF device implementation is that it +impacts only Layer 3 and above so L2 tools (e.g., LLDP) are not affected +(ie., they do not need to be run in each VRF). The design also allows +the use of higher priority ip rules (Policy Based Routing, PBR) to take +precedence over the VRF device rules directing specific traffic as desired. + +In addition, VRF devices allow VRFs to be nested within namespaces. For +example network namespaces provide separation of network interfaces at L1 +(Layer 1 separation), VLANs on the interfaces within a namespace provide +L2 separation and then VRF devices provide L3 separation. + +Design +------ +A VRF device is created with an associated route table. Network interfaces +are then enslaved to a VRF device: + + +-----------------------------+ + | vrf-blue | ===> route table 10 + +-----------------------------+ + | | | + +------+ +------+ +-------------+ + | eth1 | | eth2 | ... | bond1 | + +------+ +------+ +-------------+ + | | + +------+ +------+ + | eth8 | | eth9 | + +------+ +------+ + +Packets received on an enslaved device and are switched to the VRF device +using an rx_handler which gives the impression that packets flow through +the VRF device. Similarly on egress routing rules are used to send packets +to the VRF device driver before getting sent out the actual interface. This +allows tcpdump on a VRF device to capture all packets into and out of the +VRF as a whole.[1] Similiarly, netfilter [2] and tc rules can be applied +using the VRF device to specify rules that apply to the VRF domain as a whole. + +[1] Packets in the forwarded state do not flow through the device, so those + packets are not seen by tcpdump. Will revisit this limitation in a + future release. + +[2] Iptables on ingress is limited to NF_INET_PRE_ROUTING only with skb->dev + set to real ingress device and egress is limited to NF_INET_POST_ROUTING. + Will revisit this limitation in a future release. + + +Setup +----- +1. VRF device is created with an association to a FIB table. + e.g, ip link add vrf-blue type vrf table 10 + ip link set dev vrf-blue up + +2. Rules are added that send lookups to the associated FIB table when the + iif or oif is the VRF device. e.g., + ip ru add oif vrf-blue table 10 + ip ru add iif vrf-blue table 10 + + Set the default route for the table (and hence default route for the VRF). + e.g, ip route add table 10 prohibit default + +3. Enslave L3 interfaces to a VRF device. + e.g, ip link set dev eth1 master vrf-blue + + Local and connected routes for enslaved devices are automatically moved to + the table associated with VRF device. Any additional routes depending on + the enslaved device will need to be reinserted following the enslavement. + +4. Additional VRF routes are added to associated table. + e.g., ip route add table 10 ... + + +Applications +------------ +Applications that are to work within a VRF need to bind their socket to the +VRF device: + + setsockopt(sd, SOL_SOCKET, SO_BINDTODEVICE, dev, strlen(dev)+1); + +or to specify the output device using cmsg and IP_PKTINFO. + + +Limitations +----------- +VRF device currently only works for IPv4. Support for IPv6 is under development. + +Index of original ingress interface is not available via cmsg. Will address +soon. diff --git a/MAINTAINERS b/MAINTAINERS index 310da4295c70..d4d9d4f6d271 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -11243,6 +11243,7 @@ L: netdev@vger.kernel.org S: Maintained F: drivers/net/vrf.c F: include/net/vrf.h +F: Documentation/networking/vrf.txt VT1211 HARDWARE MONITOR DRIVER M: Juerg Haefliger From 2e64126bb0fb581b55e52e51ec451ebb0d6769c8 Mon Sep 17 00:00:00 2001 From: Phil Sutter Date: Tue, 15 Sep 2015 10:33:07 +0200 Subject: [PATCH 180/647] net: qdisc: enhance default_qdisc documentation Aside from some lingual cleanup, point out which interfaces are not or partly covered by this setting. Signed-off-by: Phil Sutter Acked-by: Cong Wang Signed-off-by: David S. Miller --- Documentation/sysctl/net.txt | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/Documentation/sysctl/net.txt b/Documentation/sysctl/net.txt index 6294b5186ae5..809ab6efcc74 100644 --- a/Documentation/sysctl/net.txt +++ b/Documentation/sysctl/net.txt @@ -54,13 +54,15 @@ default_qdisc -------------- The default queuing discipline to use for network devices. This allows -overriding the default queue discipline of pfifo_fast with an -alternative. Since the default queuing discipline is created with the -no additional parameters so is best suited to queuing disciplines that -work well without configuration like stochastic fair queue (sfq), -CoDel (codel) or fair queue CoDel (fq_codel). Don't use queuing disciplines -like Hierarchical Token Bucket or Deficit Round Robin which require setting -up classes and bandwidths. +overriding the default of pfifo_fast with an alternative. Since the default +queuing discipline is created without additional parameters so is best suited +to queuing disciplines that work well without configuration like stochastic +fair queue (sfq), CoDel (codel) or fair queue CoDel (fq_codel). Don't use +queuing disciplines like Hierarchical Token Bucket or Deficit Round Robin +which require setting up classes and bandwidths. Note that physical multiqueue +interfaces still use mq as root qdisc, which in turn uses this default for its +leaves. Virtual devices (like e.g. lo or veth) ignore this setting and instead +default to noqueue. Default: pfifo_fast busy_read From d828755eae637c6ca39e30702e98abaf34cac146 Mon Sep 17 00:00:00 2001 From: Hariprasad Shenai Date: Tue, 15 Sep 2015 17:20:09 +0530 Subject: [PATCH 181/647] cxgb4: add device ID for few T5 adapters Signed-off-by: Hariprasad Shenai Signed-off-by: David S. Miller --- drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h index 8353a6cbfcc2..03ed00c49823 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h @@ -157,6 +157,11 @@ CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN CH_PCI_ID_TABLE_FENTRY(0x5090), /* Custom T540-CR */ CH_PCI_ID_TABLE_FENTRY(0x5091), /* Custom T522-CR */ CH_PCI_ID_TABLE_FENTRY(0x5092), /* Custom T520-CR */ + CH_PCI_ID_TABLE_FENTRY(0x5093), /* Custom T580-LP-CR */ + CH_PCI_ID_TABLE_FENTRY(0x5094), /* Custom T540-CR */ + CH_PCI_ID_TABLE_FENTRY(0x5095), /* Custom T540-CR-SO */ + CH_PCI_ID_TABLE_FENTRY(0x5096), /* Custom T580-CR */ + CH_PCI_ID_TABLE_FENTRY(0x5097), /* Custom T520-KR */ /* T6 adapters: */ From 58189ca7b27411c3dc9a5cb9eeee0906da684c59 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Tue, 15 Sep 2015 15:10:50 -0700 Subject: [PATCH 182/647] net: Fix vti use case with oif in dst lookups Steffen reported that the recent change to add oif to dst lookups breaks the VTI use case. The problem is that with the oif set in the flow struct the comparison to the nh_oif is triggered. Fix by splitting the FLOWI_FLAG_VRFSRC into 2 flags -- one that triggers the vrf device cache bypass (FLOWI_FLAG_VRFSRC) and another telling the lookup to not compare nh oif (FLOWI_FLAG_SKIP_NH_OIF). Fixes: 42a7b32b73d6 ("xfrm: Add oif to dst lookups") Signed-off-by: David Ahern Acked-by: Steffen Klassert Signed-off-by: David S. Miller --- drivers/net/vrf.c | 3 ++- include/net/flow.h | 1 + include/net/route.h | 2 +- net/ipv4/fib_trie.c | 2 +- net/ipv4/udp.c | 3 ++- net/ipv4/xfrm4_policy.c | 2 ++ 6 files changed, 9 insertions(+), 4 deletions(-) diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index e7094fbd7568..488c6f50df73 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -193,7 +193,8 @@ static netdev_tx_t vrf_process_v4_outbound(struct sk_buff *skb, .flowi4_oif = vrf_dev->ifindex, .flowi4_iif = LOOPBACK_IFINDEX, .flowi4_tos = RT_TOS(ip4h->tos), - .flowi4_flags = FLOWI_FLAG_ANYSRC | FLOWI_FLAG_VRFSRC, + .flowi4_flags = FLOWI_FLAG_ANYSRC | FLOWI_FLAG_VRFSRC | + FLOWI_FLAG_SKIP_NH_OIF, .daddr = ip4h->daddr, }; diff --git a/include/net/flow.h b/include/net/flow.h index acd6a096250e..9b85db85f13c 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -35,6 +35,7 @@ struct flowi_common { #define FLOWI_FLAG_ANYSRC 0x01 #define FLOWI_FLAG_KNOWN_NH 0x02 #define FLOWI_FLAG_VRFSRC 0x04 +#define FLOWI_FLAG_SKIP_NH_OIF 0x08 __u32 flowic_secid; struct flowi_tunnel flowic_tun_key; }; diff --git a/include/net/route.h b/include/net/route.h index cc61cb95f059..f46af256880c 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -255,7 +255,7 @@ static inline void ip_route_connect_init(struct flowi4 *fl4, __be32 dst, __be32 flow_flags |= FLOWI_FLAG_ANYSRC; if (netif_index_is_vrf(sock_net(sk), oif)) - flow_flags |= FLOWI_FLAG_VRFSRC; + flow_flags |= FLOWI_FLAG_VRFSRC | FLOWI_FLAG_SKIP_NH_OIF; flowi4_init_output(fl4, oif, sk->sk_mark, tos, RT_SCOPE_UNIVERSE, protocol, flow_flags, dst, src, dport, sport); diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 26d6ffb6d23c..6c2af797f2f9 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -1426,7 +1426,7 @@ found: nh->nh_flags & RTNH_F_LINKDOWN && !(fib_flags & FIB_LOOKUP_IGNORE_LINKSTATE)) continue; - if (!(flp->flowi4_flags & FLOWI_FLAG_VRFSRC)) { + if (!(flp->flowi4_flags & FLOWI_FLAG_SKIP_NH_OIF)) { if (flp->flowi4_oif && flp->flowi4_oif != nh->nh_oif) continue; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index c0a15e7f359f..f7d1d5e19e95 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1024,7 +1024,8 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) if (netif_index_is_vrf(net, ipc.oif)) { flowi4_init_output(fl4, ipc.oif, sk->sk_mark, tos, RT_SCOPE_UNIVERSE, sk->sk_protocol, - (flow_flags | FLOWI_FLAG_VRFSRC), + (flow_flags | FLOWI_FLAG_VRFSRC | + FLOWI_FLAG_SKIP_NH_OIF), faddr, saddr, dport, inet->inet_sport); diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index bb919b28619f..c10a9ee68433 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -33,6 +33,8 @@ static struct dst_entry *__xfrm4_dst_lookup(struct net *net, struct flowi4 *fl4, if (saddr) fl4->saddr = saddr->a4; + fl4->flowi4_flags = FLOWI_FLAG_SKIP_NH_OIF; + rt = __ip_route_output_key(net, fl4); if (!IS_ERR(rt)) return &rt->dst; From 4671fc6d47e0a0108fe24a4d830347d6a6ef4aa7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 15 Sep 2015 18:29:47 -0700 Subject: [PATCH 183/647] net/mlx4_en: really allow to change RSS key When changing rss key, we do not want to overwrite user provided key by the one provided by netdev_rss_key_fill(), which is the host random key generated at boot time. Fixes: 947cbb0ac242 ("net/mlx4_en: Support for configurable RSS hash function") Signed-off-by: Eric Dumazet Cc: Eyal Perry CC: Amir Vadai Acked-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/en_rx.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index 4402a1e48c9b..0ce6ffe73ca8 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -1268,8 +1268,6 @@ int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv) rss_context->hash_fn = MLX4_RSS_HASH_TOP; memcpy(rss_context->rss_key, priv->rss_key, MLX4_EN_RSS_KEY_SIZE); - netdev_rss_key_fill(rss_context->rss_key, - MLX4_EN_RSS_KEY_SIZE); } else { en_err(priv, "Unknown RSS hash function requested\n"); err = -EINVAL; From 88c796640eac36209efabe5e42b7b47dee58603e Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Wed, 16 Sep 2015 11:11:22 +0200 Subject: [PATCH 184/647] net: ks8851: Export OF module alias information Drivers needs to export the OF id table and this be built into the module or udev won't have the necessary information to autoload the driver module when the device is registered via OF. Signed-off-by: Javier Martinez Canillas Signed-off-by: David S. Miller --- drivers/net/ethernet/micrel/ks8851.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/micrel/ks8851.c b/drivers/net/ethernet/micrel/ks8851.c index 66d4ab703f45..60f43ec22175 100644 --- a/drivers/net/ethernet/micrel/ks8851.c +++ b/drivers/net/ethernet/micrel/ks8851.c @@ -1601,6 +1601,7 @@ static const struct of_device_id ks8851_match_table[] = { { .compatible = "micrel,ks8851" }, { } }; +MODULE_DEVICE_TABLE(of, ks8851_match_table); static struct spi_driver ks8851_driver = { .driver = { From ce816eb064c82ab96276969971a561db78e66164 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 16 Sep 2015 12:35:00 +0100 Subject: [PATCH 185/647] solos-pci: Increase headroom on received packets MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A comment in include/linux/skbuff.h says that: * Various parts of the networking layer expect at least 32 bytes of * headroom, you should not reduce this. This was demonstrated by a panic when handling fragmented IPv6 packets: http://marc.info/?l=linux-netdev&m=144236093519172&w=2 It's not entirely clear if that comment is still valid — and if it is, perhaps netif_rx() ought to be enforcing it with a warning. But either way, it is rather stupid from a performance point of view for us to be receiving packets into a buffer which doesn't have enough room to prepend an Ethernet header — it means that *every* incoming packet is going to be need to be reallocated. So let's fix that. Signed-off-by: David Woodhouse Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/atm/solos-pci.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/atm/solos-pci.c b/drivers/atm/solos-pci.c index 74e18b0a6d89..3d7fb6516f74 100644 --- a/drivers/atm/solos-pci.c +++ b/drivers/atm/solos-pci.c @@ -805,7 +805,12 @@ static void solos_bh(unsigned long card_arg) continue; } - skb = alloc_skb(size + 1, GFP_ATOMIC); + /* Use netdev_alloc_skb() because it adds NET_SKB_PAD of + * headroom, and ensures we can route packets back out an + * Ethernet interface (for example) without having to + * reallocate. Adding NET_IP_ALIGN also ensures that both + * PPPoATM and PPPoEoBR2684 packets end up aligned. */ + skb = netdev_alloc_skb_ip_align(NULL, size + 1); if (!skb) { if (net_ratelimit()) dev_warn(&card->dev->dev, "Failed to allocate sk_buff for RX\n"); @@ -869,7 +874,10 @@ static void solos_bh(unsigned long card_arg) /* Allocate RX skbs for any ports which need them */ if (card->using_dma && card->atmdev[port] && !card->rx_skb[port]) { - struct sk_buff *skb = alloc_skb(RX_DMA_SIZE, GFP_ATOMIC); + /* Unlike the MMIO case (qv) we can't add NET_IP_ALIGN + * here; the FPGA can only DMA to addresses which are + * aligned to 4 bytes. */ + struct sk_buff *skb = dev_alloc_skb(RX_DMA_SIZE); if (skb) { SKB_CB(skb)->dma_addr = dma_map_single(&card->dev->dev, skb->data, From 1d325d217c7f190a42fb620ead20bb240fc16af0 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 16 Sep 2015 17:26:14 +0200 Subject: [PATCH 186/647] ipv6: ip6_fragment: fix headroom tests and skb leak David Woodhouse reports skb_under_panic when we try to push ethernet header to fragmented ipv6 skbs: skbuff: skb_under_panic: text:c1277f1e len:1294 put:14 head:dec98000 data:dec97ffc tail:0xdec9850a end:0xdec98f40 dev:br-lan [..] ip6_finish_output2+0x196/0x4da David further debugged this: [..] offending fragments were arriving here with skb_headroom(skb)==10. Which is reasonable, being the Solos ADSL card's header of 8 bytes followed by 2 bytes of PPP frame type. The problem is that if netfilter ipv6 defragmentation is used, skb_cow() in ip6_forward will only see reassembled skb. Therefore, headroom is overestimated by 8 bytes (we pulled fragment header) and we don't check the skbs in the frag_list either. We can't do these checks in netfilter defrag since outdev isn't known yet. Furthermore, existing tests in ip6_fragment did not consider the fragment or ipv6 header size when checking headroom of the fraglist skbs. While at it, also fix a skb leak on memory allocation -- ip6_fragment must consume the skb. I tested this e1000 driver hacked to not allocate additional headroom (we end up in slowpath, since LL_RESERVED_SPACE is 16). If 2 bytes of headroom are allocated, fastpath is taken (14 byte ethernet header was pulled, so 16 byte headroom available in all fragments). Reported-by: David Woodhouse Diagnosed-by: David Woodhouse Signed-off-by: Florian Westphal Tested-by: David Woodhouse Signed-off-by: David S. Miller --- net/ipv6/ip6_output.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 26ea47930740..92b1aa38f121 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -586,20 +586,22 @@ int ip6_fragment(struct sock *sk, struct sk_buff *skb, frag_id = ipv6_select_ident(net, &ipv6_hdr(skb)->daddr, &ipv6_hdr(skb)->saddr); + hroom = LL_RESERVED_SPACE(rt->dst.dev); if (skb_has_frag_list(skb)) { int first_len = skb_pagelen(skb); struct sk_buff *frag2; if (first_len - hlen > mtu || ((first_len - hlen) & 7) || - skb_cloned(skb)) + skb_cloned(skb) || + skb_headroom(skb) < (hroom + sizeof(struct frag_hdr))) goto slow_path; skb_walk_frags(skb, frag) { /* Correct geometry. */ if (frag->len > mtu || ((frag->len & 7) && frag->next) || - skb_headroom(frag) < hlen) + skb_headroom(frag) < (hlen + hroom + sizeof(struct frag_hdr))) goto slow_path_clean; /* Partially cloned skb? */ @@ -616,8 +618,6 @@ int ip6_fragment(struct sock *sk, struct sk_buff *skb, err = 0; offset = 0; - frag = skb_shinfo(skb)->frag_list; - skb_frag_list_init(skb); /* BUILD HEADER */ *prevhdr = NEXTHDR_FRAGMENT; @@ -625,8 +625,11 @@ int ip6_fragment(struct sock *sk, struct sk_buff *skb, if (!tmp_hdr) { IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); - return -ENOMEM; + err = -ENOMEM; + goto fail; } + frag = skb_shinfo(skb)->frag_list; + skb_frag_list_init(skb); __skb_pull(skb, hlen); fh = (struct frag_hdr *)__skb_push(skb, sizeof(struct frag_hdr)); @@ -723,7 +726,6 @@ slow_path: */ *prevhdr = NEXTHDR_FRAGMENT; - hroom = LL_RESERVED_SPACE(rt->dst.dev); troom = rt->dst.dev->needed_tailroom; /* From 34f5b0066435ffb793049b84fafd29fa195bcf90 Mon Sep 17 00:00:00 2001 From: Sasha Levin Date: Wed, 16 Sep 2015 15:30:21 -0400 Subject: [PATCH 187/647] atm: deal with setting entry before mkip was called If we didn't call ATMARP_MKIP before ATMARP_ENCAP the VCC descriptor is non-existant and we'll end up dereferencing a NULL ptr: [1033173.491930] kasan: GPF could be caused by NULL-ptr deref or user memory accessirq event stamp: 123386 [1033173.493678] general protection fault: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC KASAN [1033173.493689] Modules linked in: [1033173.493697] CPU: 9 PID: 23815 Comm: trinity-c64 Not tainted 4.2.0-next-20150911-sasha-00043-g353d875-dirty #2545 [1033173.493706] task: ffff8800630c4000 ti: ffff880063110000 task.ti: ffff880063110000 [1033173.493823] RIP: clip_ioctl (net/atm/clip.c:320 net/atm/clip.c:689) [1033173.493826] RSP: 0018:ffff880063117a88 EFLAGS: 00010203 [1033173.493828] RAX: dffffc0000000000 RBX: 0000000000000000 RCX: 000000000000000c [1033173.493830] RDX: 0000000000000002 RSI: ffffffffb3f10720 RDI: 0000000000000014 [1033173.493832] RBP: ffff880063117b80 R08: ffff88047574d9a4 R09: 0000000000000000 [1033173.493834] R10: 0000000000000000 R11: 0000000000000000 R12: 1ffff1000c622f53 [1033173.493836] R13: ffff8800cb905500 R14: ffff8808d6da2000 R15: 00000000fffffdfd [1033173.493840] FS: 00007fa56b92d700(0000) GS:ffff880478000000(0000) knlGS:0000000000000000 [1033173.493843] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [1033173.493845] CR2: 0000000000000000 CR3: 00000000630e8000 CR4: 00000000000006a0 [1033173.493855] Stack: [1033173.493862] ffffffffb0b60444 000000000000eaea 0000000041b58ab3 ffffffffb3c3ce32 [1033173.493867] ffffffffb0b6f3e0 ffffffffb0b60444 ffffffffb5ea2e50 1ffff1000c622f5e [1033173.493873] ffff8800630c4cd8 00000000000ee09a ffffffffb3ec4888 ffffffffb5ea2de8 [1033173.493874] Call Trace: [1033173.494108] do_vcc_ioctl (net/atm/ioctl.c:170) [1033173.494113] vcc_ioctl (net/atm/ioctl.c:189) [1033173.494116] svc_ioctl (net/atm/svc.c:605) [1033173.494200] sock_do_ioctl (net/socket.c:874) [1033173.494204] sock_ioctl (net/socket.c:958) [1033173.494244] do_vfs_ioctl (fs/ioctl.c:43 fs/ioctl.c:607) [1033173.494290] SyS_ioctl (fs/ioctl.c:622 fs/ioctl.c:613) [1033173.494295] entry_SYSCALL_64_fastpath (arch/x86/entry/entry_64.S:186) [1033173.494362] Code: fa 48 c1 ea 03 80 3c 02 00 0f 85 50 09 00 00 49 8b 9e 60 06 00 00 48 b8 00 00 00 00 00 fc ff df 48 8d 7b 14 48 89 fa 48 c1 ea 03 <0f> b6 04 02 48 89 fa 83 e2 07 38 d0 7f 08 84 c0 0f 85 14 09 00 All code ======== 0: fa cli 1: 48 c1 ea 03 shr $0x3,%rdx 5: 80 3c 02 00 cmpb $0x0,(%rdx,%rax,1) 9: 0f 85 50 09 00 00 jne 0x95f f: 49 8b 9e 60 06 00 00 mov 0x660(%r14),%rbx 16: 48 b8 00 00 00 00 00 movabs $0xdffffc0000000000,%rax 1d: fc ff df 20: 48 8d 7b 14 lea 0x14(%rbx),%rdi 24: 48 89 fa mov %rdi,%rdx 27: 48 c1 ea 03 shr $0x3,%rdx 2b:* 0f b6 04 02 movzbl (%rdx,%rax,1),%eax <-- trapping instruction 2f: 48 89 fa mov %rdi,%rdx 32: 83 e2 07 and $0x7,%edx 35: 38 d0 cmp %dl,%al 37: 7f 08 jg 0x41 39: 84 c0 test %al,%al 3b: 0f 85 14 09 00 00 jne 0x955 Code starting with the faulting instruction =========================================== 0: 0f b6 04 02 movzbl (%rdx,%rax,1),%eax 4: 48 89 fa mov %rdi,%rdx 7: 83 e2 07 and $0x7,%edx a: 38 d0 cmp %dl,%al c: 7f 08 jg 0x16 e: 84 c0 test %al,%al 10: 0f 85 14 09 00 00 jne 0x92a [1033173.494366] RIP clip_ioctl (net/atm/clip.c:320 net/atm/clip.c:689) [1033173.494368] RSP Signed-off-by: Sasha Levin Signed-off-by: David S. Miller --- net/atm/clip.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/atm/clip.c b/net/atm/clip.c index 17e55dfecbe2..e07f551a863c 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -317,6 +317,9 @@ static int clip_constructor(struct neighbour *neigh) static int clip_encap(struct atm_vcc *vcc, int mode) { + if (!CLIP_VCC(vcc)) + return -EBADFD; + CLIP_VCC(vcc)->encap = mode; return 0; } From 980137a20317055451a73547cf07be4ddea039ee Mon Sep 17 00:00:00 2001 From: Michael Grzeschik Date: Thu, 17 Sep 2015 15:18:34 +0200 Subject: [PATCH 188/647] ARCNET: fix hard_header_len limit For arcnet the bare minimum header only contains the 4 bytes to specify source, dest and offset (1, 1 and 2 bytes respectively). The corresponding struct is struct arc_hardware. The struct archdr contains additionally a union of possible soft headers. When doing $insertusecasehere packets might well include short (or even no?) soft headers. For this reason only use arc_hardware instead of archdr to determine the hard_header_len for an arcnet device. Signed-off-by: Michael Grzeschik Signed-off-by: David S. Miller --- drivers/net/arcnet/arcnet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/arcnet/arcnet.c b/drivers/net/arcnet/arcnet.c index 10f71c732b59..816d0e94961c 100644 --- a/drivers/net/arcnet/arcnet.c +++ b/drivers/net/arcnet/arcnet.c @@ -326,7 +326,7 @@ static void arcdev_setup(struct net_device *dev) dev->type = ARPHRD_ARCNET; dev->netdev_ops = &arcnet_netdev_ops; dev->header_ops = &arcnet_header_ops; - dev->hard_header_len = sizeof(struct archdr); + dev->hard_header_len = sizeof(struct arc_hardware); dev->mtu = choose_mtu(); dev->addr_len = ARCNET_ALEN; From c38f6ac74c99801360705d97244ff222ae18dc97 Mon Sep 17 00:00:00 2001 From: Michael Grzeschik Date: Thu, 17 Sep 2015 15:26:16 +0200 Subject: [PATCH 189/647] MAINTAINERS: add arcnet and take maintainership Add entry for arcnet to MAINTAINERS file and add myself as the maintainer of the subsystem. Signed-off-by: Michael Grzeschik Cc: davem@davemloft.net Cc: joe@perches.com Signed-off-by: David S. Miller --- MAINTAINERS | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index d4d9d4f6d271..c6aca87cd48a 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -808,6 +808,13 @@ S: Maintained F: drivers/video/fbdev/arcfb.c F: drivers/video/fbdev/core/fb_defio.c +ARCNET NETWORK LAYER +M: Michael Grzeschik +L: netdev@vger.kernel.org +S: Maintained +F: drivers/net/arcnet/ +F: include/uapi/linux/if_arcnet.h + ARM MFM AND FLOPPY DRIVERS M: Ian Molton S: Maintained From 9dc2ad1008c9f91f55ec6c89ec0f8639dfc91596 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Thu, 17 Sep 2015 16:11:10 +0200 Subject: [PATCH 190/647] vxlan: set needed headroom correctly vxlan_setup is called when allocating the net_device, i.e. way before vxlan_newlink (or vxlan_dev_configure) is called. This means vxlan->default_dst is actually unset in vxlan_setup and the condition that sets needed_headroom always takes the else branch. Set the needed_headrom at the point when we have the information about the address family available. Fixes: e4c7ed415387c ("vxlan: add ipv6 support") Fixes: 2853af6a2ea1a ("vxlan: use dev->needed_headroom instead of dev->hard_header_len") CC: Cong Wang Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- drivers/net/vxlan.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index cf8b7f0473b3..6ebe562af04e 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2392,10 +2392,6 @@ static void vxlan_setup(struct net_device *dev) eth_hw_addr_random(dev); ether_setup(dev); - if (vxlan->default_dst.remote_ip.sa.sa_family == AF_INET6) - dev->needed_headroom = ETH_HLEN + VXLAN6_HEADROOM; - else - dev->needed_headroom = ETH_HLEN + VXLAN_HEADROOM; dev->netdev_ops = &vxlan_netdev_ops; dev->destructor = free_netdev; @@ -2670,8 +2666,12 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev, dev->needed_headroom = lowerdev->hard_header_len + (use_ipv6 ? VXLAN6_HEADROOM : VXLAN_HEADROOM); - } else if (use_ipv6) + } else if (use_ipv6) { vxlan->flags |= VXLAN_F_IPV6; + dev->needed_headroom = ETH_HLEN + VXLAN6_HEADROOM; + } else { + dev->needed_headroom = ETH_HLEN + VXLAN_HEADROOM; + } memcpy(&vxlan->cfg, conf, sizeof(*conf)); if (!vxlan->cfg.dst_port) From 057ba29bbe85e9587635e3128b26fa30fe849af9 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Thu, 17 Sep 2015 16:11:11 +0200 Subject: [PATCH 191/647] vxlan: reject IPv6 addresses if IPv6 is not configured When IPv6 address is set without IPv6 configured, the vxlan socket is mostly treated as an IPv4 one but various lookus in fdb etc. still take the AF_INET6 into account. This creates incosistencies with weird consequences. Just reject IPv6 addresses in such case. Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- drivers/net/vxlan.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 6ebe562af04e..bbac1d35ed4e 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2636,8 +2636,11 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev, dst->remote_ip.sa.sa_family = AF_INET; if (dst->remote_ip.sa.sa_family == AF_INET6 || - vxlan->cfg.saddr.sa.sa_family == AF_INET6) + vxlan->cfg.saddr.sa.sa_family == AF_INET6) { + if (!IS_ENABLED(CONFIG_IPV6)) + return -EPFNOSUPPORT; use_ipv6 = true; + } if (conf->remote_ifindex) { struct net_device *lowerdev From 378fddc281072a10b621341b9f78c902303a8fe7 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Thu, 17 Sep 2015 16:11:12 +0200 Subject: [PATCH 192/647] qlcnic: track vxlan port count The callback for adding vxlan port can be called with the same port for both IPv4 and IPv6. Do not disable the offloading when the same port for both protocols is added and later one of them removed. Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qlcnic/qlcnic.h | 1 + .../net/ethernet/qlogic/qlcnic/qlcnic_main.c | 18 +++++++++++++----- 2 files changed, 14 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h index 06bcc734fe8d..d6696cfa11d2 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h @@ -536,6 +536,7 @@ struct qlcnic_hardware_context { u8 extend_lb_time; u8 phys_port_id[ETH_ALEN]; u8 lb_mode; + u8 vxlan_port_count; u16 vxlan_port; struct device *hwmon_dev; u32 post_mode; diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c index 8b08b20e8b30..d4481454b5f8 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c @@ -483,11 +483,17 @@ static void qlcnic_add_vxlan_port(struct net_device *netdev, /* Adapter supports only one VXLAN port. Use very first port * for enabling offload */ - if (!qlcnic_encap_rx_offload(adapter) || ahw->vxlan_port) + if (!qlcnic_encap_rx_offload(adapter)) return; + if (!ahw->vxlan_port_count) { + ahw->vxlan_port_count = 1; + ahw->vxlan_port = ntohs(port); + adapter->flags |= QLCNIC_ADD_VXLAN_PORT; + return; + } + if (ahw->vxlan_port == ntohs(port)) + ahw->vxlan_port_count++; - ahw->vxlan_port = ntohs(port); - adapter->flags |= QLCNIC_ADD_VXLAN_PORT; } static void qlcnic_del_vxlan_port(struct net_device *netdev, @@ -496,11 +502,13 @@ static void qlcnic_del_vxlan_port(struct net_device *netdev, struct qlcnic_adapter *adapter = netdev_priv(netdev); struct qlcnic_hardware_context *ahw = adapter->ahw; - if (!qlcnic_encap_rx_offload(adapter) || !ahw->vxlan_port || + if (!qlcnic_encap_rx_offload(adapter) || !ahw->vxlan_port_count || (ahw->vxlan_port != ntohs(port))) return; - adapter->flags |= QLCNIC_DEL_VXLAN_PORT; + ahw->vxlan_port_count--; + if (!ahw->vxlan_port_count) + adapter->flags |= QLCNIC_DEL_VXLAN_PORT; } static netdev_features_t qlcnic_features_check(struct sk_buff *skb, From 1e5b311ab2cc3baf73cdff066ab39c7466bf166b Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Thu, 17 Sep 2015 16:11:13 +0200 Subject: [PATCH 193/647] be2net: allow offloading with the same port for IPv4 and IPv6 The callback for adding vxlan port can be called with the same port for both IPv4 and IPv6. Do not disable the offloading if this occurs. Signed-off-by: Jiri Benc Acked-by: Sathya Perla Signed-off-by: David S. Miller --- drivers/net/ethernet/emulex/benet/be.h | 1 + drivers/net/ethernet/emulex/benet/be_main.c | 10 ++++++++++ 2 files changed, 11 insertions(+) diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index 0a27805cbbbd..821540913343 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -582,6 +582,7 @@ struct be_adapter { u16 pvid; __be16 vxlan_port; int vxlan_port_count; + int vxlan_port_aliases; struct phy_info phy; u8 wol_cap; bool wol_en; diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 12687bf52b95..7bf51a1a0a77 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -5176,6 +5176,11 @@ static void be_add_vxlan_port(struct net_device *netdev, sa_family_t sa_family, if (lancer_chip(adapter) || BEx_chip(adapter) || be_is_mc(adapter)) return; + if (adapter->vxlan_port == port && adapter->vxlan_port_count) { + adapter->vxlan_port_aliases++; + return; + } + if (adapter->flags & BE_FLAGS_VXLAN_OFFLOADS) { dev_info(dev, "Only one UDP port supported for VxLAN offloads\n"); @@ -5226,6 +5231,11 @@ static void be_del_vxlan_port(struct net_device *netdev, sa_family_t sa_family, if (adapter->vxlan_port != port) goto done; + if (adapter->vxlan_port_aliases) { + adapter->vxlan_port_aliases--; + return; + } + be_disable_vxlan_offloads(adapter); dev_info(&adapter->pdev->dev, From ac7eccd4d48fcc70d9fd6e4d10657bcde0a73f9f Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Thu, 17 Sep 2015 16:11:14 +0200 Subject: [PATCH 194/647] bnx2x: track vxlan port count The callback for adding vxlan port can be called with the same port for both IPv4 and IPv6. Do not disable the offloading when the same port for both protocols is added and later one of them removed. Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 1 + drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 14 ++++++++++++-- 2 files changed, 13 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index ba936635322a..b5e64b02200c 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -1946,6 +1946,7 @@ struct bnx2x { u16 vlan_cnt; u16 vlan_credit; u16 vxlan_dst_port; + u8 vxlan_dst_port_count; bool accept_any_vlan; }; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 89a174fa1300..f1d62d5dbaff 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -10108,12 +10108,18 @@ static void __bnx2x_add_vxlan_port(struct bnx2x *bp, u16 port) if (!netif_running(bp->dev)) return; - if (bp->vxlan_dst_port || !IS_PF(bp)) { + if (bp->vxlan_dst_port_count && bp->vxlan_dst_port == port) { + bp->vxlan_dst_port_count++; + return; + } + + if (bp->vxlan_dst_port_count || !IS_PF(bp)) { DP(BNX2X_MSG_SP, "Vxlan destination port limit reached\n"); return; } bp->vxlan_dst_port = port; + bp->vxlan_dst_port_count = 1; bnx2x_schedule_sp_rtnl(bp, BNX2X_SP_RTNL_ADD_VXLAN_PORT, 0); } @@ -10128,10 +10134,14 @@ static void bnx2x_add_vxlan_port(struct net_device *netdev, static void __bnx2x_del_vxlan_port(struct bnx2x *bp, u16 port) { - if (!bp->vxlan_dst_port || bp->vxlan_dst_port != port || !IS_PF(bp)) { + if (!bp->vxlan_dst_port_count || bp->vxlan_dst_port != port || + !IS_PF(bp)) { DP(BNX2X_MSG_SP, "Invalid vxlan port\n"); return; } + bp->vxlan_dst_port--; + if (bp->vxlan_dst_port) + return; if (netif_running(bp->dev)) { bnx2x_schedule_sp_rtnl(bp, BNX2X_SP_RTNL_DEL_VXLAN_PORT, 0); From 6cf35642147103195f126d13bdc1d4c32c7c6666 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Thu, 17 Sep 2015 16:28:31 +0200 Subject: [PATCH 195/647] MAINTAINERS: remove bouncing email address for qlcnic I got this automated message from when submitting a qlcnic patch: > Shahed Shaikh is no longer with QLogic. If you require assistance please > contact Ariel Elior Ariel.Elior@qlogic.com There's no point in having a bouncing address in MAINTAINERS. CC: Dept-GELinuxNICDev@qlogic.com CC: Ariel Elior Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index c6aca87cd48a..e1c9fbb8bf92 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8497,7 +8497,6 @@ F: Documentation/networking/LICENSE.qla3xxx F: drivers/net/ethernet/qlogic/qla3xxx.* QLOGIC QLCNIC (1/10)Gb ETHERNET DRIVER -M: Shahed Shaikh M: Dept-GELinuxNICDev@qlogic.com L: netdev@vger.kernel.org S: Supported From adb094e5e7285385770eb7a7c122bfc663c5e174 Mon Sep 17 00:00:00 2001 From: Taku Izumi Date: Thu, 17 Sep 2015 23:21:21 +0900 Subject: [PATCH 196/647] fjes: fix off-by-one error at fjes_hw_update_zone_task() Dan Carpenter reported off-by-one error of fjes at http://www.mail-archive.com/netdev@vger.kernel.org/msg77520.html Actually this is a bug. ep_shm_info[epidx].{es_status, zone} should be update inside for loop. This patch fixes this bug. Reported-by: Dan Carpenter Signed-off-by: Taku Izumi Signed-off-by: David S. Miller --- drivers/net/fjes/fjes_hw.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/fjes/fjes_hw.c b/drivers/net/fjes/fjes_hw.c index b5f4a78da828..2d3848c9dc35 100644 --- a/drivers/net/fjes/fjes_hw.c +++ b/drivers/net/fjes/fjes_hw.c @@ -1011,11 +1011,11 @@ static void fjes_hw_update_zone_task(struct work_struct *work) set_bit(epidx, &irq_bit); break; } + + hw->ep_shm_info[epidx].es_status = + info[epidx].es_status; + hw->ep_shm_info[epidx].zone = info[epidx].zone; } - - hw->ep_shm_info[epidx].es_status = info[epidx].es_status; - hw->ep_shm_info[epidx].zone = info[epidx].zone; - break; } From c2e7204d180f8efc80f27959ca9cf16fa17f67db Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 17 Sep 2015 08:38:00 -0700 Subject: [PATCH 197/647] tcp_cubic: do not set epoch_start in the future Tracking idle time in bictcp_cwnd_event() is imprecise, as epoch_start is normally set at ACK processing time, not at send time. Doing a proper fix would need to add an additional state variable, and does not seem worth the trouble, given CUBIC bug has been there forever before Jana noticed it. Let's simply not set epoch_start in the future, otherwise bictcp_update() could overflow and CUBIC would again grow cwnd too fast. This was detected thanks to a packetdrill test Neal wrote that was flaky before applying this fix. Fixes: 30927520dbae ("tcp_cubic: better follow cubic curve after idle period") Signed-off-by: Eric Dumazet Signed-off-by: Neal Cardwell Signed-off-by: Yuchung Cheng Cc: Jana Iyengar Signed-off-by: David S. Miller --- net/ipv4/tcp_cubic.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c index c6ded6b2a79f..448c2615fece 100644 --- a/net/ipv4/tcp_cubic.c +++ b/net/ipv4/tcp_cubic.c @@ -154,14 +154,20 @@ static void bictcp_init(struct sock *sk) static void bictcp_cwnd_event(struct sock *sk, enum tcp_ca_event event) { if (event == CA_EVENT_TX_START) { - s32 delta = tcp_time_stamp - tcp_sk(sk)->lsndtime; struct bictcp *ca = inet_csk_ca(sk); + u32 now = tcp_time_stamp; + s32 delta; + + delta = now - tcp_sk(sk)->lsndtime; /* We were application limited (idle) for a while. * Shift epoch_start to keep cwnd growth to cubic curve. */ - if (ca->epoch_start && delta > 0) + if (ca->epoch_start && delta > 0) { ca->epoch_start += delta; + if (after(ca->epoch_start, now)) + ca->epoch_start = now; + } return; } } From de9b8f5dcbd94bfb1d249907a635f1fb1968e19c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 13 Aug 2015 23:09:29 +0200 Subject: [PATCH 198/647] sched: Fix crash trying to dequeue/enqueue the idle thread Sasha reports that his virtual machine tries to schedule the idle thread since commit 6c37067e2786 ("sched: Change the sched_class::set_cpus_allowed() calling context"). Hit trace shows this happening from idle_thread_get()->init_idle(), which is the _second_ init_idle() invocation on that task_struct, the first being done through idle_init()->fork_idle(). (this code is insane...) Because we call init_idle() twice in a row, its ->sched_class == &idle_sched_class and ->on_rq = TASK_ON_RQ_QUEUED. This means do_set_cpus_allowed() think we're queued and will call dequeue_task(), which is implemented with BUG() for the idle class, seeing how dequeueing the idle task is a daft thing. Aside of the whole insanity of calling init_idle() _twice_, change the code to call set_cpus_allowed_common() instead as this is 'obviously' before the idle task gets ran etc.. Reported-by: Sasha Levin Tested-by: Sasha Levin Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Fixes: 6c37067e2786 ("sched: Change the sched_class::set_cpus_allowed() calling context") Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 97d276ff1edb..f0d043ec0182 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -4927,7 +4927,15 @@ void init_idle(struct task_struct *idle, int cpu) idle->state = TASK_RUNNING; idle->se.exec_start = sched_clock(); - do_set_cpus_allowed(idle, cpumask_of(cpu)); +#ifdef CONFIG_SMP + /* + * Its possible that init_idle() gets called multiple times on a task, + * in that case do_set_cpus_allowed() will not do the right thing. + * + * And since this is boot we can forgo the serialization. + */ + set_cpus_allowed_common(idle, cpumask_of(cpu)); +#endif /* * We're having a chicken and egg problem, even though we are * holding rq->lock, the cpu isn't yet set to this cpu so the @@ -4944,7 +4952,7 @@ void init_idle(struct task_struct *idle, int cpu) rq->curr = rq->idle = idle; idle->on_rq = TASK_ON_RQ_QUEUED; -#if defined(CONFIG_SMP) +#ifdef CONFIG_SMP idle->on_cpu = 1; #endif raw_spin_unlock(&rq->lock); @@ -4959,7 +4967,7 @@ void init_idle(struct task_struct *idle, int cpu) idle->sched_class = &idle_sched_class; ftrace_graph_init_idle_task(idle, cpu); vtime_init_idle(idle, cpu); -#if defined(CONFIG_SMP) +#ifdef CONFIG_SMP sprintf(idle->comm, "%s/%d", INIT_TASK_COMM, cpu); #endif } From 5e176213a6b2bc5146820c79542d37290434a3c4 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Mon, 14 Sep 2015 14:47:02 -0700 Subject: [PATCH 199/647] perf/x86/intel: Make the CYCLE_ACTIVITY.* constraint on Broadwell more specific The counter constraint for CYCLE_ACTIVITY.* on Broadwell covered all CYCLE_ACTIVITY.* sub events, and forced them on counter 2. But actually only one sub event (umask 8) needs to be on counter 2, all others do not have any constraint. Only force that subevent. This fixes groups with multiple CYCLE_ACTIVITY.* events, for example: % perf stat -x, -e '{cpu/event=0xa3,umask=0x6,cmask=6/,\ cpu/event=0xa2,umask=0x8/,\ cpu/event=0xa3,umask=0x4,cmask=4/,cpu/event=0xb1,umask=0x1,cmask=1/}' true 122150,,cpu/event=0xa3,umask=0x6,cmask=6/,846486,100.00 16483,,cpu/event=0xa2,umask=0x8/,846486,100.00 252280,,cpu/event=0xa3,umask=0x4,cmask=4/,846486,100.00 233604,,cpu/event=0xb1,umask=0x1,cmask=1/,846486,100.00 % Without this patch the third result would be Signed-off-by: Andi Kleen Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1442267222-16464-1-git-send-email-andi@firstfloor.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_intel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 3fefebfbdf4b..1d84b41ba932 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -250,7 +250,7 @@ struct event_constraint intel_bdw_event_constraints[] = { FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */ INTEL_UEVENT_CONSTRAINT(0x148, 0x4), /* L1D_PEND_MISS.PENDING */ - INTEL_EVENT_CONSTRAINT(0xa3, 0x4), /* CYCLE_ACTIVITY.* */ + INTEL_UEVENT_CONSTRAINT(0x8a3, 0x4), /* CYCLE_ACTIVITY.CYCLES_L1D_MISS */ EVENT_CONSTRAINT_END }; From d0dc8494cd6904f8ad035d9ad97f313948f35d0c Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 9 Sep 2015 14:53:59 -0700 Subject: [PATCH 200/647] perf/x86/intel/pebs: Add PEBS frontend profiling for Skylake Skylake has a new FRONTEND_LATENCY PEBS event to accurately profile frontend problems (like ITLB or decoding issues). The new event is configured through a separate MSR, which selects a range of sub events. Define the extra MSR as a extra reg and export support for it through sysfs. To avoid duplicating the existing tables use a new function to add new entries to existing tables. Signed-off-by: Andi Kleen Signed-off-by: Peter Zijlstra (Intel) Acked-by: Thomas Gleixner Cc: Linus Torvalds Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1435707205-6676-4-git-send-email-andi@firstfloor.org Signed-off-by: Ingo Molnar --- arch/x86/include/asm/msr-index.h | 2 ++ arch/x86/kernel/cpu/perf_event.h | 1 + arch/x86/kernel/cpu/perf_event_intel.c | 11 ++++++++++- 3 files changed, 13 insertions(+), 1 deletion(-) diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index c1c0a1c14344..54390bc140dd 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -141,6 +141,8 @@ #define DEBUGCTLMSR_BTS_OFF_USR (1UL << 10) #define DEBUGCTLMSR_FREEZE_LBRS_ON_PMI (1UL << 11) +#define MSR_PEBS_FRONTEND 0x000003f7 + #define MSR_IA32_POWER_CTL 0x000001fc #define MSR_IA32_MC0_CTL 0x00000400 diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h index 5edf6d868fc1..165be83a7fa4 100644 --- a/arch/x86/kernel/cpu/perf_event.h +++ b/arch/x86/kernel/cpu/perf_event.h @@ -47,6 +47,7 @@ enum extra_reg_type { EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */ EXTRA_REG_LBR = 2, /* lbr_select */ EXTRA_REG_LDLAT = 3, /* ld_lat_threshold */ + EXTRA_REG_FE = 4, /* fe_* */ EXTRA_REG_MAX /* number of entries needed */ }; diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 1d84b41ba932..ef74c9d05f82 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -205,6 +205,7 @@ static struct extra_reg intel_skl_extra_regs[] __read_mostly = { INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x3fffff8fffull, RSP_0), INTEL_UEVENT_EXTRA_REG(0x01bb, MSR_OFFCORE_RSP_1, 0x3fffff8fffull, RSP_1), INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd), + INTEL_UEVENT_EXTRA_REG(0x01c6, MSR_PEBS_FRONTEND, 0x3fff17, FE), EVENT_EXTRA_END }; @@ -2891,6 +2892,8 @@ PMU_FORMAT_ATTR(offcore_rsp, "config1:0-63"); PMU_FORMAT_ATTR(ldlat, "config1:0-15"); +PMU_FORMAT_ATTR(frontend, "config1:0-23"); + static struct attribute *intel_arch3_formats_attr[] = { &format_attr_event.attr, &format_attr_umask.attr, @@ -2907,6 +2910,11 @@ static struct attribute *intel_arch3_formats_attr[] = { NULL, }; +static struct attribute *skl_format_attr[] = { + &format_attr_frontend.attr, + NULL, +}; + static __initconst const struct x86_pmu core_pmu = { .name = "core", .handle_irq = x86_pmu_handle_irq, @@ -3516,7 +3524,8 @@ __init int intel_pmu_init(void) x86_pmu.hw_config = hsw_hw_config; x86_pmu.get_event_constraints = hsw_get_event_constraints; - x86_pmu.cpu_events = hsw_events_attrs; + x86_pmu.format_attrs = merge_attr(intel_arch3_formats_attr, + skl_format_attr); WARN_ON(!x86_pmu.format_attrs); x86_pmu.cpu_events = hsw_events_attrs; pr_cont("Skylake events, "); From dfe1f3cb312624928052413928d88b0ee3492216 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 9 Sep 2015 14:54:00 -0700 Subject: [PATCH 201/647] perf/x86/intel: Fix Skylake FRONTEND MSR extrareg mask Stephane pointed out that the extrareg mask was one bit too short. The bubble width field was truncated by one bit. Fix that here. Also add some extra comments on the reserved bits inside the event select code. Reported-by: Stephane Eranian Signed-off-by: Andi Kleen Signed-off-by: Peter Zijlstra (Intel) Acked-by: Thomas Gleixner Cc: Linus Torvalds Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1441835640-21347-3-git-send-email-andi@firstfloor.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_intel.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index ef74c9d05f82..f63360be2238 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -205,7 +205,11 @@ static struct extra_reg intel_skl_extra_regs[] __read_mostly = { INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x3fffff8fffull, RSP_0), INTEL_UEVENT_EXTRA_REG(0x01bb, MSR_OFFCORE_RSP_1, 0x3fffff8fffull, RSP_1), INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd), - INTEL_UEVENT_EXTRA_REG(0x01c6, MSR_PEBS_FRONTEND, 0x3fff17, FE), + /* + * Note the low 8 bits eventsel code is not a continuous field, containing + * some #GPing bits. These are masked out. + */ + INTEL_UEVENT_EXTRA_REG(0x01c6, MSR_PEBS_FRONTEND, 0x7fff17, FE), EVENT_EXTRA_END }; From f55fc2a57cc9ca3b1bb4fb8eb25b6e1989e5b993 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 9 Sep 2015 19:06:33 +0200 Subject: [PATCH 202/647] perf: Restructure perf syscall point of no return The exclusive_event_installable() stuff only works because its exclusive with the grouping bits. Rework the code such that there is a sane place to error out before we go do things we cannot undo. Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar --- kernel/events/core.c | 49 +++++++++++++++++++++++++++++--------------- 1 file changed, 32 insertions(+), 17 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index f548f69c4299..39679f749500 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -8297,13 +8297,30 @@ SYSCALL_DEFINE5(perf_event_open, if (move_group) { gctx = group_leader->ctx; + mutex_lock_double(&gctx->mutex, &ctx->mutex); + } else { + mutex_lock(&ctx->mutex); + } + /* + * Must be under the same ctx::mutex as perf_install_in_context(), + * because we need to serialize with concurrent event creation. + */ + if (!exclusive_event_installable(event, ctx)) { + /* exclusive and group stuff are assumed mutually exclusive */ + WARN_ON_ONCE(move_group); + + err = -EBUSY; + goto err_locked; + } + + WARN_ON_ONCE(ctx->parent_ctx); + + if (move_group) { /* * See perf_event_ctx_lock() for comments on the details * of swizzling perf_event::ctx. */ - mutex_lock_double(&gctx->mutex, &ctx->mutex); - perf_remove_from_context(group_leader, false); list_for_each_entry(sibling, &group_leader->sibling_list, @@ -8311,13 +8328,7 @@ SYSCALL_DEFINE5(perf_event_open, perf_remove_from_context(sibling, false); put_ctx(gctx); } - } else { - mutex_lock(&ctx->mutex); - } - WARN_ON_ONCE(ctx->parent_ctx); - - if (move_group) { /* * Wait for everybody to stop referencing the events through * the old lists, before installing it on new lists. @@ -8349,22 +8360,20 @@ SYSCALL_DEFINE5(perf_event_open, perf_event__state_init(group_leader); perf_install_in_context(ctx, group_leader, group_leader->cpu); get_ctx(ctx); - } - if (!exclusive_event_installable(event, ctx)) { - err = -EBUSY; - mutex_unlock(&ctx->mutex); - fput(event_file); - goto err_context; + /* + * Now that all events are installed in @ctx, nothing + * references @gctx anymore, so drop the last reference we have + * on it. + */ + put_ctx(gctx); } perf_install_in_context(ctx, event, event->cpu); perf_unpin_context(ctx); - if (move_group) { + if (move_group) mutex_unlock(&gctx->mutex); - put_ctx(gctx); - } mutex_unlock(&ctx->mutex); put_online_cpus(); @@ -8391,6 +8400,12 @@ SYSCALL_DEFINE5(perf_event_open, fd_install(event_fd, event_file); return event_fd; +err_locked: + if (move_group) + mutex_unlock(&gctx->mutex); + mutex_unlock(&ctx->mutex); +/* err_file: */ + fput(event_file); err_context: perf_unpin_context(ctx); put_ctx(ctx); From a723968c0ed36db676478c3d26078f13484fe01c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 9 Sep 2015 19:06:33 +0200 Subject: [PATCH 203/647] perf: Fix u16 overflows Vince reported that its possible to overflow the various size fields and get weird stuff if you stick too many events in a group. Put a lid on this by requiring the fixed record size not exceed 16k. This is still a fair amount of events (silly amount really) and leaves plenty room for callchains and stack dwarves while also avoiding overflowing the u16 variables. Reported-by: Vince Weaver Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar --- kernel/events/core.c | 50 +++++++++++++++++++++++++++++++++++--------- 1 file changed, 40 insertions(+), 10 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 39679f749500..dbb5329b6a3a 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1243,11 +1243,7 @@ static inline void perf_event__state_init(struct perf_event *event) PERF_EVENT_STATE_INACTIVE; } -/* - * Called at perf_event creation and when events are attached/detached from a - * group. - */ -static void perf_event__read_size(struct perf_event *event) +static void __perf_event_read_size(struct perf_event *event, int nr_siblings) { int entry = sizeof(u64); /* value */ int size = 0; @@ -1263,7 +1259,7 @@ static void perf_event__read_size(struct perf_event *event) entry += sizeof(u64); if (event->attr.read_format & PERF_FORMAT_GROUP) { - nr += event->group_leader->nr_siblings; + nr += nr_siblings; size += sizeof(u64); } @@ -1271,14 +1267,11 @@ static void perf_event__read_size(struct perf_event *event) event->read_size = size; } -static void perf_event__header_size(struct perf_event *event) +static void __perf_event_header_size(struct perf_event *event, u64 sample_type) { struct perf_sample_data *data; - u64 sample_type = event->attr.sample_type; u16 size = 0; - perf_event__read_size(event); - if (sample_type & PERF_SAMPLE_IP) size += sizeof(data->ip); @@ -1303,6 +1296,17 @@ static void perf_event__header_size(struct perf_event *event) event->header_size = size; } +/* + * Called at perf_event creation and when events are attached/detached from a + * group. + */ +static void perf_event__header_size(struct perf_event *event) +{ + __perf_event_read_size(event, + event->group_leader->nr_siblings); + __perf_event_header_size(event, event->attr.sample_type); +} + static void perf_event__id_header_size(struct perf_event *event) { struct perf_sample_data *data; @@ -1330,6 +1334,27 @@ static void perf_event__id_header_size(struct perf_event *event) event->id_header_size = size; } +static bool perf_event_validate_size(struct perf_event *event) +{ + /* + * The values computed here will be over-written when we actually + * attach the event. + */ + __perf_event_read_size(event, event->group_leader->nr_siblings + 1); + __perf_event_header_size(event, event->attr.sample_type & ~PERF_SAMPLE_READ); + perf_event__id_header_size(event); + + /* + * Sum the lot; should not exceed the 64k limit we have on records. + * Conservative limit to allow for callchains and other variable fields. + */ + if (event->read_size + event->header_size + + event->id_header_size + sizeof(struct perf_event_header) >= 16*1024) + return false; + + return true; +} + static void perf_group_attach(struct perf_event *event) { struct perf_event *group_leader = event->group_leader, *pos; @@ -8302,6 +8327,11 @@ SYSCALL_DEFINE5(perf_event_open, mutex_lock(&ctx->mutex); } + if (!perf_event_validate_size(event)) { + err = -E2BIG; + goto err_locked; + } + /* * Must be under the same ctx::mutex as perf_install_in_context(), * because we need to serialize with concurrent event creation. From f73e22ab450140830005581c2c7ec389791a1b8d Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 9 Sep 2015 20:48:22 +0200 Subject: [PATCH 204/647] perf: Fix races in computing the header sizes There are two races with the current code: - Another event can join the group and compute a larger header_size concurrently, if the smaller store wins we'll have an incorrect header_size set. - We compute the header_size after the event becomes active, therefore its possible to use the size before its computed. Remedy the first by moving the computation inside the ctx::mutex lock, and the second by placing it _before_ perf_install_in_context(). Signed-off-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar --- kernel/events/core.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index dbb5329b6a3a..b11756f9b6dc 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -8399,6 +8399,15 @@ SYSCALL_DEFINE5(perf_event_open, put_ctx(gctx); } + /* + * Precalculate sample_data sizes; do while holding ctx::mutex such + * that we're serialized against further additions and before + * perf_install_in_context() which is the point the event is active and + * can use these values. + */ + perf_event__header_size(event); + perf_event__id_header_size(event); + perf_install_in_context(ctx, event, event->cpu); perf_unpin_context(ctx); @@ -8414,12 +8423,6 @@ SYSCALL_DEFINE5(perf_event_open, list_add_tail(&event->owner_entry, ¤t->perf_event_list); mutex_unlock(¤t->perf_event_mutex); - /* - * Precalculate sample_data sizes - */ - perf_event__header_size(event); - perf_event__id_header_size(event); - /* * Drop the reference on the group_event after placing the * new event on the sibling_list. This ensures destruction From ebae871a509d3c24b32ff67af2671dadffc58770 Mon Sep 17 00:00:00 2001 From: Igor Mammedov Date: Fri, 18 Sep 2015 15:39:05 +0200 Subject: [PATCH 205/647] kvm: svm: reset mmu on VCPU reset MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When INIT/SIPI sequence is sent to VCPU which before that was in use by OS, VMRUN might fail with: KVM: entry failed, hardware error 0xffffffff EAX=00000000 EBX=00000000 ECX=00000000 EDX=000006d3 ESI=00000000 EDI=00000000 EBP=00000000 ESP=00000000 EIP=00000000 EFL=00000002 [-------] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0000 00000000 0000ffff 00009300 CS =9a00 0009a000 0000ffff 00009a00 [...] CR0=60000010 CR2=b6f3e000 CR3=01942000 CR4=000007e0 [...] EFER=0000000000000000 with corresponding SVM error: KVM: FAILED VMRUN WITH VMCB: [...] cpl: 0 efer: 0000000000001000 cr0: 0000000080010010 cr2: 00007fd7fe85bf90 cr3: 0000000187d0c000 cr4: 0000000000000020 [...] What happens is that VCPU state right after offlinig: CR0: 0x80050033 EFER: 0xd01 CR4: 0x7e0 -> long mode with CR3 pointing to longmode page tables and when VCPU gets INIT/SIPI following transition happens CR0: 0 -> 0x60000010 EFER: 0x0 CR4: 0x7e0 -> paging disabled with stale CR3 However SVM under the hood puts VCPU in Paged Real Mode* which effectively translates CR0 0x60000010 -> 80010010 after svm_vcpu_reset() -> init_vmcb() -> kvm_set_cr0() -> svm_set_cr0() but from kvm_set_cr0() perspective CR0: 0 -> 0x60000010 only caching bits are changed and commit d81135a57aa6 ("KVM: x86: do not reset mmu if CR0.CD and CR0.NW are changed")' regressed svm_vcpu_reset() which relied on MMU being reset. As result VMRUN after svm_vcpu_reset() tries to run VCPU in Paged Real Mode with stale MMU context (longmode page tables), which causes some AMD CPUs** to bail out with VMEXIT_INVALID. Fix issue by unconditionally resetting MMU context at init_vmcb() time. * AMD64 Architecture Programmer’s Manual, Volume 2: System Programming, rev: 3.25 15.19 Paged Real Mode ** Opteron 1216 Signed-off-by: Igor Mammedov Fixes: d81135a57aa6 Cc: stable@vger.kernel.org Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index fdb8cb63a6c0..89173af53c84 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1264,6 +1264,7 @@ static void init_vmcb(struct vcpu_svm *svm, bool init_event) * It also updates the guest-visible cr0 value. */ (void)kvm_set_cr0(&svm->vcpu, X86_CR0_NW | X86_CR0_CD | X86_CR0_ET); + kvm_mmu_reset_context(&svm->vcpu); save->cr4 = X86_CR4_PAE; /* rdx = ?? */ From ab1fffe3a73c694d698645451ba61255ec4ba5e6 Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Fri, 18 Sep 2015 15:02:50 +0300 Subject: [PATCH 206/647] ASoC: davinci-mcasp: Fix devm_kasprintf format string The '\n' at the end of the format string is not needed. It adds an extra line break when doing cat /proc/interrupts Signed-off-by: Peter Ujfalusi Signed-off-by: Mark Brown --- sound/soc/davinci/davinci-mcasp.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/sound/soc/davinci/davinci-mcasp.c b/sound/soc/davinci/davinci-mcasp.c index 2c5a2c5444ab..7d45d98a861f 100644 --- a/sound/soc/davinci/davinci-mcasp.c +++ b/sound/soc/davinci/davinci-mcasp.c @@ -1685,7 +1685,7 @@ static int davinci_mcasp_probe(struct platform_device *pdev) irq = platform_get_irq_byname(pdev, "common"); if (irq >= 0) { - irq_name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "%s_common\n", + irq_name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "%s_common", dev_name(&pdev->dev)); ret = devm_request_threaded_irq(&pdev->dev, irq, NULL, davinci_mcasp_common_irq_handler, @@ -1702,7 +1702,7 @@ static int davinci_mcasp_probe(struct platform_device *pdev) irq = platform_get_irq_byname(pdev, "rx"); if (irq >= 0) { - irq_name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "%s_rx\n", + irq_name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "%s_rx", dev_name(&pdev->dev)); ret = devm_request_threaded_irq(&pdev->dev, irq, NULL, davinci_mcasp_rx_irq_handler, @@ -1717,7 +1717,7 @@ static int davinci_mcasp_probe(struct platform_device *pdev) irq = platform_get_irq_byname(pdev, "tx"); if (irq >= 0) { - irq_name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "%s_tx\n", + irq_name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "%s_tx", dev_name(&pdev->dev)); ret = devm_request_threaded_irq(&pdev->dev, irq, NULL, davinci_mcasp_tx_irq_handler, From 381c02f6d8ccad8ed574630f879c40fb59715124 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Wed, 16 Sep 2015 18:18:49 +0100 Subject: [PATCH 207/647] perf record: Avoid infinite loop at buildid processing with no samples If a session contains no events, we can get stuck in an infinite loop in __perf_session__process_events, with a non-zero file_size and data_offset, but a zero data_size. In this case, we can mmap the entirety of the file (consisting of the file and attribute headers), and fetch_mmaped_event will correctly refuse to read any (unmapped and non-existent) event headers. This causes __perf_session__process_events to unmap the file and retry with the exact same parameters, getting stuck in an infinite loop. This has been observed to result in an exit-time hang when counting rare/unschedulable events with perf record, and can be triggered artificially with the script below: ---- #!/bin/sh printf "REPRO: launching perf\n"; ./perf record -e software/config=9/ sleep 1 & PERF_PID=$!; sleep 0.002; kill -2 $PERF_PID; printf "REPRO: waiting for perf (%d) to exit...\n" "$PERF_PID"; wait $PERF_PID; printf "REPRO: perf exited\n"; ---- To avoid this, have __perf_session__process_events bail out early when the file has no data (i.e. it has no events). Commiter note: I only managed to reproduce this when setting /proc/sys/kernel/kptr_restrict to '1' and changing the code to purposefully not process any samples and no synthesized samples, i.e. kptr_restrict prevents 'record' from synthesizing the kernel mmaps for vmlinux + modules and since it is a workload started from perf, we don't synthesize mmap/comm records for existing threads. Adrian Hunter managed to reproduce it in his environment tho. Signed-off-by: Mark Rutland Tested-by: Arnaldo Carvalho de Melo Tested-by: Adrian Hunter Cc: Jiri Olsa Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1442423929-12253-1-git-send-email-mark.rutland@arm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/session.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index 8a4537ee9bc3..fc3f7c922f99 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -1580,7 +1580,10 @@ static int __perf_session__process_events(struct perf_session *session, file_offset = page_offset; head = data_offset - page_offset; - if (data_size && (data_offset + data_size < file_size)) + if (data_size == 0) + goto out; + + if (data_offset + data_size < file_size) file_size = data_offset + data_size; ui_progress__init(&prog, file_size, "Processing events..."); From ee92cfb030c16ddb01f6543968f13bcb61ed9da5 Mon Sep 17 00:00:00 2001 From: Zidan Wang Date: Fri, 18 Sep 2015 17:19:25 +0800 Subject: [PATCH 208/647] ASoC: wm8962: remove 64k sample rate support wm8962 can't support 64k sample rate. When playing a 64KHz wave file, 'Unsupported rate 64000Hz' will be prompted. Signed-off-by: Zidan Wang Acked-by: Charles Keepax Signed-off-by: Mark Brown --- sound/soc/codecs/wm8962.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/sound/soc/codecs/wm8962.c b/sound/soc/codecs/wm8962.c index b4eb975da981..293e47a6ff59 100644 --- a/sound/soc/codecs/wm8962.c +++ b/sound/soc/codecs/wm8962.c @@ -2944,7 +2944,8 @@ static int wm8962_mute(struct snd_soc_dai *dai, int mute) WM8962_DAC_MUTE, val); } -#define WM8962_RATES SNDRV_PCM_RATE_8000_96000 +#define WM8962_RATES (SNDRV_PCM_RATE_8000_48000 |\ + SNDRV_PCM_RATE_88200 | SNDRV_PCM_RATE_96000) #define WM8962_FORMATS (SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S20_3LE |\ SNDRV_PCM_FMTBIT_S24_LE | SNDRV_PCM_FMTBIT_S32_LE) From 8524bb0c7ac688a3cd6ba12dae6104c54d0566b9 Mon Sep 17 00:00:00 2001 From: Zidan Wang Date: Fri, 18 Sep 2015 17:19:43 +0800 Subject: [PATCH 209/647] ASoC: wm8960: correct the max register value of mic boost pga the max register value of mic boost pga should be 3. Signed-off-by: Zidan Wang Acked-by: Charles Keepax Signed-off-by: Mark Brown --- sound/soc/codecs/wm8960.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/codecs/wm8960.c b/sound/soc/codecs/wm8960.c index 70816f02ea8c..d9b43b00244e 100644 --- a/sound/soc/codecs/wm8960.c +++ b/sound/soc/codecs/wm8960.c @@ -234,9 +234,9 @@ SOC_SINGLE_TLV("Left Input Boost Mixer LINPUT3 Volume", SOC_SINGLE_TLV("Left Input Boost Mixer LINPUT2 Volume", WM8960_INBMIX2, 1, 7, 0, lineinboost_tlv), SOC_SINGLE_TLV("Right Input Boost Mixer RINPUT1 Volume", - WM8960_RINPATH, 4, 4, 0, micboost_tlv), + WM8960_RINPATH, 4, 3, 0, micboost_tlv), SOC_SINGLE_TLV("Left Input Boost Mixer LINPUT1 Volume", - WM8960_LINPATH, 4, 4, 0, micboost_tlv), + WM8960_LINPATH, 4, 3, 0, micboost_tlv), SOC_DOUBLE_R_TLV("Playback Volume", WM8960_LDAC, WM8960_RDAC, 0, 255, 0, dac_tlv), From d702ffd4d1df73b9c620af1654af42ff5b8d5c09 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 19:09:07 +0200 Subject: [PATCH 210/647] regulator: anatop: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: Mark Brown --- drivers/regulator/anatop-regulator.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/regulator/anatop-regulator.c b/drivers/regulator/anatop-regulator.c index 738adfa5332b..52ea605f8130 100644 --- a/drivers/regulator/anatop-regulator.c +++ b/drivers/regulator/anatop-regulator.c @@ -318,6 +318,7 @@ static const struct of_device_id of_anatop_regulator_match_tbl[] = { { .compatible = "fsl,anatop-regulator", }, { /* end */ } }; +MODULE_DEVICE_TABLE(of, of_anatop_regulator_match_tbl); static struct platform_driver anatop_regulator_driver = { .driver = { From 2f9481e7dc0d3aacbaa07701f3ee2527f5d48301 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 19:09:24 +0200 Subject: [PATCH 211/647] regulator: gpio: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: Mark Brown --- drivers/regulator/gpio-regulator.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/regulator/gpio-regulator.c b/drivers/regulator/gpio-regulator.c index 464018de7e97..7bba8b747f30 100644 --- a/drivers/regulator/gpio-regulator.c +++ b/drivers/regulator/gpio-regulator.c @@ -394,6 +394,7 @@ static const struct of_device_id regulator_gpio_of_match[] = { { .compatible = "regulator-gpio", }, {}, }; +MODULE_DEVICE_TABLE(of, regulator_gpio_of_match); #endif static struct platform_driver gpio_regulator_driver = { From 7209fee89f435b69051bb6bffe7f191336ac2a5e Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 19:09:51 +0200 Subject: [PATCH 212/647] regulator: vexpress: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: Mark Brown --- drivers/regulator/vexpress.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/regulator/vexpress.c b/drivers/regulator/vexpress.c index bed9d3ee4198..c810cbbd463f 100644 --- a/drivers/regulator/vexpress.c +++ b/drivers/regulator/vexpress.c @@ -103,6 +103,7 @@ static const struct of_device_id vexpress_regulator_of_match[] = { { .compatible = "arm,vexpress-volt", }, { } }; +MODULE_DEVICE_TABLE(of, vexpress_regulator_of_match); static struct platform_driver vexpress_regulator_driver = { .probe = vexpress_regulator_probe, From c9e97b3cb2b80deb94c092a2022a6d385b838d84 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 19:42:17 +0200 Subject: [PATCH 213/647] spi: meson: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: Mark Brown --- drivers/spi/spi-meson-spifc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/spi/spi-meson-spifc.c b/drivers/spi/spi-meson-spifc.c index 5468fc70dbf8..2465259f6241 100644 --- a/drivers/spi/spi-meson-spifc.c +++ b/drivers/spi/spi-meson-spifc.c @@ -444,6 +444,7 @@ static const struct of_device_id meson_spifc_dt_match[] = { { .compatible = "amlogic,meson6-spifc", }, { }, }; +MODULE_DEVICE_TABLE(of, meson_spifc_dt_match); static struct platform_driver meson_spifc_driver = { .probe = meson_spifc_probe, From 2ace47be5a315def8f493ca77aa59c077ade30a1 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Fri, 18 Sep 2015 16:02:20 +0530 Subject: [PATCH 214/647] ASoC: wm0010: fix memory leak We have requested for the firmware but we have missed releasing it both on success and on error path. While checking the code it turned out that the requested firmware is not even used. More over the same firmware is being loaded by wm0010_stage2_load(). Signed-off-by: Sudip Mukherjee Acked-by: Charles Keepax Signed-off-by: Mark Brown --- sound/soc/codecs/wm0010.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/sound/soc/codecs/wm0010.c b/sound/soc/codecs/wm0010.c index 9d370a45abe8..7a1bc40538ca 100644 --- a/sound/soc/codecs/wm0010.c +++ b/sound/soc/codecs/wm0010.c @@ -577,7 +577,6 @@ static int wm0010_boot(struct snd_soc_codec *codec) struct wm0010_priv *wm0010 = snd_soc_codec_get_drvdata(codec); unsigned long flags; int ret; - const struct firmware *fw; struct spi_message m; struct spi_transfer t; struct dfw_pllrec pll_rec; @@ -623,14 +622,6 @@ static int wm0010_boot(struct snd_soc_codec *codec) wm0010->state = WM0010_OUT_OF_RESET; spin_unlock_irqrestore(&wm0010->irq_lock, flags); - /* First the bootloader */ - ret = request_firmware(&fw, "wm0010_stage2.bin", codec->dev); - if (ret != 0) { - dev_err(codec->dev, "Failed to request stage2 loader: %d\n", - ret); - goto abort; - } - if (!wait_for_completion_timeout(&wm0010->boot_completion, msecs_to_jiffies(20))) dev_err(codec->dev, "Failed to get interrupt from DSP\n"); From f072f91aa7517386344476813ca0799e08fd0c35 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Fri, 18 Sep 2015 16:02:21 +0530 Subject: [PATCH 215/647] ASoC: wm0010: fix error path Fix the error path so that we can free the allocated memory on the error path instead of releasing them individually on each error. Signed-off-by: Sudip Mukherjee Acked-by: Charles Keepax Signed-off-by: Mark Brown --- sound/soc/codecs/wm0010.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) diff --git a/sound/soc/codecs/wm0010.c b/sound/soc/codecs/wm0010.c index 7a1bc40538ca..76b2f88d9d9b 100644 --- a/sound/soc/codecs/wm0010.c +++ b/sound/soc/codecs/wm0010.c @@ -663,10 +663,8 @@ static int wm0010_boot(struct snd_soc_codec *codec) } img_swap = kzalloc(len, GFP_KERNEL | GFP_DMA); - if (!img_swap) { - kfree(out); - goto abort; - } + if (!img_swap) + goto abort_out; /* We need to re-order for 0010 */ byte_swap_64((u64 *)&pll_rec, img_swap, len); @@ -681,20 +679,16 @@ static int wm0010_boot(struct snd_soc_codec *codec) spi_message_add_tail(&t, &m); ret = spi_sync(spi, &m); - if (ret != 0) { + if (ret) { dev_err(codec->dev, "First PLL write failed: %d\n", ret); - kfree(img_swap); - kfree(out); - goto abort; + goto abort_swap; } /* Use a second send of the message to get the return status */ ret = spi_sync(spi, &m); - if (ret != 0) { + if (ret) { dev_err(codec->dev, "Second PLL write failed: %d\n", ret); - kfree(img_swap); - kfree(out); - goto abort; + goto abort_swap; } p = (u32 *)out; @@ -727,6 +721,10 @@ static int wm0010_boot(struct snd_soc_codec *codec) return 0; +abort_swap: + kfree(img_swap); +abort_out: + kfree(out); abort: /* Put the chip back into reset */ wm0010_halt(codec); From 5b64c173cdea21105eb4794487b3d593f0a2e6c3 Mon Sep 17 00:00:00 2001 From: Adam Thomson Date: Wed, 16 Sep 2015 10:13:19 +0100 Subject: [PATCH 216/647] ASoC: fsl_ssi: Fix checking of dai format for AC97 mode Current code incorrectly treats dai format for AC97 as bit mask whereas it's actually an integer value. This causes DAI formats other than AC97 (e.g. DSP_B) to trigger AC97 related code, which is incorrect and breaks functionality. This patch fixes the code to correctly compare values to determine AC97 or not. Signed-off-by: Adam Thomson Acked-by: Timur Tabi Signed-off-by: Mark Brown --- sound/soc/fsl/fsl_ssi.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index 8ec6fb208ea0..37c5cd4d0e59 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -249,7 +249,8 @@ MODULE_DEVICE_TABLE(of, fsl_ssi_ids); static bool fsl_ssi_is_ac97(struct fsl_ssi_private *ssi_private) { - return !!(ssi_private->dai_fmt & SND_SOC_DAIFMT_AC97); + return (ssi_private->dai_fmt & SND_SOC_DAIFMT_FORMAT_MASK) == + SND_SOC_DAIFMT_AC97; } static bool fsl_ssi_is_i2s_master(struct fsl_ssi_private *ssi_private) @@ -947,7 +948,7 @@ static int _fsl_ssi_set_dai_fmt(struct device *dev, CCSR_SSI_SCR_TCH_EN); } - if (fmt & SND_SOC_DAIFMT_AC97) + if ((fmt & SND_SOC_DAIFMT_FORMAT_MASK) == SND_SOC_DAIFMT_AC97) fsl_ssi_setup_ac97(ssi_private); return 0; From b0379d7aa415249ce1dab8dd2554f8845d6822ab Mon Sep 17 00:00:00 2001 From: Dudley Du Date: Sat, 19 Sep 2015 10:39:32 -0700 Subject: [PATCH 217/647] Input: cyapa - fix address of Gen3 devices in device tree documentation All of the Gen3 touchpads are fixed with I2C address 0x67, so correct the reg value description from 0x24 to 0x67. Signed-off-by: Dudley Du Signed-off-by: Dmitry Torokhov --- Documentation/devicetree/bindings/input/cypress,cyapa.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/input/cypress,cyapa.txt b/Documentation/devicetree/bindings/input/cypress,cyapa.txt index 635a3b036630..8d91ba9ff2fd 100644 --- a/Documentation/devicetree/bindings/input/cypress,cyapa.txt +++ b/Documentation/devicetree/bindings/input/cypress,cyapa.txt @@ -25,7 +25,7 @@ Example: /* Cypress Gen3 touchpad */ touchpad@67 { compatible = "cypress,cyapa"; - reg = <0x24>; + reg = <0x67>; interrupt-parent = <&gpio>; interrupts = <2 IRQ_TYPE_EDGE_FALLING>; /* GPIO 2 */ wakeup-source; From b1452723cf23c908eed2bf6baf0c23943eb0e8bf Mon Sep 17 00:00:00 2001 From: Daniel Martin Date: Sat, 19 Sep 2015 11:27:19 -0700 Subject: [PATCH 218/647] Input: fix typo in MT documentation Section "Event Computation" had this: ... ABS_MT_TOOL_X := C_X ABS_MT_TOOL_X := C_Y Replace the second ABS_MT_TOOL_X with ABS_MT_TOOL_Y. Signed-off-by: Daniel Martin Acked-by: Benjamin Tissoires Signed-off-by: Dmitry Torokhov --- Documentation/input/multi-touch-protocol.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/input/multi-touch-protocol.txt b/Documentation/input/multi-touch-protocol.txt index b85d000faeb4..c51f1146f3bd 100644 --- a/Documentation/input/multi-touch-protocol.txt +++ b/Documentation/input/multi-touch-protocol.txt @@ -361,7 +361,7 @@ For win8 devices with both T and C coordinates, the position mapping is ABS_MT_POSITION_X := T_X ABS_MT_POSITION_Y := T_Y ABS_MT_TOOL_X := C_X - ABS_MT_TOOL_X := C_Y + ABS_MT_TOOL_Y := C_Y Unfortunately, there is not enough information to specify both the touching ellipse and the tool ellipse, so one has to resort to approximations. One From b9ab471b71900ca8a670ecf4f1cc65b626953655 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Mon, 14 Sep 2015 10:38:31 -0700 Subject: [PATCH 219/647] Input: pm8941-pwrkey - remove unneded semicolon It's not needed and is just creating a null statement so remove it. Signed-off-by: Javier Martinez Canillas Signed-off-by: Dmitry Torokhov --- drivers/input/misc/pm8941-pwrkey.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/misc/pm8941-pwrkey.c b/drivers/input/misc/pm8941-pwrkey.c index 867db8a91372..e317b75357a0 100644 --- a/drivers/input/misc/pm8941-pwrkey.c +++ b/drivers/input/misc/pm8941-pwrkey.c @@ -93,7 +93,7 @@ static int pm8941_reboot_notify(struct notifier_block *nb, default: reset_type = PON_PS_HOLD_TYPE_HARD_RESET; break; - }; + } error = regmap_update_bits(pwrkey->regmap, pwrkey->baseaddr + PON_PS_HOLD_RST_CTL, From 86a39bffc4e94f80527c14fe00a3acb432910ef3 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Mon, 14 Sep 2015 10:38:39 -0700 Subject: [PATCH 220/647] Input: mms114 - remove unneded semicolons They aren't needed and are just creating null statements so remove it. Signed-off-by: Javier Martinez Canillas Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/mms114.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/input/touchscreen/mms114.c b/drivers/input/touchscreen/mms114.c index 7cce87650fc8..1fafc9f57af6 100644 --- a/drivers/input/touchscreen/mms114.c +++ b/drivers/input/touchscreen/mms114.c @@ -394,12 +394,12 @@ static struct mms114_platform_data *mms114_parse_dt(struct device *dev) if (of_property_read_u32(np, "x-size", &pdata->x_size)) { dev_err(dev, "failed to get x-size property\n"); return NULL; - }; + } if (of_property_read_u32(np, "y-size", &pdata->y_size)) { dev_err(dev, "failed to get y-size property\n"); return NULL; - }; + } of_property_read_u32(np, "contact-threshold", &pdata->contact_threshold); From 46b018fa95003f8159f5fcf9b8cd89acaea34e31 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Sat, 5 Sep 2015 10:19:38 -0700 Subject: [PATCH 221/647] Input: walkera0701 - fix abs() calculations on 64 bit values abs() function can not be used with 64 bit values, so let's switch to abs64(). From include/linux/kernel.h: /* * abs() handles unsigned and signed longs, ints, shorts and chars. * For all input types abs() returns a signed long. * abs() should not be used for 64-bit types (s64, u64, long long) * - use abs64() for those. */ Reported-by: Joe Perches Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/walkera0701.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/input/joystick/walkera0701.c b/drivers/input/joystick/walkera0701.c index b76ac580703c..a8bc2fe170dd 100644 --- a/drivers/input/joystick/walkera0701.c +++ b/drivers/input/joystick/walkera0701.c @@ -150,7 +150,7 @@ static void walkera0701_irq_handler(void *handler_data) if (w->counter == 24) { /* full frame */ walkera0701_parse_frame(w); w->counter = NO_SYNC; - if (abs(pulse_time - SYNC_PULSE) < RESERVE) /* new frame sync */ + if (abs64(pulse_time - SYNC_PULSE) < RESERVE) /* new frame sync */ w->counter = 0; } else { if ((pulse_time > (ANALOG_MIN_PULSE - RESERVE) @@ -161,7 +161,7 @@ static void walkera0701_irq_handler(void *handler_data) } else w->counter = NO_SYNC; } - } else if (abs(pulse_time - SYNC_PULSE - BIN0_PULSE) < + } else if (abs64(pulse_time - SYNC_PULSE - BIN0_PULSE) < RESERVE + BIN1_PULSE - BIN0_PULSE) /* frame sync .. */ w->counter = 0; From 6cc527b05847984990a09ef028b2f670bbc72c46 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Mon, 14 Sep 2015 10:36:35 -0700 Subject: [PATCH 222/647] Input: imx6ul_tsc - propagate the errors imx6ul_adc_init() may fail in two cases, so we should better propagate the errors and make sure that the callers of this function also check and propagate the errors accordingly. Signed-off-by: Fabio Estevam Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/imx6ul_tsc.c | 28 +++++++++++++++++--------- 1 file changed, 19 insertions(+), 9 deletions(-) diff --git a/drivers/input/touchscreen/imx6ul_tsc.c b/drivers/input/touchscreen/imx6ul_tsc.c index ff0b75813daa..0a96b76f6021 100644 --- a/drivers/input/touchscreen/imx6ul_tsc.c +++ b/drivers/input/touchscreen/imx6ul_tsc.c @@ -94,7 +94,7 @@ struct imx6ul_tsc { * TSC module need ADC to get the measure value. So * before config TSC, we should initialize ADC module. */ -static void imx6ul_adc_init(struct imx6ul_tsc *tsc) +static int imx6ul_adc_init(struct imx6ul_tsc *tsc) { int adc_hc = 0; int adc_gc; @@ -122,17 +122,23 @@ static void imx6ul_adc_init(struct imx6ul_tsc *tsc) timeout = wait_for_completion_timeout (&tsc->completion, ADC_TIMEOUT); - if (timeout == 0) + if (timeout == 0) { dev_err(tsc->dev, "Timeout for adc calibration\n"); + return -ETIMEDOUT; + } adc_gs = readl(tsc->adc_regs + REG_ADC_GS); - if (adc_gs & ADC_CALF) + if (adc_gs & ADC_CALF) { dev_err(tsc->dev, "ADC calibration failed\n"); + return -EINVAL; + } /* TSC need the ADC work in hardware trigger */ adc_cfg = readl(tsc->adc_regs + REG_ADC_CFG); adc_cfg |= ADC_HARDWARE_TRIGGER; writel(adc_cfg, tsc->adc_regs + REG_ADC_CFG); + + return 0; } /* @@ -188,11 +194,17 @@ static void imx6ul_tsc_set(struct imx6ul_tsc *tsc) writel(start, tsc->tsc_regs + REG_TSC_FLOW_CONTROL); } -static void imx6ul_tsc_init(struct imx6ul_tsc *tsc) +static int imx6ul_tsc_init(struct imx6ul_tsc *tsc) { - imx6ul_adc_init(tsc); + int err; + + err = imx6ul_adc_init(tsc); + if (err) + return err; imx6ul_tsc_channel_config(tsc); imx6ul_tsc_set(tsc); + + return 0; } static void imx6ul_tsc_disable(struct imx6ul_tsc *tsc) @@ -311,9 +323,7 @@ static int imx6ul_tsc_open(struct input_dev *input_dev) return err; } - imx6ul_tsc_init(tsc); - - return 0; + return imx6ul_tsc_init(tsc); } static void imx6ul_tsc_close(struct input_dev *input_dev) @@ -491,7 +501,7 @@ static int __maybe_unused imx6ul_tsc_resume(struct device *dev) goto out; } - imx6ul_tsc_init(tsc); + retval = imx6ul_tsc_init(tsc); } out: From 3905de62b2624a4574776b3b7ddfa97758b75edc Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Mon, 14 Sep 2015 10:37:08 -0700 Subject: [PATCH 223/647] Input: imx6ul_tsc - check for negative return value We should check for negative values returned by platform_get_irq(). Signed-off-by: Fabio Estevam Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/imx6ul_tsc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/touchscreen/imx6ul_tsc.c b/drivers/input/touchscreen/imx6ul_tsc.c index 0a96b76f6021..4957d8b6ca59 100644 --- a/drivers/input/touchscreen/imx6ul_tsc.c +++ b/drivers/input/touchscreen/imx6ul_tsc.c @@ -416,7 +416,7 @@ static int imx6ul_tsc_probe(struct platform_device *pdev) } adc_irq = platform_get_irq(pdev, 1); - if (adc_irq <= 0) { + if (adc_irq < 0) { dev_err(&pdev->dev, "no adc irq resource?\n"); return adc_irq; } From 5eab3cf3e48cf658f3432e8ba31436d5a4f6a219 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Mon, 14 Sep 2015 10:37:31 -0700 Subject: [PATCH 224/647] Input: imx6ul_tsc - use the preferred method for kzalloc() According to Documentation/CodingStyle: "The preferred form for passing a size of a struct is the following: p = kmalloc(sizeof(*p), ...);" so do as suggested. Signed-off-by: Fabio Estevam Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/imx6ul_tsc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/touchscreen/imx6ul_tsc.c b/drivers/input/touchscreen/imx6ul_tsc.c index 4957d8b6ca59..67c73f2e7e3a 100644 --- a/drivers/input/touchscreen/imx6ul_tsc.c +++ b/drivers/input/touchscreen/imx6ul_tsc.c @@ -347,7 +347,7 @@ static int imx6ul_tsc_probe(struct platform_device *pdev) int tsc_irq; int adc_irq; - tsc = devm_kzalloc(&pdev->dev, sizeof(struct imx6ul_tsc), GFP_KERNEL); + tsc = devm_kzalloc(&pdev->dev, sizeof(*tsc), GFP_KERNEL); if (!tsc) return -ENOMEM; From 002801fc5372ecb725f0d7939f88ca752ce1d499 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Mon, 14 Sep 2015 10:37:55 -0700 Subject: [PATCH 225/647] Input: imx6ul_tsc - fix controller name We should better write "Touchscreen". Signed-off-by: Fabio Estevam Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/imx6ul_tsc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/touchscreen/imx6ul_tsc.c b/drivers/input/touchscreen/imx6ul_tsc.c index 67c73f2e7e3a..8275267eac25 100644 --- a/drivers/input/touchscreen/imx6ul_tsc.c +++ b/drivers/input/touchscreen/imx6ul_tsc.c @@ -355,7 +355,7 @@ static int imx6ul_tsc_probe(struct platform_device *pdev) if (!input_dev) return -ENOMEM; - input_dev->name = "iMX6UL TouchScreen Controller"; + input_dev->name = "iMX6UL Touchscreen Controller"; input_dev->id.bustype = BUS_HOST; input_dev->open = imx6ul_tsc_open; From 79234c3db6842a3de03817211d891e0c2878f756 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 18 Sep 2015 15:53:24 -0400 Subject: [PATCH 226/647] SUNRPC: Lock the transport layer on shutdown Avoid all races with the connect/disconnect handlers by taking the transport lock. Reported-by:"Suzuki K. Poulose" Acked-by: Jeff Layton Signed-off-by: Trond Myklebust --- net/sunrpc/xprt.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index ab5dd621ae0c..2e98f4a243e5 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -614,6 +614,7 @@ static void xprt_autoclose(struct work_struct *work) clear_bit(XPRT_CLOSE_WAIT, &xprt->state); xprt->ops->close(xprt); xprt_release_write(xprt, NULL); + wake_up_bit(&xprt->state, XPRT_LOCKED); } /** @@ -723,6 +724,7 @@ void xprt_unlock_connect(struct rpc_xprt *xprt, void *cookie) xprt->ops->release_xprt(xprt, NULL); out: spin_unlock_bh(&xprt->transport_lock); + wake_up_bit(&xprt->state, XPRT_LOCKED); } /** @@ -1394,6 +1396,10 @@ out: static void xprt_destroy(struct rpc_xprt *xprt) { dprintk("RPC: destroying transport %p\n", xprt); + + /* Exclude transport connect/disconnect handlers */ + wait_on_bit_lock(&xprt->state, XPRT_LOCKED, TASK_UNINTERRUPTIBLE); + del_timer_sync(&xprt->timer); rpc_xprt_debugfs_unregister(xprt); From 4b0ab51db32eba0f48b7618254742f143364a28d Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 18 Sep 2015 09:52:07 -0400 Subject: [PATCH 227/647] SUNRPC: xs_sock_mark_closed() does not need to trigger socket autoclose Under all conditions, it should be quite sufficient just to mark the socket as disconnected. It will then be closed by the transport shutdown or reconnect code. Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 1 - 1 file changed, 1 deletion(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index c35038511686..1a85e0ed0b48 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -777,7 +777,6 @@ static void xs_sock_mark_closed(struct rpc_xprt *xprt) xs_sock_reset_connection_flags(xprt); /* Mark transport as closed and wake up all pending tasks */ xprt_disconnect_done(xprt); - xprt_force_disconnect(xprt); } /** From 97584d1838b7e2545c3b10aacef3327fcaa9531b Mon Sep 17 00:00:00 2001 From: Javi Merino Date: Mon, 14 Sep 2015 14:23:54 +0100 Subject: [PATCH 228/647] thermal: power_allocator: exit early if there are no cooling devices Don't waste cycles in the power allocator governor's throttle function if there are no cooling devices and exit early. This commit doesn't change any functionality, but should provide better performance for the odd case of a thermal zone with trip points but without cooling devices. Cc: Zhang Rui Cc: Eduardo Valentin Reviewed-by: Daniel Kurtz Signed-off-by: Javi Merino Signed-off-by: Eduardo Valentin --- drivers/thermal/power_allocator.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/thermal/power_allocator.c b/drivers/thermal/power_allocator.c index 718363f5be40..7ff96270c933 100644 --- a/drivers/thermal/power_allocator.c +++ b/drivers/thermal/power_allocator.c @@ -346,6 +346,11 @@ static int allocate_power(struct thermal_zone_device *tz, } } + if (!num_actors) { + ret = -ENODEV; + goto unlock; + } + /* * We need to allocate five arrays of the same size: * req_power, max_power, granted_power, extra_actor_power and From 6f29b9bba7b08c6b1d6f2cc4cf750b342fc1946c Mon Sep 17 00:00:00 2001 From: Kinglong Mee Date: Sun, 20 Sep 2015 23:03:28 +0800 Subject: [PATCH 229/647] NFS: Do cleanup before resetting pageio read/write to mds There is a reference leak of layout segment after resetting pageio read/write to mds. Signed-off-by: Kinglong Mee Cc: stable@vger.kernel.org # v4.0+ Signed-off-by: Trond Myklebust --- fs/nfs/read.c | 3 +++ fs/nfs/write.c | 3 +++ 2 files changed, 6 insertions(+) diff --git a/fs/nfs/read.c b/fs/nfs/read.c index ae0ff7a11b40..01b8cc8e8cfc 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -72,6 +72,9 @@ void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio) { struct nfs_pgio_mirror *mirror; + if (pgio->pg_ops && pgio->pg_ops->pg_cleanup) + pgio->pg_ops->pg_cleanup(pgio); + pgio->pg_ops = &nfs_pgio_rw_ops; /* read path should never have more than one mirror */ diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 388f48079c43..72624dc4a623 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1351,6 +1351,9 @@ void nfs_pageio_reset_write_mds(struct nfs_pageio_descriptor *pgio) { struct nfs_pgio_mirror *mirror; + if (pgio->pg_ops && pgio->pg_ops->pg_cleanup) + pgio->pg_ops->pg_cleanup(pgio); + pgio->pg_ops = &nfs_pgio_rw_ops; nfs_pageio_stop_mirroring(pgio); From 8714d46dc5b95a6a898ec8f1e67237c7995adfc6 Mon Sep 17 00:00:00 2001 From: Kinglong Mee Date: Sun, 20 Sep 2015 23:04:22 +0800 Subject: [PATCH 230/647] NFS: Fix an infinite loop when layoutget fail with BAD_STATEID If layouget fail with BAD_STATEID, restart should not using the old stateid. But, nfs client choose the layout stateid at first, and then the open stateid. To avoid the infinite loop of using bad stateid for layoutget, this patch sets the layout flag'ss NFS_LAYOUT_INVALID_STID bit to skip choosing the bad layout stateid. Signed-off-by: Kinglong Mee Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 693b903b48bd..10ba6c1968e2 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -7820,6 +7820,7 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata) * Mark the bad layout state as invalid, then retry * with the current stateid. */ + set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags); pnfs_mark_matching_lsegs_invalid(lo, &head, NULL); spin_unlock(&inode->i_lock); pnfs_free_lseg_list(&head); From a42fb351ca1f340f8307468be765e3f77ddedda9 Mon Sep 17 00:00:00 2001 From: Knuth Posern Date: Sun, 20 Sep 2015 21:25:22 +0200 Subject: [PATCH 231/647] thunderbolt: Allow loading of module on recent Apple MacBooks with thunderbolt 2 controller The pci device ids listed in the thunderbolt driver are to restrictive, which prevents the driver from being loaded on recent Apple MacBooks using a thunderbolt 2 controller. In particular this prevented any hot-plugging functionality for thunderbolt based ethernet dongles (i.e. Apples thunderbolt gigabit ethernet broadcom tg3 based dongle Model A1433 EMC 2590). Changing the subvendor and subdevice to PCI_ANY_ID the thunderbolt driver loads and binds to the pci device 07:00.0 System peripheral: Intel Corporation Device 156c which is the thunderbolt 2 controller on the MacBookPro12,1. Successfully tested on MacBookPro12,1. With the patch the thunderbolt module gets now loaded on boot. And it provides hot-plugging support both for a cold-plugged and a warm-plugged ethernet dongle. Signed-off-by: Andreas Noever Acked-by: Knuth Posern Signed-off-by: Greg Kroah-Hartman --- drivers/thunderbolt/nhi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/thunderbolt/nhi.c b/drivers/thunderbolt/nhi.c index c68fe1222c16..20a41f7de76f 100644 --- a/drivers/thunderbolt/nhi.c +++ b/drivers/thunderbolt/nhi.c @@ -643,7 +643,7 @@ static struct pci_device_id nhi_ids[] = { { .class = PCI_CLASS_SYSTEM_OTHER << 8, .class_mask = ~0, .vendor = PCI_VENDOR_ID_INTEL, .device = 0x156c, - .subvendor = 0x2222, .subdevice = 0x1111, + .subvendor = PCI_ANY_ID, .subdevice = PCI_ANY_ID, }, { 0,} }; From 5fc3e64f941197d5b77abccbe0144b25c2c36f2f Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Fri, 18 Sep 2015 13:13:44 +1000 Subject: [PATCH 232/647] KVM: PPC: Book3S HV: Fix handling of interrupted VCPUs This fixes a bug which results in stale vcore pointers being left in the per-cpu preempted vcore lists when a VM is destroyed. The result of the stale vcore pointers is usually either a crash or a lockup inside collect_piggybacks() when another VM is run. A typical lockup message looks like: [ 472.161074] NMI watchdog: BUG: soft lockup - CPU#24 stuck for 22s! [qemu-system-ppc:7039] [ 472.161204] Modules linked in: kvm_hv kvm_pr kvm xt_CHECKSUM ipt_MASQUERADE nf_nat_masquerade_ipv4 tun ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 xt_conntrack ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_security ip6table_raw ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw ses enclosure shpchp rtc_opal i2c_opal powernv_rng binfmt_misc dm_service_time scsi_dh_alua radeon i2c_algo_bit drm_kms_helper ttm drm tg3 ptp pps_core cxgb3 ipr i2c_core mdio dm_multipath [last unloaded: kvm_hv] [ 472.162111] CPU: 24 PID: 7039 Comm: qemu-system-ppc Not tainted 4.2.0-kvm+ #49 [ 472.162187] task: c000001e38512750 ti: c000001e41bfc000 task.ti: c000001e41bfc000 [ 472.162262] NIP: c00000000096b094 LR: c00000000096b08c CTR: c000000000111130 [ 472.162337] REGS: c000001e41bff520 TRAP: 0901 Not tainted (4.2.0-kvm+) [ 472.162399] MSR: 9000000100009033 CR: 24848844 XER: 00000000 [ 472.162588] CFAR: c00000000096b0ac SOFTE: 1 GPR00: c000000000111170 c000001e41bff7a0 c00000000127df00 0000000000000001 GPR04: 0000000000000003 0000000000000001 0000000000000000 0000000000874821 GPR08: c000001e41bff8e0 0000000000000001 0000000000000000 d00000000efde740 GPR12: c000000000111130 c00000000fdae400 [ 472.163053] NIP [c00000000096b094] _raw_spin_lock_irqsave+0xa4/0x130 [ 472.163117] LR [c00000000096b08c] _raw_spin_lock_irqsave+0x9c/0x130 [ 472.163179] Call Trace: [ 472.163206] [c000001e41bff7a0] [c000001e41bff7f0] 0xc000001e41bff7f0 (unreliable) [ 472.163295] [c000001e41bff7e0] [c000000000111170] __wake_up+0x40/0x90 [ 472.163375] [c000001e41bff830] [d00000000efd6fc0] kvmppc_run_core+0x1240/0x1950 [kvm_hv] [ 472.163465] [c000001e41bffa30] [d00000000efd8510] kvmppc_vcpu_run_hv+0x5a0/0xd90 [kvm_hv] [ 472.163559] [c000001e41bffb70] [d00000000e9318a4] kvmppc_vcpu_run+0x44/0x60 [kvm] [ 472.163653] [c000001e41bffba0] [d00000000e92e674] kvm_arch_vcpu_ioctl_run+0x64/0x170 [kvm] [ 472.163745] [c000001e41bffbe0] [d00000000e9263a8] kvm_vcpu_ioctl+0x538/0x7b0 [kvm] [ 472.163834] [c000001e41bffd40] [c0000000002d0f50] do_vfs_ioctl+0x480/0x7c0 [ 472.163910] [c000001e41bffde0] [c0000000002d1364] SyS_ioctl+0xd4/0xf0 [ 472.163986] [c000001e41bffe30] [c000000000009260] system_call+0x38/0xd0 [ 472.164060] Instruction dump: [ 472.164098] ebc1fff0 ebe1fff8 7c0803a6 4e800020 60000000 60000000 60420000 8bad02e2 [ 472.164224] 7fc3f378 4b6a57c1 60000000 7c210b78 89290009 792affe3 40820070 The bug is that kvmppc_run_vcpu does not correctly handle the case where a vcpu task receives a signal while its guest vcpu is executing in the guest as a result of being piggy-backed onto the execution of another vcore. In that case we need to wait for the vcpu to finish executing inside the guest, and then remove this vcore from the preempted vcores list. That way, we avoid leaving this vcpu's vcore on the preempted vcores list when the vcpu gets interrupted. Fixes: ec2571650826 Reported-by: Thomas Huth Tested-by: Thomas Huth Signed-off-by: Paul Mackerras --- arch/powerpc/kvm/book3s_hv.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c index 9754e6815e52..228049786888 100644 --- a/arch/powerpc/kvm/book3s_hv.c +++ b/arch/powerpc/kvm/book3s_hv.c @@ -2692,9 +2692,13 @@ static int kvmppc_run_vcpu(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) while (vcpu->arch.state == KVMPPC_VCPU_RUNNABLE && (vc->vcore_state == VCORE_RUNNING || - vc->vcore_state == VCORE_EXITING)) + vc->vcore_state == VCORE_EXITING || + vc->vcore_state == VCORE_PIGGYBACK)) kvmppc_wait_for_exec(vc, vcpu, TASK_UNINTERRUPTIBLE); + if (vc->vcore_state == VCORE_PREEMPT && vc->runner == NULL) + kvmppc_vcore_end_preempt(vc); + if (vcpu->arch.state == KVMPPC_VCPU_RUNNABLE) { kvmppc_remove_runnable(vc, vcpu); vcpu->stat.signal_exits++; From 7e022e717f54897e396504306d0c9b61452adf4e Mon Sep 17 00:00:00 2001 From: "Gautham R. Shenoy" Date: Thu, 21 May 2015 13:57:04 +0530 Subject: [PATCH 233/647] KVM: PPC: Book3S HV: Pass the correct trap argument to kvmhv_commence_exit In guest_exit_cont we call kvmhv_commence_exit which expects the trap number as the argument. However r3 doesn't contain the trap number at this point and as a result we would be calling the function with a spurious trap number. Fix this by copying r12 into r3 before calling kvmhv_commence_exit as r12 contains the trap number. Cc: stable@vger.kernel.org # v4.1+ Fixes: eddb60fb1443 Signed-off-by: Gautham R. Shenoy Signed-off-by: Paul Mackerras --- arch/powerpc/kvm/book3s_hv_rmhandlers.S | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S index 2273dcacef39..b98889e9851d 100644 --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S @@ -1257,6 +1257,7 @@ mc_cont: bl kvmhv_accumulate_time #endif + mr r3, r12 /* Increment exit count, poke other threads to exit */ bl kvmhv_commence_exit nop From 3eb4ee68254235e4f47bc0410538fcdaede39589 Mon Sep 17 00:00:00 2001 From: Thomas Huth Date: Fri, 18 Sep 2015 08:57:28 +0200 Subject: [PATCH 234/647] KVM: PPC: Book3S: Take the kvm->srcu lock in kvmppc_h_logical_ci_load/store() Access to the kvm->buses (like with the kvm_io_bus_read() and -write() functions) has to be protected via the kvm->srcu lock. The kvmppc_h_logical_ci_load() and -store() functions are missing this lock so far, so let's add it there, too. This fixes the problem that the kernel reports "suspicious RCU usage" when lock debugging is enabled. Cc: stable@vger.kernel.org # v4.1+ Fixes: 99342cf8044420eebdf9297ca03a14cb6a7085a1 Signed-off-by: Thomas Huth Signed-off-by: Paul Mackerras --- arch/powerpc/kvm/book3s.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c index cf009167d208..099c79d8c160 100644 --- a/arch/powerpc/kvm/book3s.c +++ b/arch/powerpc/kvm/book3s.c @@ -829,12 +829,15 @@ int kvmppc_h_logical_ci_load(struct kvm_vcpu *vcpu) unsigned long size = kvmppc_get_gpr(vcpu, 4); unsigned long addr = kvmppc_get_gpr(vcpu, 5); u64 buf; + int srcu_idx; int ret; if (!is_power_of_2(size) || (size > sizeof(buf))) return H_TOO_HARD; + srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); ret = kvm_io_bus_read(vcpu, KVM_MMIO_BUS, addr, size, &buf); + srcu_read_unlock(&vcpu->kvm->srcu, srcu_idx); if (ret != 0) return H_TOO_HARD; @@ -869,6 +872,7 @@ int kvmppc_h_logical_ci_store(struct kvm_vcpu *vcpu) unsigned long addr = kvmppc_get_gpr(vcpu, 5); unsigned long val = kvmppc_get_gpr(vcpu, 6); u64 buf; + int srcu_idx; int ret; switch (size) { @@ -892,7 +896,9 @@ int kvmppc_h_logical_ci_store(struct kvm_vcpu *vcpu) return H_TOO_HARD; } + srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); ret = kvm_io_bus_write(vcpu, KVM_MMIO_BUS, addr, size, &buf); + srcu_read_unlock(&vcpu->kvm->srcu, srcu_idx); if (ret != 0) return H_TOO_HARD; From 3245acbce952cda75b2fafa8e40452ef0c4756c1 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Thu, 17 Sep 2015 18:09:11 +0200 Subject: [PATCH 235/647] hwmon: (abx500) Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: Guenter Roeck --- drivers/hwmon/abx500.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hwmon/abx500.c b/drivers/hwmon/abx500.c index 6cb89c0ebab6..1fd46859ed29 100644 --- a/drivers/hwmon/abx500.c +++ b/drivers/hwmon/abx500.c @@ -470,6 +470,7 @@ static const struct of_device_id abx500_temp_match[] = { { .compatible = "stericsson,abx500-temp" }, {}, }; +MODULE_DEVICE_TABLE(of, abx500_temp_match); #endif static struct platform_driver abx500_temp_driver = { From fe5152882a45a2545d44d39fe29badc8e4f66ad3 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Thu, 17 Sep 2015 18:09:28 +0200 Subject: [PATCH 236/647] hwmon: (gpio-fan) Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: Guenter Roeck --- drivers/hwmon/gpio-fan.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hwmon/gpio-fan.c b/drivers/hwmon/gpio-fan.c index a3dae6d0082a..82de3deeb18a 100644 --- a/drivers/hwmon/gpio-fan.c +++ b/drivers/hwmon/gpio-fan.c @@ -539,6 +539,7 @@ static const struct of_device_id of_gpio_fan_match[] = { { .compatible = "gpio-fan", }, {}, }; +MODULE_DEVICE_TABLE(of, of_gpio_fan_match); #endif /* CONFIG_OF_GPIO */ static int gpio_fan_probe(struct platform_device *pdev) From f491e70ccffa5d19aa51c958909320fa1f3905ed Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Thu, 17 Sep 2015 18:09:55 +0200 Subject: [PATCH 237/647] hwmon: (pwm-fan) Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: Guenter Roeck --- drivers/hwmon/pwm-fan.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hwmon/pwm-fan.c b/drivers/hwmon/pwm-fan.c index 2d9a712699ff..3e23003f78b0 100644 --- a/drivers/hwmon/pwm-fan.c +++ b/drivers/hwmon/pwm-fan.c @@ -323,6 +323,7 @@ static const struct of_device_id of_pwm_fan_match[] = { { .compatible = "pwm-fan", }, {}, }; +MODULE_DEVICE_TABLE(of, of_pwm_fan_match); static struct platform_driver pwm_fan_driver = { .probe = pwm_fan_probe, From 24311f884189d42d40354a6f38ca218eb9aeb811 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 20 Sep 2015 10:50:17 -0400 Subject: [PATCH 238/647] NFSv4: Recovery of recalled read delegations is broken When a read delegation is being recalled, and we're reclaiming the cached opens, we need to make sure that we only reclaim read-only modes. A previous attempt to do this, relied on retrieving the delegation type from the nfs4_opendata structure. Unfortunately, as Kinglong pointed out, this field can only be set when performing reboot recovery. Furthermore, if we call nfs4_open_recover(), then we end up clobbering the state->flags for all modes that we're not recovering... The fix is to have the delegation recall code pass this information to the recovery call, and then refactor the recovery code so that nfs4_open_delegation_recall() does not need to call nfs4_open_recover(). Reported-by: Kinglong Mee Fixes: 39f897fdbd46 ("NFSv4: When returning a delegation, don't...") Tested-by: Kinglong Mee Cc: NeilBrown Cc: stable@vger.kernel.org # v4.2+ Signed-off-by: Trond Myklebust --- fs/nfs/delegation.c | 8 +++-- fs/nfs/delegation.h | 2 +- fs/nfs/nfs4proc.c | 81 ++++++++++++++++++++++++++------------------- 3 files changed, 53 insertions(+), 38 deletions(-) diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 2714ef835bdd..be806ead7f4d 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -113,7 +113,8 @@ out: return status; } -static int nfs_delegation_claim_opens(struct inode *inode, const nfs4_stateid *stateid) +static int nfs_delegation_claim_opens(struct inode *inode, + const nfs4_stateid *stateid, fmode_t type) { struct nfs_inode *nfsi = NFS_I(inode); struct nfs_open_context *ctx; @@ -140,7 +141,7 @@ again: /* Block nfs4_proc_unlck */ mutex_lock(&sp->so_delegreturn_mutex); seq = raw_seqcount_begin(&sp->so_reclaim_seqcount); - err = nfs4_open_delegation_recall(ctx, state, stateid); + err = nfs4_open_delegation_recall(ctx, state, stateid, type); if (!err) err = nfs_delegation_claim_locks(ctx, state, stateid); if (!err && read_seqcount_retry(&sp->so_reclaim_seqcount, seq)) @@ -411,7 +412,8 @@ static int nfs_end_delegation_return(struct inode *inode, struct nfs_delegation do { if (test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) break; - err = nfs_delegation_claim_opens(inode, &delegation->stateid); + err = nfs_delegation_claim_opens(inode, &delegation->stateid, + delegation->type); if (!issync || err != -EAGAIN) break; /* diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h index a44829173e57..333063e032f0 100644 --- a/fs/nfs/delegation.h +++ b/fs/nfs/delegation.h @@ -54,7 +54,7 @@ void nfs_delegation_reap_unclaimed(struct nfs_client *clp); /* NFSv4 delegation-related procedures */ int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid, int issync); -int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid); +int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid, fmode_t type); int nfs4_lock_delegation_recall(struct file_lock *fl, struct nfs4_state *state, const nfs4_stateid *stateid); bool nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode, fmode_t flags); diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 10ba6c1968e2..e129347b656f 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1127,6 +1127,21 @@ static int nfs4_wait_for_completion_rpc_task(struct rpc_task *task) return ret; } +static bool nfs4_mode_match_open_stateid(struct nfs4_state *state, + fmode_t fmode) +{ + switch(fmode & (FMODE_READ|FMODE_WRITE)) { + case FMODE_READ|FMODE_WRITE: + return state->n_rdwr != 0; + case FMODE_WRITE: + return state->n_wronly != 0; + case FMODE_READ: + return state->n_rdonly != 0; + } + WARN_ON_ONCE(1); + return false; +} + static int can_open_cached(struct nfs4_state *state, fmode_t mode, int open_mode) { int ret = 0; @@ -1571,17 +1586,13 @@ static struct nfs4_opendata *nfs4_open_recoverdata_alloc(struct nfs_open_context return opendata; } -static int nfs4_open_recover_helper(struct nfs4_opendata *opendata, fmode_t fmode, struct nfs4_state **res) +static int nfs4_open_recover_helper(struct nfs4_opendata *opendata, + fmode_t fmode) { struct nfs4_state *newstate; int ret; - if ((opendata->o_arg.claim == NFS4_OPEN_CLAIM_DELEGATE_CUR || - opendata->o_arg.claim == NFS4_OPEN_CLAIM_DELEG_CUR_FH) && - (opendata->o_arg.u.delegation_type & fmode) != fmode) - /* This mode can't have been delegated, so we must have - * a valid open_stateid to cover it - not need to reclaim. - */ + if (!nfs4_mode_match_open_stateid(opendata->state, fmode)) return 0; opendata->o_arg.open_flags = 0; opendata->o_arg.fmode = fmode; @@ -1597,14 +1608,14 @@ static int nfs4_open_recover_helper(struct nfs4_opendata *opendata, fmode_t fmod newstate = nfs4_opendata_to_nfs4_state(opendata); if (IS_ERR(newstate)) return PTR_ERR(newstate); + if (newstate != opendata->state) + ret = -ESTALE; nfs4_close_state(newstate, fmode); - *res = newstate; - return 0; + return ret; } static int nfs4_open_recover(struct nfs4_opendata *opendata, struct nfs4_state *state) { - struct nfs4_state *newstate; int ret; /* Don't trigger recovery in nfs_test_and_clear_all_open_stateid */ @@ -1615,27 +1626,15 @@ static int nfs4_open_recover(struct nfs4_opendata *opendata, struct nfs4_state * clear_bit(NFS_DELEGATED_STATE, &state->flags); clear_bit(NFS_OPEN_STATE, &state->flags); smp_rmb(); - if (state->n_rdwr != 0) { - ret = nfs4_open_recover_helper(opendata, FMODE_READ|FMODE_WRITE, &newstate); - if (ret != 0) - return ret; - if (newstate != state) - return -ESTALE; - } - if (state->n_wronly != 0) { - ret = nfs4_open_recover_helper(opendata, FMODE_WRITE, &newstate); - if (ret != 0) - return ret; - if (newstate != state) - return -ESTALE; - } - if (state->n_rdonly != 0) { - ret = nfs4_open_recover_helper(opendata, FMODE_READ, &newstate); - if (ret != 0) - return ret; - if (newstate != state) - return -ESTALE; - } + ret = nfs4_open_recover_helper(opendata, FMODE_READ|FMODE_WRITE); + if (ret != 0) + return ret; + ret = nfs4_open_recover_helper(opendata, FMODE_WRITE); + if (ret != 0) + return ret; + ret = nfs4_open_recover_helper(opendata, FMODE_READ); + if (ret != 0) + return ret; /* * We may have performed cached opens for all three recoveries. * Check if we need to update the current stateid. @@ -1759,18 +1758,32 @@ static int nfs4_handle_delegation_recall_error(struct nfs_server *server, struct return err; } -int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid) +int nfs4_open_delegation_recall(struct nfs_open_context *ctx, + struct nfs4_state *state, const nfs4_stateid *stateid, + fmode_t type) { struct nfs_server *server = NFS_SERVER(state->inode); struct nfs4_opendata *opendata; - int err; + int err = 0; opendata = nfs4_open_recoverdata_alloc(ctx, state, NFS4_OPEN_CLAIM_DELEG_CUR_FH); if (IS_ERR(opendata)) return PTR_ERR(opendata); nfs4_stateid_copy(&opendata->o_arg.u.delegation, stateid); - err = nfs4_open_recover(opendata, state); + clear_bit(NFS_DELEGATED_STATE, &state->flags); + switch (type & (FMODE_READ|FMODE_WRITE)) { + case FMODE_READ|FMODE_WRITE: + case FMODE_WRITE: + err = nfs4_open_recover_helper(opendata, FMODE_READ|FMODE_WRITE); + if (err) + break; + err = nfs4_open_recover_helper(opendata, FMODE_WRITE); + if (err) + break; + case FMODE_READ: + err = nfs4_open_recover_helper(opendata, FMODE_READ); + } nfs4_opendata_put(opendata); return nfs4_handle_delegation_recall_error(server, state, stateid, err); } From 2259f960b3a9b1020dccbf948c97311ce586db1b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 20 Sep 2015 13:30:30 -0400 Subject: [PATCH 239/647] NFSv4.x/pnfs: Don't try to recover stateids twice in layoutget If the current open or layout stateid doesn't match the stateid used in the layoutget RPC call, then don't try to recover it. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 28 +++++++++++++++++----------- 1 file changed, 17 insertions(+), 11 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index e129347b656f..ef6b46e08ac6 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -7813,20 +7813,23 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata) dprintk("%s: NFS4ERR_RECALLCONFLICT waiting %lu\n", __func__, delay); rpc_delay(task, delay); - task->tk_status = 0; - rpc_restart_call_prepare(task); - goto out; /* Do not call nfs4_async_handle_error() */ + /* Do not call nfs4_async_handle_error() */ + goto out_restart; } break; case -NFS4ERR_EXPIRED: case -NFS4ERR_BAD_STATEID: spin_lock(&inode->i_lock); - lo = NFS_I(inode)->layout; - if (!lo || list_empty(&lo->plh_segs)) { + if (nfs4_stateid_match(&lgp->args.stateid, + &lgp->args.ctx->state->stateid)) { spin_unlock(&inode->i_lock); /* If the open stateid was bad, then recover it. */ state = lgp->args.ctx->state; - } else { + break; + } + lo = NFS_I(inode)->layout; + if (lo && nfs4_stateid_match(&lgp->args.stateid, + &lo->plh_stateid)) { LIST_HEAD(head); /* @@ -7837,16 +7840,19 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata) pnfs_mark_matching_lsegs_invalid(lo, &head, NULL); spin_unlock(&inode->i_lock); pnfs_free_lseg_list(&head); - - task->tk_status = 0; - rpc_restart_call_prepare(task); - } + } else + spin_unlock(&inode->i_lock); + goto out_restart; } if (nfs4_async_handle_error(task, server, state, NULL) == -EAGAIN) - rpc_restart_call_prepare(task); + goto out_restart; out: dprintk("<-- %s\n", __func__); return; +out_restart: + task->tk_status = 0; + rpc_restart_call_prepare(task); + return; out_overflow: task->tk_status = -EOVERFLOW; goto out; From 66e8c57da6bf6b847a48a5a6fda59512f733ed78 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 25 Aug 2015 20:45:18 +0200 Subject: [PATCH 240/647] rcu: Change _wait_rcu_gp() to work around GCC bug 67055 Code like this in inline functions confuses some recent versions of gcc: const int n = const-expr; whatever_t array[n]; For more details, see: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67055#c13 This compiler bug results in the following failure after 114b7fd4b (rcu: Create rcu_sync infrastructure): In file included from include/linux/rcupdate.h:429:0, from include/linux/rcu_sync.h:5, from kernel/rcu/sync.c:1: include/linux/rcutiny.h: In function 'rcu_barrier_sched': include/linux/rcutiny.h:55:20: internal compiler error: Segmentation fault static inline void rcu_barrier_sched(void) This commit therefore eliminates the constant local variable in favor of direct use of the expression. Reported-and-tested-by: Mark Salter Reported-by: Guenter Roeck Signed-off-by: Oleg Nesterov Signed-off-by: Paul E. McKenney --- include/linux/rcupdate.h | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index ff476515f716..581abf848566 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -230,12 +230,11 @@ void __wait_rcu_gp(bool checktiny, int n, call_rcu_func_t *crcu_array, struct rcu_synchronize *rs_array); #define _wait_rcu_gp(checktiny, ...) \ -do { \ - call_rcu_func_t __crcu_array[] = { __VA_ARGS__ }; \ - const int __n = ARRAY_SIZE(__crcu_array); \ - struct rcu_synchronize __rs_array[__n]; \ - \ - __wait_rcu_gp(checktiny, __n, __crcu_array, __rs_array); \ +do { \ + call_rcu_func_t __crcu_array[] = { __VA_ARGS__ }; \ + struct rcu_synchronize __rs_array[ARRAY_SIZE(__crcu_array)]; \ + __wait_rcu_gp(checktiny, ARRAY_SIZE(__crcu_array), \ + __crcu_array, __rs_array); \ } while (0) #define wait_rcu_gp(...) _wait_rcu_gp(false, __VA_ARGS__) From 19a5ecde086a6a5287978b12ae948fa691b197b7 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 20 Sep 2015 21:01:22 -0700 Subject: [PATCH 241/647] rcu: Suppress lockdep false positive for rcp->exp_funnel_mutex In kernels built with CONFIG_PREEMPT=y, synchronize_rcu_expedited() invokes synchronize_sched_expedited() while holding RCU-preempt's root rcu_node structure's ->exp_funnel_mutex, which is acquired after the rcu_data structure's ->exp_funnel_mutex. The first thing that synchronize_sched_expedited() will do is acquire RCU-sched's rcu_data structure's ->exp_funnel_mutex. There is no danger of an actual deadlock because the locking order is always from RCU-preempt's expedited mutexes to those of RCU-sched. Unfortunately, lockdep considers both rcu_data structures' ->exp_funnel_mutex to be in the same lock class and therefore reports a deadlock cycle. This commit silences this false positive by placing RCU-sched's rcu_data structures' ->exp_funnel_mutex locks into their own lock class. Reported-by: Sasha Levin Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 9f75f25cc5d9..775d36cc0050 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -3868,6 +3868,7 @@ static void rcu_init_new_rnp(struct rcu_node *rnp_leaf) static void __init rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) { + static struct lock_class_key rcu_exp_sched_rdp_class; unsigned long flags; struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp = rcu_get_root(rsp); @@ -3883,6 +3884,10 @@ rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) mutex_init(&rdp->exp_funnel_mutex); rcu_boot_init_nocb_percpu_data(rdp); raw_spin_unlock_irqrestore(&rnp->lock, flags); + if (rsp == &rcu_sched_state) + lockdep_set_class_and_name(&rdp->exp_funnel_mutex, + &rcu_exp_sched_rdp_class, + "rcu_data_exp_sched"); } /* From ba5ca7848be05db6235aeb703586b821aa00e381 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Wed, 16 Sep 2015 15:27:43 +0200 Subject: [PATCH 242/647] bna: check for dma mapping errors Check for DMA mapping errors, recover from them and register them in ethtool stats like other errors. Cc: Rasesh Mody Signed-off-by: Ivan Vecera Acked-by: Rasesh Mody Signed-off-by: David S. Miller --- drivers/net/ethernet/brocade/bna/bna_tx_rx.c | 2 ++ drivers/net/ethernet/brocade/bna/bna_types.h | 1 + drivers/net/ethernet/brocade/bna/bnad.c | 29 ++++++++++++++++++- drivers/net/ethernet/brocade/bna/bnad.h | 2 ++ .../net/ethernet/brocade/bna/bnad_ethtool.c | 4 +++ 5 files changed, 37 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/brocade/bna/bna_tx_rx.c b/drivers/net/ethernet/brocade/bna/bna_tx_rx.c index 5d0753cc7e73..04b0d16b210e 100644 --- a/drivers/net/ethernet/brocade/bna/bna_tx_rx.c +++ b/drivers/net/ethernet/brocade/bna/bna_tx_rx.c @@ -2400,6 +2400,7 @@ bna_rx_create(struct bna *bna, struct bnad *bnad, q0->rcb->id = 0; q0->rx_packets = q0->rx_bytes = 0; q0->rx_packets_with_error = q0->rxbuf_alloc_failed = 0; + q0->rxbuf_map_failed = 0; bna_rxq_qpt_setup(q0, rxp, dpage_count, PAGE_SIZE, &dqpt_mem[i], &dsqpt_mem[i], &dpage_mem[i]); @@ -2428,6 +2429,7 @@ bna_rx_create(struct bna *bna, struct bnad *bnad, : rx_cfg->q1_buf_size; q1->rx_packets = q1->rx_bytes = 0; q1->rx_packets_with_error = q1->rxbuf_alloc_failed = 0; + q1->rxbuf_map_failed = 0; bna_rxq_qpt_setup(q1, rxp, hpage_count, PAGE_SIZE, &hqpt_mem[i], &hsqpt_mem[i], diff --git a/drivers/net/ethernet/brocade/bna/bna_types.h b/drivers/net/ethernet/brocade/bna/bna_types.h index e0e797f2ea14..c438d032e8bf 100644 --- a/drivers/net/ethernet/brocade/bna/bna_types.h +++ b/drivers/net/ethernet/brocade/bna/bna_types.h @@ -587,6 +587,7 @@ struct bna_rxq { u64 rx_bytes; u64 rx_packets_with_error; u64 rxbuf_alloc_failed; + u64 rxbuf_map_failed; }; /* RxQ pair */ diff --git a/drivers/net/ethernet/brocade/bna/bnad.c b/drivers/net/ethernet/brocade/bna/bnad.c index 506047c38607..21a0cfc3e7ec 100644 --- a/drivers/net/ethernet/brocade/bna/bnad.c +++ b/drivers/net/ethernet/brocade/bna/bnad.c @@ -399,7 +399,13 @@ bnad_rxq_refill_page(struct bnad *bnad, struct bna_rcb *rcb, u32 nalloc) } dma_addr = dma_map_page(&bnad->pcidev->dev, page, page_offset, - unmap_q->map_size, DMA_FROM_DEVICE); + unmap_q->map_size, DMA_FROM_DEVICE); + if (dma_mapping_error(&bnad->pcidev->dev, dma_addr)) { + put_page(page); + BNAD_UPDATE_CTR(bnad, rxbuf_map_failed); + rcb->rxq->rxbuf_map_failed++; + goto finishing; + } unmap->page = page; unmap->page_offset = page_offset; @@ -454,8 +460,15 @@ bnad_rxq_refill_skb(struct bnad *bnad, struct bna_rcb *rcb, u32 nalloc) rcb->rxq->rxbuf_alloc_failed++; goto finishing; } + dma_addr = dma_map_single(&bnad->pcidev->dev, skb->data, buff_sz, DMA_FROM_DEVICE); + if (dma_mapping_error(&bnad->pcidev->dev, dma_addr)) { + dev_kfree_skb_any(skb); + BNAD_UPDATE_CTR(bnad, rxbuf_map_failed); + rcb->rxq->rxbuf_map_failed++; + goto finishing; + } unmap->skb = skb; dma_unmap_addr_set(&unmap->vector, dma_addr, dma_addr); @@ -3025,6 +3038,11 @@ bnad_start_xmit(struct sk_buff *skb, struct net_device *netdev) unmap = head_unmap; dma_addr = dma_map_single(&bnad->pcidev->dev, skb->data, len, DMA_TO_DEVICE); + if (dma_mapping_error(&bnad->pcidev->dev, dma_addr)) { + dev_kfree_skb_any(skb); + BNAD_UPDATE_CTR(bnad, tx_skb_map_failed); + return NETDEV_TX_OK; + } BNA_SET_DMA_ADDR(dma_addr, &txqent->vector[0].host_addr); txqent->vector[0].length = htons(len); dma_unmap_addr_set(&unmap->vectors[0], dma_addr, dma_addr); @@ -3056,6 +3074,15 @@ bnad_start_xmit(struct sk_buff *skb, struct net_device *netdev) dma_addr = skb_frag_dma_map(&bnad->pcidev->dev, frag, 0, size, DMA_TO_DEVICE); + if (dma_mapping_error(&bnad->pcidev->dev, dma_addr)) { + /* Undo the changes starting at tcb->producer_index */ + bnad_tx_buff_unmap(bnad, unmap_q, q_depth, + tcb->producer_index); + dev_kfree_skb_any(skb); + BNAD_UPDATE_CTR(bnad, tx_skb_map_failed); + return NETDEV_TX_OK; + } + dma_unmap_len_set(&unmap->vectors[vect_id], dma_len, size); BNA_SET_DMA_ADDR(dma_addr, &txqent->vector[vect_id].host_addr); txqent->vector[vect_id].length = htons(size); diff --git a/drivers/net/ethernet/brocade/bna/bnad.h b/drivers/net/ethernet/brocade/bna/bnad.h index faedbf24777e..f4ed816b93ee 100644 --- a/drivers/net/ethernet/brocade/bna/bnad.h +++ b/drivers/net/ethernet/brocade/bna/bnad.h @@ -175,6 +175,7 @@ struct bnad_drv_stats { u64 tx_skb_headlen_zero; u64 tx_skb_frag_zero; u64 tx_skb_len_mismatch; + u64 tx_skb_map_failed; u64 hw_stats_updates; u64 netif_rx_dropped; @@ -189,6 +190,7 @@ struct bnad_drv_stats { u64 rx_unmap_q_alloc_failed; u64 rxbuf_alloc_failed; + u64 rxbuf_map_failed; }; /* Complete driver stats */ diff --git a/drivers/net/ethernet/brocade/bna/bnad_ethtool.c b/drivers/net/ethernet/brocade/bna/bnad_ethtool.c index 2bdfc5dff4b1..0e4fdc3dd729 100644 --- a/drivers/net/ethernet/brocade/bna/bnad_ethtool.c +++ b/drivers/net/ethernet/brocade/bna/bnad_ethtool.c @@ -90,6 +90,7 @@ static const char *bnad_net_stats_strings[BNAD_ETHTOOL_STATS_NUM] = { "tx_skb_headlen_zero", "tx_skb_frag_zero", "tx_skb_len_mismatch", + "tx_skb_map_failed", "hw_stats_updates", "netif_rx_dropped", @@ -102,6 +103,7 @@ static const char *bnad_net_stats_strings[BNAD_ETHTOOL_STATS_NUM] = { "tx_unmap_q_alloc_failed", "rx_unmap_q_alloc_failed", "rxbuf_alloc_failed", + "rxbuf_map_failed", "mac_stats_clr_cnt", "mac_frame_64", @@ -807,6 +809,7 @@ bnad_per_q_stats_fill(struct bnad *bnad, u64 *buf, int bi) rx_packets_with_error; buf[bi++] = rcb->rxq-> rxbuf_alloc_failed; + buf[bi++] = rcb->rxq->rxbuf_map_failed; buf[bi++] = rcb->producer_index; buf[bi++] = rcb->consumer_index; } @@ -821,6 +824,7 @@ bnad_per_q_stats_fill(struct bnad *bnad, u64 *buf, int bi) rx_packets_with_error; buf[bi++] = rcb->rxq-> rxbuf_alloc_failed; + buf[bi++] = rcb->rxq->rxbuf_map_failed; buf[bi++] = rcb->producer_index; buf[bi++] = rcb->consumer_index; } From 0315e382704817b279e5693dca8ab9d89aa20b3f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nikola=20Forr=C3=B3?= Date: Thu, 17 Sep 2015 16:01:32 +0200 Subject: [PATCH 243/647] net: Fix behaviour of unreachable, blackhole and prohibit routes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Man page of ip-route(8) says following about route types: unreachable - these destinations are unreachable. Packets are dis‐ carded and the ICMP message host unreachable is generated. The local senders get an EHOSTUNREACH error. blackhole - these destinations are unreachable. Packets are dis‐ carded silently. The local senders get an EINVAL error. prohibit - these destinations are unreachable. Packets are discarded and the ICMP message communication administratively prohibited is generated. The local senders get an EACCES error. In the inet6 address family, this was correct, except the local senders got ENETUNREACH error instead of EHOSTUNREACH in case of unreachable route. In the inet address family, all three route types generated ICMP message net unreachable, and the local senders got ENETUNREACH error. In both address families all three route types now behave consistently with documentation. Signed-off-by: Nikola Forró Signed-off-by: David S. Miller --- include/net/ip_fib.h | 30 +++++++++++++++++++----------- net/ipv4/route.c | 6 ++++-- net/ipv6/route.c | 4 +++- 3 files changed, 26 insertions(+), 14 deletions(-) diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index a37d0432bebd..727d6e9a9685 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -236,8 +236,11 @@ static inline int fib_lookup(struct net *net, const struct flowi4 *flp, rcu_read_lock(); tb = fib_get_table(net, RT_TABLE_MAIN); - if (tb && !fib_table_lookup(tb, flp, res, flags | FIB_LOOKUP_NOREF)) - err = 0; + if (tb) + err = fib_table_lookup(tb, flp, res, flags | FIB_LOOKUP_NOREF); + + if (err == -EAGAIN) + err = -ENETUNREACH; rcu_read_unlock(); @@ -258,7 +261,7 @@ static inline int fib_lookup(struct net *net, struct flowi4 *flp, struct fib_result *res, unsigned int flags) { struct fib_table *tb; - int err; + int err = -ENETUNREACH; flags |= FIB_LOOKUP_NOREF; if (net->ipv4.fib_has_custom_rules) @@ -268,15 +271,20 @@ static inline int fib_lookup(struct net *net, struct flowi4 *flp, res->tclassid = 0; - for (err = 0; !err; err = -ENETUNREACH) { - tb = rcu_dereference_rtnl(net->ipv4.fib_main); - if (tb && !fib_table_lookup(tb, flp, res, flags)) - break; + tb = rcu_dereference_rtnl(net->ipv4.fib_main); + if (tb) + err = fib_table_lookup(tb, flp, res, flags); - tb = rcu_dereference_rtnl(net->ipv4.fib_default); - if (tb && !fib_table_lookup(tb, flp, res, flags)) - break; - } + if (!err) + goto out; + + tb = rcu_dereference_rtnl(net->ipv4.fib_default); + if (tb) + err = fib_table_lookup(tb, flp, res, flags); + +out: + if (err == -EAGAIN) + err = -ENETUNREACH; rcu_read_unlock(); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 5f4a5565ad8b..c6ad99ad0ffb 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2045,6 +2045,7 @@ struct rtable *__ip_route_output_key(struct net *net, struct flowi4 *fl4) struct fib_result res; struct rtable *rth; int orig_oif; + int err = -ENETUNREACH; res.tclassid = 0; res.fi = NULL; @@ -2153,7 +2154,8 @@ struct rtable *__ip_route_output_key(struct net *net, struct flowi4 *fl4) goto make_route; } - if (fib_lookup(net, fl4, &res, 0)) { + err = fib_lookup(net, fl4, &res, 0); + if (err) { res.fi = NULL; res.table = NULL; if (fl4->flowi4_oif) { @@ -2181,7 +2183,7 @@ struct rtable *__ip_route_output_key(struct net *net, struct flowi4 *fl4) res.type = RTN_UNICAST; goto make_route; } - rth = ERR_PTR(-ENETUNREACH); + rth = ERR_PTR(err); goto out; } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index d5fa50297f80..f204089e854c 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1885,9 +1885,11 @@ int ip6_route_info_create(struct fib6_config *cfg, struct rt6_info **rt_ret) rt->dst.input = ip6_pkt_prohibit; break; case RTN_THROW: + case RTN_UNREACHABLE: default: rt->dst.error = (cfg->fc_type == RTN_THROW) ? -EAGAIN - : -ENETUNREACH; + : (cfg->fc_type == RTN_UNREACHABLE) + ? -EHOSTUNREACH : -ENETUNREACH; rt->dst.output = ip6_pkt_discard_out; rt->dst.input = ip6_pkt_discard; break; From fc27bd115b334e3ebdc682a42a47c3aea2566dcc Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 18 Sep 2015 00:19:08 +0100 Subject: [PATCH 244/647] 8139cp: Use dev_kfree_skb_any() instead of dev_kfree_skb() in cp_clean_rings() This can be called from cp_tx_timeout() with interrupts disabled. Spotted by Francois Romieu Signed-off-by: David Woodhouse Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/8139cp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index d79e33b3c191..52a533445f30 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -1151,7 +1151,7 @@ static void cp_clean_rings (struct cp_private *cp) desc = cp->rx_ring + i; dma_unmap_single(&cp->pdev->dev,le64_to_cpu(desc->addr), cp->rx_buf_sz, PCI_DMA_FROMDEVICE); - dev_kfree_skb(cp->rx_skb[i]); + dev_kfree_skb_any(cp->rx_skb[i]); } } @@ -1164,7 +1164,7 @@ static void cp_clean_rings (struct cp_private *cp) le32_to_cpu(desc->opts1) & 0xffff, PCI_DMA_TODEVICE); if (le32_to_cpu(desc->opts1) & LastFrag) - dev_kfree_skb(skb); + dev_kfree_skb_any(skb); cp->dev->stats.tx_dropped++; } } From 7a8a8e75d505147358b225173e890ada43a267e2 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 18 Sep 2015 00:21:54 +0100 Subject: [PATCH 245/647] 8139cp: Call __cp_set_rx_mode() from cp_tx_timeout() Unless we reset the RX config, on real hardware I don't seem to receive any packets after a TX timeout. Signed-off-by: David Woodhouse Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/8139cp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index 52a533445f30..ba3dab721806 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -1261,6 +1261,7 @@ static void cp_tx_timeout(struct net_device *dev) cp_clean_rings(cp); rc = cp_init_rings(cp); cp_start_hw(cp); + __cp_set_rx_mode(dev); cp_enable_irq(cp); netif_wake_queue(dev); From aab0c0e62ec4af224d1b6b40fca65055d403400b Mon Sep 17 00:00:00 2001 From: Kevin Hao Date: Fri, 18 Sep 2015 15:42:30 +0800 Subject: [PATCH 246/647] Revert "net/phy: Add Vitesse 8641 phy ID" This reverts commit 1298267b548a78840bd4b3e030993ff8747ca5e6. That commit claim that the Vitesse VSC8641 is compatible with Vitesse 82xx. But this is not true. It seems that all the registers used in Vitesse phy driver are not compatible between 8641 and 82xx. It does cause malfunction of the Ethernet on p1010rdb-pa board. So we definitely need a rework in order to support the 8641 phy in this driver. Signed-off-by: Kevin Hao Signed-off-by: David S. Miller --- drivers/net/phy/vitesse.c | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/drivers/net/phy/vitesse.c b/drivers/net/phy/vitesse.c index 17cad185169d..76cad712ddb2 100644 --- a/drivers/net/phy/vitesse.c +++ b/drivers/net/phy/vitesse.c @@ -66,7 +66,6 @@ #define PHY_ID_VSC8244 0x000fc6c0 #define PHY_ID_VSC8514 0x00070670 #define PHY_ID_VSC8574 0x000704a0 -#define PHY_ID_VSC8641 0x00070431 #define PHY_ID_VSC8662 0x00070660 #define PHY_ID_VSC8221 0x000fc550 #define PHY_ID_VSC8211 0x000fc4b0 @@ -272,18 +271,6 @@ static struct phy_driver vsc82xx_driver[] = { .ack_interrupt = &vsc824x_ack_interrupt, .config_intr = &vsc82xx_config_intr, .driver = { .owner = THIS_MODULE,}, -}, { - .phy_id = PHY_ID_VSC8641, - .name = "Vitesse VSC8641", - .phy_id_mask = 0x000ffff0, - .features = PHY_GBIT_FEATURES, - .flags = PHY_HAS_INTERRUPT, - .config_init = &vsc824x_config_init, - .config_aneg = &vsc82x4_config_aneg, - .read_status = &genphy_read_status, - .ack_interrupt = &vsc824x_ack_interrupt, - .config_intr = &vsc82xx_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_VSC8662, .name = "Vitesse VSC8662", @@ -331,7 +318,6 @@ static struct mdio_device_id __maybe_unused vitesse_tbl[] = { { PHY_ID_VSC8244, 0x000fffc0 }, { PHY_ID_VSC8514, 0x000ffff0 }, { PHY_ID_VSC8574, 0x000ffff0 }, - { PHY_ID_VSC8641, 0x000ffff0 }, { PHY_ID_VSC8662, 0x000ffff0 }, { PHY_ID_VSC8221, 0x000ffff0 }, { PHY_ID_VSC8211, 0x000ffff0 }, From 4e3ae00100945d39e1f83b7c0179a114ccf55759 Mon Sep 17 00:00:00 2001 From: Erik Hugne Date: Fri, 18 Sep 2015 10:46:31 +0200 Subject: [PATCH 247/647] tipc: reinitialize pointer after skb linearize MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The msg pointer into header may change after skb linearization. We must reinitialize it after calling skb_linearize to prevent operating on a freed or invalid pointer. Signed-off-by: Erik Hugne Reported-by: Tamás Végh Acked-by: Ying Xue Signed-off-by: David S. Miller --- net/tipc/msg.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/tipc/msg.c b/net/tipc/msg.c index 562c926a51cc..c5ac436235e0 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -539,6 +539,7 @@ bool tipc_msg_lookup_dest(struct net *net, struct sk_buff *skb, int *err) *err = -TIPC_ERR_NO_NAME; if (skb_linearize(skb)) return false; + msg = buf_msg(skb); if (msg_reroute_cnt(msg)) return false; dnode = addr_domain(net, msg_lookup_scope(msg)); From bc22a0e2ea03b75b51a1f722f93821744b5b5ff1 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Fri, 18 Sep 2015 11:47:40 +0200 Subject: [PATCH 248/647] iptunnel: make rx/tx bytes counters consistent This was already done a long time ago in commit 64194c31a0b6 ("inet: Make tunnel RX/TX byte counters more consistent") but tx path was broken (at least since 3.10). Before the patch the gre header was included on tx. After the patch: $ ping -c1 192.168.0.121 ; ip -s l ls dev gre1 PING 192.168.0.121 (192.168.0.121) 56(84) bytes of data. 64 bytes from 192.168.0.121: icmp_req=1 ttl=64 time=2.95 ms --- 192.168.0.121 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 2.955/2.955/2.955/0.000 ms 7: gre1@NONE: mtu 1468 qdisc noqueue state UNKNOWN mode DEFAULT group default link/gre 10.16.0.249 peer 10.16.0.121 RX: bytes packets errors dropped overrun mcast 84 1 0 0 0 0 TX: bytes packets errors dropped carrier collsns 84 1 0 0 0 0 Reported-by: Julien Meunier Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller --- net/ipv4/ip_tunnel_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 29ed6c5a5185..9b97204b8c81 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -51,7 +51,7 @@ int iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, __be32 src, __be32 dst, __u8 proto, __u8 tos, __u8 ttl, __be16 df, bool xnet) { - int pkt_len = skb->len; + int pkt_len = skb->len - skb_inner_network_offset(skb); struct iphdr *iph; int err; From 83cf9a2521b0934a5f9d04082c9bb4f554fddcd4 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Fri, 18 Sep 2015 11:47:41 +0200 Subject: [PATCH 249/647] ip6tunnel: make rx/tx bytes counters consistent Like the previous patch, which fixes ipv4 tunnels, here is the ipv6 part. Before the patch, the external ipv6 header + gre header were included on tx. After the patch: $ ping -c1 192.168.6.121 ; ip -s l ls dev ip6gre1 PING 192.168.6.121 (192.168.6.121) 56(84) bytes of data. 64 bytes from 192.168.6.121: icmp_req=1 ttl=64 time=1.92 ms --- 192.168.6.121 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 1.923/1.923/1.923/0.000 ms 7: ip6gre1@NONE: mtu 1440 qdisc noqueue state UNKNOWN mode DEFAULT group default link/gre6 20:01:06:60:30:08:c1:c3:00:00:00:00:00:00:01:23 peer 20:01:06:60:30:08:c1:c3:00:00:00:00:00:00:01:21 RX: bytes packets errors dropped overrun mcast 84 1 0 0 0 0 TX: bytes packets errors dropped carrier collsns 84 1 0 0 0 0 $ ping -c1 192.168.1.121 ; ip -s l ls dev ip6tnl1 PING 192.168.1.121 (192.168.1.121) 56(84) bytes of data. 64 bytes from 192.168.1.121: icmp_req=1 ttl=64 time=2.28 ms --- 192.168.1.121 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 2.288/2.288/2.288/0.000 ms 8: ip6tnl1@NONE: mtu 1452 qdisc noqueue state UNKNOWN mode DEFAULT group default link/tunnel6 2001:660:3008:c1c3::123 peer 2001:660:3008:c1c3::121 RX: bytes packets errors dropped overrun mcast 84 1 0 0 0 0 TX: bytes packets errors dropped carrier collsns 84 1 0 0 0 0 Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller --- include/net/ip6_tunnel.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h index 65c2a9397b3c..fa915fa0f703 100644 --- a/include/net/ip6_tunnel.h +++ b/include/net/ip6_tunnel.h @@ -86,7 +86,7 @@ static inline void ip6tunnel_xmit(struct sock *sk, struct sk_buff *skb, struct net_device_stats *stats = &dev->stats; int pkt_len, err; - pkt_len = skb->len; + pkt_len = skb->len - skb_inner_network_offset(skb); err = ip6_local_out_sk(sk, skb); if (net_xmit_eval(err) == 0) { From 3afb1121800128aae9f5722e50097fcf1a9d4d88 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Fri, 18 Sep 2015 17:33:04 +0200 Subject: [PATCH 250/647] KVM: x86: trap AMD MSRs for the TSeg base and mask These have roughly the same purpose as the SMRR, which we do not need to implement in KVM. However, Linux accesses MSR_K8_TSEG_ADDR at boot, which causes problems when running a Xen dom0 under KVM. Just return 0, meaning that processor protection of SMRAM is not in effect. Reported-by: M A Young Cc: stable@vger.kernel.org Acked-by: Borislav Petkov Signed-off-by: Paolo Bonzini --- arch/x86/include/asm/msr-index.h | 1 + arch/x86/kvm/x86.c | 2 ++ 2 files changed, 3 insertions(+) diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index c1c0a1c14344..b98b471a3b7e 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -331,6 +331,7 @@ /* C1E active bits in int pending message */ #define K8_INTP_C1E_ACTIVE_MASK 0x18000000 #define MSR_K8_TSEG_ADDR 0xc0010112 +#define MSR_K8_TSEG_MASK 0xc0010113 #define K8_MTRRFIXRANGE_DRAM_ENABLE 0x00040000 /* MtrrFixDramEn bit */ #define K8_MTRRFIXRANGE_DRAM_MODIFY 0x00080000 /* MtrrFixDramModEn bit */ #define K8_MTRR_RDMEM_WRMEM_MASK 0x18181818 /* Mask: RdMem|WrMem */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 6bbb0dfb99d0..991466bf8dee 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2190,6 +2190,8 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info) case MSR_IA32_LASTINTFROMIP: case MSR_IA32_LASTINTTOIP: case MSR_K8_SYSCFG: + case MSR_K8_TSEG_ADDR: + case MSR_K8_TSEG_MASK: case MSR_K7_HWCR: case MSR_VM_HSAVE_PA: case MSR_K8_INT_PENDING_MSG: From 3ea79249e81e5ed051f2e6480cbde896d99046e8 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Fri, 18 Sep 2015 13:41:09 +0300 Subject: [PATCH 251/647] macvtap: fix TUNSETSNDBUF values > 64k Upon TUNSETSNDBUF, macvtap reads the requested sndbuf size into a local variable u. commit 39ec7de7092b ("macvtap: fix uninitialized access on TUNSETIFF") changed its type to u16 (which is the right thing to do for all other macvtap ioctls), breaking all values > 64k. The value of TUNSETSNDBUF is actually a signed 32 bit integer, so the right thing to do is to read it into an int. Cc: David S. Miller Fixes: 39ec7de7092b ("macvtap: fix uninitialized access on TUNSETIFF") Reported-by: Mark A. Peloquin Bisected-by: Matthew Rosato Reported-by: Christian Borntraeger Signed-off-by: Michael S. Tsirkin Tested-by: Matthew Rosato Acked-by: Christian Borntraeger Signed-off-by: David S. Miller --- drivers/net/macvtap.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index edd77342773a..248478c6f6e4 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -1111,10 +1111,10 @@ static long macvtap_ioctl(struct file *file, unsigned int cmd, return 0; case TUNSETSNDBUF: - if (get_user(u, up)) + if (get_user(s, sp)) return -EFAULT; - q->sk.sk_sndbuf = u; + q->sk.sk_sndbuf = s; return 0; case TUNGETVNETHDRSZ: From 5964db08767fb93376c0a0ddcf975fec1c0d1bd9 Mon Sep 17 00:00:00 2001 From: Tomas Winkler Date: Mon, 24 Aug 2015 15:27:37 +0300 Subject: [PATCH 252/647] mei: fix debugfs files leak on error path if dbgfs_dir is not set then debugfs_remove_recursive is not called on the error path Signed-off-by: Tomas Winkler Signed-off-by: Greg Kroah-Hartman --- drivers/misc/mei/debugfs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/misc/mei/debugfs.c b/drivers/misc/mei/debugfs.c index 4b469cf9e60f..8504dbeacd3b 100644 --- a/drivers/misc/mei/debugfs.c +++ b/drivers/misc/mei/debugfs.c @@ -204,6 +204,8 @@ int mei_dbgfs_register(struct mei_device *dev, const char *name) if (!dir) return -ENOMEM; + dev->dbgfs_dir = dir; + f = debugfs_create_file("meclients", S_IRUSR, dir, dev, &mei_dbgfs_fops_meclients); if (!f) { @@ -228,7 +230,6 @@ int mei_dbgfs_register(struct mei_device *dev, const char *name) dev_err(dev->dev, "allow_fixed_address: registration failed\n"); goto err; } - dev->dbgfs_dir = dir; return 0; err: mei_dbgfs_deregister(dev); From ca1c4b745779e20711322b3338f3a942a3c1224a Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Thu, 13 Aug 2015 17:07:03 -0700 Subject: [PATCH 253/647] Drivers: hv: vmbus: fix init_vp_index() for reloading hv_netvsc This fixes the recent commit 3b71107d73b16074afa7658f3f0fcf837aabfe24: Drivers: hv: vmbus: Further improve CPU affiliation logic Without the fix, reloading hv_netvsc hangs the guest. Signed-off-by: Dexuan Cui Signed-off-by: K. Y. Srinivasan Signed-off-by: Greg Kroah-Hartman --- drivers/hv/channel_mgmt.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c index 2f9aead4ecfc..652afd11a9ef 100644 --- a/drivers/hv/channel_mgmt.c +++ b/drivers/hv/channel_mgmt.c @@ -204,6 +204,8 @@ void hv_process_channel_removal(struct vmbus_channel *channel, u32 relid) spin_lock_irqsave(&vmbus_connection.channel_lock, flags); list_del(&channel->listentry); spin_unlock_irqrestore(&vmbus_connection.channel_lock, flags); + + primary_channel = channel; } else { primary_channel = channel->primary_channel; spin_lock_irqsave(&primary_channel->lock, flags); @@ -211,6 +213,14 @@ void hv_process_channel_removal(struct vmbus_channel *channel, u32 relid) primary_channel->num_sc--; spin_unlock_irqrestore(&primary_channel->lock, flags); } + + /* + * We need to free the bit for init_vp_index() to work in the case + * of sub-channel, when we reload drivers like hv_netvsc. + */ + cpumask_clear_cpu(channel->target_cpu, + &primary_channel->alloced_cpus_in_node); + free_channel(channel); } @@ -458,6 +468,13 @@ static void init_vp_index(struct vmbus_channel *channel, const uuid_le *type_gui continue; } + /* + * NOTE: in the case of sub-channel, we clear the sub-channel + * related bit(s) in primary->alloced_cpus_in_node in + * hv_process_channel_removal(), so when we reload drivers + * like hv_netvsc in SMP guest, here we're able to re-allocate + * bit from primary->alloced_cpus_in_node. + */ if (!cpumask_test_cpu(cur_cpu, &primary->alloced_cpus_in_node)) { cpumask_set_cpu(cur_cpu, From 1f770c0a09da855a2b51af6d19de97fb955eca85 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 18 Sep 2015 19:16:50 +0800 Subject: [PATCH 254/647] netlink: Fix autobind race condition that leads to zero port ID The commit c0bb07df7d981e4091432754e30c9c720e2c0c78 ("netlink: Reset portid after netlink_insert failure") introduced a race condition where if two threads try to autobind the same socket one of them may end up with a zero port ID. This led to kernel deadlocks that were observed by multiple people. This patch reverts that commit and instead fixes it by introducing a separte rhash_portid variable so that the real portid is only set after the socket has been successfully hashed. Fixes: c0bb07df7d98 ("netlink: Reset portid after netlink_insert failure") Reported-by: Tejun Heo Reported-by: Linus Torvalds Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 12 +++++++----- net/netlink/af_netlink.h | 1 + 2 files changed, 8 insertions(+), 5 deletions(-) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 4cad99d6c68b..9f51608b968a 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1031,7 +1031,7 @@ static inline int netlink_compare(struct rhashtable_compare_arg *arg, const struct netlink_compare_arg *x = arg->key; const struct netlink_sock *nlk = ptr; - return nlk->portid != x->portid || + return nlk->rhash_portid != x->portid || !net_eq(sock_net(&nlk->sk), read_pnet(&x->pnet)); } @@ -1057,7 +1057,7 @@ static int __netlink_insert(struct netlink_table *table, struct sock *sk) { struct netlink_compare_arg arg; - netlink_compare_arg_init(&arg, sock_net(sk), nlk_sk(sk)->portid); + netlink_compare_arg_init(&arg, sock_net(sk), nlk_sk(sk)->rhash_portid); return rhashtable_lookup_insert_key(&table->hash, &arg, &nlk_sk(sk)->node, netlink_rhashtable_params); @@ -1119,7 +1119,7 @@ static int netlink_insert(struct sock *sk, u32 portid) unlikely(atomic_read(&table->hash.nelems) >= UINT_MAX)) goto err; - nlk_sk(sk)->portid = portid; + nlk_sk(sk)->rhash_portid = portid; sock_hold(sk); err = __netlink_insert(table, sk); @@ -1131,10 +1131,12 @@ static int netlink_insert(struct sock *sk, u32 portid) err = -EOVERFLOW; if (err == -EEXIST) err = -EADDRINUSE; - nlk_sk(sk)->portid = 0; sock_put(sk); + goto err; } + nlk_sk(sk)->portid = portid; + err: release_sock(sk); return err; @@ -3271,7 +3273,7 @@ static inline u32 netlink_hash(const void *data, u32 len, u32 seed) const struct netlink_sock *nlk = data; struct netlink_compare_arg arg; - netlink_compare_arg_init(&arg, sock_net(&nlk->sk), nlk->portid); + netlink_compare_arg_init(&arg, sock_net(&nlk->sk), nlk->rhash_portid); return jhash2((u32 *)&arg, netlink_compare_arg_len / sizeof(u32), seed); } diff --git a/net/netlink/af_netlink.h b/net/netlink/af_netlink.h index df9a06090db6..80b2b7526dfd 100644 --- a/net/netlink/af_netlink.h +++ b/net/netlink/af_netlink.h @@ -25,6 +25,7 @@ struct netlink_ring { struct netlink_sock { /* struct sock has to be the first member of netlink_sock */ struct sock sk; + u32 rhash_portid; u32 portid; u32 dst_portid; u32 dst_group; From f4513b065f7dbd37224226ef6e44b09eff742776 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Mon, 24 Aug 2015 00:35:36 +0200 Subject: [PATCH 255/647] extcon: Fix attached value returned by is_extcon_changed is_extcon_changed should only check the idx-th bit of new, not the entirety of new when setting attached. This fixes extcon sending notifications that a cable was inserted when it gets removed while another cable is still connected. Cc: stable@vger.kernel.org Signed-off-by: Hans de Goede Signed-off-by: Chanwoo Choi --- drivers/extcon/extcon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/extcon/extcon.c b/drivers/extcon/extcon.c index a07addde297b..8dd0af1d50bc 100644 --- a/drivers/extcon/extcon.c +++ b/drivers/extcon/extcon.c @@ -159,7 +159,7 @@ static int find_cable_index_by_name(struct extcon_dev *edev, const char *name) static bool is_extcon_changed(u32 prev, u32 new, int idx, bool *attached) { if (((prev >> idx) & 0x1) != ((new >> idx) & 0x1)) { - *attached = new ? true : false; + *attached = ((new >> idx) & 0x1) ? true : false; return true; } From 95d0be61286bcc3fc987043fe3a9553adca02919 Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Tue, 15 Sep 2015 06:04:36 -0700 Subject: [PATCH 256/647] drm/i915: fix kernel-doc warnings in intel_audio.c Fix the following 'make htmldocs' warnings: .//drivers/gpu/drm/i915/intel_audio.c:439: warning: No description found for parameter 'intel_encoder' .//drivers/gpu/drm/i915/intel_audio.c:439: warning: Excess function parameter 'encoder' description in 'intel_audio_codec_disable' .//drivers/gpu/drm/i915/intel_audio.c:439: warning: No description found for parameter 'intel_encoder' .//drivers/gpu/drm/i915/intel_audio.c:439: warning: Excess function parameter 'encoder' description in 'intel_audio_codec_disable' Signed-off-by: Geliang Tang Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_audio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c index 89c1a8ce1f98..2a5c76faf9f8 100644 --- a/drivers/gpu/drm/i915/intel_audio.c +++ b/drivers/gpu/drm/i915/intel_audio.c @@ -430,7 +430,7 @@ void intel_audio_codec_enable(struct intel_encoder *intel_encoder) /** * intel_audio_codec_disable - Disable the audio codec for HD audio - * @encoder: encoder on which to disable audio + * @intel_encoder: encoder on which to disable audio * * The disable sequences must be performed before disabling the transcoder or * port. From 793b8bf9ca17aee3b995c095058e6c4e7bd72e02 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Wed, 16 Sep 2015 21:21:50 +1000 Subject: [PATCH 257/647] powerpc: Wire up sys_membarrier() The selftest passes on 64-bit LE & BE, and 32-bit. Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/systbl.h | 1 + arch/powerpc/include/asm/unistd.h | 2 +- arch/powerpc/include/uapi/asm/unistd.h | 1 + 3 files changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/include/asm/systbl.h b/arch/powerpc/include/asm/systbl.h index 4d65499ee1c1..126d0c4f9b7d 100644 --- a/arch/powerpc/include/asm/systbl.h +++ b/arch/powerpc/include/asm/systbl.h @@ -369,3 +369,4 @@ SYSCALL_SPU(bpf) COMPAT_SYS(execveat) PPC64ONLY(switch_endian) SYSCALL_SPU(userfaultfd) +SYSCALL_SPU(membarrier) diff --git a/arch/powerpc/include/asm/unistd.h b/arch/powerpc/include/asm/unistd.h index 4a055b6c2a64..13411be86041 100644 --- a/arch/powerpc/include/asm/unistd.h +++ b/arch/powerpc/include/asm/unistd.h @@ -12,7 +12,7 @@ #include -#define __NR_syscalls 365 +#define __NR_syscalls 366 #define __NR__exit __NR_exit #define NR_syscalls __NR_syscalls diff --git a/arch/powerpc/include/uapi/asm/unistd.h b/arch/powerpc/include/uapi/asm/unistd.h index 6ad58d4c879b..6337738018aa 100644 --- a/arch/powerpc/include/uapi/asm/unistd.h +++ b/arch/powerpc/include/uapi/asm/unistd.h @@ -387,5 +387,6 @@ #define __NR_execveat 362 #define __NR_switch_endian 363 #define __NR_userfaultfd 364 +#define __NR_membarrier 365 #endif /* _UAPI_ASM_POWERPC_UNISTD_H_ */ From cfcd2271a9076a9891014bc8e18d4fd48acccffe Mon Sep 17 00:00:00 2001 From: Thomas Petazzoni Date: Fri, 18 Sep 2015 17:25:36 +0200 Subject: [PATCH 258/647] crypto: marvell - properly handle CRYPTO_TFM_REQ_MAY_BACKLOG-flagged requests The mv_cesa_queue_req() function calls crypto_enqueue_request() to enqueue a request. In the normal case (i.e the queue isn't full), this function returns -EINPROGRESS. The current Marvell CESA crypto driver takes this into account and cleans up the request only if an error occured, i.e if the return value is not -EINPROGRESS. Unfortunately this causes problems with CRYPTO_TFM_REQ_MAY_BACKLOG-flagged requests. When such a request is passed to crypto_enqueue_request() and the queue is full, crypto_enqueue_request() will return -EBUSY, but will keep the request enqueued nonetheless. This situation was not properly handled by the Marvell CESA driver, which was anyway cleaning up the request in such a situation. When later on the request was taken out of the backlog and actually processed, a kernel crash occured due to the internal driver data structures for this structure having been cleaned up. To avoid this situation, this commit adds a mv_cesa_req_needs_cleanup() helper function which indicates if the request needs to be cleaned up or not after a call to crypto_enqueue_request(). This helper allows to do the cleanup only in the appropriate cases, and all call sites of mv_cesa_queue_req() are fixed to use this new helper function. Reported-by: Vincent Donnefort Fixes: db509a45339fd ("crypto: marvell/cesa - add TDMA support") Cc: # v4.2+ Signed-off-by: Thomas Petazzoni Acked-by: Boris Brezillon Tested-by: Vincent Donnefort Signed-off-by: Herbert Xu --- drivers/crypto/marvell/cesa.h | 27 +++++++++++++++++++++++++++ drivers/crypto/marvell/cipher.c | 7 +++---- drivers/crypto/marvell/hash.c | 8 +++----- 3 files changed, 33 insertions(+), 9 deletions(-) diff --git a/drivers/crypto/marvell/cesa.h b/drivers/crypto/marvell/cesa.h index b60698b30d30..bc2a55bc35e4 100644 --- a/drivers/crypto/marvell/cesa.h +++ b/drivers/crypto/marvell/cesa.h @@ -687,6 +687,33 @@ static inline u32 mv_cesa_get_int_mask(struct mv_cesa_engine *engine) int mv_cesa_queue_req(struct crypto_async_request *req); +/* + * Helper function that indicates whether a crypto request needs to be + * cleaned up or not after being enqueued using mv_cesa_queue_req(). + */ +static inline int mv_cesa_req_needs_cleanup(struct crypto_async_request *req, + int ret) +{ + /* + * The queue still had some space, the request was queued + * normally, so there's no need to clean it up. + */ + if (ret == -EINPROGRESS) + return false; + + /* + * The queue had not space left, but since the request is + * flagged with CRYPTO_TFM_REQ_MAY_BACKLOG, it was added to + * the backlog and will be processed later. There's no need to + * clean it up. + */ + if (ret == -EBUSY && req->flags & CRYPTO_TFM_REQ_MAY_BACKLOG) + return false; + + /* Request wasn't queued, we need to clean it up */ + return true; +} + /* TDMA functions */ static inline void mv_cesa_req_dma_iter_init(struct mv_cesa_dma_iter *iter, diff --git a/drivers/crypto/marvell/cipher.c b/drivers/crypto/marvell/cipher.c index 0745cf3b9c0e..3df2f4e7adb2 100644 --- a/drivers/crypto/marvell/cipher.c +++ b/drivers/crypto/marvell/cipher.c @@ -189,7 +189,6 @@ static inline void mv_cesa_ablkcipher_prepare(struct crypto_async_request *req, { struct ablkcipher_request *ablkreq = ablkcipher_request_cast(req); struct mv_cesa_ablkcipher_req *creq = ablkcipher_request_ctx(ablkreq); - creq->req.base.engine = engine; if (creq->req.base.type == CESA_DMA_REQ) @@ -431,7 +430,7 @@ static int mv_cesa_des_op(struct ablkcipher_request *req, return ret; ret = mv_cesa_queue_req(&req->base); - if (ret && ret != -EINPROGRESS) + if (mv_cesa_req_needs_cleanup(&req->base, ret)) mv_cesa_ablkcipher_cleanup(req); return ret; @@ -551,7 +550,7 @@ static int mv_cesa_des3_op(struct ablkcipher_request *req, return ret; ret = mv_cesa_queue_req(&req->base); - if (ret && ret != -EINPROGRESS) + if (mv_cesa_req_needs_cleanup(&req->base, ret)) mv_cesa_ablkcipher_cleanup(req); return ret; @@ -693,7 +692,7 @@ static int mv_cesa_aes_op(struct ablkcipher_request *req, return ret; ret = mv_cesa_queue_req(&req->base); - if (ret && ret != -EINPROGRESS) + if (mv_cesa_req_needs_cleanup(&req->base, ret)) mv_cesa_ablkcipher_cleanup(req); return ret; diff --git a/drivers/crypto/marvell/hash.c b/drivers/crypto/marvell/hash.c index ae9272eb9c1a..e8d0d7128137 100644 --- a/drivers/crypto/marvell/hash.c +++ b/drivers/crypto/marvell/hash.c @@ -739,10 +739,8 @@ static int mv_cesa_ahash_update(struct ahash_request *req) return 0; ret = mv_cesa_queue_req(&req->base); - if (ret && ret != -EINPROGRESS) { + if (mv_cesa_req_needs_cleanup(&req->base, ret)) mv_cesa_ahash_cleanup(req); - return ret; - } return ret; } @@ -766,7 +764,7 @@ static int mv_cesa_ahash_final(struct ahash_request *req) return 0; ret = mv_cesa_queue_req(&req->base); - if (ret && ret != -EINPROGRESS) + if (mv_cesa_req_needs_cleanup(&req->base, ret)) mv_cesa_ahash_cleanup(req); return ret; @@ -791,7 +789,7 @@ static int mv_cesa_ahash_finup(struct ahash_request *req) return 0; ret = mv_cesa_queue_req(&req->base); - if (ret && ret != -EINPROGRESS) + if (mv_cesa_req_needs_cleanup(&req->base, ret)) mv_cesa_ahash_cleanup(req); return ret; From 17762c5acee2f1f065291d2a74ebea4d0168063b Mon Sep 17 00:00:00 2001 From: Conor McLoughlin Date: Fri, 18 Sep 2015 08:59:47 -0700 Subject: [PATCH 259/647] crypto: qat - VF should never trigger SBR on PH Don't allow to trigger SBR from a VF running in VMM context. Cc: stable@vger.kernel.org Signed-off-by: Conor McLoughlin Signed-off-by: Tadeusz Struk Signed-off-by: Herbert Xu --- drivers/crypto/qat/qat_common/adf_aer.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/crypto/qat/qat_common/adf_aer.c b/drivers/crypto/qat/qat_common/adf_aer.c index a57b4194de28..0a5ca0ba5d64 100644 --- a/drivers/crypto/qat/qat_common/adf_aer.c +++ b/drivers/crypto/qat/qat_common/adf_aer.c @@ -88,6 +88,9 @@ static void adf_dev_restore(struct adf_accel_dev *accel_dev) struct pci_dev *parent = pdev->bus->self; uint16_t bridge_ctl = 0; + if (accel_dev->is_vf) + return; + dev_info(&GET_DEV(accel_dev), "Resetting device qat_dev%d\n", accel_dev->accel_id); From 09185e2756a83763bcbfe51b308b85979f0de027 Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Mon, 21 Sep 2015 15:33:42 +0200 Subject: [PATCH 260/647] hwrng: xgene - fix handling platform_get_irq The function can return negative value. The problem has been detected using proposed semantic patch scripts/coccinelle/tests/unsigned_lesser_than_zero.cocci [1]. [1]: http://permalink.gmane.org/gmane.linux.kernel/2038576 Signed-off-by: Andrzej Hajda Signed-off-by: Herbert Xu --- drivers/char/hw_random/xgene-rng.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/char/hw_random/xgene-rng.c b/drivers/char/hw_random/xgene-rng.c index c37cf754a985..3c77645405e5 100644 --- a/drivers/char/hw_random/xgene-rng.c +++ b/drivers/char/hw_random/xgene-rng.c @@ -344,11 +344,12 @@ static int xgene_rng_probe(struct platform_device *pdev) if (IS_ERR(ctx->csr_base)) return PTR_ERR(ctx->csr_base); - ctx->irq = platform_get_irq(pdev, 0); - if (ctx->irq < 0) { + rc = platform_get_irq(pdev, 0); + if (rc < 0) { dev_err(&pdev->dev, "No IRQ resource\n"); - return ctx->irq; + return rc; } + ctx->irq = rc; dev_dbg(&pdev->dev, "APM X-Gene RNG BASE %p ALARM IRQ %d", ctx->csr_base, ctx->irq); From 9d7b03f863f2fc8bd80e83d3a2d7e521094d290e Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Sat, 19 Sep 2015 09:49:45 -0700 Subject: [PATCH 261/647] Input: elan_i2c - don't require known iap version The Asus X456UA has an ELAN1000 touchpad with IAP version 0xe. This is unknown to elan_get_fwinfo() so driver probe fails and I am left with an unusable touchpad. However, the fwinfo is not required for general driver usage, it is only needed if the user decides to upload new firmware. Adjust the driver so that we do not abort probe when we encounter unexpected IAP version, but rather warn user that firmware update feature of the driver will not work. Signed-off-by: Daniel Drake Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/elan_i2c_core.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/input/mouse/elan_i2c_core.c b/drivers/input/mouse/elan_i2c_core.c index fa945304b9a5..7cad81942b8c 100644 --- a/drivers/input/mouse/elan_i2c_core.c +++ b/drivers/input/mouse/elan_i2c_core.c @@ -266,11 +266,10 @@ static int elan_query_device_info(struct elan_tp_data *data) error = elan_get_fwinfo(data->iap_version, &data->fw_validpage_count, &data->fw_signature_address); - if (error) { - dev_err(&data->client->dev, - "unknown iap version %d\n", data->iap_version); - return error; - } + if (error) + dev_warn(&data->client->dev, + "unexpected iap version %#04x (ic type: %#04x), firmware update will not work\n", + data->iap_version, data->ic_type); return 0; } @@ -486,6 +485,9 @@ static ssize_t elan_sysfs_update_fw(struct device *dev, const u8 *fw_signature; static const u8 signature[] = {0xAA, 0x55, 0xCC, 0x33, 0xFF, 0xFF}; + if (data->fw_validpage_count == 0) + return -EINVAL; + /* Look for a firmware with the product id appended. */ fw_name = kasprintf(GFP_KERNEL, ETP_FW_NAME, data->product_id); if (!fw_name) { From c84333a1c55a2d3329e8b472f59e9d1ccf5ddc1b Mon Sep 17 00:00:00 2001 From: Duson Lin Date: Mon, 21 Sep 2015 09:24:46 -0700 Subject: [PATCH 262/647] Input: elan_i2c - add ic type 0x03 The 0x03 is valid 3000 serial ic type too. Signed-off-by: Duson Lin Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/elan_i2c_core.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/mouse/elan_i2c_core.c b/drivers/input/mouse/elan_i2c_core.c index 7cad81942b8c..0f8ab3170055 100644 --- a/drivers/input/mouse/elan_i2c_core.c +++ b/drivers/input/mouse/elan_i2c_core.c @@ -102,6 +102,7 @@ static int elan_get_fwinfo(u8 iap_version, u16 *validpage_count, *validpage_count = 512; break; case 0x09: + case 0x03: *validpage_count = 768; break; case 0x0D: From ed75a14eb554995c522a3b861d4c86a7eddb2ad7 Mon Sep 17 00:00:00 2001 From: Duson Lin Date: Mon, 21 Sep 2015 09:26:46 -0700 Subject: [PATCH 263/647] Input: elan_i2c - expand maximum product_id form 0xFF to 0xFFFF In order to support more projects in the future, we expand the maximum product_id value form 0xFF to 0xFFFF. Signed-off by: Duson Lin Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/elan_i2c.h | 2 +- drivers/input/mouse/elan_i2c_core.c | 4 ++-- drivers/input/mouse/elan_i2c_i2c.c | 4 ++-- drivers/input/mouse/elan_i2c_smbus.c | 4 ++-- 4 files changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/input/mouse/elan_i2c.h b/drivers/input/mouse/elan_i2c.h index 73670f2aebfd..c0ec26118732 100644 --- a/drivers/input/mouse/elan_i2c.h +++ b/drivers/input/mouse/elan_i2c.h @@ -60,7 +60,7 @@ struct elan_transport_ops { int (*get_sm_version)(struct i2c_client *client, u8* ic_type, u8 *version); int (*get_checksum)(struct i2c_client *client, bool iap, u16 *csum); - int (*get_product_id)(struct i2c_client *client, u8 *id); + int (*get_product_id)(struct i2c_client *client, u16 *id); int (*get_max)(struct i2c_client *client, unsigned int *max_x, unsigned int *max_y); diff --git a/drivers/input/mouse/elan_i2c_core.c b/drivers/input/mouse/elan_i2c_core.c index 0f8ab3170055..b51062622050 100644 --- a/drivers/input/mouse/elan_i2c_core.c +++ b/drivers/input/mouse/elan_i2c_core.c @@ -40,7 +40,7 @@ #include "elan_i2c.h" #define DRIVER_NAME "elan_i2c" -#define ELAN_DRIVER_VERSION "1.6.0" +#define ELAN_DRIVER_VERSION "1.6.1" #define ETP_MAX_PRESSURE 255 #define ETP_FWIDTH_REDUCE 90 #define ETP_FINGER_WIDTH 15 @@ -76,7 +76,7 @@ struct elan_tp_data { unsigned int x_res; unsigned int y_res; - u8 product_id; + u16 product_id; u8 fw_version; u8 sm_version; u8 iap_version; diff --git a/drivers/input/mouse/elan_i2c_i2c.c b/drivers/input/mouse/elan_i2c_i2c.c index 683c840c9dd7..a679e56c44cd 100644 --- a/drivers/input/mouse/elan_i2c_i2c.c +++ b/drivers/input/mouse/elan_i2c_i2c.c @@ -276,7 +276,7 @@ static int elan_i2c_get_sm_version(struct i2c_client *client, return 0; } -static int elan_i2c_get_product_id(struct i2c_client *client, u8 *id) +static int elan_i2c_get_product_id(struct i2c_client *client, u16 *id) { int error; u8 val[3]; @@ -287,7 +287,7 @@ static int elan_i2c_get_product_id(struct i2c_client *client, u8 *id) return error; } - *id = val[0]; + *id = le16_to_cpup((__le16 *)val); return 0; } diff --git a/drivers/input/mouse/elan_i2c_smbus.c b/drivers/input/mouse/elan_i2c_smbus.c index ff36a366b2aa..cb6aecbc1dc2 100644 --- a/drivers/input/mouse/elan_i2c_smbus.c +++ b/drivers/input/mouse/elan_i2c_smbus.c @@ -183,7 +183,7 @@ static int elan_smbus_get_sm_version(struct i2c_client *client, return 0; } -static int elan_smbus_get_product_id(struct i2c_client *client, u8 *id) +static int elan_smbus_get_product_id(struct i2c_client *client, u16 *id) { int error; u8 val[3]; @@ -195,7 +195,7 @@ static int elan_smbus_get_product_id(struct i2c_client *client, u8 *id) return error; } - *id = val[1]; + *id = be16_to_cpup((__be16 *)val); return 0; } From 72d4736253af74147b1fa68145b2f4c61d1f37e1 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Sat, 19 Sep 2015 11:22:57 -0700 Subject: [PATCH 264/647] Input: uinput - fix crash when using ABS events Commit b6d30968d86c45a7bb599eaca13ff048d3fa576c (Input: uinput - switch to using for_each_set_bit()) switched driver to use for_each_set_bit(). However during initial write of the uinput structure that contains min/max data for all possible axes none of them are reflected in dev->absbit yet and so we were skipping over all of them and were not allocating absinfo memory which caused crash later when driver tried to sens EV_ABS events: <1>[ 15.064330] BUG: unable to handle kernel NULL pointer dereference at 0000000000000024 <1>[ 15.064336] IP: [] input_handle_event+0x232/0x4e0 <4>[ 15.064343] PGD 0 <4>[ 15.064345] Oops: 0000 [#1] SMP Fixes: b6d30968d86c45a7bb599eaca13ff048d3fa576c Cc: stable@vger.kernel.org Reported-by: Stephen Chandler Paul Tested-by: Stephen Chandler Paul Signed-off-by: Dmitry Torokhov --- drivers/input/misc/uinput.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/misc/uinput.c b/drivers/input/misc/uinput.c index 345df9b03aed..5adbcedcb81c 100644 --- a/drivers/input/misc/uinput.c +++ b/drivers/input/misc/uinput.c @@ -414,7 +414,7 @@ static int uinput_setup_device(struct uinput_device *udev, dev->id.product = user_dev->id.product; dev->id.version = user_dev->id.version; - for_each_set_bit(i, dev->absbit, ABS_CNT) { + for (i = 0; i < ABS_CNT; i++) { input_abs_set_max(dev, i, user_dev->absmax[i]); input_abs_set_min(dev, i, user_dev->absmin[i]); input_abs_set_fuzz(dev, i, user_dev->absfuzz[i]); From cf261fd1a444e87894c2ed8f481606ead7916fab Mon Sep 17 00:00:00 2001 From: Sylvain Rochet Date: Fri, 18 Sep 2015 16:58:28 +0200 Subject: [PATCH 265/647] usb: gadget: atmel_usba_udc: add ep capabilities support on device tree binding The recently added endpoint capabilities flags verification breaks Atmel USBA because the endpoint configuration was only added when the driver is bound using the legacy pdata interface. Convert endpoint configuration to new capabilities model when driver is bound to a device tree as well. Signed-off-by: Sylvain Rochet Fixes: 47bef3865115 ("usb: gadget: atmel_usba_udc: add ep capabilities support") Signed-off-by: Nicolas Ferre Signed-off-by: Felipe Balbi --- drivers/usb/gadget/udc/atmel_usba_udc.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/usb/gadget/udc/atmel_usba_udc.c b/drivers/usb/gadget/udc/atmel_usba_udc.c index 3dfada8d6061..f0f2b066ac08 100644 --- a/drivers/usb/gadget/udc/atmel_usba_udc.c +++ b/drivers/usb/gadget/udc/atmel_usba_udc.c @@ -2002,6 +2002,17 @@ static struct usba_ep * atmel_udc_of_init(struct platform_device *pdev, ep->udc = udc; INIT_LIST_HEAD(&ep->queue); + if (ep->index == 0) { + ep->ep.caps.type_control = true; + } else { + ep->ep.caps.type_iso = ep->can_isoc; + ep->ep.caps.type_bulk = true; + ep->ep.caps.type_int = true; + } + + ep->ep.caps.dir_in = true; + ep->ep.caps.dir_out = true; + if (i) list_add_tail(&ep->ep.ep_list, &udc->gadget.ep_list); From b8239dcc03afbd0886c1d9b91ba8fee7c6c9a6cb Mon Sep 17 00:00:00 2001 From: Bin Liu Date: Wed, 16 Sep 2015 14:49:28 -0500 Subject: [PATCH 266/647] usb: musb: dsps: fix polling in device-only mode Fix the regression caused by commit ad78c918602 ("usb: musb: dsps: just start polling already") which causes polling the ID pin status even in device-only mode. Fixes: ad78c918602c ("usb: musb: dsps: just start polling already") Signed-off-by: Bin Liu Signed-off-by: Felipe Balbi --- drivers/usb/musb/musb_dsps.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/usb/musb/musb_dsps.c b/drivers/usb/musb/musb_dsps.c index a0cfead6150f..84512d1d5eee 100644 --- a/drivers/usb/musb/musb_dsps.c +++ b/drivers/usb/musb/musb_dsps.c @@ -225,8 +225,11 @@ static void dsps_musb_enable(struct musb *musb) dsps_writel(reg_base, wrp->epintr_set, epmask); dsps_writel(reg_base, wrp->coreintr_set, coremask); - /* start polling for ID change. */ - mod_timer(&glue->timer, jiffies + msecs_to_jiffies(wrp->poll_timeout)); + /* start polling for ID change in dual-role idle mode */ + if (musb->xceiv->otg->state == OTG_STATE_B_IDLE && + musb->port_mode == MUSB_PORT_MODE_DUAL_ROLE) + mod_timer(&glue->timer, jiffies + + msecs_to_jiffies(wrp->poll_timeout)); dsps_musb_try_idle(musb, 0); } From 21c3ee93867694e8c7382ff77b4645b50d3233e9 Mon Sep 17 00:00:00 2001 From: Igor Kotrasinski Date: Tue, 15 Sep 2015 16:55:29 +0200 Subject: [PATCH 267/647] usb: gadget: dummy_hcd: emulate sending zlp in packet logic currently, when a zlp flag is set and an urb/usb_request buffer is filled without a short packet, transfer() leaves its status at -EINPROGRESS and does not rescan for short packet. In a scenario where ep.maxpacket bytes are copied, URB_ZERO_PACKET is set, urb buffer is filled and usb_request buffer is not, transfer() returns with an urb with -EINPROGRESS status, which dummy_hcd treats as incomplete transfer. Check for zlp and rescan appropriately. Signed-off-by: Igor Kotrasinski Signed-off-by: Felipe Balbi --- drivers/usb/gadget/udc/dummy_hcd.c | 25 +++++++++++++++++-------- 1 file changed, 17 insertions(+), 8 deletions(-) diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c index 1379ad40d864..93b38f88e7b2 100644 --- a/drivers/usb/gadget/udc/dummy_hcd.c +++ b/drivers/usb/gadget/udc/dummy_hcd.c @@ -1429,15 +1429,24 @@ top: req->req.status = 0; } - /* many requests terminate without a short packet */ + /* + * many requests terminate without a short packet. + * send a zlp if demanded by flags. + */ } else { - if (req->req.length == req->req.actual - && !req->req.zero) - req->req.status = 0; - if (urb->transfer_buffer_length == urb->actual_length - && !(urb->transfer_flags - & URB_ZERO_PACKET)) - *status = 0; + if (req->req.length == req->req.actual) { + if (req->req.zero && to_host) + rescan = 1; + else + req->req.status = 0; + } + if (urb->transfer_buffer_length == urb->actual_length) { + if (urb->transfer_flags & URB_ZERO_PACKET && + !to_host) + rescan = 1; + else + *status = 0; + } } /* device side completion --> continuable */ From 5dda5be9d501084e8a6242e6dbeb8eea1daf01c8 Mon Sep 17 00:00:00 2001 From: Igor Kotrasinski Date: Tue, 15 Sep 2015 16:55:30 +0200 Subject: [PATCH 268/647] usb: gadget: dummy_hcd: fix unneeded else-if condition We already know at this point that to_host is false. Signed-off-by: Igor Kotrasinski Signed-off-by: Felipe Balbi --- drivers/usb/gadget/udc/dummy_hcd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c index 93b38f88e7b2..69a0b3fa4737 100644 --- a/drivers/usb/gadget/udc/dummy_hcd.c +++ b/drivers/usb/gadget/udc/dummy_hcd.c @@ -1421,7 +1421,7 @@ top: *status = -EOVERFLOW; else *status = 0; - } else if (!to_host) { + } else { *status = 0; if (host_len > dev_len) req->req.status = -EOVERFLOW; From e42bd6a54b97e2a39b5004deac66a0fcd6ebbe75 Mon Sep 17 00:00:00 2001 From: Igor Kotrasinski Date: Tue, 15 Sep 2015 16:55:31 +0200 Subject: [PATCH 269/647] usb: gadget: dummy_hcd: fix rescan logic for transfer transfer() schedules a rescan for transfers larger than maxpacket, which is wrong for transfers that are multiples of maxpacket. Rewrite to fix and clarify packet multiple / remainder transfer logic. Signed-off-by: Igor Kotrasinski Signed-off-by: Felipe Balbi --- drivers/usb/gadget/udc/dummy_hcd.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c index 69a0b3fa4737..ab7e015e2f55 100644 --- a/drivers/usb/gadget/udc/dummy_hcd.c +++ b/drivers/usb/gadget/udc/dummy_hcd.c @@ -1385,12 +1385,15 @@ top: if (len == 0) break; - /* use an extra pass for the final short packet */ - if (len > ep->ep.maxpacket) { - rescan = 1; - len -= (len % ep->ep.maxpacket); + /* send multiple of maxpacket first, then remainder */ + if (len >= ep->ep.maxpacket) { + is_short = 0; + if (len % ep->ep.maxpacket) + rescan = 1; + len -= len % ep->ep.maxpacket; + } else { + is_short = 1; } - is_short = (len % ep->ep.maxpacket) != 0; len = dummy_perform_transfer(urb, req, len); From 9a9ce1dfaef9aa15980cec22b806b39a65a9467e Mon Sep 17 00:00:00 2001 From: Igor Kotrasinski Date: Tue, 15 Sep 2015 16:55:32 +0200 Subject: [PATCH 270/647] usb: gadget: dummy_hcd: in transfer(), return data sent, not limit dummy_timer uses transfer() to update transfer limit. However, limit passed to dummy_timer changes depending on transfer type, so the actual limit is overwritten. This can cause unpredictably slow / fast bulk transfers when coupled with control / interrupt transfers. Fix by returning actual amount of data sent in transfer() and substracting from total. Signed-off-by: Igor Kotrasinski Signed-off-by: Felipe Balbi --- drivers/usb/gadget/udc/dummy_hcd.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c index ab7e015e2f55..27af0f008b57 100644 --- a/drivers/usb/gadget/udc/dummy_hcd.c +++ b/drivers/usb/gadget/udc/dummy_hcd.c @@ -1348,6 +1348,7 @@ static int transfer(struct dummy_hcd *dum_hcd, struct urb *urb, { struct dummy *dum = dum_hcd->dum; struct dummy_request *req; + int sent = 0; top: /* if there's no request queued, the device is NAKing; return */ @@ -1402,6 +1403,7 @@ top: req->req.status = len; } else { limit -= len; + sent += len; urb->actual_length += len; req->req.actual += len; } @@ -1472,7 +1474,7 @@ top: if (rescan) goto top; } - return limit; + return sent; } static int periodic_bytes(struct dummy *dum, struct dummy_ep *ep) @@ -1902,7 +1904,7 @@ restart: default: treat_control_like_bulk: ep->last_io = jiffies; - total = transfer(dum_hcd, urb, ep, limit, &status); + total -= transfer(dum_hcd, urb, ep, limit, &status); break; } From 51b91b7e6c1516c7d3ea70acc91aac9b32ae3e72 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 13 Sep 2015 14:15:09 +0200 Subject: [PATCH 271/647] usb: gadget: drop null test before destroy functions Remove unneeded NULL test. The semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // @@ expression x; @@ -if (x != NULL) \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x); // Signed-off-by: Julia Lawall Signed-off-by: Felipe Balbi --- drivers/usb/gadget/udc/bdc/bdc_core.c | 3 +-- drivers/usb/gadget/udc/gr_udc.c | 3 +-- drivers/usb/gadget/udc/mv_u3d_core.c | 3 +-- drivers/usb/gadget/udc/mv_udc_core.c | 3 +-- 4 files changed, 4 insertions(+), 8 deletions(-) diff --git a/drivers/usb/gadget/udc/bdc/bdc_core.c b/drivers/usb/gadget/udc/bdc/bdc_core.c index 5c8f4effb62a..ccb9c213cc9f 100644 --- a/drivers/usb/gadget/udc/bdc/bdc_core.c +++ b/drivers/usb/gadget/udc/bdc/bdc_core.c @@ -324,8 +324,7 @@ static void bdc_mem_free(struct bdc *bdc) bdc->scratchpad.buff, bdc->scratchpad.sp_dma); /* Destroy the dma pools */ - if (bdc->bd_table_pool) - dma_pool_destroy(bdc->bd_table_pool); + dma_pool_destroy(bdc->bd_table_pool); /* Free the bdc_ep array */ kfree(bdc->bdc_ep_array); diff --git a/drivers/usb/gadget/udc/gr_udc.c b/drivers/usb/gadget/udc/gr_udc.c index 8aa2593c2c36..b9429bc42511 100644 --- a/drivers/usb/gadget/udc/gr_udc.c +++ b/drivers/usb/gadget/udc/gr_udc.c @@ -2117,8 +2117,7 @@ static int gr_remove(struct platform_device *pdev) return -EBUSY; gr_dfs_delete(dev); - if (dev->desc_pool) - dma_pool_destroy(dev->desc_pool); + dma_pool_destroy(dev->desc_pool); platform_set_drvdata(pdev, NULL); gr_free_request(&dev->epi[0].ep, &dev->ep0reqi->req); diff --git a/drivers/usb/gadget/udc/mv_u3d_core.c b/drivers/usb/gadget/udc/mv_u3d_core.c index 4c489692745e..dafe74eb9ade 100644 --- a/drivers/usb/gadget/udc/mv_u3d_core.c +++ b/drivers/usb/gadget/udc/mv_u3d_core.c @@ -1767,8 +1767,7 @@ static int mv_u3d_remove(struct platform_device *dev) usb_del_gadget_udc(&u3d->gadget); /* free memory allocated in probe */ - if (u3d->trb_pool) - dma_pool_destroy(u3d->trb_pool); + dma_pool_destroy(u3d->trb_pool); if (u3d->ep_context) dma_free_coherent(&dev->dev, u3d->ep_context_size, diff --git a/drivers/usb/gadget/udc/mv_udc_core.c b/drivers/usb/gadget/udc/mv_udc_core.c index 339af51df57d..81b6229c7805 100644 --- a/drivers/usb/gadget/udc/mv_udc_core.c +++ b/drivers/usb/gadget/udc/mv_udc_core.c @@ -2100,8 +2100,7 @@ static int mv_udc_remove(struct platform_device *pdev) } /* free memory allocated in probe */ - if (udc->dtd_pool) - dma_pool_destroy(udc->dtd_pool); + dma_pool_destroy(udc->dtd_pool); if (udc->ep_dqh) dma_free_coherent(&pdev->dev, udc->ep_dqh_size, From 8fb7ab504381b8ce4f443129e102a713bd76dfe2 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Sat, 12 Sep 2015 10:54:26 +0200 Subject: [PATCH 272/647] usb: phy: isp1301: Export I2C module alias information The I2C core always reports the MODALIAS uevent as "i2c: Signed-off-by: Felipe Balbi --- drivers/usb/phy/phy-isp1301.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/phy/phy-isp1301.c b/drivers/usb/phy/phy-isp1301.c index 8a55b37d1a02..db68156568e6 100644 --- a/drivers/usb/phy/phy-isp1301.c +++ b/drivers/usb/phy/phy-isp1301.c @@ -31,6 +31,7 @@ static const struct i2c_device_id isp1301_id[] = { { "isp1301", 0 }, { } }; +MODULE_DEVICE_TABLE(i2c, isp1301_id); static struct i2c_client *isp1301_i2c_client; From b431ba8803666e56c1d178a421b3cbc36e8d3d33 Mon Sep 17 00:00:00 2001 From: Bin Liu Date: Mon, 24 Aug 2015 15:28:37 -0500 Subject: [PATCH 273/647] usb: musb: fix cppi channel teardown for isoch transfer After a few iterations of start/stop UVC camera streaming, the streaming stops. This patch adds 250us delay in the cppi channel abort path to let cppi drain properly. Using 50us delay seems to be too aggressive, some webcams are still broken. 250us is the original value used in TI 3.2 kernel. Signed-off-by: Bin Liu Signed-off-by: Felipe Balbi --- drivers/usb/musb/musb_cppi41.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/musb/musb_cppi41.c b/drivers/usb/musb/musb_cppi41.c index d07cafb7d5f5..e499b862a946 100644 --- a/drivers/usb/musb/musb_cppi41.c +++ b/drivers/usb/musb/musb_cppi41.c @@ -551,6 +551,9 @@ static int cppi41_dma_channel_abort(struct dma_channel *channel) } else { cppi41_set_autoreq_mode(cppi41_channel, EP_MODE_AUTOREQ_NONE); + /* delay to drain to cppi dma pipeline for isoch */ + udelay(250); + csr = musb_readw(epio, MUSB_RXCSR); csr &= ~(MUSB_RXCSR_H_REQPKT | MUSB_RXCSR_DMAENAB); musb_writew(epio, MUSB_RXCSR, csr); From a66c275b3d5d8467d770dabd30927f5d5e857294 Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Mon, 21 Sep 2015 11:08:36 +0300 Subject: [PATCH 274/647] usb: dwc3: gadget: Fix BUG in RT config Using spin_lock() in hard irq handler is pointless and causes a BUG() in RT (real-time) configuration so get rid of it. The reason it's pointless is because the driver is basically accessing register which is, anyways, atomic. Signed-off-by: Roger Quadros Signed-off-by: Felipe Balbi --- drivers/usb/dwc3/gadget.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index 0c25704dcb6b..1e8bdf817811 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -2665,8 +2665,6 @@ static irqreturn_t dwc3_interrupt(int irq, void *_dwc) int i; irqreturn_t ret = IRQ_NONE; - spin_lock(&dwc->lock); - for (i = 0; i < dwc->num_event_buffers; i++) { irqreturn_t status; @@ -2675,8 +2673,6 @@ static irqreturn_t dwc3_interrupt(int irq, void *_dwc) ret = status; } - spin_unlock(&dwc->lock); - return ret; } From 50745b0a7f46f68574cd2b9ae24566bf026e7ebd Mon Sep 17 00:00:00 2001 From: chandan Date: Fri, 28 Aug 2015 21:10:13 +0530 Subject: [PATCH 275/647] Btrfs: Direct I/O: Fix space accounting The following call trace is seen when generic/095 test is executed, WARNING: CPU: 3 PID: 2769 at /home/chandan/code/repos/linux/fs/btrfs/inode.c:8967 btrfs_destroy_inode+0x284/0x2a0() Modules linked in: CPU: 3 PID: 2769 Comm: umount Not tainted 4.2.0-rc5+ #31 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20150306_163512-brownie 04/01/2014 ffffffff81c08150 ffff8802ec9cbce8 ffffffff81984058 ffff8802ffd8feb0 0000000000000000 ffff8802ec9cbd28 ffffffff81050385 ffff8802ec9cbd38 ffff8802d12f8588 ffff8802d12f8588 ffff8802f15ab000 ffff8800bb96c0b0 Call Trace: [] dump_stack+0x45/0x57 [] warn_slowpath_common+0x85/0xc0 [] warn_slowpath_null+0x15/0x20 [] btrfs_destroy_inode+0x284/0x2a0 [] destroy_inode+0x37/0x60 [] evict+0x109/0x170 [] dispose_list+0x35/0x50 [] evict_inodes+0xaa/0x100 [] generic_shutdown_super+0x47/0xf0 [] kill_anon_super+0x11/0x20 [] btrfs_kill_super+0x13/0x110 [] deactivate_locked_super+0x39/0x70 [] deactivate_super+0x5f/0x70 [] cleanup_mnt+0x3e/0x90 [] __cleanup_mnt+0xd/0x10 [] task_work_run+0x96/0xb0 [] do_notify_resume+0x3d/0x50 [] int_signal+0x12/0x17 This means that the inode had non-zero "outstanding extents" during eviction. This occurs because, during direct I/O a task which successfully used up its reserved data space would set BTRFS_INODE_DIO_READY bit and does not clear the bit after finishing the DIO write. A future DIO write could actually fail and the unused reserve space won't be freed because of the previously set BTRFS_INODE_DIO_READY bit. Clearing the BTRFS_INODE_DIO_READY bit in btrfs_direct_IO() caused the following issue, |-----------------------------------+-------------------------------------| | Task A | Task B | |-----------------------------------+-------------------------------------| | Start direct i/o write on inode X.| | | reserve space | | | Allocate ordered extent | | | release reserved space | | | Set BTRFS_INODE_DIO_READY bit. | | | | splice() | | | Transfer data from pipe buffer to | | | destination file. | | | - kmap(pipe buffer page) | | | - Start direct i/o write on | | | inode X. | | | - reserve space | | | - dio_refill_pages() | | | - sdio->blocks_available == 0 | | | - Since a kernel address is | | | being passed instead of a | | | user space address, | | | iov_iter_get_pages() returns | | | -EFAULT. | | | - Since BTRFS_INODE_DIO_READY is | | | set, we don't release reserved | | | space. | | | - Clear BTRFS_INODE_DIO_READY bit.| | -EIOCBQUEUED is returned. | | |-----------------------------------+-------------------------------------| Hence this commit introduces "struct btrfs_dio_data" to track the usage of reserved data space. The remaining unused "reserve space" can now be freed reliably. Signed-off-by: Chandan Rajendra Reviewed-by: Liu Bo Signed-off-by: Chris Mason --- fs/btrfs/btrfs_inode.h | 2 -- fs/btrfs/inode.c | 42 +++++++++++++++++++++--------------------- 2 files changed, 21 insertions(+), 23 deletions(-) diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 81220b2203c6..0ef5cc13fae2 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -44,8 +44,6 @@ #define BTRFS_INODE_IN_DELALLOC_LIST 9 #define BTRFS_INODE_READDIO_NEED_LOCK 10 #define BTRFS_INODE_HAS_PROPS 11 -/* DIO is ready to submit */ -#define BTRFS_INODE_DIO_READY 12 /* * The following 3 bits are meant only for the btree inode. * When any of them is set, it means an error happened while writing an diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 3d983dea57af..b7e439bf5e4f 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7405,6 +7405,10 @@ static struct extent_map *create_pinned_em(struct inode *inode, u64 start, return em; } +struct btrfs_dio_data { + u64 outstanding_extents; + u64 reserve; +}; static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) @@ -7412,10 +7416,10 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, struct extent_map *em; struct btrfs_root *root = BTRFS_I(inode)->root; struct extent_state *cached_state = NULL; + struct btrfs_dio_data *dio_data = NULL; u64 start = iblock << inode->i_blkbits; u64 lockstart, lockend; u64 len = bh_result->b_size; - u64 *outstanding_extents = NULL; int unlock_bits = EXTENT_LOCKED; int ret = 0; @@ -7433,7 +7437,7 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, * that anything that needs to check if there's a transction doesn't get * confused. */ - outstanding_extents = current->journal_info; + dio_data = current->journal_info; current->journal_info = NULL; } @@ -7565,17 +7569,18 @@ unlock: * within our reservation, otherwise we need to adjust our inode * counter appropriately. */ - if (*outstanding_extents) { - (*outstanding_extents)--; + if (dio_data->outstanding_extents) { + (dio_data->outstanding_extents)--; } else { spin_lock(&BTRFS_I(inode)->lock); BTRFS_I(inode)->outstanding_extents++; spin_unlock(&BTRFS_I(inode)->lock); } - current->journal_info = outstanding_extents; btrfs_free_reserved_data_space(inode, len); - set_bit(BTRFS_INODE_DIO_READY, &BTRFS_I(inode)->runtime_flags); + WARN_ON(dio_data->reserve < len); + dio_data->reserve -= len; + current->journal_info = dio_data; } /* @@ -7598,8 +7603,8 @@ unlock: unlock_err: clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, lockend, unlock_bits, 1, 0, &cached_state, GFP_NOFS); - if (outstanding_extents) - current->journal_info = outstanding_extents; + if (dio_data) + current->journal_info = dio_data; return ret; } @@ -8329,7 +8334,8 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; - u64 outstanding_extents = 0; + struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_dio_data dio_data = { 0 }; size_t count = 0; int flags = 0; bool wakeup = true; @@ -8367,7 +8373,7 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, ret = btrfs_delalloc_reserve_space(inode, count); if (ret) goto out; - outstanding_extents = div64_u64(count + + dio_data.outstanding_extents = div64_u64(count + BTRFS_MAX_EXTENT_SIZE - 1, BTRFS_MAX_EXTENT_SIZE); @@ -8376,7 +8382,8 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, * do the accounting properly if we go over the number we * originally calculated. Abuse current->journal_info for this. */ - current->journal_info = &outstanding_extents; + dio_data.reserve = round_up(count, root->sectorsize); + current->journal_info = &dio_data; } else if (test_bit(BTRFS_INODE_READDIO_NEED_LOCK, &BTRFS_I(inode)->runtime_flags)) { inode_dio_end(inode); @@ -8391,16 +8398,9 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, if (iov_iter_rw(iter) == WRITE) { current->journal_info = NULL; if (ret < 0 && ret != -EIOCBQUEUED) { - /* - * If the error comes from submitting stage, - * btrfs_get_blocsk_direct() has free'd data space, - * and metadata space will be handled by - * finish_ordered_fn, don't do that again to make - * sure bytes_may_use is correct. - */ - if (!test_and_clear_bit(BTRFS_INODE_DIO_READY, - &BTRFS_I(inode)->runtime_flags)) - btrfs_delalloc_release_space(inode, count); + if (dio_data.reserve) + btrfs_delalloc_release_space(inode, + dio_data.reserve); } else if (ret >= 0 && (size_t)ret < count) btrfs_delalloc_release_space(inode, count - (size_t)ret); From c7d778fa74a88e7b5a09ba498d66bd67a7dc7df0 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 17:54:00 +0200 Subject: [PATCH 276/647] net: arc: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: David S. Miller --- drivers/net/ethernet/arc/emac_arc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/arc/emac_arc.c b/drivers/net/ethernet/arc/emac_arc.c index f9cb99bfb511..ffd180570920 100644 --- a/drivers/net/ethernet/arc/emac_arc.c +++ b/drivers/net/ethernet/arc/emac_arc.c @@ -78,6 +78,7 @@ static const struct of_device_id emac_arc_dt_ids[] = { { .compatible = "snps,arc-emac" }, { /* Sentinel */ } }; +MODULE_DEVICE_TABLE(of, emac_arc_dt_ids); static struct platform_driver emac_arc_driver = { .probe = emac_arc_probe, From 46d5a3431f2dbb2392b501be9b9f9c2ca17737d8 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 17:54:30 +0200 Subject: [PATCH 277/647] net: systemport: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Acked-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bcmsysport.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index b9a5a97ed4dd..f1b5364f3521 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -2079,6 +2079,7 @@ static const struct of_device_id bcm_sysport_of_match[] = { { .compatible = "brcm,systemport" }, { /* sentinel */ } }; +MODULE_DEVICE_TABLE(of, bcm_sysport_of_match); static struct platform_driver bcm_sysport_driver = { .probe = bcm_sysport_probe, From e8048e5595bedb1f4b20597d059d6877a710f12d Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 17:55:02 +0200 Subject: [PATCH 278/647] net: bcmgenet: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Acked-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index fadbd0088d3e..3bc701e4c59e 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -3155,6 +3155,7 @@ static const struct of_device_id bcmgenet_match[] = { { .compatible = "brcm,genet-v4", .data = (void *)GENET_V4 }, { }, }; +MODULE_DEVICE_TABLE(of, bcmgenet_match); static int bcmgenet_probe(struct platform_device *pdev) { From 23860063706f2037008fb423820cdbd0dcd558bb Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 17:55:27 +0200 Subject: [PATCH 279/647] net: gianfar_ptp: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Acked-by: Richard Cochran Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/gianfar_ptp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/freescale/gianfar_ptp.c b/drivers/net/ethernet/freescale/gianfar_ptp.c index 8e3cd77aa347..664d0c261269 100644 --- a/drivers/net/ethernet/freescale/gianfar_ptp.c +++ b/drivers/net/ethernet/freescale/gianfar_ptp.c @@ -557,6 +557,7 @@ static const struct of_device_id match_table[] = { { .compatible = "fsl,etsec-ptp" }, {}, }; +MODULE_DEVICE_TABLE(of, match_table); static struct platform_driver gianfar_ptp_driver = { .driver = { From ebd8ebf078879973d0e8d2642253c091c23bd7d2 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 17:56:21 +0200 Subject: [PATCH 280/647] net: moxa: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: David S. Miller --- drivers/net/ethernet/moxa/moxart_ether.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/moxa/moxart_ether.c b/drivers/net/ethernet/moxa/moxart_ether.c index becbb5f1f5a7..a10c928bbd6b 100644 --- a/drivers/net/ethernet/moxa/moxart_ether.c +++ b/drivers/net/ethernet/moxa/moxart_ether.c @@ -552,6 +552,7 @@ static const struct of_device_id moxart_mac_match[] = { { .compatible = "moxa,moxart-mac" }, { } }; +MODULE_DEVICE_TABLE(of, moxart_mac_match); static struct platform_driver moxart_mac_driver = { .probe = moxart_mac_probe, From 2f90a3070690ad80c38d650e91b96b5dcbdfe2fd Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 18:16:12 +0200 Subject: [PATCH 281/647] net: phy: mdio-bcm-unimac: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Acked-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/mdio-bcm-unimac.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/phy/mdio-bcm-unimac.c b/drivers/net/phy/mdio-bcm-unimac.c index 6a52a7f0fa0d..4bde5e728fe0 100644 --- a/drivers/net/phy/mdio-bcm-unimac.c +++ b/drivers/net/phy/mdio-bcm-unimac.c @@ -244,6 +244,7 @@ static const struct of_device_id unimac_mdio_ids[] = { { .compatible = "brcm,unimac-mdio", }, { /* sentinel */ }, }; +MODULE_DEVICE_TABLE(of, unimac_mdio_ids); static struct platform_driver unimac_mdio_driver = { .driver = { From 1ccb141e31d0dc4c1234b0886cdc8b4399112c59 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Fri, 18 Sep 2015 18:16:29 +0200 Subject: [PATCH 282/647] net: phy: mdio-gpio: Fix module autoload for OF platform driver This platform driver has a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Signed-off-by: David S. Miller --- drivers/net/phy/mdio-gpio.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/phy/mdio-gpio.c b/drivers/net/phy/mdio-gpio.c index 7dc21e56a7aa..3bc9f03349f3 100644 --- a/drivers/net/phy/mdio-gpio.c +++ b/drivers/net/phy/mdio-gpio.c @@ -261,6 +261,7 @@ static const struct of_device_id mdio_gpio_of_match[] = { { .compatible = "virtual,mdio-gpio", }, { /* sentinel */ } }; +MODULE_DEVICE_TABLE(of, mdio_gpio_of_match); static struct platform_driver mdio_gpio_driver = { .probe = mdio_gpio_probe, From 4a476bd6d1d923922ec950ddc4c27b279f6901eb Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Sun, 20 Sep 2015 02:25:38 -0700 Subject: [PATCH 283/647] usbnet: New driver for QinHeng CH9200 devices There's a bunch of cheap USB 10/100 devices based on QinHeng chipsets. The vendor driver supports the CH9100 and CH9200 devices, but the majority of the code is of the if (ch9100) {} else {} form, with the most significant difference being that CH9200 provides a real MII interface but CH9100 fakes one with a bunch of global variables and magic commands. I don't have a CH9100, so it's probably better if someone who does provides an independent driver for it. In any case, this is a lightly cleaned up version of the vendor driver with all the CH9100 code dropped. Signed-off-by: Matthew Garrett Signed-off-by: David S. Miller --- drivers/net/usb/Kconfig | 11 + drivers/net/usb/Makefile | 2 +- drivers/net/usb/ch9200.c | 443 +++++++++++++++++++++++++++++++++++++++ 3 files changed, 455 insertions(+), 1 deletion(-) create mode 100644 drivers/net/usb/ch9200.c diff --git a/drivers/net/usb/Kconfig b/drivers/net/usb/Kconfig index 1610b79ae386..fbb9325d1f6e 100644 --- a/drivers/net/usb/Kconfig +++ b/drivers/net/usb/Kconfig @@ -583,4 +583,15 @@ config USB_VL600 http://ubuntuforums.org/showpost.php?p=10589647&postcount=17 +config USB_NET_CH9200 + tristate "QingHeng CH9200 USB ethernet support" + depends on USB_USBNET + select MII + help + Choose this option if you have a USB ethernet adapter with a QinHeng + CH9200 chipset. + + To compile this driver as a module, choose M here: the + module will be called ch9200. + endif # USB_NET_DRIVERS diff --git a/drivers/net/usb/Makefile b/drivers/net/usb/Makefile index cf6a0e610a7f..b5f04068dbe4 100644 --- a/drivers/net/usb/Makefile +++ b/drivers/net/usb/Makefile @@ -38,4 +38,4 @@ obj-$(CONFIG_USB_NET_HUAWEI_CDC_NCM) += huawei_cdc_ncm.o obj-$(CONFIG_USB_VL600) += lg-vl600.o obj-$(CONFIG_USB_NET_QMI_WWAN) += qmi_wwan.o obj-$(CONFIG_USB_NET_CDC_MBIM) += cdc_mbim.o - +obj-$(CONFIG_USB_NET_CH9200) += ch9200.o diff --git a/drivers/net/usb/ch9200.c b/drivers/net/usb/ch9200.c new file mode 100644 index 000000000000..cabb670eb55c --- /dev/null +++ b/drivers/net/usb/ch9200.c @@ -0,0 +1,443 @@ +/* + * USB 10M/100M ethernet adapter + * + * This file is licensed under the terms of the GNU General Public License + * version 2. This program is licensed "as is" without any warranty of any + * kind, whether express or implied + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define CH9200_VID 0x1A86 +#define CH9200_PID_E092 0xE092 + +#define CTRL_TIMEOUT_MS 1000 + +#define CONTROL_TIMEOUT_MS 1000 + +#define REQUEST_READ 0x0E +#define REQUEST_WRITE 0x0F + +/* Address space: + * 00-63 : MII + * 64-128: MAC + * + * Note: all accesses must be 16-bit + */ + +#define MAC_REG_CTRL 64 +#define MAC_REG_STATUS 66 +#define MAC_REG_INTERRUPT_MASK 68 +#define MAC_REG_PHY_COMMAND 70 +#define MAC_REG_PHY_DATA 72 +#define MAC_REG_STATION_L 74 +#define MAC_REG_STATION_M 76 +#define MAC_REG_STATION_H 78 +#define MAC_REG_HASH_L 80 +#define MAC_REG_HASH_M1 82 +#define MAC_REG_HASH_M2 84 +#define MAC_REG_HASH_H 86 +#define MAC_REG_THRESHOLD 88 +#define MAC_REG_FIFO_DEPTH 90 +#define MAC_REG_PAUSE 92 +#define MAC_REG_FLOW_CONTROL 94 + +/* Control register bits + * + * Note: bits 13 and 15 are reserved + */ +#define LOOPBACK (0x01 << 14) +#define BASE100X (0x01 << 12) +#define MBPS_10 (0x01 << 11) +#define DUPLEX_MODE (0x01 << 10) +#define PAUSE_FRAME (0x01 << 9) +#define PROMISCUOUS (0x01 << 8) +#define MULTICAST (0x01 << 7) +#define BROADCAST (0x01 << 6) +#define HASH (0x01 << 5) +#define APPEND_PAD (0x01 << 4) +#define APPEND_CRC (0x01 << 3) +#define TRANSMITTER_ACTION (0x01 << 2) +#define RECEIVER_ACTION (0x01 << 1) +#define DMA_ACTION (0x01 << 0) + +/* Status register bits + * + * Note: bits 7-15 are reserved + */ +#define ALIGNMENT (0x01 << 6) +#define FIFO_OVER_RUN (0x01 << 5) +#define FIFO_UNDER_RUN (0x01 << 4) +#define RX_ERROR (0x01 << 3) +#define RX_COMPLETE (0x01 << 2) +#define TX_ERROR (0x01 << 1) +#define TX_COMPLETE (0x01 << 0) + +/* FIFO depth register bits + * + * Note: bits 6 and 14 are reserved + */ + +#define ETH_TXBD (0x01 << 15) +#define ETN_TX_FIFO_DEPTH (0x01 << 8) +#define ETH_RXBD (0x01 << 7) +#define ETH_RX_FIFO_DEPTH (0x01 << 0) + +static int control_read(struct usbnet *dev, + unsigned char request, unsigned short value, + unsigned short index, void *data, unsigned short size, + int timeout) +{ + unsigned char *buf = NULL; + unsigned char request_type; + int err = 0; + + if (request == REQUEST_READ) + request_type = (USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_OTHER); + else + request_type = (USB_DIR_IN | USB_TYPE_VENDOR | + USB_RECIP_DEVICE); + + netdev_dbg(dev->net, "Control_read() index=0x%02x size=%d\n", + index, size); + + buf = kmalloc(size, GFP_KERNEL); + if (!buf) { + err = -ENOMEM; + goto err_out; + } + + err = usb_control_msg(dev->udev, + usb_rcvctrlpipe(dev->udev, 0), + request, request_type, value, index, buf, size, + timeout); + if (err == size) + memcpy(data, buf, size); + else if (err >= 0) + err = -EINVAL; + kfree(buf); + + return err; + +err_out: + return err; +} + +static int control_write(struct usbnet *dev, unsigned char request, + unsigned short value, unsigned short index, + void *data, unsigned short size, int timeout) +{ + unsigned char *buf = NULL; + unsigned char request_type; + int err = 0; + + if (request == REQUEST_WRITE) + request_type = (USB_DIR_OUT | USB_TYPE_VENDOR | + USB_RECIP_OTHER); + else + request_type = (USB_DIR_OUT | USB_TYPE_VENDOR | + USB_RECIP_DEVICE); + + netdev_dbg(dev->net, "Control_write() index=0x%02x size=%d\n", + index, size); + + if (data) { + buf = kmalloc(size, GFP_KERNEL); + if (!buf) { + err = -ENOMEM; + goto err_out; + } + memcpy(buf, data, size); + } + + err = usb_control_msg(dev->udev, + usb_sndctrlpipe(dev->udev, 0), + request, request_type, value, index, buf, size, + timeout); + if (err >= 0 && err < size) + err = -EINVAL; + kfree(buf); + + return 0; + +err_out: + return err; +} + +static int ch9200_mdio_read(struct net_device *netdev, int phy_id, int loc) +{ + struct usbnet *dev = netdev_priv(netdev); + unsigned char buff[2]; + + netdev_dbg(netdev, "ch9200_mdio_read phy_id:%02x loc:%02x\n", + phy_id, loc); + + if (phy_id != 0) + return -ENODEV; + + control_read(dev, REQUEST_READ, 0, loc * 2, buff, 0x02, + CONTROL_TIMEOUT_MS); + + return (buff[0] | buff[1] << 8); +} + +static void ch9200_mdio_write(struct net_device *netdev, + int phy_id, int loc, int val) +{ + struct usbnet *dev = netdev_priv(netdev); + unsigned char buff[2]; + + netdev_dbg(netdev, "ch9200_mdio_write() phy_id=%02x loc:%02x\n", + phy_id, loc); + + if (phy_id != 0) + return; + + buff[0] = (unsigned char)val; + buff[1] = (unsigned char)(val >> 8); + + control_write(dev, REQUEST_WRITE, 0, loc * 2, buff, 0x02, + CONTROL_TIMEOUT_MS); +} + +static int ch9200_link_reset(struct usbnet *dev) +{ + struct ethtool_cmd ecmd; + + mii_check_media(&dev->mii, 1, 1); + mii_ethtool_gset(&dev->mii, &ecmd); + + netdev_dbg(dev->net, "link_reset() speed:%d duplex:%d\n", + ecmd.speed, ecmd.duplex); + + return 0; +} + +static void ch9200_status(struct usbnet *dev, struct urb *urb) +{ + int link; + unsigned char *buf; + + if (urb->actual_length < 16) + return; + + buf = urb->transfer_buffer; + link = !!(buf[0] & 0x01); + + if (link) { + netif_carrier_on(dev->net); + usbnet_defer_kevent(dev, EVENT_LINK_RESET); + } else { + netif_carrier_off(dev->net); + } +} + +static struct sk_buff *ch9200_tx_fixup(struct usbnet *dev, struct sk_buff *skb, + gfp_t flags) +{ + int i = 0; + int len = 0; + int tx_overhead = 0; + + tx_overhead = 0x40; + + len = skb->len; + if (skb_headroom(skb) < tx_overhead) { + struct sk_buff *skb2; + + skb2 = skb_copy_expand(skb, tx_overhead, 0, flags); + dev_kfree_skb_any(skb); + skb = skb2; + if (!skb) + return NULL; + } + + __skb_push(skb, tx_overhead); + /* usbnet adds padding if length is a multiple of packet size + * if so, adjust length value in header + */ + if ((skb->len % dev->maxpacket) == 0) + len++; + + skb->data[0] = len; + skb->data[1] = len >> 8; + skb->data[2] = 0x00; + skb->data[3] = 0x80; + + for (i = 4; i < 48; i++) + skb->data[i] = 0x00; + + skb->data[48] = len; + skb->data[49] = len >> 8; + skb->data[50] = 0x00; + skb->data[51] = 0x80; + + for (i = 52; i < 64; i++) + skb->data[i] = 0x00; + + return skb; +} + +static int ch9200_rx_fixup(struct usbnet *dev, struct sk_buff *skb) +{ + int len = 0; + int rx_overhead = 0; + + rx_overhead = 64; + + if (unlikely(skb->len < rx_overhead)) { + dev_err(&dev->udev->dev, "unexpected tiny rx frame\n"); + return 0; + } + + len = (skb->data[skb->len - 16] | skb->data[skb->len - 15] << 8); + skb_trim(skb, len); + + return 1; +} + +static int get_mac_address(struct usbnet *dev, unsigned char *data) +{ + int err = 0; + unsigned char mac_addr[0x06]; + int rd_mac_len = 0; + + netdev_dbg(dev->net, "get_mac_address:\n\tusbnet VID:%0x PID:%0x\n", + dev->udev->descriptor.idVendor, + dev->udev->descriptor.idProduct); + + memset(mac_addr, 0, sizeof(mac_addr)); + rd_mac_len = control_read(dev, REQUEST_READ, 0, + MAC_REG_STATION_L, mac_addr, 0x02, + CONTROL_TIMEOUT_MS); + rd_mac_len += control_read(dev, REQUEST_READ, 0, MAC_REG_STATION_M, + mac_addr + 2, 0x02, CONTROL_TIMEOUT_MS); + rd_mac_len += control_read(dev, REQUEST_READ, 0, MAC_REG_STATION_H, + mac_addr + 4, 0x02, CONTROL_TIMEOUT_MS); + if (rd_mac_len != ETH_ALEN) + err = -EINVAL; + + data[0] = mac_addr[5]; + data[1] = mac_addr[4]; + data[2] = mac_addr[3]; + data[3] = mac_addr[2]; + data[4] = mac_addr[1]; + data[5] = mac_addr[0]; + + return err; +} + +static int ch9200_bind(struct usbnet *dev, struct usb_interface *intf) +{ + int retval = 0; + unsigned char data[2]; + + retval = usbnet_get_endpoints(dev, intf); + if (retval) + return retval; + + dev->mii.dev = dev->net; + dev->mii.mdio_read = ch9200_mdio_read; + dev->mii.mdio_write = ch9200_mdio_write; + dev->mii.reg_num_mask = 0x1f; + + dev->mii.phy_id_mask = 0x1f; + + dev->hard_mtu = dev->net->mtu + dev->net->hard_header_len; + dev->rx_urb_size = 24 * 64 + 16; + mii_nway_restart(&dev->mii); + + data[0] = 0x01; + data[1] = 0x0F; + retval = control_write(dev, REQUEST_WRITE, 0, MAC_REG_THRESHOLD, data, + 0x02, CONTROL_TIMEOUT_MS); + + data[0] = 0xA0; + data[1] = 0x90; + retval = control_write(dev, REQUEST_WRITE, 0, MAC_REG_FIFO_DEPTH, data, + 0x02, CONTROL_TIMEOUT_MS); + + data[0] = 0x30; + data[1] = 0x00; + retval = control_write(dev, REQUEST_WRITE, 0, MAC_REG_PAUSE, data, + 0x02, CONTROL_TIMEOUT_MS); + + data[0] = 0x17; + data[1] = 0xD8; + retval = control_write(dev, REQUEST_WRITE, 0, MAC_REG_FLOW_CONTROL, + data, 0x02, CONTROL_TIMEOUT_MS); + + /* Undocumented register */ + data[0] = 0x01; + data[1] = 0x00; + retval = control_write(dev, REQUEST_WRITE, 0, 254, data, 0x02, + CONTROL_TIMEOUT_MS); + + data[0] = 0x5F; + data[1] = 0x0D; + retval = control_write(dev, REQUEST_WRITE, 0, MAC_REG_CTRL, data, 0x02, + CONTROL_TIMEOUT_MS); + + retval = get_mac_address(dev, dev->net->dev_addr); + + return retval; +} + +static const struct driver_info ch9200_info = { + .description = "CH9200 USB to Network Adaptor", + .flags = FLAG_ETHER, + .bind = ch9200_bind, + .rx_fixup = ch9200_rx_fixup, + .tx_fixup = ch9200_tx_fixup, + .status = ch9200_status, + .link_reset = ch9200_link_reset, + .reset = ch9200_link_reset, +}; + +static const struct usb_device_id ch9200_products[] = { + { + USB_DEVICE(0x1A86, 0xE092), + .driver_info = (unsigned long)&ch9200_info, + }, + {}, +}; + +MODULE_DEVICE_TABLE(usb, ch9200_products); + +static struct usb_driver ch9200_driver = { + .name = "ch9200", + .id_table = ch9200_products, + .probe = usbnet_probe, + .disconnect = usbnet_disconnect, + .suspend = usbnet_suspend, + .resume = usbnet_resume, +}; + +static int __init ch9200_init(void) +{ + return usb_register(&ch9200_driver); +} + +static void __exit ch9200_exit(void) +{ + usb_deregister(&ch9200_driver); +} + +module_init(ch9200_init); +module_exit(ch9200_exit); + +MODULE_DESCRIPTION("QinHeng CH9200 USB Network device"); +MODULE_LICENSE("GPL"); From 5eb8f289ac3020a9abad1c3c532ceca83284b6ed Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Fri, 18 Sep 2015 16:20:32 -0400 Subject: [PATCH 284/647] geneve: remove vlan-related feature assignment The code handling vlan tag insertion was dropped in commit 371bd1061d29 ("geneve: Consolidate Geneve functionality in single module."). Now we need to drop the related vlan feature bits in the netdev structure. Signed-off-by: John W. Linville Reviewed-by: Jesse Gross Signed-off-by: David S. Miller --- drivers/net/geneve.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index da3259ce7c8d..8f1e9150341a 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -748,12 +748,8 @@ static void geneve_setup(struct net_device *dev) dev->features |= NETIF_F_RXCSUM; dev->features |= NETIF_F_GSO_SOFTWARE; - dev->vlan_features = dev->features; - dev->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX; - dev->hw_features |= NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_RXCSUM; dev->hw_features |= NETIF_F_GSO_SOFTWARE; - dev->hw_features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX; netif_keep_dst(dev); dev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_NO_QUEUE; From 4c5d283acc997a1bd7bc37cddcf7d284521cffff Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Fri, 18 Sep 2015 17:47:55 -0400 Subject: [PATCH 285/647] sunvnet: Invoke SET_NETDEV_DEV() to set up the vdev in vnet_new() `ls /sys/devices/channel-devices/vnet-port-0-0/net' is missing without this change, and applications like NetworkManager are looking in sysfs for the information. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller --- drivers/net/ethernet/sun/sunvnet.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/sun/sunvnet.c b/drivers/net/ethernet/sun/sunvnet.c index 53fe200e0b79..cc106d892e29 100644 --- a/drivers/net/ethernet/sun/sunvnet.c +++ b/drivers/net/ethernet/sun/sunvnet.c @@ -1756,7 +1756,8 @@ static const struct net_device_ops vnet_ops = { #endif }; -static struct vnet *vnet_new(const u64 *local_mac) +static struct vnet *vnet_new(const u64 *local_mac, + struct vio_dev *vdev) { struct net_device *dev; struct vnet *vp; @@ -1790,6 +1791,8 @@ static struct vnet *vnet_new(const u64 *local_mac) NETIF_F_HW_CSUM | NETIF_F_SG; dev->features = dev->hw_features; + SET_NETDEV_DEV(dev, &vdev->dev); + err = register_netdev(dev); if (err) { pr_err("Cannot register net device, aborting\n"); @@ -1808,7 +1811,8 @@ err_out_free_dev: return ERR_PTR(err); } -static struct vnet *vnet_find_or_create(const u64 *local_mac) +static struct vnet *vnet_find_or_create(const u64 *local_mac, + struct vio_dev *vdev) { struct vnet *iter, *vp; @@ -1821,7 +1825,7 @@ static struct vnet *vnet_find_or_create(const u64 *local_mac) } } if (!vp) - vp = vnet_new(local_mac); + vp = vnet_new(local_mac, vdev); mutex_unlock(&vnet_list_mutex); return vp; @@ -1848,7 +1852,8 @@ static void vnet_cleanup(void) static const char *local_mac_prop = "local-mac-address"; static struct vnet *vnet_find_parent(struct mdesc_handle *hp, - u64 port_node) + u64 port_node, + struct vio_dev *vdev) { const u64 *local_mac = NULL; u64 a; @@ -1869,7 +1874,7 @@ static struct vnet *vnet_find_parent(struct mdesc_handle *hp, if (!local_mac) return ERR_PTR(-ENODEV); - return vnet_find_or_create(local_mac); + return vnet_find_or_create(local_mac, vdev); } static struct ldc_channel_config vnet_ldc_cfg = { @@ -1923,7 +1928,7 @@ static int vnet_port_probe(struct vio_dev *vdev, const struct vio_device_id *id) hp = mdesc_grab(); - vp = vnet_find_parent(hp, vdev->mp); + vp = vnet_find_parent(hp, vdev->mp, vdev); if (IS_ERR(vp)) { pr_err("Cannot find port parent vnet\n"); err = PTR_ERR(vp); From ed2e923945892a8372ab70d2f61d364b0b6d9054 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 19 Sep 2015 09:08:34 -0700 Subject: [PATCH 286/647] tcp/dccp: fix timewait races in timer handling When creating a timewait socket, we need to arm the timer before allowing other cpus to find it. The signal allowing cpus to find the socket is setting tw_refcnt to non zero value. As we set tw_refcnt in __inet_twsk_hashdance(), we therefore need to call inet_twsk_schedule() first. This also means we need to remove tw_refcnt changes from inet_twsk_schedule() and let the caller handle it. Note that because we use mod_timer_pinned(), we have the guarantee the timer wont expire before we set tw_refcnt as we run in BH context. To make things more readable I introduced inet_twsk_reschedule() helper. When rearming the timer, we can use mod_timer_pending() to make sure we do not rearm a canceled timer. Note: This bug can possibly trigger if packets of a flow can hit multiple cpus. This does not normally happen, unless flow steering is broken somehow. This explains this bug was spotted ~5 months after its introduction. A similar fix is needed for SYN_RECV sockets in reqsk_queue_hash_req(), but will be provided in a separate patch for proper tracking. Fixes: 789f558cfb36 ("tcp/dccp: get rid of central timewait timer") Signed-off-by: Eric Dumazet Reported-by: Ying Cai Signed-off-by: David S. Miller --- include/net/inet_timewait_sock.h | 14 +++++++++++++- net/dccp/minisocks.c | 4 ++-- net/ipv4/inet_timewait_sock.c | 16 ++++++++++------ net/ipv4/tcp_minisocks.c | 13 ++++++------- 4 files changed, 31 insertions(+), 16 deletions(-) diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h index 879d6e5a973b..186f3a1e1b1f 100644 --- a/include/net/inet_timewait_sock.h +++ b/include/net/inet_timewait_sock.h @@ -110,7 +110,19 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk, void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk, struct inet_hashinfo *hashinfo); -void inet_twsk_schedule(struct inet_timewait_sock *tw, const int timeo); +void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo, + bool rearm); + +static void inline inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo) +{ + __inet_twsk_schedule(tw, timeo, false); +} + +static void inline inet_twsk_reschedule(struct inet_timewait_sock *tw, int timeo) +{ + __inet_twsk_schedule(tw, timeo, true); +} + void inet_twsk_deschedule_put(struct inet_timewait_sock *tw); void inet_twsk_purge(struct inet_hashinfo *hashinfo, diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 30addee2dd03..838f524cf11a 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -48,8 +48,6 @@ void dccp_time_wait(struct sock *sk, int state, int timeo) tw->tw_ipv6only = sk->sk_ipv6only; } #endif - /* Linkage updates. */ - __inet_twsk_hashdance(tw, sk, &dccp_hashinfo); /* Get the TIME_WAIT timeout firing. */ if (timeo < rto) @@ -60,6 +58,8 @@ void dccp_time_wait(struct sock *sk, int state, int timeo) timeo = DCCP_TIMEWAIT_LEN; inet_twsk_schedule(tw, timeo); + /* Linkage updates. */ + __inet_twsk_hashdance(tw, sk, &dccp_hashinfo); inet_twsk_put(tw); } else { /* Sorry, if we're out of memory, just CLOSE this diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index ae22cc24fbe8..c67f9bd7699c 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -123,13 +123,15 @@ void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk, /* * Step 2: Hash TW into tcp ehash chain. * Notes : - * - tw_refcnt is set to 3 because : + * - tw_refcnt is set to 4 because : * - We have one reference from bhash chain. * - We have one reference from ehash chain. + * - We have one reference from timer. + * - One reference for ourself (our caller will release it). * We can use atomic_set() because prior spin_lock()/spin_unlock() * committed into memory all tw fields. */ - atomic_set(&tw->tw_refcnt, 1 + 1 + 1); + atomic_set(&tw->tw_refcnt, 4); inet_twsk_add_node_rcu(tw, &ehead->chain); /* Step 3: Remove SK from hash chain */ @@ -217,7 +219,7 @@ void inet_twsk_deschedule_put(struct inet_timewait_sock *tw) } EXPORT_SYMBOL(inet_twsk_deschedule_put); -void inet_twsk_schedule(struct inet_timewait_sock *tw, const int timeo) +void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo, bool rearm) { /* timeout := RTO * 3.5 * @@ -245,12 +247,14 @@ void inet_twsk_schedule(struct inet_timewait_sock *tw, const int timeo) */ tw->tw_kill = timeo <= 4*HZ; - if (!mod_timer_pinned(&tw->tw_timer, jiffies + timeo)) { - atomic_inc(&tw->tw_refcnt); + if (!rearm) { + BUG_ON(mod_timer_pinned(&tw->tw_timer, jiffies + timeo)); atomic_inc(&tw->tw_dr->tw_count); + } else { + mod_timer_pending(&tw->tw_timer, jiffies + timeo); } } -EXPORT_SYMBOL_GPL(inet_twsk_schedule); +EXPORT_SYMBOL_GPL(__inet_twsk_schedule); void inet_twsk_purge(struct inet_hashinfo *hashinfo, struct inet_timewait_death_row *twdr, int family) diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 6d8795b066ac..def765911ff8 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -162,9 +162,9 @@ kill_with_rst: if (tcp_death_row.sysctl_tw_recycle && tcptw->tw_ts_recent_stamp && tcp_tw_remember_stamp(tw)) - inet_twsk_schedule(tw, tw->tw_timeout); + inet_twsk_reschedule(tw, tw->tw_timeout); else - inet_twsk_schedule(tw, TCP_TIMEWAIT_LEN); + inet_twsk_reschedule(tw, TCP_TIMEWAIT_LEN); return TCP_TW_ACK; } @@ -201,7 +201,7 @@ kill: return TCP_TW_SUCCESS; } } - inet_twsk_schedule(tw, TCP_TIMEWAIT_LEN); + inet_twsk_reschedule(tw, TCP_TIMEWAIT_LEN); if (tmp_opt.saw_tstamp) { tcptw->tw_ts_recent = tmp_opt.rcv_tsval; @@ -251,7 +251,7 @@ kill: * Do not reschedule in the last case. */ if (paws_reject || th->ack) - inet_twsk_schedule(tw, TCP_TIMEWAIT_LEN); + inet_twsk_reschedule(tw, TCP_TIMEWAIT_LEN); return tcp_timewait_check_oow_rate_limit( tw, skb, LINUX_MIB_TCPACKSKIPPEDTIMEWAIT); @@ -322,9 +322,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo) } while (0); #endif - /* Linkage updates. */ - __inet_twsk_hashdance(tw, sk, &tcp_hashinfo); - /* Get the TIME_WAIT timeout firing. */ if (timeo < rto) timeo = rto; @@ -338,6 +335,8 @@ void tcp_time_wait(struct sock *sk, int state, int timeo) } inet_twsk_schedule(tw, timeo); + /* Linkage updates. */ + __inet_twsk_hashdance(tw, sk, &tcp_hashinfo); inet_twsk_put(tw); } else { /* Sorry, if we're out of memory, just CLOSE this From 29c6852602e259d2c1882f320b29d5c3fec0de04 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 19 Sep 2015 09:48:04 -0700 Subject: [PATCH 287/647] inet: fix races in reqsk_queue_hash_req() Before allowing lockless LISTEN processing, we need to make sure to arm the SYN_RECV timer before the req socket is visible in hash tables. Also, req->rsk_hash should be written before we set rsk_refcnt to a non zero value. Fixes: fa76ce7328b2 ("inet: get rid of central tcp/dccp listener timer") Signed-off-by: Eric Dumazet Cc: Ying Cai Signed-off-by: David S. Miller --- net/ipv4/inet_connection_sock.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 134957159c27..7bb9c39e0a4d 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -685,20 +685,20 @@ void reqsk_queue_hash_req(struct request_sock_queue *queue, req->num_timeout = 0; req->sk = NULL; + setup_timer(&req->rsk_timer, reqsk_timer_handler, (unsigned long)req); + mod_timer_pinned(&req->rsk_timer, jiffies + timeout); + req->rsk_hash = hash; + /* before letting lookups find us, make sure all req fields * are committed to memory and refcnt initialized. */ smp_wmb(); atomic_set(&req->rsk_refcnt, 2); - setup_timer(&req->rsk_timer, reqsk_timer_handler, (unsigned long)req); - req->rsk_hash = hash; spin_lock(&queue->syn_wait_lock); req->dl_next = lopt->syn_table[hash]; lopt->syn_table[hash] = req; spin_unlock(&queue->syn_wait_lock); - - mod_timer_pinned(&req->rsk_timer, jiffies + timeout); } EXPORT_SYMBOL(reqsk_queue_hash_req); From ff30cbc8da425754e8ab96904db1d295bd034f27 Mon Sep 17 00:00:00 2001 From: Mathias Nyman Date: Mon, 21 Sep 2015 17:46:09 +0300 Subject: [PATCH 288/647] usb: Use the USB_SS_MULT() macro to get the burst multiplier. Bits 1:0 of the bmAttributes are used for the burst multiplier. The rest of the bits used to be reserved (zero), but USB3.1 takes bit 7 into use. Use the existing USB_SS_MULT() macro instead to make sure the mult value and hence max packet calculations are correct for USB3.1 devices. Note that burst multiplier in bmAttributes is zero based and that the USB_SS_MULT() macro adds one. Cc: Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/config.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/usb/core/config.c b/drivers/usb/core/config.c index b2a540b43f97..b9ddf0c1ffe5 100644 --- a/drivers/usb/core/config.c +++ b/drivers/usb/core/config.c @@ -112,7 +112,7 @@ static void usb_parse_ss_endpoint_companion(struct device *ddev, int cfgno, cfgno, inum, asnum, ep->desc.bEndpointAddress); ep->ss_ep_comp.bmAttributes = 16; } else if (usb_endpoint_xfer_isoc(&ep->desc) && - desc->bmAttributes > 2) { + USB_SS_MULT(desc->bmAttributes) > 3) { dev_warn(ddev, "Isoc endpoint has Mult of %d in " "config %d interface %d altsetting %d ep %d: " "setting to 3\n", desc->bmAttributes + 1, @@ -121,7 +121,8 @@ static void usb_parse_ss_endpoint_companion(struct device *ddev, int cfgno, } if (usb_endpoint_xfer_isoc(&ep->desc)) - max_tx = (desc->bMaxBurst + 1) * (desc->bmAttributes + 1) * + max_tx = (desc->bMaxBurst + 1) * + (USB_SS_MULT(desc->bmAttributes)) * usb_endpoint_maxp(&ep->desc); else if (usb_endpoint_xfer_int(&ep->desc)) max_tx = usb_endpoint_maxp(&ep->desc) * From a6809ffd1687b3a8c192960e69add559b9d32649 Mon Sep 17 00:00:00 2001 From: Mathias Nyman Date: Mon, 21 Sep 2015 17:46:10 +0300 Subject: [PATCH 289/647] xhci: give command abortion one more chance before killing xhci We want to give the command abortion an additional try to stop the command ring before we completely hose xhci. Cc: Tested-by: Vincent Pelletier Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci-ring.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index a47a1e897086..1c61e5e597a8 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -302,6 +302,15 @@ static int xhci_abort_cmd_ring(struct xhci_hcd *xhci) ret = xhci_handshake(&xhci->op_regs->cmd_ring, CMD_RING_RUNNING, 0, 5 * 1000 * 1000); if (ret < 0) { + /* we are about to kill xhci, give it one more chance */ + xhci_write_64(xhci, temp_64 | CMD_RING_ABORT, + &xhci->op_regs->cmd_ring); + udelay(1000); + ret = xhci_handshake(&xhci->op_regs->cmd_ring, + CMD_RING_RUNNING, 0, 3 * 1000 * 1000); + if (ret == 0) + return 0; + xhci_err(xhci, "Stopped the command ring failed, " "maybe the host is dead\n"); xhci->xhc_state |= XHCI_STATE_DYING; From 2b7627b73e81e5d23d5ae1490fe8e690af86e053 Mon Sep 17 00:00:00 2001 From: Tomer Barletz Date: Mon, 21 Sep 2015 17:46:11 +0300 Subject: [PATCH 290/647] xhci: Move xhci_pme_quirk() behind #ifdef CONFIG_PM xhci_pme_quirk() is only used when CONFIG_PM is defined. Compiling a kernel without PM complains about this function [reworded commit message -Mathias] Cc: Signed-off-by: Tomer Barletz Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci-pci.c | 90 ++++++++++++++++++------------------- 1 file changed, 45 insertions(+), 45 deletions(-) diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c index 5590eac2b22d..c79d33676672 100644 --- a/drivers/usb/host/xhci-pci.c +++ b/drivers/usb/host/xhci-pci.c @@ -180,51 +180,6 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci) "QUIRK: Resetting on resume"); } -/* - * In some Intel xHCI controllers, in order to get D3 working, - * through a vendor specific SSIC CONFIG register at offset 0x883c, - * SSIC PORT need to be marked as "unused" before putting xHCI - * into D3. After D3 exit, the SSIC port need to be marked as "used". - * Without this change, xHCI might not enter D3 state. - * Make sure PME works on some Intel xHCI controllers by writing 1 to clear - * the Internal PME flag bit in vendor specific PMCTRL register at offset 0x80a4 - */ -static void xhci_pme_quirk(struct usb_hcd *hcd, bool suspend) -{ - struct xhci_hcd *xhci = hcd_to_xhci(hcd); - struct pci_dev *pdev = to_pci_dev(hcd->self.controller); - u32 val; - void __iomem *reg; - - if (pdev->vendor == PCI_VENDOR_ID_INTEL && - pdev->device == PCI_DEVICE_ID_INTEL_CHERRYVIEW_XHCI) { - - reg = (void __iomem *) xhci->cap_regs + PORT2_SSIC_CONFIG_REG2; - - /* Notify SSIC that SSIC profile programming is not done */ - val = readl(reg) & ~PROG_DONE; - writel(val, reg); - - /* Mark SSIC port as unused(suspend) or used(resume) */ - val = readl(reg); - if (suspend) - val |= SSIC_PORT_UNUSED; - else - val &= ~SSIC_PORT_UNUSED; - writel(val, reg); - - /* Notify SSIC that SSIC profile programming is done */ - val = readl(reg) | PROG_DONE; - writel(val, reg); - readl(reg); - } - - reg = (void __iomem *) xhci->cap_regs + 0x80a4; - val = readl(reg); - writel(val | BIT(28), reg); - readl(reg); -} - #ifdef CONFIG_ACPI static void xhci_pme_acpi_rtd3_enable(struct pci_dev *dev) { @@ -345,6 +300,51 @@ static void xhci_pci_remove(struct pci_dev *dev) } #ifdef CONFIG_PM +/* + * In some Intel xHCI controllers, in order to get D3 working, + * through a vendor specific SSIC CONFIG register at offset 0x883c, + * SSIC PORT need to be marked as "unused" before putting xHCI + * into D3. After D3 exit, the SSIC port need to be marked as "used". + * Without this change, xHCI might not enter D3 state. + * Make sure PME works on some Intel xHCI controllers by writing 1 to clear + * the Internal PME flag bit in vendor specific PMCTRL register at offset 0x80a4 + */ +static void xhci_pme_quirk(struct usb_hcd *hcd, bool suspend) +{ + struct xhci_hcd *xhci = hcd_to_xhci(hcd); + struct pci_dev *pdev = to_pci_dev(hcd->self.controller); + u32 val; + void __iomem *reg; + + if (pdev->vendor == PCI_VENDOR_ID_INTEL && + pdev->device == PCI_DEVICE_ID_INTEL_CHERRYVIEW_XHCI) { + + reg = (void __iomem *) xhci->cap_regs + PORT2_SSIC_CONFIG_REG2; + + /* Notify SSIC that SSIC profile programming is not done */ + val = readl(reg) & ~PROG_DONE; + writel(val, reg); + + /* Mark SSIC port as unused(suspend) or used(resume) */ + val = readl(reg); + if (suspend) + val |= SSIC_PORT_UNUSED; + else + val &= ~SSIC_PORT_UNUSED; + writel(val, reg); + + /* Notify SSIC that SSIC profile programming is done */ + val = readl(reg) | PROG_DONE; + writel(val, reg); + readl(reg); + } + + reg = (void __iomem *) xhci->cap_regs + 0x80a4; + val = readl(reg); + writel(val | BIT(28), reg); + readl(reg); +} + static int xhci_pci_suspend(struct usb_hcd *hcd, bool do_wakeup) { struct xhci_hcd *xhci = hcd_to_xhci(hcd); From 85ac90f8953a58f6a057b727bc9db97721e3fb8e Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Mon, 21 Sep 2015 17:46:12 +0300 Subject: [PATCH 291/647] usb: xhci: lock mutex on xhci_stop Else it races with xhci_setup_device Cc: Signed-off-by: Roger Quadros Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 6b0f4a47e402..f560c4100e2f 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -677,8 +677,11 @@ void xhci_stop(struct usb_hcd *hcd) u32 temp; struct xhci_hcd *xhci = hcd_to_xhci(hcd); + mutex_lock(&xhci->mutex); + if (!usb_hcd_is_primary_hcd(hcd)) { xhci_only_stop_hcd(xhci->shared_hcd); + mutex_unlock(&xhci->mutex); return; } @@ -717,6 +720,7 @@ void xhci_stop(struct usb_hcd *hcd) xhci_dbg_trace(xhci, trace_xhci_dbg_init, "xhci_stop completed - status = %x", readl(&xhci->op_regs->status)); + mutex_unlock(&xhci->mutex); } /* From e5bfeab0ad515b4f6df39fe716603e9dc6d3dfd0 Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Mon, 21 Sep 2015 17:46:13 +0300 Subject: [PATCH 292/647] usb: xhci: Clear XHCI_STATE_DYING on start For whatever reason if XHCI died in the previous instant then it will never recover on the next xhci_start unless we clear the DYING flag. Cc: Signed-off-by: Roger Quadros Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index f560c4100e2f..5fe24196c88b 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -146,7 +146,8 @@ static int xhci_start(struct xhci_hcd *xhci) "waited %u microseconds.\n", XHCI_MAX_HALT_USEC); if (!ret) - xhci->xhc_state &= ~XHCI_STATE_HALTED; + xhci->xhc_state &= ~(XHCI_STATE_HALTED | XHCI_STATE_DYING); + return ret; } From 8c24d6d7b09deee3036ddc4f2b81b53b28c8f877 Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Mon, 21 Sep 2015 17:46:14 +0300 Subject: [PATCH 293/647] usb: xhci: stop everything on the first call to xhci_stop xhci_stop will be called twice, once for the shared hcd and again for the primary hcd. We stop the XHCI controller in any case so clean up everything on the first call else we can timeout waiting for pending requests to complete. Cc: Signed-off-by: Roger Quadros Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci.c | 20 +++++--------------- 1 file changed, 5 insertions(+), 15 deletions(-) diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 5fe24196c88b..f881d5a85486 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -655,15 +655,6 @@ int xhci_run(struct usb_hcd *hcd) } EXPORT_SYMBOL_GPL(xhci_run); -static void xhci_only_stop_hcd(struct usb_hcd *hcd) -{ - struct xhci_hcd *xhci = hcd_to_xhci(hcd); - - spin_lock_irq(&xhci->lock); - xhci_halt(xhci); - spin_unlock_irq(&xhci->lock); -} - /* * Stop xHCI driver. * @@ -678,15 +669,14 @@ void xhci_stop(struct usb_hcd *hcd) u32 temp; struct xhci_hcd *xhci = hcd_to_xhci(hcd); - mutex_lock(&xhci->mutex); - - if (!usb_hcd_is_primary_hcd(hcd)) { - xhci_only_stop_hcd(xhci->shared_hcd); - mutex_unlock(&xhci->mutex); + if (xhci->xhc_state & XHCI_STATE_HALTED) return; - } + mutex_lock(&xhci->mutex); spin_lock_irq(&xhci->lock); + xhci->xhc_state |= XHCI_STATE_HALTED; + xhci->cmd_ring_state = CMD_RING_STATE_STOPPED; + /* Make sure the xHC is halted for a USB3 roothub * (xhci_stop() could be called as part of failed init). */ From 448116bfa856d3c076fa7178ed96661a008a5d45 Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Mon, 21 Sep 2015 17:46:15 +0300 Subject: [PATCH 294/647] usb: xhci: exit early in xhci_setup_device() if we're halted or dying During quick plug/removal of OTG adapter during dual-role testing it can happen that xhci_alloc_device() is called for the newly detected device after the DRD library has called xhci_stop to remove the HCD. If that is the case, just fail early to prevent the following warning. [ 154.732649] hub 4-0:1.0: USB hub found [ 154.742204] hub 4-0:1.0: 1 port detected [ 154.824458] hub 3-0:1.0: state 7 ports 1 chg 0002 evt 0000 [ 154.854609] hub 4-0:1.0: state 7 ports 1 chg 0000 evt 0000 [ 154.944430] usb 3-1: new high-speed USB device number 2 using xhci-hcd [ 154.951009] xhci-hcd xhci-hcd.0.auto: xhci_setup_device [ 155.038191] xhci-hcd xhci-hcd.0.auto: remove, state 4 [ 155.043315] usb usb4: USB disconnect, device number 1 [ 155.055270] xhci-hcd xhci-hcd.0.auto: xhci_stop [ 155.060094] xhci-hcd xhci-hcd.0.auto: USB bus 4 deregistered [ 155.066576] xhci-hcd xhci-hcd.0.auto: remove, state 1 [ 155.071710] usb usb3: USB disconnect, device number 1 [ 155.077124] xhci-hcd xhci-hcd.0.auto: xhci_setup_device [ 155.082389] ------------[ cut here ]------------ [ 155.087690] WARNING: CPU: 0 PID: 72 at drivers/usb/host/xhci.c:3800 xhci_setup_device+0x410/0x484 [xhci_hcd]() [ 155.097861] Modules linked in: sd_mod usb_storage scsi_mod usb_f_ss_lb g_zero libcomposite ipv6 xhci_plat_hcd xhci_hcd usbcore dwc3 udc_core evdev ti_am335x_adc joydev kfifo_buf industrialio snd_soc_simple_cc [ 155.146734] CPU: 0 PID: 72 Comm: kworker/0:3 Tainted: G W 4.1.4-00834-gcd9380b-dirty #50 [ 155.156073] Hardware name: Generic AM43 (Flattened Device Tree) [ 155.162117] Workqueue: usb_hub_wq hub_event [usbcore] [ 155.167249] Backtrace: [ 155.169751] [] (dump_backtrace) from [] (show_stack+0x18/0x1c) [ 155.177390] r6:c089d4a4 r5:ffffffff r4:00000000 r3:ee46c000 [ 155.183137] [] (show_stack) from [] (dump_stack+0x84/0xd0) [ 155.190446] [] (dump_stack) from [] (warn_slowpath_common+0x80/0xbc) [ 155.198605] r7:00000009 r6:00000ed8 r5:bf27eb70 r4:00000000 [ 155.204348] [] (warn_slowpath_common) from [] (warn_slowpath_null+0x24/0x2c) [ 155.213202] r8:ee49f000 r7:ee7c0004 r6:00000000 r5:ee7c0158 r4:ee7c0000 [ 155.220051] [] (warn_slowpath_null) from [] (xhci_setup_device+0x410/0x484 [xhci_hcd]) [ 155.229816] [] (xhci_setup_device [xhci_hcd]) from [] (xhci_address_device+0x14/0x18 [xhci_hcd]) [ 155.240415] r10:ee598200 r9:00000001 r8:00000002 r7:00000001 r6:00000003 r5:00000002 [ 155.248363] r4:ee49f000 [ 155.250978] [] (xhci_address_device [xhci_hcd]) from [] (hub_port_init+0x1b8/0xa9c [usbcore]) [ 155.261403] [] (hub_port_init [usbcore]) from [] (hub_event+0x738/0x1020 [usbcore]) [ 155.270874] r10:ee598200 r9:ee7c0000 r8:ee7c0038 r7:ee518800 r6:ee49f000 r5:00000001 [ 155.278822] r4:00000000 [ 155.281426] [] (hub_event [usbcore]) from [] (process_one_work+0x128/0x340) [ 155.290196] r10:00000000 r9:00000003 r8:00000000 r7:fedfa000 r6:eeec5400 r5:ee598314 [ 155.298151] r4:ee434380 [ 155.300718] [] (process_one_work) from [] (worker_thread+0x158/0x49c) [ 155.308963] r10:ee434380 r9:00000003 r8:eeec5400 r7:00000008 r6:ee434398 r5:eeec5400 [ 155.316913] r4:eeec5414 [ 155.319482] [] (worker_thread) from [] (kthread+0xdc/0xf8) [ 155.326765] r10:00000000 r9:00000000 r8:00000000 r7:c00577a0 r6:ee434380 r5:ee4441c0 [ 155.334713] r4:00000000 r3:00000000 [ 155.338341] [] (kthread) from [] (ret_from_fork+0x14/0x2c) [ 155.345626] r7:00000000 r6:00000000 r5:c005cb64 r4:ee4441c0 [ 155.356108] ---[ end trace a58d34c223b190e6 ]--- [ 155.360783] xhci-hcd xhci-hcd.0.auto: Virt dev invalid for slot_id 0x1! [ 155.574404] xhci-hcd xhci-hcd.0.auto: xhci_setup_device [ 155.579667] ------------[ cut here ]------------ Cc: Signed-off-by: Roger Quadros Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index f881d5a85486..9957bd96d4bc 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -3788,6 +3788,9 @@ static int xhci_setup_device(struct usb_hcd *hcd, struct usb_device *udev, mutex_lock(&xhci->mutex); + if (xhci->xhc_state) /* dying or halted */ + goto out; + if (!udev->slot_id) { xhci_dbg_trace(xhci, trace_xhci_dbg_address, "Bad Slot ID %d", udev->slot_id); From dca7794539eff04b786fb6907186989e5eaaa9c2 Mon Sep 17 00:00:00 2001 From: Mathias Nyman Date: Mon, 21 Sep 2015 17:46:16 +0300 Subject: [PATCH 295/647] xhci: change xhci 1.0 only restrictions to support xhci 1.1 Some changes between xhci 0.96 and xhci 1.0 specifications forced us to check the hci version in code, some of these checks were implemented as hci_version == 1.0, which will not work with new xhci 1.1 controllers. xhci 1.1 behaves similar to xhci 1.0 in these cases, so change these checks to hci_version >= 1.0 Cc: Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci-mem.c | 6 +++--- drivers/usb/host/xhci-ring.c | 4 ++-- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 9a8c936cd42c..8497fb8f6193 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -1498,10 +1498,10 @@ int xhci_endpoint_init(struct xhci_hcd *xhci, * use Event Data TRBs, and we don't chain in a link TRB on short * transfers, we're basically dividing by 1. * - * xHCI 1.0 specification indicates that the Average TRB Length should - * be set to 8 for control endpoints. + * xHCI 1.0 and 1.1 specification indicates that the Average TRB Length + * should be set to 8 for control endpoints. */ - if (usb_endpoint_xfer_control(&ep->desc) && xhci->hci_version == 0x100) + if (usb_endpoint_xfer_control(&ep->desc) && xhci->hci_version >= 0x100) ep_ctx->tx_info |= cpu_to_le32(AVG_TRB_LENGTH_FOR_EP(8)); else ep_ctx->tx_info |= diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 1c61e5e597a8..43291f93afeb 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -3470,8 +3470,8 @@ int xhci_queue_ctrl_tx(struct xhci_hcd *xhci, gfp_t mem_flags, if (start_cycle == 0) field |= 0x1; - /* xHCI 1.0 6.4.1.2.1: Transfer Type field */ - if (xhci->hci_version == 0x100) { + /* xHCI 1.0/1.1 6.4.1.2.1: Transfer Type field */ + if (xhci->hci_version >= 0x100) { if (urb->transfer_buffer_length > 0) { if (setup->bRequestType & USB_DIR_IN) field |= TRB_TX_TYPE(TRB_DATA_IN); From cc8e4fc0c3b5e8340bc8358990515d116a3c274c Mon Sep 17 00:00:00 2001 From: Mathias Nyman Date: Mon, 21 Sep 2015 17:46:17 +0300 Subject: [PATCH 296/647] xhci: init command timeout timer earlier to avoid deleting it uninitialized Don't check if timer is running with a timer_pending() before deleting it with del_timer_sync(), this defies the whole point of the sync part and can cause a possible race. Instead we just want to make sure the timer is initialized early enough before we have a chance to delete it. Cc: Reported-by: Oliver Neukum Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci-mem.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 8497fb8f6193..41f841fa6c4d 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -1792,8 +1792,7 @@ void xhci_mem_cleanup(struct xhci_hcd *xhci) int size; int i, j, num_ports; - if (timer_pending(&xhci->cmd_timer)) - del_timer_sync(&xhci->cmd_timer); + del_timer_sync(&xhci->cmd_timer); /* Free the Event Ring Segment Table and the actual Event Ring */ size = sizeof(struct xhci_erst_entry)*(xhci->erst.num_entries); @@ -2321,6 +2320,10 @@ int xhci_mem_init(struct xhci_hcd *xhci, gfp_t flags) INIT_LIST_HEAD(&xhci->cmd_list); + /* init command timeout timer */ + setup_timer(&xhci->cmd_timer, xhci_handle_command_timeout, + (unsigned long)xhci); + page_size = readl(&xhci->op_regs->page_size); xhci_dbg_trace(xhci, trace_xhci_dbg_init, "Supported page size register = 0x%x", page_size); @@ -2505,10 +2508,6 @@ int xhci_mem_init(struct xhci_hcd *xhci, gfp_t flags) "Wrote ERST address to ir_set 0."); xhci_print_ir_set(xhci, 0); - /* init command timeout timer */ - setup_timer(&xhci->cmd_timer, xhci_handle_command_timeout, - (unsigned long)xhci); - /* * XXX: Might need to set the Interrupter Moderation Register to * something other than the default (~1ms minimum between interrupts). From 208473c1f3ac3eccec097021eec3890f5e20fcc7 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 17 Sep 2015 14:13:44 +0100 Subject: [PATCH 297/647] ARM: wire up new syscalls Wire up the new userfaultfd and membarrier syscalls for ARM. Signed-off-by: Russell King --- arch/arm/include/asm/unistd.h | 2 +- arch/arm/include/uapi/asm/unistd.h | 2 ++ arch/arm/kernel/calls.S | 2 ++ 3 files changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/arm/include/asm/unistd.h b/arch/arm/include/asm/unistd.h index 32640c431a08..7cba573c2cc9 100644 --- a/arch/arm/include/asm/unistd.h +++ b/arch/arm/include/asm/unistd.h @@ -19,7 +19,7 @@ * This may need to be greater than __NR_last_syscall+1 in order to * account for the padding in the syscall table */ -#define __NR_syscalls (388) +#define __NR_syscalls (392) /* * *NOTE*: This is a ghost syscall private to the kernel. Only the diff --git a/arch/arm/include/uapi/asm/unistd.h b/arch/arm/include/uapi/asm/unistd.h index 0c3f5a0dafd3..7a2a32a1d5a8 100644 --- a/arch/arm/include/uapi/asm/unistd.h +++ b/arch/arm/include/uapi/asm/unistd.h @@ -414,6 +414,8 @@ #define __NR_memfd_create (__NR_SYSCALL_BASE+385) #define __NR_bpf (__NR_SYSCALL_BASE+386) #define __NR_execveat (__NR_SYSCALL_BASE+387) +#define __NR_userfaultfd (__NR_SYSCALL_BASE+388) +#define __NR_membarrier (__NR_SYSCALL_BASE+389) /* * The following SWIs are ARM private. diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S index 05745eb838c5..fde6c88d560c 100644 --- a/arch/arm/kernel/calls.S +++ b/arch/arm/kernel/calls.S @@ -397,6 +397,8 @@ /* 385 */ CALL(sys_memfd_create) CALL(sys_bpf) CALL(sys_execveat) + CALL(sys_userfaultfd) + CALL(sys_membarrier) #ifndef syscalls_counted .equ syscalls_padding, ((NR_syscalls + 3) & ~3) - NR_syscalls #define syscalls_counted From 2df1b131b54f431877a6665139dac805ba5ce1a5 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 21 Aug 2015 14:07:13 +0200 Subject: [PATCH 298/647] mac80211: fix VHT MCS mask array overrun The HT MCS mask has 9 bytes, the VHT one only has 8 streams. Split the loops to handle this correctly. Reported-by: Dan Carpenter Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 17b1fe961c5d..a30ec3ce3d25 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -2518,15 +2518,17 @@ static int ieee80211_set_bitrate_mask(struct wiphy *wiphy, continue; for (j = 0; j < IEEE80211_HT_MCS_MASK_LEN; j++) { - if (~sdata->rc_rateidx_mcs_mask[i][j]) + if (~sdata->rc_rateidx_mcs_mask[i][j]) { sdata->rc_has_mcs_mask[i] = true; - - if (~sdata->rc_rateidx_vht_mcs_mask[i][j]) - sdata->rc_has_vht_mcs_mask[i] = true; - - if (sdata->rc_has_mcs_mask[i] && - sdata->rc_has_vht_mcs_mask[i]) break; + } + } + + for (j = 0; j < NL80211_VHT_NSS_MAX; j++) { + if (~sdata->rc_rateidx_vht_mcs_mask[i][j]) { + sdata->rc_has_vht_mcs_mask[i] = true; + break; + } } } From babc305e21ea3811d98b67437299360904ac1b6a Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Mon, 21 Sep 2015 15:47:40 +0300 Subject: [PATCH 299/647] mac80211: reset CQM history upon reconfiguration The current behavior of notifying CQM events is inconsistent: Upon first configuration there is a cqm event with the current status according to threshold configured, regardless of signal stability. When there is reconfiguration no event is sent unless there is a significant change to the signal level according to the new configuration. Since the current reconfiguration behavior might cause missing CQM events in case the current signal did not change but is on the other side of the new threshold, fix that by resetting the stored signal level upon reconfiguration. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index a30ec3ce3d25..7a77a1470f25 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -2474,6 +2474,7 @@ static int ieee80211_set_cqm_rssi_config(struct wiphy *wiphy, bss_conf->cqm_rssi_thold = rssi_thold; bss_conf->cqm_rssi_hyst = rssi_hyst; + sdata->u.mgd.last_cqm_event_signal = 0; /* tell the driver upon association, unless already associated */ if (sdata->u.mgd.associated && From d32dc9aa10c739363c775baf4499416b2e0dc11f Mon Sep 17 00:00:00 2001 From: Ludovic Desroches Date: Mon, 21 Sep 2015 15:46:04 +0200 Subject: [PATCH 300/647] irqchip/atmel-aic5: Use per chip mask caches in mask/unmask() When masking/unmasking interrupts, mask_cache is updated and used later for suspend/resume. Unfortunately, it always was the mask_cache associated with the first irq chip which was updated. So when performing resume, only irqs 0-31 could be enabled. Fixes: b1479ebb7720 ("irqchip: atmel-aic: Add atmel AIC/AIC5 drivers") Signed-off-by: Ludovic Desroches Cc: Cc: Cc: Cc: Cc: Cc: Cc: Cc: Cc: stable@vger.kernel.org #3.18 Link: http://lkml.kernel.org/r/1442843173-2390-1-git-send-email-ludovic.desroches@atmel.com Signed-off-by: Thomas Gleixner --- drivers/irqchip/irq-atmel-aic5.c | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) diff --git a/drivers/irqchip/irq-atmel-aic5.c b/drivers/irqchip/irq-atmel-aic5.c index 9da9942ac83c..f6d680485bee 100644 --- a/drivers/irqchip/irq-atmel-aic5.c +++ b/drivers/irqchip/irq-atmel-aic5.c @@ -88,28 +88,36 @@ static void aic5_mask(struct irq_data *d) { struct irq_domain *domain = d->domain; struct irq_domain_chip_generic *dgc = domain->gc; - struct irq_chip_generic *gc = dgc->gc[0]; + struct irq_chip_generic *bgc = dgc->gc[0]; + struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d); - /* Disable interrupt on AIC5 */ - irq_gc_lock(gc); + /* + * Disable interrupt on AIC5. We always take the lock of the + * first irq chip as all chips share the same registers. + */ + irq_gc_lock(bgc); irq_reg_writel(gc, d->hwirq, AT91_AIC5_SSR); irq_reg_writel(gc, 1, AT91_AIC5_IDCR); gc->mask_cache &= ~d->mask; - irq_gc_unlock(gc); + irq_gc_unlock(bgc); } static void aic5_unmask(struct irq_data *d) { struct irq_domain *domain = d->domain; struct irq_domain_chip_generic *dgc = domain->gc; - struct irq_chip_generic *gc = dgc->gc[0]; + struct irq_chip_generic *bgc = dgc->gc[0]; + struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d); - /* Enable interrupt on AIC5 */ - irq_gc_lock(gc); + /* + * Enable interrupt on AIC5. We always take the lock of the + * first irq chip as all chips share the same registers. + */ + irq_gc_lock(bgc); irq_reg_writel(gc, d->hwirq, AT91_AIC5_SSR); irq_reg_writel(gc, 1, AT91_AIC5_IECR); gc->mask_cache |= d->mask; - irq_gc_unlock(gc); + irq_gc_unlock(bgc); } static int aic5_retrigger(struct irq_data *d) From c2e4b24ff848bb180f9b9cd873a38327cd219ad2 Mon Sep 17 00:00:00 2001 From: Kapileshwar Singh Date: Tue, 22 Sep 2015 14:22:03 +0100 Subject: [PATCH 301/647] tools lib traceevent: Fix string handling in heterogeneous arch environments When a trace recorded on a 32-bit device is processed with a 64-bit binary, the higher 32-bits of the address need to ignored. The lack of this results in the output of the 64-bit pointer value to the trace as the 32-bit address lookup fails in find_printk(). Before: burn-1778 [003] 548.600305: bputs: 0xc0046db2s: 2cec5c058d98c After: burn-1778 [003] 548.600305: bputs: 0xc0046db2s: RT throttling activated The problem occurs in PRINT_FIELD when the field is recognized as a pointer to a string (of the type const char *) Heterogeneous architectures cases below can arise and should be handled: * Traces recorded using 32-bit addresses processed on a 64-bit machine * Traces recorded using 64-bit addresses processed on a 32-bit machine Reported-by: Juri Lelli Signed-off-by: Kapileshwar Singh Reviewed-by: Steven Rostedt Cc: David Ahern Cc: Javi Merino Cc: Jiri Olsa Cc: Namhyung Kim Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1442928123-13824-1-git-send-email-kapileshwar.singh@arm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/lib/traceevent/event-parse.c | 23 ++++++++++++++++++++--- 1 file changed, 20 insertions(+), 3 deletions(-) diff --git a/tools/lib/traceevent/event-parse.c b/tools/lib/traceevent/event-parse.c index 4d885934b919..cf42b090477b 100644 --- a/tools/lib/traceevent/event-parse.c +++ b/tools/lib/traceevent/event-parse.c @@ -3795,7 +3795,7 @@ static void print_str_arg(struct trace_seq *s, void *data, int size, struct format_field *field; struct printk_map *printk; long long val, fval; - unsigned long addr; + unsigned long long addr; char *str; unsigned char *hex; int print; @@ -3828,13 +3828,30 @@ static void print_str_arg(struct trace_seq *s, void *data, int size, */ if (!(field->flags & FIELD_IS_ARRAY) && field->size == pevent->long_size) { - addr = *(unsigned long *)(data + field->offset); + + /* Handle heterogeneous recording and processing + * architectures + * + * CASE I: + * Traces recorded on 32-bit devices (32-bit + * addressing) and processed on 64-bit devices: + * In this case, only 32 bits should be read. + * + * CASE II: + * Traces recorded on 64 bit devices and processed + * on 32-bit devices: + * In this case, 64 bits must be read. + */ + addr = (pevent->long_size == 8) ? + *(unsigned long long *)(data + field->offset) : + (unsigned long long)*(unsigned int *)(data + field->offset); + /* Check if it matches a print format */ printk = find_printk(pevent, addr); if (printk) trace_seq_puts(s, printk->printk); else - trace_seq_printf(s, "%lx", addr); + trace_seq_printf(s, "%llx", addr); break; } str = malloc(len + 1); From fc2ca674470bbfe11d72a20a3f19fd3dc43bfca0 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sun, 30 Aug 2015 21:19:58 -0700 Subject: [PATCH 302/647] MIPS: Fix console output for Fulong2e system Commit 3adeb2566b9b ("MIPS: Loongson: Improve LEFI firmware interface") made the number of UARTs dynamic if LEFI_FIRMWARE_INTERFACE is configured. Unfortunately, it did not initialize the number of UARTs if LEFI_FIRMWARE_INTERFACE is not configured. As a result, the Fulong2e system has no console. Fixes: 3adeb2566b9b ("MIPS: Loongson: Improve LEFI firmware interface") Acked-by: Huacai Chen Signed-off-by: Guenter Roeck Tested-by: Markos Chandras Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/11076/ Signed-off-by: Ralf Baechle --- arch/mips/loongson64/common/env.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/mips/loongson64/common/env.c b/arch/mips/loongson64/common/env.c index f6c44dd332e2..d6d07ad56180 100644 --- a/arch/mips/loongson64/common/env.c +++ b/arch/mips/loongson64/common/env.c @@ -64,6 +64,9 @@ void __init prom_init_env(void) } if (memsize == 0) memsize = 256; + + loongson_sysconf.nr_uarts = 1; + pr_info("memsize=%u, highmemsize=%u\n", memsize, highmemsize); #else struct boot_params *boot_p; From 88d3426942d748b90b051b7ef2d5d765f5f3054c Mon Sep 17 00:00:00 2001 From: Alexander Sverdlin Date: Thu, 3 Sep 2015 08:36:35 +0200 Subject: [PATCH 303/647] MIPS: bootmem: Fix mapstart calculation for contiguous maps Commit a6335fa1 fixed the case with gap between initrd and next usable PFN zone, but broken the case when initrd is combined with usable memory into one region (in add_memory_region()). Restore the fixup initially brought in by f9a7febd. ---- error message ---- Unpacking initramfs... Initramfs unpacking failed: junk in compressed archive BUG: Bad page state in process swapper pfn:00261 page:81004c20 count:0 mapcount:-127 mapping: (null) index:0x2 flags: 0x0() page dumped because: nonzero mapcount CPU: 0 PID: 1 Comm: swapper Not tainted 4.2.0+ #1782 ----------------------- Signed-off-by: Alexander Sverdlin Reported-by: Tony Wu Tested-by: Tony Wu Cc: David Daney Cc: Zubair Lutfullah Kakakhel Cc: Huacai Chen Cc: Joe Perches Cc: Steven J. Hill Cc: Aaro Koskinen Cc: stable@vger.kernel.org Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/11086/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/setup.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c index 35b8316002f8..479515109e5b 100644 --- a/arch/mips/kernel/setup.c +++ b/arch/mips/kernel/setup.c @@ -338,7 +338,7 @@ static void __init bootmem_init(void) if (end <= reserved_end) continue; #ifdef CONFIG_BLK_DEV_INITRD - /* mapstart should be after initrd_end */ + /* Skip zones before initrd and initrd itself */ if (initrd_end && end <= (unsigned long)PFN_UP(__pa(initrd_end))) continue; #endif @@ -371,6 +371,14 @@ static void __init bootmem_init(void) max_low_pfn = PFN_DOWN(HIGHMEM_START); } +#ifdef CONFIG_BLK_DEV_INITRD + /* + * mapstart should be after initrd_end + */ + if (initrd_end) + mapstart = max(mapstart, (unsigned long)PFN_UP(__pa(initrd_end))); +#endif + /* * Initialize the boot-time allocator with low memory only. */ From f7a7651fcd40a2bf29357ffeaee176fb159bcddb Mon Sep 17 00:00:00 2001 From: Jonathan McDowell Date: Mon, 21 Sep 2015 21:23:47 +0100 Subject: [PATCH 304/647] tty: serial: Add missing module license for 8250_base.ko The split of the 8250 driver into a 8250_base/8250.ko resulted in a lack of a license for the 8250_base.ko module. This caused the module to fail to load and the kernel to be tainted. Add the appropriate MODULE_LICENSE to 8250_port.c, which is always compiled into 8250_base.ko Signed-off-by: Jonathan McDowell Reported-by: Mikael Pettersson Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_port.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index 54e6c8ddef5d..b1e0ba3e525b 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -2910,3 +2910,5 @@ int serial8250_console_setup(struct uart_port *port, char *options, bool probe) } #endif /* CONFIG_SERIAL_8250_CONSOLE */ + +MODULE_LICENSE("GPL"); From b0b4855099e301c8603ea37da9a0103a96c2e0b1 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Tue, 22 Sep 2015 14:32:03 +0300 Subject: [PATCH 305/647] spi: xtensa-xtfpga: fix register endianness XTFPGA SPI controller has native endian registers. Fix register acessors so that they work in big-endian configurations. Signed-off-by: Max Filippov Signed-off-by: Mark Brown Cc: stable@vger.kernel.org --- drivers/spi/spi-xtensa-xtfpga.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/spi/spi-xtensa-xtfpga.c b/drivers/spi/spi-xtensa-xtfpga.c index 2e32ea2f194f..be6155cba9de 100644 --- a/drivers/spi/spi-xtensa-xtfpga.c +++ b/drivers/spi/spi-xtensa-xtfpga.c @@ -34,13 +34,13 @@ struct xtfpga_spi { static inline void xtfpga_spi_write32(const struct xtfpga_spi *spi, unsigned addr, u32 val) { - iowrite32(val, spi->regs + addr); + __raw_writel(val, spi->regs + addr); } static inline unsigned int xtfpga_spi_read32(const struct xtfpga_spi *spi, unsigned addr) { - return ioread32(spi->regs + addr); + return __raw_readl(spi->regs + addr); } static inline void xtfpga_spi_wait_busy(struct xtfpga_spi *xspi) From f0e03dbd2d61d991bdd2d76b4e84681fe3077176 Mon Sep 17 00:00:00 2001 From: Richard Fitzgerald Date: Tue, 22 Sep 2015 15:30:33 +0100 Subject: [PATCH 306/647] MAINTAINERS: Update website and git repo for Wolfson Microelectronics Support for Wolfson Microelectronics devices is now part of Cirrus Logic and the relevant parts of the old opensource.wolfsonmicro.com site have moved to the Cirrus Logic GitHub area. This patch updates the website and git repo links, and also removes an obsolete website link for the voltage and current drivers. Signed-off-by: Richard Fitzgerald Signed-off-by: Mark Brown --- MAINTAINERS | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 7ba7ab749c85..dffea210e439 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -11239,7 +11239,6 @@ VOLTAGE AND CURRENT REGULATOR FRAMEWORK M: Liam Girdwood M: Mark Brown L: linux-kernel@vger.kernel.org -W: http://opensource.wolfsonmicro.com/node/15 W: http://www.slimlogic.co.uk/?p=48 T: git git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator.git S: Supported @@ -11368,17 +11367,15 @@ WM97XX TOUCHSCREEN DRIVERS M: Mark Brown M: Liam Girdwood L: linux-input@vger.kernel.org -T: git git://opensource.wolfsonmicro.com/linux-2.6-touch -W: http://opensource.wolfsonmicro.com/node/7 +W: https://github.com/CirrusLogic/linux-drivers/wiki S: Supported F: drivers/input/touchscreen/*wm97* F: include/linux/wm97xx.h WOLFSON MICROELECTRONICS DRIVERS L: patches@opensource.wolfsonmicro.com -T: git git://opensource.wolfsonmicro.com/linux-2.6-asoc -T: git git://opensource.wolfsonmicro.com/linux-2.6-audioplus -W: http://opensource.wolfsonmicro.com/content/linux-drivers-wolfson-devices +T: git https://github.com/CirrusLogic/linux-drivers.git +W: https://github.com/CirrusLogic/linux-drivers/wiki S: Supported F: Documentation/hwmon/wm83?? F: arch/arm/mach-s3c64xx/mach-crag6410* From faa9724a674e5e52316bb0d173aed16bd17d536c Mon Sep 17 00:00:00 2001 From: Aurelien Jarno Date: Sat, 5 Sep 2015 18:46:56 +0200 Subject: [PATCH 307/647] MIPS: BPF: Avoid unreachable code on little endian On little endian, avoid generating the big endian version of the code by using #else in addition to #ifdef #endif. Also fix one alignment issue wrt delay slot. Signed-off-by: Aurelien Jarno Reviewed-by: Markos Chandras Cc: stable@vger.kernel.org # v4.2+ Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/11097/ Signed-off-by: Ralf Baechle --- arch/mips/net/bpf_jit_asm.S | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/arch/mips/net/bpf_jit_asm.S b/arch/mips/net/bpf_jit_asm.S index e92726099be0..4f54cb18f104 100644 --- a/arch/mips/net/bpf_jit_asm.S +++ b/arch/mips/net/bpf_jit_asm.S @@ -151,9 +151,10 @@ NESTED(bpf_slow_path_word, (6 * SZREG), $r_sp) wsbh t0, $r_s0 jr $r_ra rotr $r_A, t0, 16 -#endif +#else jr $r_ra - move $r_A, $r_s0 + move $r_A, $r_s0 +#endif END(bpf_slow_path_word) @@ -162,9 +163,10 @@ NESTED(bpf_slow_path_half, (6 * SZREG), $r_sp) #ifdef CONFIG_CPU_LITTLE_ENDIAN jr $r_ra wsbh $r_A, $r_s0 -#endif +#else jr $r_ra move $r_A, $r_s0 +#endif END(bpf_slow_path_half) From b259e51f2e29390518021f9b8df55a3de42f371b Mon Sep 17 00:00:00 2001 From: Aurelien Jarno Date: Sat, 5 Sep 2015 18:46:57 +0200 Subject: [PATCH 308/647] MIPS: BPF: Fix build on pre-R2 little endian CPUs The rotr, seh and wsbh instructions have been introduced with the R2 ISA. Thus the current BPF code fails to build on pre-R2 little endian CPUs: CC arch/mips/net/bpf_jit.o AS arch/mips/net/bpf_jit_asm.o /home/aurel32/linux-4.2/arch/mips/net/bpf_jit_asm.S: Assembler messages: /home/aurel32/linux-4.2/arch/mips/net/bpf_jit_asm.S:67: Error: opcode not supported on this processor: mips32 (mips32) `wsbh $8,$19' /home/aurel32/linux-4.2/arch/mips/net/bpf_jit_asm.S:68: Error: opcode not supported on this processor: mips32 (mips32) `rotr $19,$8,16' /home/aurel32/linux-4.2/arch/mips/net/bpf_jit_asm.S:83: Error: opcode not supported on this processor: mips32 (mips32) `wsbh $8,$19' /home/aurel32/linux-4.2/arch/mips/net/bpf_jit_asm.S:84: Error: opcode not supported on this processor: mips32 (mips32) `seh $19,$8' /home/aurel32/linux-4.2/arch/mips/net/bpf_jit_asm.S:151: Error: opcode not supported on this processor: mips32 (mips32) `wsbh $8,$12' /home/aurel32/linux-4.2/arch/mips/net/bpf_jit_asm.S:153: Error: opcode not supported on this processor: mips32 (mips32) `rotr $19,$8,16' /home/aurel32/linux-4.2/arch/mips/net/bpf_jit_asm.S:164: Error: opcode not supported on this processor: mips32 (mips32) `wsbh $19,$12' /home/aurel32/linux-4.2/scripts/Makefile.build:294: recipe for target 'arch/mips/net/bpf_jit_asm.o' failed Fix that by providing equivalent code for these CPUs. Signed-off-by: Aurelien Jarno Reviewed-by: Markos Chandras Cc: stable@vger.kernel.org # v4.2+ Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/11098/ Signed-off-by: Ralf Baechle --- arch/mips/net/bpf_jit_asm.S | 42 +++++++++++++++++++++++++++++++++++++ 1 file changed, 42 insertions(+) diff --git a/arch/mips/net/bpf_jit_asm.S b/arch/mips/net/bpf_jit_asm.S index 4f54cb18f104..dabf4179cd7e 100644 --- a/arch/mips/net/bpf_jit_asm.S +++ b/arch/mips/net/bpf_jit_asm.S @@ -64,8 +64,20 @@ sk_load_word_positive: PTR_ADDU t1, $r_skb_data, offset lw $r_A, 0(t1) #ifdef CONFIG_CPU_LITTLE_ENDIAN +# if defined(__mips_isa_rev) && (__mips_isa_rev >= 2) wsbh t0, $r_A rotr $r_A, t0, 16 +# else + sll t0, $r_A, 24 + srl t1, $r_A, 24 + srl t2, $r_A, 8 + or t0, t0, t1 + andi t2, t2, 0xff00 + andi t1, $r_A, 0xff00 + or t0, t0, t2 + sll t1, t1, 8 + or $r_A, t0, t1 +# endif #endif jr $r_ra move $r_ret, zero @@ -80,8 +92,16 @@ sk_load_half_positive: PTR_ADDU t1, $r_skb_data, offset lh $r_A, 0(t1) #ifdef CONFIG_CPU_LITTLE_ENDIAN +# if defined(__mips_isa_rev) && (__mips_isa_rev >= 2) wsbh t0, $r_A seh $r_A, t0 +# else + sll t0, $r_A, 24 + andi t1, $r_A, 0xff00 + sra t0, t0, 16 + srl t1, t1, 8 + or $r_A, t0, t1 +# endif #endif jr $r_ra move $r_ret, zero @@ -148,9 +168,22 @@ sk_load_byte_positive: NESTED(bpf_slow_path_word, (6 * SZREG), $r_sp) bpf_slow_path_common(4) #ifdef CONFIG_CPU_LITTLE_ENDIAN +# if defined(__mips_isa_rev) && (__mips_isa_rev >= 2) wsbh t0, $r_s0 jr $r_ra rotr $r_A, t0, 16 +# else + sll t0, $r_s0, 24 + srl t1, $r_s0, 24 + srl t2, $r_s0, 8 + or t0, t0, t1 + andi t2, t2, 0xff00 + andi t1, $r_s0, 0xff00 + or t0, t0, t2 + sll t1, t1, 8 + jr $r_ra + or $r_A, t0, t1 +# endif #else jr $r_ra move $r_A, $r_s0 @@ -161,8 +194,17 @@ NESTED(bpf_slow_path_word, (6 * SZREG), $r_sp) NESTED(bpf_slow_path_half, (6 * SZREG), $r_sp) bpf_slow_path_common(2) #ifdef CONFIG_CPU_LITTLE_ENDIAN +# if defined(__mips_isa_rev) && (__mips_isa_rev >= 2) jr $r_ra wsbh $r_A, $r_s0 +# else + sll t0, $r_s0, 8 + andi t1, $r_s0, 0xff00 + andi t0, t0, 0xff00 + srl t1, t1, 8 + jr $r_ra + or $r_A, t0, t1 +# endif #else jr $r_ra move $r_A, $r_s0 From 84dedd71cf3bc61cc65ca43a48f7252344a1bb68 Mon Sep 17 00:00:00 2001 From: Alexander Couzens Date: Sat, 19 Sep 2015 06:26:19 +0200 Subject: [PATCH 309/647] MIPS: ATH79: Set missing irq ack handler for ar7100-misc-intc irq chip The irq ack handler was forgotten while introducing OF support. Only ar71xx and ar933x based devices require it. Signed-off-by: Alexander Couzens Acked-by: Alban Bedel Acked-by: Thomas Gleixner Cc: linux-mips@linux-mips.org Cc: Rob Herring Cc: Pawel Moll Cc: Mark Rutland Cc: Ian Campbell Cc: Kumar Gala Cc: devicetree@vger.kernel.org Cc: Jason Cooper Cc: Marc Zyngier Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/11163/ Signed-off-by: Ralf Baechle --- arch/mips/ath79/irq.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/arch/mips/ath79/irq.c b/arch/mips/ath79/irq.c index 15ecb4831e12..bf1651cc7e64 100644 --- a/arch/mips/ath79/irq.c +++ b/arch/mips/ath79/irq.c @@ -293,8 +293,16 @@ static int __init ath79_misc_intc_of_init( return 0; } -IRQCHIP_DECLARE(ath79_misc_intc, "qca,ar7100-misc-intc", - ath79_misc_intc_of_init); + +static int __init ar7100_misc_intc_of_init( + struct device_node *node, struct device_node *parent) +{ + ath79_misc_irq_chip.irq_mask_ack = ar71xx_misc_irq_mask; + return ath79_misc_intc_of_init(node, parent); +} + +IRQCHIP_DECLARE(ar7100_misc_intc, "qca,ar7100-misc-intc", + ar7100_misc_intc_of_init); static int __init ar79_cpu_intc_of_init( struct device_node *node, struct device_node *parent) From 19446da415e0f01d56364b700fe984cda78bca50 Mon Sep 17 00:00:00 2001 From: Alexander Couzens Date: Sat, 19 Sep 2015 06:26:20 +0200 Subject: [PATCH 310/647] MIPS: ATH79: Add irq chip ar7240-misc-intc The ar7240 misc irq chip use ack handler instead of ack_mask handler. All new ath79 chips use the ar7240 misc irq chip Signed-off-by: Alexander Couzens Acked-by: Alban Bedel Acked-by: Thomas Gleixner Cc: linux-mips@linux-mips.org Cc: Rob Herring Cc: Pawel Moll Cc: Mark Rutland Cc: Ian Campbell Cc: Kumar Gala Cc: devicetree@vger.kernel.org Cc: Jason Cooper Cc: Marc Zyngier Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/11164/ Signed-off-by: Ralf Baechle --- .../qca,ath79-misc-intc.txt | 20 +++++++++++++++++-- arch/mips/ath79/irq.c | 10 ++++++++++ 2 files changed, 28 insertions(+), 2 deletions(-) diff --git a/Documentation/devicetree/bindings/interrupt-controller/qca,ath79-misc-intc.txt b/Documentation/devicetree/bindings/interrupt-controller/qca,ath79-misc-intc.txt index 391717a68f3b..ec96b1f01478 100644 --- a/Documentation/devicetree/bindings/interrupt-controller/qca,ath79-misc-intc.txt +++ b/Documentation/devicetree/bindings/interrupt-controller/qca,ath79-misc-intc.txt @@ -4,8 +4,8 @@ The MISC interrupt controller is a secondary controller for lower priority interrupt. Required Properties: -- compatible: has to be "qca,-cpu-intc", "qca,ar7100-misc-intc" - as fallback +- compatible: has to be "qca,-cpu-intc", "qca,ar7100-misc-intc" or + "qca,-cpu-intc", "qca,ar7240-misc-intc" - reg: Base address and size of the controllers memory area - interrupt-parent: phandle of the parent interrupt controller. - interrupts: Interrupt specifier for the controllers interrupt. @@ -13,6 +13,9 @@ Required Properties: - #interrupt-cells : Specifies the number of cells needed to encode interrupt source, should be 1 +Compatible fallback depends on the SoC. Use ar7100 for ar71xx and ar913x, +use ar7240 for all other SoCs. + Please refer to interrupts.txt in this directory for details of the common Interrupt Controllers bindings used by client devices. @@ -28,3 +31,16 @@ Example: interrupt-controller; #interrupt-cells = <1>; }; + +Another example: + + interrupt-controller@18060010 { + compatible = "qca,ar9331-misc-intc", qca,ar7240-misc-intc"; + reg = <0x18060010 0x4>; + + interrupt-parent = <&cpuintc>; + interrupts = <6>; + + interrupt-controller; + #interrupt-cells = <1>; + }; diff --git a/arch/mips/ath79/irq.c b/arch/mips/ath79/irq.c index bf1651cc7e64..eeb3953ed8ac 100644 --- a/arch/mips/ath79/irq.c +++ b/arch/mips/ath79/irq.c @@ -304,6 +304,16 @@ static int __init ar7100_misc_intc_of_init( IRQCHIP_DECLARE(ar7100_misc_intc, "qca,ar7100-misc-intc", ar7100_misc_intc_of_init); +static int __init ar7240_misc_intc_of_init( + struct device_node *node, struct device_node *parent) +{ + ath79_misc_irq_chip.irq_ack = ar724x_misc_irq_ack; + return ath79_misc_intc_of_init(node, parent); +} + +IRQCHIP_DECLARE(ar7240_misc_intc, "qca,ar7240-misc-intc", + ar7240_misc_intc_of_init); + static int __init ar79_cpu_intc_of_init( struct device_node *node, struct device_node *parent) { From 2b9dbef272b63c561aab0a5be34fd428f7b710f5 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 15 Sep 2015 10:07:04 -0400 Subject: [PATCH 311/647] Btrfs: keep dropped roots in cache until transaction commit MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When dropping a snapshot we need to account for the qgroup changes. If we drop the snapshot in all one go then the backref code will fail to find blocks from the snapshot we dropped since it won't be able to find the root in the fs root cache. This can lead to us failing to find refs from other roots that pointed at blocks in the now deleted root. To handle this we need to not remove the fs roots from the cache until after we process the qgroup operations. Do this by adding dropped roots to a list on the transaction, and letting the transaction remove the roots at the same time it drops the commit roots. This will keep all of the backref searching code in sync properly, and fixes a problem Mark was seeing with snapshot delete and qgroups. Thanks, Signed-off-by: Josef Bacik Tested-by: Holger Hoffstätte Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 2 +- fs/btrfs/transaction.c | 32 ++++++++++++++++++++++++++++++++ fs/btrfs/transaction.h | 5 ++++- 3 files changed, 37 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 3686315bbd66..9f9604201333 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -8665,7 +8665,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root, } if (test_bit(BTRFS_ROOT_IN_RADIX, &root->state)) { - btrfs_drop_and_free_fs_root(tree_root->fs_info, root); + btrfs_add_dropped_root(trans, root); } else { free_extent_buffer(root->node); free_extent_buffer(root->commit_root); diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 68ad89e23713..a2d6f7bcef6c 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -117,6 +117,18 @@ static noinline void switch_commit_roots(struct btrfs_transaction *trans, btrfs_unpin_free_ino(root); clear_btree_io_tree(&root->dirty_log_pages); } + + /* We can free old roots now. */ + spin_lock(&trans->dropped_roots_lock); + while (!list_empty(&trans->dropped_roots)) { + root = list_first_entry(&trans->dropped_roots, + struct btrfs_root, root_list); + list_del_init(&root->root_list); + spin_unlock(&trans->dropped_roots_lock); + btrfs_drop_and_free_fs_root(fs_info, root); + spin_lock(&trans->dropped_roots_lock); + } + spin_unlock(&trans->dropped_roots_lock); up_write(&fs_info->commit_root_sem); } @@ -255,11 +267,13 @@ loop: INIT_LIST_HEAD(&cur_trans->pending_ordered); INIT_LIST_HEAD(&cur_trans->dirty_bgs); INIT_LIST_HEAD(&cur_trans->io_bgs); + INIT_LIST_HEAD(&cur_trans->dropped_roots); mutex_init(&cur_trans->cache_write_mutex); cur_trans->num_dirty_bgs = 0; spin_lock_init(&cur_trans->dirty_bgs_lock); INIT_LIST_HEAD(&cur_trans->deleted_bgs); spin_lock_init(&cur_trans->deleted_bgs_lock); + spin_lock_init(&cur_trans->dropped_roots_lock); list_add_tail(&cur_trans->list, &fs_info->trans_list); extent_io_tree_init(&cur_trans->dirty_pages, fs_info->btree_inode->i_mapping); @@ -336,6 +350,24 @@ static int record_root_in_trans(struct btrfs_trans_handle *trans, } +void btrfs_add_dropped_root(struct btrfs_trans_handle *trans, + struct btrfs_root *root) +{ + struct btrfs_transaction *cur_trans = trans->transaction; + + /* Add ourselves to the transaction dropped list */ + spin_lock(&cur_trans->dropped_roots_lock); + list_add_tail(&root->root_list, &cur_trans->dropped_roots); + spin_unlock(&cur_trans->dropped_roots_lock); + + /* Make sure we don't try to update the root at commit time */ + spin_lock(&root->fs_info->fs_roots_radix_lock); + radix_tree_tag_clear(&root->fs_info->fs_roots_radix, + (unsigned long)root->root_key.objectid, + BTRFS_ROOT_TRANS_TAG); + spin_unlock(&root->fs_info->fs_roots_radix_lock); +} + int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans, struct btrfs_root *root) { diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index edc2fbc262d7..87964bf8892d 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -65,6 +65,7 @@ struct btrfs_transaction { struct list_head switch_commits; struct list_head dirty_bgs; struct list_head io_bgs; + struct list_head dropped_roots; u64 num_dirty_bgs; /* @@ -76,6 +77,7 @@ struct btrfs_transaction { spinlock_t dirty_bgs_lock; struct list_head deleted_bgs; spinlock_t deleted_bgs_lock; + spinlock_t dropped_roots_lock; struct btrfs_delayed_ref_root delayed_refs; int aborted; int dirty_bg_run; @@ -216,5 +218,6 @@ int btrfs_transaction_blocked(struct btrfs_fs_info *info); int btrfs_transaction_in_commit(struct btrfs_fs_info *info); void btrfs_put_transaction(struct btrfs_transaction *transaction); void btrfs_apply_pending_changes(struct btrfs_fs_info *fs_info); - +void btrfs_add_dropped_root(struct btrfs_trans_handle *trans, + struct btrfs_root *root); #endif From 2f6f31363cb7890784458d7805140687b4de5b59 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Thu, 17 Sep 2015 17:49:20 +0100 Subject: [PATCH 312/647] MIPS: cpu-features: Add cpu_has_ftlb Add cpu_has_ftlb, which specifies that an FTLB is present in addition to the VTLB, probed based on whether Config.MT == 4 (rather than 1 for standard JTLB). This is necessary since MIPS release 6 removes Config4.MMUExtDef, so the presence of the FTLB fields in Config4 must be determined from Config.MT instead. Signed-off-by: James Hogan Cc: Markos Chandras Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/11159/ Signed-off-by: Ralf Baechle --- arch/mips/include/asm/cpu-features.h | 3 +++ arch/mips/include/asm/cpu.h | 1 + arch/mips/include/asm/mipsregs.h | 2 ++ arch/mips/kernel/cpu-probe.c | 8 +++++--- 4 files changed, 11 insertions(+), 3 deletions(-) diff --git a/arch/mips/include/asm/cpu-features.h b/arch/mips/include/asm/cpu-features.h index 9801ac982655..fe67f12ac239 100644 --- a/arch/mips/include/asm/cpu-features.h +++ b/arch/mips/include/asm/cpu-features.h @@ -20,6 +20,9 @@ #ifndef cpu_has_tlb #define cpu_has_tlb (cpu_data[0].options & MIPS_CPU_TLB) #endif +#ifndef cpu_has_ftlb +#define cpu_has_ftlb (cpu_data[0].options & MIPS_CPU_FTLB) +#endif #ifndef cpu_has_tlbinv #define cpu_has_tlbinv (cpu_data[0].options & MIPS_CPU_TLBINV) #endif diff --git a/arch/mips/include/asm/cpu.h b/arch/mips/include/asm/cpu.h index cd89e9855775..82ad15f11049 100644 --- a/arch/mips/include/asm/cpu.h +++ b/arch/mips/include/asm/cpu.h @@ -385,6 +385,7 @@ enum cpu_type_enum { #define MIPS_CPU_CDMM 0x4000000000ull /* CPU has Common Device Memory Map */ #define MIPS_CPU_BP_GHIST 0x8000000000ull /* R12K+ Branch Prediction Global History */ #define MIPS_CPU_SP 0x10000000000ull /* Small (1KB) page support */ +#define MIPS_CPU_FTLB 0x20000000000ull /* CPU has Fixed-page-size TLB */ /* * CPU ASE encodings diff --git a/arch/mips/include/asm/mipsregs.h b/arch/mips/include/asm/mipsregs.h index d3cd8eac81e3..c64781cf649f 100644 --- a/arch/mips/include/asm/mipsregs.h +++ b/arch/mips/include/asm/mipsregs.h @@ -487,6 +487,8 @@ /* Bits specific to the MIPS32/64 PRA. */ #define MIPS_CONF_MT (_ULCAST_(7) << 7) +#define MIPS_CONF_MT_TLB (_ULCAST_(1) << 7) +#define MIPS_CONF_MT_FTLB (_ULCAST_(4) << 7) #define MIPS_CONF_AR (_ULCAST_(7) << 10) #define MIPS_CONF_AT (_ULCAST_(3) << 13) #define MIPS_CONF_M (_ULCAST_(1) << 31) diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c index 571a8e6ea5bd..397551cf2e7b 100644 --- a/arch/mips/kernel/cpu-probe.c +++ b/arch/mips/kernel/cpu-probe.c @@ -410,16 +410,18 @@ static int set_ftlb_enable(struct cpuinfo_mips *c, int enable) static inline unsigned int decode_config0(struct cpuinfo_mips *c) { unsigned int config0; - int isa; + int isa, mt; config0 = read_c0_config(); /* * Look for Standard TLB or Dual VTLB and FTLB */ - if ((((config0 & MIPS_CONF_MT) >> 7) == 1) || - (((config0 & MIPS_CONF_MT) >> 7) == 4)) + mt = config0 & MIPS_CONF_MT; + if (mt == MIPS_CONF_MT_TLB) c->options |= MIPS_CPU_TLB; + else if (mt == MIPS_CONF_MT_FTLB) + c->options |= MIPS_CPU_TLB | MIPS_CPU_FTLB; isa = (config0 & MIPS_CONF_AT) >> 13; switch (isa) { From 43d104db596977a8fddc1e71245859a7fe85a658 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Thu, 17 Sep 2015 17:49:21 +0100 Subject: [PATCH 313/647] MIPS: Fix FTLB detection for R6 R6 removed the Config4.MMUExtDef field, with the low 16 bits only allowed to contain FTLB fields, and commit e87569cd6c57 ("MIPS: cpu-probe: Fix VTLB/FTLB configuration for R6") updated the probing of this field to assume an FTLB is always present for R6. However the FTLB may still be absent. The presence of those fields is actually specified by the MMU type in the Config.MT field, so use that (the new cpu_has_ftlb) to determine whether the FTLB is actually present. Fixes: e87569cd6c57 ("MIPS: cpu-probe: Fix VTLB/FTLB configuration for R6") Signed-off-by: James Hogan Cc: Ralf Baechle Cc: Markos Chandras Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/11160/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/cpu-probe.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c index 397551cf2e7b..09a51d091941 100644 --- a/arch/mips/kernel/cpu-probe.c +++ b/arch/mips/kernel/cpu-probe.c @@ -561,15 +561,18 @@ static inline unsigned int decode_config4(struct cpuinfo_mips *c) if (cpu_has_tlb) { if (((config4 & MIPS_CONF4_IE) >> 29) == 2) c->options |= MIPS_CPU_TLBINV; + /* - * This is a bit ugly. R6 has dropped that field from - * config4 and the only valid configuration is VTLB+FTLB so - * set a good value for mmuextdef for that case. + * R6 has dropped the MMUExtDef field from config4. + * On R6 the fields always describe the FTLB, and only if it is + * present according to Config.MT. */ - if (cpu_has_mips_r6) + if (!cpu_has_mips_r6) + mmuextdef = config4 & MIPS_CONF4_MMUEXTDEF; + else if (cpu_has_ftlb) mmuextdef = MIPS_CONF4_MMUEXTDEF_VTLBSIZEEXT; else - mmuextdef = config4 & MIPS_CONF4_MMUEXTDEF; + mmuextdef = 0; switch (mmuextdef) { case MIPS_CONF4_MMUEXTDEF_MMUSIZEEXT: From e0ddde9d44e37fbc21ce893553094ecf1a633ab5 Mon Sep 17 00:00:00 2001 From: Steve French Date: Tue, 22 Sep 2015 09:29:38 -0500 Subject: [PATCH 314/647] disabling oplocks/leases via module parm enable_oplocks broken for SMB3 leases (oplocks) were always requested for SMB2/SMB3 even when oplocks disabled in the cifs.ko module. Signed-off-by: Steve French Reviewed-by: Chandrika Srinivasan CC: Stable --- fs/cifs/smb2ops.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index df91bcf56d67..18da19f4f811 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -50,9 +50,13 @@ change_conf(struct TCP_Server_Info *server) break; default: server->echoes = true; - server->oplocks = true; + if (enable_oplocks) { + server->oplocks = true; + server->oplock_credits = 1; + } else + server->oplocks = false; + server->echo_credits = 1; - server->oplock_credits = 1; } server->credits -= server->echo_credits + server->oplock_credits; return 0; From 98ce94c8df762d413b3ecb849e2b966b21606d04 Mon Sep 17 00:00:00 2001 From: Peter Seiderer Date: Thu, 17 Sep 2015 21:40:12 +0200 Subject: [PATCH 315/647] cifs: use server timestamp for ntlmv2 authentication Linux cifs mount with ntlmssp against an Mac OS X (Yosemite 10.10.5) share fails in case the clocks differ more than +/-2h: digest-service: digest-request: od failed with 2 proto=ntlmv2 digest-service: digest-request: kdc failed with -1561745592 proto=ntlmv2 Fix this by (re-)using the given server timestamp for the ntlmv2 authentication (as Windows 7 does). A related problem was also reported earlier by Namjae Jaen (see below): Windows machine has extended security feature which refuse to allow authentication when there is time difference between server time and client time when ntlmv2 negotiation is used. This problem is prevalent in embedded enviornment where system time is set to default 1970. Modern servers send the server timestamp in the TargetInfo Av_Pair structure in the challenge message [see MS-NLMP 2.2.2.1] In [MS-NLMP 3.1.5.1.2] it is explicitly mentioned that the client must use the server provided timestamp if present OR current time if it is not Reported-by: Namjae Jeon Signed-off-by: Peter Seiderer Signed-off-by: Steve French CC: Stable --- fs/cifs/cifsencrypt.c | 53 +++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 51 insertions(+), 2 deletions(-) diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index aa0dc2573374..afa09fce8151 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -444,6 +444,48 @@ find_domain_name(struct cifs_ses *ses, const struct nls_table *nls_cp) return 0; } +/* Server has provided av pairs/target info in the type 2 challenge + * packet and we have plucked it and stored within smb session. + * We parse that blob here to find the server given timestamp + * as part of ntlmv2 authentication (or local current time as + * default in case of failure) + */ +static __le64 +find_timestamp(struct cifs_ses *ses) +{ + unsigned int attrsize; + unsigned int type; + unsigned int onesize = sizeof(struct ntlmssp2_name); + unsigned char *blobptr; + unsigned char *blobend; + struct ntlmssp2_name *attrptr; + + if (!ses->auth_key.len || !ses->auth_key.response) + return 0; + + blobptr = ses->auth_key.response; + blobend = blobptr + ses->auth_key.len; + + while (blobptr + onesize < blobend) { + attrptr = (struct ntlmssp2_name *) blobptr; + type = le16_to_cpu(attrptr->type); + if (type == NTLMSSP_AV_EOL) + break; + blobptr += 2; /* advance attr type */ + attrsize = le16_to_cpu(attrptr->length); + blobptr += 2; /* advance attr size */ + if (blobptr + attrsize > blobend) + break; + if (type == NTLMSSP_AV_TIMESTAMP) { + if (attrsize == sizeof(u64)) + return *((__le64 *)blobptr); + } + blobptr += attrsize; /* advance attr value */ + } + + return cpu_to_le64(cifs_UnixTimeToNT(CURRENT_TIME)); +} + static int calc_ntlmv2_hash(struct cifs_ses *ses, char *ntlmv2_hash, const struct nls_table *nls_cp) { @@ -641,6 +683,7 @@ setup_ntlmv2_rsp(struct cifs_ses *ses, const struct nls_table *nls_cp) struct ntlmv2_resp *ntlmv2; char ntlmv2_hash[16]; unsigned char *tiblob = NULL; /* target info blob */ + __le64 rsp_timestamp; if (ses->server->negflavor == CIFS_NEGFLAVOR_EXTENDED) { if (!ses->domainName) { @@ -659,6 +702,12 @@ setup_ntlmv2_rsp(struct cifs_ses *ses, const struct nls_table *nls_cp) } } + /* Must be within 5 minutes of the server (or in range +/-2h + * in case of Mac OS X), so simply carry over server timestamp + * (as Windows 7 does) + */ + rsp_timestamp = find_timestamp(ses); + baselen = CIFS_SESS_KEY_SIZE + sizeof(struct ntlmv2_resp); tilen = ses->auth_key.len; tiblob = ses->auth_key.response; @@ -675,8 +724,8 @@ setup_ntlmv2_rsp(struct cifs_ses *ses, const struct nls_table *nls_cp) (ses->auth_key.response + CIFS_SESS_KEY_SIZE); ntlmv2->blob_signature = cpu_to_le32(0x00000101); ntlmv2->reserved = 0; - /* Must be within 5 minutes of the server */ - ntlmv2->time = cpu_to_le64(cifs_UnixTimeToNT(CURRENT_TIME)); + ntlmv2->time = rsp_timestamp; + get_random_bytes(&ntlmv2->client_chal, sizeof(ntlmv2->client_chal)); ntlmv2->reserved2 = 0; From fc57a7c68020dcf954428869eafd934c0ab1536f Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Sun, 20 Sep 2015 16:32:04 -0700 Subject: [PATCH 316/647] x86/paravirt: Replace the paravirt nop with a bona fide empty function PARAVIRT_ADJUST_EXCEPTION_FRAME generates this code (using nmi as an example, trimmed for readability): ff 15 00 00 00 00 callq *0x0(%rip) # 2796 2792: R_X86_64_PC32 pv_irq_ops+0x2c That's a call through a function pointer to regular C function that does nothing on native boots, but that function isn't protected against kprobes, isn't marked notrace, and is certainly not guaranteed to preserve any registers if the compiler is feeling perverse. This is bad news for a CLBR_NONE operation. Of course, if everything works correctly, once paravirt ops are patched, it gets nopped out, but what if we hit this code before paravirt ops are patched in? This can potentially cause breakage that is very difficult to debug. A more subtle failure is possible here, too: if _paravirt_nop uses the stack at all (even just to push RBP), it will overwrite the "NMI executing" variable if it's called in the NMI prologue. The Xen case, perhaps surprisingly, is fine, because it's already written in asm. Fix all of the cases that default to paravirt_nop (including adjust_exception_frame) with a big hammer: replace paravirt_nop with an asm function that is just a ret instruction. The Xen case may have other problems, so document them. This is part of a fix for some random crashes that Sasha saw. Reported-and-tested-by: Sasha Levin Signed-off-by: Andy Lutomirski Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/8f5d2ba295f9d73751c33d97fda03e0495d9ade0.1442791737.git.luto@kernel.org Signed-off-by: Thomas Gleixner --- arch/x86/entry/entry_64.S | 11 +++++++++++ arch/x86/kernel/paravirt.c | 16 ++++++++++++---- 2 files changed, 23 insertions(+), 4 deletions(-) diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S index d3033183ed70..404ca97c4715 100644 --- a/arch/x86/entry/entry_64.S +++ b/arch/x86/entry/entry_64.S @@ -1128,7 +1128,18 @@ END(error_exit) /* Runs on exception stack */ ENTRY(nmi) + /* + * Fix up the exception frame if we're on Xen. + * PARAVIRT_ADJUST_EXCEPTION_FRAME is guaranteed to push at most + * one value to the stack on native, so it may clobber the rdx + * scratch slot, but it won't clobber any of the important + * slots past it. + * + * Xen is a different story, because the Xen frame itself overlaps + * the "NMI executing" variable. + */ PARAVIRT_ADJUST_EXCEPTION_FRAME + /* * We allow breakpoints in NMIs. If a breakpoint occurs, then * the iretq it performs will take us out of NMI context. diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index f68e48f5f6c2..c2130aef3f9d 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -41,10 +41,18 @@ #include #include -/* nop stub */ -void _paravirt_nop(void) -{ -} +/* + * nop stub, which must not clobber anything *including the stack* to + * avoid confusing the entry prologues. + */ +extern void _paravirt_nop(void); +asm (".pushsection .entry.text, \"ax\"\n" + ".global _paravirt_nop\n" + "_paravirt_nop:\n\t" + "ret\n\t" + ".size _paravirt_nop, . - _paravirt_nop\n\t" + ".type _paravirt_nop, @function\n\t" + ".popsection"); /* identity function, which can be inlined */ u32 _paravirt_ident_32(u32 x) From 83c133cf11fb0e68a51681447e372489f052d40e Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Sun, 20 Sep 2015 16:32:05 -0700 Subject: [PATCH 317/647] x86/nmi/64: Fix a paravirt stack-clobbering bug in the NMI code The NMI entry code that switches to the normal kernel stack needs to be very careful not to clobber any extra stack slots on the NMI stack. The code is fine under the assumption that SWAPGS is just a normal instruction, but that assumption isn't really true. Use SWAPGS_UNSAFE_STACK instead. This is part of a fix for some random crashes that Sasha saw. Fixes: 9b6e6a8334d5 ("x86/nmi/64: Switch stacks on userspace NMI entry") Reported-and-tested-by: Sasha Levin Signed-off-by: Andy Lutomirski Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/974bc40edffdb5c2950a5c4977f821a446b76178.1442791737.git.luto@kernel.org Signed-off-by: Thomas Gleixner --- arch/x86/entry/entry_64.S | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S index 404ca97c4715..055a01de7c8d 100644 --- a/arch/x86/entry/entry_64.S +++ b/arch/x86/entry/entry_64.S @@ -1190,9 +1190,12 @@ ENTRY(nmi) * we don't want to enable interrupts, because then we'll end * up in an awkward situation in which IRQs are on but NMIs * are off. + * + * We also must not push anything to the stack before switching + * stacks lest we corrupt the "NMI executing" variable. */ - SWAPGS + SWAPGS_UNSAFE_STACK cld movq %rsp, %rdx movq PER_CPU_VAR(cpu_current_top_of_stack), %rsp From 8811191fdf7ed02ee07cb8469428158572d355a2 Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Tue, 22 Sep 2015 21:20:22 +0200 Subject: [PATCH 318/647] ASoC: pxa: pxa2xx-ac97: fix dma requestor lines PCM receive and transmit DMA requestor lines were reverted, breaking the PCM playback interface for PXA platforms using the sound/soc/ variant instead of the sound/arm variant. The commit below shows the inversion in the requestor lines. Fixes: d65a14587a9b ("ASoC: pxa: use snd_dmaengine_dai_dma_data") Signed-off-by: Robert Jarzmik Signed-off-by: Mark Brown Cc: stable@vger.kernel.org --- sound/soc/pxa/pxa2xx-ac97.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/pxa/pxa2xx-ac97.c b/sound/soc/pxa/pxa2xx-ac97.c index 1f6054650991..9e4b04e0fbd1 100644 --- a/sound/soc/pxa/pxa2xx-ac97.c +++ b/sound/soc/pxa/pxa2xx-ac97.c @@ -49,7 +49,7 @@ static struct snd_ac97_bus_ops pxa2xx_ac97_ops = { .reset = pxa2xx_ac97_cold_reset, }; -static unsigned long pxa2xx_ac97_pcm_stereo_in_req = 12; +static unsigned long pxa2xx_ac97_pcm_stereo_in_req = 11; static struct snd_dmaengine_dai_dma_data pxa2xx_ac97_pcm_stereo_in = { .addr = __PREG(PCDR), .addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES, @@ -57,7 +57,7 @@ static struct snd_dmaengine_dai_dma_data pxa2xx_ac97_pcm_stereo_in = { .filter_data = &pxa2xx_ac97_pcm_stereo_in_req, }; -static unsigned long pxa2xx_ac97_pcm_stereo_out_req = 11; +static unsigned long pxa2xx_ac97_pcm_stereo_out_req = 12; static struct snd_dmaengine_dai_dma_data pxa2xx_ac97_pcm_stereo_out = { .addr = __PREG(PCDR), .addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES, From b838b39e930aa1cfd099ea82ac40ed6d6413af26 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Tue, 22 Sep 2015 17:03:54 -0500 Subject: [PATCH 319/647] PCI: Clear IORESOURCE_UNSET when clipping a bridge window c770cb4cb505 ("PCI: Mark invalid BARs as unassigned") sets IORESOURCE_UNSET if we fail to claim a resource. If we tried to claim a bridge window, failed, clipped the window, and tried to claim the clipped window, we failed again because of IORESOURCE_UNSET: pci_bus 0000:00: root bus resource [mem 0xc0000000-0xffffffff window] pci 0000:00:01.0: can't claim BAR 15 [mem 0xbdf00000-0xddefffff 64bit pref]: no compatible bridge window pci 0000:00:01.0: [mem size 0x20000000 64bit pref] clipped to [mem size 0x1df00000 64bit pref] pci 0000:00:01.0: bridge window [mem size 0x1df00000 64bit pref] pci 0000:00:01.0: can't claim BAR 15 [mem size 0x1df00000 64bit pref]: no address assigned The 00:01.0 window started as [mem 0xbdf00000-0xddefffff 64bit pref]. That starts before the host bridge window [mem 0xc0000000-0xffffffff window], so we clipped the 00:01.0 window to [mem 0xc0000000-0xddefffff 64bit pref]. But we left it marked IORESOURCE_UNSET, so the second claim failed when it should have succeeded. This means downstream devices will also fail for lack of resources, e.g., in the bugzilla below, radeon 0000:01:00.0: Fatal error during GPU init Clear IORESOURCE_UNSET when we clip a bridge window. Also clear IORESOURCE_UNSET in our copy of the unclipped window so we can see exactly what the original window was and how it now fits inside the upstream window. Fixes: c770cb4cb505 ("PCI: Mark invalid BARs as unassigned") Link: https://bugzilla.kernel.org/show_bug.cgi?id=85491#c47 Based-on-patch-by: Lorenzo Pieralisi Based-on-patch-by: Yinghai Lu Tested-by: Lorenzo Pieralisi Signed-off-by: Bjorn Helgaas Reviewed-by: Lorenzo Pieralisi Acked-by: Yinghai Lu CC: stable@vger.kernel.org # v4.1+ --- drivers/pci/bus.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c index 6fbd3f2b5992..d3346d23963b 100644 --- a/drivers/pci/bus.c +++ b/drivers/pci/bus.c @@ -256,6 +256,8 @@ bool pci_bus_clip_resource(struct pci_dev *dev, int idx) res->start = start; res->end = end; + res->flags &= ~IORESOURCE_UNSET; + orig_res.flags &= ~IORESOURCE_UNSET; dev_printk(KERN_DEBUG, &dev->dev, "%pR clipped to %pR\n", &orig_res, res); From 834e465bba38f2768747bccb5f00e951e72d2bf5 Mon Sep 17 00:00:00 2001 From: Kinglong Mee Date: Tue, 22 Sep 2015 06:54:47 +0800 Subject: [PATCH 320/647] NFS: Skip checking ds_cinfo.buckets when lseg's commit_through_mds is set When lseg's commit_through_mds is set, pnfs client always WARN once in nfs_direct_select_verf after checking ds_cinfo.nbuckets. nfs should use the DS verf except commit_through_mds is set for layout segment where nbuckets is zero. [17844.666094] ------------[ cut here ]------------ [17844.667071] WARNING: CPU: 0 PID: 21758 at /root/source/linux-pnfs/fs/nfs/direct.c:174 nfs_direct_select_verf+0x5a/0x70 [nfs]() [17844.668650] Modules linked in: nfs_layout_nfsv41_files(OE) nfsv4(OE) nfs(OE) fscache(E) nfsd(OE) xfs libcrc32c btrfs ppdev coretemp crct10dif_pclmul auth_rpcgss crc32_pclmul crc32c_intel nfs_acl ghash_clmulni_intel lockd vmw_balloon xor vmw_vmci grace raid6_pq shpchp sunrpc parport_pc i2c_piix4 parport vmwgfx drm_kms_helper ttm drm serio_raw mptspi e1000 scsi_transport_spi mptscsih mptbase ata_generic pata_acpi [last unloaded: fscache] [17844.686676] CPU: 0 PID: 21758 Comm: kworker/0:1 Tainted: G W OE 4.3.0-rc1-pnfs+ #245 [17844.687352] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 05/20/2014 [17844.698502] Workqueue: nfsiod rpc_async_release [sunrpc] [17844.699212] 0000000000000009 0000000043e58010 ffff8800454fbc10 ffffffff813680c4 [17844.699990] ffff8800454fbc48 ffffffff8108b49d ffff88004eb20000 ffff88004eb20000 [17844.700844] ffff880062e26000 0000000000000000 0000000000000001 ffff8800454fbc58 [17844.701637] Call Trace: [17844.725252] [] dump_stack+0x19/0x25 [17844.732693] [] warn_slowpath_common+0x7d/0xb0 [17844.733855] [] warn_slowpath_null+0x1a/0x20 [17844.735015] [] nfs_direct_select_verf+0x5a/0x70 [nfs] [17844.735999] [] nfs_direct_set_hdr_verf+0x23/0x90 [nfs] [17844.736846] [] nfs_direct_write_completion+0x227/0x260 [nfs] [17844.737782] [] nfs_pgio_release+0x1c/0x20 [nfs] [17844.738597] [] pnfs_generic_rw_release+0x23/0x30 [nfsv4] [17844.739486] [] rpc_free_task+0x2a/0x70 [sunrpc] [17844.740326] [] rpc_async_release+0x15/0x20 [sunrpc] [17844.741173] [] process_one_work+0x21c/0x4c0 [17844.741984] [] ? process_one_work+0x16d/0x4c0 [17844.742837] [] worker_thread+0x4a/0x440 [17844.743639] [] ? process_one_work+0x4c0/0x4c0 [17844.744399] [] ? process_one_work+0x4c0/0x4c0 [17844.745176] [] kthread+0xf5/0x110 [17844.745927] [] ? kthread_create_on_node+0x240/0x240 [17844.747105] [] ret_from_fork+0x3f/0x70 [17844.747856] [] ? kthread_create_on_node+0x240/0x240 [17844.748642] ---[ end trace 336a2845d42b83f0 ]--- Signed-off-by: Kinglong Mee Signed-off-by: Trond Myklebust --- fs/nfs/direct.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 38678d9a5cc4..4b1d08f56aba 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -166,8 +166,11 @@ nfs_direct_select_verf(struct nfs_direct_req *dreq, struct nfs_writeverf *verfp = &dreq->verf; #ifdef CONFIG_NFS_V4_1 - if (ds_clp) { - /* pNFS is in use, use the DS verf */ + /* + * pNFS is in use, use the DS verf except commit_through_mds is set + * for layout segment where nbuckets is zero. + */ + if (ds_clp && dreq->ds_cinfo.nbuckets > 0) { if (commit_idx >= 0 && commit_idx < dreq->ds_cinfo.nbuckets) verfp = &dreq->ds_cinfo.buckets[commit_idx].direct_verf; else From ac5be6b47e8bd25b62bed2c82cda7398999f59e9 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Tue, 22 Sep 2015 14:58:49 -0700 Subject: [PATCH 321/647] userfaultfd: revert "userfaultfd: waitqueue: add nr wake parameter to __wake_up_locked_key" This reverts commit 51360155eccb907ff8635bd10fc7de876408c2e0 and adapts fs/userfaultfd.c to use the old version of that function. It didn't look robust to call __wake_up_common with "nr == 1" when we absolutely require wakeall semantics, but we've full control of what we insert in the two waitqueue heads of the blocked userfaults. No exclusive waitqueue risks to be inserted into those two waitqueue heads so we can as well stick to "nr == 1" of the old code and we can rely purely on the fact no waitqueue inserted in one of the two waitqueue heads we must enforce as wakeall, has wait->flags WQ_FLAG_EXCLUSIVE set. Signed-off-by: Andrea Arcangeli Cc: Dr. David Alan Gilbert Cc: Michael Ellerman Cc: Shuah Khan Cc: Thierry Reding Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/userfaultfd.c | 8 ++++---- include/linux/wait.h | 5 ++--- kernel/sched/wait.c | 7 +++---- net/sunrpc/sched.c | 2 +- 4 files changed, 10 insertions(+), 12 deletions(-) diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index f9aeb40a7197..50311703135b 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -467,8 +467,8 @@ static int userfaultfd_release(struct inode *inode, struct file *file) * the fault_*wqh. */ spin_lock(&ctx->fault_pending_wqh.lock); - __wake_up_locked_key(&ctx->fault_pending_wqh, TASK_NORMAL, 0, &range); - __wake_up_locked_key(&ctx->fault_wqh, TASK_NORMAL, 0, &range); + __wake_up_locked_key(&ctx->fault_pending_wqh, TASK_NORMAL, &range); + __wake_up_locked_key(&ctx->fault_wqh, TASK_NORMAL, &range); spin_unlock(&ctx->fault_pending_wqh.lock); wake_up_poll(&ctx->fd_wqh, POLLHUP); @@ -650,10 +650,10 @@ static void __wake_userfault(struct userfaultfd_ctx *ctx, spin_lock(&ctx->fault_pending_wqh.lock); /* wake all in the range and autoremove */ if (waitqueue_active(&ctx->fault_pending_wqh)) - __wake_up_locked_key(&ctx->fault_pending_wqh, TASK_NORMAL, 0, + __wake_up_locked_key(&ctx->fault_pending_wqh, TASK_NORMAL, range); if (waitqueue_active(&ctx->fault_wqh)) - __wake_up_locked_key(&ctx->fault_wqh, TASK_NORMAL, 0, range); + __wake_up_locked_key(&ctx->fault_wqh, TASK_NORMAL, range); spin_unlock(&ctx->fault_pending_wqh.lock); } diff --git a/include/linux/wait.h b/include/linux/wait.h index d3d077228d4c..1e1bf9f963a9 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -147,8 +147,7 @@ __remove_wait_queue(wait_queue_head_t *head, wait_queue_t *old) typedef int wait_bit_action_f(struct wait_bit_key *); void __wake_up(wait_queue_head_t *q, unsigned int mode, int nr, void *key); -void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, int nr, - void *key); +void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, void *key); void __wake_up_sync_key(wait_queue_head_t *q, unsigned int mode, int nr, void *key); void __wake_up_locked(wait_queue_head_t *q, unsigned int mode, int nr); void __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr); @@ -180,7 +179,7 @@ wait_queue_head_t *bit_waitqueue(void *, int); #define wake_up_poll(x, m) \ __wake_up(x, TASK_NORMAL, 1, (void *) (m)) #define wake_up_locked_poll(x, m) \ - __wake_up_locked_key((x), TASK_NORMAL, 1, (void *) (m)) + __wake_up_locked_key((x), TASK_NORMAL, (void *) (m)) #define wake_up_interruptible_poll(x, m) \ __wake_up(x, TASK_INTERRUPTIBLE, 1, (void *) (m)) #define wake_up_interruptible_sync_poll(x, m) \ diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c index 272d9322bc5d..052e02672d12 100644 --- a/kernel/sched/wait.c +++ b/kernel/sched/wait.c @@ -106,10 +106,9 @@ void __wake_up_locked(wait_queue_head_t *q, unsigned int mode, int nr) } EXPORT_SYMBOL_GPL(__wake_up_locked); -void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, int nr, - void *key) +void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, void *key) { - __wake_up_common(q, mode, nr, 0, key); + __wake_up_common(q, mode, 1, 0, key); } EXPORT_SYMBOL_GPL(__wake_up_locked_key); @@ -284,7 +283,7 @@ void abort_exclusive_wait(wait_queue_head_t *q, wait_queue_t *wait, if (!list_empty(&wait->task_list)) list_del_init(&wait->task_list); else if (waitqueue_active(q)) - __wake_up_locked_key(q, mode, 1, key); + __wake_up_locked_key(q, mode, key); spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL(abort_exclusive_wait); diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index b140c092d226..337ca851a350 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -297,7 +297,7 @@ static int rpc_complete_task(struct rpc_task *task) clear_bit(RPC_TASK_ACTIVE, &task->tk_runstate); ret = atomic_dec_and_test(&task->tk_count); if (waitqueue_active(wq)) - __wake_up_locked_key(wq, TASK_NORMAL, 1, &k); + __wake_up_locked_key(wq, TASK_NORMAL, &k); spin_unlock_irqrestore(&wq->lock, flags); return ret; } From d0a871141d07929b559f5eae9c3fc4b63d16866b Mon Sep 17 00:00:00 2001 From: Thierry Reding Date: Tue, 22 Sep 2015 14:58:52 -0700 Subject: [PATCH 322/647] userfaultfd: selftests: vm: pick up sanitized kernel headers Add the usr/include subdirectory of the top-level tree to the include path, and make sure to include headers without relative paths to make sure the sanitized headers get picked up. Otherwise the compiler will not be able to find the linux/compiler.h header included by the non- sanitized include/uapi/linux/userfaultfd.h. While at it, make sure to only hardcode the syscall numbers on x86 and PowerPC if they haven't been properly picked up from the headers. Signed-off-by: Thierry Reding Acked-by: Michael Ellerman Cc: Shuah Khan Signed-off-by: Andrea Arcangeli Cc: Dr. David Alan Gilbert Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- tools/testing/selftests/vm/Makefile | 2 +- tools/testing/selftests/vm/userfaultfd.c | 4 +++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/vm/Makefile b/tools/testing/selftests/vm/Makefile index d36fab7d8ebd..949e275f11a6 100644 --- a/tools/testing/selftests/vm/Makefile +++ b/tools/testing/selftests/vm/Makefile @@ -1,6 +1,6 @@ # Makefile for vm selftests -CFLAGS = -Wall +CFLAGS = -Wall -I ../../../../usr/include $(EXTRA_CFLAGS) BINARIES = compaction_test BINARIES += hugepage-mmap BINARIES += hugepage-shm diff --git a/tools/testing/selftests/vm/userfaultfd.c b/tools/testing/selftests/vm/userfaultfd.c index 7c1d958857d2..59d145f20918 100644 --- a/tools/testing/selftests/vm/userfaultfd.c +++ b/tools/testing/selftests/vm/userfaultfd.c @@ -64,8 +64,9 @@ #include #include #include -#include "../../../../include/uapi/linux/userfaultfd.h" +#include +#ifndef __NR_userfaultfd #ifdef __x86_64__ #define __NR_userfaultfd 323 #elif defined(__i386__) @@ -77,6 +78,7 @@ #else #error "missing __NR_userfaultfd definition" #endif +#endif static unsigned long nr_cpus, nr_pages, nr_pages_per_cpu, page_size; From 67f6a029b2ccf3399783a0ff2f812666f290d94f Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Tue, 22 Sep 2015 14:58:55 -0700 Subject: [PATCH 323/647] userfaultfd: selftest: headers fixup Depend on "make headers_install" to create proper headers to include and provide syscall numbers. Signed-off-by: Andrea Arcangeli Cc: Dr. David Alan Gilbert Cc: Michael Ellerman Cc: Shuah Khan Cc: Thierry Reding Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- tools/testing/selftests/vm/Makefile | 7 +++++-- tools/testing/selftests/vm/userfaultfd.c | 10 ---------- 2 files changed, 5 insertions(+), 12 deletions(-) diff --git a/tools/testing/selftests/vm/Makefile b/tools/testing/selftests/vm/Makefile index 949e275f11a6..3c53cac15de1 100644 --- a/tools/testing/selftests/vm/Makefile +++ b/tools/testing/selftests/vm/Makefile @@ -12,8 +12,11 @@ BINARIES += userfaultfd all: $(BINARIES) %: %.c $(CC) $(CFLAGS) -o $@ $^ -lrt -userfaultfd: userfaultfd.c - $(CC) $(CFLAGS) -O2 -o $@ $^ -lpthread +userfaultfd: userfaultfd.c ../../../../usr/include/linux/kernel.h + $(CC) $(CFLAGS) -O2 -o $@ $< -lpthread + +../../../../usr/include/linux/kernel.h: + make -C ../../../.. headers_install TEST_PROGS := run_vmtests TEST_FILES := $(BINARIES) diff --git a/tools/testing/selftests/vm/userfaultfd.c b/tools/testing/selftests/vm/userfaultfd.c index 59d145f20918..a9e0b9143f06 100644 --- a/tools/testing/selftests/vm/userfaultfd.c +++ b/tools/testing/selftests/vm/userfaultfd.c @@ -67,18 +67,8 @@ #include #ifndef __NR_userfaultfd -#ifdef __x86_64__ -#define __NR_userfaultfd 323 -#elif defined(__i386__) -#define __NR_userfaultfd 374 -#elif defined(__powewrpc__) -#define __NR_userfaultfd 364 -#elif defined(__s390__) -#define __NR_userfaultfd 355 -#else #error "missing __NR_userfaultfd definition" #endif -#endif static unsigned long nr_cpus, nr_pages, nr_pages_per_cpu, page_size; From 56ed8f169e225dce1f9e40f6eee2e2dabe7d06fc Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 22 Sep 2015 14:58:58 -0700 Subject: [PATCH 324/647] userfaultfd: selftest: only warn if __NR_userfaultfd is undefined If __NR_userfaultfd is not yet defined by the arch, warn but still build and run the userfaultfd selftest successfully. Signed-off-by: Michael Ellerman Signed-off-by: Andrea Arcangeli Cc: Dr. David Alan Gilbert Cc: Shuah Khan Cc: Thierry Reding Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- tools/testing/selftests/vm/userfaultfd.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/vm/userfaultfd.c b/tools/testing/selftests/vm/userfaultfd.c index a9e0b9143f06..0671ae1d1052 100644 --- a/tools/testing/selftests/vm/userfaultfd.c +++ b/tools/testing/selftests/vm/userfaultfd.c @@ -66,9 +66,7 @@ #include #include -#ifndef __NR_userfaultfd -#error "missing __NR_userfaultfd definition" -#endif +#ifdef __NR_userfaultfd static unsigned long nr_cpus, nr_pages, nr_pages_per_cpu, page_size; @@ -631,3 +629,15 @@ int main(int argc, char **argv) nr_pages, nr_pages_per_cpu); return userfaultfd_stress(); } + +#else /* __NR_userfaultfd */ + +#warning "missing __NR_userfaultfd definition" + +int main(void) +{ + printf("skip: Skipping userfaultfd test (missing __NR_userfaultfd)\n"); + return 0; +} + +#endif /* __NR_userfaultfd */ From 1f5fee2cf232f9fac05b65f21107d2cf3c32092c Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Tue, 22 Sep 2015 14:59:00 -0700 Subject: [PATCH 325/647] userfaultfd: selftest: avoid my_bcmp false positives with powerpc Keep a non-zero placeholder after the count, for the my_bcmp comparison of the page against the zeropage. The lockless increment between 255 to 256 against a lockless my_bcmp could otherwise return false positives on ppc32le. Signed-off-by: Andrea Arcangeli Tested-by: Michael Ellerman Cc: Dr. David Alan Gilbert Cc: Shuah Khan Cc: Thierry Reding Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- tools/testing/selftests/vm/userfaultfd.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/vm/userfaultfd.c b/tools/testing/selftests/vm/userfaultfd.c index 0671ae1d1052..10897092823d 100644 --- a/tools/testing/selftests/vm/userfaultfd.c +++ b/tools/testing/selftests/vm/userfaultfd.c @@ -465,6 +465,14 @@ static int userfaultfd_stress(void) *area_mutex(area_src, nr) = (pthread_mutex_t) PTHREAD_MUTEX_INITIALIZER; count_verify[nr] = *area_count(area_src, nr) = 1; + /* + * In the transition between 255 to 256, powerpc will + * read out of order in my_bcmp and see both bytes as + * zero, so leave a placeholder below always non-zero + * after the count, to avoid my_bcmp to trigger false + * positives. + */ + *(area_count(area_src, nr) + 1) = 1; } pipefd = malloc(sizeof(int) * nr_cpus * 2); @@ -610,8 +618,8 @@ int main(int argc, char **argv) fprintf(stderr, "Usage: \n"), exit(1); nr_cpus = sysconf(_SC_NPROCESSORS_ONLN); page_size = sysconf(_SC_PAGE_SIZE); - if ((unsigned long) area_count(NULL, 0) + sizeof(unsigned long long) > - page_size) + if ((unsigned long) area_count(NULL, 0) + sizeof(unsigned long long) * 2 + > page_size) fprintf(stderr, "Impossible to run this test\n"), exit(2); nr_pages_per_cpu = atol(argv[1]) * 1024*1024 / page_size / nr_cpus; From a5932bf5737f0b5caf6deaa92b062e4fe66cf5b2 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Tue, 22 Sep 2015 14:59:03 -0700 Subject: [PATCH 326/647] userfaultfd: selftest: return an error if BOUNCE_VERIFY fails This will report the error in the exit code, in addition of the fprintf. Signed-off-by: Andrea Arcangeli Cc: Dr. David Alan Gilbert Cc: Michael Ellerman Cc: Shuah Khan Cc: Thierry Reding Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- tools/testing/selftests/vm/userfaultfd.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/vm/userfaultfd.c b/tools/testing/selftests/vm/userfaultfd.c index 10897092823d..174f2fc8d257 100644 --- a/tools/testing/selftests/vm/userfaultfd.c +++ b/tools/testing/selftests/vm/userfaultfd.c @@ -422,7 +422,7 @@ static int userfaultfd_stress(void) struct uffdio_register uffdio_register; struct uffdio_api uffdio_api; unsigned long cpu; - int uffd_flags; + int uffd_flags, err; unsigned long userfaults[nr_cpus]; if (posix_memalign(&area, page_size, nr_pages * page_size)) { @@ -499,6 +499,7 @@ static int userfaultfd_stress(void) pthread_attr_init(&attr); pthread_attr_setstacksize(&attr, 16*1024*1024); + err = 0; while (bounces--) { unsigned long expected_ioctls; @@ -583,8 +584,9 @@ static int userfaultfd_stress(void) area_dst + nr * page_size, sizeof(pthread_mutex_t))) { fprintf(stderr, - "error mutex 2 %lu\n", + "error mutex %lu\n", nr); + err = 1; bounces = 0; } if (*area_count(area_dst, nr) != count_verify[nr]) { @@ -593,6 +595,7 @@ static int userfaultfd_stress(void) *area_count(area_src, nr), count_verify[nr], nr); + err = 1; bounces = 0; } } @@ -609,7 +612,7 @@ static int userfaultfd_stress(void) printf("\n"); } - return 0; + return err; } int main(int argc, char **argv) From 5dd01be14565df814408327971775f36e55bf5e3 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Tue, 22 Sep 2015 14:59:06 -0700 Subject: [PATCH 327/647] userfaultfd: selftest: don't error out if pthread_mutex_t isn't identical On ppc big endian this check fails, the mutex doesn't necessarily need to be identical for all pages after pthread_mutex_lock/unlock cycles. The count verification (outside of the pthread_mutex_t structure) suffices and that is retained. Signed-off-by: Andrea Arcangeli Cc: Dr. David Alan Gilbert Cc: Michael Ellerman Cc: Shuah Khan Cc: Thierry Reding Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- tools/testing/selftests/vm/userfaultfd.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/tools/testing/selftests/vm/userfaultfd.c b/tools/testing/selftests/vm/userfaultfd.c index 174f2fc8d257..d77ed41b2094 100644 --- a/tools/testing/selftests/vm/userfaultfd.c +++ b/tools/testing/selftests/vm/userfaultfd.c @@ -580,15 +580,6 @@ static int userfaultfd_stress(void) /* verification */ if (bounces & BOUNCE_VERIFY) { for (nr = 0; nr < nr_pages; nr++) { - if (my_bcmp(area_dst, - area_dst + nr * page_size, - sizeof(pthread_mutex_t))) { - fprintf(stderr, - "error mutex %lu\n", - nr); - err = 1; - bounces = 0; - } if (*area_count(area_dst, nr) != count_verify[nr]) { fprintf(stderr, "error area_count %Lu %Lu %lu\n", From 09f7298100ea9767324298ab0c7979f6d7463183 Mon Sep 17 00:00:00 2001 From: "Dr. David Alan Gilbert" Date: Tue, 22 Sep 2015 14:59:09 -0700 Subject: [PATCH 328/647] userfaultfd: register uapi generic syscall (aarch64) Add the userfaultfd syscalls to uapi asm-generic, it was tested with postcopy live migration on aarch64 with both 4k and 64k pagesize kernels. Signed-off-by: Dr. David Alan Gilbert Signed-off-by: Andrea Arcangeli Cc: Michael Ellerman Cc: Shuah Khan Cc: Thierry Reding Cc: Mathieu Desnoyers Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/uapi/asm-generic/unistd.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/include/uapi/asm-generic/unistd.h b/include/uapi/asm-generic/unistd.h index 8da542a2874d..ee124009e12a 100644 --- a/include/uapi/asm-generic/unistd.h +++ b/include/uapi/asm-generic/unistd.h @@ -709,17 +709,19 @@ __SYSCALL(__NR_memfd_create, sys_memfd_create) __SYSCALL(__NR_bpf, sys_bpf) #define __NR_execveat 281 __SC_COMP(__NR_execveat, sys_execveat, compat_sys_execveat) -#define __NR_membarrier 282 +#define __NR_userfaultfd 282 +__SYSCALL(__NR_userfaultfd, sys_userfaultfd) +#define __NR_membarrier 283 __SYSCALL(__NR_membarrier, sys_membarrier) #undef __NR_syscalls -#define __NR_syscalls 283 +#define __NR_syscalls 284 /* * All syscalls below here should go away really, * these are provided for both review and as a porting * help for the C library version. -* + * * Last chance: are any of these important enough to * enable by default? */ From 8a04446ab0cf4f35d9f583cd6adcbf7c534e4995 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Tue, 22 Sep 2015 14:59:12 -0700 Subject: [PATCH 329/647] mm, dax: VMA with vm_ops->pfn_mkwrite wants to be write-notified For VM_PFNMAP and VM_MIXEDMAP we use vm_ops->pfn_mkwrite instead of vm_ops->page_mkwrite to notify abort write access. This means we want vma->vm_page_prot to be write-protected if the VMA provides this vm_ops. A theoretical scenario that will cause these missed events is: On writable mapping with vm_ops->pfn_mkwrite, but without vm_ops->page_mkwrite: read fault followed by write access to the pfn. Writable pte will be set up on read fault and write fault will not be generated. I found it examining Dave's complaint on generic/080: http://lkml.kernel.org/g/20150831233803.GO3902@dastard Although I don't think it's the reason. It shouldn't be a problem for ext2/ext4 as they provide both pfn_mkwrite and page_mkwrite. [akpm@linux-foundation.org: add local vm_ops to avoid 80-cols mess] Signed-off-by: Kirill A. Shutemov Cc: Yigal Korman Acked-by: Boaz Harrosh Cc: Matthew Wilcox Cc: Jan Kara Cc: Dave Chinner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mmap.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/mm/mmap.c b/mm/mmap.c index c739d6db7193..79bcc9f92e48 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1490,13 +1490,14 @@ SYSCALL_DEFINE1(old_mmap, struct mmap_arg_struct __user *, arg) int vma_wants_writenotify(struct vm_area_struct *vma) { vm_flags_t vm_flags = vma->vm_flags; + const struct vm_operations_struct *vm_ops = vma->vm_ops; /* If it was private or non-writable, the write bit is already clear */ if ((vm_flags & (VM_WRITE|VM_SHARED)) != ((VM_WRITE|VM_SHARED))) return 0; /* The backer wishes to know when pages are first written to? */ - if (vma->vm_ops && vma->vm_ops->page_mkwrite) + if (vm_ops && (vm_ops->page_mkwrite || vm_ops->pfn_mkwrite)) return 1; /* The open routine did something to the protections that pgprot_modify From 3aaa76e125c1dd58c9b599baa8c6021896874c12 Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Tue, 22 Sep 2015 14:59:14 -0700 Subject: [PATCH 330/647] mm: migrate: hugetlb: putback destination hugepage to active list Since commit bcc54222309c ("mm: hugetlb: introduce page_huge_active") each hugetlb page maintains its active flag to avoid a race condition betwe= en multiple calls of isolate_huge_page(), but current kernel doesn't set the f= lag on a hugepage allocated by migration because the proper putback routine isn= 't called. This means that users could still encounter the race referred to by bcc54222309c in this special case, so this patch fixes it. Fixes: bcc54222309c ("mm: hugetlb: introduce page_huge_active") Signed-off-by: Naoya Horiguchi Cc: Michal Hocko Cc: Andi Kleen Cc: Hugh Dickins Cc: [4.1.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/migrate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/migrate.c b/mm/migrate.c index c3cb566af3e2..7452a00bbb50 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1075,7 +1075,7 @@ out: if (rc != MIGRATEPAGE_SUCCESS && put_new_page) put_new_page(new_hpage, private); else - put_page(new_hpage); + putback_active_hugepage(new_hpage); if (result) { if (rc) From 769a8089c1fd2fe94c13e66fe6e03d7820953ee3 Mon Sep 17 00:00:00 2001 From: Andrey Ryabinin Date: Tue, 22 Sep 2015 14:59:17 -0700 Subject: [PATCH 331/647] x86, efi, kasan: #undef memset/memcpy/memmove per arch In not-instrumented code KASAN replaces instrumented memset/memcpy/memmove with not-instrumented analogues __memset/__memcpy/__memove. However, on x86 the EFI stub is not linked with the kernel. It uses not-instrumented mem*() functions from arch/x86/boot/compressed/string.c So we don't replace them with __mem*() variants in EFI stub. On ARM64 the EFI stub is linked with the kernel, so we should replace mem*() functions with __mem*(), because the EFI stub runs before KASAN sets up early shadow. So let's move these #undef mem* into arch's asm/efi.h which is also included by the EFI stub. Also, this will fix the warning in 32-bit build reported by kbuild test robot: efi-stub-helper.c:599:2: warning: implicit declaration of function 'memcpy' [akpm@linux-foundation.org: use 80 cols in comment] Signed-off-by: Andrey Ryabinin Reported-by: Fengguang Wu Cc: Will Deacon Cc: Catalin Marinas Cc: Matt Fleming Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/include/asm/efi.h | 10 ++++++++++ drivers/firmware/efi/libstub/efistub.h | 4 ---- 2 files changed, 10 insertions(+), 4 deletions(-) diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index 155162ea0e00..ab5f1d447ef9 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -86,6 +86,16 @@ extern u64 asmlinkage efi_call(void *fp, ...); extern void __iomem *__init efi_ioremap(unsigned long addr, unsigned long size, u32 type, u64 attribute); +/* + * CONFIG_KASAN may redefine memset to __memset. __memset function is present + * only in kernel binary. Since the EFI stub linked into a separate binary it + * doesn't have __memset(). So we should use standard memset from + * arch/x86/boot/compressed/string.c. The same applies to memcpy and memmove. + */ +#undef memcpy +#undef memset +#undef memmove + #endif /* CONFIG_X86_32 */ extern struct efi_scratch efi_scratch; diff --git a/drivers/firmware/efi/libstub/efistub.h b/drivers/firmware/efi/libstub/efistub.h index e334a01cf92f..6b6548fda089 100644 --- a/drivers/firmware/efi/libstub/efistub.h +++ b/drivers/firmware/efi/libstub/efistub.h @@ -5,10 +5,6 @@ /* error code which can't be mistaken for valid address */ #define EFI_ERROR (~0UL) -#undef memcpy -#undef memset -#undef memmove - void efi_char16_printk(efi_system_table_t *, efi_char16_t *); efi_status_t efi_open_volume(efi_system_table_t *sys_table_arg, void *__image, From d046b770c9fc36ccb19c27afdb8322220108cbc7 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Tue, 22 Sep 2015 14:59:20 -0700 Subject: [PATCH 332/647] lib/iommu-common.c: do not try to deref a null iommu->lazy_flush() pointer when n < pool->hint The check for invoking iommu->lazy_flush() from iommu_tbl_range_alloc() has to be refactored so that we only call ->lazy_flush() if it is non-null. I had a sparc kernel that was crashing when I was trying to process some very large perf.data files- the crash happens when the scsi driver calls into dma_4v_map_sg and thus the iommu_tbl_range_alloc(). Signed-off-by: Sowmini Varadhan Cc: Benjamin Herrenschmidt Cc: Guenter Roeck Cc: David S. Miller Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- lib/iommu-common.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/iommu-common.c b/lib/iommu-common.c index ff19f66d3f7f..b1c93e94ca7a 100644 --- a/lib/iommu-common.c +++ b/lib/iommu-common.c @@ -21,8 +21,7 @@ static DEFINE_PER_CPU(unsigned int, iommu_hash_common); static inline bool need_flush(struct iommu_map_table *iommu) { - return (iommu->lazy_flush != NULL && - (iommu->flags & IOMMU_NEED_FLUSH) != 0); + return ((iommu->flags & IOMMU_NEED_FLUSH) != 0); } static inline void set_flush(struct iommu_map_table *iommu) @@ -211,7 +210,8 @@ unsigned long iommu_tbl_range_alloc(struct device *dev, goto bail; } } - if (n < pool->hint || need_flush(iommu)) { + if (iommu->lazy_flush && + (n < pool->hint || need_flush(iommu))) { clear_flush(iommu); iommu->lazy_flush(iommu); } From d5028f9f7d8de5c375c52b98976b6f310e73398f Mon Sep 17 00:00:00 2001 From: Vladimir Davydov Date: Tue, 22 Sep 2015 14:59:20 -0700 Subject: [PATCH 333/647] vmscan: fix sane_reclaim helper for legacy memcg The sane_reclaim() helper is supposed to return false for memcg reclaim if the legacy hierarchy is used, because the latter lacks dirty throttling mechanism, and so it did before it was accidentally broken by commit 33398cf2f360c ("memcg: export struct mem_cgroup"). Fix it. Fixes: 33398cf2f360c ("memcg: export struct mem_cgroup") Signed-off-by: Vladimir Davydov Acked-by: Tejun Heo Acked-by: Michal Hocko Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmscan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index 2d978b28a410..7f63a9381f71 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -175,7 +175,7 @@ static bool sane_reclaim(struct scan_control *sc) if (!memcg) return true; #ifdef CONFIG_CGROUP_WRITEBACK - if (memcg->css.cgroup) + if (cgroup_on_dfl(memcg->css.cgroup)) return true; #endif return false; From 7a07b503bf249986a1eeef0351d66cac0d8bf721 Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Tue, 22 Sep 2015 14:59:20 -0700 Subject: [PATCH 334/647] membarrier: clean up selftest We don't need to specify an explicit rule in the Makefile, the implicit one will do the same. The "__EXPORTED_HEADERS__" define is not needed, because we build the test against the installed kernel headers, not the in-tree kernel headers. Re-use "$(TEST_PROGS)" in the clean target rather than spelling the executable name twice. Include rather than the rather specific . Include rather than . In both cases, the former header is located in a standard location and includes the latter. Signed-off-by: Mathieu Desnoyers Acked-by: Michael Ellerman Cc: Pranith Kumar Cc: Shuah Khan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- tools/testing/selftests/membarrier/Makefile | 7 +++---- tools/testing/selftests/membarrier/membarrier_test.c | 5 +---- 2 files changed, 4 insertions(+), 8 deletions(-) diff --git a/tools/testing/selftests/membarrier/Makefile b/tools/testing/selftests/membarrier/Makefile index 877a50355d7f..a1a97085847d 100644 --- a/tools/testing/selftests/membarrier/Makefile +++ b/tools/testing/selftests/membarrier/Makefile @@ -1,11 +1,10 @@ CFLAGS += -g -I../../../../usr/include/ -all: - $(CC) $(CFLAGS) membarrier_test.c -o membarrier_test - TEST_PROGS := membarrier_test +all: $(TEST_PROGS) + include ../lib.mk clean: - $(RM) membarrier_test + $(RM) $(TEST_PROGS) diff --git a/tools/testing/selftests/membarrier/membarrier_test.c b/tools/testing/selftests/membarrier/membarrier_test.c index dde312508007..535f0fef4d0b 100644 --- a/tools/testing/selftests/membarrier/membarrier_test.c +++ b/tools/testing/selftests/membarrier/membarrier_test.c @@ -1,9 +1,6 @@ #define _GNU_SOURCE -#define __EXPORTED_HEADERS__ - #include -#include -#include +#include #include #include #include From 012572d4fc2e4ddd5c8ec8614d51414ec6cae02a Mon Sep 17 00:00:00 2001 From: Joseph Qi Date: Tue, 22 Sep 2015 14:59:20 -0700 Subject: [PATCH 335/647] ocfs2/dlm: fix deadlock when dispatch assert master The order of the following three spinlocks should be: dlm_domain_lock < dlm_ctxt->spinlock < dlm_lock_resource->spinlock But dlm_dispatch_assert_master() is called while holding dlm_ctxt->spinlock and dlm_lock_resource->spinlock, and then it calls dlm_grab() which will take dlm_domain_lock. Once another thread (for example, dlm_query_join_handler) has already taken dlm_domain_lock, and tries to take dlm_ctxt->spinlock deadlock happens. Signed-off-by: Joseph Qi Cc: Joel Becker Cc: Mark Fasheh Cc: "Junxiao Bi" Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ocfs2/dlm/dlmmaster.c | 9 ++++++--- fs/ocfs2/dlm/dlmrecovery.c | 8 ++++++-- 2 files changed, 12 insertions(+), 5 deletions(-) diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 46b8b2bbc95a..ee5aa4daaea0 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -1439,6 +1439,7 @@ int dlm_master_request_handler(struct o2net_msg *msg, u32 len, void *data, int found, ret; int set_maybe; int dispatch_assert = 0; + int dispatched = 0; if (!dlm_grab(dlm)) return DLM_MASTER_RESP_NO; @@ -1658,15 +1659,18 @@ send_response: mlog(ML_ERROR, "failed to dispatch assert master work\n"); response = DLM_MASTER_RESP_ERROR; dlm_lockres_put(res); - } else + } else { + dispatched = 1; __dlm_lockres_grab_inflight_worker(dlm, res); + } spin_unlock(&res->spinlock); } else { if (res) dlm_lockres_put(res); } - dlm_put(dlm); + if (!dispatched) + dlm_put(dlm); return response; } @@ -2090,7 +2094,6 @@ int dlm_dispatch_assert_master(struct dlm_ctxt *dlm, /* queue up work for dlm_assert_master_worker */ - dlm_grab(dlm); /* get an extra ref for the work item */ dlm_init_work_item(dlm, item, dlm_assert_master_worker, NULL); item->u.am.lockres = res; /* already have a ref */ /* can optionally ignore node numbers higher than this node */ diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index ce12e0b1a31f..3d90ad7ff91f 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -1694,6 +1694,7 @@ int dlm_master_requery_handler(struct o2net_msg *msg, u32 len, void *data, unsigned int hash; int master = DLM_LOCK_RES_OWNER_UNKNOWN; u32 flags = DLM_ASSERT_MASTER_REQUERY; + int dispatched = 0; if (!dlm_grab(dlm)) { /* since the domain has gone away on this @@ -1719,8 +1720,10 @@ int dlm_master_requery_handler(struct o2net_msg *msg, u32 len, void *data, dlm_put(dlm); /* sender will take care of this and retry */ return ret; - } else + } else { + dispatched = 1; __dlm_lockres_grab_inflight_worker(dlm, res); + } spin_unlock(&res->spinlock); } else { /* put.. incase we are not the master */ @@ -1730,7 +1733,8 @@ int dlm_master_requery_handler(struct o2net_msg *msg, u32 len, void *data, } spin_unlock(&dlm->spinlock); - dlm_put(dlm); + if (!dispatched) + dlm_put(dlm); return master; } From 08399efc631960c827cebcfe83ad7ed54ebc012b Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Mon, 21 Sep 2015 10:29:09 -0400 Subject: [PATCH 336/647] geneve: ensure ECN info is handled properly in all tx/rx paths Partially due to a pre-exising "thinko", the new metadata-based tx/rx paths were handling ECN propagation differently than the traditional tx/rx paths. This patch removes the "thinko" (involving multiple ip_hdr assignments) on the rx path and corrects the ECN handling on both the rx and tx paths. Signed-off-by: John W. Linville Reviewed-by: Jesse Gross Signed-off-by: David S. Miller --- drivers/net/geneve.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 8f1e9150341a..4ce8b7b90c7c 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -126,6 +126,8 @@ static void geneve_rx(struct geneve_sock *gs, struct sk_buff *skb) __be32 addr; int err; + iph = ip_hdr(skb); /* outer IP header... */ + if (gs->collect_md) { static u8 zero_vni[3]; @@ -133,7 +135,6 @@ static void geneve_rx(struct geneve_sock *gs, struct sk_buff *skb) addr = 0; } else { vni = gnvh->vni; - iph = ip_hdr(skb); /* Still outer IP header... */ addr = iph->saddr; } @@ -178,7 +179,6 @@ static void geneve_rx(struct geneve_sock *gs, struct sk_buff *skb) skb_reset_network_header(skb); - iph = ip_hdr(skb); /* Now inner IP header... */ err = IP_ECN_decapsulate(iph, skb); if (unlikely(err)) { @@ -626,6 +626,7 @@ static netdev_tx_t geneve_xmit(struct sk_buff *skb, struct net_device *dev) struct geneve_sock *gs = geneve->sock; struct ip_tunnel_info *info = NULL; struct rtable *rt = NULL; + const struct iphdr *iip; /* interior IP header */ struct flowi4 fl4; __u8 tos, ttl; __be16 sport; @@ -653,6 +654,8 @@ static netdev_tx_t geneve_xmit(struct sk_buff *skb, struct net_device *dev) sport = udp_flow_src_port(geneve->net, skb, 1, USHRT_MAX, true); skb_reset_mac_header(skb); + iip = ip_hdr(skb); + if (info) { const struct ip_tunnel_key *key = &info->key; u8 *opts = NULL; @@ -668,19 +671,16 @@ static netdev_tx_t geneve_xmit(struct sk_buff *skb, struct net_device *dev) if (unlikely(err)) goto err; - tos = key->tos; + tos = ip_tunnel_ecn_encap(key->tos, iip, skb); ttl = key->ttl; df = key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0; } else { - const struct iphdr *iip; /* interior IP header */ - udp_csum = false; err = geneve_build_skb(rt, skb, 0, geneve->vni, 0, NULL, udp_csum); if (unlikely(err)) goto err; - iip = ip_hdr(skb); tos = ip_tunnel_ecn_encap(fl4.flowi4_tos, iip, skb); ttl = geneve->ttl; if (!ttl && IN_MULTICAST(ntohl(fl4.daddr))) From 53adc9e83028d9e35b6408231ebaf62a94a16e4d Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 21 Sep 2015 21:42:59 +0100 Subject: [PATCH 337/647] net: dsa: actually force the speed on the CPU port Commit 54d792f257c6 ("net: dsa: Centralise global and port setup code into mv88e6xxx.") merged in the 4.2 merge window broke the link speed forcing for the CPU port of Marvell DSA switches. The original code was: /* MAC Forcing register: don't force link, speed, duplex * or flow control state to any particular values on physical * ports, but force the CPU port and all DSA ports to 1000 Mb/s * full duplex. */ if (dsa_is_cpu_port(ds, p) || ds->dsa_port_mask & (1 << p)) REG_WRITE(addr, 0x01, 0x003e); else REG_WRITE(addr, 0x01, 0x0003); but the new code does a read-modify-write: reg = _mv88e6xxx_reg_read(ds, REG_PORT(port), PORT_PCS_CTRL); if (dsa_is_cpu_port(ds, port) || ds->dsa_port_mask & (1 << port)) { reg |= PORT_PCS_CTRL_FORCE_LINK | PORT_PCS_CTRL_LINK_UP | PORT_PCS_CTRL_DUPLEX_FULL | PORT_PCS_CTRL_FORCE_DUPLEX; if (mv88e6xxx_6065_family(ds)) reg |= PORT_PCS_CTRL_100; else reg |= PORT_PCS_CTRL_1000; The link speed in the PCS control register is a two bit field. Forcing the link speed in this way doesn't ensure that the bit field is set to the correct value - on the hardware I have here, the speed bitfield remains set to 0x03, resulting in the speed not being forced to gigabit. We must clear both bits before forcing the link speed. Fixes: 54d792f257c6 ("net: dsa: Centralise global and port setup code into mv88e6xxx.") Signed-off-by: Russell King Acked-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/dsa/mv88e6xxx.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 6f13f7206762..f8baa897d1a0 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -2000,6 +2000,7 @@ static int mv88e6xxx_setup_port(struct dsa_switch *ds, int port) */ reg = _mv88e6xxx_reg_read(ds, REG_PORT(port), PORT_PCS_CTRL); if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) { + reg &= ~PORT_PCS_CTRL_UNFORCED; reg |= PORT_PCS_CTRL_FORCE_LINK | PORT_PCS_CTRL_LINK_UP | PORT_PCS_CTRL_DUPLEX_FULL | From ae5f2fb1d51fa128a460bcfbe3c56d7ab8bf6a43 Mon Sep 17 00:00:00 2001 From: Jesse Gross Date: Mon, 21 Sep 2015 20:21:20 -0700 Subject: [PATCH 338/647] openvswitch: Zero flows on allocation. When support for megaflows was introduced, OVS needed to start installing flows with a mask applied to them. Since masking is an expensive operation, OVS also had an optimization that would only take the parts of the flow keys that were covered by a non-zero mask. The values stored in the remaining pieces should not matter because they are masked out. While this works fine for the purposes of matching (which must always look at the mask), serialization to netlink can be problematic. Since the flow and the mask are serialized separately, the uninitialized portions of the flow can be encoded with whatever values happen to be present. In terms of functionality, this has little effect since these fields will be masked out by definition. However, it leaks kernel memory to userspace, which is a potential security vulnerability. It is also possible that other code paths could look at the masked key and get uninitialized data, although this does not currently appear to be an issue in practice. This removes the mask optimization for flows that are being installed. This was always intended to be the case as the mask optimizations were really targetting per-packet flow operations. Fixes: 03f0d916 ("openvswitch: Mega flow implementation") Signed-off-by: Jesse Gross Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/datapath.c | 4 ++-- net/openvswitch/flow_table.c | 23 ++++++++++++----------- net/openvswitch/flow_table.h | 2 +- 3 files changed, 15 insertions(+), 14 deletions(-) diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 6fbd2decb19e..b816ff871528 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -952,7 +952,7 @@ static int ovs_flow_cmd_new(struct sk_buff *skb, struct genl_info *info) if (error) goto err_kfree_flow; - ovs_flow_mask_key(&new_flow->key, &key, &mask); + ovs_flow_mask_key(&new_flow->key, &key, true, &mask); /* Extract flow identifier. */ error = ovs_nla_get_identifier(&new_flow->id, a[OVS_FLOW_ATTR_UFID], @@ -1080,7 +1080,7 @@ static struct sw_flow_actions *get_flow_actions(struct net *net, struct sw_flow_key masked_key; int error; - ovs_flow_mask_key(&masked_key, key, mask); + ovs_flow_mask_key(&masked_key, key, true, mask); error = ovs_nla_copy_actions(net, a, &masked_key, &acts, log); if (error) { OVS_NLERR(log, diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c index d22d8e948d0f..f2ea83ba4763 100644 --- a/net/openvswitch/flow_table.c +++ b/net/openvswitch/flow_table.c @@ -57,20 +57,21 @@ static u16 range_n_bytes(const struct sw_flow_key_range *range) } void ovs_flow_mask_key(struct sw_flow_key *dst, const struct sw_flow_key *src, - const struct sw_flow_mask *mask) + bool full, const struct sw_flow_mask *mask) { - const long *m = (const long *)((const u8 *)&mask->key + - mask->range.start); - const long *s = (const long *)((const u8 *)src + - mask->range.start); - long *d = (long *)((u8 *)dst + mask->range.start); + int start = full ? 0 : mask->range.start; + int len = full ? sizeof *dst : range_n_bytes(&mask->range); + const long *m = (const long *)((const u8 *)&mask->key + start); + const long *s = (const long *)((const u8 *)src + start); + long *d = (long *)((u8 *)dst + start); int i; - /* The memory outside of the 'mask->range' are not set since - * further operations on 'dst' only uses contents within - * 'mask->range'. + /* If 'full' is true then all of 'dst' is fully initialized. Otherwise, + * if 'full' is false the memory outside of the 'mask->range' is left + * uninitialized. This can be used as an optimization when further + * operations on 'dst' only use contents within 'mask->range'. */ - for (i = 0; i < range_n_bytes(&mask->range); i += sizeof(long)) + for (i = 0; i < len; i += sizeof(long)) *d++ = *s++ & *m++; } @@ -475,7 +476,7 @@ static struct sw_flow *masked_flow_lookup(struct table_instance *ti, u32 hash; struct sw_flow_key masked_key; - ovs_flow_mask_key(&masked_key, unmasked, mask); + ovs_flow_mask_key(&masked_key, unmasked, false, mask); hash = flow_hash(&masked_key, &mask->range); head = find_bucket(ti, hash); hlist_for_each_entry_rcu(flow, head, flow_table.node[ti->node_ver]) { diff --git a/net/openvswitch/flow_table.h b/net/openvswitch/flow_table.h index 616eda10d955..2dd9900f533d 100644 --- a/net/openvswitch/flow_table.h +++ b/net/openvswitch/flow_table.h @@ -86,5 +86,5 @@ struct sw_flow *ovs_flow_tbl_lookup_ufid(struct flow_table *, bool ovs_flow_cmp(const struct sw_flow *, const struct sw_flow_match *); void ovs_flow_mask_key(struct sw_flow_key *dst, const struct sw_flow_key *src, - const struct sw_flow_mask *mask); + bool full, const struct sw_flow_mask *mask); #endif /* flow_table.h */ From 23eedbc2435ddd226717603c4f3c8efec7bdbb4d Mon Sep 17 00:00:00 2001 From: Tobias Klauser Date: Tue, 22 Sep 2015 09:29:49 +0200 Subject: [PATCH 339/647] ch9200: Convert to use module_usb_driver Converts the ch9200 driver to use the module_usb_driver() macro which makes the code smaller and a bit simpler. Signed-off-by: Tobias Klauser Acked-by: Matthew Garrett Signed-off-by: David S. Miller --- drivers/net/usb/ch9200.c | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-) diff --git a/drivers/net/usb/ch9200.c b/drivers/net/usb/ch9200.c index cabb670eb55c..5e151e6a3e09 100644 --- a/drivers/net/usb/ch9200.c +++ b/drivers/net/usb/ch9200.c @@ -426,18 +426,7 @@ static struct usb_driver ch9200_driver = { .resume = usbnet_resume, }; -static int __init ch9200_init(void) -{ - return usb_register(&ch9200_driver); -} - -static void __exit ch9200_exit(void) -{ - usb_deregister(&ch9200_driver); -} - -module_init(ch9200_init); -module_exit(ch9200_exit); +module_usb_driver(ch9200_driver); MODULE_DESCRIPTION("QinHeng CH9200 USB Network device"); MODULE_LICENSE("GPL"); From 7def0f952eccdd0edb3c504f4dab35ee0d3aba1f Mon Sep 17 00:00:00 2001 From: Dmitriy Vyukov Date: Tue, 22 Sep 2015 10:51:52 +0200 Subject: [PATCH 340/647] lib: fix data race in rhashtable_rehash_one rhashtable_rehash_one() uses complex logic to update entry->next field, after INIT_RHT_NULLS_HEAD and NULLS_MARKER expansion: entry->next = 1 | ((base + off) << 1) This can be compiled along the lines of: entry->next = base + off entry->next <<= 1 entry->next |= 1 Which will break concurrent readers. NULLS value recomputation is not needed here, so just remove the complex logic. The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov Acked-by: Eric Dumazet Acked-by: Thomas Graf Acked-by: Herbert Xu Signed-off-by: David S. Miller --- lib/rhashtable.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/lib/rhashtable.c b/lib/rhashtable.c index cc0c69710dcf..a54ff8949f91 100644 --- a/lib/rhashtable.c +++ b/lib/rhashtable.c @@ -187,10 +187,7 @@ static int rhashtable_rehash_one(struct rhashtable *ht, unsigned int old_hash) head = rht_dereference_bucket(new_tbl->buckets[new_hash], new_tbl, new_hash); - if (rht_is_a_nulls(head)) - INIT_RHT_NULLS_HEAD(entry->next, ht, new_hash); - else - RCU_INIT_POINTER(entry->next, head); + RCU_INIT_POINTER(entry->next, head); rcu_assign_pointer(new_tbl->buckets[new_hash], entry); spin_unlock(new_bucket_lock); From fbd03513bf36c4e5c2942f436f05c8eb99a3cc9e Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Tue, 22 Sep 2015 11:28:14 +0200 Subject: [PATCH 341/647] net: dsa: Fix Marvell Egress Trailer check The Marvell Egress rx trailer check must be fixed to correctly detect bad bits in the third byte of the Eggress trailer as described in the Table 28 of the 88E6060 datasheet. The current code incorrectly omits to check the third byte and checks the fourth byte twice. Signed-off-by: Neil Armstrong Acked-by: Guenter Roeck Signed-off-by: David S. Miller --- net/dsa/tag_trailer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/dsa/tag_trailer.c b/net/dsa/tag_trailer.c index d25efc93d8f1..b6ca0890d018 100644 --- a/net/dsa/tag_trailer.c +++ b/net/dsa/tag_trailer.c @@ -78,7 +78,7 @@ static int trailer_rcv(struct sk_buff *skb, struct net_device *dev, trailer = skb_tail_pointer(skb) - 4; if (trailer[0] != 0x80 || (trailer[1] & 0xf8) != 0x00 || - (trailer[3] & 0xef) != 0x00 || trailer[3] != 0x00) + (trailer[2] & 0xef) != 0x00 || trailer[3] != 0x00) goto out_drop; source_port = trailer[1] & 7; From 41b578fb0e8b930f2470d3f673b0fa279e77a7b8 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Tue, 22 Sep 2015 12:15:54 -0700 Subject: [PATCH 342/647] drm/i915: workaround bad DSL readout v3 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On HSW at least (still testing other platforms, but should be harmless elsewhere), the DSL reg reads back as 0 when read around vblank start time. This ends up confusing the atomic start/end checking code, since it causes the update to appear as if it crossed a frame count boundary. Avoid the problem by making sure we don't return scanline_offset from the get_crtc_scanline function. In moving the code there, I add to add an additional delay since it could be called and have a legitimate 0 result for some time (depending on the pixel clock). v2: move hsw dsl read hack to get_crtc_scanline (Ville) v3: use break instead of goto (Ville) update comment with workaround details (Ville) References: https://bugs.freedesktop.org/show_bug.cgi?id=91579 Signed-off-by: Jesse Barnes Reviewed-by: Ville Syrjälä Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_irq.c | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 5a244ab9395b..39d73dbc1c47 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -639,6 +639,32 @@ static int __intel_get_crtc_scanline(struct intel_crtc *crtc) else position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN3; + /* + * On HSW, the DSL reg (0x70000) appears to return 0 if we + * read it just before the start of vblank. So try it again + * so we don't accidentally end up spanning a vblank frame + * increment, causing the pipe_update_end() code to squak at us. + * + * The nature of this problem means we can't simply check the ISR + * bit and return the vblank start value; nor can we use the scanline + * debug register in the transcoder as it appears to have the same + * problem. We may need to extend this to include other platforms, + * but so far testing only shows the problem on HSW. + */ + if (IS_HASWELL(dev) && !position) { + int i, temp; + + for (i = 0; i < 100; i++) { + udelay(1); + temp = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & + DSL_LINEMASK_GEN3; + if (temp != position) { + position = temp; + break; + } + } + } + /* * See update_scanline_offset() for the details on the * scanline_offset adjustment. From 21199f27b430576552b26210b3194a363d7f05cd Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Sep 2015 16:10:40 +0200 Subject: [PATCH 343/647] locking/lockdep: Fix hlock->pin_count reset on lock stack rebuilds Various people reported hitting the "unpinning an unpinned lock" warning. As it turns out there are 2 places where we take a lock out of the middle of a stack, and in those cases it would fail to preserve the pin_count when rebuilding the lock stack. Reported-by: Sasha Levin Reported-by: Tim Spriggs Tested-by: Sasha Levin Signed-off-by: Peter Zijlstra (Intel) Cc: Andrew Morton Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: davej@codemonkey.org.uk Link: http://lkml.kernel.org/r/20150916141040.GA11639@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- kernel/locking/lockdep.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 8acfbf773e06..4e49cc4c9952 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -3068,7 +3068,7 @@ static int __lock_is_held(struct lockdep_map *lock); static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, int trylock, int read, int check, int hardirqs_off, struct lockdep_map *nest_lock, unsigned long ip, - int references) + int references, int pin_count) { struct task_struct *curr = current; struct lock_class *class = NULL; @@ -3157,7 +3157,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, hlock->waittime_stamp = 0; hlock->holdtime_stamp = lockstat_clock(); #endif - hlock->pin_count = 0; + hlock->pin_count = pin_count; if (check && !mark_irqflags(curr, hlock)) return 0; @@ -3343,7 +3343,7 @@ found_it: hlock_class(hlock)->subclass, hlock->trylock, hlock->read, hlock->check, hlock->hardirqs_off, hlock->nest_lock, hlock->acquire_ip, - hlock->references)) + hlock->references, hlock->pin_count)) return 0; } @@ -3433,7 +3433,7 @@ found_it: hlock_class(hlock)->subclass, hlock->trylock, hlock->read, hlock->check, hlock->hardirqs_off, hlock->nest_lock, hlock->acquire_ip, - hlock->references)) + hlock->references, hlock->pin_count)) return 0; } @@ -3583,7 +3583,7 @@ void lock_acquire(struct lockdep_map *lock, unsigned int subclass, current->lockdep_recursion = 1; trace_lock_acquire(lock, subclass, trylock, read, check, nest_lock, ip); __lock_acquire(lock, subclass, trylock, read, check, - irqs_disabled_flags(flags), nest_lock, ip, 0); + irqs_disabled_flags(flags), nest_lock, ip, 0, 0); current->lockdep_recursion = 0; raw_local_irq_restore(flags); } From a7adb91b13c104e5ad950fbe1795aa2722f2ea0a Mon Sep 17 00:00:00 2001 From: Kristen Carlson Accardi Date: Tue, 22 Sep 2015 10:51:36 -0700 Subject: [PATCH 344/647] x86/cpufeatures: Correct spelling of the HWP_NOTIFY flag Because noitification just isn't right. Signed-off-by: Kristen Carlson Accardi Acked-by: Rafael J. Wysocki Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Cc: rjw@rjwysocki.net Link: http://lkml.kernel.org/r/1442944296-11737-1-git-send-email-kristen@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/cpufeature.h | 2 +- arch/x86/kernel/cpu/scattered.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index e6cf2ad350d1..9727b3b48bd1 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -193,7 +193,7 @@ #define X86_FEATURE_HW_PSTATE ( 7*32+ 8) /* AMD HW-PState */ #define X86_FEATURE_PROC_FEEDBACK ( 7*32+ 9) /* AMD ProcFeedbackInterface */ #define X86_FEATURE_HWP ( 7*32+ 10) /* "hwp" Intel HWP */ -#define X86_FEATURE_HWP_NOITFY ( 7*32+ 11) /* Intel HWP_NOTIFY */ +#define X86_FEATURE_HWP_NOTIFY ( 7*32+ 11) /* Intel HWP_NOTIFY */ #define X86_FEATURE_HWP_ACT_WINDOW ( 7*32+ 12) /* Intel HWP_ACT_WINDOW */ #define X86_FEATURE_HWP_EPP ( 7*32+13) /* Intel HWP_EPP */ #define X86_FEATURE_HWP_PKG_REQ ( 7*32+14) /* Intel HWP_PKG_REQ */ diff --git a/arch/x86/kernel/cpu/scattered.c b/arch/x86/kernel/cpu/scattered.c index 3d423a101fae..608fb26c7254 100644 --- a/arch/x86/kernel/cpu/scattered.c +++ b/arch/x86/kernel/cpu/scattered.c @@ -37,7 +37,7 @@ void init_scattered_cpuid_features(struct cpuinfo_x86 *c) { X86_FEATURE_PLN, CR_EAX, 4, 0x00000006, 0 }, { X86_FEATURE_PTS, CR_EAX, 6, 0x00000006, 0 }, { X86_FEATURE_HWP, CR_EAX, 7, 0x00000006, 0 }, - { X86_FEATURE_HWP_NOITFY, CR_EAX, 8, 0x00000006, 0 }, + { X86_FEATURE_HWP_NOTIFY, CR_EAX, 8, 0x00000006, 0 }, { X86_FEATURE_HWP_ACT_WINDOW, CR_EAX, 9, 0x00000006, 0 }, { X86_FEATURE_HWP_EPP, CR_EAX,10, 0x00000006, 0 }, { X86_FEATURE_HWP_PKG_REQ, CR_EAX,11, 0x00000006, 0 }, From 721a09f7393de6c28a07516dccd654c6e995944a Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Tue, 15 Sep 2015 14:28:54 +0200 Subject: [PATCH 345/647] drm/i915: Add primary plane to mask if it's visible This fixes the warnings like "plane A assertion failure, should be disabled but not" that on the initial modeset during boot. This can happen if the primary plane is enabled by the firmware, but inheriting it fails because the DMAR is active or for other reasons. Most likely caused by commit 36750f284b3a4f19b304fda1bb7d6e9e1275ea8d Author: Maarten Lankhorst Date: Mon Jun 1 12:49:54 2015 +0200 drm/i915: update plane state during init Reported-by: Andreas Reis Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91429 Reported-and-tested-by: Emil Renner Berthing Tested-by: Andreas Reis Reviewed-by: Daniel Vetter Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_display.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 8cc9264f7809..cf418be7d30a 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -15087,9 +15087,12 @@ static void readout_plane_state(struct intel_crtc *crtc, plane_state = to_intel_plane_state(p->base.state); - if (p->base.type == DRM_PLANE_TYPE_PRIMARY) + if (p->base.type == DRM_PLANE_TYPE_PRIMARY) { plane_state->visible = primary_get_hw_state(crtc); - else { + if (plane_state->visible) + crtc->base.state->plane_mask |= + 1 << drm_plane_index(&p->base); + } else { if (active) p->disable_plane(&p->base, &crtc->base); From d6eb71a6d2eda21c8cd7a4dcd6207a0d94eb6ae7 Mon Sep 17 00:00:00 2001 From: Vaibhav Jain Date: Wed, 23 Sep 2015 08:37:59 +0530 Subject: [PATCH 346/647] cxl: Fix lockdep warning while creating afu_err_buff attribute Presently a lockdep warning is reported during creation of afu_err_buff bin_attribute for the afu. This is caused due to the variable attr.key not pointing to a static class key, hence the function lockdep_init_map reports this warning: BUG: key not in .data! The patch fixes this issue by calling sysfs_attr_init on the attr_eb.attr structure before populating it with the afu_err_buff file details. This will populate the attr.key variable with a static class key so that lockdep_init_map stops complaining about the lockdep key not being static. Reported-by: Daniel Axtens Signed-off-by: Vaibhav Jain Acked-by: Ian Munsie Reviewed-by: Daniel Axtens Signed-off-by: Michael Ellerman --- drivers/misc/cxl/sysfs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/misc/cxl/sysfs.c b/drivers/misc/cxl/sysfs.c index 25868c2ec03e..02006f7109a8 100644 --- a/drivers/misc/cxl/sysfs.c +++ b/drivers/misc/cxl/sysfs.c @@ -592,6 +592,8 @@ int cxl_sysfs_afu_add(struct cxl_afu *afu) /* conditionally create the add the binary file for error info buffer */ if (afu->eb_len) { + sysfs_attr_init(&afu->attr_eb.attr); + afu->attr_eb.attr.name = "afu_err_buff"; afu->attr_eb.attr.mode = S_IRUGO; afu->attr_eb.size = afu->eb_len; From 500d701f336b2771d34e46da7875a4782515a652 Mon Sep 17 00:00:00 2001 From: Peng Tao Date: Tue, 22 Sep 2015 11:35:22 +0800 Subject: [PATCH 347/647] NFS41: make close wait for layoutreturn If we send a layoutreturn asynchronously before close, the close might reach server first and layoutreturn would fail with BADSTATEID because there is nothing keeping the layout stateid alive. Also do not pretend sending layoutreturn if we are not. Signed-off-by: Peng Tao Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 17 +++++++++++++++++ fs/nfs/pnfs.c | 35 +++++++++++++++++++++++++---------- fs/nfs/pnfs.h | 7 +++++++ 3 files changed, 49 insertions(+), 10 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index ef6b46e08ac6..f93b9cdb4934 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2658,6 +2658,15 @@ out: return err; } +static bool +nfs4_wait_on_layoutreturn(struct inode *inode, struct rpc_task *task) +{ + if (inode == NULL || !nfs_have_layout(inode)) + return false; + + return pnfs_wait_on_layoutreturn(inode, task); +} + struct nfs4_closedata { struct inode *inode; struct nfs4_state *state; @@ -2776,6 +2785,11 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) goto out_no_action; } + if (nfs4_wait_on_layoutreturn(inode, task)) { + nfs_release_seqid(calldata->arg.seqid); + goto out_wait; + } + if (calldata->arg.fmode == 0) task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE]; if (calldata->roc) @@ -5321,6 +5335,9 @@ static void nfs4_delegreturn_prepare(struct rpc_task *task, void *data) d_data = (struct nfs4_delegreturndata *)data; + if (nfs4_wait_on_layoutreturn(d_data->inode, task)) + return; + if (d_data->roc) pnfs_roc_get_barrier(d_data->inode, &d_data->roc_barrier); diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index ba1246433794..8abe27165ad0 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1104,20 +1104,15 @@ bool pnfs_roc(struct inode *ino) mark_lseg_invalid(lseg, &tmp_list); found = true; } - /* pnfs_prepare_layoutreturn() grabs lo ref and it will be put - * in pnfs_roc_release(). We don't really send a layoutreturn but - * still want others to view us like we are sending one! - * - * If pnfs_prepare_layoutreturn() fails, it means someone else is doing - * LAYOUTRETURN, so we proceed like there are no layouts to return. - * - * ROC in three conditions: + /* ROC in two conditions: * 1. there are ROC lsegs * 2. we don't send layoutreturn - * 3. no others are sending layoutreturn */ - if (found && !layoutreturn && pnfs_prepare_layoutreturn(lo)) + if (found && !layoutreturn) { + /* lo ref dropped in pnfs_roc_release() */ + pnfs_get_layout_hdr(lo); roc = true; + } out_noroc: spin_unlock(&ino->i_lock); @@ -1172,6 +1167,26 @@ void pnfs_roc_get_barrier(struct inode *ino, u32 *barrier) spin_unlock(&ino->i_lock); } +bool pnfs_wait_on_layoutreturn(struct inode *ino, struct rpc_task *task) +{ + struct nfs_inode *nfsi = NFS_I(ino); + struct pnfs_layout_hdr *lo; + bool sleep = false; + + /* we might not have grabbed lo reference. so need to check under + * i_lock */ + spin_lock(&ino->i_lock); + lo = nfsi->layout; + if (lo && test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) + sleep = true; + spin_unlock(&ino->i_lock); + + if (sleep) + rpc_sleep_on(&NFS_SERVER(ino)->roc_rpcwaitq, task, NULL); + + return sleep; +} + /* * Compare two layout segments for sorting into layout cache. * We want to preferentially return RW over RO layouts, so ensure those diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index 78c9351ff117..d1990e90e7a0 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -270,6 +270,7 @@ bool pnfs_roc(struct inode *ino); void pnfs_roc_release(struct inode *ino); void pnfs_roc_set_barrier(struct inode *ino, u32 barrier); void pnfs_roc_get_barrier(struct inode *ino, u32 *barrier); +bool pnfs_wait_on_layoutreturn(struct inode *ino, struct rpc_task *task); void pnfs_set_layoutcommit(struct inode *, struct pnfs_layout_segment *, loff_t); void pnfs_cleanup_layoutcommit(struct nfs4_layoutcommit_data *data); int pnfs_layoutcommit_inode(struct inode *inode, bool sync); @@ -639,6 +640,12 @@ pnfs_roc_get_barrier(struct inode *ino, u32 *barrier) { } +static inline bool +pnfs_wait_on_layoutreturn(struct inode *ino, struct rpc_task *task) +{ + return false; +} + static inline void set_pnfs_layoutdriver(struct nfs_server *s, const struct nfs_fh *mntfh, u32 id) { From cd67d226ebd909d239d2c6e5a6abd6e2a338d1cd Mon Sep 17 00:00:00 2001 From: Jani Nikula Date: Thu, 17 Sep 2015 16:42:07 +0300 Subject: [PATCH 348/647] drm/i915/bios: handle MIPI Sequence Block v3+ gracefully The VBT MIPI Sequence Block version 3 has forward incompatible changes: First, the block size in the header has been specified reserved, and the actual size is a separate 32-bit value within the block. The current find_section() function to will only look at the size in the block header, and, depending on what's in that now reserved size field, continue looking for other sections in the wrong place. Fix this by taking the new block size field into account. This will ensure that the lookups for other sections will work properly, as long as the new 32-bit size does not go beyond the opregion VBT mailbox size. Second, the contents of the block have been completely changed. Gracefully refuse parsing the yet unknown data version. Cc: Deepak M Cc: stable@vger.kernel.org Reviewed-by: Deepak M Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_bios.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c index b3e437b3bb54..c19e669ffe50 100644 --- a/drivers/gpu/drm/i915/intel_bios.c +++ b/drivers/gpu/drm/i915/intel_bios.c @@ -42,7 +42,7 @@ find_section(const void *_bdb, int section_id) const struct bdb_header *bdb = _bdb; const u8 *base = _bdb; int index = 0; - u16 total, current_size; + u32 total, current_size; u8 current_id; /* skip to first section */ @@ -57,6 +57,10 @@ find_section(const void *_bdb, int section_id) current_size = *((const u16 *)(base + index)); index += 2; + /* The MIPI Sequence Block v3+ has a separate size field. */ + if (current_id == BDB_MIPI_SEQUENCE && *(base + index) >= 3) + current_size = *((const u32 *)(base + index + 1)); + if (index + current_size > total) return NULL; @@ -799,6 +803,12 @@ parse_mipi(struct drm_i915_private *dev_priv, const struct bdb_header *bdb) return; } + /* Fail gracefully for forward incompatible sequence block. */ + if (sequence->version >= 3) { + DRM_ERROR("Unable to parse MIPI Sequence Block v3+\n"); + return; + } + DRM_DEBUG_DRIVER("Found MIPI sequence block\n"); block_size = get_blocksize(sequence); From f929d42ceb18a8acfd47e0e7b7d90b5d49bd9258 Mon Sep 17 00:00:00 2001 From: Roger Pau Monne Date: Fri, 4 Sep 2015 12:08:07 +0200 Subject: [PATCH 349/647] xen/blkback: free requests on disconnection MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is due to commit 86839c56dee28c315a4c19b7bfee450ccd84cd25 "xen/block: add multi-page ring support" When using an guest under UEFI - after the domain is destroyed the following warning comes from blkback. ------------[ cut here ]------------ WARNING: CPU: 2 PID: 95 at /home/julien/works/linux/drivers/block/xen-blkback/xenbus.c:274 xen_blkif_deferred_free+0x1f4/0x1f8() Modules linked in: CPU: 2 PID: 95 Comm: kworker/2:1 Tainted: G W 4.2.0 #85 Hardware name: APM X-Gene Mustang board (DT) Workqueue: events xen_blkif_deferred_free Call trace: [] dump_backtrace+0x0/0x124 [] show_stack+0x10/0x1c [] dump_stack+0x78/0x98 [] warn_slowpath_common+0x9c/0xd4 [] warn_slowpath_null+0x14/0x20 [] xen_blkif_deferred_free+0x1f0/0x1f8 [] process_one_work+0x160/0x3b4 [] worker_thread+0x140/0x494 [] kthread+0xd8/0xf0 ---[ end trace 6f859b7883c88cdd ]--- Request allocation has been moved to connect_ring, which is called every time blkback connects to the frontend (this can happen multiple times during a blkback instance life cycle). On the other hand, request freeing has not been moved, so it's only called when destroying the backend instance. Due to this mismatch, blkback can allocate the request pool multiple times, without freeing it. In order to fix it, move the freeing of requests to xen_blkif_disconnect to restore the symmetry between request allocation and freeing. Reported-by: Julien Grall Signed-off-by: Roger Pau Monné Tested-by: Julien Grall Cc: Konrad Rzeszutek Wilk Cc: Boris Ostrovsky Cc: David Vrabel Cc: xen-devel@lists.xenproject.org CC: stable@vger.kernel.org # 4.2 Signed-off-by: Konrad Rzeszutek Wilk --- drivers/block/xen-blkback/xenbus.c | 40 ++++++++++++++++-------------- 1 file changed, 21 insertions(+), 19 deletions(-) diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c index deb3f001791f..767657565de6 100644 --- a/drivers/block/xen-blkback/xenbus.c +++ b/drivers/block/xen-blkback/xenbus.c @@ -212,6 +212,9 @@ static int xen_blkif_map(struct xen_blkif *blkif, grant_ref_t *gref, static int xen_blkif_disconnect(struct xen_blkif *blkif) { + struct pending_req *req, *n; + int i = 0, j; + if (blkif->xenblkd) { kthread_stop(blkif->xenblkd); wake_up(&blkif->shutdown_wq); @@ -238,25 +241,6 @@ static int xen_blkif_disconnect(struct xen_blkif *blkif) /* Remove all persistent grants and the cache of ballooned pages. */ xen_blkbk_free_caches(blkif); - return 0; -} - -static void xen_blkif_free(struct xen_blkif *blkif) -{ - struct pending_req *req, *n; - int i = 0, j; - - xen_blkif_disconnect(blkif); - xen_vbd_free(&blkif->vbd); - - /* Make sure everything is drained before shutting down */ - BUG_ON(blkif->persistent_gnt_c != 0); - BUG_ON(atomic_read(&blkif->persistent_gnt_in_use) != 0); - BUG_ON(blkif->free_pages_num != 0); - BUG_ON(!list_empty(&blkif->persistent_purge_list)); - BUG_ON(!list_empty(&blkif->free_pages)); - BUG_ON(!RB_EMPTY_ROOT(&blkif->persistent_gnts)); - /* Check that there is no request in use */ list_for_each_entry_safe(req, n, &blkif->pending_free, free_list) { list_del(&req->free_list); @@ -272,6 +256,24 @@ static void xen_blkif_free(struct xen_blkif *blkif) } WARN_ON(i != (XEN_BLKIF_REQS_PER_PAGE * blkif->nr_ring_pages)); + blkif->nr_ring_pages = 0; + + return 0; +} + +static void xen_blkif_free(struct xen_blkif *blkif) +{ + + xen_blkif_disconnect(blkif); + xen_vbd_free(&blkif->vbd); + + /* Make sure everything is drained before shutting down */ + BUG_ON(blkif->persistent_gnt_c != 0); + BUG_ON(atomic_read(&blkif->persistent_gnt_in_use) != 0); + BUG_ON(blkif->free_pages_num != 0); + BUG_ON(!list_empty(&blkif->persistent_purge_list)); + BUG_ON(!list_empty(&blkif->free_pages)); + BUG_ON(!RB_EMPTY_ROOT(&blkif->persistent_gnts)); kmem_cache_free(xen_blkif_cachep, blkif); } From 50b956f3d85cdea130866f33613416431d60f396 Mon Sep 17 00:00:00 2001 From: Andy Gross Date: Fri, 11 Sep 2015 16:01:16 -0500 Subject: [PATCH 350/647] firmware: qcom: scm: Add function stubs for ARM64 This patch adds stubs for the SCM functions exposed in the QCOM SCM API. Signed-off-by: Andy Gross Acked-by: Russell King Acked-by: Bjorn Andersson --- drivers/firmware/Kconfig | 8 +++++ drivers/firmware/Makefile | 3 +- drivers/firmware/qcom_scm-64.c | 63 ++++++++++++++++++++++++++++++++++ 3 files changed, 73 insertions(+), 1 deletion(-) create mode 100644 drivers/firmware/qcom_scm-64.c diff --git a/drivers/firmware/Kconfig b/drivers/firmware/Kconfig index d8de6a8dd4de..665efca59487 100644 --- a/drivers/firmware/Kconfig +++ b/drivers/firmware/Kconfig @@ -139,6 +139,14 @@ config QCOM_SCM bool depends on ARM || ARM64 +config QCOM_SCM_32 + def_bool y + depends on QCOM_SCM && ARM + +config QCOM_SCM_64 + def_bool y + depends on QCOM_SCM && ARM64 + source "drivers/firmware/broadcom/Kconfig" source "drivers/firmware/google/Kconfig" source "drivers/firmware/efi/Kconfig" diff --git a/drivers/firmware/Makefile b/drivers/firmware/Makefile index 000830fc6707..2ee83474a3c1 100644 --- a/drivers/firmware/Makefile +++ b/drivers/firmware/Makefile @@ -13,7 +13,8 @@ obj-$(CONFIG_ISCSI_IBFT_FIND) += iscsi_ibft_find.o obj-$(CONFIG_ISCSI_IBFT) += iscsi_ibft.o obj-$(CONFIG_FIRMWARE_MEMMAP) += memmap.o obj-$(CONFIG_QCOM_SCM) += qcom_scm.o -obj-$(CONFIG_QCOM_SCM) += qcom_scm-32.o +obj-$(CONFIG_QCOM_SCM_64) += qcom_scm-64.o +obj-$(CONFIG_QCOM_SCM_32) += qcom_scm-32.o CFLAGS_qcom_scm-32.o :=$(call as-instr,.arch_extension sec,-DREQUIRES_SEC=1) obj-y += broadcom/ diff --git a/drivers/firmware/qcom_scm-64.c b/drivers/firmware/qcom_scm-64.c new file mode 100644 index 000000000000..bb6555f6d63b --- /dev/null +++ b/drivers/firmware/qcom_scm-64.c @@ -0,0 +1,63 @@ +/* Copyright (c) 2015, The Linux Foundation. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 and + * only version 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include +#include +#include + +/** + * qcom_scm_set_cold_boot_addr() - Set the cold boot address for cpus + * @entry: Entry point function for the cpus + * @cpus: The cpumask of cpus that will use the entry point + * + * Set the cold boot address of the cpus. Any cpu outside the supported + * range would be removed from the cpu present mask. + */ +int __qcom_scm_set_cold_boot_addr(void *entry, const cpumask_t *cpus) +{ + return -ENOTSUPP; +} + +/** + * qcom_scm_set_warm_boot_addr() - Set the warm boot address for cpus + * @entry: Entry point function for the cpus + * @cpus: The cpumask of cpus that will use the entry point + * + * Set the Linux entry point for the SCM to transfer control to when coming + * out of a power down. CPU power down may be executed on cpuidle or hotplug. + */ +int __qcom_scm_set_warm_boot_addr(void *entry, const cpumask_t *cpus) +{ + return -ENOTSUPP; +} + +/** + * qcom_scm_cpu_power_down() - Power down the cpu + * @flags - Flags to flush cache + * + * This is an end point to power down cpu. If there was a pending interrupt, + * the control would return from this function, otherwise, the cpu jumps to the + * warm boot entry point set for this cpu upon reset. + */ +void __qcom_scm_cpu_power_down(u32 flags) +{ +} + +int __qcom_scm_is_call_available(u32 svc_id, u32 cmd_id) +{ + return -ENOTSUPP; +} + +int __qcom_scm_hdcp_req(struct qcom_scm_hdcp_req *req, u32 req_cnt, u32 *resp) +{ + return -ENOTSUPP; +} From 21cb13e72b02dbbb5a02477d4dd46bc2bc1cfd08 Mon Sep 17 00:00:00 2001 From: Oder Chiou Date: Wed, 23 Sep 2015 14:35:28 +0800 Subject: [PATCH 351/647] ASoC: rt5645: Use the type SOC_DAPM_SINGLE_AUTODISABLE to prevent the weird sound in runtime of power up Signed-off-by: Oder Chiou Signed-off-by: Mark Brown --- sound/soc/codecs/rt5645.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index dbc1d76d9d4e..25c34fbae307 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -732,14 +732,14 @@ static const struct snd_kcontrol_new rt5645_mono_adc_r_mix[] = { static const struct snd_kcontrol_new rt5645_dac_l_mix[] = { SOC_DAPM_SINGLE("Stereo ADC Switch", RT5645_AD_DA_MIXER, RT5645_M_ADCMIX_L_SFT, 1, 1), - SOC_DAPM_SINGLE("DAC1 Switch", RT5645_AD_DA_MIXER, + SOC_DAPM_SINGLE_AUTODISABLE("DAC1 Switch", RT5645_AD_DA_MIXER, RT5645_M_DAC1_L_SFT, 1, 1), }; static const struct snd_kcontrol_new rt5645_dac_r_mix[] = { SOC_DAPM_SINGLE("Stereo ADC Switch", RT5645_AD_DA_MIXER, RT5645_M_ADCMIX_R_SFT, 1, 1), - SOC_DAPM_SINGLE("DAC1 Switch", RT5645_AD_DA_MIXER, + SOC_DAPM_SINGLE_AUTODISABLE("DAC1 Switch", RT5645_AD_DA_MIXER, RT5645_M_DAC1_R_SFT, 1, 1), }; From 4f4794124e0421b080a0f1f5f1207636ba55eb85 Mon Sep 17 00:00:00 2001 From: Oder Chiou Date: Wed, 23 Sep 2015 14:35:29 +0800 Subject: [PATCH 352/647] ASoC: rt5645: Increase the delay time to remove the pop sound Signed-off-by: Oder Chiou Signed-off-by: Mark Brown --- sound/soc/codecs/rt5645.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index 25c34fbae307..61384ee989fc 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -1381,7 +1381,7 @@ static void hp_amp_power(struct snd_soc_codec *codec, int on) regmap_write(rt5645->regmap, RT5645_PR_BASE + RT5645_MAMP_INT_REG2, 0xfc00); snd_soc_write(codec, RT5645_DEPOP_M2, 0x1140); - mdelay(5); + msleep(40); rt5645->hp_on = true; } else { /* depop parameters */ From fce97b4d70ad632dd9c6058622492501377bbaaa Mon Sep 17 00:00:00 2001 From: Oder Chiou Date: Wed, 23 Sep 2015 14:35:30 +0800 Subject: [PATCH 353/647] ASoC: rt5645: Prevent the pop sound in case of playback and the jack is plugging Signed-off-by: Oder Chiou Signed-off-by: Mark Brown --- sound/soc/codecs/rt5645.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index 61384ee989fc..268a28bd1df4 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -2832,6 +2832,9 @@ static int rt5645_jack_detect(struct snd_soc_codec *codec, int jack_insert) } else { /* jack out */ rt5645->jack_type = 0; + regmap_update_bits(rt5645->regmap, RT5645_HP_VOL, + RT5645_L_MUTE | RT5645_R_MUTE, + RT5645_L_MUTE | RT5645_R_MUTE); regmap_update_bits(rt5645->regmap, RT5645_IN1_CTRL2, RT5645_CBJ_MN_JD, RT5645_CBJ_MN_JD); regmap_update_bits(rt5645->regmap, RT5645_IN1_CTRL1, From cbb4be652d374f64661137756b8f357a1827d6a4 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Wed, 23 Sep 2015 11:41:42 -0700 Subject: [PATCH 354/647] USB: whiteheat: fix potential null-deref at probe Fix potential null-pointer dereference at probe by making sure that the required endpoints are present. The whiteheat driver assumes there are at least five pairs of bulk endpoints, of which the final pair is used for the "command port". An attempt to bind to an interface with fewer bulk endpoints would currently lead to an oops. Fixes CVE-2015-5257. Reported-by: Moein Ghasemzadeh Cc: stable Signed-off-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/whiteheat.c | 31 +++++++++++++++++++++++++++++++ 1 file changed, 31 insertions(+) diff --git a/drivers/usb/serial/whiteheat.c b/drivers/usb/serial/whiteheat.c index 6c3734d2b45a..d3ea90bef84d 100644 --- a/drivers/usb/serial/whiteheat.c +++ b/drivers/usb/serial/whiteheat.c @@ -80,6 +80,8 @@ static int whiteheat_firmware_download(struct usb_serial *serial, static int whiteheat_firmware_attach(struct usb_serial *serial); /* function prototypes for the Connect Tech WhiteHEAT serial converter */ +static int whiteheat_probe(struct usb_serial *serial, + const struct usb_device_id *id); static int whiteheat_attach(struct usb_serial *serial); static void whiteheat_release(struct usb_serial *serial); static int whiteheat_port_probe(struct usb_serial_port *port); @@ -116,6 +118,7 @@ static struct usb_serial_driver whiteheat_device = { .description = "Connect Tech - WhiteHEAT", .id_table = id_table_std, .num_ports = 4, + .probe = whiteheat_probe, .attach = whiteheat_attach, .release = whiteheat_release, .port_probe = whiteheat_port_probe, @@ -217,6 +220,34 @@ static int whiteheat_firmware_attach(struct usb_serial *serial) /***************************************************************************** * Connect Tech's White Heat serial driver functions *****************************************************************************/ + +static int whiteheat_probe(struct usb_serial *serial, + const struct usb_device_id *id) +{ + struct usb_host_interface *iface_desc; + struct usb_endpoint_descriptor *endpoint; + size_t num_bulk_in = 0; + size_t num_bulk_out = 0; + size_t min_num_bulk; + unsigned int i; + + iface_desc = serial->interface->cur_altsetting; + + for (i = 0; i < iface_desc->desc.bNumEndpoints; i++) { + endpoint = &iface_desc->endpoint[i].desc; + if (usb_endpoint_is_bulk_in(endpoint)) + ++num_bulk_in; + if (usb_endpoint_is_bulk_out(endpoint)) + ++num_bulk_out; + } + + min_num_bulk = COMMAND_PORT + 1; + if (num_bulk_in < min_num_bulk || num_bulk_out < min_num_bulk) + return -ENODEV; + + return 0; +} + static int whiteheat_attach(struct usb_serial *serial) { struct usb_serial_port *command_port; From bda4e0fb3126aca15586d165b5a15a37edc0a984 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Thu, 3 Sep 2015 08:18:17 -0600 Subject: [PATCH 355/647] NVMe: Set affinity after allocating request queues MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The asynchronous namespace scanning caused affinity hints to be set before its tagset initialized, so there was no cpu mask to set the hint. This patch moves the affinity hint setting to after namespaces are scanned. Reported-by: 김경산 Signed-off-by: Keith Busch Signed-off-by: Jens Axboe --- drivers/block/nvme-core.c | 36 +++++++++++++++++------------------- 1 file changed, 17 insertions(+), 19 deletions(-) diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index b97fc3fe0916..30758bdf69ea 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -2439,6 +2439,22 @@ static void nvme_scan_namespaces(struct nvme_dev *dev, unsigned nn) list_sort(NULL, &dev->namespaces, ns_cmp); } +static void nvme_set_irq_hints(struct nvme_dev *dev) +{ + struct nvme_queue *nvmeq; + int i; + + for (i = 0; i < dev->online_queues; i++) { + nvmeq = dev->queues[i]; + + if (!nvmeq->tags || !(*nvmeq->tags)) + continue; + + irq_set_affinity_hint(dev->entry[nvmeq->cq_vector].vector, + blk_mq_tags_cpumask(*nvmeq->tags)); + } +} + static void nvme_dev_scan(struct work_struct *work) { struct nvme_dev *dev = container_of(work, struct nvme_dev, scan_work); @@ -2450,6 +2466,7 @@ static void nvme_dev_scan(struct work_struct *work) return; nvme_scan_namespaces(dev, le32_to_cpup(&ctrl->nn)); kfree(ctrl); + nvme_set_irq_hints(dev); } /* @@ -2953,22 +2970,6 @@ static const struct file_operations nvme_dev_fops = { .compat_ioctl = nvme_dev_ioctl, }; -static void nvme_set_irq_hints(struct nvme_dev *dev) -{ - struct nvme_queue *nvmeq; - int i; - - for (i = 0; i < dev->online_queues; i++) { - nvmeq = dev->queues[i]; - - if (!nvmeq->tags || !(*nvmeq->tags)) - continue; - - irq_set_affinity_hint(dev->entry[nvmeq->cq_vector].vector, - blk_mq_tags_cpumask(*nvmeq->tags)); - } -} - static int nvme_dev_start(struct nvme_dev *dev) { int result; @@ -3010,8 +3011,6 @@ static int nvme_dev_start(struct nvme_dev *dev) if (result) goto free_tags; - nvme_set_irq_hints(dev); - dev->event_limit = 1; return result; @@ -3062,7 +3061,6 @@ static int nvme_dev_resume(struct nvme_dev *dev) } else { nvme_unfreeze_queues(dev); nvme_dev_add(dev); - nvme_set_irq_hints(dev); } return 0; } From 3daea9e3d3ecd217a63f35e63f18ea7138f2ae17 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Sat, 5 Sep 2015 11:12:27 +0200 Subject: [PATCH 356/647] drm/amdgpu: add option to disable semaphores MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Provide module parameter to enable/disable them. Still enabled by default. Signed-off-by: Christian König Reviewed-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 4 ++++ drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c | 17 ++++++++++++++++- 3 files changed, 21 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 668939a14206..e587e20074d3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -82,6 +82,7 @@ extern int amdgpu_vm_block_size; extern int amdgpu_enable_scheduler; extern int amdgpu_sched_jobs; extern int amdgpu_sched_hw_submission; +extern int amdgpu_enable_semaphores; #define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS 3000 #define AMDGPU_MAX_USEC_TIMEOUT 100000 /* 100 ms */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 0fcc0bd1622c..adb48353f2e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -79,6 +79,7 @@ int amdgpu_exp_hw_support = 0; int amdgpu_enable_scheduler = 0; int amdgpu_sched_jobs = 16; int amdgpu_sched_hw_submission = 2; +int amdgpu_enable_semaphores = 1; MODULE_PARM_DESC(vramlimit, "Restrict VRAM for testing, in megabytes"); module_param_named(vramlimit, amdgpu_vram_limit, int, 0600); @@ -152,6 +153,9 @@ module_param_named(sched_jobs, amdgpu_sched_jobs, int, 0444); MODULE_PARM_DESC(sched_hw_submission, "the max number of HW submissions (default 2)"); module_param_named(sched_hw_submission, amdgpu_sched_hw_submission, int, 0444); +MODULE_PARM_DESC(enable_semaphores, "Enable semaphores (1 = enable (default), 0 = disable)"); +module_param_named(enable_semaphores, amdgpu_enable_semaphores, int, 0644); + static struct pci_device_id pciidlist[] = { #ifdef CONFIG_DRM_AMDGPU_CIK /* Kaveri */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c index 068aeaff7183..cfd2999acc59 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c @@ -251,6 +251,20 @@ int amdgpu_sync_wait(struct amdgpu_sync *sync) fence_put(e->fence); kfree(e); } + + if (amdgpu_enable_semaphores) + return 0; + + for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { + struct amdgpu_fence *fence = sync->sync_to[i]; + if (!fence) + continue; + + r = fence_wait(&fence->base, false); + if (r) + return r; + } + return 0; } @@ -285,7 +299,8 @@ int amdgpu_sync_rings(struct amdgpu_sync *sync, return -EINVAL; } - if (amdgpu_enable_scheduler || (count >= AMDGPU_NUM_SYNCS)) { + if (amdgpu_enable_scheduler || !amdgpu_enable_semaphores || + (count >= AMDGPU_NUM_SYNCS)) { /* not enough room, wait manually */ r = fence_wait(&fence->base, false); if (r) From 22c01cc48301f6974868bd4a7b03e29883da1103 Mon Sep 17 00:00:00 2001 From: Anatoli Antonovitch Date: Thu, 3 Sep 2015 11:13:31 -0400 Subject: [PATCH 357/647] drm/amdgpu: execution barrier after fence v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Insert wait for reg mem after EOP to fix potential issue with vm context switch v2: move wait to vm_flush() use equal instead of greater than. Signed-off-by: Anatoli Antonovitch Signed-off-by: Christian König Reviewed-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 53f07439a512..0f979abca235 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -3965,6 +3965,7 @@ static void gfx_v8_0_ring_emit_fence_gfx(struct amdgpu_ring *ring, u64 addr, DATA_SEL(write64bit ? 2 : 1) | INT_SEL(int_sel ? 2 : 0)); amdgpu_ring_write(ring, lower_32_bits(seq)); amdgpu_ring_write(ring, upper_32_bits(seq)); + } /** @@ -4044,6 +4045,17 @@ static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring, unsigned vm_id, uint64_t pd_addr) { int usepfp = (ring->type == AMDGPU_RING_TYPE_GFX); + uint32_t seq = ring->fence_drv.sync_seq[ring->idx]; + uint64_t addr = ring->fence_drv.gpu_addr; + + amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); + amdgpu_ring_write(ring, (WAIT_REG_MEM_MEM_SPACE(1) | /* memory */ + WAIT_REG_MEM_FUNCTION(3))); /* equal */ + amdgpu_ring_write(ring, addr & 0xfffffffc); + amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff); + amdgpu_ring_write(ring, seq); + amdgpu_ring_write(ring, 0xffffffff); + amdgpu_ring_write(ring, 4); /* poll interval */ amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) | From 20a85ff846ffed84fba8637abbb6b1c96436c0ac Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Sat, 5 Sep 2015 11:59:50 +0200 Subject: [PATCH 358/647] drm/amdgpu: use write confirm for vm_flush() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Make sure the CP waits for the write to be confirmed before invalidating. Signed-off-by: Christian König Reviewed-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 0f979abca235..d6d330bc6daa 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -4059,7 +4059,8 @@ static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring, amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) | - WRITE_DATA_DST_SEL(0))); + WRITE_DATA_DST_SEL(0)) | + WR_CONFIRM); if (vm_id < 8) { amdgpu_ring_write(ring, (mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR + vm_id)); From 353da3c520b47272b9e3ddbc70b81be285c0b933 Mon Sep 17 00:00:00 2001 From: Chunming Zhou Date: Mon, 7 Sep 2015 16:06:53 +0800 Subject: [PATCH 359/647] drm/amdgpu: add tracepoint for scheduler (v2) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit track sched job status like the length of job queue and hw job queue. v2: fix build after rebase Signed-off-by: Chunming Zhou Reviewed-by: Jammy Zhou Reviewed-by: Christian König --- .../gpu/drm/amd/scheduler/gpu_sched_trace.h | 41 +++++++++++++++++++ drivers/gpu/drm/amd/scheduler/gpu_scheduler.c | 5 ++- 2 files changed, 45 insertions(+), 1 deletion(-) create mode 100644 drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h diff --git a/drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h b/drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h new file mode 100644 index 000000000000..a1f4ece58a24 --- /dev/null +++ b/drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h @@ -0,0 +1,41 @@ +#if !defined(_GPU_SCHED_TRACE_H) || defined(TRACE_HEADER_MULTI_READ) +#define _GPU_SCHED_TRACE_H_ + +#include +#include +#include + +#include + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM gpu_sched +#define TRACE_INCLUDE_FILE gpu_sched_trace + +TRACE_EVENT(amd_sched_job, + TP_PROTO(struct amd_sched_job *sched_job), + TP_ARGS(sched_job), + TP_STRUCT__entry( + __field(struct amd_sched_entity *, entity) + __field(u32, ring_id) + __field(u32, job_count) + __field(int, hw_job_count) + ), + + TP_fast_assign( + __entry->entity = sched_job->s_entity; + __entry->ring_id = sched_job->sched->ring_id; + __entry->job_count = kfifo_len( + &sched_job->s_entity->job_queue) / sizeof(sched_job); + __entry->hw_job_count = atomic_read( + &sched_job->sched->hw_rq_count); + ), + TP_printk("entity=%p, ring=%u, job count:%u, hw job count:%d", + __entry->entity, __entry->ring_id, __entry->job_count, + __entry->hw_job_count) +); +#endif + +/* This part must be outside protection */ +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#include diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index 9259f1b6664c..5ad7fa83f501 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -27,6 +27,9 @@ #include #include "gpu_scheduler.h" +#define CREATE_TRACE_POINTS +#include "gpu_sched_trace.h" + static struct amd_sched_job * amd_sched_entity_pop_job(struct amd_sched_entity *entity); static void amd_sched_wakeup(struct amd_gpu_scheduler *sched); @@ -273,7 +276,7 @@ int amd_sched_entity_push_job(struct amd_sched_job *sched_job) wait_event(entity->scheduler->job_scheduled, amd_sched_entity_in(sched_job)); - + trace_amd_sched_job(sched_job); return 0; } From 27439fcac03632f2b1fd85268dc61af828c77e7b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Wed, 2 Sep 2015 12:03:06 +0200 Subject: [PATCH 360/647] drm/amdgpu: signal scheduler fence when hw submission fails v3 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Otherwise the resource blocked by it will never be reclaimed. v2: add DRM_ERROR. v3: fix typo in commit message Signed-off-by: Christian König Reviewed-by: Junwei Zhang Reviewed-by: Chunming Zhou Reviewed-by: Jammy Zhou --- drivers/gpu/drm/amd/scheduler/gpu_scheduler.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index 5ad7fa83f501..905fd3089a53 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -362,6 +362,9 @@ static int amd_sched_main(void *param) else if (r) DRM_ERROR("fence add callback failed (%d)\n", r); fence_put(fence); + } else { + DRM_ERROR("Failed to run job!\n"); + amd_sched_process_job(NULL, &job->cb); } count = kfifo_out(&entity->job_queue, &job, sizeof(job)); From 258f3f99d514172aa5a9df15e6d6ebe33aad2f55 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 31 Aug 2015 17:02:52 +0200 Subject: [PATCH 361/647] drm/amdgpu: move scheduler fence callback into fence v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit And call the processed callback directly after submitting the job. v2: split adding error handling into separate patch. Signed-off-by: Christian König Reviewed-by: Chunming Zhou Reviewed-by: Junwei Zhang Reviewed-by: Jammy Zhou --- drivers/gpu/drm/amd/scheduler/gpu_scheduler.c | 21 ++++++++++--------- drivers/gpu/drm/amd/scheduler/gpu_scheduler.h | 2 +- 2 files changed, 12 insertions(+), 11 deletions(-) diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index 905fd3089a53..92b000d4307e 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -319,15 +319,13 @@ amd_sched_select_job(struct amd_gpu_scheduler *sched) static void amd_sched_process_job(struct fence *f, struct fence_cb *cb) { - struct amd_sched_job *sched_job = - container_of(cb, struct amd_sched_job, cb); - struct amd_gpu_scheduler *sched; + struct amd_sched_fence *s_fence = + container_of(cb, struct amd_sched_fence, cb); + struct amd_gpu_scheduler *sched = s_fence->scheduler; - sched = sched_job->sched; - amd_sched_fence_signal(sched_job->s_fence); atomic_dec(&sched->hw_rq_count); - fence_put(&sched_job->s_fence->base); - sched->ops->process_job(sched_job); + amd_sched_fence_signal(s_fence); + fence_put(&s_fence->base); wake_up_interruptible(&sched->wake_up_worker); } @@ -341,6 +339,7 @@ static int amd_sched_main(void *param) while (!kthread_should_stop()) { struct amd_sched_entity *entity; + struct amd_sched_fence *s_fence; struct amd_sched_job *job; struct fence *fence; @@ -352,19 +351,21 @@ static int amd_sched_main(void *param) continue; entity = job->s_entity; + s_fence = job->s_fence; atomic_inc(&sched->hw_rq_count); fence = sched->ops->run_job(job); + sched->ops->process_job(job); if (fence) { - r = fence_add_callback(fence, &job->cb, + r = fence_add_callback(fence, &s_fence->cb, amd_sched_process_job); if (r == -ENOENT) - amd_sched_process_job(fence, &job->cb); + amd_sched_process_job(fence, &s_fence->cb); else if (r) DRM_ERROR("fence add callback failed (%d)\n", r); fence_put(fence); } else { DRM_ERROR("Failed to run job!\n"); - amd_sched_process_job(NULL, &job->cb); + amd_sched_process_job(NULL, &s_fence->cb); } count = kfifo_out(&entity->job_queue, &job, sizeof(job)); diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h index 2af0e4d4d817..7a0552fb6ba4 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h @@ -62,13 +62,13 @@ struct amd_sched_rq { struct amd_sched_fence { struct fence base; + struct fence_cb cb; struct amd_gpu_scheduler *scheduler; spinlock_t lock; void *owner; }; struct amd_sched_job { - struct fence_cb cb; struct amd_gpu_scheduler *sched; struct amd_sched_entity *s_entity; struct amd_sched_fence *s_fence; From 1886d1a9caed20f457dd69a926c7f8b54c2d5f48 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 31 Aug 2015 17:28:28 +0200 Subject: [PATCH 362/647] drm/amdgpu: remove process_job callback from the scheduler MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Just free the resources immediately after submitting the job. Signed-off-by: Christian König Reviewed-by: Chunming Zhou Reviewed-by: Junwei Zhang Reviewed-by: Jammy Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c | 29 +++++-------------- drivers/gpu/drm/amd/scheduler/gpu_scheduler.c | 1 - drivers/gpu/drm/amd/scheduler/gpu_scheduler.h | 1 - 3 files changed, 7 insertions(+), 24 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c index de98fbd2971e..5724a81fbf5e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c @@ -35,8 +35,8 @@ static struct fence *amdgpu_sched_dependency(struct amd_sched_job *job) static struct fence *amdgpu_sched_run_job(struct amd_sched_job *job) { + struct amdgpu_fence *fence = NULL; struct amdgpu_job *sched_job; - struct amdgpu_fence *fence; int r; if (!job) { @@ -49,41 +49,26 @@ static struct fence *amdgpu_sched_run_job(struct amd_sched_job *job) sched_job->num_ibs, sched_job->ibs, sched_job->base.owner); - if (r) + if (r) { + DRM_ERROR("Error scheduling IBs (%d)\n", r); goto err; + } + fence = amdgpu_fence_ref(sched_job->ibs[sched_job->num_ibs - 1].fence); +err: if (sched_job->free_job) sched_job->free_job(sched_job); mutex_unlock(&sched_job->job_lock); - return &fence->base; - -err: - DRM_ERROR("Run job error\n"); - mutex_unlock(&sched_job->job_lock); - job->sched->ops->process_job(job); - return NULL; -} - -static void amdgpu_sched_process_job(struct amd_sched_job *job) -{ - struct amdgpu_job *sched_job; - - if (!job) { - DRM_ERROR("job is null\n"); - return; - } - sched_job = (struct amdgpu_job *)job; - /* after processing job, free memory */ fence_put(&sched_job->base.s_fence->base); kfree(sched_job); + return fence ? &fence->base : NULL; } struct amd_sched_backend_ops amdgpu_sched_ops = { .dependency = amdgpu_sched_dependency, .run_job = amdgpu_sched_run_job, - .process_job = amdgpu_sched_process_job }; int amdgpu_sched_ib_submit_kernel_helper(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index 92b000d4307e..191fd513d979 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -354,7 +354,6 @@ static int amd_sched_main(void *param) s_fence = job->s_fence; atomic_inc(&sched->hw_rq_count); fence = sched->ops->run_job(job); - sched->ops->process_job(job); if (fence) { r = fence_add_callback(fence, &s_fence->cb, amd_sched_process_job); diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h index 7a0552fb6ba4..ac56d9211f7c 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h @@ -93,7 +93,6 @@ static inline struct amd_sched_fence *to_amd_sched_fence(struct fence *f) struct amd_sched_backend_ops { struct fence *(*dependency)(struct amd_sched_job *job); struct fence *(*run_job)(struct amd_sched_job *job); - void (*process_job)(struct amd_sched_job *job); }; /** From b7d698d7fd7d132c6ebe56d230584f2cae6c94ee Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 7 Sep 2015 12:32:09 +0200 Subject: [PATCH 363/647] drm/amdgpu: fix overflow on 32bit systems MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit mem->start is a long, so this can overflow on 32bit systems. Signed-off-by: Christian König Reviewed-by: Jammy Zhou Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index f68b7cdc370a..b68dcb9edacd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -855,7 +855,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, int r; if (mem) { - addr = mem->start << PAGE_SHIFT; + addr = (u64)mem->start << PAGE_SHIFT; if (mem->mem_type != TTM_PL_TT) addr += adev->vm_manager.vram_base_offset; } else { From 72d7668b5ba5180b651e8a07dd6ed62e4e26f207 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Thu, 3 Sep 2015 17:34:59 +0200 Subject: [PATCH 364/647] drm/amdgpu: export reservation_object from dmabuf to ttm (v2) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Adds an extra argument to amdgpu_bo_create, which is only used in amdgpu_prime.c. Port of radeon commit 831b6966a60fe72d85ae3576056b4e4e0775b112. v2: fix up kfd. Signed-off-by: Christian König Reviewed-by: Jammy Zhou Reviewed-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c | 6 ++++-- drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c | 5 +++-- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 5 +++-- drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 3 ++- drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 15 ++++++++------- drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c | 5 ++++- drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/amdgpu_test.c | 8 +++++--- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c | 6 +++--- drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/cz_smc.c | 6 ++++-- drivers/gpu/drm/amd/amdgpu/fiji_smc.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c | 19 +++++++++++-------- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 10 +++++----- drivers/gpu/drm/amd/amdgpu/iceland_smc.c | 2 +- drivers/gpu/drm/amd/amdgpu/tonga_smc.c | 4 ++-- 24 files changed, 70 insertions(+), 52 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 496ed2192eba..84d68d658f8a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -183,7 +183,7 @@ int alloc_gtt_mem(struct kgd_dev *kgd, size_t size, return -ENOMEM; r = amdgpu_bo_create(rdev, size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GTT, - AMDGPU_GEM_CREATE_CPU_GTT_USWC, NULL, &(*mem)->bo); + AMDGPU_GEM_CREATE_CPU_GTT_USWC, NULL, NULL, &(*mem)->bo); if (r) { dev_err(rdev->dev, "failed to allocate BO for amdkfd (%d)\n", r); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c index 98d59ee640ce..cd639c362df3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c @@ -79,7 +79,8 @@ static void amdgpu_benchmark_move(struct amdgpu_device *adev, unsigned size, int time; n = AMDGPU_BENCHMARK_ITERATIONS; - r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, sdomain, 0, NULL, &sobj); + r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, sdomain, 0, NULL, + NULL, &sobj); if (r) { goto out_cleanup; } @@ -91,7 +92,8 @@ static void amdgpu_benchmark_move(struct amdgpu_device *adev, unsigned size, if (r) { goto out_cleanup; } - r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, ddomain, 0, NULL, &dobj); + r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, ddomain, 0, NULL, + NULL, &dobj); if (r) { goto out_cleanup; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c index 6b1243f9f86d..1c3fc99c5465 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c @@ -86,7 +86,7 @@ static int amdgpu_cgs_gmap_kmem(void *cgs_device, void *kmem, struct sg_table *sg = drm_prime_pages_to_sg(&kmem_page, npages); ret = amdgpu_bo_create(adev, size, PAGE_SIZE, false, - AMDGPU_GEM_DOMAIN_GTT, 0, sg, &bo); + AMDGPU_GEM_DOMAIN_GTT, 0, sg, NULL, &bo); if (ret) return ret; ret = amdgpu_bo_reserve(bo, false); @@ -197,7 +197,8 @@ static int amdgpu_cgs_alloc_gpu_mem(void *cgs_device, ret = amdgpu_bo_create_restricted(adev, size, PAGE_SIZE, true, domain, flags, - NULL, &placement, &obj); + NULL, &placement, NULL, + &obj); if (ret) { DRM_ERROR("(%d) bo create failed\n", ret); return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 6ff6ae945794..2d569eccf41f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -246,7 +246,7 @@ static int amdgpu_vram_scratch_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, AMDGPU_GPU_PAGE_SIZE, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->vram_scratch.robj); + NULL, NULL, &adev->vram_scratch.robj); if (r) { return r; } @@ -449,7 +449,8 @@ static int amdgpu_wb_init(struct amdgpu_device *adev) if (adev->wb.wb_obj == NULL) { r = amdgpu_bo_create(adev, AMDGPU_MAX_WB * 4, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, &adev->wb.wb_obj); + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, + &adev->wb.wb_obj); if (r) { dev_warn(adev->dev, "(%d) create WB bo failed\n", r); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c index cbd3a486c5c2..7312d729d300 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c @@ -127,7 +127,7 @@ int amdgpu_gart_table_vram_alloc(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, adev->gart.table_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->gart.robj); + NULL, NULL, &adev->gart.robj); if (r) { return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 5839fab374bf..b75c3b235832 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -69,7 +69,8 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size, } } retry: - r = amdgpu_bo_create(adev, size, alignment, kernel, initial_domain, flags, NULL, &robj); + r = amdgpu_bo_create(adev, size, alignment, kernel, initial_domain, + flags, NULL, NULL, &robj); if (r) { if (r != -ERESTARTSYS) { if (initial_domain == AMDGPU_GEM_DOMAIN_VRAM) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c index 5c8a803acedc..534fc04e80fd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c @@ -43,7 +43,7 @@ static int amdgpu_ih_ring_alloc(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, adev->irq.ih.ring_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GTT, 0, - NULL, &adev->irq.ih.ring_obj); + NULL, NULL, &adev->irq.ih.ring_obj); if (r) { DRM_ERROR("amdgpu: failed to create ih ring buffer (%d).\n", r); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 08b09d55b96f..f25cfed33927 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -215,6 +215,7 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev, bool kernel, u32 domain, u64 flags, struct sg_table *sg, struct ttm_placement *placement, + struct reservation_object *resv, struct amdgpu_bo **bo_ptr) { struct amdgpu_bo *bo; @@ -261,7 +262,7 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev, /* Kernel allocation are uninterruptible */ r = ttm_bo_init(&adev->mman.bdev, &bo->tbo, size, type, &bo->placement, page_align, !kernel, NULL, - acc_size, sg, NULL, &amdgpu_ttm_bo_destroy); + acc_size, sg, resv, &amdgpu_ttm_bo_destroy); if (unlikely(r != 0)) { return r; } @@ -275,7 +276,9 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev, int amdgpu_bo_create(struct amdgpu_device *adev, unsigned long size, int byte_align, bool kernel, u32 domain, u64 flags, - struct sg_table *sg, struct amdgpu_bo **bo_ptr) + struct sg_table *sg, + struct reservation_object *resv, + struct amdgpu_bo **bo_ptr) { struct ttm_placement placement = {0}; struct ttm_place placements[AMDGPU_GEM_DOMAIN_MAX + 1]; @@ -286,11 +289,9 @@ int amdgpu_bo_create(struct amdgpu_device *adev, amdgpu_ttm_placement_init(adev, &placement, placements, domain, flags); - return amdgpu_bo_create_restricted(adev, size, byte_align, - kernel, domain, flags, - sg, - &placement, - bo_ptr); + return amdgpu_bo_create_restricted(adev, size, byte_align, kernel, + domain, flags, sg, &placement, + resv, bo_ptr); } int amdgpu_bo_kmap(struct amdgpu_bo *bo, void **ptr) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 6ea18dcec561..3c2ff4567798 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -129,12 +129,14 @@ int amdgpu_bo_create(struct amdgpu_device *adev, unsigned long size, int byte_align, bool kernel, u32 domain, u64 flags, struct sg_table *sg, + struct reservation_object *resv, struct amdgpu_bo **bo_ptr); int amdgpu_bo_create_restricted(struct amdgpu_device *adev, unsigned long size, int byte_align, bool kernel, u32 domain, u64 flags, struct sg_table *sg, struct ttm_placement *placement, + struct reservation_object *resv, struct amdgpu_bo **bo_ptr); int amdgpu_bo_kmap(struct amdgpu_bo *bo, void **ptr); void amdgpu_bo_kunmap(struct amdgpu_bo *bo); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c index d9652fe32d6a..59f735a933a9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c @@ -61,12 +61,15 @@ struct drm_gem_object *amdgpu_gem_prime_import_sg_table(struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sg) { + struct reservation_object *resv = attach->dmabuf->resv; struct amdgpu_device *adev = dev->dev_private; struct amdgpu_bo *bo; int ret; + ww_mutex_lock(&resv->lock, NULL); ret = amdgpu_bo_create(adev, attach->dmabuf->size, PAGE_SIZE, false, - AMDGPU_GEM_DOMAIN_GTT, 0, sg, &bo); + AMDGPU_GEM_DOMAIN_GTT, 0, sg, resv, &bo); + ww_mutex_unlock(&resv->lock); if (ret) return ERR_PTR(ret); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 9bec91484c24..130292dcf243 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -407,7 +407,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, if (ring->ring_obj == NULL) { r = amdgpu_bo_create(adev, ring->ring_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GTT, 0, - NULL, &ring->ring_obj); + NULL, NULL, &ring->ring_obj); if (r) { dev_err(adev->dev, "(%d) ring create failed\n", r); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c index 74dad270362c..b70ce106ca90 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c @@ -64,8 +64,8 @@ int amdgpu_sa_bo_manager_init(struct amdgpu_device *adev, INIT_LIST_HEAD(&sa_manager->flist[i]); } - r = amdgpu_bo_create(adev, size, align, true, - domain, 0, NULL, &sa_manager->bo); + r = amdgpu_bo_create(adev, size, align, true, domain, + 0, NULL, NULL, &sa_manager->bo); if (r) { dev_err(adev->dev, "(%d) failed to allocate bo for manager\n", r); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c index f80b1a43be8a..4865615e9c06 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c @@ -59,8 +59,9 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev) goto out_cleanup; } - r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, 0, - NULL, &vram_obj); + r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, + AMDGPU_GEM_DOMAIN_VRAM, 0, + NULL, NULL, &vram_obj); if (r) { DRM_ERROR("Failed to create VRAM object\n"); goto out_cleanup; @@ -80,7 +81,8 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev) struct fence *fence = NULL; r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, gtt_obj + i); + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, + NULL, gtt_obj + i); if (r) { DRM_ERROR("Failed to create GTT object %d\n", i); goto out_lclean; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index b5abd5cde413..364cbe975332 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -861,7 +861,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, 256 * 1024, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->stollen_vga_memory); + NULL, NULL, &adev->stollen_vga_memory); if (r) { return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index 482e66797ae6..5cc95f1a7dab 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -247,7 +247,7 @@ int amdgpu_ucode_init_bo(struct amdgpu_device *adev) const struct common_firmware_header *header = NULL; err = amdgpu_bo_create(adev, adev->firmware.fw_size, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, bo); + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, bo); if (err) { dev_err(adev->dev, "(%d) Firmware buffer allocate failed\n", err); err = -ENOMEM; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 2cf6c6b06e3b..3cc86631d5d6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -156,7 +156,7 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, bo_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->uvd.vcpu_bo); + NULL, NULL, &adev->uvd.vcpu_bo); if (r) { dev_err(adev->dev, "(%d) failed to allocate UVD bo\n", r); return r; @@ -905,7 +905,7 @@ int amdgpu_uvd_get_create_msg(struct amdgpu_ring *ring, uint32_t handle, r = amdgpu_bo_create(adev, 1024, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &bo); + NULL, NULL, &bo); if (r) return r; @@ -954,7 +954,7 @@ int amdgpu_uvd_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle, r = amdgpu_bo_create(adev, 1024, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &bo); + NULL, NULL, &bo); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index 3cab96c42aa8..c90caf810e4a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -143,7 +143,7 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size) r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->vce.vcpu_bo); + NULL, NULL, &adev->vce.vcpu_bo); if (r) { dev_err(adev->dev, "(%d) failed to allocate VCE bo\n", r); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index b68dcb9edacd..8927dc6dec7c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -1101,7 +1101,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev, AMDGPU_GPU_PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_NO_CPU_ACCESS, - NULL, &pt); + NULL, NULL, &pt); if (r) goto error_free; @@ -1303,7 +1303,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm) r = amdgpu_bo_create(adev, pd_size, align, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_NO_CPU_ACCESS, - NULL, &vm->page_directory); + NULL, NULL, &vm->page_directory); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/cz_smc.c b/drivers/gpu/drm/amd/amdgpu/cz_smc.c index a72ffc7d6c26..e33180d3314a 100644 --- a/drivers/gpu/drm/amd/amdgpu/cz_smc.c +++ b/drivers/gpu/drm/amd/amdgpu/cz_smc.c @@ -814,7 +814,8 @@ int cz_smu_init(struct amdgpu_device *adev) * 3. map kernel virtual address */ ret = amdgpu_bo_create(adev, priv->toc_buffer.data_size, PAGE_SIZE, - true, AMDGPU_GEM_DOMAIN_GTT, 0, NULL, toc_buf); + true, AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, + toc_buf); if (ret) { dev_err(adev->dev, "(%d) SMC TOC buffer allocation failed\n", ret); @@ -822,7 +823,8 @@ int cz_smu_init(struct amdgpu_device *adev) } ret = amdgpu_bo_create(adev, priv->smu_buffer.data_size, PAGE_SIZE, - true, AMDGPU_GEM_DOMAIN_GTT, 0, NULL, smu_buf); + true, AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, + smu_buf); if (ret) { dev_err(adev->dev, "(%d) SMC Internal buffer allocation failed\n", ret); diff --git a/drivers/gpu/drm/amd/amdgpu/fiji_smc.c b/drivers/gpu/drm/amd/amdgpu/fiji_smc.c index 322edea65857..bda1249eb871 100644 --- a/drivers/gpu/drm/amd/amdgpu/fiji_smc.c +++ b/drivers/gpu/drm/amd/amdgpu/fiji_smc.c @@ -764,7 +764,7 @@ int fiji_smu_init(struct amdgpu_device *adev) ret = amdgpu_bo_create(adev, image_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, toc_buf); + NULL, NULL, toc_buf); if (ret) { DRM_ERROR("Failed to allocate memory for TOC buffer\n"); return -ENOMEM; @@ -774,7 +774,7 @@ int fiji_smu_init(struct amdgpu_device *adev) ret = amdgpu_bo_create(adev, smu_internal_buffer_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, smu_buf); + NULL, NULL, smu_buf); if (ret) { DRM_ERROR("Failed to allocate memory for SMU internal buffer\n"); return -ENOMEM; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 4bd1e5cf65ca..392ec10cc475 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -3206,7 +3206,7 @@ static int gfx_v7_0_mec_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, adev->gfx.mec.num_mec *adev->gfx.mec.num_pipe * MEC_HPD_SIZE * 2, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, &adev->gfx.mec.hpd_eop_obj); if (r) { dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r); @@ -3373,7 +3373,7 @@ static int gfx_v7_0_cp_compute_resume(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, sizeof(struct bonaire_mqd), PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, &ring->mqd_obj); if (r) { dev_warn(adev->dev, "(%d) create MQD bo failed\n", r); @@ -3788,7 +3788,8 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, dws * 4, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->gfx.rlc.save_restore_obj); + NULL, NULL, + &adev->gfx.rlc.save_restore_obj); if (r) { dev_warn(adev->dev, "(%d) create RLC sr bo failed\n", r); return r; @@ -3831,7 +3832,8 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, dws * 4, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->gfx.rlc.clear_state_obj); + NULL, NULL, + &adev->gfx.rlc.clear_state_obj); if (r) { dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r); gfx_v7_0_rlc_fini(adev); @@ -3870,7 +3872,8 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, adev->gfx.rlc.cp_table_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->gfx.rlc.cp_table_obj); + NULL, NULL, + &adev->gfx.rlc.cp_table_obj); if (r) { dev_warn(adev->dev, "(%d) create RLC cp table bo failed\n", r); gfx_v7_0_rlc_fini(adev); @@ -4851,21 +4854,21 @@ static int gfx_v7_0_sw_init(void *handle) r = amdgpu_bo_create(adev, adev->gds.mem.gfx_partition_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GDS, 0, - NULL, &adev->gds.gds_gfx_bo); + NULL, NULL, &adev->gds.gds_gfx_bo); if (r) return r; r = amdgpu_bo_create(adev, adev->gds.gws.gfx_partition_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GWS, 0, - NULL, &adev->gds.gws_gfx_bo); + NULL, NULL, &adev->gds.gws_gfx_bo); if (r) return r; r = amdgpu_bo_create(adev, adev->gds.oa.gfx_partition_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_OA, 0, - NULL, &adev->gds.oa_gfx_bo); + NULL, NULL, &adev->gds.oa_gfx_bo); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index d6d330bc6daa..78e5900d71cd 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -868,7 +868,7 @@ static int gfx_v8_0_mec_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, adev->gfx.mec.num_mec *adev->gfx.mec.num_pipe * MEC_HPD_SIZE * 2, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, &adev->gfx.mec.hpd_eop_obj); if (r) { dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r); @@ -995,21 +995,21 @@ static int gfx_v8_0_sw_init(void *handle) /* reserve GDS, GWS and OA resource for gfx */ r = amdgpu_bo_create(adev, adev->gds.mem.gfx_partition_size, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GDS, 0, + AMDGPU_GEM_DOMAIN_GDS, 0, NULL, NULL, &adev->gds.gds_gfx_bo); if (r) return r; r = amdgpu_bo_create(adev, adev->gds.gws.gfx_partition_size, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GWS, 0, + AMDGPU_GEM_DOMAIN_GWS, 0, NULL, NULL, &adev->gds.gws_gfx_bo); if (r) return r; r = amdgpu_bo_create(adev, adev->gds.oa.gfx_partition_size, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_OA, 0, + AMDGPU_GEM_DOMAIN_OA, 0, NULL, NULL, &adev->gds.oa_gfx_bo); if (r) return r; @@ -3106,7 +3106,7 @@ static int gfx_v8_0_cp_compute_resume(struct amdgpu_device *adev) sizeof(struct vi_mqd), PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GTT, 0, NULL, - &ring->mqd_obj); + NULL, &ring->mqd_obj); if (r) { dev_warn(adev->dev, "(%d) create MQD bo failed\n", r); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/iceland_smc.c b/drivers/gpu/drm/amd/amdgpu/iceland_smc.c index c900aa942ade..966d4b2ed9da 100644 --- a/drivers/gpu/drm/amd/amdgpu/iceland_smc.c +++ b/drivers/gpu/drm/amd/amdgpu/iceland_smc.c @@ -625,7 +625,7 @@ int iceland_smu_init(struct amdgpu_device *adev) ret = amdgpu_bo_create(adev, image_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, toc_buf); + NULL, NULL, toc_buf); if (ret) { DRM_ERROR("Failed to allocate memory for TOC buffer\n"); return -ENOMEM; diff --git a/drivers/gpu/drm/amd/amdgpu/tonga_smc.c b/drivers/gpu/drm/amd/amdgpu/tonga_smc.c index 1f5ac941a610..5421309c1862 100644 --- a/drivers/gpu/drm/amd/amdgpu/tonga_smc.c +++ b/drivers/gpu/drm/amd/amdgpu/tonga_smc.c @@ -763,7 +763,7 @@ int tonga_smu_init(struct amdgpu_device *adev) ret = amdgpu_bo_create(adev, image_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, toc_buf); + NULL, NULL, toc_buf); if (ret) { DRM_ERROR("Failed to allocate memory for TOC buffer\n"); return -ENOMEM; @@ -773,7 +773,7 @@ int tonga_smu_init(struct amdgpu_device *adev) ret = amdgpu_bo_create(adev, smu_internal_buffer_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, smu_buf); + NULL, NULL, smu_buf); if (ret) { DRM_ERROR("Failed to allocate memory for SMU internal buffer\n"); return -ENOMEM; From a5b750583eb4af69da1e659c7684b6d370b2ae97 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Thu, 3 Sep 2015 16:40:39 +0200 Subject: [PATCH 365/647] drm/amdgpu: validate duplicates in the CS as well MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This allows for multiple BOs to have the same reservation object. Signed-off-by: Christian König Reviewed-by: Jammy Zhou Reviewed-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 34 +++++++++++++++----------- 1 file changed, 20 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 3b355aeb62fd..4b92e3836684 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -321,25 +321,17 @@ static u64 amdgpu_cs_get_threshold_for_moves(struct amdgpu_device *adev) return max(bytes_moved_threshold, 1024*1024ull); } -int amdgpu_cs_list_validate(struct amdgpu_cs_parser *p) +int amdgpu_cs_list_validate(struct amdgpu_device *adev, + struct amdgpu_vm *vm, + struct list_head *validated) { - struct amdgpu_fpriv *fpriv = p->filp->driver_priv; - struct amdgpu_vm *vm = &fpriv->vm; - struct amdgpu_device *adev = p->adev; struct amdgpu_bo_list_entry *lobj; - struct list_head duplicates; struct amdgpu_bo *bo; u64 bytes_moved = 0, initial_bytes_moved; u64 bytes_moved_threshold = amdgpu_cs_get_threshold_for_moves(adev); int r; - INIT_LIST_HEAD(&duplicates); - r = ttm_eu_reserve_buffers(&p->ticket, &p->validated, true, &duplicates); - if (unlikely(r != 0)) { - return r; - } - - list_for_each_entry(lobj, &p->validated, tv.head) { + list_for_each_entry(lobj, validated, tv.head) { bo = lobj->robj; if (!bo->pin_count) { u32 domain = lobj->prefered_domains; @@ -373,7 +365,6 @@ int amdgpu_cs_list_validate(struct amdgpu_cs_parser *p) domain = lobj->allowed_domains; goto retry; } - ttm_eu_backoff_reservation(&p->ticket, &p->validated); return r; } } @@ -386,6 +377,7 @@ static int amdgpu_cs_parser_relocs(struct amdgpu_cs_parser *p) { struct amdgpu_fpriv *fpriv = p->filp->driver_priv; struct amdgpu_cs_buckets buckets; + struct list_head duplicates; bool need_mmap_lock = false; int i, r; @@ -405,8 +397,22 @@ static int amdgpu_cs_parser_relocs(struct amdgpu_cs_parser *p) if (need_mmap_lock) down_read(¤t->mm->mmap_sem); - r = amdgpu_cs_list_validate(p); + INIT_LIST_HEAD(&duplicates); + r = ttm_eu_reserve_buffers(&p->ticket, &p->validated, true, &duplicates); + if (unlikely(r != 0)) + goto error_reserve; + r = amdgpu_cs_list_validate(p->adev, &fpriv->vm, &p->validated); + if (r) + goto error_validate; + + r = amdgpu_cs_list_validate(p->adev, &fpriv->vm, &duplicates); + +error_validate: + if (r) + ttm_eu_backoff_reservation(&p->ticket, &p->validated); + +error_reserve: if (need_mmap_lock) up_read(¤t->mm->mmap_sem); From bf60efd353f68e5dec1a177b5cbe4da07c819569 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Fri, 4 Sep 2015 10:47:56 +0200 Subject: [PATCH 366/647] drm/amdgpu: use only one reservation object for each VM v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Reduces the locking and fencing overhead. v2: add comment why we need the duplicates list in the GEM op. Signed-off-by: Christian König Reviewed-by: Alex Deucher Reviewed-by: Jammy Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 6 +++-- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 33 ++++--------------------- 2 files changed, 9 insertions(+), 30 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index b75c3b235832..2f39fea10bd1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -455,11 +455,12 @@ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev, struct ttm_validate_buffer tv, *entry; struct amdgpu_bo_list_entry *vm_bos; struct ww_acquire_ctx ticket; - struct list_head list; + struct list_head list, duplicates; unsigned domain; int r; INIT_LIST_HEAD(&list); + INIT_LIST_HEAD(&duplicates); tv.bo = &bo_va->bo->tbo; tv.shared = true; @@ -469,7 +470,8 @@ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev, if (!vm_bos) return; - r = ttm_eu_reserve_buffers(&ticket, &list, true, NULL); + /* Provide duplicates to avoid -EALREADY */ + r = ttm_eu_reserve_buffers(&ticket, &list, true, &duplicates); if (r) goto error_free; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 8927dc6dec7c..d30fbecac26e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -685,31 +685,6 @@ static int amdgpu_vm_update_ptes(struct amdgpu_device *adev, return 0; } -/** - * amdgpu_vm_fence_pts - fence page tables after an update - * - * @vm: requested vm - * @start: start of GPU address range - * @end: end of GPU address range - * @fence: fence to use - * - * Fence the page tables in the range @start - @end (cayman+). - * - * Global and local mutex must be locked! - */ -static void amdgpu_vm_fence_pts(struct amdgpu_vm *vm, - uint64_t start, uint64_t end, - struct fence *fence) -{ - unsigned i; - - start >>= amdgpu_vm_block_size; - end >>= amdgpu_vm_block_size; - - for (i = start; i <= end; ++i) - amdgpu_bo_fence(vm->page_tables[i].bo, fence, true); -} - /** * amdgpu_vm_bo_update_mapping - update a mapping in the vm page table * @@ -813,8 +788,7 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, if (r) goto error_free; - amdgpu_vm_fence_pts(vm, mapping->it.start, - mapping->it.last + 1, f); + amdgpu_bo_fence(vm->page_directory, f, true); if (fence) { fence_put(*fence); *fence = fence_get(f); @@ -1089,6 +1063,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev, /* walk over the address space and allocate the page tables */ for (pt_idx = saddr; pt_idx <= eaddr; ++pt_idx) { + struct reservation_object *resv = vm->page_directory->tbo.resv; struct amdgpu_bo *pt; if (vm->page_tables[pt_idx].bo) @@ -1097,11 +1072,13 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev, /* drop mutex to allocate and clear page table */ mutex_unlock(&vm->mutex); + ww_mutex_lock(&resv->lock, NULL); r = amdgpu_bo_create(adev, AMDGPU_VM_PTE_COUNT * 8, AMDGPU_GPU_PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_NO_CPU_ACCESS, - NULL, NULL, &pt); + NULL, resv, &pt); + ww_mutex_unlock(&resv->lock); if (r) goto error_free; From 4c7eb91cae88fd2aa101750d6825b4176f85ffb2 Mon Sep 17 00:00:00 2001 From: Junwei Zhang Date: Wed, 9 Sep 2015 09:05:55 +0800 Subject: [PATCH 367/647] drm/amdgpu: refine the job naming for amdgpu_job and amdgpu_sched_job MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Use consistent naming across functions. Reviewed-by: Christian König Reviewed-by: David Zhou Signed-off-by: Junwei Zhang --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 14 ++-- drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c | 36 +++++------ drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c | 6 +- drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c | 6 +- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 8 +-- drivers/gpu/drm/amd/scheduler/gpu_scheduler.c | 64 ++++++++++--------- drivers/gpu/drm/amd/scheduler/gpu_scheduler.h | 4 +- 8 files changed, 71 insertions(+), 69 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index e587e20074d3..58fe9fab8192 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1275,7 +1275,7 @@ struct amdgpu_job { uint32_t num_ibs; struct mutex job_lock; struct amdgpu_user_fence uf; - int (*free_job)(struct amdgpu_job *sched_job); + int (*free_job)(struct amdgpu_job *job); }; static inline u32 amdgpu_get_ib_value(struct amdgpu_cs_parser *p, uint32_t ib_idx, int idx) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 4b92e3836684..4c1698861c42 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -778,15 +778,15 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev, return 0; } -static int amdgpu_cs_free_job(struct amdgpu_job *sched_job) +static int amdgpu_cs_free_job(struct amdgpu_job *job) { int i; - if (sched_job->ibs) - for (i = 0; i < sched_job->num_ibs; i++) - amdgpu_ib_free(sched_job->adev, &sched_job->ibs[i]); - kfree(sched_job->ibs); - if (sched_job->uf.bo) - drm_gem_object_unreference_unlocked(&sched_job->uf.bo->gem_base); + if (job->ibs) + for (i = 0; i < job->num_ibs; i++) + amdgpu_ib_free(job->adev, &job->ibs[i]); + kfree(job->ibs); + if (job->uf.bo) + drm_gem_object_unreference_unlocked(&job->uf.bo->gem_base); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c index 5724a81fbf5e..af1a3dabd190 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c @@ -27,42 +27,42 @@ #include #include "amdgpu.h" -static struct fence *amdgpu_sched_dependency(struct amd_sched_job *job) +static struct fence *amdgpu_sched_dependency(struct amd_sched_job *sched_job) { - struct amdgpu_job *sched_job = (struct amdgpu_job *)job; - return amdgpu_sync_get_fence(&sched_job->ibs->sync); + struct amdgpu_job *job = (struct amdgpu_job *)sched_job; + return amdgpu_sync_get_fence(&job->ibs->sync); } -static struct fence *amdgpu_sched_run_job(struct amd_sched_job *job) +static struct fence *amdgpu_sched_run_job(struct amd_sched_job *sched_job) { struct amdgpu_fence *fence = NULL; - struct amdgpu_job *sched_job; + struct amdgpu_job *job; int r; - if (!job) { + if (!sched_job) { DRM_ERROR("job is null\n"); return NULL; } - sched_job = (struct amdgpu_job *)job; - mutex_lock(&sched_job->job_lock); - r = amdgpu_ib_schedule(sched_job->adev, - sched_job->num_ibs, - sched_job->ibs, - sched_job->base.owner); + job = (struct amdgpu_job *)sched_job; + mutex_lock(&job->job_lock); + r = amdgpu_ib_schedule(job->adev, + job->num_ibs, + job->ibs, + job->base.owner); if (r) { DRM_ERROR("Error scheduling IBs (%d)\n", r); goto err; } - fence = amdgpu_fence_ref(sched_job->ibs[sched_job->num_ibs - 1].fence); + fence = amdgpu_fence_ref(job->ibs[job->num_ibs - 1].fence); err: - if (sched_job->free_job) - sched_job->free_job(sched_job); + if (job->free_job) + job->free_job(job); - mutex_unlock(&sched_job->job_lock); - fence_put(&sched_job->base.s_fence->base); - kfree(sched_job); + mutex_unlock(&job->job_lock); + fence_put(&job->base.s_fence->base); + kfree(job); return fence ? &fence->base : NULL; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 3cc86631d5d6..1a8e43b3039e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -805,10 +805,10 @@ int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, uint32_t ib_idx) } static int amdgpu_uvd_free_job( - struct amdgpu_job *sched_job) + struct amdgpu_job *job) { - amdgpu_ib_free(sched_job->adev, sched_job->ibs); - kfree(sched_job->ibs); + amdgpu_ib_free(job->adev, job->ibs); + kfree(job->ibs); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index c90caf810e4a..74f2038ac747 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -342,10 +342,10 @@ void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp) } static int amdgpu_vce_free_job( - struct amdgpu_job *sched_job) + struct amdgpu_job *job) { - amdgpu_ib_free(sched_job->adev, sched_job->ibs); - kfree(sched_job->ibs); + amdgpu_ib_free(job->adev, job->ibs); + kfree(job->ibs); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index d30fbecac26e..1e14531353e0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -316,12 +316,12 @@ static void amdgpu_vm_update_pages(struct amdgpu_device *adev, } } -int amdgpu_vm_free_job(struct amdgpu_job *sched_job) +int amdgpu_vm_free_job(struct amdgpu_job *job) { int i; - for (i = 0; i < sched_job->num_ibs; i++) - amdgpu_ib_free(sched_job->adev, &sched_job->ibs[i]); - kfree(sched_job->ibs); + for (i = 0; i < job->num_ibs; i++) + amdgpu_ib_free(job->adev, &job->ibs[i]); + kfree(job->ibs); return 0; } diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index 191fd513d979..5c8dcf89297a 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -68,29 +68,29 @@ static struct amd_sched_job * amd_sched_rq_select_job(struct amd_sched_rq *rq) { struct amd_sched_entity *entity; - struct amd_sched_job *job; + struct amd_sched_job *sched_job; spin_lock(&rq->lock); entity = rq->current_entity; if (entity) { list_for_each_entry_continue(entity, &rq->entities, list) { - job = amd_sched_entity_pop_job(entity); - if (job) { + sched_job = amd_sched_entity_pop_job(entity); + if (sched_job) { rq->current_entity = entity; spin_unlock(&rq->lock); - return job; + return sched_job; } } } list_for_each_entry(entity, &rq->entities, list) { - job = amd_sched_entity_pop_job(entity); - if (job) { + sched_job = amd_sched_entity_pop_job(entity); + if (sched_job) { rq->current_entity = entity; spin_unlock(&rq->lock); - return job; + return sched_job; } if (entity == rq->current_entity) @@ -208,15 +208,15 @@ static struct amd_sched_job * amd_sched_entity_pop_job(struct amd_sched_entity *entity) { struct amd_gpu_scheduler *sched = entity->scheduler; - struct amd_sched_job *job; + struct amd_sched_job *sched_job; if (ACCESS_ONCE(entity->dependency)) return NULL; - if (!kfifo_out_peek(&entity->job_queue, &job, sizeof(job))) + if (!kfifo_out_peek(&entity->job_queue, &sched_job, sizeof(sched_job))) return NULL; - while ((entity->dependency = sched->ops->dependency(job))) { + while ((entity->dependency = sched->ops->dependency(sched_job))) { if (fence_add_callback(entity->dependency, &entity->cb, amd_sched_entity_wakeup)) @@ -225,32 +225,33 @@ amd_sched_entity_pop_job(struct amd_sched_entity *entity) return NULL; } - return job; + return sched_job; } /** * Helper to submit a job to the job queue * - * @job The pointer to job required to submit + * @sched_job The pointer to job required to submit * * Returns true if we could submit the job. */ -static bool amd_sched_entity_in(struct amd_sched_job *job) +static bool amd_sched_entity_in(struct amd_sched_job *sched_job) { - struct amd_sched_entity *entity = job->s_entity; + struct amd_sched_entity *entity = sched_job->s_entity; bool added, first = false; spin_lock(&entity->queue_lock); - added = kfifo_in(&entity->job_queue, &job, sizeof(job)) == sizeof(job); + added = kfifo_in(&entity->job_queue, &sched_job, + sizeof(sched_job)) == sizeof(sched_job); - if (added && kfifo_len(&entity->job_queue) == sizeof(job)) + if (added && kfifo_len(&entity->job_queue) == sizeof(sched_job)) first = true; spin_unlock(&entity->queue_lock); /* first job wakes up scheduler */ if (first) - amd_sched_wakeup(job->sched); + amd_sched_wakeup(sched_job->sched); return added; } @@ -258,7 +259,7 @@ static bool amd_sched_entity_in(struct amd_sched_job *job) /** * Submit a job to the job queue * - * @job The pointer to job required to submit + * @sched_job The pointer to job required to submit * * Returns 0 for success, negative error code otherwise. */ @@ -304,17 +305,17 @@ static void amd_sched_wakeup(struct amd_gpu_scheduler *sched) static struct amd_sched_job * amd_sched_select_job(struct amd_gpu_scheduler *sched) { - struct amd_sched_job *job; + struct amd_sched_job *sched_job; if (!amd_sched_ready(sched)) return NULL; /* Kernel run queue has higher priority than normal run queue*/ - job = amd_sched_rq_select_job(&sched->kernel_rq); - if (job == NULL) - job = amd_sched_rq_select_job(&sched->sched_rq); + sched_job = amd_sched_rq_select_job(&sched->kernel_rq); + if (sched_job == NULL) + sched_job = amd_sched_rq_select_job(&sched->sched_rq); - return job; + return sched_job; } static void amd_sched_process_job(struct fence *f, struct fence_cb *cb) @@ -340,20 +341,20 @@ static int amd_sched_main(void *param) while (!kthread_should_stop()) { struct amd_sched_entity *entity; struct amd_sched_fence *s_fence; - struct amd_sched_job *job; + struct amd_sched_job *sched_job; struct fence *fence; wait_event_interruptible(sched->wake_up_worker, kthread_should_stop() || - (job = amd_sched_select_job(sched))); + (sched_job = amd_sched_select_job(sched))); - if (!job) + if (!sched_job) continue; - entity = job->s_entity; - s_fence = job->s_fence; + entity = sched_job->s_entity; + s_fence = sched_job->s_fence; atomic_inc(&sched->hw_rq_count); - fence = sched->ops->run_job(job); + fence = sched->ops->run_job(sched_job); if (fence) { r = fence_add_callback(fence, &s_fence->cb, amd_sched_process_job); @@ -367,8 +368,9 @@ static int amd_sched_main(void *param) amd_sched_process_job(NULL, &s_fence->cb); } - count = kfifo_out(&entity->job_queue, &job, sizeof(job)); - WARN_ON(count != sizeof(job)); + count = kfifo_out(&entity->job_queue, &sched_job, + sizeof(sched_job)); + WARN_ON(count != sizeof(sched_job)); wake_up(&sched->job_scheduled); } return 0; diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h index ac56d9211f7c..f33df6c70ca0 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h @@ -91,8 +91,8 @@ static inline struct amd_sched_fence *to_amd_sched_fence(struct fence *f) * these functions should be implemented in driver side */ struct amd_sched_backend_ops { - struct fence *(*dependency)(struct amd_sched_job *job); - struct fence *(*run_job)(struct amd_sched_job *job); + struct fence *(*dependency)(struct amd_sched_job *sched_job); + struct fence *(*run_job)(struct amd_sched_job *sched_job); }; /** From a6db8a33e164ae72fb5429ab637e8cfee057a722 Mon Sep 17 00:00:00 2001 From: Junwei Zhang Date: Wed, 9 Sep 2015 09:21:19 +0800 Subject: [PATCH 368/647] drm/amdgpu: refine the scheduler job type conversion MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Use container_of rather than casting. Reviewed-by: Christian König Reviewed-by: David Zhou Signed-off-by: Junwei Zhang --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c | 6 +++--- 3 files changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 58fe9fab8192..dbe061bf88cc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1277,6 +1277,8 @@ struct amdgpu_job { struct amdgpu_user_fence uf; int (*free_job)(struct amdgpu_job *job); }; +#define to_amdgpu_job(sched_job) \ + container_of((sched_job), struct amdgpu_job, base) static inline u32 amdgpu_get_ib_value(struct amdgpu_cs_parser *p, uint32_t ib_idx, int idx) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 4c1698861c42..546968a83b08 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -863,7 +863,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) job->free_job = amdgpu_cs_free_job; mutex_lock(&job->job_lock); - r = amd_sched_entity_push_job((struct amd_sched_job *)job); + r = amd_sched_entity_push_job(&job->base); if (r) { mutex_unlock(&job->job_lock); amdgpu_cs_free_job(job); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c index af1a3dabd190..58408da122c5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c @@ -29,7 +29,7 @@ static struct fence *amdgpu_sched_dependency(struct amd_sched_job *sched_job) { - struct amdgpu_job *job = (struct amdgpu_job *)sched_job; + struct amdgpu_job *job = to_amdgpu_job(sched_job); return amdgpu_sync_get_fence(&job->ibs->sync); } @@ -43,7 +43,7 @@ static struct fence *amdgpu_sched_run_job(struct amd_sched_job *sched_job) DRM_ERROR("job is null\n"); return NULL; } - job = (struct amdgpu_job *)sched_job; + job = to_amdgpu_job(sched_job); mutex_lock(&job->job_lock); r = amdgpu_ib_schedule(job->adev, job->num_ibs, @@ -94,7 +94,7 @@ int amdgpu_sched_ib_submit_kernel_helper(struct amdgpu_device *adev, mutex_init(&job->job_lock); job->free_job = free_job; mutex_lock(&job->job_lock); - r = amd_sched_entity_push_job((struct amd_sched_job *)job); + r = amd_sched_entity_push_job(&job->base); if (r) { mutex_unlock(&job->job_lock); kfree(job); From 0f75aee75112934bcaf42410df5c51d7194b5896 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 7 Sep 2015 18:07:14 +0200 Subject: [PATCH 369/647] drm/amdgpu: cleanup entity init MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Reorder the fields and properly return the kfifo_alloc error code. Signed-off-by: Christian König Reviewed-by: Junwei Zhang Reviewed-by: Chunming Zhou --- drivers/gpu/drm/amd/scheduler/gpu_scheduler.c | 30 +++++++++++-------- drivers/gpu/drm/amd/scheduler/gpu_scheduler.h | 12 ++++---- drivers/gpu/drm/amd/scheduler/sched_fence.c | 2 +- 3 files changed, 25 insertions(+), 19 deletions(-) diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index 5c8dcf89297a..a9d582634d8e 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -118,23 +118,27 @@ int amd_sched_entity_init(struct amd_gpu_scheduler *sched, struct amd_sched_rq *rq, uint32_t jobs) { + int r; + if (!(sched && entity && rq)) return -EINVAL; memset(entity, 0, sizeof(struct amd_sched_entity)); - entity->belongto_rq = rq; - entity->scheduler = sched; - entity->fence_context = fence_context_alloc(1); - if(kfifo_alloc(&entity->job_queue, - jobs * sizeof(void *), - GFP_KERNEL)) - return -EINVAL; + INIT_LIST_HEAD(&entity->list); + entity->rq = rq; + entity->sched = sched; spin_lock_init(&entity->queue_lock); + r = kfifo_alloc(&entity->job_queue, jobs * sizeof(void *), GFP_KERNEL); + if (r) + return r; + atomic_set(&entity->fence_seq, 0); + entity->fence_context = fence_context_alloc(1); /* Add the entity to the run queue */ amd_sched_rq_add_entity(rq, entity); + return 0; } @@ -149,8 +153,8 @@ int amd_sched_entity_init(struct amd_gpu_scheduler *sched, static bool amd_sched_entity_is_initialized(struct amd_gpu_scheduler *sched, struct amd_sched_entity *entity) { - return entity->scheduler == sched && - entity->belongto_rq != NULL; + return entity->sched == sched && + entity->rq != NULL; } /** @@ -180,7 +184,7 @@ static bool amd_sched_entity_is_idle(struct amd_sched_entity *entity) void amd_sched_entity_fini(struct amd_gpu_scheduler *sched, struct amd_sched_entity *entity) { - struct amd_sched_rq *rq = entity->belongto_rq; + struct amd_sched_rq *rq = entity->rq; if (!amd_sched_entity_is_initialized(sched, entity)) return; @@ -201,13 +205,13 @@ static void amd_sched_entity_wakeup(struct fence *f, struct fence_cb *cb) container_of(cb, struct amd_sched_entity, cb); entity->dependency = NULL; fence_put(f); - amd_sched_wakeup(entity->scheduler); + amd_sched_wakeup(entity->sched); } static struct amd_sched_job * amd_sched_entity_pop_job(struct amd_sched_entity *entity) { - struct amd_gpu_scheduler *sched = entity->scheduler; + struct amd_gpu_scheduler *sched = entity->sched; struct amd_sched_job *sched_job; if (ACCESS_ONCE(entity->dependency)) @@ -275,7 +279,7 @@ int amd_sched_entity_push_job(struct amd_sched_job *sched_job) fence_get(&fence->base); sched_job->s_fence = fence; - wait_event(entity->scheduler->job_scheduled, + wait_event(entity->sched->job_scheduled, amd_sched_entity_in(sched_job)); trace_amd_sched_job(sched_job); return 0; diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h index f33df6c70ca0..c4fe24e2cb21 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h @@ -38,13 +38,15 @@ struct amd_sched_rq; */ struct amd_sched_entity { struct list_head list; - struct amd_sched_rq *belongto_rq; - atomic_t fence_seq; - /* the job_queue maintains the jobs submitted by clients */ - struct kfifo job_queue; + struct amd_sched_rq *rq; + struct amd_gpu_scheduler *sched; + spinlock_t queue_lock; - struct amd_gpu_scheduler *scheduler; + struct kfifo job_queue; + + atomic_t fence_seq; uint64_t fence_context; + struct fence *dependency; struct fence_cb cb; }; diff --git a/drivers/gpu/drm/amd/scheduler/sched_fence.c b/drivers/gpu/drm/amd/scheduler/sched_fence.c index e62c37920e11..733522f7a0ea 100644 --- a/drivers/gpu/drm/amd/scheduler/sched_fence.c +++ b/drivers/gpu/drm/amd/scheduler/sched_fence.c @@ -36,7 +36,7 @@ struct amd_sched_fence *amd_sched_fence_create(struct amd_sched_entity *s_entity if (fence == NULL) return NULL; fence->owner = owner; - fence->scheduler = s_entity->scheduler; + fence->scheduler = s_entity->sched; spin_lock_init(&fence->lock); seq = atomic_inc_return(&s_entity->fence_seq); From 9b398fa5c24eb05fc60fafd8543cc03e9170f054 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 7 Sep 2015 18:16:49 +0200 Subject: [PATCH 370/647] drm/amdgpu: rename fence->scheduler to sched v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Just to be consistent with the other members. v2: rename the ring member as well. Signed-off-by: Christian König Reviewed-by: Junwei Zhang (v1) Reviewed-by: Chunming Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 10 +++++----- drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c | 2 +- drivers/gpu/drm/amd/scheduler/gpu_scheduler.c | 2 +- drivers/gpu/drm/amd/scheduler/gpu_scheduler.h | 2 +- drivers/gpu/drm/amd/scheduler/sched_fence.c | 4 ++-- 10 files changed, 22 insertions(+), 22 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index dbe061bf88cc..9108b7c7d4a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -891,7 +891,7 @@ struct amdgpu_ring { struct amdgpu_device *adev; const struct amdgpu_ring_funcs *funcs; struct amdgpu_fence_driver fence_drv; - struct amd_gpu_scheduler *scheduler; + struct amd_gpu_scheduler *sched; spinlock_t fence_lock; struct mutex *ring_lock; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 546968a83b08..6f39b2d2106d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -848,7 +848,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) job = kzalloc(sizeof(struct amdgpu_job), GFP_KERNEL); if (!job) return -ENOMEM; - job->base.sched = ring->scheduler; + job->base.sched = ring->sched; job->base.s_entity = &parser->ctx->rings[ring->idx].entity; job->adev = parser->adev; job->ibs = parser->ibs; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 20cbc4eb5a6f..5494831e1a24 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -43,10 +43,10 @@ int amdgpu_ctx_init(struct amdgpu_device *adev, bool kernel, for (i = 0; i < adev->num_rings; i++) { struct amd_sched_rq *rq; if (kernel) - rq = &adev->rings[i]->scheduler->kernel_rq; + rq = &adev->rings[i]->sched->kernel_rq; else - rq = &adev->rings[i]->scheduler->sched_rq; - r = amd_sched_entity_init(adev->rings[i]->scheduler, + rq = &adev->rings[i]->sched->sched_rq; + r = amd_sched_entity_init(adev->rings[i]->sched, &ctx->rings[i].entity, rq, amdgpu_sched_jobs); if (r) @@ -55,7 +55,7 @@ int amdgpu_ctx_init(struct amdgpu_device *adev, bool kernel, if (i < adev->num_rings) { for (j = 0; j < i; j++) - amd_sched_entity_fini(adev->rings[j]->scheduler, + amd_sched_entity_fini(adev->rings[j]->sched, &ctx->rings[j].entity); kfree(ctx); return r; @@ -75,7 +75,7 @@ void amdgpu_ctx_fini(struct amdgpu_ctx *ctx) if (amdgpu_enable_scheduler) { for (i = 0; i < adev->num_rings; i++) - amd_sched_entity_fini(adev->rings[i]->scheduler, + amd_sched_entity_fini(adev->rings[i]->sched, &ctx->rings[i].entity); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 1be2bd6d07ea..8e8cd09d36c1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -626,11 +626,11 @@ void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring) ring->fence_drv.ring = ring; if (amdgpu_enable_scheduler) { - ring->scheduler = amd_sched_create(&amdgpu_sched_ops, - ring->idx, - amdgpu_sched_hw_submission, - (void *)ring->adev); - if (!ring->scheduler) + ring->sched = amd_sched_create(&amdgpu_sched_ops, + ring->idx, + amdgpu_sched_hw_submission, + (void *)ring->adev); + if (!ring->sched) DRM_ERROR("Failed to create scheduler on ring %d.\n", ring->idx); } @@ -681,8 +681,8 @@ void amdgpu_fence_driver_fini(struct amdgpu_device *adev) wake_up_all(&ring->fence_drv.fence_queue); amdgpu_irq_put(adev, ring->fence_drv.irq_src, ring->fence_drv.irq_type); - if (ring->scheduler) - amd_sched_destroy(ring->scheduler); + if (ring->sched) + amd_sched_destroy(ring->sched); ring->fence_drv.initialized = false; } mutex_unlock(&adev->ring_lock); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c index b70ce106ca90..7cf5405afe4e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c @@ -146,7 +146,7 @@ static uint32_t amdgpu_sa_get_ring_from_fence(struct fence *f) s_fence = to_amd_sched_fence(f); if (s_fence) - return s_fence->scheduler->ring_id; + return s_fence->sched->ring_id; a_fence = to_amdgpu_fence(f); if (a_fence) return a_fence->ring->idx; @@ -437,7 +437,7 @@ void amdgpu_sa_bo_dump_debug_info(struct amdgpu_sa_manager *sa_manager, if (s_fence) seq_printf(m, " protected by 0x%016x on ring %d", s_fence->base.seqno, - s_fence->scheduler->ring_id); + s_fence->sched->ring_id); } seq_printf(m, "\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c index 58408da122c5..d1984fc5dfc4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c @@ -85,7 +85,7 @@ int amdgpu_sched_ib_submit_kernel_helper(struct amdgpu_device *adev, kzalloc(sizeof(struct amdgpu_job), GFP_KERNEL); if (!job) return -ENOMEM; - job->base.sched = ring->scheduler; + job->base.sched = ring->sched; job->base.s_entity = &adev->kernel_ctx.rings[ring->idx].entity; job->adev = adev; job->ibs = ibs; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c index cfd2999acc59..b57ca10a8533 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c @@ -66,7 +66,7 @@ static bool amdgpu_sync_same_dev(struct amdgpu_device *adev, struct fence *f) if (a_fence) return a_fence->ring->adev == adev; if (s_fence) - return (struct amdgpu_device *)s_fence->scheduler->priv == adev; + return (struct amdgpu_device *)s_fence->sched->priv == adev; return false; } diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index a9d582634d8e..ec4842e58fd7 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -326,7 +326,7 @@ static void amd_sched_process_job(struct fence *f, struct fence_cb *cb) { struct amd_sched_fence *s_fence = container_of(cb, struct amd_sched_fence, cb); - struct amd_gpu_scheduler *sched = s_fence->scheduler; + struct amd_gpu_scheduler *sched = s_fence->sched; atomic_dec(&sched->hw_rq_count); amd_sched_fence_signal(s_fence); diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h index c4fe24e2cb21..89d977dd30ac 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h @@ -65,7 +65,7 @@ struct amd_sched_rq { struct amd_sched_fence { struct fence base; struct fence_cb cb; - struct amd_gpu_scheduler *scheduler; + struct amd_gpu_scheduler *sched; spinlock_t lock; void *owner; }; diff --git a/drivers/gpu/drm/amd/scheduler/sched_fence.c b/drivers/gpu/drm/amd/scheduler/sched_fence.c index 733522f7a0ea..d802638094f4 100644 --- a/drivers/gpu/drm/amd/scheduler/sched_fence.c +++ b/drivers/gpu/drm/amd/scheduler/sched_fence.c @@ -36,7 +36,7 @@ struct amd_sched_fence *amd_sched_fence_create(struct amd_sched_entity *s_entity if (fence == NULL) return NULL; fence->owner = owner; - fence->scheduler = s_entity->sched; + fence->sched = s_entity->sched; spin_lock_init(&fence->lock); seq = atomic_inc_return(&s_entity->fence_seq); @@ -63,7 +63,7 @@ static const char *amd_sched_fence_get_driver_name(struct fence *fence) static const char *amd_sched_fence_get_timeline_name(struct fence *f) { struct amd_sched_fence *fence = to_amd_sched_fence(f); - return (const char *)fence->scheduler->name; + return (const char *)fence->sched->name; } static bool amd_sched_fence_enable_signaling(struct fence *f) From 5ec92a7692872d656cffe010920fb49c4f51d75f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 7 Sep 2015 18:43:02 +0200 Subject: [PATCH 371/647] drm/amdgpu: cleanup fence queue init v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move the fence related stuff into amdgpu_fence.c v2: rework commit message, cause this is actually not a bug Signed-off-by: Christian König Reviewed-by: Chunming Zhou Reviewed-by: Junwei Zhang --- drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 2 -- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 8e8cd09d36c1..7f2d85e7e77a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -625,6 +625,8 @@ void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring) amdgpu_fence_check_lockup); ring->fence_drv.ring = ring; + init_waitqueue_head(&ring->fence_drv.fence_queue); + if (amdgpu_enable_scheduler) { ring->sched = amd_sched_create(&amdgpu_sched_ops, ring->idx, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 130292dcf243..6e735431676d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -360,8 +360,6 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, amdgpu_fence_driver_init_ring(ring); } - init_waitqueue_head(&ring->fence_drv.fence_queue); - r = amdgpu_wb_get(adev, &ring->rptr_offs); if (r) { dev_err(adev->dev, "(%d) ring rptr_offs wb alloc failed\n", r); From 4f839a243d3b0d8b1a14f4778a87ec4d8ddbf15f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Tue, 8 Sep 2015 20:22:31 +0200 Subject: [PATCH 372/647] drm/amdgpu: more scheduler cleanups v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Embed the scheduler into the ring structure instead of allocating it. Use the ring name directly instead of the id. v2: rebased, whitespace cleanup Signed-off-by: Christian König Reviewed-by: Junwei Zhang Reviewed-by: Chunming Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 4 +- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 10 ++--- drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 23 +++++----- drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 4 +- drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c | 43 +++++++++++++------ drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c | 10 ++++- .../gpu/drm/amd/scheduler/gpu_sched_trace.h | 8 ++-- drivers/gpu/drm/amd/scheduler/gpu_scheduler.c | 37 ++++++---------- drivers/gpu/drm/amd/scheduler/gpu_scheduler.h | 20 ++++----- 11 files changed, 87 insertions(+), 76 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 9108b7c7d4a3..57b427f958da 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -433,7 +433,7 @@ int amdgpu_fence_driver_init(struct amdgpu_device *adev); void amdgpu_fence_driver_fini(struct amdgpu_device *adev); void amdgpu_fence_driver_force_completion(struct amdgpu_device *adev); -void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring); +int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring); int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring, struct amdgpu_irq_src *irq_src, unsigned irq_type); @@ -891,7 +891,7 @@ struct amdgpu_ring { struct amdgpu_device *adev; const struct amdgpu_ring_funcs *funcs; struct amdgpu_fence_driver fence_drv; - struct amd_gpu_scheduler *sched; + struct amd_gpu_scheduler sched; spinlock_t fence_lock; struct mutex *ring_lock; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 6f39b2d2106d..b74b6a8e80a6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -848,7 +848,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) job = kzalloc(sizeof(struct amdgpu_job), GFP_KERNEL); if (!job) return -ENOMEM; - job->base.sched = ring->sched; + job->base.sched = &ring->sched; job->base.s_entity = &parser->ctx->rings[ring->idx].entity; job->adev = parser->adev; job->ibs = parser->ibs; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 5494831e1a24..e0b80ccdfe8a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -43,10 +43,10 @@ int amdgpu_ctx_init(struct amdgpu_device *adev, bool kernel, for (i = 0; i < adev->num_rings; i++) { struct amd_sched_rq *rq; if (kernel) - rq = &adev->rings[i]->sched->kernel_rq; + rq = &adev->rings[i]->sched.kernel_rq; else - rq = &adev->rings[i]->sched->sched_rq; - r = amd_sched_entity_init(adev->rings[i]->sched, + rq = &adev->rings[i]->sched.sched_rq; + r = amd_sched_entity_init(&adev->rings[i]->sched, &ctx->rings[i].entity, rq, amdgpu_sched_jobs); if (r) @@ -55,7 +55,7 @@ int amdgpu_ctx_init(struct amdgpu_device *adev, bool kernel, if (i < adev->num_rings) { for (j = 0; j < i; j++) - amd_sched_entity_fini(adev->rings[j]->sched, + amd_sched_entity_fini(&adev->rings[j]->sched, &ctx->rings[j].entity); kfree(ctx); return r; @@ -75,7 +75,7 @@ void amdgpu_ctx_fini(struct amdgpu_ctx *ctx) if (amdgpu_enable_scheduler) { for (i = 0; i < adev->num_rings; i++) - amd_sched_entity_fini(adev->rings[i]->sched, + amd_sched_entity_fini(&adev->rings[i]->sched, &ctx->rings[i].entity); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 7f2d85e7e77a..b3fc26c59787 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -609,9 +609,9 @@ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring, * Init the fence driver for the requested ring (all asics). * Helper function for amdgpu_fence_driver_init(). */ -void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring) +int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring) { - int i; + int i, r; ring->fence_drv.cpu_addr = NULL; ring->fence_drv.gpu_addr = 0; @@ -628,14 +628,16 @@ void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring) init_waitqueue_head(&ring->fence_drv.fence_queue); if (amdgpu_enable_scheduler) { - ring->sched = amd_sched_create(&amdgpu_sched_ops, - ring->idx, - amdgpu_sched_hw_submission, - (void *)ring->adev); - if (!ring->sched) - DRM_ERROR("Failed to create scheduler on ring %d.\n", - ring->idx); + r = amd_sched_init(&ring->sched, &amdgpu_sched_ops, + amdgpu_sched_hw_submission, ring->name); + if (r) { + DRM_ERROR("Failed to create scheduler on ring %s.\n", + ring->name); + return r; + } } + + return 0; } /** @@ -683,8 +685,7 @@ void amdgpu_fence_driver_fini(struct amdgpu_device *adev) wake_up_all(&ring->fence_drv.fence_queue); amdgpu_irq_put(adev, ring->fence_drv.irq_src, ring->fence_drv.irq_type); - if (ring->sched) - amd_sched_destroy(ring->sched); + amd_sched_fini(&ring->sched); ring->fence_drv.initialized = false; } mutex_unlock(&adev->ring_lock); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 6e735431676d..30dce235ddeb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -357,7 +357,9 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, ring->adev = adev; ring->idx = adev->num_rings++; adev->rings[ring->idx] = ring; - amdgpu_fence_driver_init_ring(ring); + r = amdgpu_fence_driver_init_ring(ring); + if (r) + return r; } r = amdgpu_wb_get(adev, &ring->rptr_offs); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c index 7cf5405afe4e..e90712443fe9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c @@ -145,8 +145,13 @@ static uint32_t amdgpu_sa_get_ring_from_fence(struct fence *f) struct amd_sched_fence *s_fence; s_fence = to_amd_sched_fence(f); - if (s_fence) - return s_fence->sched->ring_id; + if (s_fence) { + struct amdgpu_ring *ring; + + ring = container_of(s_fence->sched, struct amdgpu_ring, sched); + return ring->idx; + } + a_fence = to_amdgpu_fence(f); if (a_fence) return a_fence->ring->idx; @@ -412,6 +417,26 @@ void amdgpu_sa_bo_free(struct amdgpu_device *adev, struct amdgpu_sa_bo **sa_bo, } #if defined(CONFIG_DEBUG_FS) + +static void amdgpu_sa_bo_dump_fence(struct fence *fence, struct seq_file *m) +{ + struct amdgpu_fence *a_fence = to_amdgpu_fence(fence); + struct amd_sched_fence *s_fence = to_amd_sched_fence(fence); + + if (a_fence) + seq_printf(m, " protected by 0x%016llx on ring %d", + a_fence->seq, a_fence->ring->idx); + + if (s_fence) { + struct amdgpu_ring *ring; + + + ring = container_of(s_fence->sched, struct amdgpu_ring, sched); + seq_printf(m, " protected by 0x%016x on ring %d", + s_fence->base.seqno, ring->idx); + } +} + void amdgpu_sa_bo_dump_debug_info(struct amdgpu_sa_manager *sa_manager, struct seq_file *m) { @@ -428,18 +453,8 @@ void amdgpu_sa_bo_dump_debug_info(struct amdgpu_sa_manager *sa_manager, } seq_printf(m, "[0x%010llx 0x%010llx] size %8lld", soffset, eoffset, eoffset - soffset); - if (i->fence) { - struct amdgpu_fence *a_fence = to_amdgpu_fence(i->fence); - struct amd_sched_fence *s_fence = to_amd_sched_fence(i->fence); - if (a_fence) - seq_printf(m, " protected by 0x%016llx on ring %d", - a_fence->seq, a_fence->ring->idx); - if (s_fence) - seq_printf(m, " protected by 0x%016x on ring %d", - s_fence->base.seqno, - s_fence->sched->ring_id); - - } + if (i->fence) + amdgpu_sa_bo_dump_fence(i->fence, m); seq_printf(m, "\n"); } spin_unlock(&sa_manager->wq.lock); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c index d1984fc5dfc4..2e946b2cad88 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c @@ -85,7 +85,7 @@ int amdgpu_sched_ib_submit_kernel_helper(struct amdgpu_device *adev, kzalloc(sizeof(struct amdgpu_job), GFP_KERNEL); if (!job) return -ENOMEM; - job->base.sched = ring->sched; + job->base.sched = &ring->sched; job->base.s_entity = &adev->kernel_ctx.rings[ring->idx].entity; job->adev = adev; job->ibs = ibs; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c index b57ca10a8533..4921de15b451 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c @@ -65,8 +65,14 @@ static bool amdgpu_sync_same_dev(struct amdgpu_device *adev, struct fence *f) if (a_fence) return a_fence->ring->adev == adev; - if (s_fence) - return (struct amdgpu_device *)s_fence->sched->priv == adev; + + if (s_fence) { + struct amdgpu_ring *ring; + + ring = container_of(s_fence->sched, struct amdgpu_ring, sched); + return ring->adev == adev; + } + return false; } diff --git a/drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h b/drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h index a1f4ece58a24..144f50acc971 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h +++ b/drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h @@ -16,21 +16,21 @@ TRACE_EVENT(amd_sched_job, TP_ARGS(sched_job), TP_STRUCT__entry( __field(struct amd_sched_entity *, entity) - __field(u32, ring_id) + __field(const char *, name) __field(u32, job_count) __field(int, hw_job_count) ), TP_fast_assign( __entry->entity = sched_job->s_entity; - __entry->ring_id = sched_job->sched->ring_id; + __entry->name = sched_job->sched->name; __entry->job_count = kfifo_len( &sched_job->s_entity->job_queue) / sizeof(sched_job); __entry->hw_job_count = atomic_read( &sched_job->sched->hw_rq_count); ), - TP_printk("entity=%p, ring=%u, job count:%u, hw job count:%d", - __entry->entity, __entry->ring_id, __entry->job_count, + TP_printk("entity=%p, ring=%s, job count:%u, hw job count:%d", + __entry->entity, __entry->name, __entry->job_count, __entry->hw_job_count) ); #endif diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index ec4842e58fd7..3697eeeecf82 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -381,56 +381,45 @@ static int amd_sched_main(void *param) } /** - * Create a gpu scheduler + * Init a gpu scheduler instance * + * @sched The pointer to the scheduler * @ops The backend operations for this scheduler. - * @ring The the ring id for the scheduler. * @hw_submissions Number of hw submissions to do. + * @name Name used for debugging * - * Return the pointer to scheduler for success, otherwise return NULL + * Return 0 on success, otherwise error code. */ -struct amd_gpu_scheduler *amd_sched_create(struct amd_sched_backend_ops *ops, - unsigned ring, unsigned hw_submission, - void *priv) +int amd_sched_init(struct amd_gpu_scheduler *sched, + struct amd_sched_backend_ops *ops, + unsigned hw_submission, const char *name) { - struct amd_gpu_scheduler *sched; - - sched = kzalloc(sizeof(struct amd_gpu_scheduler), GFP_KERNEL); - if (!sched) - return NULL; - sched->ops = ops; - sched->ring_id = ring; sched->hw_submission_limit = hw_submission; - sched->priv = priv; - snprintf(sched->name, sizeof(sched->name), "amdgpu[%d]", ring); + sched->name = name; amd_sched_rq_init(&sched->sched_rq); amd_sched_rq_init(&sched->kernel_rq); init_waitqueue_head(&sched->wake_up_worker); init_waitqueue_head(&sched->job_scheduled); atomic_set(&sched->hw_rq_count, 0); + /* Each scheduler will run on a seperate kernel thread */ sched->thread = kthread_run(amd_sched_main, sched, sched->name); if (IS_ERR(sched->thread)) { - DRM_ERROR("Failed to create scheduler for id %d.\n", ring); - kfree(sched); - return NULL; + DRM_ERROR("Failed to create scheduler for %s.\n", name); + return PTR_ERR(sched->thread); } - return sched; + return 0; } /** * Destroy a gpu scheduler * * @sched The pointer to the scheduler - * - * return 0 if succeed. -1 if failed. */ -int amd_sched_destroy(struct amd_gpu_scheduler *sched) +void amd_sched_fini(struct amd_gpu_scheduler *sched) { kthread_stop(sched->thread); - kfree(sched); - return 0; } diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h index 89d977dd30ac..80b64dc22214 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h @@ -101,23 +101,21 @@ struct amd_sched_backend_ops { * One scheduler is implemented for each hardware ring */ struct amd_gpu_scheduler { - struct task_struct *thread; + struct amd_sched_backend_ops *ops; + uint32_t hw_submission_limit; + const char *name; struct amd_sched_rq sched_rq; struct amd_sched_rq kernel_rq; - atomic_t hw_rq_count; - struct amd_sched_backend_ops *ops; - uint32_t ring_id; wait_queue_head_t wake_up_worker; wait_queue_head_t job_scheduled; - uint32_t hw_submission_limit; - char name[20]; - void *priv; + atomic_t hw_rq_count; + struct task_struct *thread; }; -struct amd_gpu_scheduler * -amd_sched_create(struct amd_sched_backend_ops *ops, - uint32_t ring, uint32_t hw_submission, void *priv); -int amd_sched_destroy(struct amd_gpu_scheduler *sched); +int amd_sched_init(struct amd_gpu_scheduler *sched, + struct amd_sched_backend_ops *ops, + uint32_t hw_submission, const char *name); +void amd_sched_fini(struct amd_gpu_scheduler *sched); int amd_sched_entity_init(struct amd_gpu_scheduler *sched, struct amd_sched_entity *entity, From 1ee4478a26cf55c8f8a6219d7e99f2b48959394d Mon Sep 17 00:00:00 2001 From: Leo Liu Date: Thu, 10 Sep 2015 13:41:38 -0400 Subject: [PATCH 373/647] drm/amdgpu: Disable UVD PG MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This causes problems with multiple suspend/resume cycles. Signed-off-by: Leo Liu Reviewed-by: Christian König Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/vi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index 552d9e75ad1b..b55ceb14fdcd 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -1400,7 +1400,8 @@ static int vi_common_early_init(void *handle) case CHIP_CARRIZO: adev->has_uvd = true; adev->cg_flags = 0; - adev->pg_flags = AMDGPU_PG_SUPPORT_UVD | AMDGPU_PG_SUPPORT_VCE; + /* Disable UVD pg */ + adev->pg_flags = /* AMDGPU_PG_SUPPORT_UVD | */AMDGPU_PG_SUPPORT_VCE; adev->external_rev_id = adev->rev_id + 0x1; if (amdgpu_smc_load_fw && smc_enabled) adev->firmware.smu_load = true; From 5146419e6feb99cfbc8dbf005dd2f62603e15efb Mon Sep 17 00:00:00 2001 From: Leo Liu Date: Tue, 15 Sep 2015 10:38:38 -0400 Subject: [PATCH 374/647] drm/amdgpu: make UVD handle checking more strict MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Invalid messages can crash the hw otherwise Ported from radeon commit a1b403da70e038ca6c6c6fe434d1d873546873a3 Signed-off-by: Leo Liu Reviewed-by: Christian König Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c | 68 +++++++++++++++---------- 1 file changed, 41 insertions(+), 27 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 1a8e43b3039e..d0312364d950 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -543,46 +543,60 @@ static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx, return -EINVAL; } - if (msg_type == 1) { + switch (msg_type) { + case 0: + /* it's a create msg, calc image size (width * height) */ + amdgpu_bo_kunmap(bo); + + /* try to alloc a new handle */ + for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) { + if (atomic_read(&adev->uvd.handles[i]) == handle) { + DRM_ERROR("Handle 0x%x already in use!\n", handle); + return -EINVAL; + } + + if (!atomic_cmpxchg(&adev->uvd.handles[i], 0, handle)) { + adev->uvd.filp[i] = ctx->parser->filp; + return 0; + } + } + + DRM_ERROR("No more free UVD handles!\n"); + return -EINVAL; + + case 1: /* it's a decode msg, calc buffer sizes */ r = amdgpu_uvd_cs_msg_decode(msg, ctx->buf_sizes); amdgpu_bo_kunmap(bo); if (r) return r; - } else if (msg_type == 2) { + /* validate the handle */ + for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) { + if (atomic_read(&adev->uvd.handles[i]) == handle) { + if (adev->uvd.filp[i] != ctx->parser->filp) { + DRM_ERROR("UVD handle collision detected!\n"); + return -EINVAL; + } + return 0; + } + } + + DRM_ERROR("Invalid UVD handle 0x%x!\n", handle); + return -ENOENT; + + case 2: /* it's a destroy msg, free the handle */ for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) atomic_cmpxchg(&adev->uvd.handles[i], handle, 0); amdgpu_bo_kunmap(bo); return 0; - } else { - /* it's a create msg */ - amdgpu_bo_kunmap(bo); - if (msg_type != 0) { - DRM_ERROR("Illegal UVD message type (%d)!\n", msg_type); - return -EINVAL; - } - - /* it's a create msg, no special handling needed */ + default: + DRM_ERROR("Illegal UVD message type (%d)!\n", msg_type); + return -EINVAL; } - - /* create or decode, validate the handle */ - for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) { - if (atomic_read(&adev->uvd.handles[i]) == handle) - return 0; - } - - /* handle not found try to alloc a new one */ - for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) { - if (!atomic_cmpxchg(&adev->uvd.handles[i], 0, handle)) { - adev->uvd.filp[i] = ctx->parser->filp; - return 0; - } - } - - DRM_ERROR("No more free UVD handles!\n"); + BUG(); return -EINVAL; } From 2bd188d0167227932be3cf5b033c0e600b01291f Mon Sep 17 00:00:00 2001 From: Leo Liu Date: Fri, 11 Sep 2015 14:22:18 -0400 Subject: [PATCH 375/647] drm/amdgpu: fix the UVD suspend sequence order MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes suspend issues with UVD. Signed-off-by: Leo Liu Reviewed-by: Christian König Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c index 5fac5da694f0..ed50dd725788 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c @@ -224,11 +224,11 @@ static int uvd_v4_2_suspend(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = uvd_v4_2_hw_fini(adev); + r = amdgpu_uvd_suspend(adev); if (r) return r; - r = amdgpu_uvd_suspend(adev); + r = uvd_v4_2_hw_fini(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c index 2d5c59c318af..9ad8b9906c0b 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c @@ -220,11 +220,11 @@ static int uvd_v5_0_suspend(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = uvd_v5_0_hw_fini(adev); + r = amdgpu_uvd_suspend(adev); if (r) return r; - r = amdgpu_uvd_suspend(adev); + r = uvd_v5_0_hw_fini(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c index d9f553fce531..a7622efef506 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c @@ -214,11 +214,11 @@ static int uvd_v6_0_suspend(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = uvd_v6_0_hw_fini(adev); + r = amdgpu_uvd_suspend(adev); if (r) return r; - r = amdgpu_uvd_suspend(adev); + r = uvd_v6_0_hw_fini(adev); if (r) return r; From 1f4452105ce39786be13b4636251377a30d1286a Mon Sep 17 00:00:00 2001 From: Leo Liu Date: Fri, 11 Sep 2015 17:09:57 -0400 Subject: [PATCH 376/647] drm/amdgpu: fix UVD suspend and resume for VI APU MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit User space passed the same handle before suspend and after resume, so we have remove the session and handle destroy, and keep the firmware untouched. Signed-off-by: Leo Liu Reviewed-by: Christian König Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c index a7622efef506..7e9934fa4193 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c @@ -214,10 +214,12 @@ static int uvd_v6_0_suspend(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_uvd_suspend(adev); - if (r) - return r; - + /* Skip this for APU for now */ + if (!(adev->flags & AMD_IS_APU)) { + r = amdgpu_uvd_suspend(adev); + if (r) + return r; + } r = uvd_v6_0_hw_fini(adev); if (r) return r; @@ -230,10 +232,12 @@ static int uvd_v6_0_resume(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_uvd_resume(adev); - if (r) - return r; - + /* Skip this for APU for now */ + if (!(adev->flags & AMD_IS_APU)) { + r = amdgpu_uvd_resume(adev); + if (r) + return r; + } r = uvd_v6_0_hw_init(adev); if (r) return r; From 71affda522bb0f43e205cf4f000e2c50261c01a6 Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Mon, 21 Sep 2015 17:34:39 -0400 Subject: [PATCH 377/647] drm/amdgpu: use kmemdup rather than duplicating its implementation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The patch was generated using fixed coccinelle semantic patch scripts/coccinelle/api/memdup.cocci [1]. [1]: http://permalink.gmane.org/gmane.linux.kernel/2014320 Signed-off-by: Andrzej Hajda Reviewed-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index f25cfed33927..1a7708f365f3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -536,12 +536,10 @@ int amdgpu_bo_set_metadata (struct amdgpu_bo *bo, void *metadata, if (metadata == NULL) return -EINVAL; - buffer = kzalloc(metadata_size, GFP_KERNEL); + buffer = kmemdup(metadata, metadata_size, GFP_KERNEL); if (buffer == NULL) return -ENOMEM; - memcpy(buffer, metadata, metadata_size); - kfree(bo->metadata); bo->metadata_flags = flags; bo->metadata = buffer; From 5a6adfa20b622a273205e33b20c12332aa7eb724 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 22 Sep 2015 10:06:45 -0400 Subject: [PATCH 378/647] drm/amdgpu: Fix max_vblank_count value for current display engines MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The value was much too low, which could cause the userspace visible vblank counter to move backwards when the hardware counter wrapped around. Ported from radeon commit: b0b9bb4dd51f396dcf843831905f729e74b0c8c0 Reviewed-by: Christian König Reviewed-by: Jammy Zhou Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index 0aba8e9bc8a0..7c42ff670080 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -140,7 +140,7 @@ void amdgpu_irq_preinstall(struct drm_device *dev) */ int amdgpu_irq_postinstall(struct drm_device *dev) { - dev->max_vblank_count = 0x001fffff; + dev->max_vblank_count = 0x00ffffff; return 0; } From 1d263474c4416efb6d0feca98fe6d462b0d28f56 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 23 Sep 2015 13:59:28 +0300 Subject: [PATCH 379/647] drm/amdgpu: unwind properly in amdgpu_cs_parser_init() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The amdgpu_cs_parser_init() function doesn't clean up after itself but instead the caller uses a free everything function amdgpu_cs_parser_fini() on failure. This style of error handling is often buggy. In this example, we call "drm_free_large(parser->chunks[i].kdata);" when it is an unintialized pointer or when "parser->chunks" is NULL. I fixed this bug by adding unwind code so that it frees everything that it allocates. I also mode some other very minor changes: 1) Renamed "r" to "ret". 2) Moved the chunk_array allocation to the start of the function. 3) Removed some initializers which are no longer needed. Reviewed-by: Christian König Reported-by: Ilja Van Sprundel Signed-off-by: Dan Carpenter Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 85 +++++++++++++++----------- 1 file changed, 51 insertions(+), 34 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index b74b6a8e80a6..749420f1ea6f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -154,42 +154,41 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) { union drm_amdgpu_cs *cs = data; uint64_t *chunk_array_user; - uint64_t *chunk_array = NULL; + uint64_t *chunk_array; struct amdgpu_fpriv *fpriv = p->filp->driver_priv; unsigned size, i; - int r = 0; + int ret; - if (!cs->in.num_chunks) - goto out; + if (cs->in.num_chunks == 0) + return 0; + + chunk_array = kmalloc_array(cs->in.num_chunks, sizeof(uint64_t), GFP_KERNEL); + if (!chunk_array) + return -ENOMEM; p->ctx = amdgpu_ctx_get(fpriv, cs->in.ctx_id); if (!p->ctx) { - r = -EINVAL; - goto out; + ret = -EINVAL; + goto free_chunk; } + p->bo_list = amdgpu_bo_list_get(fpriv, cs->in.bo_list_handle); /* get chunks */ INIT_LIST_HEAD(&p->validated); - chunk_array = kmalloc_array(cs->in.num_chunks, sizeof(uint64_t), GFP_KERNEL); - if (chunk_array == NULL) { - r = -ENOMEM; - goto out; - } - chunk_array_user = (uint64_t __user *)(cs->in.chunks); if (copy_from_user(chunk_array, chunk_array_user, sizeof(uint64_t)*cs->in.num_chunks)) { - r = -EFAULT; - goto out; + ret = -EFAULT; + goto put_bo_list; } p->nchunks = cs->in.num_chunks; p->chunks = kmalloc_array(p->nchunks, sizeof(struct amdgpu_cs_chunk), GFP_KERNEL); - if (p->chunks == NULL) { - r = -ENOMEM; - goto out; + if (!p->chunks) { + ret = -ENOMEM; + goto put_bo_list; } for (i = 0; i < p->nchunks; i++) { @@ -200,8 +199,9 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) chunk_ptr = (void __user *)chunk_array[i]; if (copy_from_user(&user_chunk, chunk_ptr, sizeof(struct drm_amdgpu_cs_chunk))) { - r = -EFAULT; - goto out; + ret = -EFAULT; + i--; + goto free_partial_kdata; } p->chunks[i].chunk_id = user_chunk.chunk_id; p->chunks[i].length_dw = user_chunk.length_dw; @@ -212,13 +212,14 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) p->chunks[i].kdata = drm_malloc_ab(size, sizeof(uint32_t)); if (p->chunks[i].kdata == NULL) { - r = -ENOMEM; - goto out; + ret = -ENOMEM; + i--; + goto free_partial_kdata; } size *= sizeof(uint32_t); if (copy_from_user(p->chunks[i].kdata, cdata, size)) { - r = -EFAULT; - goto out; + ret = -EFAULT; + goto free_partial_kdata; } switch (p->chunks[i].chunk_id) { @@ -238,15 +239,15 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) gobj = drm_gem_object_lookup(p->adev->ddev, p->filp, handle); if (gobj == NULL) { - r = -EINVAL; - goto out; + ret = -EINVAL; + goto free_partial_kdata; } p->uf.bo = gem_to_amdgpu_bo(gobj); p->uf.offset = fence_data->offset; } else { - r = -EINVAL; - goto out; + ret = -EINVAL; + goto free_partial_kdata; } break; @@ -254,19 +255,35 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) break; default: - r = -EINVAL; - goto out; + ret = -EINVAL; + goto free_partial_kdata; } } p->ibs = kcalloc(p->num_ibs, sizeof(struct amdgpu_ib), GFP_KERNEL); - if (!p->ibs) - r = -ENOMEM; + if (!p->ibs) { + ret = -ENOMEM; + goto free_all_kdata; + } -out: kfree(chunk_array); - return r; + return 0; + +free_all_kdata: + i = p->nchunks - 1; +free_partial_kdata: + for (; i >= 0; i--) + drm_free_large(p->chunks[i].kdata); + kfree(p->chunks); +put_bo_list: + if (p->bo_list) + amdgpu_bo_list_put(p->bo_list); + amdgpu_ctx_put(p->ctx); +free_chunk: + kfree(chunk_array); + + return ret; } /* Returns how many bytes TTM can move per IB. @@ -810,7 +827,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) r = amdgpu_cs_parser_init(parser, data); if (r) { DRM_ERROR("Failed to initialize parser !\n"); - amdgpu_cs_parser_fini(parser, r, false); + kfree(parser); up_read(&adev->exclusive_lock); r = amdgpu_cs_handle_lockup(adev, r); return r; From 0d2edd3791bb172a59d708d5c94330bbd6050f97 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 23 Sep 2015 14:00:12 +0300 Subject: [PATCH 380/647] drm/amdgpu: integer overflow in amdgpu_info_ioctl() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The "alloc_size" calculation can overflow leading to memory corruption. Reviewed-by: Christian König Reported-by: Ilja Van Sprundel Signed-off-by: Dan Carpenter Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 22367939ebf1..8c735f544b66 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -390,7 +390,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file min((size_t)size, sizeof(vram_gtt))) ? -EFAULT : 0; } case AMDGPU_INFO_READ_MMR_REG: { - unsigned n, alloc_size = info->read_mmr_reg.count * 4; + unsigned n, alloc_size; uint32_t *regs; unsigned se_num = (info->read_mmr_reg.instance >> AMDGPU_INFO_MMR_SE_INDEX_SHIFT) & @@ -406,9 +406,10 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file if (sh_num == AMDGPU_INFO_MMR_SH_INDEX_MASK) sh_num = 0xffffffff; - regs = kmalloc(alloc_size, GFP_KERNEL); + regs = kmalloc_array(info->read_mmr_reg.count, sizeof(*regs), GFP_KERNEL); if (!regs) return -ENOMEM; + alloc_size = info->read_mmr_reg.count * sizeof(*regs); for (i = 0; i < info->read_mmr_reg.count; i++) if (amdgpu_asic_read_register(adev, se_num, sh_num, From 0913eab648e4fb30ddca8882e707d0fcf5b237c6 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 23 Sep 2015 14:00:35 +0300 Subject: [PATCH 381/647] drm/amdgpu: info leak in amdgpu_gem_metadata_ioctl() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is no limit on args->data.data_size_bytes so we could read beyond the end of the args->data.data[] array. Reviewed-by: Christian König Reported-by: Ilja Van Sprundel Signed-off-by: Dan Carpenter Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 2f39fea10bd1..b82fab2cc888 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -427,6 +427,10 @@ int amdgpu_gem_metadata_ioctl(struct drm_device *dev, void *data, &args->data.data_size_bytes, &args->data.flags); } else if (args->op == AMDGPU_GEM_METADATA_OP_SET_METADATA) { + if (args->data.data_size_bytes > sizeof(args->data.data)) { + r = -EINVAL; + goto unreserve; + } r = amdgpu_bo_set_tiling_flags(robj, args->data.tiling_info); if (!r) r = amdgpu_bo_set_metadata(robj, args->data.data, @@ -434,6 +438,7 @@ int amdgpu_gem_metadata_ioctl(struct drm_device *dev, void *data, args->data.flags); } +unreserve: amdgpu_bo_unreserve(robj); out: drm_gem_object_unreference_unlocked(gobj); From 54ef0b5461c071050c61e501af5544842d61f40a Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 23 Sep 2015 14:00:59 +0300 Subject: [PATCH 382/647] drm/amdgpu: integer overflow in amdgpu_mode_dumb_create() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit args->size is a u64. arg->pitch and args->height are u32. The multiplication will overflow instead of using the high 32 bits as intended. Reviewed-by: Christian König Signed-off-by: Dan Carpenter Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index b82fab2cc888..7297ca3a0ba7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -659,7 +659,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv, int r; args->pitch = amdgpu_align_pitch(adev, args->width, args->bpp, 0) * ((args->bpp + 1) / 8); - args->size = args->pitch * args->height; + args->size = (u64)args->pitch * args->height; args->size = ALIGN(args->size, PAGE_SIZE); r = amdgpu_gem_object_create(adev, args->size, 0, From 5c3422b0b135b46c8dca9c1d909c1ae84f3561bd Mon Sep 17 00:00:00 2001 From: "monk.liu" Date: Wed, 23 Sep 2015 13:49:58 +0800 Subject: [PATCH 383/647] drm/amdgpu: sync ce and me with SWITCH_BUFFER(2) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit we used to adopt wait_reg_mem to let CE wait before DE finish page updating, but from Tonga+, CE doesn't support wait_reg_mem package so this logic no longer works. so here is another approach to do same thing: Insert two of SWITCH_BUFFER at both front and end of vm_flush can guarantee that CE not go further to process IB_const before vm_flush done. Insert two of SWITCH_BUFFER also works on CI, so remove legency method to sync CE and ME v2: Insert double SWITCH_BUFFER at front of vm flush as well. Signed-off-by: monk.liu Reviewed-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 - drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c | 55 +++++-------------------- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 58 ++++++--------------------- 3 files changed, 23 insertions(+), 92 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 57b427f958da..6647fb26ef25 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1202,8 +1202,6 @@ struct amdgpu_gfx { struct amdgpu_irq_src priv_inst_irq; /* gfx status */ uint32_t gfx_current_status; - /* sync signal for const engine */ - unsigned ce_sync_offs; /* ce ram size*/ unsigned ce_ram_size; }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 392ec10cc475..e992bf2ff66c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -3610,41 +3610,6 @@ static int gfx_v7_0_cp_resume(struct amdgpu_device *adev) return 0; } -static void gfx_v7_0_ce_sync_me(struct amdgpu_ring *ring) -{ - struct amdgpu_device *adev = ring->adev; - u64 gpu_addr = adev->wb.gpu_addr + adev->gfx.ce_sync_offs * 4; - - /* instruct DE to set a magic number */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); - amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | - WRITE_DATA_DST_SEL(5))); - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 1); - - /* let CE wait till condition satisfied */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); - amdgpu_ring_write(ring, (WAIT_REG_MEM_OPERATION(0) | /* wait */ - WAIT_REG_MEM_MEM_SPACE(1) | /* memory */ - WAIT_REG_MEM_FUNCTION(3) | /* == */ - WAIT_REG_MEM_ENGINE(2))); /* ce */ - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 1); - amdgpu_ring_write(ring, 0xffffffff); - amdgpu_ring_write(ring, 4); /* poll interval */ - - /* instruct CE to reset wb of ce_sync to zero */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); - amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(2) | - WRITE_DATA_DST_SEL(5) | - WR_CONFIRM)); - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 0); -} - /* * vm * VMID 0 is the physical GPU addresses as used by the kernel. @@ -3663,6 +3628,13 @@ static void gfx_v7_0_ring_emit_vm_flush(struct amdgpu_ring *ring, unsigned vm_id, uint64_t pd_addr) { int usepfp = (ring->type == AMDGPU_RING_TYPE_GFX); + if (usepfp) { + /* synce CE with ME to prevent CE fetch CEIB before context switch done */ + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + } amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) | @@ -3703,7 +3675,10 @@ static void gfx_v7_0_ring_emit_vm_flush(struct amdgpu_ring *ring, amdgpu_ring_write(ring, 0x0); /* synce CE with ME to prevent CE fetch CEIB before context switch done */ - gfx_v7_0_ce_sync_me(ring); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); } } @@ -4805,12 +4780,6 @@ static int gfx_v7_0_sw_init(void *handle) return r; } - r = amdgpu_wb_get(adev, &adev->gfx.ce_sync_offs); - if (r) { - DRM_ERROR("(%d) gfx.ce_sync_offs wb alloc failed\n", r); - return r; - } - for (i = 0; i < adev->gfx.num_gfx_rings; i++) { ring = &adev->gfx.gfx_ring[i]; ring->ring_obj = NULL; @@ -4889,8 +4858,6 @@ static int gfx_v7_0_sw_fini(void *handle) for (i = 0; i < adev->gfx.num_compute_rings; i++) amdgpu_ring_fini(&adev->gfx.compute_ring[i]); - amdgpu_wb_free(adev, adev->gfx.ce_sync_offs); - gfx_v7_0_cp_compute_fini(adev); gfx_v7_0_rlc_fini(adev); gfx_v7_0_mec_fini(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 78e5900d71cd..cb4f68f53f24 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -940,12 +940,6 @@ static int gfx_v8_0_sw_init(void *handle) return r; } - r = amdgpu_wb_get(adev, &adev->gfx.ce_sync_offs); - if (r) { - DRM_ERROR("(%d) gfx.ce_sync_offs wb alloc failed\n", r); - return r; - } - /* set up the gfx ring */ for (i = 0; i < adev->gfx.num_gfx_rings; i++) { ring = &adev->gfx.gfx_ring[i]; @@ -1033,8 +1027,6 @@ static int gfx_v8_0_sw_fini(void *handle) for (i = 0; i < adev->gfx.num_compute_rings; i++) amdgpu_ring_fini(&adev->gfx.compute_ring[i]); - amdgpu_wb_free(adev, adev->gfx.ce_sync_offs); - gfx_v8_0_mec_fini(adev); return 0; @@ -4006,41 +3998,6 @@ static bool gfx_v8_0_ring_emit_semaphore(struct amdgpu_ring *ring, return true; } -static void gfx_v8_0_ce_sync_me(struct amdgpu_ring *ring) -{ - struct amdgpu_device *adev = ring->adev; - u64 gpu_addr = adev->wb.gpu_addr + adev->gfx.ce_sync_offs * 4; - - /* instruct DE to set a magic number */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); - amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | - WRITE_DATA_DST_SEL(5))); - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 1); - - /* let CE wait till condition satisfied */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); - amdgpu_ring_write(ring, (WAIT_REG_MEM_OPERATION(0) | /* wait */ - WAIT_REG_MEM_MEM_SPACE(1) | /* memory */ - WAIT_REG_MEM_FUNCTION(3) | /* == */ - WAIT_REG_MEM_ENGINE(2))); /* ce */ - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 1); - amdgpu_ring_write(ring, 0xffffffff); - amdgpu_ring_write(ring, 4); /* poll interval */ - - /* instruct CE to reset wb of ce_sync to zero */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); - amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(2) | - WRITE_DATA_DST_SEL(5) | - WR_CONFIRM)); - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 0); -} - static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring, unsigned vm_id, uint64_t pd_addr) { @@ -4057,6 +4014,14 @@ static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring, amdgpu_ring_write(ring, 0xffffffff); amdgpu_ring_write(ring, 4); /* poll interval */ + if (usepfp) { + /* synce CE with ME to prevent CE fetch CEIB before context switch done */ + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + } + amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) | WRITE_DATA_DST_SEL(0)) | @@ -4096,9 +4061,10 @@ static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring, /* sync PFP to ME, otherwise we might get invalid PFP reads */ amdgpu_ring_write(ring, PACKET3(PACKET3_PFP_SYNC_ME, 0)); amdgpu_ring_write(ring, 0x0); - - /* synce CE with ME to prevent CE fetch CEIB before context switch done */ - gfx_v8_0_ce_sync_me(ring); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); } } From 6adaed5bfe4f6f0a0e027e87d5dd80bd9834d5f0 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 23 Sep 2015 20:26:45 +0200 Subject: [PATCH 384/647] drm/radeon: Sprinkle drm_modeset_lock_all to appease locking checks In commit 7a3f3d6667f5f9ffd1517f6b21d64bbf5312042c Author: Daniel Vetter Date: Thu Jul 9 23:44:28 2015 +0200 drm: Check locking in drm_for_each_connector I added locking checks to drm_for_each_connector but failed that through drm_helper_connector_dpms -> drm_helper_choose_encoder_dpms it's used in a few more places in the radeon resume/suspend code. Fix them up. Note that we could use the connector iterator macros in there too, but that's for the future. Reported-and-tested-by: Borislav Petkov Cc: Borislav Petkov Cc: Alex Deucher Signed-off-by: Daniel Vetter Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/radeon_device.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index d8319dae8358..f3f562f6d848 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1573,10 +1573,12 @@ int radeon_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon) drm_kms_helper_poll_disable(dev); + drm_modeset_lock_all(dev); /* turn off display hw */ list_for_each_entry(connector, &dev->mode_config.connector_list, head) { drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF); } + drm_modeset_unlock_all(dev); /* unpin the front buffers and cursors */ list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { @@ -1734,9 +1736,11 @@ int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon) if (fbcon) { drm_helper_resume_force_mode(dev); /* turn on display hw */ + drm_modeset_lock_all(dev); list_for_each_entry(connector, &dev->mode_config.connector_list, head) { drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON); } + drm_modeset_unlock_all(dev); } drm_kms_helper_poll_enable(dev); From 4c7fbc39b1d58d9f4113ef962743a67bcdfe6be2 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 23 Sep 2015 14:32:06 -0400 Subject: [PATCH 385/647] drm/amdgpu: Sprinkle drm_modeset_lock_all to appease locking checks In commit 7a3f3d6667f5f9ffd1517f6b21d64bbf5312042c Author: Daniel Vetter Date: Thu Jul 9 23:44:28 2015 +0200 drm: Check locking in drm_for_each_connector I added locking checks to drm_for_each_connector but failed that through drm_helper_connector_dpms -> drm_helper_choose_encoder_dpms it's used in a few more places in the amdgpu resume/suspend code. Fix them up. Note that we could use the connector iterator macros in there too, but that's for the future. Port of radeon commit: drm/radeon: Sprinkle drm_modeset_lock_all to appease locking checks Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 2d569eccf41f..6068d8207d10 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1651,9 +1651,11 @@ int amdgpu_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon) drm_kms_helper_poll_disable(dev); /* turn off display hw */ + drm_modeset_lock_all(dev); list_for_each_entry(connector, &dev->mode_config.connector_list, head) { drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF); } + drm_modeset_unlock_all(dev); /* unpin the front buffers */ list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { @@ -1748,9 +1750,11 @@ int amdgpu_resume_kms(struct drm_device *dev, bool resume, bool fbcon) if (fbcon) { drm_helper_resume_force_mode(dev); /* turn on display hw */ + drm_modeset_lock_all(dev); list_for_each_entry(connector, &dev->mode_config.connector_list, head) { drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON); } + drm_modeset_unlock_all(dev); } drm_kms_helper_poll_enable(dev); From e78654799135a788a941bacad3452fbd7083e518 Mon Sep 17 00:00:00 2001 From: Maxim Sheviakov Date: Wed, 23 Sep 2015 17:10:51 -0400 Subject: [PATCH 386/647] drm/radeon: add quirk for MSI R7 370 Just adds the quirk for MSI R7 370 Armor 2X Bug: https://bugs.freedesktop.org/show_bug.cgi?id=91294 Signed-off-by: Maxim Sheviakov Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/si_dpm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index 787cd8fd897f..e9115d3f67b0 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -2927,6 +2927,7 @@ static struct si_dpm_quirk si_dpm_quirk_list[] = { { PCI_VENDOR_ID_ATI, 0x6810, 0x1462, 0x3036, 0, 120000 }, { PCI_VENDOR_ID_ATI, 0x6811, 0x174b, 0xe271, 0, 120000 }, { PCI_VENDOR_ID_ATI, 0x6810, 0x174b, 0xe271, 85000, 90000 }, + { PCI_VENDOR_ID_ATI, 0x6811, 0x1762, 0x2015, 0, 120000 }, { 0, 0, 0, 0 }, }; From 675ee231d960af2af3606b4480324e26797eb010 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 23 Sep 2015 14:00:21 -0700 Subject: [PATCH 387/647] tcp: add proper TS val into RST packets RST packets sent on behalf of TCP connections with TS option (RFC 7323 TCP timestamps) have incorrect TS val (set to 0), but correct TS ecr. A > B: Flags [S], seq 0, win 65535, options [mss 1000,nop,nop,TS val 100 ecr 0], length 0 B > A: Flags [S.], seq 2444755794, ack 1, win 28960, options [mss 1460,nop,nop,TS val 7264344 ecr 100], length 0 A > B: Flags [.], ack 1, win 65535, options [nop,nop,TS val 110 ecr 7264344], length 0 B > A: Flags [R.], seq 1, ack 1, win 28960, options [nop,nop,TS val 0 ecr 110], length 0 We need to call skb_mstamp_get() to get proper TS val, derived from skb->skb_mstamp Note that RFC 1323 was advocating to not send TS option in RST segment, but RFC 7323 recommends the opposite : Once TSopt has been successfully negotiated, that is both and contain TSopt, the TSopt MUST be sent in every non- segment for the duration of the connection, and SHOULD be sent in an segment (see Section 5.2 for details) Note this RFC recommends to send TS val = 0, but we believe it is premature : We do not know if all TCP stacks are properly handling the receive side : When an segment is received, it MUST NOT be subjected to the PAWS check by verifying an acceptable value in SEG.TSval, and information from the Timestamps option MUST NOT be used to update connection state information. SEG.TSecr MAY be used to provide stricter acceptance checks. In 5 years, if/when all TCP stack are RFC 7323 ready, we might consider to decide to send TS val = 0, if it buys something. Fixes: 7faee5c0d514 ("tcp: remove TCP_SKB_CB(skb)->when") Signed-off-by: Eric Dumazet Acked-by: Yuchung Cheng Signed-off-by: David S. Miller --- net/ipv4/tcp_output.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f9a8a12b62ee..1100ffe4a722 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2897,6 +2897,7 @@ void tcp_send_active_reset(struct sock *sk, gfp_t priority) skb_reserve(skb, MAX_TCP_HEADER); tcp_init_nondata_skb(skb, tcp_acceptable_seq(sk), TCPHDR_ACK | TCPHDR_RST); + skb_mstamp_get(&skb->skb_mstamp); /* Send it off. */ if (tcp_transmit_skb(sk, skb, 0, priority)) NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTFAILED); From 2d8bff12699abc3a9bf886bb0b79f44d94d81496 Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Wed, 23 Sep 2015 14:57:58 -0400 Subject: [PATCH 388/647] netpoll: Close race condition between poll_one_napi and napi_disable Drivers might call napi_disable while not holding the napi instance poll_lock. In those instances, its possible for a race condition to exist between poll_one_napi and napi_disable. That is to say, poll_one_napi only tests the NAPI_STATE_SCHED bit to see if there is work to do during a poll, and as such the following may happen: CPU0 CPU1 ndo_tx_timeout napi_poll_dev napi_disable poll_one_napi test_and_set_bit (ret 0) test_bit (ret 1) reset adapter napi_poll_routine If the adapter gets a tx timeout without a napi instance scheduled, its possible for the adapter to think it has exclusive access to the hardware (as the napi instance is now scheduled via the napi_disable call), while the netpoll code thinks there is simply work to do. The result is parallel hardware access leading to corrupt data structures in the driver, and a crash. Additionaly, there is another, more critical race between netpoll and napi_disable. The disabled napi state is actually identical to the scheduled state for a given napi instance. The implication being that, if a napi instance is disabled, a netconsole instance would see the napi state of the device as having been scheduled, and poll it, likely while the driver was dong something requiring exclusive access. In the case above, its fairly clear that not having the rings in a state ready to be polled will cause any number of crashes. The fix should be pretty easy. netpoll uses its own bit to indicate that that the napi instance is in a state of being serviced by netpoll (NAPI_STATE_NPSVC). We can just gate disabling on that bit as well as the sched bit. That should prevent netpoll from conducting a napi poll if we convert its set bit to a test_and_set_bit operation to provide mutual exclusion Change notes: V2) Remove a trailing whtiespace Resubmit with proper subject prefix V3) Clean up spacing nits Signed-off-by: Neil Horman CC: "David S. Miller" CC: jmaxwell@redhat.com Tested-by: jmaxwell@redhat.com Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 + net/core/dev.c | 2 ++ net/core/netpoll.c | 10 ++++++++-- 3 files changed, 11 insertions(+), 2 deletions(-) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 88a00694eda5..2d15e3831440 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -507,6 +507,7 @@ static inline void napi_enable(struct napi_struct *n) BUG_ON(!test_bit(NAPI_STATE_SCHED, &n->state)); smp_mb__before_atomic(); clear_bit(NAPI_STATE_SCHED, &n->state); + clear_bit(NAPI_STATE_NPSVC, &n->state); } #ifdef CONFIG_SMP diff --git a/net/core/dev.c b/net/core/dev.c index 877c84834d81..6bb6470f5b7b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4713,6 +4713,8 @@ void napi_disable(struct napi_struct *n) while (test_and_set_bit(NAPI_STATE_SCHED, &n->state)) msleep(1); + while (test_and_set_bit(NAPI_STATE_NPSVC, &n->state)) + msleep(1); hrtimer_cancel(&n->timer); diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 6aa3db8dfc3b..8bdada242a7d 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -142,7 +142,7 @@ static void queue_process(struct work_struct *work) */ static int poll_one_napi(struct napi_struct *napi, int budget) { - int work; + int work = 0; /* net_rx_action's ->poll() invocations and our's are * synchronized by this test which is only made while @@ -151,7 +151,12 @@ static int poll_one_napi(struct napi_struct *napi, int budget) if (!test_bit(NAPI_STATE_SCHED, &napi->state)) return budget; - set_bit(NAPI_STATE_NPSVC, &napi->state); + /* If we set this bit but see that it has already been set, + * that indicates that napi has been disabled and we need + * to abort this operation + */ + if (test_and_set_bit(NAPI_STATE_NPSVC, &napi->state)) + goto out; work = napi->poll(napi, budget); WARN_ONCE(work > budget, "%pF exceeded budget in poll\n", napi->poll); @@ -159,6 +164,7 @@ static int poll_one_napi(struct napi_struct *napi, int budget) clear_bit(NAPI_STATE_NPSVC, &napi->state); +out: return budget - work; } From d3869efe7a8a2298516d9af4f91487cf486ca945 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 23 Sep 2015 19:45:08 +0100 Subject: [PATCH 389/647] Fix AF_PACKET ABI breakage in 4.2 Commit 7d82410950aa ("virtio: add explicit big-endian support to memory accessors") accidentally changed the virtio_net header used by AF_PACKET with PACKET_VNET_HDR from host-endian to big-endian. Since virtio_legacy_is_little_endian() is a very long identifier, define a vio_le macro and use that throughout the code instead of the hard-coded 'false' for little-endian. This restores the ABI to match 4.1 and earlier kernels, and makes my test program work again. Signed-off-by: David Woodhouse Signed-off-by: David S. Miller --- net/packet/af_packet.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 7b8e39a22387..aa4b15c35884 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -230,6 +230,8 @@ struct packet_skb_cb { } sa; }; +#define vio_le() virtio_legacy_is_little_endian() + #define PACKET_SKB_CB(__skb) ((struct packet_skb_cb *)((__skb)->cb)) #define GET_PBDQC_FROM_RB(x) ((struct tpacket_kbdq_core *)(&(x)->prb_bdqc)) @@ -2680,15 +2682,15 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) goto out_unlock; if ((vnet_hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) && - (__virtio16_to_cpu(false, vnet_hdr.csum_start) + - __virtio16_to_cpu(false, vnet_hdr.csum_offset) + 2 > - __virtio16_to_cpu(false, vnet_hdr.hdr_len))) - vnet_hdr.hdr_len = __cpu_to_virtio16(false, - __virtio16_to_cpu(false, vnet_hdr.csum_start) + - __virtio16_to_cpu(false, vnet_hdr.csum_offset) + 2); + (__virtio16_to_cpu(vio_le(), vnet_hdr.csum_start) + + __virtio16_to_cpu(vio_le(), vnet_hdr.csum_offset) + 2 > + __virtio16_to_cpu(vio_le(), vnet_hdr.hdr_len))) + vnet_hdr.hdr_len = __cpu_to_virtio16(vio_le(), + __virtio16_to_cpu(vio_le(), vnet_hdr.csum_start) + + __virtio16_to_cpu(vio_le(), vnet_hdr.csum_offset) + 2); err = -EINVAL; - if (__virtio16_to_cpu(false, vnet_hdr.hdr_len) > len) + if (__virtio16_to_cpu(vio_le(), vnet_hdr.hdr_len) > len) goto out_unlock; if (vnet_hdr.gso_type != VIRTIO_NET_HDR_GSO_NONE) { @@ -2731,7 +2733,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) hlen = LL_RESERVED_SPACE(dev); tlen = dev->needed_tailroom; skb = packet_alloc_skb(sk, hlen + tlen, hlen, len, - __virtio16_to_cpu(false, vnet_hdr.hdr_len), + __virtio16_to_cpu(vio_le(), vnet_hdr.hdr_len), msg->msg_flags & MSG_DONTWAIT, &err); if (skb == NULL) goto out_unlock; @@ -2778,8 +2780,8 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) if (po->has_vnet_hdr) { if (vnet_hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) { - u16 s = __virtio16_to_cpu(false, vnet_hdr.csum_start); - u16 o = __virtio16_to_cpu(false, vnet_hdr.csum_offset); + u16 s = __virtio16_to_cpu(vio_le(), vnet_hdr.csum_start); + u16 o = __virtio16_to_cpu(vio_le(), vnet_hdr.csum_offset); if (!skb_partial_csum_set(skb, s, o)) { err = -EINVAL; goto out_free; @@ -2787,7 +2789,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) } skb_shinfo(skb)->gso_size = - __virtio16_to_cpu(false, vnet_hdr.gso_size); + __virtio16_to_cpu(vio_le(), vnet_hdr.gso_size); skb_shinfo(skb)->gso_type = gso_type; /* Header must be checked, and gso_segs computed. */ @@ -3161,9 +3163,9 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, /* This is a hint as to how much should be linear. */ vnet_hdr.hdr_len = - __cpu_to_virtio16(false, skb_headlen(skb)); + __cpu_to_virtio16(vio_le(), skb_headlen(skb)); vnet_hdr.gso_size = - __cpu_to_virtio16(false, sinfo->gso_size); + __cpu_to_virtio16(vio_le(), sinfo->gso_size); if (sinfo->gso_type & SKB_GSO_TCPV4) vnet_hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV4; else if (sinfo->gso_type & SKB_GSO_TCPV6) @@ -3181,9 +3183,9 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, if (skb->ip_summed == CHECKSUM_PARTIAL) { vnet_hdr.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM; - vnet_hdr.csum_start = __cpu_to_virtio16(false, + vnet_hdr.csum_start = __cpu_to_virtio16(vio_le(), skb_checksum_start_offset(skb)); - vnet_hdr.csum_offset = __cpu_to_virtio16(false, + vnet_hdr.csum_offset = __cpu_to_virtio16(vio_le(), skb->csum_offset); } else if (skb->ip_summed == CHECKSUM_UNNECESSARY) { vnet_hdr.flags = VIRTIO_NET_HDR_F_DATA_VALID; From 8c85151ddec66f78fbf997e35be005a322fbb9c9 Mon Sep 17 00:00:00 2001 From: WingMan Kwok Date: Wed, 23 Sep 2015 13:37:05 -0400 Subject: [PATCH 390/647] net: netcp: ethss: fix error in calling sgmii api with incorrect offset On K2HK, sgmii module registers of slave 0 and 1 are mem mapped to one contiguous block, while those of slave 2 and 3 are mapped to another contiguous block. However, on K2E and K2L, sgmii module registers of all slaves are mem mapped to one contiguous block. SGMII APIs expect slave 0 sgmii base when API is invoked for slave 0 and 1, and slave 2 sgmii base when invoked for other slaves. Before this patch, slave 0 sgmii base is always passed to sgmii API for K2E regardless which slave is the API invoked for. This patch fixes the problem. Signed-off-by: WingMan Kwok Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/netcp_ethss.c | 47 ++++++++++++--------------- 1 file changed, 20 insertions(+), 27 deletions(-) diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index 6f16d6aaf7b7..6bff8d82ceab 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -77,6 +77,7 @@ #define GBENU_ALE_OFFSET 0x1e000 #define GBENU_HOST_PORT_NUM 0 #define GBENU_NUM_ALE_ENTRIES 1024 +#define GBENU_SGMII_MODULE_SIZE 0x100 /* 10G Ethernet SS defines */ #define XGBE_MODULE_NAME "netcp-xgbe" @@ -149,8 +150,8 @@ #define XGBE_STATS2_MODULE 2 /* s: 0-based slave_port */ -#define SGMII_BASE(s) \ - (((s) < 2) ? gbe_dev->sgmii_port_regs : gbe_dev->sgmii_port34_regs) +#define SGMII_BASE(d, s) \ + (((s) < 2) ? (d)->sgmii_port_regs : (d)->sgmii_port34_regs) #define GBE_TX_QUEUE 648 #define GBE_TXHOOK_ORDER 0 @@ -1997,13 +1998,8 @@ static void netcp_ethss_update_link_state(struct gbe_priv *gbe_dev, return; if (!SLAVE_LINK_IS_XGMII(slave)) { - if (gbe_dev->ss_version == GBE_SS_VERSION_14) - sgmii_link_state = - netcp_sgmii_get_port_link(SGMII_BASE(sp), sp); - else - sgmii_link_state = - netcp_sgmii_get_port_link( - gbe_dev->sgmii_port_regs, sp); + sgmii_link_state = + netcp_sgmii_get_port_link(SGMII_BASE(gbe_dev, sp), sp); } phy_link_state = gbe_phy_link_status(slave); @@ -2100,17 +2096,11 @@ static void gbe_port_config(struct gbe_priv *gbe_dev, struct gbe_slave *slave, static void gbe_sgmii_rtreset(struct gbe_priv *priv, struct gbe_slave *slave, bool set) { - void __iomem *sgmii_port_regs; - if (SLAVE_LINK_IS_XGMII(slave)) return; - if ((priv->ss_version == GBE_SS_VERSION_14) && (slave->slave_num >= 2)) - sgmii_port_regs = priv->sgmii_port34_regs; - else - sgmii_port_regs = priv->sgmii_port_regs; - - netcp_sgmii_rtreset(sgmii_port_regs, slave->slave_num, set); + netcp_sgmii_rtreset(SGMII_BASE(priv, slave->slave_num), + slave->slave_num, set); } static void gbe_slave_stop(struct gbe_intf *intf) @@ -2136,17 +2126,12 @@ static void gbe_slave_stop(struct gbe_intf *intf) static void gbe_sgmii_config(struct gbe_priv *priv, struct gbe_slave *slave) { - void __iomem *sgmii_port_regs; + if (SLAVE_LINK_IS_XGMII(slave)) + return; - sgmii_port_regs = priv->sgmii_port_regs; - if ((priv->ss_version == GBE_SS_VERSION_14) && (slave->slave_num >= 2)) - sgmii_port_regs = priv->sgmii_port34_regs; - - if (!SLAVE_LINK_IS_XGMII(slave)) { - netcp_sgmii_reset(sgmii_port_regs, slave->slave_num); - netcp_sgmii_config(sgmii_port_regs, slave->slave_num, - slave->link_interface); - } + netcp_sgmii_reset(SGMII_BASE(priv, slave->slave_num), slave->slave_num); + netcp_sgmii_config(SGMII_BASE(priv, slave->slave_num), slave->slave_num, + slave->link_interface); } static int gbe_slave_open(struct gbe_intf *gbe_intf) @@ -2997,6 +2982,14 @@ static int set_gbenu_ethss_priv(struct gbe_priv *gbe_dev, gbe_dev->switch_regs = regs; gbe_dev->sgmii_port_regs = gbe_dev->ss_regs + GBENU_SGMII_MODULE_OFFSET; + + /* Although sgmii modules are mem mapped to one contiguous + * region on GBENU devices, setting sgmii_port34_regs allows + * consistent code when accessing sgmii api + */ + gbe_dev->sgmii_port34_regs = gbe_dev->sgmii_port_regs + + (2 * GBENU_SGMII_MODULE_SIZE); + gbe_dev->host_port_regs = gbe_dev->switch_regs + GBENU_HOST_PORT_OFFSET; for (i = 0; i < (gbe_dev->max_num_ports); i++) From 156e3c21f89655f099228577005a6c656b3ceb3d Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Wed, 23 Sep 2015 13:37:06 -0400 Subject: [PATCH 391/647] net: netcp: remove dead code from the driver netcp_core is the first driver that will get initialized and the modules (ethss, pa etc) will then get initialized. So the code at the end of netcp_probe() that iterate over the modules is a dead code as the module list will be always be empty. So remove this code. Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/netcp_core.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index 1a5aca55ea9f..c0bc4b9658fe 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -2040,7 +2040,6 @@ static int netcp_probe(struct platform_device *pdev) struct device_node *child, *interfaces; struct netcp_device *netcp_device; struct device *dev = &pdev->dev; - struct netcp_module *module; int ret; if (!node) { @@ -2087,14 +2086,6 @@ static int netcp_probe(struct platform_device *pdev) /* Add the device instance to the list */ list_add_tail(&netcp_device->device_list, &netcp_devices); - /* Probe & attach any modules already registered */ - mutex_lock(&netcp_modules_lock); - for_each_netcp_module(module) { - ret = netcp_module_probe(netcp_device, module); - if (ret < 0) - dev_err(dev, "module(%s) probe failed\n", module->name); - } - mutex_unlock(&netcp_modules_lock); return 0; probe_quit_interface: From 736532a0705ffc27c14f712fa2758a7f8b15e8b4 Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Wed, 23 Sep 2015 13:37:07 -0400 Subject: [PATCH 392/647] net: netcp: move netcp_register_interface() to after attach module The netcp interface is not fully initialized before attach the module to the interface. For example, the tx pipe/rx pipe is initialized in ethss module as part of attach(). So until this is complete, the interface can't be registered. So move registration of interface to net device outside the current loop that attaches the modules to the interface. Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/netcp_core.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index c0bc4b9658fe..cf693dec2e44 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -291,13 +291,6 @@ static int netcp_module_probe(struct netcp_device *netcp_device, interface_list) { struct netcp_intf_modpriv *intf_modpriv; - /* If interface not registered then register now */ - if (!netcp_intf->netdev_registered) - ret = netcp_register_interface(netcp_intf); - - if (ret) - return -ENODEV; - intf_modpriv = devm_kzalloc(dev, sizeof(*intf_modpriv), GFP_KERNEL); if (!intf_modpriv) @@ -323,6 +316,18 @@ static int netcp_module_probe(struct netcp_device *netcp_device, continue; } } + + /* Now register the interface with netdev */ + list_for_each_entry(netcp_intf, + &netcp_device->interface_head, + interface_list) { + /* If interface not registered then register now */ + if (!netcp_intf->netdev_registered) { + ret = netcp_register_interface(netcp_intf); + if (ret) + return -ENODEV; + } + } return 0; } From e558b1fbf5f43da83f91a31e595a6d65e663b100 Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Wed, 23 Sep 2015 13:37:08 -0400 Subject: [PATCH 393/647] net: netcp: add error check to netcp_allocate_rx_buf() Currently, if netcp_allocate_rx_buf() fails due no descriptors in the rx free descriptor queue, inside the netcp_rxpool_refill() function the iterative loop to fill buffers doesn't terminate right away. So modify the netcp_allocate_rx_buf() to return an error code and use it break the loop when there is error. Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/netcp_core.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index cf693dec2e44..97e2629c10b9 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -801,7 +801,7 @@ static void netcp_rxpool_free(struct netcp_intf *netcp) netcp->rx_pool = NULL; } -static void netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq) +static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq) { struct knav_dma_desc *hwdesc; unsigned int buf_len, dma_sz; @@ -815,7 +815,7 @@ static void netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq) hwdesc = knav_pool_desc_get(netcp->rx_pool); if (IS_ERR_OR_NULL(hwdesc)) { dev_dbg(netcp->ndev_dev, "out of rx pool desc\n"); - return; + return -ENOMEM; } if (likely(fdq == 0)) { @@ -867,25 +867,26 @@ static void netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq) knav_pool_desc_map(netcp->rx_pool, hwdesc, sizeof(*hwdesc), &dma, &dma_sz); knav_queue_push(netcp->rx_fdq[fdq], dma, sizeof(*hwdesc), 0); - return; + return 0; fail: knav_pool_desc_put(netcp->rx_pool, hwdesc); + return -ENOMEM; } /* Refill Rx FDQ with descriptors & attached buffers */ static void netcp_rxpool_refill(struct netcp_intf *netcp) { u32 fdq_deficit[KNAV_DMA_FDQ_PER_CHAN] = {0}; - int i; + int i, ret = 0; /* Calculate the FDQ deficit and refill */ for (i = 0; i < KNAV_DMA_FDQ_PER_CHAN && netcp->rx_fdq[i]; i++) { fdq_deficit[i] = netcp->rx_queue_depths[i] - knav_queue_get_count(netcp->rx_fdq[i]); - while (fdq_deficit[i]--) - netcp_allocate_rx_buf(netcp, i); + while (fdq_deficit[i]-- && !ret) + ret = netcp_allocate_rx_buf(netcp, i); } /* end for fdqs */ } From 915c5857874fc211874de1363e88f902e581e6eb Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Wed, 23 Sep 2015 13:37:09 -0400 Subject: [PATCH 394/647] net: netcp: check for interface handle in netcp_module_probe() Currently netcp_module_probe() doesn't check the return value of of_parse_phandle() that points to the interface data for the module and then pass the node ptr to the module which is incorrect. Check for return value and free the intf_modpriv if there is error. Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/netcp_core.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index 97e2629c10b9..d39dce3b37a7 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -299,6 +299,11 @@ static int netcp_module_probe(struct netcp_device *netcp_device, interface = of_parse_phandle(netcp_intf->node_interface, module->name, 0); + if (!interface) { + devm_kfree(dev, intf_modpriv); + continue; + } + intf_modpriv->netcp_priv = netcp_intf; intf_modpriv->netcp_module = module; list_add_tail(&intf_modpriv->intf_list, From 99f8ef5dc6546ac28cc7a03ff8301bc72fe5527e Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Wed, 23 Sep 2015 13:37:10 -0400 Subject: [PATCH 395/647] net: netcp: allocate buffers to desc before re-enable interrupt Currently netcp_rxpool_refill() that refill descriptors and attached buffers to fdq while interrupt is enabled as part of NAPI poll. Doing it while interrupt is disabled could be beneficial as hardware will not be starved when CPU is busy with processing interrupt. Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/netcp_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index d39dce3b37a7..8026daaf940b 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -904,12 +904,12 @@ static int netcp_rx_poll(struct napi_struct *napi, int budget) packets = netcp_process_rx_packets(netcp, budget); + netcp_rxpool_refill(netcp); if (packets < budget) { napi_complete(&netcp->rx_napi); knav_queue_enable_notify(netcp->rx_queue); } - netcp_rxpool_refill(netcp); return packets; } From 8ceaf361ffd131e835aef1e6cdb1d5ba70702617 Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Wed, 23 Sep 2015 13:37:11 -0400 Subject: [PATCH 396/647] net: netcp: fix deadlock reported by lockup detector A deadlock trace is seen in netcp driver with lockup detector enabled. The trace log is provided below for reference. This patch fixes the bug by removing the usage of netcp_modules_lock within ndo_ops functions. ndo_{open/close/ioctl)() is already called with rtnl_lock held. So there is no need to hold another mutex for serialization across processes on multiple cores. So remove use of netcp_modules_lock mutex from these ndo ops functions. ndo_set_rx_mode() shouldn't be using a mutex as it is called from atomic context. In the case of ndo_set_rx_mode(), there can be call to this API without rtnl_lock held from an atomic context. As the underlying modules are expected to add address to a hardware table, it is to be protected across concurrent updates and hence a spin lock is used to synchronize the access. Same with ndo_vlan_rx_add_vid() & ndo_vlan_rx_kill_vid(). Probably the netcp_modules_lock is used to protect the module not being removed as part of rmmod. Currently this is not fully implemented and assumes the interface is brought down before doing rmmod of modules. The support for rmmmod while interface is up is expected in a future patch set when additional modules such as pa, qos are added. For now all of the tests such as if up/down, reboot, iperf works fine with this patch applied. Deadlock trace seen with lockup detector enabled is shown below for reference. [ 16.863014] ====================================================== [ 16.869183] [ INFO: possible circular locking dependency detected ] [ 16.875441] 4.1.6-01265-gfb1e101 #1 Tainted: G W [ 16.881176] ------------------------------------------------------- [ 16.887432] ifconfig/1662 is trying to acquire lock: [ 16.892386] (netcp_modules_lock){+.+.+.}, at: [] netcp_ndo_open+0x168/0x518 [ 16.900321] [ 16.900321] but task is already holding lock: [ 16.906144] (rtnl_mutex){+.+.+.}, at: [] devinet_ioctl+0xf8/0x7e4 [ 16.913206] [ 16.913206] which lock already depends on the new lock. [ 16.913206] [ 16.921372] [ 16.921372] the existing dependency chain (in reverse order) is: [ 16.928844] -> #1 (rtnl_mutex){+.+.+.}: [ 16.932865] [] mutex_lock_nested+0x68/0x4a8 [ 16.938521] [] register_netdev+0xc/0x24 [ 16.943831] [] netcp_module_probe+0x214/0x2ec [ 16.949660] [] netcp_register_module+0xd4/0x140 [ 16.955663] [] keystone_gbe_init+0x10/0x28 [ 16.961233] [] do_one_initcall+0xb8/0x1f8 [ 16.966714] [] kernel_init_freeable+0x148/0x1e8 [ 16.972720] [] kernel_init+0xc/0xe8 [ 16.977682] [] ret_from_fork+0x14/0x3c [ 16.982905] -> #0 (netcp_modules_lock){+.+.+.}: [ 16.987619] [] lock_acquire+0x118/0x320 [ 16.992928] [] mutex_lock_nested+0x68/0x4a8 [ 16.998582] [] netcp_ndo_open+0x168/0x518 [ 17.004064] [] __dev_open+0xa8/0x10c [ 17.009112] [] __dev_change_flags+0x94/0x144 [ 17.014853] [] dev_change_flags+0x18/0x48 [ 17.020334] [] devinet_ioctl+0x6dc/0x7e4 [ 17.025729] [] sock_ioctl+0x1d0/0x2a8 [ 17.030865] [] do_vfs_ioctl+0x41c/0x688 [ 17.036173] [] SyS_ioctl+0x34/0x5c [ 17.041046] [] ret_fast_syscall+0x0/0x54 [ 17.046441] [ 17.046441] other info that might help us debug this: [ 17.046441] [ 17.054434] Possible unsafe locking scenario: [ 17.054434] [ 17.060343] CPU0 CPU1 [ 17.064862] ---- ---- [ 17.069381] lock(rtnl_mutex); [ 17.072522] lock(netcp_modules_lock); [ 17.078875] lock(rtnl_mutex); [ 17.084532] lock(netcp_modules_lock); [ 17.088366] [ 17.088366] *** DEADLOCK *** [ 17.088366] [ 17.094279] 1 lock held by ifconfig/1662: [ 17.098278] #0: (rtnl_mutex){+.+.+.}, at: [] devinet_ioctl+0xf8/0x7e4 [ 17.105774] [ 17.105774] stack backtrace: [ 17.110124] CPU: 1 PID: 1662 Comm: ifconfig Tainted: G W 4.1.6-01265-gfb1e101 #1 [ 17.118637] Hardware name: Keystone [ 17.122123] [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [ 17.129862] [] (show_stack) from [] (dump_stack+0x84/0xc4) [ 17.137079] [] (dump_stack) from [] (print_circular_bug+0x210/0x330) [ 17.145161] [] (print_circular_bug) from [] (validate_chain.isra.35+0xf98/0x13ac) [ 17.154372] [] (validate_chain.isra.35) from [] (__lock_acquire+0x52c/0xcc0) [ 17.163149] [] (__lock_acquire) from [] (lock_acquire+0x118/0x320) [ 17.171058] [] (lock_acquire) from [] (mutex_lock_nested+0x68/0x4a8) [ 17.179140] [] (mutex_lock_nested) from [] (netcp_ndo_open+0x168/0x518) [ 17.187484] [] (netcp_ndo_open) from [] (__dev_open+0xa8/0x10c) [ 17.195133] [] (__dev_open) from [] (__dev_change_flags+0x94/0x144) [ 17.203129] [] (__dev_change_flags) from [] (dev_change_flags+0x18/0x48) [ 17.211560] [] (dev_change_flags) from [] (devinet_ioctl+0x6dc/0x7e4) [ 17.219729] [] (devinet_ioctl) from [] (sock_ioctl+0x1d0/0x2a8) [ 17.227378] [] (sock_ioctl) from [] (do_vfs_ioctl+0x41c/0x688) [ 17.234939] [] (do_vfs_ioctl) from [] (SyS_ioctl+0x34/0x5c) [ 17.242242] [] (SyS_ioctl) from [] (ret_fast_syscall+0x0/0x54) [ 17.258855] netcp-1.0 2620110.netcp eth0: Link is Up - 1Gbps/Full - flow control off [ 17.271282] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:616 [ 17.279712] in_atomic(): 1, irqs_disabled(): 0, pid: 1662, name: ifconfig [ 17.286500] INFO: lockdep is turned off. [ 17.290413] Preemption disabled at:[< (null)>] (null) [ 17.295728] [ 17.297214] CPU: 1 PID: 1662 Comm: ifconfig Tainted: G W 4.1.6-01265-gfb1e101 #1 [ 17.305735] Hardware name: Keystone [ 17.309223] [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [ 17.316970] [] (show_stack) from [] (dump_stack+0x84/0xc4) [ 17.324194] [] (dump_stack) from [] (mutex_lock_nested+0x28/0x4a8) [ 17.332112] [] (mutex_lock_nested) from [] (netcp_set_rx_mode+0x160/0x210) [ 17.340724] [] (netcp_set_rx_mode) from [] (dev_set_rx_mode+0x1c/0x28) [ 17.348982] [] (dev_set_rx_mode) from [] (__dev_open+0xc4/0x10c) [ 17.356724] [] (__dev_open) from [] (__dev_change_flags+0x94/0x144) [ 17.364729] [] (__dev_change_flags) from [] (dev_change_flags+0x18/0x48) [ 17.373166] [] (dev_change_flags) from [] (devinet_ioctl+0x6dc/0x7e4) [ 17.381344] [] (devinet_ioctl) from [] (sock_ioctl+0x1d0/0x2a8) [ 17.388994] [] (sock_ioctl) from [] (do_vfs_ioctl+0x41c/0x688) [ 17.396563] [] (do_vfs_ioctl) from [] (SyS_ioctl+0x34/0x5c) [ 17.403873] [] (SyS_ioctl) from [] (ret_fast_syscall+0x0/0x54) [ 17.413772] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready udhcpc (v1.20.2) started Sending discover... [ 18.690666] netcp-1.0 2620110.netcp eth0: Link is Up - 1Gbps/Full - flow control off Sending discover... [ 22.250972] netcp-1.0 2620110.netcp eth0: Link is Up - 1Gbps/Full - flow control off [ 22.258721] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready [ 22.265458] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:616 [ 22.273896] in_atomic(): 1, irqs_disabled(): 0, pid: 342, name: kworker/1:1 [ 22.280854] INFO: lockdep is turned off. [ 22.284767] Preemption disabled at:[< (null)>] (null) [ 22.290074] [ 22.291568] CPU: 1 PID: 342 Comm: kworker/1:1 Tainted: G W 4.1.6-01265-gfb1e101 #1 [ 22.300255] Hardware name: Keystone [ 22.303750] Workqueue: ipv6_addrconf addrconf_dad_work [ 22.308895] [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [ 22.316643] [] (show_stack) from [] (dump_stack+0x84/0xc4) [ 22.323867] [] (dump_stack) from [] (mutex_lock_nested+0x28/0x4a8) [ 22.331786] [] (mutex_lock_nested) from [] (netcp_set_rx_mode+0x160/0x210) [ 22.340394] [] (netcp_set_rx_mode) from [] (__dev_mc_add+0x54/0x68) [ 22.348401] [] (__dev_mc_add) from [] (igmp6_group_added+0x168/0x1b4) [ 22.356580] [] (igmp6_group_added) from [] (ipv6_dev_mc_inc+0x4f0/0x5a8) [ 22.365019] [] (ipv6_dev_mc_inc) from [] (addrconf_dad_work+0x21c/0x33c) [ 22.373460] [] (addrconf_dad_work) from [] (process_one_work+0x214/0x8d0) [ 22.381986] [] (process_one_work) from [] (worker_thread+0x48/0x4bc) [ 22.390071] [] (worker_thread) from [] (kthread+0xf0/0x108) [ 22.397381] [] (kthread) from [] Trace related to incorrect usage of mutex inside ndo_set_rx_mode [ 24.086066] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:616 [ 24.094506] in_atomic(): 1, irqs_disabled(): 0, pid: 1682, name: ifconfig [ 24.101291] INFO: lockdep is turned off. [ 24.105203] Preemption disabled at:[< (null)>] (null) [ 24.110511] [ 24.112005] CPU: 2 PID: 1682 Comm: ifconfig Tainted: G W 4.1.6-01265-gfb1e101 #1 [ 24.120518] Hardware name: Keystone [ 24.124018] [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [ 24.131772] [] (show_stack) from [] (dump_stack+0x84/0xc4) [ 24.138989] [] (dump_stack) from [] (mutex_lock_nested+0x28/0x4a8) [ 24.146908] [] (mutex_lock_nested) from [] (netcp_set_rx_mode+0x160/0x210) [ 24.155523] [] (netcp_set_rx_mode) from [] (dev_set_rx_mode+0x1c/0x28) [ 24.163787] [] (dev_set_rx_mode) from [] (__dev_open+0xc4/0x10c) [ 24.171531] [] (__dev_open) from [] (__dev_change_flags+0x94/0x144) [ 24.179528] [] (__dev_change_flags) from [] (dev_change_flags+0x18/0x48) [ 24.187966] [] (dev_change_flags) from [] (devinet_ioctl+0x6dc/0x7e4) [ 24.196145] [] (devinet_ioctl) from [] (sock_ioctl+0x1d0/0x2a8) [ 24.203803] [] (sock_ioctl) from [] (do_vfs_ioctl+0x41c/0x688) [ 24.211373] [] (do_vfs_ioctl) from [] (SyS_ioctl+0x34/0x5c) [ 24.218676] [] (SyS_ioctl) from [] (ret_fast_syscall+0x0/0x54) [ 24.227156] IPv6: ADDRCONF(NETDEV_UP): eth1: link is not ready Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/netcp_core.c | 26 ++++++++++---------------- 1 file changed, 10 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index 8026daaf940b..9f9832f0dea9 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -367,7 +367,6 @@ int netcp_register_module(struct netcp_module *module) if (ret < 0) goto fail; } - mutex_unlock(&netcp_modules_lock); return 0; @@ -1395,7 +1394,6 @@ static void netcp_addr_sweep_del(struct netcp_intf *netcp) continue; dev_dbg(netcp->ndev_dev, "deleting address %pM, type %x\n", naddr->addr, naddr->type); - mutex_lock(&netcp_modules_lock); for_each_module(netcp, priv) { module = priv->netcp_module; if (!module->del_addr) @@ -1404,7 +1402,6 @@ static void netcp_addr_sweep_del(struct netcp_intf *netcp) naddr); WARN_ON(error); } - mutex_unlock(&netcp_modules_lock); netcp_addr_del(netcp, naddr); } } @@ -1421,7 +1418,7 @@ static void netcp_addr_sweep_add(struct netcp_intf *netcp) continue; dev_dbg(netcp->ndev_dev, "adding address %pM, type %x\n", naddr->addr, naddr->type); - mutex_lock(&netcp_modules_lock); + for_each_module(netcp, priv) { module = priv->netcp_module; if (!module->add_addr) @@ -1429,7 +1426,6 @@ static void netcp_addr_sweep_add(struct netcp_intf *netcp) error = module->add_addr(priv->module_priv, naddr); WARN_ON(error); } - mutex_unlock(&netcp_modules_lock); } } @@ -1443,6 +1439,7 @@ static void netcp_set_rx_mode(struct net_device *ndev) ndev->flags & IFF_ALLMULTI || netdev_mc_count(ndev) > NETCP_MAX_MCAST_ADDR); + spin_lock(&netcp->lock); /* first clear all marks */ netcp_addr_clear_mark(netcp); @@ -1461,6 +1458,7 @@ static void netcp_set_rx_mode(struct net_device *ndev) /* finally sweep and callout into modules */ netcp_addr_sweep_del(netcp); netcp_addr_sweep_add(netcp); + spin_unlock(&netcp->lock); } static void netcp_free_navigator_resources(struct netcp_intf *netcp) @@ -1625,7 +1623,6 @@ static int netcp_ndo_open(struct net_device *ndev) goto fail; } - mutex_lock(&netcp_modules_lock); for_each_module(netcp, intf_modpriv) { module = intf_modpriv->netcp_module; if (module->open) { @@ -1636,7 +1633,6 @@ static int netcp_ndo_open(struct net_device *ndev) } } } - mutex_unlock(&netcp_modules_lock); napi_enable(&netcp->rx_napi); napi_enable(&netcp->tx_napi); @@ -1653,7 +1649,6 @@ fail_open: if (module->close) module->close(intf_modpriv->module_priv, ndev); } - mutex_unlock(&netcp_modules_lock); fail: netcp_free_navigator_resources(netcp); @@ -1677,7 +1672,6 @@ static int netcp_ndo_stop(struct net_device *ndev) napi_disable(&netcp->rx_napi); napi_disable(&netcp->tx_napi); - mutex_lock(&netcp_modules_lock); for_each_module(netcp, intf_modpriv) { module = intf_modpriv->netcp_module; if (module->close) { @@ -1686,7 +1680,6 @@ static int netcp_ndo_stop(struct net_device *ndev) dev_err(netcp->ndev_dev, "Close failed\n"); } } - mutex_unlock(&netcp_modules_lock); /* Recycle Rx descriptors from completion queue */ netcp_empty_rx_queue(netcp); @@ -1714,7 +1707,6 @@ static int netcp_ndo_ioctl(struct net_device *ndev, if (!netif_running(ndev)) return -EINVAL; - mutex_lock(&netcp_modules_lock); for_each_module(netcp, intf_modpriv) { module = intf_modpriv->netcp_module; if (!module->ioctl) @@ -1730,7 +1722,6 @@ static int netcp_ndo_ioctl(struct net_device *ndev, } out: - mutex_unlock(&netcp_modules_lock); return (ret == 0) ? 0 : err; } @@ -1765,11 +1756,12 @@ static int netcp_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid) struct netcp_intf *netcp = netdev_priv(ndev); struct netcp_intf_modpriv *intf_modpriv; struct netcp_module *module; + unsigned long flags; int err = 0; dev_dbg(netcp->ndev_dev, "adding rx vlan id: %d\n", vid); - mutex_lock(&netcp_modules_lock); + spin_lock_irqsave(&netcp->lock, flags); for_each_module(netcp, intf_modpriv) { module = intf_modpriv->netcp_module; if ((module->add_vid) && (vid != 0)) { @@ -1781,7 +1773,8 @@ static int netcp_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid) } } } - mutex_unlock(&netcp_modules_lock); + spin_unlock_irqrestore(&netcp->lock, flags); + return err; } @@ -1790,11 +1783,12 @@ static int netcp_rx_kill_vid(struct net_device *ndev, __be16 proto, u16 vid) struct netcp_intf *netcp = netdev_priv(ndev); struct netcp_intf_modpriv *intf_modpriv; struct netcp_module *module; + unsigned long flags; int err = 0; dev_dbg(netcp->ndev_dev, "removing rx vlan id: %d\n", vid); - mutex_lock(&netcp_modules_lock); + spin_lock_irqsave(&netcp->lock, flags); for_each_module(netcp, intf_modpriv) { module = intf_modpriv->netcp_module; if (module->del_vid) { @@ -1806,7 +1800,7 @@ static int netcp_rx_kill_vid(struct net_device *ndev, __be16 proto, u16 vid) } } } - mutex_unlock(&netcp_modules_lock); + spin_unlock_irqrestore(&netcp->lock, flags); return err; } From aaa0062ecf4877a26dea66bee1039c6eaf906c94 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 23 Sep 2015 09:43:41 +0100 Subject: [PATCH 397/647] 8139cp: Do not re-enable RX interrupts in cp_tx_timeout() If an RX interrupt was already received but NAPI has not yet run when the RX timeout happens, we end up in cp_tx_timeout() with RX interrupts already disabled. Blindly re-enabling them will cause an IRQ storm. (This is made particularly horrid by the fact that cp_interrupt() always returns that it's handled the interrupt, even when it hasn't actually done anything. If it didn't do that, the core IRQ code would have detected the storm and handled it, I'd have had a clear smoking gun backtrace instead of just a spontaneously resetting router, and I'd have at *least* two days of my life back. Changing the return value of cp_interrupt() will be argued about under separate cover.) Unconditionally leave RX interrupts disabled after the reset, and schedule NAPI to check the receive ring and re-enable them. Signed-off-by: David Woodhouse Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/8139cp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index ba3dab721806..947932d4b79e 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -1262,9 +1262,10 @@ static void cp_tx_timeout(struct net_device *dev) rc = cp_init_rings(cp); cp_start_hw(cp); __cp_set_rx_mode(dev); - cp_enable_irq(cp); + cpw16_f(IntrMask, cp_norx_intr_mask); netif_wake_queue(dev); + napi_schedule_irqoff(&cp->napi); spin_unlock_irqrestore(&cp->lock, flags); } From 26b0bad6ac3a0167792dc4ffb276c29bc597d239 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 23 Sep 2015 09:44:06 +0100 Subject: [PATCH 398/647] 8139cp: Fix tx_queued debug message to print correct slot numbers After a certain amount of staring at the debug output of this driver, I realised it was lying to me. Signed-off-by: David Woodhouse Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/8139cp.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index 947932d4b79e..cbca0de1e857 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -786,7 +786,8 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, wmb(); cp->tx_skb[entry] = skb; - entry = NEXT_TX(entry); + netif_dbg(cp, tx_queued, cp->dev, "tx queued, slot %d, skblen %d\n", + entry, skb->len); } else { struct cp_desc *txd; u32 first_len, first_eor; @@ -805,7 +806,6 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, goto out_dma_error; cp->tx_skb[entry] = skb; - entry = NEXT_TX(entry); for (frag = 0; frag < skb_shinfo(skb)->nr_frags; frag++) { const skb_frag_t *this_frag = &skb_shinfo(skb)->frags[frag]; @@ -813,6 +813,8 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, u32 ctrl; dma_addr_t mapping; + entry = NEXT_TX(entry); + len = skb_frag_size(this_frag); mapping = dma_map_single(&cp->pdev->dev, skb_frag_address(this_frag), @@ -848,9 +850,7 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, txd->opts1 = cpu_to_le32(ctrl); wmb(); - cp->tx_skb[entry] = skb; - entry = NEXT_TX(entry); } txd = &cp->tx_ring[first_entry]; @@ -873,12 +873,13 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, txd->opts1 = cpu_to_le32(first_eor | first_len | FirstFrag | DescOwn); wmb(); + + netif_dbg(cp, tx_queued, cp->dev, "tx queued, slots %d-%d, skblen %d\n", + first_entry, entry, skb->len); } - cp->tx_head = entry; + cp->tx_head = NEXT_TX(entry); netdev_sent_queue(dev, skb->len); - netif_dbg(cp, tx_queued, cp->dev, "tx queued, slot %d, skblen %d\n", - entry, skb->len); if (TX_BUFFS_AVAIL(cp) <= (MAX_SKB_FRAGS + 1)) netif_stop_queue(dev); From a3b804043f490aeec57d8ca5baccdd35e6250857 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 23 Sep 2015 09:44:38 +0100 Subject: [PATCH 399/647] 8139cp: Fix TSO/scatter-gather descriptor setup When sending a TSO frame in multiple buffers, we were neglecting to set the first descriptor up in TSO mode. Signed-off-by: David Woodhouse Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/8139cp.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index cbca0de1e857..75a8cee6b1d6 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -790,7 +790,7 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, entry, skb->len); } else { struct cp_desc *txd; - u32 first_len, first_eor; + u32 first_len, first_eor, ctrl; dma_addr_t first_mapping; int frag, first_entry = entry; const struct iphdr *ip = ip_hdr(skb); @@ -810,7 +810,6 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, for (frag = 0; frag < skb_shinfo(skb)->nr_frags; frag++) { const skb_frag_t *this_frag = &skb_shinfo(skb)->frags[frag]; u32 len; - u32 ctrl; dma_addr_t mapping; entry = NEXT_TX(entry); @@ -858,20 +857,19 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, txd->addr = cpu_to_le64(first_mapping); wmb(); - if (skb->ip_summed == CHECKSUM_PARTIAL) { + ctrl = first_eor | first_len | FirstFrag | DescOwn; + if (mss) + ctrl |= LargeSend | ((mss & MSSMask) << MSSShift); + else if (skb->ip_summed == CHECKSUM_PARTIAL) { if (ip->protocol == IPPROTO_TCP) - txd->opts1 = cpu_to_le32(first_eor | first_len | - FirstFrag | DescOwn | - IPCS | TCPCS); + ctrl |= IPCS | TCPCS; else if (ip->protocol == IPPROTO_UDP) - txd->opts1 = cpu_to_le32(first_eor | first_len | - FirstFrag | DescOwn | - IPCS | UDPCS); + ctrl |= IPCS | UDPCS; else BUG(); - } else - txd->opts1 = cpu_to_le32(first_eor | first_len | - FirstFrag | DescOwn); + } + + txd->opts1 = cpu_to_le32(ctrl); wmb(); netif_dbg(cp, tx_queued, cp->dev, "tx queued, slots %d-%d, skblen %d\n", From 0a5aeee0b79fa99d8e04c98dd4e87d4f52aa497b Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 23 Sep 2015 09:44:57 +0100 Subject: [PATCH 400/647] 8139cp: Reduce duplicate csum/tso code in cp_start_xmit() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We calculate the value of the opts1 descriptor field in three different places. With two different behaviours when given an invalid packet to be checksummed — none of them correct. Sort that out. Signed-off-by: David Woodhouse Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/8139cp.c | 61 +++++++++------------------ 1 file changed, 20 insertions(+), 41 deletions(-) diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index 75a8cee6b1d6..b3bd8b103593 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -733,7 +733,7 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, { struct cp_private *cp = netdev_priv(dev); unsigned entry; - u32 eor, flags; + u32 eor, opts1; unsigned long intr_flags; __le32 opts2; int mss = 0; @@ -753,6 +753,21 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, mss = skb_shinfo(skb)->gso_size; opts2 = cpu_to_le32(cp_tx_vlan_tag(skb)); + opts1 = DescOwn; + if (mss) + opts1 |= LargeSend | ((mss & MSSMask) << MSSShift); + else if (skb->ip_summed == CHECKSUM_PARTIAL) { + const struct iphdr *ip = ip_hdr(skb); + if (ip->protocol == IPPROTO_TCP) + opts1 |= IPCS | TCPCS; + else if (ip->protocol == IPPROTO_UDP) + opts1 |= IPCS | UDPCS; + else { + WARN_ONCE(1, + "Net bug: asked to checksum invalid Legacy IP packet\n"); + goto out_dma_error; + } + } if (skb_shinfo(skb)->nr_frags == 0) { struct cp_desc *txd = &cp->tx_ring[entry]; @@ -768,21 +783,9 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, txd->addr = cpu_to_le64(mapping); wmb(); - flags = eor | len | DescOwn | FirstFrag | LastFrag; + opts1 |= eor | len | FirstFrag | LastFrag; - if (mss) - flags |= LargeSend | ((mss & MSSMask) << MSSShift); - else if (skb->ip_summed == CHECKSUM_PARTIAL) { - const struct iphdr *ip = ip_hdr(skb); - if (ip->protocol == IPPROTO_TCP) - flags |= IPCS | TCPCS; - else if (ip->protocol == IPPROTO_UDP) - flags |= IPCS | UDPCS; - else - WARN_ON(1); /* we need a WARN() */ - } - - txd->opts1 = cpu_to_le32(flags); + txd->opts1 = cpu_to_le32(opts1); wmb(); cp->tx_skb[entry] = skb; @@ -793,7 +796,6 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, u32 first_len, first_eor, ctrl; dma_addr_t first_mapping; int frag, first_entry = entry; - const struct iphdr *ip = ip_hdr(skb); /* We must give this initial chunk to the device last. * Otherwise we could race with the device. @@ -825,19 +827,7 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, eor = (entry == (CP_TX_RING_SIZE - 1)) ? RingEnd : 0; - ctrl = eor | len | DescOwn; - - if (mss) - ctrl |= LargeSend | - ((mss & MSSMask) << MSSShift); - else if (skb->ip_summed == CHECKSUM_PARTIAL) { - if (ip->protocol == IPPROTO_TCP) - ctrl |= IPCS | TCPCS; - else if (ip->protocol == IPPROTO_UDP) - ctrl |= IPCS | UDPCS; - else - BUG(); - } + ctrl = opts1 | eor | len; if (frag == skb_shinfo(skb)->nr_frags - 1) ctrl |= LastFrag; @@ -857,18 +847,7 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, txd->addr = cpu_to_le64(first_mapping); wmb(); - ctrl = first_eor | first_len | FirstFrag | DescOwn; - if (mss) - ctrl |= LargeSend | ((mss & MSSMask) << MSSShift); - else if (skb->ip_summed == CHECKSUM_PARTIAL) { - if (ip->protocol == IPPROTO_TCP) - ctrl |= IPCS | TCPCS; - else if (ip->protocol == IPPROTO_UDP) - ctrl |= IPCS | UDPCS; - else - BUG(); - } - + ctrl = opts1 | first_eor | first_len | FirstFrag; txd->opts1 = cpu_to_le32(ctrl); wmb(); From 7f4c685633e2df9ba10d49a31dda13715745db37 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 23 Sep 2015 09:45:16 +0100 Subject: [PATCH 401/647] 8139cp: Fix DMA unmapping of transmitted buffers The low 16 bits of the 'opts1' field in the TX descriptor are supposed to still contain the buffer length when the descriptor is handed back to us. In practice, at least on my hardware, they don't. So stash the original value of the opts1 field and get the length to unmap from there. There are other ways we could have worked out the length, but I actually want a stash of the opts1 field anyway so that I can dump it alongside the contents of the descriptor ring when we suffer a TX timeout. Signed-off-by: David Woodhouse Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/8139cp.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index b3bd8b103593..ec6bd864b263 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -341,6 +341,7 @@ struct cp_private { unsigned tx_tail; struct cp_desc *tx_ring; struct sk_buff *tx_skb[CP_TX_RING_SIZE]; + u32 tx_opts[CP_TX_RING_SIZE]; unsigned rx_buf_sz; unsigned wol_enabled : 1; /* Is Wake-on-LAN enabled? */ @@ -665,7 +666,7 @@ static void cp_tx (struct cp_private *cp) BUG_ON(!skb); dma_unmap_single(&cp->pdev->dev, le64_to_cpu(txd->addr), - le32_to_cpu(txd->opts1) & 0xffff, + cp->tx_opts[tx_tail] & 0xffff, PCI_DMA_TODEVICE); if (status & LastFrag) { @@ -789,6 +790,7 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, wmb(); cp->tx_skb[entry] = skb; + cp->tx_opts[entry] = opts1; netif_dbg(cp, tx_queued, cp->dev, "tx queued, slot %d, skblen %d\n", entry, skb->len); } else { @@ -839,6 +841,8 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, txd->opts1 = cpu_to_le32(ctrl); wmb(); + + cp->tx_opts[entry] = ctrl; cp->tx_skb[entry] = skb; } @@ -851,6 +855,7 @@ static netdev_tx_t cp_start_xmit (struct sk_buff *skb, txd->opts1 = cpu_to_le32(ctrl); wmb(); + cp->tx_opts[first_entry] = ctrl; netif_dbg(cp, tx_queued, cp->dev, "tx queued, slots %d-%d, skblen %d\n", first_entry, entry, skb->len); } @@ -1093,6 +1098,7 @@ static int cp_init_rings (struct cp_private *cp) { memset(cp->tx_ring, 0, sizeof(struct cp_desc) * CP_TX_RING_SIZE); cp->tx_ring[CP_TX_RING_SIZE - 1].opts1 = cpu_to_le32(RingEnd); + memset(cp->tx_opts, 0, sizeof(cp->tx_opts)); cp_init_rings_index(cp); @@ -1150,6 +1156,7 @@ static void cp_clean_rings (struct cp_private *cp) memset(cp->rx_ring, 0, sizeof(struct cp_desc) * CP_RX_RING_SIZE); memset(cp->tx_ring, 0, sizeof(struct cp_desc) * CP_TX_RING_SIZE); + memset(cp->tx_opts, 0, sizeof(cp->tx_opts)); memset(cp->rx_skb, 0, sizeof(struct sk_buff *) * CP_RX_RING_SIZE); memset(cp->tx_skb, 0, sizeof(struct sk_buff *) * CP_TX_RING_SIZE); From 41b976414c88016e2c9d9b2f6667ee67a998d388 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 23 Sep 2015 09:45:31 +0100 Subject: [PATCH 402/647] 8139cp: Dump contents of descriptor ring on TX timeout We are seeing unexplained TX timeouts under heavy load. Let's try to get a better idea of what's going on. Signed-off-by: David Woodhouse Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/8139cp.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index ec6bd864b263..686334f4588d 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -157,6 +157,7 @@ enum { NWayAdvert = 0x66, /* MII ADVERTISE */ NWayLPAR = 0x68, /* MII LPA */ NWayExpansion = 0x6A, /* MII Expansion */ + TxDmaOkLowDesc = 0x82, /* Low 16 bit address of a Tx descriptor. */ Config5 = 0xD8, /* Config5 */ TxPoll = 0xD9, /* Tell chip to check Tx descriptors for work */ RxMaxSize = 0xDA, /* Max size of an Rx packet (8169 only) */ @@ -1234,7 +1235,7 @@ static void cp_tx_timeout(struct net_device *dev) { struct cp_private *cp = netdev_priv(dev); unsigned long flags; - int rc; + int rc, i; netdev_warn(dev, "Transmit timeout, status %2x %4x %4x %4x\n", cpr8(Cmd), cpr16(CpCmd), @@ -1242,6 +1243,17 @@ static void cp_tx_timeout(struct net_device *dev) spin_lock_irqsave(&cp->lock, flags); + netif_dbg(cp, tx_err, cp->dev, "TX ring head %d tail %d desc %x\n", + cp->tx_head, cp->tx_tail, cpr16(TxDmaOkLowDesc)); + for (i = 0; i < CP_TX_RING_SIZE; i++) { + netif_dbg(cp, tx_err, cp->dev, + "TX slot %d @%p: %08x (%08x) %08x %llx %p\n", + i, &cp->tx_ring[i], le32_to_cpu(cp->tx_ring[i].opts1), + cp->tx_opts[i], le32_to_cpu(cp->tx_ring[i].opts2), + le64_to_cpu(cp->tx_ring[i].addr), + cp->tx_skb[i]); + } + cp_stop_hw(cp); cp_clean_rings(cp); rc = cp_init_rings(cp); From 69e5d3f893e19613486f300fd6e631810338aa4b Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Mon, 14 Sep 2015 10:28:34 +1000 Subject: [PATCH 403/647] drm/qxl: only report first monitor as connected if we have no state If the server isn't new enough to give us state, report the first monitor as always connected, otherwise believe the server side. Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/qxl/qxl_display.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c index 7c6225c84ba6..dd845f82cc24 100644 --- a/drivers/gpu/drm/qxl/qxl_display.c +++ b/drivers/gpu/drm/qxl/qxl_display.c @@ -886,13 +886,15 @@ static enum drm_connector_status qxl_conn_detect( drm_connector_to_qxl_output(connector); struct drm_device *ddev = connector->dev; struct qxl_device *qdev = ddev->dev_private; - int connected; + bool connected = false; /* The first monitor is always connected */ - connected = (output->index == 0) || - (qdev->client_monitors_config && - qdev->client_monitors_config->count > output->index && - qxl_head_enabled(&qdev->client_monitors_config->heads[output->index])); + if (!qdev->client_monitors_config) { + if (output->index == 0) + connected = true; + } else + connected = qdev->client_monitors_config->count > output->index && + qxl_head_enabled(&qdev->client_monitors_config->heads[output->index]); DRM_DEBUG("#%d connected: %d\n", output->index, connected); if (!connected) From aec9e12953e777f62acdab069656ebd9bcb6c9ba Mon Sep 17 00:00:00 2001 From: Archit Taneja Date: Thu, 17 Sep 2015 16:30:54 +0530 Subject: [PATCH 404/647] drm/mgag200: Fix error handling paths in fbdev driver Set up error handling in mgag200_fbdev_init and mgag200fb_create such that they release the things they allocate, rather than relying on someone calling mga_fbdev_destroy. Based on a patch by Sudip Mukherjee Link: http://lkml.kernel.org/r/55F6E68D.8070800@codeaurora.org Reported-by: Ingo Molnar Cc: Daniel Vetter Cc: Dave Airlie Cc: David Airlie Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Sudip Mukherjee Cc: Thomas Gleixner Cc: dri-devel Signed-off-by: Archit Taneja Signed-off-by: Dave Airlie --- drivers/gpu/drm/mgag200/mgag200_fb.c | 31 +++++++++++++++++++++------- 1 file changed, 23 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/mgag200/mgag200_fb.c b/drivers/gpu/drm/mgag200/mgag200_fb.c index 87de15ea1f93..b35b5b2db4ec 100644 --- a/drivers/gpu/drm/mgag200/mgag200_fb.c +++ b/drivers/gpu/drm/mgag200/mgag200_fb.c @@ -186,17 +186,19 @@ static int mgag200fb_create(struct drm_fb_helper *helper, sysram = vmalloc(size); if (!sysram) - return -ENOMEM; + goto err_sysram; info = drm_fb_helper_alloc_fbi(helper); - if (IS_ERR(info)) - return PTR_ERR(info); + if (IS_ERR(info)) { + ret = PTR_ERR(info); + goto err_alloc_fbi; + } info->par = mfbdev; ret = mgag200_framebuffer_init(dev, &mfbdev->mfb, &mode_cmd, gobj); if (ret) - return ret; + goto err_framebuffer_init; mfbdev->sysram = sysram; mfbdev->size = size; @@ -225,7 +227,17 @@ static int mgag200fb_create(struct drm_fb_helper *helper, DRM_DEBUG_KMS("allocated %dx%d\n", fb->width, fb->height); + return 0; + +err_framebuffer_init: + drm_fb_helper_release_fbi(helper); +err_alloc_fbi: + vfree(sysram); +err_sysram: + drm_gem_object_unreference_unlocked(gobj); + + return ret; } static int mga_fbdev_destroy(struct drm_device *dev, @@ -276,23 +288,26 @@ int mgag200_fbdev_init(struct mga_device *mdev) ret = drm_fb_helper_init(mdev->dev, &mfbdev->helper, mdev->num_crtc, MGAG200FB_CONN_LIMIT); if (ret) - return ret; + goto err_fb_helper; ret = drm_fb_helper_single_add_all_connectors(&mfbdev->helper); if (ret) - goto fini; + goto err_fb_setup; /* disable all the possible outputs/crtcs before entering KMS mode */ drm_helper_disable_unused_functions(mdev->dev); ret = drm_fb_helper_initial_config(&mfbdev->helper, bpp_sel); if (ret) - goto fini; + goto err_fb_setup; return 0; -fini: +err_fb_setup: drm_fb_helper_fini(&mfbdev->helper); +err_fb_helper: + mdev->mfbdev = NULL; + return ret; } From 728f86607d47f7d2d24d61fd30852faa66ca5aa9 Mon Sep 17 00:00:00 2001 From: Archit Taneja Date: Thu, 17 Sep 2015 16:30:55 +0530 Subject: [PATCH 405/647] drm/mgag200: Fix driver_load error handling mgag200_driver_load's error path just calls the drm driver's driver_unload op. It isn't safe to call this because it doesn't handle things well if driver_load fails somewhere mid way. Replace the call to mgag200_driver_unload with a more finegrained error handling path. Link: http://lkml.kernel.org/r/55F6E68D.8070800@codeaurora.org Reported-by: Ingo Molnar Cc: Daniel Vetter Cc: Dave Airlie Cc: David Airlie Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Sudip Mukherjee Cc: Thomas Gleixner Cc: dri-devel Signed-off-by: Archit Taneja Signed-off-by: Dave Airlie --- drivers/gpu/drm/mgag200/mgag200_main.c | 34 ++++++++++++++------------ 1 file changed, 19 insertions(+), 15 deletions(-) diff --git a/drivers/gpu/drm/mgag200/mgag200_main.c b/drivers/gpu/drm/mgag200/mgag200_main.c index de06388069e7..b1a0f5656175 100644 --- a/drivers/gpu/drm/mgag200/mgag200_main.c +++ b/drivers/gpu/drm/mgag200/mgag200_main.c @@ -220,7 +220,7 @@ int mgag200_driver_load(struct drm_device *dev, unsigned long flags) } r = mgag200_mm_init(mdev); if (r) - goto out; + goto err_mm; drm_mode_config_init(dev); dev->mode_config.funcs = (void *)&mga_mode_funcs; @@ -233,7 +233,7 @@ int mgag200_driver_load(struct drm_device *dev, unsigned long flags) r = mgag200_modeset_init(mdev); if (r) { dev_err(&dev->pdev->dev, "Fatal error during modeset init: %d\n", r); - goto out; + goto err_modeset; } /* Make small buffers to store a hardware cursor (double buffered icon updates) */ @@ -241,20 +241,24 @@ int mgag200_driver_load(struct drm_device *dev, unsigned long flags) &mdev->cursor.pixels_1); mgag200_bo_create(dev, roundup(48*64, PAGE_SIZE), 0, 0, &mdev->cursor.pixels_2); - if (!mdev->cursor.pixels_2 || !mdev->cursor.pixels_1) - goto cursor_nospace; - mdev->cursor.pixels_current = mdev->cursor.pixels_1; - mdev->cursor.pixels_prev = mdev->cursor.pixels_2; - goto cursor_done; - cursor_nospace: - mdev->cursor.pixels_1 = NULL; - mdev->cursor.pixels_2 = NULL; - dev_warn(&dev->pdev->dev, "Could not allocate space for cursors. Not doing hardware cursors.\n"); - cursor_done: + if (!mdev->cursor.pixels_2 || !mdev->cursor.pixels_1) { + mdev->cursor.pixels_1 = NULL; + mdev->cursor.pixels_2 = NULL; + dev_warn(&dev->pdev->dev, + "Could not allocate space for cursors. Not doing hardware cursors.\n"); + } else { + mdev->cursor.pixels_current = mdev->cursor.pixels_1; + mdev->cursor.pixels_prev = mdev->cursor.pixels_2; + } + + return 0; + +err_modeset: + drm_mode_config_cleanup(dev); + mgag200_mm_fini(mdev); +err_mm: + dev->dev_private = NULL; -out: - if (r) - mgag200_driver_unload(dev); return r; } From 14d11b8dca17f6dedf4b62d7391b9b294e3414ff Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Mon, 21 Sep 2015 15:33:47 +0200 Subject: [PATCH 406/647] drm/layerscape: fix handling fsl_dcu_drm_plane_index result The function can return negative value. The problem has been detected using proposed semantic patch scripts/coccinelle/tests/unsigned_lesser_than_zero.cocci [1]. [1]: http://permalink.gmane.org/gmane.linux.kernel/2038576 Signed-off-by: Andrzej Hajda Signed-off-by: Dave Airlie --- drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c index 82be6b86a168..d1e300dcd544 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c @@ -58,7 +58,8 @@ static void fsl_dcu_drm_plane_atomic_disable(struct drm_plane *plane, struct drm_plane_state *old_state) { struct fsl_dcu_drm_device *fsl_dev = plane->dev->dev_private; - unsigned int index, value, ret; + unsigned int value; + int index, ret; index = fsl_dcu_drm_plane_index(plane); if (index < 0) From 7bbe33ff1896d225588b37c574c0d80dbc63c657 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Tue, 22 Sep 2015 13:09:32 -0400 Subject: [PATCH 407/647] geneve: use network byte order for destination port config parameter This is primarily for consistancy with vxlan and other tunnels which use network byte order for similar parameters. Signed-off-by: John W. Linville Signed-off-by: David S. Miller --- drivers/net/geneve.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 4ce8b7b90c7c..8f5c02eed47d 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -815,7 +815,7 @@ static struct geneve_dev *geneve_find_dev(struct geneve_net *gn, static int geneve_configure(struct net *net, struct net_device *dev, __be32 rem_addr, __u32 vni, __u8 ttl, __u8 tos, - __u16 dst_port, bool metadata) + __be16 dst_port, bool metadata) { struct geneve_net *gn = net_generic(net, geneve_net_id); struct geneve_dev *t, *geneve = netdev_priv(dev); @@ -840,10 +840,10 @@ static int geneve_configure(struct net *net, struct net_device *dev, geneve->ttl = ttl; geneve->tos = tos; - geneve->dst_port = htons(dst_port); + geneve->dst_port = dst_port; geneve->collect_md = metadata; - t = geneve_find_dev(gn, htons(dst_port), rem_addr, geneve->vni, + t = geneve_find_dev(gn, dst_port, rem_addr, geneve->vni, &tun_on_same_port, &tun_collect_md); if (t) return -EBUSY; @@ -867,7 +867,7 @@ static int geneve_configure(struct net *net, struct net_device *dev, static int geneve_newlink(struct net *net, struct net_device *dev, struct nlattr *tb[], struct nlattr *data[]) { - __u16 dst_port = GENEVE_UDP_PORT; + __be16 dst_port = htons(GENEVE_UDP_PORT); __u8 ttl = 0, tos = 0; bool metadata = false; __be32 rem_addr; @@ -886,7 +886,7 @@ static int geneve_newlink(struct net *net, struct net_device *dev, tos = nla_get_u8(data[IFLA_GENEVE_TOS]); if (data[IFLA_GENEVE_PORT]) - dst_port = nla_get_u16(data[IFLA_GENEVE_PORT]); + dst_port = nla_get_be16(data[IFLA_GENEVE_PORT]); if (data[IFLA_GENEVE_COLLECT_METADATA]) metadata = true; @@ -909,7 +909,7 @@ static size_t geneve_get_size(const struct net_device *dev) nla_total_size(sizeof(struct in_addr)) + /* IFLA_GENEVE_REMOTE */ nla_total_size(sizeof(__u8)) + /* IFLA_GENEVE_TTL */ nla_total_size(sizeof(__u8)) + /* IFLA_GENEVE_TOS */ - nla_total_size(sizeof(__u16)) + /* IFLA_GENEVE_PORT */ + nla_total_size(sizeof(__be16)) + /* IFLA_GENEVE_PORT */ nla_total_size(0) + /* IFLA_GENEVE_COLLECT_METADATA */ 0; } @@ -931,7 +931,7 @@ static int geneve_fill_info(struct sk_buff *skb, const struct net_device *dev) nla_put_u8(skb, IFLA_GENEVE_TOS, geneve->tos)) goto nla_put_failure; - if (nla_put_u16(skb, IFLA_GENEVE_PORT, ntohs(geneve->dst_port))) + if (nla_put_be16(skb, IFLA_GENEVE_PORT, geneve->dst_port)) goto nla_put_failure; if (geneve->collect_md) { @@ -971,7 +971,7 @@ struct net_device *geneve_dev_create_fb(struct net *net, const char *name, if (IS_ERR(dev)) return dev; - err = geneve_configure(net, dev, 0, 0, 0, 0, dst_port, true); + err = geneve_configure(net, dev, 0, 0, 0, 0, htons(dst_port), true); if (err) { free_netdev(dev); return ERR_PTR(err); From ceb1b0b9b4d1089e9f2731a314689ae17784c861 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 24 Sep 2015 00:52:37 -0500 Subject: [PATCH 408/647] [SMB3] Fix sec=krb5 on smb3 mounts Kerberos, which is very important for security, was only enabled for CIFS not SMB2/SMB3 mounts (e.g. vers=3.0) Patch based on the information detailed in http://thread.gmane.org/gmane.linux.kernel.cifs/10081/focus=10307 to enable Kerberized SMB2/SMB3 a) SMB2_negotiate: enable/use decode_negTokenInit in SMB2_negotiate b) SMB2_sess_setup: handle Kerberos sectype and replicate Kerberos SMB1 processing done in sess_auth_kerberos Signed-off-by: Noel Power Signed-off-by: Jim McDonough CC: Stable Signed-off-by: Steve French --- fs/cifs/smb2pdu.c | 76 +++++++++++++++++++++++++++++++++++++---------- 1 file changed, 61 insertions(+), 15 deletions(-) diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 070fb2ad85ce..e543431db916 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -46,6 +46,7 @@ #include "smb2status.h" #include "smb2glob.h" #include "cifspdu.h" +#include "cifs_spnego.h" /* * The following table defines the expected "StructureSize" of SMB2 requests @@ -486,19 +487,15 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses) cifs_dbg(FYI, "missing security blob on negprot\n"); rc = cifs_enable_signing(server, ses->sign); -#ifdef CONFIG_SMB2_ASN1 /* BB REMOVEME when updated asn1.c ready */ if (rc) goto neg_exit; - if (blob_length) + if (blob_length) { rc = decode_negTokenInit(security_blob, blob_length, server); - if (rc == 1) - rc = 0; - else if (rc == 0) { - rc = -EIO; - goto neg_exit; + if (rc == 1) + rc = 0; + else if (rc == 0) + rc = -EIO; } -#endif - neg_exit: free_rsp_buf(resp_buftype, rsp); return rc; @@ -592,7 +589,8 @@ SMB2_sess_setup(const unsigned int xid, struct cifs_ses *ses, __le32 phase = NtLmNegotiate; /* NTLMSSP, if needed, is multistage */ struct TCP_Server_Info *server = ses->server; u16 blob_length = 0; - char *security_blob; + struct key *spnego_key = NULL; + char *security_blob = NULL; char *ntlmssp_blob = NULL; bool use_spnego = false; /* else use raw ntlmssp */ @@ -620,7 +618,8 @@ SMB2_sess_setup(const unsigned int xid, struct cifs_ses *ses, ses->ntlmssp->sesskey_per_smbsess = true; /* FIXME: allow for other auth types besides NTLMSSP (e.g. krb5) */ - ses->sectype = RawNTLMSSP; + if (ses->sectype != Kerberos && ses->sectype != RawNTLMSSP) + ses->sectype = RawNTLMSSP; ssetup_ntlmssp_authenticate: if (phase == NtLmChallenge) @@ -649,7 +648,48 @@ ssetup_ntlmssp_authenticate: iov[0].iov_base = (char *)req; /* 4 for rfc1002 length field and 1 for pad */ iov[0].iov_len = get_rfc1002_length(req) + 4 - 1; - if (phase == NtLmNegotiate) { + + if (ses->sectype == Kerberos) { +#ifdef CONFIG_CIFS_UPCALL + struct cifs_spnego_msg *msg; + + spnego_key = cifs_get_spnego_key(ses); + if (IS_ERR(spnego_key)) { + rc = PTR_ERR(spnego_key); + spnego_key = NULL; + goto ssetup_exit; + } + + msg = spnego_key->payload.data; + /* + * check version field to make sure that cifs.upcall is + * sending us a response in an expected form + */ + if (msg->version != CIFS_SPNEGO_UPCALL_VERSION) { + cifs_dbg(VFS, + "bad cifs.upcall version. Expected %d got %d", + CIFS_SPNEGO_UPCALL_VERSION, msg->version); + rc = -EKEYREJECTED; + goto ssetup_exit; + } + ses->auth_key.response = kmemdup(msg->data, msg->sesskey_len, + GFP_KERNEL); + if (!ses->auth_key.response) { + cifs_dbg(VFS, + "Kerberos can't allocate (%u bytes) memory", + msg->sesskey_len); + rc = -ENOMEM; + goto ssetup_exit; + } + ses->auth_key.len = msg->sesskey_len; + blob_length = msg->secblob_len; + iov[1].iov_base = msg->data + msg->sesskey_len; + iov[1].iov_len = blob_length; +#else + rc = -EOPNOTSUPP; + goto ssetup_exit; +#endif /* CONFIG_CIFS_UPCALL */ + } else if (phase == NtLmNegotiate) { /* if not krb5 must be ntlmssp */ ntlmssp_blob = kmalloc(sizeof(struct _NEGOTIATE_MESSAGE), GFP_KERNEL); if (ntlmssp_blob == NULL) { @@ -672,6 +712,8 @@ ssetup_ntlmssp_authenticate: /* with raw NTLMSSP we don't encapsulate in SPNEGO */ security_blob = ntlmssp_blob; } + iov[1].iov_base = security_blob; + iov[1].iov_len = blob_length; } else if (phase == NtLmAuthenticate) { req->hdr.SessionId = ses->Suid; ntlmssp_blob = kzalloc(sizeof(struct _NEGOTIATE_MESSAGE) + 500, @@ -699,6 +741,8 @@ ssetup_ntlmssp_authenticate: } else { security_blob = ntlmssp_blob; } + iov[1].iov_base = security_blob; + iov[1].iov_len = blob_length; } else { cifs_dbg(VFS, "illegal ntlmssp phase\n"); rc = -EIO; @@ -710,8 +754,6 @@ ssetup_ntlmssp_authenticate: cpu_to_le16(sizeof(struct smb2_sess_setup_req) - 1 /* pad */ - 4 /* rfc1001 len */); req->SecurityBufferLength = cpu_to_le16(blob_length); - iov[1].iov_base = security_blob; - iov[1].iov_len = blob_length; inc_rfc1001_len(req, blob_length - 1 /* pad */); @@ -722,6 +764,7 @@ ssetup_ntlmssp_authenticate: kfree(security_blob); rsp = (struct smb2_sess_setup_rsp *)iov[0].iov_base; + ses->Suid = rsp->hdr.SessionId; if (resp_buftype != CIFS_NO_BUFFER && rsp->hdr.Status == STATUS_MORE_PROCESSING_REQUIRED) { if (phase != NtLmNegotiate) { @@ -739,7 +782,6 @@ ssetup_ntlmssp_authenticate: /* NTLMSSP Negotiate sent now processing challenge (response) */ phase = NtLmChallenge; /* process ntlmssp challenge */ rc = 0; /* MORE_PROCESSING is not an error here but expected */ - ses->Suid = rsp->hdr.SessionId; rc = decode_ntlmssp_challenge(rsp->Buffer, le16_to_cpu(rsp->SecurityBufferLength), ses); } @@ -796,6 +838,10 @@ keygen_exit: kfree(ses->auth_key.response); ses->auth_key.response = NULL; } + if (spnego_key) { + key_invalidate(spnego_key); + key_put(spnego_key); + } kfree(ses->ntlmssp); return rc; From 88627148400e37b4bff197285d348fc05c9de013 Mon Sep 17 00:00:00 2001 From: Steve French Date: Tue, 22 Sep 2015 03:16:27 -0500 Subject: [PATCH 409/647] fix encryption error checks on mount Signed-off-by: Steve French --- fs/cifs/smb2pdu.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index e543431db916..ce83e2edbe0a 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -922,6 +922,12 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree, if (tcon && tcon->bad_network_name) return -ENOENT; + if ((tcon->seal) && + ((ses->server->capabilities & SMB2_GLOBAL_CAP_ENCRYPTION) == 0)) { + cifs_dbg(VFS, "encryption requested but no server support"); + return -EOPNOTSUPP; + } + unc_path = kmalloc(MAX_SHARENAME_LENGTH * 2, GFP_KERNEL); if (unc_path == NULL) return -ENOMEM; @@ -1001,6 +1007,8 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree, ((tcon->share_flags & SHI1005_FLAGS_DFS) == 0)) cifs_dbg(VFS, "DFS capability contradicts DFS flag\n"); init_copy_chunk_defaults(tcon); + if (tcon->share_flags & SHI1005_FLAGS_ENCRYPT_DATA) + cifs_dbg(VFS, "Encrypted shares not supported"); if (tcon->ses->server->ops->validate_negotiate) rc = tcon->ses->server->ops->validate_negotiate(xid, tcon); tcon_exit: From 49558b471369e790650df8706b5608fee27af42c Mon Sep 17 00:00:00 2001 From: Christian Engelmayer Date: Sat, 19 Sep 2015 00:31:10 +0200 Subject: [PATCH 410/647] drm/vmwgfx: Fix uninitialized return in vmw_cotable_unbind() Function vmw_cotable_unbind() uses the uninitialized variable ret as return value. Make the result deterministic and directly return as the variable is unused anyway. Detected by Coverity CID 1324256. Signed-off-by: Christian Engelmayer Reviewed-by: Sinclair Yeh Reviewed-by: Thomas Hellstrom --- drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c index ce659a125f2b..092ea81eeff7 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c @@ -311,7 +311,6 @@ static int vmw_cotable_unbind(struct vmw_resource *res, struct vmw_private *dev_priv = res->dev_priv; struct ttm_buffer_object *bo = val_buf->bo; struct vmw_fence_obj *fence; - int ret; if (list_empty(&res->mob_head)) return 0; @@ -328,7 +327,7 @@ static int vmw_cotable_unbind(struct vmw_resource *res, if (likely(fence != NULL)) vmw_fence_obj_unreference(&fence); - return ret; + return 0; } /** From f3b8c0caca02001565cb14b845bf90f59dea8213 Mon Sep 17 00:00:00 2001 From: Christian Engelmayer Date: Sat, 19 Sep 2015 00:32:24 +0200 Subject: [PATCH 411/647] drm/vmwgfx: Fix uninitialized return in vmw_kms_helper_dirty() Function vmw_kms_helper_dirty() uses the uninitialized variable ret as return value. Make the result deterministic and directly return as the variable is unused anyway. Detected by Coverity CID 1324255. Signed-off-by: Christian Engelmayer Reviewed-by: Sinclair Yeh Reviewed-by: Thomas Hellstrom --- drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index 61fb7f3de311..15a6c01cd016 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -1685,7 +1685,6 @@ int vmw_kms_helper_dirty(struct vmw_private *dev_priv, struct drm_crtc *crtc; u32 num_units = 0; u32 i, k; - int ret; dirty->dev_priv = dev_priv; @@ -1711,7 +1710,7 @@ int vmw_kms_helper_dirty(struct vmw_private *dev_priv, if (!dirty->cmd) { DRM_ERROR("Couldn't reserve fifo space " "for dirty blits.\n"); - return ret; + return -ENOMEM; } memset(dirty->cmd, 0, dirty->fifo_reserve_size); } From 30c64664f110f76064e364cb5dd385edc3751ba5 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Tue, 15 Sep 2015 01:11:42 -0700 Subject: [PATCH 412/647] drm: Allow also control clients to check the drm version This should be harmless. Vmware will, due to old infrastructure reasons, be using a privileged control client to supply GUI layout information rather than obtaining it from the device. That control client will be needing access to DRM version information. Signed-off-by: Thomas Hellstrom Reviewed-by: Brian Paul Reviewed-by: Sinclair Yeh Acked-by: David Herrmann --- drivers/gpu/drm/drm_ioctl.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c index 9a860ca1e9d7..d93e7378c077 100644 --- a/drivers/gpu/drm/drm_ioctl.c +++ b/drivers/gpu/drm/drm_ioctl.c @@ -520,7 +520,8 @@ EXPORT_SYMBOL(drm_ioctl_permit); /** Ioctl table */ static const struct drm_ioctl_desc drm_ioctls[] = { - DRM_IOCTL_DEF(DRM_IOCTL_VERSION, drm_version, DRM_UNLOCKED|DRM_RENDER_ALLOW), + DRM_IOCTL_DEF(DRM_IOCTL_VERSION, drm_version, + DRM_UNLOCKED|DRM_RENDER_ALLOW|DRM_CONTROL_ALLOW), DRM_IOCTL_DEF(DRM_IOCTL_GET_UNIQUE, drm_getunique, 0), DRM_IOCTL_DEF(DRM_IOCTL_GET_MAGIC, drm_getmagic, 0), DRM_IOCTL_DEF(DRM_IOCTL_IRQ_BUSID, drm_irq_by_busid, DRM_MASTER|DRM_ROOT_ONLY), From 274e91b81ed22957b510ad2988359584eea95dae Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 23 Sep 2015 11:06:30 +0100 Subject: [PATCH 413/647] ARM: alignment: fix alignment handling for uaccess changes Jonathan Liu reports that the recent addition of CPU_SW_DOMAIN_PAN causes wpa_supplicant to die due to the following kernel oops: Unhandled fault: page domain fault (0x81b) at 0x001017a2 pgd = ee1b8000 [001017a2] *pgd=6ebee831, *pte=6c35475f, *ppte=6c354c7f Internal error: : 81b [#1] SMP ARM Modules linked in: rt2800usb rt2x00usb rt2800librt2x00lib crc_ccitt mac80211 CPU: 1 PID: 202 Comm: wpa_supplicant Not tainted 4.3.0-rc2 #1 Hardware name: Allwinner sun7i (A20) Family task: ec872f80 ti: ee364000 task.ti: ee364000 PC is at do_alignment_ldmstm+0x1d4/0x238 LR is at 0x0 pc : [] lr : [<00000000>] psr: 600c0113 sp : ee365e18 ip : 00000000 fp : 00000002 r10: 001017a2 r9 : 00000002 r8 : 001017aa r7 : ee365fb0 r6 : e8820018 r5 : 001017a2 r4 : 00000003 r3 : d49e30e0 r2 : 00000000 r1 : ee365fbc r0 : 00000000 Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none[ 34.393106] Control: 10c5387d Table: 6e1b806a DAC: 00000051 Process wpa_supplicant (pid: 202, stack limit = 0xee364210) Stack: (0xee365e18 to 0xee366000) ... [] (do_alignment_ldmstm) from [] (do_alignment+0x1f0/0x904) [] (do_alignment) from [] (do_DataAbort+0x38/0xb4) [] (do_DataAbort) from [] (__dabt_usr+0x3c/0x40) Exception stack(0xee365fb0 to 0xee365ff8) 5fa0: 00000000 56c728c0 001017a2 d49e30e0 5fc0: 775448d2 597d4e74 00200800 7a9e1625 00802001 00000021 b6deec84 00000100 5fe0: 08020200 be9f4f20 0c0b0d0a b6d9b3e0 600c0010 ffffffff Code: e1a0a005 e1a0000c 1affffe8 e5913000 (e4ea3001) ---[ end trace 0acd3882fcfdf9dd ]--- This is caused by the alignment handler not being fixed up for the uaccess changes, and userspace issuing an unaligned LDM instruction. So, fix the problem by adding the necessary fixups. Reported-by: Jonathan Liu Tested-by: Jonathan Liu Signed-off-by: Russell King --- arch/arm/mm/alignment.c | 30 +++++++++++++++++++++++++----- 1 file changed, 25 insertions(+), 5 deletions(-) diff --git a/arch/arm/mm/alignment.c b/arch/arm/mm/alignment.c index 9769f1eefe3b..00b7f7de28a1 100644 --- a/arch/arm/mm/alignment.c +++ b/arch/arm/mm/alignment.c @@ -365,15 +365,21 @@ do_alignment_ldrhstrh(unsigned long addr, unsigned long instr, struct pt_regs *r user: if (LDST_L_BIT(instr)) { unsigned long val; + unsigned int __ua_flags = uaccess_save_and_enable(); + get16t_unaligned_check(val, addr); + uaccess_restore(__ua_flags); /* signed half-word? */ if (instr & 0x40) val = (signed long)((signed short) val); regs->uregs[rd] = val; - } else + } else { + unsigned int __ua_flags = uaccess_save_and_enable(); put16t_unaligned_check(regs->uregs[rd], addr); + uaccess_restore(__ua_flags); + } return TYPE_LDST; @@ -420,14 +426,21 @@ do_alignment_ldrdstrd(unsigned long addr, unsigned long instr, user: if (load) { - unsigned long val; + unsigned long val, val2; + unsigned int __ua_flags = uaccess_save_and_enable(); + get32t_unaligned_check(val, addr); + get32t_unaligned_check(val2, addr + 4); + + uaccess_restore(__ua_flags); + regs->uregs[rd] = val; - get32t_unaligned_check(val, addr + 4); - regs->uregs[rd2] = val; + regs->uregs[rd2] = val2; } else { + unsigned int __ua_flags = uaccess_save_and_enable(); put32t_unaligned_check(regs->uregs[rd], addr); put32t_unaligned_check(regs->uregs[rd2], addr + 4); + uaccess_restore(__ua_flags); } return TYPE_LDST; @@ -458,10 +471,15 @@ do_alignment_ldrstr(unsigned long addr, unsigned long instr, struct pt_regs *reg trans: if (LDST_L_BIT(instr)) { unsigned int val; + unsigned int __ua_flags = uaccess_save_and_enable(); get32t_unaligned_check(val, addr); + uaccess_restore(__ua_flags); regs->uregs[rd] = val; - } else + } else { + unsigned int __ua_flags = uaccess_save_and_enable(); put32t_unaligned_check(regs->uregs[rd], addr); + uaccess_restore(__ua_flags); + } return TYPE_LDST; fault: @@ -531,6 +549,7 @@ do_alignment_ldmstm(unsigned long addr, unsigned long instr, struct pt_regs *reg #endif if (user_mode(regs)) { + unsigned int __ua_flags = uaccess_save_and_enable(); for (regbits = REGMASK_BITS(instr), rd = 0; regbits; regbits >>= 1, rd += 1) if (regbits & 1) { @@ -542,6 +561,7 @@ do_alignment_ldmstm(unsigned long addr, unsigned long instr, struct pt_regs *reg put32t_unaligned_check(regs->uregs[rd], eaddr); eaddr += 4; } + uaccess_restore(__ua_flags); } else { for (regbits = REGMASK_BITS(instr), rd = 0; regbits; regbits >>= 1, rd += 1) From 83510441bc08bee201c0ded9d81da6dfd008d69a Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 24 Sep 2015 11:00:18 +0200 Subject: [PATCH 414/647] ALSA: hda/tegra - async probe for avoiding module loading deadlock The Tegra HD-audio controller driver causes deadlocks when loaded as a module since the driver invokes request_module() at binding with the codec driver. This patch works around it by deferring the probe in a work like Intel HD-audio controller driver does. Although hovering the codec probe stuff into udev would be a better solution, it may cause other regressions, so let's try this band-aid fix until the more proper solution gets landed. Reported-by: Thierry Reding Tested-by: Thierry Reding Cc: Signed-off-by: Takashi Iwai --- sound/pci/hda/hda_tegra.c | 30 +++++++++++++++++++++++++----- 1 file changed, 25 insertions(+), 5 deletions(-) diff --git a/sound/pci/hda/hda_tegra.c b/sound/pci/hda/hda_tegra.c index 477742cb70a2..58c0aad37284 100644 --- a/sound/pci/hda/hda_tegra.c +++ b/sound/pci/hda/hda_tegra.c @@ -73,6 +73,7 @@ struct hda_tegra { struct clk *hda2codec_2x_clk; struct clk *hda2hdmi_clk; void __iomem *regs; + struct work_struct probe_work; }; #ifdef CONFIG_PM @@ -294,7 +295,9 @@ static int hda_tegra_dev_disconnect(struct snd_device *device) static int hda_tegra_dev_free(struct snd_device *device) { struct azx *chip = device->device_data; + struct hda_tegra *hda = container_of(chip, struct hda_tegra, chip); + cancel_work_sync(&hda->probe_work); if (azx_bus(chip)->chip_init) { azx_stop_all_streams(chip); azx_stop_chip(chip); @@ -426,6 +429,9 @@ static int hda_tegra_first_init(struct azx *chip, struct platform_device *pdev) /* * constructor */ + +static void hda_tegra_probe_work(struct work_struct *work); + static int hda_tegra_create(struct snd_card *card, unsigned int driver_caps, struct hda_tegra *hda) @@ -452,6 +458,8 @@ static int hda_tegra_create(struct snd_card *card, chip->single_cmd = false; chip->snoop = true; + INIT_WORK(&hda->probe_work, hda_tegra_probe_work); + err = azx_bus_init(chip, NULL, &hda_tegra_io_ops); if (err < 0) return err; @@ -499,6 +507,21 @@ static int hda_tegra_probe(struct platform_device *pdev) card->private_data = chip; dev_set_drvdata(&pdev->dev, card); + schedule_work(&hda->probe_work); + + return 0; + +out_free: + snd_card_free(card); + return err; +} + +static void hda_tegra_probe_work(struct work_struct *work) +{ + struct hda_tegra *hda = container_of(work, struct hda_tegra, probe_work); + struct azx *chip = &hda->chip; + struct platform_device *pdev = to_platform_device(hda->dev); + int err; err = hda_tegra_first_init(chip, pdev); if (err < 0) @@ -520,11 +543,8 @@ static int hda_tegra_probe(struct platform_device *pdev) chip->running = 1; snd_hda_set_power_save(&chip->bus, power_save * 1000); - return 0; - -out_free: - snd_card_free(card); - return err; + out_free: + return; /* no error return from async probe */ } static int hda_tegra_remove(struct platform_device *pdev) From 7f57d803ee03730d570dc59a9e3e4842b58dd5cc Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 24 Sep 2015 17:36:51 +0200 Subject: [PATCH 415/647] ALSA: hda - Disable power_save_node for Thinkpads Lenovo Thinkpads with recent Realtek codecs seem suffering from click noises at power transition since the introduction of widget power saving in 4.1 kernel. Although this might be solved by some delays in appropriate points, as a quick workaround, just disable the power_save_node feature for now. The gain it gives is relatively small, and this makes the situation back to pre 4.1 time. This patch ended up with a bit more code changes than usual because the existing fixup for Thinkpads is highly chained. Instead of adding yet another chain, combine a few of them into a single fixup entry, as a gratis cleanup. Bugzilla: https://bugzilla.suse.com/show_bug.cgi?id=943982 Cc: # v4.1+ Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 31 +++++++++++++++++++------------ 1 file changed, 19 insertions(+), 12 deletions(-) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index a75b5611d1e4..afec6dc9f91f 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -4188,6 +4188,24 @@ static void alc_fixup_disable_aamix(struct hda_codec *codec, } } +/* fixup for Thinkpad docks: add dock pins, avoid HP parser fixup */ +static void alc_fixup_tpt440_dock(struct hda_codec *codec, + const struct hda_fixup *fix, int action) +{ + static const struct hda_pintbl pincfgs[] = { + { 0x16, 0x21211010 }, /* dock headphone */ + { 0x19, 0x21a11010 }, /* dock mic */ + { } + }; + struct alc_spec *spec = codec->spec; + + if (action == HDA_FIXUP_ACT_PRE_PROBE) { + spec->parse_flags = HDA_PINCFG_NO_HP_FIXUP; + codec->power_save_node = 0; /* avoid click noises */ + snd_hda_apply_pincfgs(codec, pincfgs); + } +} + static void alc_shutup_dell_xps13(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; @@ -4562,7 +4580,6 @@ enum { ALC255_FIXUP_HEADSET_MODE_NO_HP_MIC, ALC293_FIXUP_DELL1_MIC_NO_PRESENCE, ALC292_FIXUP_TPT440_DOCK, - ALC292_FIXUP_TPT440_DOCK2, ALC283_FIXUP_BXBT2807_MIC, ALC255_FIXUP_DELL_WMI_MIC_MUTE_LED, ALC282_FIXUP_ASPIRE_V5_PINS, @@ -5029,17 +5046,7 @@ static const struct hda_fixup alc269_fixups[] = { }, [ALC292_FIXUP_TPT440_DOCK] = { .type = HDA_FIXUP_FUNC, - .v.func = alc269_fixup_pincfg_no_hp_to_lineout, - .chained = true, - .chain_id = ALC292_FIXUP_TPT440_DOCK2 - }, - [ALC292_FIXUP_TPT440_DOCK2] = { - .type = HDA_FIXUP_PINS, - .v.pins = (const struct hda_pintbl[]) { - { 0x16, 0x21211010 }, /* dock headphone */ - { 0x19, 0x21a11010 }, /* dock mic */ - { } - }, + .v.func = alc_fixup_tpt440_dock, .chained = true, .chain_id = ALC269_FIXUP_LIMIT_INT_MIC_BOOST }, From da314c9923fed553a007785a901fd395b7eb6c19 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 22 Sep 2015 11:38:56 +0800 Subject: [PATCH 416/647] netlink: Replace rhash_portid with bound On Mon, Sep 21, 2015 at 02:20:22PM -0400, Tejun Heo wrote: > > store_release and load_acquire are different from the usual memory > barriers and can't be paired this way. You have to pair store_release > and load_acquire. Besides, it isn't a particularly good idea to OK I've decided to drop the acquire/release helpers as they don't help us at all and simply pessimises the code by using full memory barriers (on some architectures) where only a write or read barrier is needed. > depend on memory barriers embedded in other data structures like the > above. Here, especially, rhashtable_insert() would have write barrier > *before* the entry is hashed not necessarily *after*, which means that > in the above case, a socket which appears to have set bound to a > reader might not visible when the reader tries to look up the socket > on the hashtable. But you are right we do need an explicit write barrier here to ensure that the hashing is visible. > There's no reason to be overly smart here. This isn't a crazy hot > path, write barriers tend to be very cheap, store_release more so. > Please just do smp_store_release() and note what it's paired with. It's not about being overly smart. It's about actually understanding what's going on with the code. I've seen too many instances of people simply sprinkling synchronisation primitives around without any knowledge of what is happening underneath, which is just a recipe for creating hard-to-debug races. > > @@ -1539,7 +1546,7 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, > > } > > } > > > > - if (!nlk->portid) { > > + if (!nlk->bound) { > > I don't think you can skip load_acquire here just because this is the > second deref of the variable. That doesn't change anything. Race > condition could still happen between the first and second tests and > skipping the second would lead to the same kind of bug. The reason this one is OK is because we do not use nlk->portid or try to get nlk from the hash table before we return to user-space. However, there is a real bug here that none of these acquire/release helpers discovered. The two bound tests here used to be a single one. Now that they are separate it is entirely possible for another thread to come in the middle and bind the socket. So we need to repeat the portid check in order to maintain consistency. > > @@ -1587,7 +1594,7 @@ static int netlink_connect(struct socket *sock, struct sockaddr *addr, > > !netlink_allowed(sock, NL_CFG_F_NONROOT_SEND)) > > return -EPERM; > > > > - if (!nlk->portid) > > + if (!nlk->bound) > > Don't we need load_acquire here too? Is this path holding a lock > which makes that unnecessary? Ditto. ---8<--- The commit 1f770c0a09da855a2b51af6d19de97fb955eca85 ("netlink: Fix autobind race condition that leads to zero port ID") created some new races that can occur due to inconcsistencies between the two port IDs. Tejun is right that a barrier is unavoidable. Therefore I am reverting to the original patch that used a boolean to indicate that a user netlink socket has been bound. Barriers have been added where necessary to ensure that a valid portid and the hashed socket is visible. I have also changed netlink_insert to only return EBUSY if the socket is bound to a portid different to the requested one. This combined with only reading nlk->bound once in netlink_bind fixes a race where two threads that bind the socket at the same time with different port IDs may both succeed. Fixes: 1f770c0a09da ("netlink: Fix autobind race condition that leads to zero port ID") Reported-by: Tejun Heo Reported-by: Linus Torvalds Signed-off-by: Herbert Xu Nacked-by: Tejun Heo Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 39 ++++++++++++++++++++++++++++----------- net/netlink/af_netlink.h | 2 +- 2 files changed, 29 insertions(+), 12 deletions(-) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 9f51608b968a..8f060d7f9a0e 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1031,7 +1031,7 @@ static inline int netlink_compare(struct rhashtable_compare_arg *arg, const struct netlink_compare_arg *x = arg->key; const struct netlink_sock *nlk = ptr; - return nlk->rhash_portid != x->portid || + return nlk->portid != x->portid || !net_eq(sock_net(&nlk->sk), read_pnet(&x->pnet)); } @@ -1057,7 +1057,7 @@ static int __netlink_insert(struct netlink_table *table, struct sock *sk) { struct netlink_compare_arg arg; - netlink_compare_arg_init(&arg, sock_net(sk), nlk_sk(sk)->rhash_portid); + netlink_compare_arg_init(&arg, sock_net(sk), nlk_sk(sk)->portid); return rhashtable_lookup_insert_key(&table->hash, &arg, &nlk_sk(sk)->node, netlink_rhashtable_params); @@ -1110,8 +1110,8 @@ static int netlink_insert(struct sock *sk, u32 portid) lock_sock(sk); - err = -EBUSY; - if (nlk_sk(sk)->portid) + err = nlk_sk(sk)->portid == portid ? 0 : -EBUSY; + if (nlk_sk(sk)->bound) goto err; err = -ENOMEM; @@ -1119,7 +1119,7 @@ static int netlink_insert(struct sock *sk, u32 portid) unlikely(atomic_read(&table->hash.nelems) >= UINT_MAX)) goto err; - nlk_sk(sk)->rhash_portid = portid; + nlk_sk(sk)->portid = portid; sock_hold(sk); err = __netlink_insert(table, sk); @@ -1135,7 +1135,9 @@ static int netlink_insert(struct sock *sk, u32 portid) goto err; } - nlk_sk(sk)->portid = portid; + /* We need to ensure that the socket is hashed and visible. */ + smp_wmb(); + nlk_sk(sk)->bound = portid; err: release_sock(sk); @@ -1521,6 +1523,7 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, struct sockaddr_nl *nladdr = (struct sockaddr_nl *)addr; int err; long unsigned int groups = nladdr->nl_groups; + bool bound; if (addr_len < sizeof(struct sockaddr_nl)) return -EINVAL; @@ -1537,9 +1540,14 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, return err; } - if (nlk->portid) + bound = nlk->bound; + if (bound) { + /* Ensure nlk->portid is up-to-date. */ + smp_rmb(); + if (nladdr->nl_pid != nlk->portid) return -EINVAL; + } if (nlk->netlink_bind && groups) { int group; @@ -1555,7 +1563,10 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, } } - if (!nlk->portid) { + /* No need for barriers here as we return to user-space without + * using any of the bound attributes. + */ + if (!bound) { err = nladdr->nl_pid ? netlink_insert(sk, nladdr->nl_pid) : netlink_autobind(sock); @@ -1603,7 +1614,10 @@ static int netlink_connect(struct socket *sock, struct sockaddr *addr, !netlink_allowed(sock, NL_CFG_F_NONROOT_SEND)) return -EPERM; - if (!nlk->portid) + /* No need for barriers here as we return to user-space without + * using any of the bound attributes. + */ + if (!nlk->bound) err = netlink_autobind(sock); if (err == 0) { @@ -2444,10 +2458,13 @@ static int netlink_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) dst_group = nlk->dst_group; } - if (!nlk->portid) { + if (!nlk->bound) { err = netlink_autobind(sock); if (err) goto out; + } else { + /* Ensure nlk is hashed and visible. */ + smp_rmb(); } /* It's a really convoluted way for userland to ask for mmaped @@ -3273,7 +3290,7 @@ static inline u32 netlink_hash(const void *data, u32 len, u32 seed) const struct netlink_sock *nlk = data; struct netlink_compare_arg arg; - netlink_compare_arg_init(&arg, sock_net(&nlk->sk), nlk->rhash_portid); + netlink_compare_arg_init(&arg, sock_net(&nlk->sk), nlk->portid); return jhash2((u32 *)&arg, netlink_compare_arg_len / sizeof(u32), seed); } diff --git a/net/netlink/af_netlink.h b/net/netlink/af_netlink.h index 80b2b7526dfd..14437d9b1965 100644 --- a/net/netlink/af_netlink.h +++ b/net/netlink/af_netlink.h @@ -25,7 +25,6 @@ struct netlink_ring { struct netlink_sock { /* struct sock has to be the first member of netlink_sock */ struct sock sk; - u32 rhash_portid; u32 portid; u32 dst_portid; u32 dst_group; @@ -36,6 +35,7 @@ struct netlink_sock { unsigned long state; size_t max_recvmsg_len; wait_queue_head_t wait; + bool bound; bool cb_running; struct netlink_callback cb; struct mutex *cb_mutex; From 9badce000e2ce68ba74838a3cd356dde58221c2f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 23 Sep 2015 17:07:29 -0400 Subject: [PATCH 417/647] cgroup, writeback: don't enable cgroup writeback on traditional hierarchies inode_cgwb_enabled() gates cgroup writeback support. If it returns true, each inode is attached to the corresponding memory domain which gets mapped to io domain. It currently only tests whether the filesystem and bdi support cgroup writeback; however, cgroup writeback support doesn't work on traditional hierarchies and thus it should also test whether memcg and iocg are on the default hierarchy. This caused traditional hierarchy setups to hit the cgroup writeback path inadvertently and ended up creating separate writeback domains for each memcg and mapping them all to the root iocg uncovering a couple issues in the cgroup writeback path. cgroup writeback was never meant to be enabled on traditional hierarchies. Make inode_cgwb_enabled() test whether both memcg and iocg are on the default hierarchy. Signed-off-by: Tejun Heo Reported-by: Artem Bityutskiy Reported-by: Dexuan Cui Link: http://lkml.kernel.org/g/1443012552.19983.209.camel@gmail.com Link: http://lkml.kernel.org/g/f30d4a6aa8a546ff88f73021d026a453@SIXPR30MB031.064d.mgd.msft.net --- include/linux/backing-dev.h | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 5a5d79ee256f..d5eb4ad1c534 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -252,13 +253,19 @@ int inode_congested(struct inode *inode, int cong_bits); * @inode: inode of interest * * cgroup writeback requires support from both the bdi and filesystem. - * Test whether @inode has both. + * Also, both memcg and iocg have to be on the default hierarchy. Test + * whether all conditions are met. + * + * Note that the test result may change dynamically on the same inode + * depending on how memcg and iocg are configured. */ static inline bool inode_cgwb_enabled(struct inode *inode) { struct backing_dev_info *bdi = inode_to_bdi(inode); - return bdi_cap_account_dirty(bdi) && + return cgroup_on_dfl(mem_cgroup_root_css->cgroup) && + cgroup_on_dfl(blkcg_root_css->cgroup) && + bdi_cap_account_dirty(bdi) && (bdi->capabilities & BDI_CAP_CGROUP_WRITEBACK) && (inode->i_sb->s_iflags & SB_I_CGROUPWB); } From 6ae459bdaaeebc632b16e54dcbabb490c6931d61 Mon Sep 17 00:00:00 2001 From: Pravin B Shelar Date: Tue, 22 Sep 2015 12:57:53 -0700 Subject: [PATCH 418/647] skbuff: Fix skb checksum flag on skb pull VXLAN device can receive skb with checksum partial. But the checksum offset could be in outer header which is pulled on receive. This results in negative checksum offset for the skb. Such skb can cause the assert failure in skb_checksum_help(). Following patch fixes the bug by setting checksum-none while pulling outer header. Following is the kernel panic msg from old kernel hitting the bug. ------------[ cut here ]------------ kernel BUG at net/core/dev.c:1906! RIP: 0010:[] skb_checksum_help+0x144/0x150 Call Trace: [] queue_userspace_packet+0x408/0x470 [openvswitch] [] ovs_dp_upcall+0x5d/0x60 [openvswitch] [] ovs_dp_process_packet_with_key+0xe6/0x100 [openvswitch] [] ovs_dp_process_received_packet+0x4b/0x80 [openvswitch] [] ovs_vport_receive+0x2a/0x30 [openvswitch] [] vxlan_rcv+0x53/0x60 [openvswitch] [] vxlan_udp_encap_recv+0x8b/0xf0 [openvswitch] [] udp_queue_rcv_skb+0x2dc/0x3b0 [] __udp4_lib_rcv+0x1cf/0x6c0 [] udp_rcv+0x1a/0x20 [] ip_local_deliver_finish+0xdd/0x280 [] ip_local_deliver+0x88/0x90 [] ip_rcv_finish+0x10d/0x370 [] ip_rcv+0x235/0x300 [] __netif_receive_skb+0x55d/0x620 [] netif_receive_skb+0x80/0x90 [] virtnet_poll+0x555/0x6f0 [] net_rx_action+0x134/0x290 [] __do_softirq+0xa8/0x210 [] call_softirq+0x1c/0x30 [] do_softirq+0x65/0xa0 [] irq_exit+0x8e/0xb0 [] do_IRQ+0x63/0xe0 [] common_interrupt+0x6e/0x6e Reported-by: Anupam Chanda Signed-off-by: Pravin B Shelar Acked-by: Tom Herbert Signed-off-by: David S. Miller --- include/linux/skbuff.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 9987af080fa0..2b0a30a6e31c 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -2707,6 +2707,9 @@ static inline void skb_postpull_rcsum(struct sk_buff *skb, { if (skb->ip_summed == CHECKSUM_COMPLETE) skb->csum = csum_sub(skb->csum, csum_partial(start, len, 0)); + else if (skb->ip_summed == CHECKSUM_PARTIAL && + skb_checksum_start_offset(skb) <= len) + skb->ip_summed = CHECKSUM_NONE; } unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len); From d5b8d6404395641987db76e28334cae4cef771ae Mon Sep 17 00:00:00 2001 From: Sudeep Holla Date: Mon, 21 Sep 2015 16:47:09 +0100 Subject: [PATCH 419/647] net: gianfar: remove misuse of IRQF_NO_SUSPEND flag The device is set as wakeup capable using proper wakeup API but the driver misuses IRQF_NO_SUSPEND to set the interrupt as wakeup source which is incorrect. This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead. Cc: "David S. Miller" Cc: Claudiu Manoil Cc: Kevin Hao Cc: netdev@vger.kernel.org Signed-off-by: Sudeep Holla Acked-by: Claudiu Manoil Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/gianfar.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index 4b69d061d90f..803ed4c93503 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -1970,8 +1970,7 @@ static int register_grp_irqs(struct gfar_priv_grp *grp) /* Install our interrupt handlers for Error, * Transmit, and Receive */ - err = request_irq(gfar_irq(grp, ER)->irq, gfar_error, - IRQF_NO_SUSPEND, + err = request_irq(gfar_irq(grp, ER)->irq, gfar_error, 0, gfar_irq(grp, ER)->name, grp); if (err < 0) { netif_err(priv, intr, dev, "Can't get IRQ %d\n", @@ -1979,6 +1978,8 @@ static int register_grp_irqs(struct gfar_priv_grp *grp) goto err_irq_fail; } + enable_irq_wake(gfar_irq(grp, ER)->irq); + err = request_irq(gfar_irq(grp, TX)->irq, gfar_transmit, 0, gfar_irq(grp, TX)->name, grp); if (err < 0) { @@ -1994,14 +1995,14 @@ static int register_grp_irqs(struct gfar_priv_grp *grp) goto rx_irq_fail; } } else { - err = request_irq(gfar_irq(grp, TX)->irq, gfar_interrupt, - IRQF_NO_SUSPEND, + err = request_irq(gfar_irq(grp, TX)->irq, gfar_interrupt, 0, gfar_irq(grp, TX)->name, grp); if (err < 0) { netif_err(priv, intr, dev, "Can't get IRQ %d\n", gfar_irq(grp, TX)->irq); goto err_irq_fail; } + enable_irq_wake(gfar_irq(grp, TX)->irq); } return 0; From 63d008a4e9ee86614ca5671b7f3ba447df007190 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Tue, 22 Sep 2015 18:12:11 +0200 Subject: [PATCH 420/647] ipv4: send arp replies to the correct tunnel When using ip lwtunnels, the additional data for xmit (basically, the actual tunnel to use) are carried in ip_tunnel_info either in dst->lwtstate or in metadata dst. When replying to ARP requests, we need to send the reply to the same tunnel the request came from. This means we need to construct proper metadata dst for ARP replies. We could perform another route lookup to get a dst entry with the correct lwtstate. However, this won't always ensure that the outgoing tunnel is the same as the incoming one, and it won't work anyway for IPv4 duplicate address detection. The only thing to do is to "reverse" the ip_tunnel_info. Signed-off-by: Jiri Benc Acked-by: Thomas Graf Signed-off-by: David S. Miller --- include/net/ip_tunnels.h | 2 ++ net/ipv4/arp.c | 39 +++++++++++++++++++++++++-------------- net/ipv4/ip_tunnel_core.c | 28 ++++++++++++++++++++++++++++ 3 files changed, 55 insertions(+), 14 deletions(-) diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index 9a6a3ba888e8..f6dafec9102c 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -276,6 +276,8 @@ int iptunnel_pull_header(struct sk_buff *skb, int hdr_len, __be16 inner_proto); int iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, __be32 src, __be32 dst, u8 proto, u8 tos, u8 ttl, __be16 df, bool xnet); +struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md, + gfp_t flags); struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb, bool gre_csum, int gso_type_mask); diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 30409b75e925..f03db8b7abee 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -113,6 +113,8 @@ #include #include #include +#include +#include #include @@ -296,7 +298,8 @@ static void arp_send_dst(int type, int ptype, __be32 dest_ip, struct net_device *dev, __be32 src_ip, const unsigned char *dest_hw, const unsigned char *src_hw, - const unsigned char *target_hw, struct sk_buff *oskb) + const unsigned char *target_hw, + struct dst_entry *dst) { struct sk_buff *skb; @@ -309,9 +312,7 @@ static void arp_send_dst(int type, int ptype, __be32 dest_ip, if (!skb) return; - if (oskb) - skb_dst_copy(skb, oskb); - + skb_dst_set(skb, dst); arp_xmit(skb); } @@ -333,6 +334,7 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb) __be32 target = *(__be32 *)neigh->primary_key; int probes = atomic_read(&neigh->probes); struct in_device *in_dev; + struct dst_entry *dst = NULL; rcu_read_lock(); in_dev = __in_dev_get_rcu(dev); @@ -381,9 +383,10 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb) } } + if (skb && !(dev->priv_flags & IFF_XMIT_DST_RELEASE)) + dst = dst_clone(skb_dst(skb)); arp_send_dst(ARPOP_REQUEST, ETH_P_ARP, target, dev, saddr, - dst_hw, dev->dev_addr, NULL, - dev->priv_flags & IFF_XMIT_DST_RELEASE ? NULL : skb); + dst_hw, dev->dev_addr, NULL, dst); } static int arp_ignore(struct in_device *in_dev, __be32 sip, __be32 tip) @@ -649,6 +652,7 @@ static int arp_process(struct sock *sk, struct sk_buff *skb) int addr_type; struct neighbour *n; struct net *net = dev_net(dev); + struct dst_entry *reply_dst = NULL; bool is_garp = false; /* arp_rcv below verifies the ARP header and verifies the device @@ -749,13 +753,18 @@ static int arp_process(struct sock *sk, struct sk_buff *skb) * cache. */ + if (arp->ar_op == htons(ARPOP_REQUEST) && skb_metadata_dst(skb)) + reply_dst = (struct dst_entry *) + iptunnel_metadata_reply(skb_metadata_dst(skb), + GFP_ATOMIC); + /* Special case: IPv4 duplicate address detection packet (RFC2131) */ if (sip == 0) { if (arp->ar_op == htons(ARPOP_REQUEST) && inet_addr_type_dev_table(net, dev, tip) == RTN_LOCAL && !arp_ignore(in_dev, sip, tip)) - arp_send(ARPOP_REPLY, ETH_P_ARP, sip, dev, tip, sha, - dev->dev_addr, sha); + arp_send_dst(ARPOP_REPLY, ETH_P_ARP, sip, dev, tip, + sha, dev->dev_addr, sha, reply_dst); goto out; } @@ -774,9 +783,10 @@ static int arp_process(struct sock *sk, struct sk_buff *skb) if (!dont_send) { n = neigh_event_ns(&arp_tbl, sha, &sip, dev); if (n) { - arp_send(ARPOP_REPLY, ETH_P_ARP, sip, - dev, tip, sha, dev->dev_addr, - sha); + arp_send_dst(ARPOP_REPLY, ETH_P_ARP, + sip, dev, tip, sha, + dev->dev_addr, sha, + reply_dst); neigh_release(n); } } @@ -794,9 +804,10 @@ static int arp_process(struct sock *sk, struct sk_buff *skb) if (NEIGH_CB(skb)->flags & LOCALLY_ENQUEUED || skb->pkt_type == PACKET_HOST || NEIGH_VAR(in_dev->arp_parms, PROXY_DELAY) == 0) { - arp_send(ARPOP_REPLY, ETH_P_ARP, sip, - dev, tip, sha, dev->dev_addr, - sha); + arp_send_dst(ARPOP_REPLY, ETH_P_ARP, + sip, dev, tip, sha, + dev->dev_addr, sha, + reply_dst); } else { pneigh_enqueue(&arp_tbl, in_dev->arp_parms, skb); diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 9b97204b8c81..ce3a1e728606 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -46,6 +46,7 @@ #include #include #include +#include int iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, __be32 src, __be32 dst, __u8 proto, @@ -119,6 +120,33 @@ int iptunnel_pull_header(struct sk_buff *skb, int hdr_len, __be16 inner_proto) } EXPORT_SYMBOL_GPL(iptunnel_pull_header); +struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md, + gfp_t flags) +{ + struct metadata_dst *res; + struct ip_tunnel_info *dst, *src; + + if (!md || md->u.tun_info.mode & IP_TUNNEL_INFO_TX) + return NULL; + + res = metadata_dst_alloc(0, flags); + if (!res) + return NULL; + + dst = &res->u.tun_info; + src = &md->u.tun_info; + dst->key.tun_id = src->key.tun_id; + if (src->mode & IP_TUNNEL_INFO_IPV6) + memcpy(&dst->key.u.ipv6.dst, &src->key.u.ipv6.src, + sizeof(struct in6_addr)); + else + dst->key.u.ipv4.dst = src->key.u.ipv4.src; + dst->mode = src->mode | IP_TUNNEL_INFO_TX; + + return res; +} +EXPORT_SYMBOL_GPL(iptunnel_metadata_reply); + struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb, bool csum_help, int gso_type_mask) From b194f30c61efb0767a98f47a64530baa8b731670 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Tue, 22 Sep 2015 18:12:12 +0200 Subject: [PATCH 421/647] lwtunnel: remove source and destination UDP port config option The UDP tunnel config is asymmetric wrt. to the ports used. The source and destination ports from one direction of the tunnel are not related to the ports of the other direction. We need to be able to respond to ARP requests using the correct ports without involving routing. As the consequence, UDP ports need to be fixed property of the tunnel interface and cannot be set per route. Remove the ability to set ports per route. This is still okay to do, as no kernel has been released with these attributes yet. Note that the ability to specify source and destination ports is preserved for other users of the lwtunnel API which don't use routes for tunnel key specification (like openvswitch). If in the future we rework ARP handling to allow port specification, the attributes can be added back. Signed-off-by: Jiri Benc Acked-by: Thomas Graf Signed-off-by: David S. Miller --- include/uapi/linux/lwtunnel.h | 4 ---- net/ipv4/ip_tunnel_core.c | 24 ------------------------ 2 files changed, 28 deletions(-) diff --git a/include/uapi/linux/lwtunnel.h b/include/uapi/linux/lwtunnel.h index 34141a5dfe74..f8b01887a495 100644 --- a/include/uapi/linux/lwtunnel.h +++ b/include/uapi/linux/lwtunnel.h @@ -21,8 +21,6 @@ enum lwtunnel_ip_t { LWTUNNEL_IP_SRC, LWTUNNEL_IP_TTL, LWTUNNEL_IP_TOS, - LWTUNNEL_IP_SPORT, - LWTUNNEL_IP_DPORT, LWTUNNEL_IP_FLAGS, __LWTUNNEL_IP_MAX, }; @@ -36,8 +34,6 @@ enum lwtunnel_ip6_t { LWTUNNEL_IP6_SRC, LWTUNNEL_IP6_HOPLIMIT, LWTUNNEL_IP6_TC, - LWTUNNEL_IP6_SPORT, - LWTUNNEL_IP6_DPORT, LWTUNNEL_IP6_FLAGS, __LWTUNNEL_IP6_MAX, }; diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index ce3a1e728606..84dce6a92f93 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -226,8 +226,6 @@ static const struct nla_policy ip_tun_policy[LWTUNNEL_IP_MAX + 1] = { [LWTUNNEL_IP_SRC] = { .type = NLA_U32 }, [LWTUNNEL_IP_TTL] = { .type = NLA_U8 }, [LWTUNNEL_IP_TOS] = { .type = NLA_U8 }, - [LWTUNNEL_IP_SPORT] = { .type = NLA_U16 }, - [LWTUNNEL_IP_DPORT] = { .type = NLA_U16 }, [LWTUNNEL_IP_FLAGS] = { .type = NLA_U16 }, }; @@ -267,12 +265,6 @@ static int ip_tun_build_state(struct net_device *dev, struct nlattr *attr, if (tb[LWTUNNEL_IP_TOS]) tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP_TOS]); - if (tb[LWTUNNEL_IP_SPORT]) - tun_info->key.tp_src = nla_get_be16(tb[LWTUNNEL_IP_SPORT]); - - if (tb[LWTUNNEL_IP_DPORT]) - tun_info->key.tp_dst = nla_get_be16(tb[LWTUNNEL_IP_DPORT]); - if (tb[LWTUNNEL_IP_FLAGS]) tun_info->key.tun_flags = nla_get_u16(tb[LWTUNNEL_IP_FLAGS]); @@ -294,8 +286,6 @@ static int ip_tun_fill_encap_info(struct sk_buff *skb, nla_put_be32(skb, LWTUNNEL_IP_SRC, tun_info->key.u.ipv4.src) || nla_put_u8(skb, LWTUNNEL_IP_TOS, tun_info->key.tos) || nla_put_u8(skb, LWTUNNEL_IP_TTL, tun_info->key.ttl) || - nla_put_u16(skb, LWTUNNEL_IP_SPORT, tun_info->key.tp_src) || - nla_put_u16(skb, LWTUNNEL_IP_DPORT, tun_info->key.tp_dst) || nla_put_u16(skb, LWTUNNEL_IP_FLAGS, tun_info->key.tun_flags)) return -ENOMEM; @@ -309,8 +299,6 @@ static int ip_tun_encap_nlsize(struct lwtunnel_state *lwtstate) + nla_total_size(4) /* LWTUNNEL_IP_SRC */ + nla_total_size(1) /* LWTUNNEL_IP_TOS */ + nla_total_size(1) /* LWTUNNEL_IP_TTL */ - + nla_total_size(2) /* LWTUNNEL_IP_SPORT */ - + nla_total_size(2) /* LWTUNNEL_IP_DPORT */ + nla_total_size(2); /* LWTUNNEL_IP_FLAGS */ } @@ -333,8 +321,6 @@ static const struct nla_policy ip6_tun_policy[LWTUNNEL_IP6_MAX + 1] = { [LWTUNNEL_IP6_SRC] = { .len = sizeof(struct in6_addr) }, [LWTUNNEL_IP6_HOPLIMIT] = { .type = NLA_U8 }, [LWTUNNEL_IP6_TC] = { .type = NLA_U8 }, - [LWTUNNEL_IP6_SPORT] = { .type = NLA_U16 }, - [LWTUNNEL_IP6_DPORT] = { .type = NLA_U16 }, [LWTUNNEL_IP6_FLAGS] = { .type = NLA_U16 }, }; @@ -374,12 +360,6 @@ static int ip6_tun_build_state(struct net_device *dev, struct nlattr *attr, if (tb[LWTUNNEL_IP6_TC]) tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP6_TC]); - if (tb[LWTUNNEL_IP6_SPORT]) - tun_info->key.tp_src = nla_get_be16(tb[LWTUNNEL_IP6_SPORT]); - - if (tb[LWTUNNEL_IP6_DPORT]) - tun_info->key.tp_dst = nla_get_be16(tb[LWTUNNEL_IP6_DPORT]); - if (tb[LWTUNNEL_IP6_FLAGS]) tun_info->key.tun_flags = nla_get_u16(tb[LWTUNNEL_IP6_FLAGS]); @@ -401,8 +381,6 @@ static int ip6_tun_fill_encap_info(struct sk_buff *skb, nla_put_in6_addr(skb, LWTUNNEL_IP6_SRC, &tun_info->key.u.ipv6.src) || nla_put_u8(skb, LWTUNNEL_IP6_HOPLIMIT, tun_info->key.tos) || nla_put_u8(skb, LWTUNNEL_IP6_TC, tun_info->key.ttl) || - nla_put_u16(skb, LWTUNNEL_IP6_SPORT, tun_info->key.tp_src) || - nla_put_u16(skb, LWTUNNEL_IP6_DPORT, tun_info->key.tp_dst) || nla_put_u16(skb, LWTUNNEL_IP6_FLAGS, tun_info->key.tun_flags)) return -ENOMEM; @@ -416,8 +394,6 @@ static int ip6_tun_encap_nlsize(struct lwtunnel_state *lwtstate) + nla_total_size(16) /* LWTUNNEL_IP6_SRC */ + nla_total_size(1) /* LWTUNNEL_IP6_HOPLIMIT */ + nla_total_size(1) /* LWTUNNEL_IP6_TC */ - + nla_total_size(2) /* LWTUNNEL_IP6_SPORT */ - + nla_total_size(2) /* LWTUNNEL_IP6_DPORT */ + nla_total_size(2); /* LWTUNNEL_IP6_FLAGS */ } From d8aecb10115497f6cdf841df8c88ebb3ba25fa28 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Tue, 22 Sep 2015 17:01:11 -0700 Subject: [PATCH 422/647] net: revert "net_sched: move tp->root allocation into fw_init()" fw filter uses tp->root==NULL to check if it is the old method, so it doesn't need allocation at all in this case. This patch reverts the offending commit and adds some comments for old method to make it obvious. Fixes: 33f8b9ecdb15 ("net_sched: move tp->root allocation into fw_init()") Reported-by: Akshat Kakkar Cc: Jamal Hadi Salim Signed-off-by: Cong Wang Acked-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- net/sched/cls_fw.c | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c index 715e01e5910a..f23a3b68bba6 100644 --- a/net/sched/cls_fw.c +++ b/net/sched/cls_fw.c @@ -33,7 +33,6 @@ struct fw_head { u32 mask; - bool mask_set; struct fw_filter __rcu *ht[HTSIZE]; struct rcu_head rcu; }; @@ -84,7 +83,7 @@ static int fw_classify(struct sk_buff *skb, const struct tcf_proto *tp, } } } else { - /* old method */ + /* Old method: classify the packet using its skb mark. */ if (id && (TC_H_MAJ(id) == 0 || !(TC_H_MAJ(id ^ tp->q->handle)))) { res->classid = id; @@ -114,14 +113,9 @@ static unsigned long fw_get(struct tcf_proto *tp, u32 handle) static int fw_init(struct tcf_proto *tp) { - struct fw_head *head; - - head = kzalloc(sizeof(struct fw_head), GFP_KERNEL); - if (head == NULL) - return -ENOBUFS; - - head->mask_set = false; - rcu_assign_pointer(tp->root, head); + /* We don't allocate fw_head here, because in the old method + * we don't need it at all. + */ return 0; } @@ -252,7 +246,7 @@ static int fw_change(struct net *net, struct sk_buff *in_skb, int err; if (!opt) - return handle ? -EINVAL : 0; + return handle ? -EINVAL : 0; /* Succeed if it is old method. */ err = nla_parse_nested(tb, TCA_FW_MAX, opt, fw_policy); if (err < 0) @@ -302,11 +296,17 @@ static int fw_change(struct net *net, struct sk_buff *in_skb, if (!handle) return -EINVAL; - if (!head->mask_set) { - head->mask = 0xFFFFFFFF; + if (!head) { + u32 mask = 0xFFFFFFFF; if (tb[TCA_FW_MASK]) - head->mask = nla_get_u32(tb[TCA_FW_MASK]); - head->mask_set = true; + mask = nla_get_u32(tb[TCA_FW_MASK]); + + head = kzalloc(sizeof(*head), GFP_KERNEL); + if (!head) + return -ENOBUFS; + head->mask = mask; + + rcu_assign_pointer(tp->root, head); } f = kzalloc(sizeof(struct fw_filter), GFP_KERNEL); From d682d2bdc30650a5c7ce9908ab83ab674b658744 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 22 Sep 2015 17:04:58 -0700 Subject: [PATCH 423/647] bnx2x: byte swap rss_key to comply to Toeplitz specs After a good amount of debugging, I found bnx2x was byte swaping the 40 bytes of rss_key. If we byte swap the key, then bnx2x generates hashes matching MSDN specs as documented in (Verifying the RSS Hash Calculation) https://msdn.microsoft.com/en-us/library/windows/hardware/ff571021% 28v=vs.85%29.aspx It is mostly a non issue, unless we want to mix different NIC in a host, and want consistent hashing among all of them, ie if they all use the boot time generated rss key, or if some application is choosing specific tuple(s) so that incoming traffic lands into known rx queue(s). Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c index c9bd7f16018e..ff702a707a91 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c @@ -4319,8 +4319,16 @@ static int bnx2x_setup_rss(struct bnx2x *bp, /* RSS keys */ if (test_bit(BNX2X_RSS_SET_SRCH, &p->rss_flags)) { - memcpy(&data->rss_key[0], &p->rss_key[0], - sizeof(data->rss_key)); + u8 *dst = (u8 *)(data->rss_key) + sizeof(data->rss_key); + const u8 *src = (const u8 *)p->rss_key; + int i; + + /* Apparently, bnx2x reads this array in reverse order + * We need to byte swap rss_key to comply with Toeplitz specs. + */ + for (i = 0; i < sizeof(data->rss_key); i++) + *--dst = *src++; + caps |= ETH_RSS_UPDATE_RAMROD_DATA_UPDATE_RSS_KEY; } From 38ea72bdb65df2f40ec77b2c9d1413e7f5e34465 Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Fri, 18 Sep 2015 15:08:54 -0600 Subject: [PATCH 424/647] PCI/MSI: Fix MSI IRQ domains for VFs on virtual buses SR-IOV creates a virtual bus where bus->self is NULL. When we add VFs and scan for an MSI domain, pci_set_bus_msi_domain() dereferences bus->self, which causes a kernel NULL pointer dereference oops. Scan up to the parent bus until we find a real bridge where we can get the MSI domain. [bhelgaas: changelog] Fixes: 44aa0c657e3e ("PCI/MSI: Add hooks to populate the msi_domain field") Tested-by: Joerg Roedel Signed-off-by: Alex Williamson Signed-off-by: Bjorn Helgaas Acked-by: Marc Zyngier --- drivers/pci/probe.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index c8cc0e62a7b9..8361d27e5eca 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -676,15 +676,20 @@ static struct irq_domain *pci_host_bridge_msi_domain(struct pci_bus *bus) static void pci_set_bus_msi_domain(struct pci_bus *bus) { struct irq_domain *d; + struct pci_bus *b; /* - * Either bus is the root, and we must obtain it from the - * firmware, or we inherit it from the bridge device. + * The bus can be a root bus, a subordinate bus, or a virtual bus + * created by an SR-IOV device. Walk up to the first bridge device + * found or derive the domain from the host bridge. */ - if (pci_is_root_bus(bus)) - d = pci_host_bridge_msi_domain(bus); - else - d = dev_get_msi_domain(&bus->self->dev); + for (b = bus, d = NULL; !d && !pci_is_root_bus(b); b = b->parent) { + if (b->self) + d = dev_get_msi_domain(&b->self->dev); + } + + if (!d) + d = pci_host_bridge_msi_domain(b); dev_set_msi_domain(&bus->dev, d); } From 9d9240756e63dd87d6cbf5da8b98ceb8f8192b55 Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Tue, 15 Sep 2015 11:17:21 -0600 Subject: [PATCH 425/647] PCI: Fix devfn for VPD access through function 0 Commit 932c435caba8 ("PCI: Add dev_flags bit to access VPD through function 0") passes PCI_SLOT(devfn) for the devfn parameter of pci_get_slot(). Generally this works because we're fairly well guaranteed that a PCIe device is at slot address 0, but for the general case, including conventional PCI, it's incorrect. We need to get the slot and then convert it back into a devfn. Fixes: 932c435caba8 ("PCI: Add dev_flags bit to access VPD through function 0") Signed-off-by: Alex Williamson Signed-off-by: Bjorn Helgaas Acked-by: Myron Stowe Acked-by: Mark Rustad CC: stable@vger.kernel.org --- drivers/pci/access.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/pci/access.c b/drivers/pci/access.c index 769f7e35f1a2..5a5f0a7ba801 100644 --- a/drivers/pci/access.c +++ b/drivers/pci/access.c @@ -442,7 +442,8 @@ static const struct pci_vpd_ops pci_vpd_pci22_ops = { static ssize_t pci_vpd_f0_read(struct pci_dev *dev, loff_t pos, size_t count, void *arg) { - struct pci_dev *tdev = pci_get_slot(dev->bus, PCI_SLOT(dev->devfn)); + struct pci_dev *tdev = pci_get_slot(dev->bus, + PCI_DEVFN(PCI_SLOT(dev->devfn), 0)); ssize_t ret; if (!tdev) @@ -456,7 +457,8 @@ static ssize_t pci_vpd_f0_read(struct pci_dev *dev, loff_t pos, size_t count, static ssize_t pci_vpd_f0_write(struct pci_dev *dev, loff_t pos, size_t count, const void *arg) { - struct pci_dev *tdev = pci_get_slot(dev->bus, PCI_SLOT(dev->devfn)); + struct pci_dev *tdev = pci_get_slot(dev->bus, + PCI_DEVFN(PCI_SLOT(dev->devfn), 0)); ssize_t ret; if (!tdev) @@ -475,7 +477,8 @@ static const struct pci_vpd_ops pci_vpd_f0_ops = { static int pci_vpd_f0_dev_check(struct pci_dev *dev) { - struct pci_dev *tdev = pci_get_slot(dev->bus, PCI_SLOT(dev->devfn)); + struct pci_dev *tdev = pci_get_slot(dev->bus, + PCI_DEVFN(PCI_SLOT(dev->devfn), 0)); int ret = 0; if (!tdev) From da2d03ea27f6ed9d2005a67b20dd021ddacf1e4d Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Tue, 15 Sep 2015 22:24:46 -0600 Subject: [PATCH 426/647] PCI: Use function 0 VPD for identical functions, regular VPD for others 932c435caba8 ("PCI: Add dev_flags bit to access VPD through function 0") added PCI_DEV_FLAGS_VPD_REF_F0. Previously, we set the flag on every non-zero function of quirked devices. If a function turned out to be different from function 0, i.e., it had a different class, vendor ID, or device ID, the flag remained set but we didn't make VPD accessible at all. Flip this around so we only set PCI_DEV_FLAGS_VPD_REF_F0 for functions that are identical to function 0, and allow regular VPD access for any other functions. [bhelgaas: changelog, stable tag] Fixes: 932c435caba8 ("PCI: Add dev_flags bit to access VPD through function 0") Signed-off-by: Alex Williamson Signed-off-by: Bjorn Helgaas Acked-by: Myron Stowe Acked-by: Mark Rustad CC: stable@vger.kernel.org --- drivers/pci/access.c | 22 ---------------------- drivers/pci/quirks.c | 20 ++++++++++++++++++-- 2 files changed, 18 insertions(+), 24 deletions(-) diff --git a/drivers/pci/access.c b/drivers/pci/access.c index 5a5f0a7ba801..59ac36fe7c42 100644 --- a/drivers/pci/access.c +++ b/drivers/pci/access.c @@ -475,23 +475,6 @@ static const struct pci_vpd_ops pci_vpd_f0_ops = { .release = pci_vpd_pci22_release, }; -static int pci_vpd_f0_dev_check(struct pci_dev *dev) -{ - struct pci_dev *tdev = pci_get_slot(dev->bus, - PCI_DEVFN(PCI_SLOT(dev->devfn), 0)); - int ret = 0; - - if (!tdev) - return -ENODEV; - if (!tdev->vpd || !tdev->multifunction || - dev->class != tdev->class || dev->vendor != tdev->vendor || - dev->device != tdev->device) - ret = -ENODEV; - - pci_dev_put(tdev); - return ret; -} - int pci_vpd_pci22_init(struct pci_dev *dev) { struct pci_vpd_pci22 *vpd; @@ -500,12 +483,7 @@ int pci_vpd_pci22_init(struct pci_dev *dev) cap = pci_find_capability(dev, PCI_CAP_ID_VPD); if (!cap) return -ENODEV; - if (dev->dev_flags & PCI_DEV_FLAGS_VPD_REF_F0) { - int ret = pci_vpd_f0_dev_check(dev); - if (ret) - return ret; - } vpd = kzalloc(sizeof(*vpd), GFP_ATOMIC); if (!vpd) return -ENOMEM; diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 6a30252cd79f..b03373fd05ca 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1907,11 +1907,27 @@ static void quirk_netmos(struct pci_dev *dev) DECLARE_PCI_FIXUP_CLASS_HEADER(PCI_VENDOR_ID_NETMOS, PCI_ANY_ID, PCI_CLASS_COMMUNICATION_SERIAL, 8, quirk_netmos); +/* + * Quirk non-zero PCI functions to route VPD access through function 0 for + * devices that share VPD resources between functions. The functions are + * expected to be identical devices. + */ static void quirk_f0_vpd_link(struct pci_dev *dev) { - if (!dev->multifunction || !PCI_FUNC(dev->devfn)) + struct pci_dev *f0; + + if (!PCI_FUNC(dev->devfn)) return; - dev->dev_flags |= PCI_DEV_FLAGS_VPD_REF_F0; + + f0 = pci_get_slot(dev->bus, PCI_DEVFN(PCI_SLOT(dev->devfn), 0)); + if (!f0) + return; + + if (f0->vpd && dev->class == f0->class && + dev->vendor == f0->vendor && dev->device == f0->device) + dev->dev_flags |= PCI_DEV_FLAGS_VPD_REF_F0; + + pci_dev_put(f0); } DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_INTEL, PCI_ANY_ID, PCI_CLASS_NETWORK_ETHERNET, 8, quirk_f0_vpd_link); From de24c18c0faca5ebd618e1cb87f5489745e40475 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sat, 12 Sep 2015 02:06:09 +0300 Subject: [PATCH 427/647] PCI: rcar: Add R8A7794 support Add Renesas R8A7794 SoC support to the Renesas R-Car gen2 PCI driver. Signed-off-by: Sergei Shtylyov Signed-off-by: Bjorn Helgaas Acked-by: Simon Horman --- Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt | 3 ++- drivers/pci/host/pci-rcar-gen2.c | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt b/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt index d8ef5bf50f11..7fab84b33531 100644 --- a/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt +++ b/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt @@ -7,7 +7,8 @@ OHCI and EHCI controllers. Required properties: - compatible: "renesas,pci-r8a7790" for the R8A7790 SoC; - "renesas,pci-r8a7791" for the R8A7791 SoC. + "renesas,pci-r8a7791" for the R8A7791 SoC; + "renesas,pci-r8a7794" for the R8A7794 SoC. - reg: A list of physical regions to access the device: the first is the operational registers for the OHCI/EHCI controllers and the second is for the bridge configuration and control registers. diff --git a/drivers/pci/host/pci-rcar-gen2.c b/drivers/pci/host/pci-rcar-gen2.c index 367e28fa7564..c4f64bfee551 100644 --- a/drivers/pci/host/pci-rcar-gen2.c +++ b/drivers/pci/host/pci-rcar-gen2.c @@ -362,6 +362,7 @@ static int rcar_pci_probe(struct platform_device *pdev) static struct of_device_id rcar_pci_of_match[] = { { .compatible = "renesas,pci-r8a7790", }, { .compatible = "renesas,pci-r8a7791", }, + { .compatible = "renesas,pci-r8a7794", }, { }, }; From 41fc014332d91ee90c32840bf161f9685b7fbf2b Mon Sep 17 00:00:00 2001 From: Wilson Kok Date: Tue, 22 Sep 2015 21:40:22 -0700 Subject: [PATCH 428/647] fib_rules: fix fib rule dumps across multiple skbs dump_rules returns skb length and not error. But when family == AF_UNSPEC, the caller of dump_rules assumes that it returns an error. Hence, when family == AF_UNSPEC, we continue trying to dump on -EMSGSIZE errors resulting in incorrect dump idx carried between skbs belonging to the same dump. This results in fib rule dump always only dumping rules that fit into the first skb. This patch fixes dump_rules to return error so that we exit correctly and idx is correctly maintained between skbs that are part of the same dump. Signed-off-by: Wilson Kok Signed-off-by: Roopa Prabhu Signed-off-by: David S. Miller --- net/core/fib_rules.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index bf77e3639ce0..365de66436ac 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -631,15 +631,17 @@ static int dump_rules(struct sk_buff *skb, struct netlink_callback *cb, { int idx = 0; struct fib_rule *rule; + int err = 0; rcu_read_lock(); list_for_each_entry_rcu(rule, &ops->rules_list, list) { if (idx < cb->args[1]) goto skip; - if (fib_nl_fill_rule(skb, rule, NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, RTM_NEWRULE, - NLM_F_MULTI, ops) < 0) + err = fib_nl_fill_rule(skb, rule, NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, RTM_NEWRULE, + NLM_F_MULTI, ops); + if (err) break; skip: idx++; @@ -648,7 +650,7 @@ skip: cb->args[1] = idx; rules_ops_put(ops); - return skb->len; + return err; } static int fib_nl_dumprule(struct sk_buff *skb, struct netlink_callback *cb) @@ -664,7 +666,9 @@ static int fib_nl_dumprule(struct sk_buff *skb, struct netlink_callback *cb) if (ops == NULL) return -EAFNOSUPPORT; - return dump_rules(skb, cb, ops); + dump_rules(skb, cb, ops); + + return skb->len; } rcu_read_lock(); From a46496ce38eeb401344d5623c1960dbf2f1769be Mon Sep 17 00:00:00 2001 From: Matt Bennett Date: Wed, 23 Sep 2015 16:58:31 +1200 Subject: [PATCH 429/647] ip6_gre: Reduce log level in ip6gre_err() to debug Currently error log messages in ip6gre_err are printed at 'warn' level. This is different to most other tunnel types which don't print any messages. These log messages don't provide any information that couldn't be deduced with networking tools. Also it can be annoying to have one end of the tunnel go down and have the logs fill with pointless messages such as "Path to destination invalid or inactive!". This patch reduces the log level of these messages to 'dbg' level to bring the visible behaviour into line with other tunnel types. Signed-off-by: Matt Bennett Signed-off-by: David S. Miller --- net/ipv6/ip6_gre.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 646512488c28..3c7b9310b33f 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -404,13 +404,13 @@ static void ip6gre_err(struct sk_buff *skb, struct inet6_skb_parm *opt, struct ipv6_tlv_tnl_enc_lim *tel; __u32 mtu; case ICMPV6_DEST_UNREACH: - net_warn_ratelimited("%s: Path to destination invalid or inactive!\n", - t->parms.name); + net_dbg_ratelimited("%s: Path to destination invalid or inactive!\n", + t->parms.name); break; case ICMPV6_TIME_EXCEED: if (code == ICMPV6_EXC_HOPLIMIT) { - net_warn_ratelimited("%s: Too small hop limit or routing loop in tunnel!\n", - t->parms.name); + net_dbg_ratelimited("%s: Too small hop limit or routing loop in tunnel!\n", + t->parms.name); } break; case ICMPV6_PARAMPROB: @@ -421,12 +421,12 @@ static void ip6gre_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (teli && teli == be32_to_cpu(info) - 2) { tel = (struct ipv6_tlv_tnl_enc_lim *) &skb->data[teli]; if (tel->encap_limit == 0) { - net_warn_ratelimited("%s: Too small encapsulation limit or routing loop in tunnel!\n", - t->parms.name); + net_dbg_ratelimited("%s: Too small encapsulation limit or routing loop in tunnel!\n", + t->parms.name); } } else { - net_warn_ratelimited("%s: Recipient unable to parse tunneled packet!\n", - t->parms.name); + net_dbg_ratelimited("%s: Recipient unable to parse tunneled packet!\n", + t->parms.name); } break; case ICMPV6_PKT_TOOBIG: From e7ae65ced7dd71aa3dc29bda7a94ac82d9ea7751 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Mon, 21 Sep 2015 14:57:25 +0200 Subject: [PATCH 430/647] gpio: mention in DT binding doc that -gpio is deprecated The gpiolib supports parsing DT properties of the form -gpio but it was only added for compatibility with older DT bindings that got it wrong and should not be used in newer bindings. The commit that added support for this was: dd34c37aa3e8 ("gpio: of: Allow -gpio suffix for property names") but didn't update the documentation to explain this so it's been a source of confusion. So let's make this clear in the GPIO DT binding doc. Signed-off-by: Javier Martinez Canillas Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/gpio/gpio.txt | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/gpio/gpio.txt b/Documentation/devicetree/bindings/gpio/gpio.txt index 5788d5cf1252..82d40e2505f6 100644 --- a/Documentation/devicetree/bindings/gpio/gpio.txt +++ b/Documentation/devicetree/bindings/gpio/gpio.txt @@ -16,7 +16,9 @@ properties, each containing a 'gpio-list': GPIO properties should be named "[-]gpios", with being the purpose of this GPIO for the device. While a non-existent is considered valid for compatibility reasons (resolving to the "gpios" property), it is not allowed -for new bindings. +for new bindings. Also, GPIO properties named "[-]gpio" are valid and old +bindings use it, but are only supported for compatibility reasons and should not +be used for newer bindings since it has been deprecated. GPIO properties can contain one or more GPIO phandles, but only in exceptional cases should they contain more than one. If your device uses several GPIOs with From a13f18f59d2646754cda3662a9e215ff43e7a7d5 Mon Sep 17 00:00:00 2001 From: Lorenzo Pieralisi Date: Thu, 24 Sep 2015 15:53:56 +0100 Subject: [PATCH 431/647] Documentation: arm: Fix typo in the idle-states bindings examples The idle-states bindings mandate that the entry-method string in the idle-states node must be "psci" for ARM v8 64-bit systems, but the examples in the bindings report a wrong entry-method string. Owing to this typo, some dts in the kernel wrongly defined the entry-method property, since they likely cut and pasted the example definition without paying attention to the bindings definitions. This patch fixes the typo in the DT idle states bindings examples and respective dts in the kernel so that the bindings and related dts files are made compliant. Signed-off-by: Lorenzo Pieralisi Cc: Howard Chen Cc: Rob Herring Cc: Mark Rutland Cc: Heiko Stuebner Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/arm/idle-states.txt | 2 +- arch/arm64/boot/dts/mediatek/mt8173.dtsi | 2 +- arch/arm64/boot/dts/rockchip/rk3368.dtsi | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/arm/idle-states.txt b/Documentation/devicetree/bindings/arm/idle-states.txt index a8274eabae2e..b8e41c148a3c 100644 --- a/Documentation/devicetree/bindings/arm/idle-states.txt +++ b/Documentation/devicetree/bindings/arm/idle-states.txt @@ -497,7 +497,7 @@ cpus { }; idle-states { - entry-method = "arm,psci"; + entry-method = "psci"; CPU_RETENTION_0_0: cpu-retention-0-0 { compatible = "arm,idle-state"; diff --git a/arch/arm64/boot/dts/mediatek/mt8173.dtsi b/arch/arm64/boot/dts/mediatek/mt8173.dtsi index d18ee4259ee5..06a15644be38 100644 --- a/arch/arm64/boot/dts/mediatek/mt8173.dtsi +++ b/arch/arm64/boot/dts/mediatek/mt8173.dtsi @@ -81,7 +81,7 @@ }; idle-states { - entry-method = "arm,psci"; + entry-method = "psci"; CPU_SLEEP_0: cpu-sleep-0 { compatible = "arm,idle-state"; diff --git a/arch/arm64/boot/dts/rockchip/rk3368.dtsi b/arch/arm64/boot/dts/rockchip/rk3368.dtsi index a712bea3bf2c..cc093a482aa4 100644 --- a/arch/arm64/boot/dts/rockchip/rk3368.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3368.dtsi @@ -106,7 +106,7 @@ }; idle-states { - entry-method = "arm,psci"; + entry-method = "psci"; cpu_sleep: cpu-sleep-0 { compatible = "arm,idle-state"; From 17a10c9215b39a5ea8faeb241759c1aee6553919 Mon Sep 17 00:00:00 2001 From: Matt Bennett Date: Fri, 25 Sep 2015 11:01:47 +1200 Subject: [PATCH 432/647] ip6_tunnel: Reduce log level in ip6_tnl_err() to debug Currently error log messages in ip6_tnl_err are printed at 'warn' level. This is different to other tunnel types which don't print any messages. These log messages don't provide any information that couldn't be deduced with networking tools. Also it can be annoying to have one end of the tunnel go down and have the logs fill with pointless messages such as "Path to destination invalid or inactive!". This patch reduces the log level of these messages to 'dbg' level to bring the visible behaviour into line with other tunnel types. Signed-off-by: Matt Bennett Signed-off-by: David S. Miller --- net/ipv6/ip6_tunnel.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 983f0d20f96d..eabffbb89795 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -569,14 +569,14 @@ ip6_tnl_err(struct sk_buff *skb, __u8 ipproto, struct inet6_skb_parm *opt, struct ipv6_tlv_tnl_enc_lim *tel; __u32 mtu; case ICMPV6_DEST_UNREACH: - net_warn_ratelimited("%s: Path to destination invalid or inactive!\n", - t->parms.name); + net_dbg_ratelimited("%s: Path to destination invalid or inactive!\n", + t->parms.name); rel_msg = 1; break; case ICMPV6_TIME_EXCEED: if ((*code) == ICMPV6_EXC_HOPLIMIT) { - net_warn_ratelimited("%s: Too small hop limit or routing loop in tunnel!\n", - t->parms.name); + net_dbg_ratelimited("%s: Too small hop limit or routing loop in tunnel!\n", + t->parms.name); rel_msg = 1; } break; @@ -588,13 +588,13 @@ ip6_tnl_err(struct sk_buff *skb, __u8 ipproto, struct inet6_skb_parm *opt, if (teli && teli == *info - 2) { tel = (struct ipv6_tlv_tnl_enc_lim *) &skb->data[teli]; if (tel->encap_limit == 0) { - net_warn_ratelimited("%s: Too small encapsulation limit or routing loop in tunnel!\n", - t->parms.name); + net_dbg_ratelimited("%s: Too small encapsulation limit or routing loop in tunnel!\n", + t->parms.name); rel_msg = 1; } } else { - net_warn_ratelimited("%s: Recipient unable to parse tunneled packet!\n", - t->parms.name); + net_dbg_ratelimited("%s: Recipient unable to parse tunneled packet!\n", + t->parms.name); } break; case ICMPV6_PKT_TOOBIG: From 9a5e3f27d1b8ca349b79e8b5fe1874eeeedb6f45 Mon Sep 17 00:00:00 2001 From: Kishon Vijay Abraham I Date: Fri, 4 Sep 2015 17:38:24 +0530 Subject: [PATCH 433/647] ARM: dts: fix omap2+ address translation for pbias "ARM: dts: : add minimal l4 bus layout with control module support" moved pbias_regulator dt node from being a child node of ocp to be the child node of 'syscon'. Since 'syscon' doesn't have the 'ranges' property, address translation fails while trying to convert the address to resource. Fix it here by populating 'ranges' property in syscon dt node. Fixes: 72b10ac00eb1 ("ARM: dts: omap24xx: add minimal l4 bus layout with control module support") Fixes: 7415b0b4c645 ("ARM: dts: omap4: add minimal l4 bus layout with control module support") Fixes: ed8509edddeb ("ARM: dts: omap5: add minimal l4 bus layout with control module support") Fixes: d919501feffa ("ARM: dts: dra7: add minimal l4 bus layout with control module support") Signed-off-by: Kishon Vijay Abraham I [tony@atomide.com: fixed omap3 pbias to work] Signed-off-by: Tony Lindgren --- arch/arm/boot/dts/dra7.dtsi | 1 + arch/arm/boot/dts/omap2430.dtsi | 1 + arch/arm/boot/dts/omap3.dtsi | 25 +++++++++++++------------ arch/arm/boot/dts/omap4.dtsi | 1 + arch/arm/boot/dts/omap5.dtsi | 1 + 5 files changed, 17 insertions(+), 12 deletions(-) diff --git a/arch/arm/boot/dts/dra7.dtsi b/arch/arm/boot/dts/dra7.dtsi index fed95a4547ee..e289c706d27d 100644 --- a/arch/arm/boot/dts/dra7.dtsi +++ b/arch/arm/boot/dts/dra7.dtsi @@ -120,6 +120,7 @@ reg = <0x0 0x1400>; #address-cells = <1>; #size-cells = <1>; + ranges = <0 0x0 0x1400>; pbias_regulator: pbias_regulator { compatible = "ti,pbias-dra7", "ti,pbias-omap"; diff --git a/arch/arm/boot/dts/omap2430.dtsi b/arch/arm/boot/dts/omap2430.dtsi index 3961a6f170e1..798dda072b2a 100644 --- a/arch/arm/boot/dts/omap2430.dtsi +++ b/arch/arm/boot/dts/omap2430.dtsi @@ -56,6 +56,7 @@ reg = <0x270 0x240>; #address-cells = <1>; #size-cells = <1>; + ranges = <0 0x270 0x240>; scm_clocks: clocks { #address-cells = <1>; diff --git a/arch/arm/boot/dts/omap3.dtsi b/arch/arm/boot/dts/omap3.dtsi index 9af9ae190299..8a2b25332b8c 100644 --- a/arch/arm/boot/dts/omap3.dtsi +++ b/arch/arm/boot/dts/omap3.dtsi @@ -113,10 +113,22 @@ }; scm_conf: scm_conf@270 { - compatible = "syscon"; + compatible = "syscon", "simple-bus"; reg = <0x270 0x330>; #address-cells = <1>; #size-cells = <1>; + ranges = <0 0x270 0x330>; + + pbias_regulator: pbias_regulator { + compatible = "ti,pbias-omap3", "ti,pbias-omap"; + reg = <0x2b0 0x4>; + syscon = <&scm_conf>; + pbias_mmc_reg: pbias_mmc_omap2430 { + regulator-name = "pbias_mmc_omap2430"; + regulator-min-microvolt = <1800000>; + regulator-max-microvolt = <3000000>; + }; + }; scm_clocks: clocks { #address-cells = <1>; @@ -202,17 +214,6 @@ dma-requests = <96>; }; - pbias_regulator: pbias_regulator { - compatible = "ti,pbias-omap3", "ti,pbias-omap"; - reg = <0x2b0 0x4>; - syscon = <&scm_conf>; - pbias_mmc_reg: pbias_mmc_omap2430 { - regulator-name = "pbias_mmc_omap2430"; - regulator-min-microvolt = <1800000>; - regulator-max-microvolt = <3000000>; - }; - }; - gpio1: gpio@48310000 { compatible = "ti,omap3-gpio"; reg = <0x48310000 0x200>; diff --git a/arch/arm/boot/dts/omap4.dtsi b/arch/arm/boot/dts/omap4.dtsi index 5aad7f37df8b..5a206c100ce2 100644 --- a/arch/arm/boot/dts/omap4.dtsi +++ b/arch/arm/boot/dts/omap4.dtsi @@ -196,6 +196,7 @@ reg = <0x5a0 0x170>; #address-cells = <1>; #size-cells = <1>; + ranges = <0 0x5a0 0x170>; pbias_regulator: pbias_regulator { compatible = "ti,pbias-omap4", "ti,pbias-omap"; diff --git a/arch/arm/boot/dts/omap5.dtsi b/arch/arm/boot/dts/omap5.dtsi index 8d5f9d26110e..4c04389dab32 100644 --- a/arch/arm/boot/dts/omap5.dtsi +++ b/arch/arm/boot/dts/omap5.dtsi @@ -185,6 +185,7 @@ reg = <0x5a0 0xec>; #address-cells = <1>; #size-cells = <1>; + ranges = <0 0x5a0 0xec>; pbias_regulator: pbias_regulator { compatible = "ti,pbias-omap5", "ti,pbias-omap"; From 79a313f5a587eae14a0f96df5d819c2fe266e3e9 Mon Sep 17 00:00:00 2001 From: Benjamin Gaignard Date: Wed, 23 Sep 2015 21:48:03 +0200 Subject: [PATCH 434/647] ARM: sti: dt: adapt DT to fix probe/bind issues in DRM driver STI drm drivers probe and bind using component framework was incorrect. In addition to drivers fix DT update is needed to make all sub-components become childs of sti-display-subsystem. Signed-off-by: Benjamin Gaignard Signed-off-by: Maxime Coquelin Signed-off-by: Olof Johansson --- arch/arm/boot/dts/stih407.dtsi | 74 +++++++++++++++++----------------- arch/arm/boot/dts/stih410.dtsi | 74 +++++++++++++++++----------------- 2 files changed, 72 insertions(+), 76 deletions(-) diff --git a/arch/arm/boot/dts/stih407.dtsi b/arch/arm/boot/dts/stih407.dtsi index 3efa3b2ebe90..6b914e4bb099 100644 --- a/arch/arm/boot/dts/stih407.dtsi +++ b/arch/arm/boot/dts/stih407.dtsi @@ -103,48 +103,46 @@ <&clk_s_d0_quadfs 0>, <&clk_s_d2_quadfs 0>, <&clk_s_d2_quadfs 0>; - ranges; + }; - sti-hdmi@8d04000 { - compatible = "st,stih407-hdmi"; - reg = <0x8d04000 0x1000>; - reg-names = "hdmi-reg"; - interrupts = ; - interrupt-names = "irq"; - clock-names = "pix", - "tmds", - "phy", - "audio", - "main_parent", - "aux_parent"; + sti-hdmi@8d04000 { + compatible = "st,stih407-hdmi"; + reg = <0x8d04000 0x1000>; + reg-names = "hdmi-reg"; + interrupts = ; + interrupt-names = "irq"; + clock-names = "pix", + "tmds", + "phy", + "audio", + "main_parent", + "aux_parent"; - clocks = <&clk_s_d2_flexgen CLK_PIX_HDMI>, - <&clk_s_d2_flexgen CLK_TMDS_HDMI>, - <&clk_s_d2_flexgen CLK_REF_HDMIPHY>, - <&clk_s_d0_flexgen CLK_PCM_0>, - <&clk_s_d2_quadfs 0>, - <&clk_s_d2_quadfs 1>; + clocks = <&clk_s_d2_flexgen CLK_PIX_HDMI>, + <&clk_s_d2_flexgen CLK_TMDS_HDMI>, + <&clk_s_d2_flexgen CLK_REF_HDMIPHY>, + <&clk_s_d0_flexgen CLK_PCM_0>, + <&clk_s_d2_quadfs 0>, + <&clk_s_d2_quadfs 1>; - hdmi,hpd-gpio = <&pio5 3>; - reset-names = "hdmi"; - resets = <&softreset STIH407_HDMI_TX_PHY_SOFTRESET>; - ddc = <&hdmiddc>; + hdmi,hpd-gpio = <&pio5 3>; + reset-names = "hdmi"; + resets = <&softreset STIH407_HDMI_TX_PHY_SOFTRESET>; + ddc = <&hdmiddc>; + }; - }; - - sti-hda@8d02000 { - compatible = "st,stih407-hda"; - reg = <0x8d02000 0x400>, <0x92b0120 0x4>; - reg-names = "hda-reg", "video-dacs-ctrl"; - clock-names = "pix", - "hddac", - "main_parent", - "aux_parent"; - clocks = <&clk_s_d2_flexgen CLK_PIX_HDDAC>, - <&clk_s_d2_flexgen CLK_HDDAC>, - <&clk_s_d2_quadfs 0>, - <&clk_s_d2_quadfs 1>; - }; + sti-hda@8d02000 { + compatible = "st,stih407-hda"; + reg = <0x8d02000 0x400>, <0x92b0120 0x4>; + reg-names = "hda-reg", "video-dacs-ctrl"; + clock-names = "pix", + "hddac", + "main_parent", + "aux_parent"; + clocks = <&clk_s_d2_flexgen CLK_PIX_HDDAC>, + <&clk_s_d2_flexgen CLK_HDDAC>, + <&clk_s_d2_quadfs 0>, + <&clk_s_d2_quadfs 1>; }; }; }; diff --git a/arch/arm/boot/dts/stih410.dtsi b/arch/arm/boot/dts/stih410.dtsi index 6f40bc99c22f..8c6e61a27234 100644 --- a/arch/arm/boot/dts/stih410.dtsi +++ b/arch/arm/boot/dts/stih410.dtsi @@ -178,48 +178,46 @@ <&clk_s_d0_quadfs 0>, <&clk_s_d2_quadfs 0>, <&clk_s_d2_quadfs 0>; - ranges; + }; - sti-hdmi@8d04000 { - compatible = "st,stih407-hdmi"; - reg = <0x8d04000 0x1000>; - reg-names = "hdmi-reg"; - interrupts = ; - interrupt-names = "irq"; - clock-names = "pix", - "tmds", - "phy", - "audio", - "main_parent", - "aux_parent"; + sti-hdmi@8d04000 { + compatible = "st,stih407-hdmi"; + reg = <0x8d04000 0x1000>; + reg-names = "hdmi-reg"; + interrupts = ; + interrupt-names = "irq"; + clock-names = "pix", + "tmds", + "phy", + "audio", + "main_parent", + "aux_parent"; - clocks = <&clk_s_d2_flexgen CLK_PIX_HDMI>, - <&clk_s_d2_flexgen CLK_TMDS_HDMI>, - <&clk_s_d2_flexgen CLK_REF_HDMIPHY>, - <&clk_s_d0_flexgen CLK_PCM_0>, - <&clk_s_d2_quadfs 0>, - <&clk_s_d2_quadfs 1>; + clocks = <&clk_s_d2_flexgen CLK_PIX_HDMI>, + <&clk_s_d2_flexgen CLK_TMDS_HDMI>, + <&clk_s_d2_flexgen CLK_REF_HDMIPHY>, + <&clk_s_d0_flexgen CLK_PCM_0>, + <&clk_s_d2_quadfs 0>, + <&clk_s_d2_quadfs 1>; - hdmi,hpd-gpio = <&pio5 3>; - reset-names = "hdmi"; - resets = <&softreset STIH407_HDMI_TX_PHY_SOFTRESET>; - ddc = <&hdmiddc>; + hdmi,hpd-gpio = <&pio5 3>; + reset-names = "hdmi"; + resets = <&softreset STIH407_HDMI_TX_PHY_SOFTRESET>; + ddc = <&hdmiddc>; + }; - }; - - sti-hda@8d02000 { - compatible = "st,stih407-hda"; - reg = <0x8d02000 0x400>, <0x92b0120 0x4>; - reg-names = "hda-reg", "video-dacs-ctrl"; - clock-names = "pix", - "hddac", - "main_parent", - "aux_parent"; - clocks = <&clk_s_d2_flexgen CLK_PIX_HDDAC>, - <&clk_s_d2_flexgen CLK_HDDAC>, - <&clk_s_d2_quadfs 0>, - <&clk_s_d2_quadfs 1>; - }; + sti-hda@8d02000 { + compatible = "st,stih407-hda"; + reg = <0x8d02000 0x400>, <0x92b0120 0x4>; + reg-names = "hda-reg", "video-dacs-ctrl"; + clock-names = "pix", + "hddac", + "main_parent", + "aux_parent"; + clocks = <&clk_s_d2_flexgen CLK_PIX_HDDAC>, + <&clk_s_d2_flexgen CLK_HDDAC>, + <&clk_s_d2_quadfs 0>, + <&clk_s_d2_quadfs 1>; }; }; From 680168a58a9315e1301f4ebb062244470d4919b0 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Tue, 22 Sep 2015 09:35:31 -0700 Subject: [PATCH 435/647] PM / OPP: of_property_count_u32_elems() can return errors of_property_count_u32_elems() will never return 0, but a -ve error value of a positive count. And so the current !count check is wrong. Also, a missing "opp-microvolt" property isn't a problem and so we need to do of_find_property() separately to confirm that. Fixes: 274659029c9d (PM / OPP: Add support to parse "operating-points-v2" bindings) Signed-off-by: Viresh Kumar Reviewed-by: Stephen Boyd Signed-off-by: Rafael J. Wysocki --- drivers/base/power/opp.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/base/power/opp.c b/drivers/base/power/opp.c index 28cd75c535b0..1194669c2bb5 100644 --- a/drivers/base/power/opp.c +++ b/drivers/base/power/opp.c @@ -892,10 +892,17 @@ static int opp_get_microvolt(struct dev_pm_opp *opp, struct device *dev) u32 microvolt[3] = {0}; int count, ret; - count = of_property_count_u32_elems(opp->np, "opp-microvolt"); - if (!count) + /* Missing property isn't a problem, but an invalid entry is */ + if (!of_find_property(opp->np, "opp-microvolt", NULL)) return 0; + count = of_property_count_u32_elems(opp->np, "opp-microvolt"); + if (count < 0) { + dev_err(dev, "%s: Invalid opp-microvolt property (%d)\n", + __func__, count); + return count; + } + /* There can be one or three elements here */ if (count != 1 && count != 3) { dev_err(dev, "%s: Invalid number of elements in opp-microvolt property (%d)\n", From a8360062ccfb4b891d3013d0e55826c8bcb02bfb Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 18 Sep 2015 03:08:40 +0200 Subject: [PATCH 436/647] PCI / PM: Update runtime PM documentation for PCI devices Section 3.2 "Device Runtime Power Management" of pci.txt has become outdated, so update it to correctly reflect the current code flow. Also update the comment in local_pci_probe() to document the fact that pm_runtime_put_noidle() is not the only runtime PM helper function that can be used to decrement the device's runtime PM usage counter in .probe(). Signed-off-by: Rafael J. Wysocki Acked-by: Alan Stern --- Documentation/power/pci.txt | 51 +++++++++++++++++++++++++++---------- drivers/pci/pci-driver.c | 7 ++--- 2 files changed, 42 insertions(+), 16 deletions(-) diff --git a/Documentation/power/pci.txt b/Documentation/power/pci.txt index 62328d76b55b..b0e911e0e8f5 100644 --- a/Documentation/power/pci.txt +++ b/Documentation/power/pci.txt @@ -979,20 +979,45 @@ every time right after the runtime_resume() callback has returned (alternatively, the runtime_suspend() callback will have to check if the device should really be suspended and return -EAGAIN if that is not the case). -The runtime PM of PCI devices is disabled by default. It is also blocked by -pci_pm_init() that runs the pm_runtime_forbid() helper function. If a PCI -driver implements the runtime PM callbacks and intends to use the runtime PM -framework provided by the PM core and the PCI subsystem, it should enable this -feature by executing the pm_runtime_enable() helper function. However, the -driver should not call the pm_runtime_allow() helper function unblocking -the runtime PM of the device. Instead, it should allow user space or some -platform-specific code to do that (user space can do it via sysfs), although -once it has called pm_runtime_enable(), it must be prepared to handle the +The runtime PM of PCI devices is enabled by default by the PCI core. PCI +device drivers do not need to enable it and should not attempt to do so. +However, it is blocked by pci_pm_init() that runs the pm_runtime_forbid() +helper function. In addition to that, the runtime PM usage counter of +each PCI device is incremented by local_pci_probe() before executing the +probe callback provided by the device's driver. + +If a PCI driver implements the runtime PM callbacks and intends to use the +runtime PM framework provided by the PM core and the PCI subsystem, it needs +to decrement the device's runtime PM usage counter in its probe callback +function. If it doesn't do that, the counter will always be different from +zero for the device and it will never be runtime-suspended. The simplest +way to do that is by calling pm_runtime_put_noidle(), but if the driver +wants to schedule an autosuspend right away, for example, it may call +pm_runtime_put_autosuspend() instead for this purpose. Generally, it +just needs to call a function that decrements the devices usage counter +from its probe routine to make runtime PM work for the device. + +It is important to remember that the driver's runtime_suspend() callback +may be executed right after the usage counter has been decremented, because +user space may already have cuased the pm_runtime_allow() helper function +unblocking the runtime PM of the device to run via sysfs, so the driver must +be prepared to cope with that. + +The driver itself should not call pm_runtime_allow(), though. Instead, it +should let user space or some platform-specific code do that (user space can +do it via sysfs as stated above), but it must be prepared to handle the runtime PM of the device correctly as soon as pm_runtime_allow() is called -(which may happen at any time). [It also is possible that user space causes -pm_runtime_allow() to be called via sysfs before the driver is loaded, so in -fact the driver has to be prepared to handle the runtime PM of the device as -soon as it calls pm_runtime_enable().] +(which may happen at any time, even before the driver is loaded). + +When the driver's remove callback runs, it has to balance the decrementation +of the device's runtime PM usage counter at the probe time. For this reason, +if it has decremented the counter in its probe callback, it must run +pm_runtime_get_noresume() in its remove callback. [Since the core carries +out a runtime resume of the device and bumps up the device's usage counter +before running the driver's remove callback, the runtime PM of the device +is effectively disabled for the duration of the remove execution and all +runtime PM helper functions incrementing the device's usage counter are +then effectively equivalent to pm_runtime_get_noresume().] The runtime PM framework works by processing requests to suspend or resume devices, or to check if they are idle (in which cases it is reasonable to diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index dd652f2ae03d..108a3118ace7 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -299,9 +299,10 @@ static long local_pci_probe(void *_ddi) * Unbound PCI devices are always put in D0, regardless of * runtime PM status. During probe, the device is set to * active and the usage count is incremented. If the driver - * supports runtime PM, it should call pm_runtime_put_noidle() - * in its probe routine and pm_runtime_get_noresume() in its - * remove routine. + * supports runtime PM, it should call pm_runtime_put_noidle(), + * or any other runtime PM helper function decrementing the usage + * count, in its probe routine and pm_runtime_get_noresume() in + * its remove routine. */ pm_runtime_get_sync(dev); pci_dev->driver = pci_drv; From aebf5a67db8dbacbc624b9c652b81f5460b15eff Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Mon, 21 Sep 2015 11:06:32 +0200 Subject: [PATCH 437/647] dmaengine: pxa_dma: fix initial list move Since the commit to have an allocated list of virtual descriptors was reverted, the pxa_dma driver is broken, as it assumes the descriptor is placed on the allocated list upon allocation. Fix the issue in pxa_dma by making an allocated virtual descriptor a singleton. Fixes: 8c8fe97b2b8a ("Revert "dmaengine: virt-dma: don't always free descriptor upon completion"") Signed-off-by: Robert Jarzmik Signed-off-by: Vinod Koul --- drivers/dma/pxa_dma.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/dma/pxa_dma.c b/drivers/dma/pxa_dma.c index 5cb61ce01036..c6723ecd5848 100644 --- a/drivers/dma/pxa_dma.c +++ b/drivers/dma/pxa_dma.c @@ -887,6 +887,7 @@ pxad_tx_prep(struct virt_dma_chan *vc, struct virt_dma_desc *vd, struct dma_async_tx_descriptor *tx; struct pxad_chan *chan = container_of(vc, struct pxad_chan, vc); + INIT_LIST_HEAD(&vd->node); tx = vchan_tx_prep(vc, vd, tx_flags); tx->tx_submit = pxad_tx_submit; dev_dbg(&chan->vc.chan.dev->device, From c1492b4c541e3a382b60f1b5879cd3c4d246ad31 Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Thu, 24 Sep 2015 16:00:17 +0200 Subject: [PATCH 438/647] dmaengine: xgene-dma: fix handling xgene_dma_get_ring_size result The function can return negative value. The problem has been detected using proposed semantic patch scripts/coccinelle/tests/assign_signed_to_unsigned.cocci [1]. [1]: http://permalink.gmane.org/gmane.linux.kernel/2046107 Signed-off-by: Andrzej Hajda Signed-off-by: Vinod Koul --- drivers/dma/xgene-dma.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/dma/xgene-dma.c b/drivers/dma/xgene-dma.c index b23e8d52d126..4c5dddaad2d5 100644 --- a/drivers/dma/xgene-dma.c +++ b/drivers/dma/xgene-dma.c @@ -1421,15 +1421,18 @@ static int xgene_dma_create_ring_one(struct xgene_dma_chan *chan, struct xgene_dma_ring *ring, enum xgene_dma_ring_cfgsize cfgsize) { + int ret; + /* Setup DMA ring descriptor variables */ ring->pdma = chan->pdma; ring->cfgsize = cfgsize; ring->num = chan->pdma->ring_num++; ring->id = XGENE_DMA_RING_ID_GET(ring->owner, ring->buf_num); - ring->size = xgene_dma_get_ring_size(chan, cfgsize); - if (ring->size <= 0) - return ring->size; + ret = xgene_dma_get_ring_size(chan, cfgsize); + if (ret <= 0) + return ret; + ring->size = ret; /* Allocate memory for DMA ring descriptor */ ring->desc_vaddr = dma_zalloc_coherent(chan->dev, ring->size, From 0b23a1ece9be2c3e04c3b8d3594a1ada1fa1ae50 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 14 Sep 2015 11:55:36 +0300 Subject: [PATCH 439/647] dmaengine: idma64: improve residue estimation The residue calculation may provide a wrong estimation when the transfer is started. There are possible scenarios we have to separate: 1) the transfer is not started yet; residue is equal to the total length; 2) the transfer is just started (first chunk is ongoing); residue is equal to the total length without already transfered bytes; 3) the transfer is ongoing and we already sent few chunks of data; residue is equal to the total length without fully transfered chunks and already sent bytes. Mistakenly the calculation in cases 2) and 3) was done in the similar way and the result is equal to -bytes that have been transfered, i.e. quite big since size_t type can't keep negative values. Rewrite the calculation algorithm to be one pass and have a correct result. Besides above in case user asks for a status of the active DMA descriptor without pausing an ongoing transfer the residue will be estimated based on the register value, though it's still racy. Since the transfer is active the value is continuously being changed. Here we have to read two registers at a time. To minimize an error make those reads close to each other. Signed-off-by: Andy Shevchenko Signed-off-by: Vinod Koul --- drivers/dma/idma64.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/dma/idma64.c b/drivers/dma/idma64.c index 18c14e1f1414..48d6d9e94f67 100644 --- a/drivers/dma/idma64.c +++ b/drivers/dma/idma64.c @@ -355,23 +355,23 @@ static size_t idma64_active_desc_size(struct idma64_chan *idma64c) struct idma64_desc *desc = idma64c->desc; struct idma64_hw_desc *hw; size_t bytes = desc->length; - u64 llp; - u32 ctlhi; + u64 llp = channel_readq(idma64c, LLP); + u32 ctlhi = channel_readl(idma64c, CTL_HI); unsigned int i = 0; - llp = channel_readq(idma64c, LLP); do { hw = &desc->hw[i]; - } while ((hw->llp != llp) && (++i < desc->ndesc)); + if (hw->llp == llp) + break; + bytes -= hw->len; + } while (++i < desc->ndesc); if (!i) return bytes; - do { - bytes -= desc->hw[--i].len; - } while (i); + /* The current chunk is not fully transfered yet */ + bytes += desc->hw[--i].len; - ctlhi = channel_readl(idma64c, CTL_HI); return bytes - IDMA64C_CTLH_BLOCK_TS(ctlhi); } From a136442131443d929d2d8d243157824de4dfbae8 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Sep 2015 20:35:52 +0100 Subject: [PATCH 440/647] phy: fix of_mdio_find_bus() device refcount leak of_mdio_find_bus() leaks a struct device refcount, caused by using class_find_device() and not realising that the device reference has its refcount incremented: * Note, you will need to drop the reference with put_device() after use. ... while ((dev = class_dev_iter_next(&iter))) { if (match(dev, data)) { get_device(dev); break; } Update the comment, and arrange for the phy code to drop this refcount when disposing of a reference to it. Signed-off-by: Russell King Signed-off-by: David S. Miller --- drivers/net/phy/mdio-mux.c | 19 +++++++++++++------ drivers/net/phy/mdio_bus.c | 4 +++- 2 files changed, 16 insertions(+), 7 deletions(-) diff --git a/drivers/net/phy/mdio-mux.c b/drivers/net/phy/mdio-mux.c index 4d4d25efc1e1..280c7c311f72 100644 --- a/drivers/net/phy/mdio-mux.c +++ b/drivers/net/phy/mdio-mux.c @@ -113,18 +113,18 @@ int mdio_mux_init(struct device *dev, if (!parent_bus_node) return -ENODEV; - parent_bus = of_mdio_find_bus(parent_bus_node); - if (parent_bus == NULL) { - ret_val = -EPROBE_DEFER; - goto err_parent_bus; - } - pb = devm_kzalloc(dev, sizeof(*pb), GFP_KERNEL); if (pb == NULL) { ret_val = -ENOMEM; goto err_parent_bus; } + parent_bus = of_mdio_find_bus(parent_bus_node); + if (parent_bus == NULL) { + ret_val = -EPROBE_DEFER; + goto err_parent_bus; + } + pb->switch_data = data; pb->switch_fn = switch_fn; pb->current_child = -1; @@ -173,6 +173,10 @@ int mdio_mux_init(struct device *dev, dev_info(dev, "Version " DRV_VERSION "\n"); return 0; } + + /* balance the reference of_mdio_find_bus() took */ + put_device(&pb->mii_bus->dev); + err_parent_bus: of_node_put(parent_bus_node); return ret_val; @@ -189,6 +193,9 @@ void mdio_mux_uninit(void *mux_handle) mdiobus_free(cb->mii_bus); cb = cb->next; } + + /* balance the reference of_mdio_find_bus() in mdio_mux_init() took */ + put_device(&pb->mii_bus->dev); } EXPORT_SYMBOL_GPL(mdio_mux_uninit); diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 02a4615b65f8..67553e13bd36 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -167,7 +167,9 @@ static int of_mdio_bus_match(struct device *dev, const void *mdio_bus_np) * of_mdio_find_bus - Given an mii_bus node, find the mii_bus. * @mdio_bus_np: Pointer to the mii_bus. * - * Returns a pointer to the mii_bus, or NULL if none found. + * Returns a reference to the mii_bus, or NULL if none found. The + * embedded struct device will have its reference count incremented, + * and this must be put once the bus is finished with. * * Because the association of a device_node and mii_bus is made via * of_mdiobus_register(), the mii_bus cannot be found before it is From e496ae690b2faff751e1849fb97b060615e21f28 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Sep 2015 20:35:57 +0100 Subject: [PATCH 441/647] net: dsa: fix of_mdio_find_bus() device refcount leak Current users of of_mdio_find_bus() leak a struct device refcount, as they fail to clean up the reference obtained inside class_find_device(). Fix the DSA code to properly refcount the returned MDIO bus by: 1. taking a reference on the struct device whenever we assign it to pd->chip[x].host_dev. 2. dropping the reference when we overwrite the existing reference. 3. dropping the reference when we free the data structure. 4. dropping the initial reference we obtained after setting up the platform data structure, or on failure. In step 2 above, where we obtain a new MDIO bus, there is no need to take a reference on it as we would only have to drop it immediately after assignment again, iow: put_device(cd->host_dev); /* drop original assignment ref */ cd->host_dev = get_device(&mdio_bus_switch->dev); /* get our ref */ put_device(&mdio_bus_switch->dev); /* drop of_mdio_find_bus ref */ Signed-off-by: Russell King Signed-off-by: David S. Miller --- net/dsa/dsa.c | 38 +++++++++++++++++++++++++++++++------- 1 file changed, 31 insertions(+), 7 deletions(-) diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 76e3800765f8..bf4ba15fb780 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -634,6 +634,10 @@ static void dsa_of_free_platform_data(struct dsa_platform_data *pd) port_index++; } kfree(pd->chip[i].rtable); + + /* Drop our reference to the MDIO bus device */ + if (pd->chip[i].host_dev) + put_device(pd->chip[i].host_dev); } kfree(pd->chip); } @@ -661,16 +665,22 @@ static int dsa_of_probe(struct device *dev) return -EPROBE_DEFER; ethernet = of_parse_phandle(np, "dsa,ethernet", 0); - if (!ethernet) - return -EINVAL; + if (!ethernet) { + ret = -EINVAL; + goto out_put_mdio; + } ethernet_dev = of_find_net_device_by_node(ethernet); - if (!ethernet_dev) - return -EPROBE_DEFER; + if (!ethernet_dev) { + ret = -EPROBE_DEFER; + goto out_put_mdio; + } pd = kzalloc(sizeof(*pd), GFP_KERNEL); - if (!pd) - return -ENOMEM; + if (!pd) { + ret = -ENOMEM; + goto out_put_mdio; + } dev->platform_data = pd; pd->of_netdev = ethernet_dev; @@ -691,7 +701,9 @@ static int dsa_of_probe(struct device *dev) cd = &pd->chip[chip_index]; cd->of_node = child; - cd->host_dev = &mdio_bus->dev; + + /* When assigning the host device, increment its refcount */ + cd->host_dev = get_device(&mdio_bus->dev); sw_addr = of_get_property(child, "reg", NULL); if (!sw_addr) @@ -711,6 +723,12 @@ static int dsa_of_probe(struct device *dev) ret = -EPROBE_DEFER; goto out_free_chip; } + + /* Drop the mdio_bus device ref, replacing the host + * device with the mdio_bus_switch device, keeping + * the refcount from of_mdio_find_bus() above. + */ + put_device(cd->host_dev); cd->host_dev = &mdio_bus_switch->dev; } @@ -744,6 +762,10 @@ static int dsa_of_probe(struct device *dev) } } + /* The individual chips hold their own refcount on the mdio bus, + * so drop ours */ + put_device(&mdio_bus->dev); + return 0; out_free_chip: @@ -751,6 +773,8 @@ out_free_chip: out_free: kfree(pd); dev->platform_data = NULL; +out_put_mdio: + put_device(&mdio_bus->dev); return ret; } From 3e3aaf649416988ca8be4ad2c52dc24d8be7b46e Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Sep 2015 20:36:02 +0100 Subject: [PATCH 442/647] phy: fix mdiobus module safety Re-implement the mdiobus module refcounting to ensure that we actually ensure that the mdiobus module code does not go away while we might call into it. The old scheme using bus->dev.driver was buggy, because bus->dev is a class device which never has a struct device_driver associated with it, and hence the associated code trying to obtain a refcount did nothing useful. Instead, take the approach that other subsystems do: pass the module when calling mdiobus_register(), and record that in the mii_bus struct. When we need to increment the module use count in the phy code, use this stored pointer. When the phy is deteched, drop the module refcount, remembering that the phy device might go away at that point. This doesn't stop the mii_bus going away while there are in-use phys - it merely stops the underlying code vanishing. Signed-off-by: Russell King Signed-off-by: David S. Miller --- drivers/net/phy/mdio_bus.c | 5 +++-- drivers/net/phy/phy_device.c | 32 ++++++++++++++++++-------------- include/linux/phy.h | 5 ++++- 3 files changed, 25 insertions(+), 17 deletions(-) diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 67553e13bd36..992406624b7c 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -244,7 +244,7 @@ static inline void of_mdiobus_link_phydev(struct mii_bus *mdio, * * Returns 0 on success or < 0 on error. */ -int mdiobus_register(struct mii_bus *bus) +int __mdiobus_register(struct mii_bus *bus, struct module *owner) { int i, err; @@ -255,6 +255,7 @@ int mdiobus_register(struct mii_bus *bus) BUG_ON(bus->state != MDIOBUS_ALLOCATED && bus->state != MDIOBUS_UNREGISTERED); + bus->owner = owner; bus->dev.parent = bus->parent; bus->dev.class = &mdio_bus_class; bus->dev.groups = NULL; @@ -296,7 +297,7 @@ error: device_del(&bus->dev); return err; } -EXPORT_SYMBOL(mdiobus_register); +EXPORT_SYMBOL(__mdiobus_register); void mdiobus_unregister(struct mii_bus *bus) { diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index c0f211127274..03adf328f49b 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -582,10 +582,15 @@ EXPORT_SYMBOL(phy_init_hw); int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, u32 flags, phy_interface_t interface) { + struct mii_bus *bus = phydev->bus; struct device *d = &phydev->dev; - struct module *bus_module; int err; + if (!try_module_get(bus->owner)) { + dev_err(&dev->dev, "failed to get the bus module\n"); + return -EIO; + } + /* Assume that if there is no driver, that it doesn't * exist, and we should use the genphy driver. */ @@ -600,20 +605,13 @@ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, err = device_bind_driver(d); if (err) - return err; + goto error; } if (phydev->attached_dev) { dev_err(&dev->dev, "PHY already attached\n"); - return -EBUSY; - } - - /* Increment the bus module reference count */ - bus_module = phydev->bus->dev.driver ? - phydev->bus->dev.driver->owner : NULL; - if (!try_module_get(bus_module)) { - dev_err(&dev->dev, "failed to get the bus module\n"); - return -EIO; + err = -EBUSY; + goto error; } phydev->attached_dev = dev; @@ -636,6 +634,10 @@ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, phy_resume(phydev); return err; + +error: + module_put(bus->owner); + return err; } EXPORT_SYMBOL(phy_attach_direct); @@ -680,11 +682,9 @@ EXPORT_SYMBOL(phy_attach); */ void phy_detach(struct phy_device *phydev) { + struct mii_bus *bus; int i; - if (phydev->bus->dev.driver) - module_put(phydev->bus->dev.driver->owner); - phydev->attached_dev->phydev = NULL; phydev->attached_dev = NULL; phy_suspend(phydev); @@ -700,6 +700,10 @@ void phy_detach(struct phy_device *phydev) break; } } + + bus = phydev->bus; + + module_put(bus->owner); } EXPORT_SYMBOL(phy_detach); diff --git a/include/linux/phy.h b/include/linux/phy.h index 962387a192f1..11bce44f6d65 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -19,6 +19,7 @@ #include #include #include +#include #include #include #include @@ -153,6 +154,7 @@ struct sk_buff; * PHYs should register using this structure */ struct mii_bus { + struct module *owner; const char *name; char id[MII_BUS_ID_SIZE]; void *priv; @@ -198,7 +200,8 @@ static inline struct mii_bus *mdiobus_alloc(void) return mdiobus_alloc_size(0); } -int mdiobus_register(struct mii_bus *bus); +int __mdiobus_register(struct mii_bus *bus, struct module *owner); +#define mdiobus_register(bus) __mdiobus_register(bus, THIS_MODULE) void mdiobus_unregister(struct mii_bus *bus); void mdiobus_free(struct mii_bus *bus); struct mii_bus *devm_mdiobus_alloc_size(struct device *dev, int sizeof_priv); From 7322967bc1bd97ac9c49ecea19e5a1f681ca27ee Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Sep 2015 20:36:08 +0100 Subject: [PATCH 443/647] phy: add proper phy struct device refcounting Take a refcount on the phy struct device when the phy device is attached to a network device, and drop it after it's detached. This ensures that a refcount is held on the phy device while the device is being used by a network device, thereby preventing the phy_device from being unexpectedly kfree()'d by phy_device_release(). Signed-off-by: Russell King Signed-off-by: David S. Miller --- drivers/net/phy/phy_device.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 03adf328f49b..97a4f52addac 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -578,6 +578,7 @@ EXPORT_SYMBOL(phy_init_hw); * generic driver is used. The phy_device is given a ptr to * the attaching device, and given a callback for link status * change. The phy_device is returned to the attaching driver. + * This function takes a reference on the phy device. */ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, u32 flags, phy_interface_t interface) @@ -591,6 +592,8 @@ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, return -EIO; } + get_device(d); + /* Assume that if there is no driver, that it doesn't * exist, and we should use the genphy driver. */ @@ -636,6 +639,7 @@ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, return err; error: + put_device(d); module_put(bus->owner); return err; } @@ -679,6 +683,9 @@ EXPORT_SYMBOL(phy_attach); /** * phy_detach - detach a PHY device from its network device * @phydev: target phy_device struct + * + * This detaches the phy device from its network device and the phy + * driver, and drops the reference count taken in phy_attach_direct(). */ void phy_detach(struct phy_device *phydev) { @@ -701,8 +708,13 @@ void phy_detach(struct phy_device *phydev) } } + /* + * The phydev might go away on the put_device() below, so avoid + * a use-after-free bug by reading the underlying bus first. + */ bus = phydev->bus; + put_device(&phydev->dev); module_put(bus->owner); } EXPORT_SYMBOL(phy_detach); From f018ae7a8c576345d56a0cd40d86c0574a2eb360 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Sep 2015 20:36:13 +0100 Subject: [PATCH 444/647] of_mdio: fix MDIO phy device refcounting bus_find_device() is defined as: * This is similar to the bus_for_each_dev() function above, but it * returns a reference to a device that is 'found' for later use, as * determined by the @match callback. and it does indeed return a reference-counted pointer to the device: while ((dev = next_device(&i))) if (match(dev, data) && get_device(dev)) ^^^^^^^^^^^^^^^ break; klist_iter_exit(&i); return dev; What that means is that when we're done with the struct device, we must drop that reference. Neither of_phy_connect() nor of_phy_attach() did this when phy_connect_direct() or phy_attach_direct() failed. With our previous patch, phy_connect_direct() and phy_attach_direct() take a new refcount on the phy device when successful, so we can drop our local reference immediatley after these functions, whether or not they succeeded. Signed-off-by: Russell King Acked-by: Rob Herring Signed-off-by: David S. Miller --- drivers/of/of_mdio.c | 27 +++++++++++++++++++++++---- 1 file changed, 23 insertions(+), 4 deletions(-) diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index 1350fa25cdb0..a87a868fed64 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -197,7 +197,8 @@ static int of_phy_match(struct device *dev, void *phy_np) * of_phy_find_device - Give a PHY node, find the phy_device * @phy_np: Pointer to the phy's device tree node * - * Returns a pointer to the phy_device. + * If successful, returns a pointer to the phy_device with the embedded + * struct device refcount incremented by one, or NULL on failure. */ struct phy_device *of_phy_find_device(struct device_node *phy_np) { @@ -217,7 +218,9 @@ EXPORT_SYMBOL(of_phy_find_device); * @hndlr: Link state callback for the network device * @iface: PHY data interface type * - * Returns a pointer to the phy_device if successful. NULL otherwise + * If successful, returns a pointer to the phy_device with the embedded + * struct device refcount incremented by one, or NULL on failure. The + * refcount must be dropped by calling phy_disconnect() or phy_detach(). */ struct phy_device *of_phy_connect(struct net_device *dev, struct device_node *phy_np, @@ -225,13 +228,19 @@ struct phy_device *of_phy_connect(struct net_device *dev, phy_interface_t iface) { struct phy_device *phy = of_phy_find_device(phy_np); + int ret; if (!phy) return NULL; phy->dev_flags = flags; - return phy_connect_direct(dev, phy, hndlr, iface) ? NULL : phy; + ret = phy_connect_direct(dev, phy, hndlr, iface); + + /* refcount is held by phy_connect_direct() on success */ + put_device(&phy->dev); + + return ret ? NULL : phy; } EXPORT_SYMBOL(of_phy_connect); @@ -241,17 +250,27 @@ EXPORT_SYMBOL(of_phy_connect); * @phy_np: Node pointer for the PHY * @flags: flags to pass to the PHY * @iface: PHY data interface type + * + * If successful, returns a pointer to the phy_device with the embedded + * struct device refcount incremented by one, or NULL on failure. The + * refcount must be dropped by calling phy_disconnect() or phy_detach(). */ struct phy_device *of_phy_attach(struct net_device *dev, struct device_node *phy_np, u32 flags, phy_interface_t iface) { struct phy_device *phy = of_phy_find_device(phy_np); + int ret; if (!phy) return NULL; - return phy_attach_direct(dev, phy, flags, iface) ? NULL : phy; + ret = phy_attach_direct(dev, phy, flags, iface); + + /* refcount is held by phy_attach_direct() on success */ + put_device(&phy->dev); + + return ret ? NULL : phy; } EXPORT_SYMBOL(of_phy_attach); From 04d53b20fe44afe635b3d4438b437f7a12927e9a Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Sep 2015 20:36:18 +0100 Subject: [PATCH 445/647] net: fix phy refcounting in a bunch of drivers of_phy_find_device() increments the phy struct device refcount, which we need to properly balance. Add code to network drivers using this function to ensure that the struct device refcount is correctly balanced. For xgene, looking back in the history, we should be able to use of_phy_connect() with a zero flags argument for the DT case as this is how the driver used to operate prior to de7b5b3d790a ("net: eth: xgene: change APM X-Gene SoC platform ethernet to support ACPI"). This leaves the Cavium Thunder BGX unfixed; fixing this driver is a complicated task, one which the maintainers need to be involved with. Signed-off-by: Russell King Signed-off-by: David S. Miller --- .../net/ethernet/apm/xgene/xgene_enet_hw.c | 24 ++++++++++++------- drivers/net/ethernet/freescale/gianfar.c | 3 +++ drivers/net/ethernet/freescale/ucc_geth.c | 8 ++++++- drivers/net/ethernet/marvell/mvneta.c | 2 ++ drivers/net/ethernet/xilinx/xilinx_emaclite.c | 2 ++ 5 files changed, 30 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index cfa37041ab71..c4bb8027b3fb 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -689,16 +689,24 @@ static int xgene_enet_phy_connect(struct net_device *ndev) netdev_dbg(ndev, "No phy-handle found in DT\n"); return -ENODEV; } - pdata->phy_dev = of_phy_find_device(phy_np); - } - phy_dev = pdata->phy_dev; + phy_dev = of_phy_connect(ndev, phy_np, &xgene_enet_adjust_link, + 0, pdata->phy_mode); + if (!phy_dev) { + netdev_err(ndev, "Could not connect to PHY\n"); + return -ENODEV; + } - if (!phy_dev || - phy_connect_direct(ndev, phy_dev, &xgene_enet_adjust_link, - pdata->phy_mode)) { - netdev_err(ndev, "Could not connect to PHY\n"); - return -ENODEV; + pdata->phy_dev = phy_dev; + } else { + phy_dev = pdata->phy_dev; + + if (!phy_dev || + phy_connect_direct(ndev, phy_dev, &xgene_enet_adjust_link, + pdata->phy_mode)) { + netdev_err(ndev, "Could not connect to PHY\n"); + return -ENODEV; + } } pdata->phy_speed = SPEED_UNKNOWN; diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index 803ed4c93503..a5cf4332d307 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -1702,6 +1702,7 @@ static void gfar_configure_serdes(struct net_device *dev) tbiphy = of_phy_find_device(priv->tbi_node); if (!tbiphy) { dev_err(&dev->dev, "error: Could not get TBI device\n"); + put_device(&tbiphy->dev); return; } @@ -1723,6 +1724,8 @@ static void gfar_configure_serdes(struct net_device *dev) phy_write(tbiphy, MII_BMCR, BMCR_ANENABLE | BMCR_ANRESTART | BMCR_FULLDPLX | BMCR_SPEED1000); + + put_device(&tbiphy->dev); } static int __gfar_is_rx_idle(struct gfar_private *priv) diff --git a/drivers/net/ethernet/freescale/ucc_geth.c b/drivers/net/ethernet/freescale/ucc_geth.c index 4dd40e057f40..650f7888e32b 100644 --- a/drivers/net/ethernet/freescale/ucc_geth.c +++ b/drivers/net/ethernet/freescale/ucc_geth.c @@ -1384,6 +1384,8 @@ static int adjust_enet_interface(struct ucc_geth_private *ugeth) value = phy_read(tbiphy, ENET_TBI_MII_CR); value &= ~0x1000; /* Turn off autonegotiation */ phy_write(tbiphy, ENET_TBI_MII_CR, value); + + put_device(&tbiphy->dev); } init_check_frame_length_mode(ug_info->lengthCheckRx, &ug_regs->maccfg2); @@ -1702,8 +1704,10 @@ static void uec_configure_serdes(struct net_device *dev) * everything for us? Resetting it takes the link down and requires * several seconds for it to come back. */ - if (phy_read(tbiphy, ENET_TBI_MII_SR) & TBISR_LSTATUS) + if (phy_read(tbiphy, ENET_TBI_MII_SR) & TBISR_LSTATUS) { + put_device(&tbiphy->dev); return; + } /* Single clk mode, mii mode off(for serdes communication) */ phy_write(tbiphy, ENET_TBI_MII_ANA, TBIANA_SETTINGS); @@ -1711,6 +1715,8 @@ static void uec_configure_serdes(struct net_device *dev) phy_write(tbiphy, ENET_TBI_MII_TBICON, TBICON_CLK_SELECT); phy_write(tbiphy, ENET_TBI_MII_CR, TBICR_SETTINGS); + + put_device(&tbiphy->dev); } /* Configure the PHY for dev. diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 09ec32e33076..514df76fc70f 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -3175,6 +3175,8 @@ static int mvneta_probe(struct platform_device *pdev) struct phy_device *phy = of_phy_find_device(dn); mvneta_fixed_link_update(pp, phy); + + put_device(&phy->dev); } return 0; diff --git a/drivers/net/ethernet/xilinx/xilinx_emaclite.c b/drivers/net/ethernet/xilinx/xilinx_emaclite.c index 6008eee01a33..cf468c87ce57 100644 --- a/drivers/net/ethernet/xilinx/xilinx_emaclite.c +++ b/drivers/net/ethernet/xilinx/xilinx_emaclite.c @@ -828,6 +828,8 @@ static int xemaclite_mdio_setup(struct net_local *lp, struct device *dev) if (!phydev) dev_info(dev, "MDIO of the phy is not registered yet\n"); + else + put_device(&phydev->dev); return 0; } From d618bf2bfd2a095644c852ebea16f5a981f9d875 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Sep 2015 20:36:23 +0100 Subject: [PATCH 446/647] phy: fixed-phy: properly validate phy in fixed_phy_update_state() Validate that the phy_device passed into fixed_phy_update_state() is a fixed-phy device before walking the list of phys for a fixed phy at the same address. Signed-off-by: Russell King Signed-off-by: David S. Miller --- drivers/net/phy/fixed_phy.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index fb1299c6326e..e23bf5b90e17 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -220,7 +220,7 @@ int fixed_phy_update_state(struct phy_device *phydev, struct fixed_mdio_bus *fmb = &platform_fmb; struct fixed_phy *fp; - if (!phydev || !phydev->bus) + if (!phydev || phydev->bus != fmb->mii_bus) return -EINVAL; list_for_each_entry(fp, &fmb->phys, node) { From 38737e490d4ea91660d3cec83ef88c4e6d360ae4 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Sep 2015 20:36:28 +0100 Subject: [PATCH 447/647] phy: add phy_device_remove() Add a phy_device_remove() function to complement phy_device_register(), which undoes the effects of phy_device_register() by removing the phy device from visibility, but not freeing it. This allows these details to be moved out of the mdio bus code into the phy code where this action belongs. Signed-off-by: Russell King Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/gianfar.c | 5 +++-- drivers/net/phy/mdio_bus.c | 15 ++++++++++----- drivers/net/phy/phy_device.c | 18 ++++++++++++++++++ include/linux/phy.h | 1 + 4 files changed, 32 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index a5cf4332d307..710715fcb23d 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -1702,7 +1702,6 @@ static void gfar_configure_serdes(struct net_device *dev) tbiphy = of_phy_find_device(priv->tbi_node); if (!tbiphy) { dev_err(&dev->dev, "error: Could not get TBI device\n"); - put_device(&tbiphy->dev); return; } @@ -1711,8 +1710,10 @@ static void gfar_configure_serdes(struct net_device *dev) * everything for us? Resetting it takes the link down and requires * several seconds for it to come back. */ - if (phy_read(tbiphy, MII_BMSR) & BMSR_LSTATUS) + if (phy_read(tbiphy, MII_BMSR) & BMSR_LSTATUS) { + put_device(&tbiphy->dev); return; + } /* Single clk mode, mii mode off(for serdes communication) */ phy_write(tbiphy, MII_TBICON, TBICON_CLK_SELECT); diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 992406624b7c..c340e412b38f 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -291,8 +291,11 @@ int __mdiobus_register(struct mii_bus *bus, struct module *owner) error: while (--i >= 0) { - if (bus->phy_map[i]) - device_unregister(&bus->phy_map[i]->dev); + struct phy_device *phydev = bus->phy_map[i]; + if (phydev) { + phy_device_remove(phydev); + phy_device_free(phydev); + } } device_del(&bus->dev); return err; @@ -307,9 +310,11 @@ void mdiobus_unregister(struct mii_bus *bus) bus->state = MDIOBUS_UNREGISTERED; for (i = 0; i < PHY_MAX_ADDR; i++) { - if (bus->phy_map[i]) - device_unregister(&bus->phy_map[i]->dev); - bus->phy_map[i] = NULL; + struct phy_device *phydev = bus->phy_map[i]; + if (phydev) { + phy_device_remove(phydev); + phy_device_free(phydev); + } } device_del(&bus->dev); } diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 97a4f52addac..f761288abe66 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -383,6 +383,24 @@ int phy_device_register(struct phy_device *phydev) } EXPORT_SYMBOL(phy_device_register); +/** + * phy_device_remove - Remove a previously registered phy device from the MDIO bus + * @phydev: phy_device structure to remove + * + * This doesn't free the phy_device itself, it merely reverses the effects + * of phy_device_register(). Use phy_device_free() to free the device + * after calling this function. + */ +void phy_device_remove(struct phy_device *phydev) +{ + struct mii_bus *bus = phydev->bus; + int addr = phydev->addr; + + device_del(&phydev->dev); + bus->phy_map[addr] = NULL; +} +EXPORT_SYMBOL(phy_device_remove); + /** * phy_find_first - finds the first PHY device on the bus * @bus: the target MII bus diff --git a/include/linux/phy.h b/include/linux/phy.h index 11bce44f6d65..4a4e3a092337 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -745,6 +745,7 @@ struct phy_device *phy_device_create(struct mii_bus *bus, int addr, int phy_id, struct phy_c45_device_ids *c45_ids); struct phy_device *get_phy_device(struct mii_bus *bus, int addr, bool is_c45); int phy_device_register(struct phy_device *phy); +void phy_device_remove(struct phy_device *phydev); int phy_init_hw(struct phy_device *phydev); int phy_suspend(struct phy_device *phydev); int phy_resume(struct phy_device *phydev); From 9861f72074c77a8a065622c1be7e9c4277e600eb Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Sep 2015 20:36:33 +0100 Subject: [PATCH 448/647] net: fix net_device refcounting of_find_net_device_by_node() uses class_find_device() internally to lookup the corresponding network device. class_find_device() returns a reference to the embedded struct device, with its refcount incremented. Add a comment to the definition in net/core/net-sysfs.c indicating the need to drop this refcount, and fix the DSA code to drop this refcount when the OF-generated platform data is cleaned up and freed. Also arrange for the ref to be dropped when handling errors. Signed-off-by: Russell King Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 9 +++++++++ net/dsa/dsa.c | 5 ++++- 2 files changed, 13 insertions(+), 1 deletion(-) diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index b279077c3089..805a95a48107 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -1481,6 +1481,15 @@ static int of_dev_node_match(struct device *dev, const void *data) return ret == 0 ? dev->of_node == data : ret; } +/* + * of_find_net_device_by_node - lookup the net device for the device node + * @np: OF device node + * + * Looks up the net_device structure corresponding with the device node. + * If successful, returns a pointer to the net_device with the embedded + * struct device refcount incremented by one, or NULL on failure. The + * refcount must be dropped when done with the net_device. + */ struct net_device *of_find_net_device_by_node(struct device_node *np) { struct device *dev; diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index bf4ba15fb780..c59fa5d9c22c 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -679,7 +679,7 @@ static int dsa_of_probe(struct device *dev) pd = kzalloc(sizeof(*pd), GFP_KERNEL); if (!pd) { ret = -ENOMEM; - goto out_put_mdio; + goto out_put_ethernet; } dev->platform_data = pd; @@ -773,6 +773,8 @@ out_free_chip: out_free: kfree(pd); dev->platform_data = NULL; +out_put_ethernet: + put_device(ðernet_dev->dev); out_put_mdio: put_device(&mdio_bus->dev); return ret; @@ -786,6 +788,7 @@ static void dsa_of_remove(struct device *dev) return; dsa_of_free_platform_data(pd); + put_device(&pd->of_netdev->dev); kfree(pd); } #else From 3ccd6e83df8a0d4a664edeecc453c4fa046395fb Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Sun, 13 Sep 2015 02:30:46 -0700 Subject: [PATCH 449/647] target: Fix PR registration + APTPL RCU conversion regression This patch fixes a v4.2+ regression introduced by commit 79dc9c9e86 where lookup of t10_pr_registration->pr_reg_deve and associated ->pr_kref get was missing from __core_scsi3_do_alloc_registration(), which is responsible for setting DEF_PR_REG_ACTIVE. This would result in REGISTER operations completing successfully, but subsequent core_scsi3_pr_seq_non_holder() checking would fail with !DEF_PR_REG_ACTIVE -> RESERVATION CONFLICT status. Update __core_scsi3_add_registration() to drop ->pr_kref reference after registration and any optional ALL_TG_PT=1 processing has completed. Update core_scsi3_decode_spec_i_port() to release the new parent local_pr_reg->pr_kref as well. Also, update __core_scsi3_check_aptpl_registration() to perform the same target_nacl_find_deve() lookup + ->pr_kref get, now that __core_scsi3_add_registration() expects to drop the reference. Finally, since there are cases when se_dev_entry->se_lun_acl can still be dereferenced in core_scsi3_lunacl_undepend_item() while holding ->pr_kref, go ahead and move explicit rcu_assign_pointer() NULL assignments within core_disable_device_list_for_node() until after orig->pr_comp finishes. Reported-by: Scott L. Lykens Tested-by: Scott L. Lykens Cc: Christoph Hellwig Cc: Hannes Reinecke Cc: Lee Duncan Cc: # v4.2+ Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_device.c | 5 +- drivers/target/target_core_pr.c | 91 +++++++++++++++++++++-------- 2 files changed, 70 insertions(+), 26 deletions(-) diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c index dcc424ac35d4..abf20763b0b6 100644 --- a/drivers/target/target_core_device.c +++ b/drivers/target/target_core_device.c @@ -427,8 +427,6 @@ void core_disable_device_list_for_node( hlist_del_rcu(&orig->link); clear_bit(DEF_PR_REG_ACTIVE, &orig->deve_flags); - rcu_assign_pointer(orig->se_lun, NULL); - rcu_assign_pointer(orig->se_lun_acl, NULL); orig->lun_flags = 0; orig->creation_time = 0; orig->attach_count--; @@ -439,6 +437,9 @@ void core_disable_device_list_for_node( kref_put(&orig->pr_kref, target_pr_kref_release); wait_for_completion(&orig->pr_comp); + rcu_assign_pointer(orig->se_lun, NULL); + rcu_assign_pointer(orig->se_lun_acl, NULL); + kfree_rcu(orig, rcu_head); core_scsi3_free_pr_reg_from_nacl(dev, nacl); diff --git a/drivers/target/target_core_pr.c b/drivers/target/target_core_pr.c index 5ab7100de17e..e7933115087a 100644 --- a/drivers/target/target_core_pr.c +++ b/drivers/target/target_core_pr.c @@ -618,7 +618,7 @@ static struct t10_pr_registration *__core_scsi3_do_alloc_registration( struct se_device *dev, struct se_node_acl *nacl, struct se_lun *lun, - struct se_dev_entry *deve, + struct se_dev_entry *dest_deve, u64 mapped_lun, unsigned char *isid, u64 sa_res_key, @@ -640,7 +640,29 @@ static struct t10_pr_registration *__core_scsi3_do_alloc_registration( INIT_LIST_HEAD(&pr_reg->pr_reg_atp_mem_list); atomic_set(&pr_reg->pr_res_holders, 0); pr_reg->pr_reg_nacl = nacl; - pr_reg->pr_reg_deve = deve; + /* + * For destination registrations for ALL_TG_PT=1 and SPEC_I_PT=1, + * the se_dev_entry->pr_ref will have been already obtained by + * core_get_se_deve_from_rtpi() or __core_scsi3_alloc_registration(). + * + * Otherwise, locate se_dev_entry now and obtain a reference until + * registration completes in __core_scsi3_add_registration(). + */ + if (dest_deve) { + pr_reg->pr_reg_deve = dest_deve; + } else { + rcu_read_lock(); + pr_reg->pr_reg_deve = target_nacl_find_deve(nacl, mapped_lun); + if (!pr_reg->pr_reg_deve) { + rcu_read_unlock(); + pr_err("Unable to locate PR deve %s mapped_lun: %llu\n", + nacl->initiatorname, mapped_lun); + kmem_cache_free(t10_pr_reg_cache, pr_reg); + return NULL; + } + kref_get(&pr_reg->pr_reg_deve->pr_kref); + rcu_read_unlock(); + } pr_reg->pr_res_mapped_lun = mapped_lun; pr_reg->pr_aptpl_target_lun = lun->unpacked_lun; pr_reg->tg_pt_sep_rtpi = lun->lun_rtpi; @@ -936,17 +958,29 @@ static int __core_scsi3_check_aptpl_registration( !(strcmp(pr_reg->pr_tport, t_port)) && (pr_reg->pr_reg_tpgt == tpgt) && (pr_reg->pr_aptpl_target_lun == target_lun)) { + /* + * Obtain the ->pr_reg_deve pointer + reference, that + * is released by __core_scsi3_add_registration() below. + */ + rcu_read_lock(); + pr_reg->pr_reg_deve = target_nacl_find_deve(nacl, mapped_lun); + if (!pr_reg->pr_reg_deve) { + pr_err("Unable to locate PR APTPL %s mapped_lun:" + " %llu\n", nacl->initiatorname, mapped_lun); + rcu_read_unlock(); + continue; + } + kref_get(&pr_reg->pr_reg_deve->pr_kref); + rcu_read_unlock(); pr_reg->pr_reg_nacl = nacl; pr_reg->tg_pt_sep_rtpi = lun->lun_rtpi; - list_del(&pr_reg->pr_reg_aptpl_list); spin_unlock(&pr_tmpl->aptpl_reg_lock); /* * At this point all of the pointers in *pr_reg will * be setup, so go ahead and add the registration. */ - __core_scsi3_add_registration(dev, nacl, pr_reg, 0, 0); /* * If this registration is the reservation holder, @@ -1044,18 +1078,11 @@ static void __core_scsi3_add_registration( __core_scsi3_dump_registration(tfo, dev, nacl, pr_reg, register_type); spin_unlock(&pr_tmpl->registration_lock); - - rcu_read_lock(); - deve = pr_reg->pr_reg_deve; - if (deve) - set_bit(DEF_PR_REG_ACTIVE, &deve->deve_flags); - rcu_read_unlock(); - /* * Skip extra processing for ALL_TG_PT=0 or REGISTER_AND_MOVE. */ if (!pr_reg->pr_reg_all_tg_pt || register_move) - return; + goto out; /* * Walk pr_reg->pr_reg_atp_list and add registrations for ALL_TG_PT=1 * allocated in __core_scsi3_alloc_registration() @@ -1075,19 +1102,31 @@ static void __core_scsi3_add_registration( __core_scsi3_dump_registration(tfo, dev, nacl_tmp, pr_reg_tmp, register_type); spin_unlock(&pr_tmpl->registration_lock); - + /* + * Drop configfs group dependency reference and deve->pr_kref + * obtained from __core_scsi3_alloc_registration() code. + */ rcu_read_lock(); deve = pr_reg_tmp->pr_reg_deve; - if (deve) + if (deve) { set_bit(DEF_PR_REG_ACTIVE, &deve->deve_flags); + core_scsi3_lunacl_undepend_item(deve); + pr_reg_tmp->pr_reg_deve = NULL; + } rcu_read_unlock(); - - /* - * Drop configfs group dependency reference from - * __core_scsi3_alloc_registration() - */ - core_scsi3_lunacl_undepend_item(pr_reg_tmp->pr_reg_deve); } +out: + /* + * Drop deve->pr_kref obtained in __core_scsi3_do_alloc_registration() + */ + rcu_read_lock(); + deve = pr_reg->pr_reg_deve; + if (deve) { + set_bit(DEF_PR_REG_ACTIVE, &deve->deve_flags); + kref_put(&deve->pr_kref, target_pr_kref_release); + pr_reg->pr_reg_deve = NULL; + } + rcu_read_unlock(); } static int core_scsi3_alloc_registration( @@ -1785,9 +1824,11 @@ core_scsi3_decode_spec_i_port( dest_node_acl->initiatorname, i_buf, (dest_se_deve) ? dest_se_deve->mapped_lun : 0); - if (!dest_se_deve) + if (!dest_se_deve) { + kref_put(&local_pr_reg->pr_reg_deve->pr_kref, + target_pr_kref_release); continue; - + } core_scsi3_lunacl_undepend_item(dest_se_deve); core_scsi3_nodeacl_undepend_item(dest_node_acl); core_scsi3_tpg_undepend_item(dest_tpg); @@ -1823,9 +1864,11 @@ out: kmem_cache_free(t10_pr_reg_cache, dest_pr_reg); - if (!dest_se_deve) + if (!dest_se_deve) { + kref_put(&local_pr_reg->pr_reg_deve->pr_kref, + target_pr_kref_release); continue; - + } core_scsi3_lunacl_undepend_item(dest_se_deve); core_scsi3_nodeacl_undepend_item(dest_node_acl); core_scsi3_tpg_undepend_item(dest_tpg); From eeeb9522231118138be418ff527dc8c9050f4707 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Tue, 15 Sep 2015 17:27:35 -0700 Subject: [PATCH 450/647] target: Propigate backend read-only to core_tpg_add_lun This patch adds a DF_READ_ONLY flag that is used by IBLOCK to signal when a backend has been set to read-only mode, in order to propigate read-only status up to core_tpg_add_lun() for all future LUN fabric exports. With this is place, existing emulation for reporting read-only in spc_emulate_modesense() and normal transport_lookup_cmd_lun() TCM_WRITE_PROTECTED status checking just works as expected. Reported-by: Joeue Deng Reported-by: Andy Grover Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_iblock.c | 2 ++ drivers/target/target_core_tpg.c | 5 ++++- include/target/target_core_base.h | 1 + 3 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/target/target_core_iblock.c b/drivers/target/target_core_iblock.c index 5a9982f5d5d6..0f19e11acac2 100644 --- a/drivers/target/target_core_iblock.c +++ b/drivers/target/target_core_iblock.c @@ -105,6 +105,8 @@ static int iblock_configure_device(struct se_device *dev) mode = FMODE_READ|FMODE_EXCL; if (!ib_dev->ibd_readonly) mode |= FMODE_WRITE; + else + dev->dev_flags |= DF_READ_ONLY; bd = blkdev_get_by_path(ib_dev->ibd_udev_path, mode, ib_dev); if (IS_ERR(bd)) { diff --git a/drivers/target/target_core_tpg.c b/drivers/target/target_core_tpg.c index 2d0381dd105c..5fb9dd7f08bb 100644 --- a/drivers/target/target_core_tpg.c +++ b/drivers/target/target_core_tpg.c @@ -668,7 +668,10 @@ int core_tpg_add_lun( list_add_tail(&lun->lun_dev_link, &dev->dev_sep_list); spin_unlock(&dev->se_port_lock); - lun->lun_access = lun_access; + if (dev->dev_flags & DF_READ_ONLY) + lun->lun_access = TRANSPORT_LUNFLAGS_READ_ONLY; + else + lun->lun_access = lun_access; if (!(dev->se_hba->hba_flags & HBA_FLAGS_INTERNAL_USE)) hlist_add_head_rcu(&lun->link, &tpg->tpg_lun_hlist); mutex_unlock(&tpg->tpg_lun_mutex); diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h index ac9bf1c0e42d..5f48754dc36a 100644 --- a/include/target/target_core_base.h +++ b/include/target/target_core_base.h @@ -730,6 +730,7 @@ struct se_device { #define DF_EMULATED_VPD_UNIT_SERIAL 0x00000004 #define DF_USING_UDEV_PATH 0x00000008 #define DF_USING_ALIAS 0x00000010 +#define DF_READ_ONLY 0x00000020 /* Physical device queue depth */ u32 queue_depth; /* Used for SPC-2 reservations enforce of ISIDs */ From f3bb467ff64b2598d023b3a07592748fa7768b6a Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Tue, 15 Sep 2015 23:07:45 -0700 Subject: [PATCH 451/647] target: Fix target_sense_desc_format NULL pointer dereference This patch allows target_sense_desc_format() to be called without a valid se_device pointer, which can occur during an early exception ahead of transport_lookup_cmd_lun() setting up se_cmd->se_device. This addresses a v4.3-rc1 specific NULL pointer dereference regression introduced by commit 4e4937e8. Cc: Sagi Grimberg Cc: Christoph Hellwig Cc: Hannes Reinecke Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_hba.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/target/target_core_hba.c b/drivers/target/target_core_hba.c index 9522960c7fdd..22390e0e046c 100644 --- a/drivers/target/target_core_hba.c +++ b/drivers/target/target_core_hba.c @@ -187,5 +187,5 @@ core_delete_hba(struct se_hba *hba) bool target_sense_desc_format(struct se_device *dev) { - return dev->transport->get_blocks(dev) > U32_MAX; + return (dev) ? dev->transport->get_blocks(dev) > U32_MAX : false; } From 8fa3a867486f85df66eba8c4df85804d3309c6ad Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Wed, 16 Sep 2015 20:23:53 -0700 Subject: [PATCH 452/647] target: Make TCM_WRITE_PROTECT failure honor D_SENSE bit This patch changes transport_lookup_cmd_lun() to obtain se_lun->lun_ref + se_cmd->se_device rcu_dereference during TCM_WRITE_PROTECT -> CHECK_CONDITION failure status. Do this to ensure the active control D_SENSE mode page bit is being honored. Reported-by: Sagi Grimberg Reviewed-by: Sagi Grimberg Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_device.c | 40 +++++++++++++++++------------ 1 file changed, 23 insertions(+), 17 deletions(-) diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c index abf20763b0b6..88ea4e4f124b 100644 --- a/drivers/target/target_core_device.c +++ b/drivers/target/target_core_device.c @@ -62,22 +62,13 @@ transport_lookup_cmd_lun(struct se_cmd *se_cmd, u64 unpacked_lun) struct se_session *se_sess = se_cmd->se_sess; struct se_node_acl *nacl = se_sess->se_node_acl; struct se_dev_entry *deve; + sense_reason_t ret = TCM_NO_SENSE; rcu_read_lock(); deve = target_nacl_find_deve(nacl, unpacked_lun); if (deve) { atomic_long_inc(&deve->total_cmds); - if ((se_cmd->data_direction == DMA_TO_DEVICE) && - (deve->lun_flags & TRANSPORT_LUNFLAGS_READ_ONLY)) { - pr_err("TARGET_CORE[%s]: Detected WRITE_PROTECTED LUN" - " Access for 0x%08llx\n", - se_cmd->se_tfo->get_fabric_name(), - unpacked_lun); - rcu_read_unlock(); - return TCM_WRITE_PROTECTED; - } - if (se_cmd->data_direction == DMA_TO_DEVICE) atomic_long_add(se_cmd->data_length, &deve->write_bytes); @@ -93,6 +84,17 @@ transport_lookup_cmd_lun(struct se_cmd *se_cmd, u64 unpacked_lun) percpu_ref_get(&se_lun->lun_ref); se_cmd->lun_ref_active = true; + + if ((se_cmd->data_direction == DMA_TO_DEVICE) && + (deve->lun_flags & TRANSPORT_LUNFLAGS_READ_ONLY)) { + pr_err("TARGET_CORE[%s]: Detected WRITE_PROTECTED LUN" + " Access for 0x%08llx\n", + se_cmd->se_tfo->get_fabric_name(), + unpacked_lun); + rcu_read_unlock(); + ret = TCM_WRITE_PROTECTED; + goto ref_dev; + } } rcu_read_unlock(); @@ -109,12 +111,6 @@ transport_lookup_cmd_lun(struct se_cmd *se_cmd, u64 unpacked_lun) unpacked_lun); return TCM_NON_EXISTENT_LUN; } - /* - * Force WRITE PROTECT for virtual LUN 0 - */ - if ((se_cmd->data_direction != DMA_FROM_DEVICE) && - (se_cmd->data_direction != DMA_NONE)) - return TCM_WRITE_PROTECTED; se_lun = se_sess->se_tpg->tpg_virt_lun0; se_cmd->se_lun = se_sess->se_tpg->tpg_virt_lun0; @@ -123,6 +119,15 @@ transport_lookup_cmd_lun(struct se_cmd *se_cmd, u64 unpacked_lun) percpu_ref_get(&se_lun->lun_ref); se_cmd->lun_ref_active = true; + + /* + * Force WRITE PROTECT for virtual LUN 0 + */ + if ((se_cmd->data_direction != DMA_FROM_DEVICE) && + (se_cmd->data_direction != DMA_NONE)) { + ret = TCM_WRITE_PROTECTED; + goto ref_dev; + } } /* * RCU reference protected by percpu se_lun->lun_ref taken above that @@ -130,6 +135,7 @@ transport_lookup_cmd_lun(struct se_cmd *se_cmd, u64 unpacked_lun) * pointer can be kfree_rcu() by the final se_lun->lun_group put via * target_core_fabric_configfs.c:target_fabric_port_release */ +ref_dev: se_cmd->se_dev = rcu_dereference_raw(se_lun->lun_se_dev); atomic_long_inc(&se_cmd->se_dev->num_cmds); @@ -140,7 +146,7 @@ transport_lookup_cmd_lun(struct se_cmd *se_cmd, u64 unpacked_lun) atomic_long_add(se_cmd->data_length, &se_cmd->se_dev->read_bytes); - return 0; + return ret; } EXPORT_SYMBOL(transport_lookup_cmd_lun); From 673681cafa99776e334c3e61cafa2cf115950c32 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Tue, 22 Sep 2015 22:32:14 -0700 Subject: [PATCH 453/647] iscsi-target: Avoid OFMarker + IFMarker negotiation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch fixes a v4.2+ regression introduced by commit c04a6091 that removed support for obsolete sync-and-steering markers usage as originally defined in RFC-3720. The regression would involve attempting to send OFMarker=No + IFMarker=No keys during opertional negotiation login phase, including when initiators did not actually propose these keys. The result for MSFT iSCSI initiators would be random junk in TCP stream after the last successful login request was been sent signaling the move to full feature phase (FFP) operation. To address this bug, go ahead and avoid negotiating these keys by default unless the initiator explicitly proposes them, but still respond to them with 'No' if they are proposed. Reported-by: Dragan Milivojević Bisected-by: Christophe Vu-Brugier Tested-by: Christophe Vu-Brugier Signed-off-by: Nicholas Bellinger --- drivers/target/iscsi/iscsi_target_parameters.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/target/iscsi/iscsi_target_parameters.c b/drivers/target/iscsi/iscsi_target_parameters.c index e8a52f7d6204..51d1734d5390 100644 --- a/drivers/target/iscsi/iscsi_target_parameters.c +++ b/drivers/target/iscsi/iscsi_target_parameters.c @@ -407,6 +407,7 @@ int iscsi_create_default_params(struct iscsi_param_list **param_list_ptr) TYPERANGE_UTF8, USE_INITIAL_ONLY); if (!param) goto out; + /* * Extra parameters for ISER from RFC-5046 */ @@ -496,9 +497,9 @@ int iscsi_set_keys_to_negotiate( } else if (!strcmp(param->name, SESSIONTYPE)) { SET_PSTATE_NEGOTIATE(param); } else if (!strcmp(param->name, IFMARKER)) { - SET_PSTATE_NEGOTIATE(param); + SET_PSTATE_REJECT(param); } else if (!strcmp(param->name, OFMARKER)) { - SET_PSTATE_NEGOTIATE(param); + SET_PSTATE_REJECT(param); } else if (!strcmp(param->name, IFMARKINT)) { SET_PSTATE_REJECT(param); } else if (!strcmp(param->name, OFMARKINT)) { From 7e5560a5648ab2bce7199c73b9c2a51b846f5541 Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Thu, 24 Sep 2015 04:48:53 -0700 Subject: [PATCH 454/647] perf/x86: Change test_aperfmperf() and test_intel() to static Fixes the following sparse warnings: arch/x86/kernel/cpu/perf_event_msr.c:13:6: warning: symbol 'test_aperfmperf' was not declared. Should it be static? arch/x86/kernel/cpu/perf_event_msr.c:18:6: warning: symbol 'test_intel' was not declared. Should it be static? Signed-off-by: Geliang Tang Cc: Arnaldo Carvalho de Melo Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/4588e8ab09638458f2451af572827108be3b4a36.1443123796.git.geliangtang@163.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_msr.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/cpu/perf_event_msr.c b/arch/x86/kernel/cpu/perf_event_msr.c index 086b12eae794..f32ac13934f2 100644 --- a/arch/x86/kernel/cpu/perf_event_msr.c +++ b/arch/x86/kernel/cpu/perf_event_msr.c @@ -10,12 +10,12 @@ enum perf_msr_id { PERF_MSR_EVENT_MAX, }; -bool test_aperfmperf(int idx) +static bool test_aperfmperf(int idx) { return boot_cpu_has(X86_FEATURE_APERFMPERF); } -bool test_intel(int idx) +static bool test_intel(int idx) { if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL || boot_cpu_data.x86 != 6) From 79a8059d244e99454e474902e4325ee9b50e9178 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Mon, 21 Sep 2015 07:46:55 +0200 Subject: [PATCH 455/647] KVM: svm: do not call kvm_set_cr0 from init_vmcb kvm_set_cr0 may want to call kvm_zap_gfn_range and thus access the memslots array (SRCU protected). Using a mini SRCU critical section is ugly, and adding it to kvm_arch_vcpu_create doesn't work because the VMX vcpu_create callback calls synchronize_srcu. Fixes this lockdep splat: =============================== [ INFO: suspicious RCU usage. ] 4.3.0-rc1+ #1 Not tainted ------------------------------- include/linux/kvm_host.h:488 suspicious rcu_dereference_check() usage! other info that might help us debug this: rcu_scheduler_active = 1, debug_locks = 0 1 lock held by qemu-system-i38/17000: #0: (&(&kvm->mmu_lock)->rlock){+.+...}, at: kvm_zap_gfn_range+0x24/0x1a0 [kvm] [...] Call Trace: dump_stack+0x4e/0x84 lockdep_rcu_suspicious+0xfd/0x130 kvm_zap_gfn_range+0x188/0x1a0 [kvm] kvm_set_cr0+0xde/0x1e0 [kvm] init_vmcb+0x760/0xad0 [kvm_amd] svm_create_vcpu+0x197/0x250 [kvm_amd] kvm_arch_vcpu_create+0x47/0x70 [kvm] kvm_vm_ioctl+0x302/0x7e0 [kvm] ? __lock_is_held+0x51/0x70 ? __fget+0x101/0x210 do_vfs_ioctl+0x2f4/0x560 ? __fget_light+0x29/0x90 SyS_ioctl+0x4c/0x90 entry_SYSCALL_64_fastpath+0x16/0x73 Reported-by: Borislav Petkov Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 89173af53c84..94b7d15db3fc 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -202,6 +202,7 @@ module_param(npt, int, S_IRUGO); static int nested = true; module_param(nested, int, S_IRUGO); +static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0); static void svm_flush_tlb(struct kvm_vcpu *vcpu); static void svm_complete_interrupts(struct vcpu_svm *svm); @@ -1263,7 +1264,7 @@ static void init_vmcb(struct vcpu_svm *svm, bool init_event) * svm_set_cr0() sets PG and WP and clears NW and CD on save->cr0. * It also updates the guest-visible cr0 value. */ - (void)kvm_set_cr0(&svm->vcpu, X86_CR0_NW | X86_CR0_CD | X86_CR0_ET); + svm_set_cr0(&svm->vcpu, X86_CR0_NW | X86_CR0_CD | X86_CR0_ET); kvm_mmu_reset_context(&svm->vcpu); save->cr4 = X86_CR4_PAE; From 6fec21449a62702a582cecbb0b351363e039c95e Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Tue, 22 Sep 2015 23:02:14 +0200 Subject: [PATCH 456/647] KVM: x86: use correct page table format to check nested page table reserved bits Intel CPUID on AMD host or vice versa is a weird case, but it can happen. Handle it by checking the host CPU vendor instead of the guest's in reset_tdp_shadow_zero_bits_mask. For speed, the check uses the fact that Intel EPT has an X (executable) bit while AMD NPT has NX. Reported-by: Borislav Petkov Tested-by: Borislav Petkov Signed-off-by: Paolo Bonzini --- arch/x86/kvm/mmu.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 69088a1ba509..6f207637e437 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -3614,7 +3614,7 @@ static void __reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, struct rsvd_bits_validate *rsvd_check, int maxphyaddr, int level, bool nx, bool gbpages, - bool pse) + bool pse, bool amd) { u64 exb_bit_rsvd = 0; u64 gbpages_bit_rsvd = 0; @@ -3631,7 +3631,7 @@ __reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, * Non-leaf PML4Es and PDPEs reserve bit 8 (which would be the G bit for * leaf entries) on AMD CPUs only. */ - if (guest_cpuid_is_amd(vcpu)) + if (amd) nonleaf_bit8_rsvd = rsvd_bits(8, 8); switch (level) { @@ -3699,7 +3699,7 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, __reset_rsvds_bits_mask(vcpu, &context->guest_rsvd_check, cpuid_maxphyaddr(vcpu), context->root_level, context->nx, guest_cpuid_has_gbpages(vcpu), - is_pse(vcpu)); + is_pse(vcpu), guest_cpuid_is_amd(vcpu)); } static void @@ -3749,13 +3749,24 @@ static void reset_rsvds_bits_mask_ept(struct kvm_vcpu *vcpu, void reset_shadow_zero_bits_mask(struct kvm_vcpu *vcpu, struct kvm_mmu *context) { + /* + * Passing "true" to the last argument is okay; it adds a check + * on bit 8 of the SPTEs which KVM doesn't use anyway. + */ __reset_rsvds_bits_mask(vcpu, &context->shadow_zero_check, boot_cpu_data.x86_phys_bits, context->shadow_root_level, context->nx, - guest_cpuid_has_gbpages(vcpu), is_pse(vcpu)); + guest_cpuid_has_gbpages(vcpu), is_pse(vcpu), + true); } EXPORT_SYMBOL_GPL(reset_shadow_zero_bits_mask); +static inline bool boot_cpu_is_amd(void) +{ + WARN_ON_ONCE(!tdp_enabled); + return shadow_x_mask == 0; +} + /* * the direct page table on host, use as much mmu features as * possible, however, kvm currently does not do execution-protection. @@ -3764,11 +3775,11 @@ static void reset_tdp_shadow_zero_bits_mask(struct kvm_vcpu *vcpu, struct kvm_mmu *context) { - if (guest_cpuid_is_amd(vcpu)) + if (boot_cpu_is_amd()) __reset_rsvds_bits_mask(vcpu, &context->shadow_zero_check, boot_cpu_data.x86_phys_bits, context->shadow_root_level, false, - cpu_has_gbpages, true); + cpu_has_gbpages, true, true); else __reset_rsvds_bits_mask_ept(&context->shadow_zero_check, boot_cpu_data.x86_phys_bits, From 58c95070da3a504fbeca7939435bbb062cb96ea3 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Tue, 22 Sep 2015 10:15:59 +0200 Subject: [PATCH 457/647] KVM: x86: fix off-by-one in reserved bits check 29ecd6601904 ("KVM: x86: avoid uninitialized variable warning", 2015-09-06) introduced a not-so-subtle problem, which probably escaped review because it was not part of the patch context. Before the patch, leaf was always equal to iterator.level. After, it is equal to iterator.level - 1 in the call to is_shadow_zero_bits_set, and when is_shadow_zero_bits_set does another "-1" the check on reserved bits becomes incorrect. Using "iterator.level" in the call fixes this call trace: WARNING: CPU: 2 PID: 17000 at arch/x86/kvm/mmu.c:3385 handle_mmio_page_fault.part.93+0x1a/0x20 [kvm]() Modules linked in: tun sha256_ssse3 sha256_generic drbg binfmt_misc ipv6 vfat fat fuse dm_crypt dm_mod kvm_amd kvm crc32_pclmul aesni_intel aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd fam15h_power amd64_edac_mod k10temp edac_core amdkfd amd_iommu_v2 radeon acpi_cpufreq [...] Call Trace: dump_stack+0x4e/0x84 warn_slowpath_common+0x95/0xe0 warn_slowpath_null+0x1a/0x20 handle_mmio_page_fault.part.93+0x1a/0x20 [kvm] tdp_page_fault+0x231/0x290 [kvm] ? emulator_pio_in_out+0x6e/0xf0 [kvm] kvm_mmu_page_fault+0x36/0x240 [kvm] ? svm_set_cr0+0x95/0xc0 [kvm_amd] pf_interception+0xde/0x1d0 [kvm_amd] handle_exit+0x181/0xa70 [kvm_amd] ? kvm_arch_vcpu_ioctl_run+0x68b/0x1730 [kvm] kvm_arch_vcpu_ioctl_run+0x6f6/0x1730 [kvm] ? kvm_arch_vcpu_ioctl_run+0x68b/0x1730 [kvm] ? preempt_count_sub+0x9b/0xf0 ? mutex_lock_killable_nested+0x26f/0x490 ? preempt_count_sub+0x9b/0xf0 kvm_vcpu_ioctl+0x358/0x710 [kvm] ? __fget+0x5/0x210 ? __fget+0x101/0x210 do_vfs_ioctl+0x2f4/0x560 ? __fget_light+0x29/0x90 SyS_ioctl+0x4c/0x90 entry_SYSCALL_64_fastpath+0x16/0x73 ---[ end trace 37901c8686d84de6 ]--- Reported-by: Borislav Petkov Tested-by: Borislav Petkov Signed-off-by: Paolo Bonzini --- arch/x86/kvm/mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 6f207637e437..ff606f507913 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -3322,7 +3322,7 @@ walk_shadow_page_get_mmio_spte(struct kvm_vcpu *vcpu, u64 addr, u64 *sptep) break; reserved |= is_shadow_zero_bits_set(&vcpu->arch.mmu, spte, - leaf); + iterator.level); } walk_shadow_page_lockless_end(vcpu); From 920552b213e3dc832a874b4e7ba29ecddbab31bc Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Fri, 18 Sep 2015 12:34:53 +0200 Subject: [PATCH 458/647] KVM: disable halt_poll_ns as default for s390x We observed some performance degradation on s390x with dynamic halt polling. Until we can provide a proper fix, let's enable halt_poll_ns as default only for supported architectures. Architectures are now free to set their own halt_poll_ns default value. Signed-off-by: David Hildenbrand Signed-off-by: Paolo Bonzini --- arch/arm/include/asm/kvm_host.h | 1 + arch/arm64/include/asm/kvm_host.h | 1 + arch/mips/include/asm/kvm_host.h | 1 + arch/powerpc/include/asm/kvm_host.h | 1 + arch/s390/include/asm/kvm_host.h | 1 + arch/x86/include/asm/kvm_host.h | 1 + virt/kvm/kvm_main.c | 4 ++-- 7 files changed, 8 insertions(+), 2 deletions(-) diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h index 3df1e975f72a..c4072d9f32c7 100644 --- a/arch/arm/include/asm/kvm_host.h +++ b/arch/arm/include/asm/kvm_host.h @@ -33,6 +33,7 @@ #define KVM_PRIVATE_MEM_SLOTS 4 #define KVM_COALESCED_MMIO_PAGE_OFFSET 1 #define KVM_HAVE_ONE_REG +#define KVM_HALT_POLL_NS_DEFAULT 500000 #define KVM_VCPU_MAX_FEATURES 2 diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index 4562459456a6..ed039688c221 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -33,6 +33,7 @@ #define KVM_USER_MEM_SLOTS 32 #define KVM_PRIVATE_MEM_SLOTS 4 #define KVM_COALESCED_MMIO_PAGE_OFFSET 1 +#define KVM_HALT_POLL_NS_DEFAULT 500000 #include #include diff --git a/arch/mips/include/asm/kvm_host.h b/arch/mips/include/asm/kvm_host.h index 3a54dbca9f7e..5a1a882e0a75 100644 --- a/arch/mips/include/asm/kvm_host.h +++ b/arch/mips/include/asm/kvm_host.h @@ -61,6 +61,7 @@ #define KVM_PRIVATE_MEM_SLOTS 0 #define KVM_COALESCED_MMIO_PAGE_OFFSET 1 +#define KVM_HALT_POLL_NS_DEFAULT 500000 diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h index 195886a583ba..827a38d7a9db 100644 --- a/arch/powerpc/include/asm/kvm_host.h +++ b/arch/powerpc/include/asm/kvm_host.h @@ -44,6 +44,7 @@ #ifdef CONFIG_KVM_MMIO #define KVM_COALESCED_MMIO_PAGE_OFFSET 1 #endif +#define KVM_HALT_POLL_NS_DEFAULT 500000 /* These values are internal and can be increased later */ #define KVM_NR_IRQCHIPS 1 diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index 6ce4a0b7e8da..8ced426091e1 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -35,6 +35,7 @@ */ #define KVM_NR_IRQCHIPS 1 #define KVM_IRQCHIP_NUM_PINS 4096 +#define KVM_HALT_POLL_NS_DEFAULT 0 #define SIGP_CTRL_C 0x80 #define SIGP_CTRL_SCN_MASK 0x3f diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 349f80a82b82..2beee0382088 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -40,6 +40,7 @@ #define KVM_PIO_PAGE_OFFSET 1 #define KVM_COALESCED_MMIO_PAGE_OFFSET 2 +#define KVM_HALT_POLL_NS_DEFAULT 500000 #define KVM_IRQCHIP_NUM_PINS KVM_IOAPIC_NUM_PINS diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 04146a2e1d81..8db1d9361993 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -66,8 +66,8 @@ MODULE_AUTHOR("Qumranet"); MODULE_LICENSE("GPL"); -/* halt polling only reduces halt latency by 5-7 us, 500us is enough */ -static unsigned int halt_poll_ns = 500000; +/* Architectures should define their poll value according to the halt latency */ +static unsigned int halt_poll_ns = KVM_HALT_POLL_NS_DEFAULT; module_param(halt_poll_ns, uint, S_IRUGO | S_IWUSR); /* Default doubles per-vcpu halt_poll_ns. */ From 266fa2b22294909ddf6e7d2f8acfe07adf9fd978 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 24 Sep 2015 11:24:18 -0300 Subject: [PATCH 459/647] perf probe: Use existing routine to look for a kernel module by dso->short_name We have map_groups__find_by_name() to look at the list of modules that are in place for a given machine, so use it instead of traversing the machine dso list, which also includes DSOs for userspace. When merging the user and kernel DSO lists a bug was introduced where 'perf probe' stopped being able to add probes to modules using its short name: # perf probe -m usbnet --add usbnet_start_xmit usbnet_start_xmit is out of .text, skip it. Error: Failed to add events. # With this fix it works again: # perf probe -m usbnet --add usbnet_start_xmit Added new event: probe:usbnet_start_xmit (on usbnet_start_xmit in usbnet) You can now use it in all perf tools, such as: perf record -e probe:usbnet_start_xmit -aR sleep 1 # Reported-by: Wang Nan Acked-by: Masami Hiramatsu Cc: Adrian Hunter Cc: Borislav Petkov Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Namhyung Kim Cc: Stephane Eranian Fixes: 3d39ac538629 ("perf machine: No need to have two DSOs lists") Link: http://lkml.kernel.org/r/20150924015008.GE1897@kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/probe-event.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index eb5f18b75402..c6f9af78f6f5 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -270,12 +270,13 @@ static int kernel_get_module_dso(const char *module, struct dso **pdso) int ret = 0; if (module) { - list_for_each_entry(dso, &host_machine->dsos.head, node) { - if (!dso->kernel) - continue; - if (strncmp(dso->short_name + 1, module, - dso->short_name_len - 2) == 0) - goto found; + char module_name[128]; + + snprintf(module_name, sizeof(module_name), "[%s]", module); + map = map_groups__find_by_name(&host_machine->kmaps, MAP__FUNCTION, module_name); + if (map) { + dso = map->dso; + goto found; } pr_debug("Failed to find module %s.\n", module); return -ENOENT; From 597ee40722bf05195f91a41e88e15b79bdab152c Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 24 Sep 2015 13:05:21 +0300 Subject: [PATCH 460/647] perf intel-pt: Remove no_force_psb from documentation no_force_psb was dropped as a late change to the kernel driver. Consequently, remove it from the documentation. Signed-off-by: Adrian Hunter Cc: Jiri Olsa Link: http://lkml.kernel.org/r/1443089122-19082-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Documentation/intel-pt.txt | 15 --------------- 1 file changed, 15 deletions(-) diff --git a/tools/perf/Documentation/intel-pt.txt b/tools/perf/Documentation/intel-pt.txt index 4a0501d7a3b4..c94c9de3173e 100644 --- a/tools/perf/Documentation/intel-pt.txt +++ b/tools/perf/Documentation/intel-pt.txt @@ -364,21 +364,6 @@ cyc_thresh Specifies how frequently CYC packets are produced - see cyc CYC packets are not requested by default. -no_force_psb This is a driver option and is not in the IA32_RTIT_CTL MSR. - - It stops the driver resetting the byte count to zero whenever - enabling the trace (for example on context switches) which in - turn results in no PSB being forced. However some processors - will produce a PSB anyway. - - In any case, there is still a PSB when the trace is enabled for - the first time. - - no_force_psb can be used to slightly decrease the trace size but - may make it harder for the decoder to recover from errors. - - no_force_psb is not selected by default. - new snapshot option ------------------- From b5cabbcbd157a4bf5a92dfc85134999a3b55342d Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 24 Sep 2015 13:05:22 +0300 Subject: [PATCH 461/647] perf tools: Fix copying of /proc/kcore A copy of /proc/kcore containing the kernel text can be made to the buildid cache. e.g. perf buildid-cache -v -k /proc/kcore To workaround objdump limitations, a copy is also made when annotating against /proc/kcore. The copying process stops working from libelf about v1.62 onwards (the problem was found with v1.63). The cause is that a call to gelf_getphdr() in kcore__add_phdr() fails because additional validation has been added to gelf_getphdr(). The use of gelf_getphdr() is a misguided attempt to get default initialization of the Gelf_Phdr structure. That should not be necessary because every member of the Gelf_Phdr structure is subsequently assigned. So just remove the call to gelf_getphdr(). Similarly, a call to gelf_getehdr() in gelf_kcore__init() can be removed also. Committer notes: Note to stable@kernel.org, from Adrian in the cover letter for this patchkit: The "Fix copying of /proc/kcore" problem goes back to v3.13 if you think it is important enough for stable. Signed-off-by: Adrian Hunter Cc: Jiri Olsa Cc: stable@kernel.org Link: http://lkml.kernel.org/r/1443089122-19082-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/symbol-elf.c | 33 ++++++++++++--------------------- 1 file changed, 12 insertions(+), 21 deletions(-) diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index f78ea3dc4c08..475d88d0a1c9 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -1271,8 +1271,6 @@ out_close: static int kcore__init(struct kcore *kcore, char *filename, int elfclass, bool temp) { - GElf_Ehdr *ehdr; - kcore->elfclass = elfclass; if (temp) @@ -1289,9 +1287,7 @@ static int kcore__init(struct kcore *kcore, char *filename, int elfclass, if (!gelf_newehdr(kcore->elf, elfclass)) goto out_end; - ehdr = gelf_getehdr(kcore->elf, &kcore->ehdr); - if (!ehdr) - goto out_end; + memset(&kcore->ehdr, 0, sizeof(GElf_Ehdr)); return 0; @@ -1348,23 +1344,18 @@ static int kcore__copy_hdr(struct kcore *from, struct kcore *to, size_t count) static int kcore__add_phdr(struct kcore *kcore, int idx, off_t offset, u64 addr, u64 len) { - GElf_Phdr gphdr; - GElf_Phdr *phdr; + GElf_Phdr phdr = { + .p_type = PT_LOAD, + .p_flags = PF_R | PF_W | PF_X, + .p_offset = offset, + .p_vaddr = addr, + .p_paddr = 0, + .p_filesz = len, + .p_memsz = len, + .p_align = page_size, + }; - phdr = gelf_getphdr(kcore->elf, idx, &gphdr); - if (!phdr) - return -1; - - phdr->p_type = PT_LOAD; - phdr->p_flags = PF_R | PF_W | PF_X; - phdr->p_offset = offset; - phdr->p_vaddr = addr; - phdr->p_paddr = 0; - phdr->p_filesz = len; - phdr->p_memsz = len; - phdr->p_align = page_size; - - if (!gelf_update_phdr(kcore->elf, idx, phdr)) + if (!gelf_update_phdr(kcore->elf, idx, &phdr)) return -1; return 0; From bb6c96d72879fe1f674a804eb95b891def4ace61 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 24 Sep 2015 10:34:21 +0300 Subject: [PATCH 462/647] xprtrdma: Replace global lkey with lkey local to PD The core API has changed so that devices that do not have a global DMA lkey automatically create an mr, per-PD, and make that lkey available. The global DMA lkey interface is going away in favor of the per-PD DMA lkey. The per-PD DMA lkey is always available. Convert xprtrdma to use the device's per-PD DMA lkey for regbufs, no matter which memory registration scheme is in use. Signed-off-by: Chuck Lever Signed-off-by: Sagi Grimberg Cc: linux-nfs Acked-by: Anna Schumaker Signed-off-by: Doug Ledford --- net/sunrpc/xprtrdma/fmr_ops.c | 19 ------------------- net/sunrpc/xprtrdma/frwr_ops.c | 5 ----- net/sunrpc/xprtrdma/physical_ops.c | 10 +--------- net/sunrpc/xprtrdma/verbs.c | 2 +- net/sunrpc/xprtrdma/xprt_rdma.h | 1 - 5 files changed, 2 insertions(+), 35 deletions(-) diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c index cb25c89da623..f1e8dafbd507 100644 --- a/net/sunrpc/xprtrdma/fmr_ops.c +++ b/net/sunrpc/xprtrdma/fmr_ops.c @@ -39,25 +39,6 @@ static int fmr_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep, struct rpcrdma_create_data_internal *cdata) { - struct ib_device_attr *devattr = &ia->ri_devattr; - struct ib_mr *mr; - - /* Obtain an lkey to use for the regbufs, which are - * protected from remote access. - */ - if (devattr->device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY) { - ia->ri_dma_lkey = ia->ri_device->local_dma_lkey; - } else { - mr = ib_get_dma_mr(ia->ri_pd, IB_ACCESS_LOCAL_WRITE); - if (IS_ERR(mr)) { - pr_err("%s: ib_get_dma_mr for failed with %lX\n", - __func__, PTR_ERR(mr)); - return -ENOMEM; - } - ia->ri_dma_lkey = ia->ri_dma_mr->lkey; - ia->ri_dma_mr = mr; - } - return 0; } diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c index d6653f5d0830..5318951b3b53 100644 --- a/net/sunrpc/xprtrdma/frwr_ops.c +++ b/net/sunrpc/xprtrdma/frwr_ops.c @@ -189,11 +189,6 @@ frwr_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep, struct ib_device_attr *devattr = &ia->ri_devattr; int depth, delta; - /* Obtain an lkey to use for the regbufs, which are - * protected from remote access. - */ - ia->ri_dma_lkey = ia->ri_device->local_dma_lkey; - ia->ri_max_frmr_depth = min_t(unsigned int, RPCRDMA_MAX_DATA_SEGS, devattr->max_fast_reg_page_list_len); diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c index 72cf8b15bbb4..617b76f22154 100644 --- a/net/sunrpc/xprtrdma/physical_ops.c +++ b/net/sunrpc/xprtrdma/physical_ops.c @@ -23,7 +23,6 @@ static int physical_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep, struct rpcrdma_create_data_internal *cdata) { - struct ib_device_attr *devattr = &ia->ri_devattr; struct ib_mr *mr; /* Obtain an rkey to use for RPC data payloads. @@ -37,15 +36,8 @@ physical_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep, __func__, PTR_ERR(mr)); return -ENOMEM; } + ia->ri_dma_mr = mr; - - /* Obtain an lkey to use for regbufs. - */ - if (devattr->device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY) - ia->ri_dma_lkey = ia->ri_device->local_dma_lkey; - else - ia->ri_dma_lkey = ia->ri_dma_mr->lkey; - return 0; } diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index 682996779970..eb081ad05e33 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -1252,7 +1252,7 @@ rpcrdma_alloc_regbuf(struct rpcrdma_ia *ia, size_t size, gfp_t flags) goto out_free; iov->length = size; - iov->lkey = ia->ri_dma_lkey; + iov->lkey = ia->ri_pd->local_dma_lkey; rb->rg_size = size; rb->rg_owner = NULL; return rb; diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h index 02512221b8bc..c09414e6f91b 100644 --- a/net/sunrpc/xprtrdma/xprt_rdma.h +++ b/net/sunrpc/xprtrdma/xprt_rdma.h @@ -65,7 +65,6 @@ struct rpcrdma_ia { struct rdma_cm_id *ri_id; struct ib_pd *ri_pd; struct ib_mr *ri_dma_mr; - u32 ri_dma_lkey; struct completion ri_done; int ri_async_rc; unsigned int ri_max_frmr_depth; From 3cffd930171518821595839c5ce7036894ef0d74 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Thu, 24 Sep 2015 10:34:22 +0300 Subject: [PATCH 463/647] IB/iser: Add module parameter for always register memory This module parameter forces memory registration even for a continuous memory region. It is true by default as sending an all-physical rkey with remote permissions might be insecure. Signed-off-by: Sagi Grimberg Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/iser/iscsi_iser.c | 5 +++++ drivers/infiniband/ulp/iser/iscsi_iser.h | 1 + drivers/infiniband/ulp/iser/iser_memory.c | 18 ++++++++++++------ drivers/infiniband/ulp/iser/iser_verbs.c | 21 +++++++++++++-------- 4 files changed, 31 insertions(+), 14 deletions(-) diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index 1ace5d83a4d7..f58ff96b6cbb 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -97,6 +97,11 @@ unsigned int iser_max_sectors = ISER_DEF_MAX_SECTORS; module_param_named(max_sectors, iser_max_sectors, uint, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(max_sectors, "Max number of sectors in a single scsi command (default:1024"); +bool iser_always_reg = true; +module_param_named(always_register, iser_always_reg, bool, S_IRUGO); +MODULE_PARM_DESC(always_register, + "Always register memory, even for continuous memory regions (default:true)"); + bool iser_pi_enable = false; module_param_named(pi_enable, iser_pi_enable, bool, S_IRUGO); MODULE_PARM_DESC(pi_enable, "Enable T10-PI offload support (default:disabled)"); diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h b/drivers/infiniband/ulp/iser/iscsi_iser.h index 86f6583485ef..a5edd6ede692 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.h +++ b/drivers/infiniband/ulp/iser/iscsi_iser.h @@ -611,6 +611,7 @@ extern int iser_debug_level; extern bool iser_pi_enable; extern int iser_pi_guard; extern unsigned int iser_max_sectors; +extern bool iser_always_reg; int iser_assign_reg_ops(struct iser_device *device); diff --git a/drivers/infiniband/ulp/iser/iser_memory.c b/drivers/infiniband/ulp/iser/iser_memory.c index 2493cc748db8..4c46d67d37a1 100644 --- a/drivers/infiniband/ulp/iser/iser_memory.c +++ b/drivers/infiniband/ulp/iser/iser_memory.c @@ -803,11 +803,12 @@ static int iser_reg_prot_sg(struct iscsi_iser_task *task, struct iser_data_buf *mem, struct iser_fr_desc *desc, + bool use_dma_key, struct iser_mem_reg *reg) { struct iser_device *device = task->iser_conn->ib_conn.device; - if (mem->dma_nents == 1) + if (use_dma_key) return iser_reg_dma(device, mem, reg); return device->reg_ops->reg_mem(task, mem, &desc->pi_ctx->rsc, reg); @@ -817,11 +818,12 @@ static int iser_reg_data_sg(struct iscsi_iser_task *task, struct iser_data_buf *mem, struct iser_fr_desc *desc, + bool use_dma_key, struct iser_mem_reg *reg) { struct iser_device *device = task->iser_conn->ib_conn.device; - if (mem->dma_nents == 1) + if (use_dma_key) return iser_reg_dma(device, mem, reg); return device->reg_ops->reg_mem(task, mem, &desc->rsc, reg); @@ -836,14 +838,17 @@ int iser_reg_rdma_mem(struct iscsi_iser_task *task, struct iser_mem_reg *reg = &task->rdma_reg[dir]; struct iser_mem_reg *data_reg; struct iser_fr_desc *desc = NULL; + bool use_dma_key; int err; err = iser_handle_unaligned_buf(task, mem, dir); if (unlikely(err)) return err; - if (mem->dma_nents != 1 || - scsi_get_prot_op(task->sc) != SCSI_PROT_NORMAL) { + use_dma_key = (mem->dma_nents == 1 && !iser_always_reg && + scsi_get_prot_op(task->sc) == SCSI_PROT_NORMAL); + + if (!use_dma_key) { desc = device->reg_ops->reg_desc_get(ib_conn); reg->mem_h = desc; } @@ -853,7 +858,7 @@ int iser_reg_rdma_mem(struct iscsi_iser_task *task, else data_reg = &task->desc.data_reg; - err = iser_reg_data_sg(task, mem, desc, data_reg); + err = iser_reg_data_sg(task, mem, desc, use_dma_key, data_reg); if (unlikely(err)) goto err_reg; @@ -866,7 +871,8 @@ int iser_reg_rdma_mem(struct iscsi_iser_task *task, if (unlikely(err)) goto err_reg; - err = iser_reg_prot_sg(task, mem, desc, prot_reg); + err = iser_reg_prot_sg(task, mem, desc, + use_dma_key, prot_reg); if (unlikely(err)) goto err_reg; } diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index ae70cc1463ac..85132d867bc8 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -133,11 +133,15 @@ static int iser_create_device_ib_res(struct iser_device *device) (unsigned long)comp); } - device->mr = ib_get_dma_mr(device->pd, IB_ACCESS_LOCAL_WRITE | - IB_ACCESS_REMOTE_WRITE | - IB_ACCESS_REMOTE_READ); - if (IS_ERR(device->mr)) - goto dma_mr_err; + if (!iser_always_reg) { + int access = IB_ACCESS_LOCAL_WRITE | + IB_ACCESS_REMOTE_WRITE | + IB_ACCESS_REMOTE_READ; + + device->mr = ib_get_dma_mr(device->pd, access); + if (IS_ERR(device->mr)) + goto dma_mr_err; + } INIT_IB_EVENT_HANDLER(&device->event_handler, device->ib_device, iser_event_handler); @@ -147,7 +151,8 @@ static int iser_create_device_ib_res(struct iser_device *device) return 0; handler_err: - ib_dereg_mr(device->mr); + if (device->mr) + ib_dereg_mr(device->mr); dma_mr_err: for (i = 0; i < device->comps_used; i++) tasklet_kill(&device->comps[i].tasklet); @@ -173,7 +178,6 @@ comps_err: static void iser_free_device_ib_res(struct iser_device *device) { int i; - BUG_ON(device->mr == NULL); for (i = 0; i < device->comps_used; i++) { struct iser_comp *comp = &device->comps[i]; @@ -184,7 +188,8 @@ static void iser_free_device_ib_res(struct iser_device *device) } (void)ib_unregister_event_handler(&device->event_handler); - (void)ib_dereg_mr(device->mr); + if (device->mr) + (void)ib_dereg_mr(device->mr); ib_dealloc_pd(device->pd); kfree(device->comps); From c6790aa9f4fdc26b1246ba36da2fd749663beb65 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Thu, 24 Sep 2015 10:34:23 +0300 Subject: [PATCH 464/647] IB/mlx5: Remove support for IB_DEVICE_LOCAL_DMA_LKEY Commit 96249d70dd70 ("IB/core: Guarantee that a local_dma_lkey is available") allows ULPs that make use of the local dma key to keep working as before by allocating a DMA MR with local permissions and converted these consumers to use the MR associated with the PD rather then device->local_dma_lkey. ConnectIB has some known issues with memory registration using the local_dma_lkey (SEND, RDMA, RECV seems to work ok). Thus don't expose support for it (remove device->local_dma_lkey setting), and take advantage of the above commit such that no regression is introduced to working systems. The local_dma_lkey support will be restored in CX4 depending on FW capability query. Signed-off-by: Sagi Grimberg Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx5/main.c | 10 +-------- drivers/net/ethernet/mellanox/mlx5/core/fw.c | 22 -------------------- include/linux/mlx5/device.h | 11 ---------- include/linux/mlx5/driver.h | 1 - 4 files changed, 1 insertion(+), 43 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 41d6911e244e..0ab9625911a1 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -245,7 +245,6 @@ static int mlx5_ib_query_device(struct ib_device *ibdev, props->device_cap_flags |= IB_DEVICE_BAD_QKEY_CNTR; if (MLX5_CAP_GEN(mdev, apm)) props->device_cap_flags |= IB_DEVICE_AUTO_PATH_MIG; - props->device_cap_flags |= IB_DEVICE_LOCAL_DMA_LKEY; if (MLX5_CAP_GEN(mdev, xrc)) props->device_cap_flags |= IB_DEVICE_XRC; props->device_cap_flags |= IB_DEVICE_MEM_MGT_EXTENSIONS; @@ -1245,18 +1244,10 @@ static int create_dev_resources(struct mlx5_ib_resources *devr) struct ib_srq_init_attr attr; struct mlx5_ib_dev *dev; struct ib_cq_init_attr cq_attr = {.cqe = 1}; - u32 rsvd_lkey; int ret = 0; dev = container_of(devr, struct mlx5_ib_dev, devr); - ret = mlx5_core_query_special_context(dev->mdev, &rsvd_lkey); - if (ret) { - pr_err("Failed to query special context %d\n", ret); - return ret; - } - dev->ib_dev.local_dma_lkey = rsvd_lkey; - devr->p0 = mlx5_ib_alloc_pd(&dev->ib_dev, NULL, NULL); if (IS_ERR(devr->p0)) { ret = PTR_ERR(devr->p0); @@ -1418,6 +1409,7 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) strlcpy(dev->ib_dev.name, "mlx5_%d", IB_DEVICE_NAME_MAX); dev->ib_dev.owner = THIS_MODULE; dev->ib_dev.node_type = RDMA_NODE_IB_CA; + dev->ib_dev.local_dma_lkey = 0 /* not supported for now */; dev->num_ports = MLX5_CAP_GEN(mdev, num_ports); dev->ib_dev.phys_port_cnt = dev->num_ports; dev->ib_dev.num_comp_vectors = diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index aa0d5ffe92d8..9335e5ae18cc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -200,25 +200,3 @@ int mlx5_cmd_teardown_hca(struct mlx5_core_dev *dev) return err; } - -int mlx5_core_query_special_context(struct mlx5_core_dev *dev, u32 *rsvd_lkey) -{ - struct mlx5_cmd_query_special_contexts_mbox_in in; - struct mlx5_cmd_query_special_contexts_mbox_out out; - int err; - - memset(&in, 0, sizeof(in)); - memset(&out, 0, sizeof(out)); - in.hdr.opcode = cpu_to_be16(MLX5_CMD_OP_QUERY_SPECIAL_CONTEXTS); - err = mlx5_cmd_exec(dev, &in, sizeof(in), &out, sizeof(out)); - if (err) - return err; - - if (out.hdr.status) - err = mlx5_cmd_status_to_err(&out.hdr); - - *rsvd_lkey = be32_to_cpu(out.resd_lkey); - - return err; -} -EXPORT_SYMBOL(mlx5_core_query_special_context); diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 8eb3b19af2a4..250b1ff8b48d 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -402,17 +402,6 @@ struct mlx5_cmd_teardown_hca_mbox_out { u8 rsvd[8]; }; -struct mlx5_cmd_query_special_contexts_mbox_in { - struct mlx5_inbox_hdr hdr; - u8 rsvd[8]; -}; - -struct mlx5_cmd_query_special_contexts_mbox_out { - struct mlx5_outbox_hdr hdr; - __be32 dump_fill_mkey; - __be32 resd_lkey; -}; - struct mlx5_cmd_layout { u8 type; u8 rsvd0[3]; diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 27b53f9a24ad..8b6d6f2154a4 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -845,7 +845,6 @@ void *mlx5_get_protocol_dev(struct mlx5_core_dev *mdev, int protocol); int mlx5_register_interface(struct mlx5_interface *intf); void mlx5_unregister_interface(struct mlx5_interface *intf); int mlx5_core_query_vendor_id(struct mlx5_core_dev *mdev, u32 *vendor_id); -int mlx5_core_query_special_context(struct mlx5_core_dev *dev, u32 *rsvd_lkey); struct mlx5_profile { u64 mask; From 81fb5e26a9d05674c048803a20cb8f08a1b1c9b8 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Thu, 24 Sep 2015 10:34:24 +0300 Subject: [PATCH 465/647] IB/mlx5: Remove pa_lkey usages Since mlx5 driver cannot rely on registration using the reserved lkey (global_dma_lkey) it used to allocate a private physical address lkey for each allocated pd. Commit 96249d70dd70 ("IB/core: Guarantee that a local_dma_lkey is available") just does it in the core layer so we can go ahead and use that. Signed-off-by: Sagi Grimberg Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx5/main.c | 57 ---------------------------- drivers/infiniband/hw/mlx5/mlx5_ib.h | 2 - drivers/infiniband/hw/mlx5/qp.c | 4 +- 3 files changed, 1 insertion(+), 62 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 0ab9625911a1..f1ccd40beae9 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -794,53 +794,6 @@ static int mlx5_ib_mmap(struct ib_ucontext *ibcontext, struct vm_area_struct *vm return 0; } -static int alloc_pa_mkey(struct mlx5_ib_dev *dev, u32 *key, u32 pdn) -{ - struct mlx5_create_mkey_mbox_in *in; - struct mlx5_mkey_seg *seg; - struct mlx5_core_mr mr; - int err; - - in = kzalloc(sizeof(*in), GFP_KERNEL); - if (!in) - return -ENOMEM; - - seg = &in->seg; - seg->flags = MLX5_PERM_LOCAL_READ | MLX5_ACCESS_MODE_PA; - seg->flags_pd = cpu_to_be32(pdn | MLX5_MKEY_LEN64); - seg->qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); - seg->start_addr = 0; - - err = mlx5_core_create_mkey(dev->mdev, &mr, in, sizeof(*in), - NULL, NULL, NULL); - if (err) { - mlx5_ib_warn(dev, "failed to create mkey, %d\n", err); - goto err_in; - } - - kfree(in); - *key = mr.key; - - return 0; - -err_in: - kfree(in); - - return err; -} - -static void free_pa_mkey(struct mlx5_ib_dev *dev, u32 key) -{ - struct mlx5_core_mr mr; - int err; - - memset(&mr, 0, sizeof(mr)); - mr.key = key; - err = mlx5_core_destroy_mkey(dev->mdev, &mr); - if (err) - mlx5_ib_warn(dev, "failed to destroy mkey 0x%x\n", key); -} - static struct ib_pd *mlx5_ib_alloc_pd(struct ib_device *ibdev, struct ib_ucontext *context, struct ib_udata *udata) @@ -866,13 +819,6 @@ static struct ib_pd *mlx5_ib_alloc_pd(struct ib_device *ibdev, kfree(pd); return ERR_PTR(-EFAULT); } - } else { - err = alloc_pa_mkey(to_mdev(ibdev), &pd->pa_lkey, pd->pdn); - if (err) { - mlx5_core_dealloc_pd(to_mdev(ibdev)->mdev, pd->pdn); - kfree(pd); - return ERR_PTR(err); - } } return &pd->ibpd; @@ -883,9 +829,6 @@ static int mlx5_ib_dealloc_pd(struct ib_pd *pd) struct mlx5_ib_dev *mdev = to_mdev(pd->device); struct mlx5_ib_pd *mpd = to_mpd(pd); - if (!pd->uobject) - free_pa_mkey(mdev, mpd->pa_lkey); - mlx5_core_dealloc_pd(mdev->mdev, mpd->pdn); kfree(mpd); diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index bb8cda79e881..22123b79d550 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -103,7 +103,6 @@ static inline struct mlx5_ib_ucontext *to_mucontext(struct ib_ucontext *ibuconte struct mlx5_ib_pd { struct ib_pd ibpd; u32 pdn; - u32 pa_lkey; }; /* Use macros here so that don't have to duplicate @@ -213,7 +212,6 @@ struct mlx5_ib_qp { int uuarn; int create_type; - u32 pa_lkey; /* Store signature errors */ bool signature_en; diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index c745c6c5e10d..6f521a3418e8 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -925,8 +925,6 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd, err = create_kernel_qp(dev, init_attr, qp, &in, &inlen); if (err) mlx5_ib_dbg(dev, "err %d\n", err); - else - qp->pa_lkey = to_mpd(pd)->pa_lkey; } if (err) @@ -2045,7 +2043,7 @@ static void set_frwr_pages(struct mlx5_wqe_data_seg *dseg, mfrpl->mapped_page_list[i] = cpu_to_be64(page_list[i] | perm); dseg->addr = cpu_to_be64(mfrpl->map); dseg->byte_count = cpu_to_be32(ALIGN(sizeof(u64) * wr->wr.fast_reg.page_list_len, 64)); - dseg->lkey = cpu_to_be32(pd->pa_lkey); + dseg->lkey = cpu_to_be32(pd->ibpd.local_dma_lkey); } static __be32 send_ieth(struct ib_send_wr *wr) From 94c4554ba07adbdde396748ee7ae01e86cf2d8d7 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 25 Sep 2015 16:30:08 +0100 Subject: [PATCH 466/647] KEYS: Fix race between key destruction and finding a keyring by name There appears to be a race between: (1) key_gc_unused_keys() which frees key->security and then calls keyring_destroy() to unlink the name from the name list (2) find_keyring_by_name() which calls key_permission(), thus accessing key->security, on a key before checking to see whether the key usage is 0 (ie. the key is dead and might be cleaned up). Fix this by calling ->destroy() before cleaning up the core key data - including key->security. Reported-by: Petr Matousek Signed-off-by: David Howells --- security/keys/gc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/security/keys/gc.c b/security/keys/gc.c index c7952375ac53..39eac1fd5706 100644 --- a/security/keys/gc.c +++ b/security/keys/gc.c @@ -134,6 +134,10 @@ static noinline void key_gc_unused_keys(struct list_head *keys) kdebug("- %u", key->serial); key_check(key); + /* Throw away the key data */ + if (key->type->destroy) + key->type->destroy(key); + security_key_free(key); /* deal with the user's key tracking and quota */ @@ -148,10 +152,6 @@ static noinline void key_gc_unused_keys(struct list_head *keys) if (test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) atomic_dec(&key->user->nikeys); - /* now throw away the key memory */ - if (key->type->destroy) - key->type->destroy(key); - key_user_put(key->user); kfree(key->description); From 292c6091353475d94e2cfb49c29906e88ee967ba Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 25 Sep 2015 16:31:45 +0100 Subject: [PATCH 467/647] KEYS: Remove unnecessary header #inclusions from extract-cert.c Remove headers #included unnecessarily from extract-cert.c lest they cause compilation of the tool to fail against an older OpenSSL library. Signed-off-by: David Howells Acked-by: David Woodhouse --- scripts/extract-cert.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/scripts/extract-cert.c b/scripts/extract-cert.c index 6ce5945a0b89..b071bf476fea 100644 --- a/scripts/extract-cert.c +++ b/scripts/extract-cert.c @@ -17,13 +17,9 @@ #include #include #include -#include #include -#include #include -#include #include -#include #include #include From e7c87bef7de2417b219d4dbfe8d33a0098a8df54 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 25 Sep 2015 16:31:46 +0100 Subject: [PATCH 468/647] X.509: Don't strip leading 00's from key ID when constructing key description Don't strip leading zeros from the crypto key ID when using it to construct the struct key description as the signature in kernels up to and including 4.2 matched this aspect of the key. This means that 1 in 256 keys won't actually match if their key ID begins with 00. The key ID is stored in the module signature as binary and so must be converted to text in order to invoke request_key() - but it isn't stripped at this point. Something like this is likely to be observed in dmesg when the key is loaded: [ 1.572423] Loaded X.509 cert 'Build time autogenerated kernel key: 62a7c3d2da278be024da4af8652c071f3fea33' followed by this when we try and use it: [ 1.646153] Request for unknown module key 'Build time autogenerated kernel key: 0062a7c3d2da278be024da4af8652c071f3fea33' err -11 The 'Loaded' line should show an extra '00' on the front of the hex string. This problem should not affect 4.3-rc1 and onwards because there the key should be matched on one of its auxiliary identities rather than the key struct's description string. Reported-by: Arjan van de Ven Reported-by: Andy Whitcroft Signed-off-by: David Howells --- crypto/asymmetric_keys/x509_public_key.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/crypto/asymmetric_keys/x509_public_key.c b/crypto/asymmetric_keys/x509_public_key.c index 6d88dd15c98d..197096632412 100644 --- a/crypto/asymmetric_keys/x509_public_key.c +++ b/crypto/asymmetric_keys/x509_public_key.c @@ -332,10 +332,6 @@ static int x509_key_preparse(struct key_preparsed_payload *prep) srlen = cert->raw_serial_size; q = cert->raw_serial; } - if (srlen > 1 && *q == 0) { - srlen--; - q++; - } ret = -ENOMEM; desc = kmalloc(sulen + 2 + srlen * 2 + 1, GFP_KERNEL); From 283e8ba2dfde54f8f27d7d0f459a07de79a39d55 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 25 Sep 2015 16:31:46 +0100 Subject: [PATCH 469/647] MODSIGN: Change from CMS to PKCS#7 signing if the openssl is too old The sign-file.c program actually uses CMS rather than PKCS#7 to sign a file since that allows the target X.509 certificate to be specified by subjectKeyId rather than by issuer + serialNumber. However, older versions of the OpenSSL crypto library (such as may be found in CentOS 5.11) don't support CMS. Assume everything prior to OpenSSL-1.0.0 doesn't support CMS and switch to using PKCS#7 in that case. Further, the pre-1.0.0 OpenSSL only supports PKCS#7 signing with SHA1, so give an error from the sign-file script if the caller requests anything other than SHA1. The compiler gives the following error with an OpenSSL crypto library that's too old: HOSTCC scripts/sign-file scripts/sign-file.c:23:25: fatal error: openssl/cms.h: No such file or directory #include Reported-by: Vinson Lee Signed-off-by: David Howells Acked-by: David Woodhouse --- Documentation/Changes | 2 +- scripts/sign-file.c | 94 +++++++++++++++++++++++++++++++++++-------- 2 files changed, 78 insertions(+), 18 deletions(-) diff --git a/Documentation/Changes b/Documentation/Changes index 6d8863004858..f447f0516f07 100644 --- a/Documentation/Changes +++ b/Documentation/Changes @@ -43,7 +43,7 @@ o udev 081 # udevd --version o grub 0.93 # grub --version || grub-install --version o mcelog 0.6 # mcelog --version o iptables 1.4.2 # iptables -V -o openssl & libcrypto 1.0.1k # openssl version +o openssl & libcrypto 1.0.0 # openssl version Kernel compilation diff --git a/scripts/sign-file.c b/scripts/sign-file.c index c3899ca4811c..250a7a645033 100755 --- a/scripts/sign-file.c +++ b/scripts/sign-file.c @@ -20,13 +20,34 @@ #include #include #include +#include #include #include #include -#include #include #include +/* + * Use CMS if we have openssl-1.0.0 or newer available - otherwise we have to + * assume that it's not available and its header file is missing and that we + * should use PKCS#7 instead. Switching to the older PKCS#7 format restricts + * the options we have on specifying the X.509 certificate we want. + * + * Further, older versions of OpenSSL don't support manually adding signers to + * the PKCS#7 message so have to accept that we get a certificate included in + * the signature message. Nor do such older versions of OpenSSL support + * signing with anything other than SHA1 - so we're stuck with that if such is + * the case. + */ +#if OPENSSL_VERSION_NUMBER < 0x10000000L +#define USE_PKCS7 +#endif +#ifndef USE_PKCS7 +#include +#else +#include +#endif + struct module_signature { uint8_t algo; /* Public-key crypto algorithm [0] */ uint8_t hash; /* Digest algorithm [0] */ @@ -110,30 +131,42 @@ int main(int argc, char **argv) struct module_signature sig_info = { .id_type = PKEY_ID_PKCS7 }; char *hash_algo = NULL; char *private_key_name, *x509_name, *module_name, *dest_name; - bool save_cms = false, replace_orig; + bool save_sig = false, replace_orig; bool sign_only = false; unsigned char buf[4096]; - unsigned long module_size, cms_size; - unsigned int use_keyid = 0, use_signed_attrs = CMS_NOATTR; + unsigned long module_size, sig_size; + unsigned int use_signed_attrs; const EVP_MD *digest_algo; EVP_PKEY *private_key; +#ifndef USE_PKCS7 CMS_ContentInfo *cms; + unsigned int use_keyid = 0; +#else + PKCS7 *pkcs7; +#endif X509 *x509; BIO *b, *bd = NULL, *bm; int opt, n; - OpenSSL_add_all_algorithms(); ERR_load_crypto_strings(); ERR_clear_error(); key_pass = getenv("KBUILD_SIGN_PIN"); +#ifndef USE_PKCS7 + use_signed_attrs = CMS_NOATTR; +#else + use_signed_attrs = PKCS7_NOATTR; +#endif + do { opt = getopt(argc, argv, "dpk"); switch (opt) { - case 'p': save_cms = true; break; - case 'd': sign_only = true; save_cms = true; break; + case 'p': save_sig = true; break; + case 'd': sign_only = true; save_sig = true; break; +#ifndef USE_PKCS7 case 'k': use_keyid = CMS_USE_KEYID; break; +#endif case -1: break; default: format(); } @@ -157,6 +190,14 @@ int main(int argc, char **argv) replace_orig = true; } +#ifdef USE_PKCS7 + if (strcmp(hash_algo, "sha1") != 0) { + fprintf(stderr, "sign-file: %s only supports SHA1 signing\n", + OPENSSL_VERSION_TEXT); + exit(3); + } +#endif + /* Read the private key and the X.509 cert the PKCS#7 message * will point to. */ @@ -213,7 +254,8 @@ int main(int argc, char **argv) bm = BIO_new_file(module_name, "rb"); ERR(!bm, "%s", module_name); - /* Load the CMS message from the digest buffer. */ +#ifndef USE_PKCS7 + /* Load the signature message from the digest buffer. */ cms = CMS_sign(NULL, NULL, NULL, NULL, CMS_NOCERTS | CMS_PARTIAL | CMS_BINARY | CMS_DETACHED | CMS_STREAM); ERR(!cms, "CMS_sign"); @@ -221,17 +263,31 @@ int main(int argc, char **argv) ERR(!CMS_add1_signer(cms, x509, private_key, digest_algo, CMS_NOCERTS | CMS_BINARY | CMS_NOSMIMECAP | use_keyid | use_signed_attrs), - "CMS_sign_add_signer"); + "CMS_add1_signer"); ERR(CMS_final(cms, bm, NULL, CMS_NOCERTS | CMS_BINARY) < 0, "CMS_final"); - if (save_cms) { - char *cms_name; +#else + pkcs7 = PKCS7_sign(x509, private_key, NULL, bm, + PKCS7_NOCERTS | PKCS7_BINARY | + PKCS7_DETACHED | use_signed_attrs); + ERR(!pkcs7, "PKCS7_sign"); +#endif - ERR(asprintf(&cms_name, "%s.p7s", module_name) < 0, "asprintf"); - b = BIO_new_file(cms_name, "wb"); - ERR(!b, "%s", cms_name); - ERR(i2d_CMS_bio_stream(b, cms, NULL, 0) < 0, "%s", cms_name); + if (save_sig) { + char *sig_file_name; + + ERR(asprintf(&sig_file_name, "%s.p7s", module_name) < 0, + "asprintf"); + b = BIO_new_file(sig_file_name, "wb"); + ERR(!b, "%s", sig_file_name); +#ifndef USE_PKCS7 + ERR(i2d_CMS_bio_stream(b, cms, NULL, 0) < 0, + "%s", sig_file_name); +#else + ERR(i2d_PKCS7_bio(b, pkcs7) < 0, + "%s", sig_file_name); +#endif BIO_free(b); } @@ -247,9 +303,13 @@ int main(int argc, char **argv) ERR(n < 0, "%s", module_name); module_size = BIO_number_written(bd); +#ifndef USE_PKCS7 ERR(i2d_CMS_bio_stream(bd, cms, NULL, 0) < 0, "%s", dest_name); - cms_size = BIO_number_written(bd) - module_size; - sig_info.sig_len = htonl(cms_size); +#else + ERR(i2d_PKCS7_bio(bd, pkcs7) < 0, "%s", dest_name); +#endif + sig_size = BIO_number_written(bd) - module_size; + sig_info.sig_len = htonl(sig_size); ERR(BIO_write(bd, &sig_info, sizeof(sig_info)) < 0, "%s", dest_name); ERR(BIO_write(bd, magic_number, sizeof(magic_number) - 1) < 0, "%s", dest_name); From bd99b2e05c4df2a428e5c9dd338289089d0e26df Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 24 Sep 2015 12:00:05 -0500 Subject: [PATCH 470/647] IB/ipoib: Expire sendonly multicast joins On neighbor expiration, check to see if the neighbor was actually a sendonly multicast join, and if so, leave the multicast group as we expire the neighbor. Signed-off-by: Christoph Lameter Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/ipoib/ipoib.h | 2 ++ drivers/infiniband/ulp/ipoib/ipoib_main.c | 18 ++++++++++++++++++ drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 4 ++-- 3 files changed, 22 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index ca2873698d75..4ff4e5290354 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -548,6 +548,8 @@ void ipoib_path_iter_read(struct ipoib_path_iter *iter, int ipoib_mcast_attach(struct net_device *dev, u16 mlid, union ib_gid *mgid, int set_qkey); +int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast); +struct ipoib_mcast *__ipoib_mcast_find(struct net_device *dev, void *mgid); int ipoib_init_qp(struct net_device *dev); int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 36536ce5a3e2..f74316e679d2 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -1149,6 +1149,9 @@ static void __ipoib_reap_neigh(struct ipoib_dev_priv *priv) unsigned long dt; unsigned long flags; int i; + LIST_HEAD(remove_list); + struct ipoib_mcast *mcast, *tmcast; + struct net_device *dev = priv->dev; if (test_bit(IPOIB_STOP_NEIGH_GC, &priv->flags)) return; @@ -1176,6 +1179,19 @@ static void __ipoib_reap_neigh(struct ipoib_dev_priv *priv) lockdep_is_held(&priv->lock))) != NULL) { /* was the neigh idle for two GC periods */ if (time_after(neigh_obsolete, neigh->alive)) { + u8 *mgid = neigh->daddr + 4; + + /* Is this multicast ? */ + if (*mgid == 0xff) { + mcast = __ipoib_mcast_find(dev, mgid); + + if (mcast && test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) { + list_del(&mcast->list); + rb_erase(&mcast->rb_node, &priv->multicast_tree); + list_add_tail(&mcast->list, &remove_list); + } + } + rcu_assign_pointer(*np, rcu_dereference_protected(neigh->hnext, lockdep_is_held(&priv->lock))); @@ -1191,6 +1207,8 @@ static void __ipoib_reap_neigh(struct ipoib_dev_priv *priv) out_unlock: spin_unlock_irqrestore(&priv->lock, flags); + list_for_each_entry_safe(mcast, tmcast, &remove_list, list) + ipoib_mcast_leave(dev, mcast); } static void ipoib_reap_neigh(struct work_struct *work) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 09a1748f9d13..33131001fa24 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -153,7 +153,7 @@ static struct ipoib_mcast *ipoib_mcast_alloc(struct net_device *dev, return mcast; } -static struct ipoib_mcast *__ipoib_mcast_find(struct net_device *dev, void *mgid) +struct ipoib_mcast *__ipoib_mcast_find(struct net_device *dev, void *mgid) { struct ipoib_dev_priv *priv = netdev_priv(dev); struct rb_node *n = priv->multicast_tree.rb_node; @@ -675,7 +675,7 @@ int ipoib_mcast_stop_thread(struct net_device *dev) return 0; } -static int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast) +int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast) { struct ipoib_dev_priv *priv = netdev_priv(dev); int ret = 0; From c3852ab0e606212de523c1fb1e15adbf9f431619 Mon Sep 17 00:00:00 2001 From: Doug Ledford Date: Fri, 25 Sep 2015 14:35:01 -0400 Subject: [PATCH 471/647] IB/ipoib: Make sendonly multicast joins create the mcast group Since IPoIB should, as much as possible, emulate how multicast sends work on Ethernet for regular TCP/IP apps, there should be no requirement to subscribe to a multicast group before your sends are properly sent. However, due to the difference in how multicast is handled on InfiniBand, we must join the appropriate multicast group before we can send to it. Previously we tried not to trigger the auto-create feature of the subnet manager when doing this because we didn't have tracking of these sendonly groups and the auto-creation might never get undone. The previous patch added timing to these sendonly joins and allows us to leave them after a reasonable idle expiration time. So supply all of the information needed to auto-create group. Signed-off-by: Doug Ledford --- .../infiniband/ulp/ipoib/ipoib_multicast.c | 22 ++++++++++--------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 33131001fa24..136cbefe00f8 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -508,17 +508,19 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast) rec.hop_limit = priv->broadcast->mcmember.hop_limit; /* - * Historically Linux IPoIB has never properly supported SEND - * ONLY join. It emulated it by not providing all the required - * attributes, which is enough to prevent group creation and - * detect if there are full members or not. A major problem - * with supporting SEND ONLY is detecting when the group is - * auto-destroyed as IPoIB will cache the MLID.. + * Send-only IB Multicast joins do not work at the core + * IB layer yet, so we can't use them here. However, + * we are emulating an Ethernet multicast send, which + * does not require a multicast subscription and will + * still send properly. The most appropriate thing to + * do is to create the group if it doesn't exist as that + * most closely emulates the behavior, from a user space + * application perspecitive, of Ethernet multicast + * operation. For now, we do a full join, maybe later + * when the core IB layers support send only joins we + * will use them. */ -#if 1 - if (test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) - comp_mask &= ~IB_SA_MCMEMBER_REC_TRAFFIC_CLASS; -#else +#if 0 if (test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) rec.join_state = 4; #endif From 357cd64c18404309aabefb82019b12773de31a12 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Sep 2015 00:07:17 +0100 Subject: [PATCH 472/647] phy: marvell: add link partner advertised modes Read the standard link partner advertisment registers and store it in phydev->lp_advertising, so ethtool can report this information to userspace via ethtool. Zero it as per genphy if autonegotiation is disabled. Tested with a Marvell 88E1512 PHY. Signed-off-by: Russell King Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/marvell.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index e6897b6a8a53..5de8d5827536 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -785,6 +785,7 @@ static int marvell_read_status(struct phy_device *phydev) int adv; int err; int lpa; + int lpagb; int status = 0; /* Update the link, but return if there @@ -802,10 +803,17 @@ static int marvell_read_status(struct phy_device *phydev) if (lpa < 0) return lpa; + lpagb = phy_read(phydev, MII_STAT1000); + if (lpagb < 0) + return lpagb; + adv = phy_read(phydev, MII_ADVERTISE); if (adv < 0) return adv; + phydev->lp_advertising = mii_stat1000_to_ethtool_lpa_t(lpagb) | + mii_lpa_to_ethtool_lpa_t(lpa); + lpa &= adv; if (status & MII_M1011_PHY_STATUS_FULLDUPLEX) @@ -853,6 +861,7 @@ static int marvell_read_status(struct phy_device *phydev) phydev->speed = SPEED_10; phydev->pause = phydev->asym_pause = 0; + phydev->lp_advertising = 0; } return 0; From 21343ac21ec7d871e94e98e288f3398a4207d9c0 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Thu, 24 Sep 2015 15:46:53 +0530 Subject: [PATCH 473/647] net: via/Kconfig: GENERIC_PCI_IOMAP required if PCI not selected The builds of allmodconfig of avr32 is failing with: drivers/net/ethernet/via/via-rhine.c:1098:2: error: implicit declaration of function 'pci_iomap' [-Werror=implicit-function-declaration] drivers/net/ethernet/via/via-rhine.c:1119:2: error: implicit declaration of function 'pci_iounmap' [-Werror=implicit-function-declaration] The generic empty pci_iomap and pci_iounmap is used only if CONFIG_PCI is not defined and CONFIG_GENERIC_PCI_IOMAP is defined. Add GENERIC_PCI_IOMAP in the dependency list for VIA_RHINE as we are getting build failure when CONFIG_PCI and CONFIG_GENERIC_PCI_IOMAP both are not defined. Signed-off-by: Sudip Mukherjee Signed-off-by: David S. Miller --- drivers/net/ethernet/via/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/via/Kconfig b/drivers/net/ethernet/via/Kconfig index 2f1264b882b9..d3d094742a7e 100644 --- a/drivers/net/ethernet/via/Kconfig +++ b/drivers/net/ethernet/via/Kconfig @@ -17,7 +17,7 @@ if NET_VENDOR_VIA config VIA_RHINE tristate "VIA Rhine support" - depends on (PCI || OF_IRQ) + depends on PCI || (OF_IRQ && GENERIC_PCI_IOMAP) depends on HAS_DMA select CRC32 select MII From 58a89ecaca53736aa465170530acea4f8be34ab4 Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Thu, 24 Sep 2015 12:54:01 +0200 Subject: [PATCH 474/647] ppp: fix lockdep splat in ppp_dev_uninit() ppp_dev_uninit() locks all_ppp_mutex while under rtnl mutex protection. ppp_create_interface() must then lock these mutexes in that same order to avoid possible deadlock. [ 120.880011] ====================================================== [ 120.880011] [ INFO: possible circular locking dependency detected ] [ 120.880011] 4.2.0 #1 Not tainted [ 120.880011] ------------------------------------------------------- [ 120.880011] ppp-apitest/15827 is trying to acquire lock: [ 120.880011] (&pn->all_ppp_mutex){+.+.+.}, at: [] ppp_dev_uninit+0x64/0xb0 [ppp_generic] [ 120.880011] [ 120.880011] but task is already holding lock: [ 120.880011] (rtnl_mutex){+.+.+.}, at: [] rtnl_lock+0x12/0x14 [ 120.880011] [ 120.880011] which lock already depends on the new lock. [ 120.880011] [ 120.880011] [ 120.880011] the existing dependency chain (in reverse order) is: [ 120.880011] [ 120.880011] -> #1 (rtnl_mutex){+.+.+.}: [ 120.880011] [] lock_acquire+0xcf/0x10e [ 120.880011] [] mutex_lock_nested+0x56/0x341 [ 120.880011] [] rtnl_lock+0x12/0x14 [ 120.880011] [] register_netdev+0x11/0x27 [ 120.880011] [] ppp_ioctl+0x289/0xc98 [ppp_generic] [ 120.880011] [] do_vfs_ioctl+0x4ea/0x532 [ 120.880011] [] SyS_ioctl+0x4e/0x7d [ 120.880011] [] entry_SYSCALL_64_fastpath+0x12/0x6f [ 120.880011] [ 120.880011] -> #0 (&pn->all_ppp_mutex){+.+.+.}: [ 120.880011] [] __lock_acquire+0xb07/0xe76 [ 120.880011] [] lock_acquire+0xcf/0x10e [ 120.880011] [] mutex_lock_nested+0x56/0x341 [ 120.880011] [] ppp_dev_uninit+0x64/0xb0 [ppp_generic] [ 120.880011] [] rollback_registered_many+0x19e/0x252 [ 120.880011] [] rollback_registered+0x29/0x38 [ 120.880011] [] unregister_netdevice_queue+0x6a/0x77 [ 120.880011] [] ppp_release+0x42/0x79 [ppp_generic] [ 120.880011] [] __fput+0xec/0x192 [ 120.880011] [] ____fput+0x9/0xb [ 120.880011] [] task_work_run+0x66/0x80 [ 120.880011] [] prepare_exit_to_usermode+0x8c/0xa7 [ 120.880011] [] syscall_return_slowpath+0xe4/0x104 [ 120.880011] [] int_ret_from_sys_call+0x25/0x9f [ 120.880011] [ 120.880011] other info that might help us debug this: [ 120.880011] [ 120.880011] Possible unsafe locking scenario: [ 120.880011] [ 120.880011] CPU0 CPU1 [ 120.880011] ---- ---- [ 120.880011] lock(rtnl_mutex); [ 120.880011] lock(&pn->all_ppp_mutex); [ 120.880011] lock(rtnl_mutex); [ 120.880011] lock(&pn->all_ppp_mutex); [ 120.880011] [ 120.880011] *** DEADLOCK *** Fixes: 8cb775bc0a34 ("ppp: fix device unregistration upon netns deletion") Reported-by: Sedat Dilek Tested-by: Sedat Dilek Signed-off-by: Guillaume Nault Signed-off-by: David S. Miller --- drivers/net/ppp/ppp_generic.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c index 0481daf9201a..ed00446759b2 100644 --- a/drivers/net/ppp/ppp_generic.c +++ b/drivers/net/ppp/ppp_generic.c @@ -2755,6 +2755,7 @@ static struct ppp *ppp_create_interface(struct net *net, int unit, */ dev_net_set(dev, net); + rtnl_lock(); mutex_lock(&pn->all_ppp_mutex); if (unit < 0) { @@ -2785,7 +2786,7 @@ static struct ppp *ppp_create_interface(struct net *net, int unit, ppp->file.index = unit; sprintf(dev->name, "ppp%d", unit); - ret = register_netdev(dev); + ret = register_netdevice(dev); if (ret != 0) { unit_put(&pn->units_idr, unit); netdev_err(ppp->dev, "PPP: couldn't register device %s (%d)\n", @@ -2797,6 +2798,7 @@ static struct ppp *ppp_create_interface(struct net *net, int unit, atomic_inc(&ppp_unit_count); mutex_unlock(&pn->all_ppp_mutex); + rtnl_unlock(); *retp = 0; return ppp; From e1a2d49cd5ef551c51be95cc037033e9e582b0cd Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Thu, 24 Sep 2015 12:28:44 -0700 Subject: [PATCH 475/647] PM / OPP: Fix typo modifcation -> modification Reported-by: Viresh Kumar Signed-off-by: Stephen Boyd Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki --- drivers/base/power/opp.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/base/power/opp.c b/drivers/base/power/opp.c index 1194669c2bb5..7ae7cd990fbf 100644 --- a/drivers/base/power/opp.c +++ b/drivers/base/power/opp.c @@ -1070,7 +1070,7 @@ EXPORT_SYMBOL_GPL(dev_pm_opp_add); * share a common logic which is isolated here. * * Return: -EINVAL for bad pointers, -ENOMEM if no memory available for the - * copy operation, returns 0 if no modifcation was done OR modification was + * copy operation, returns 0 if no modification was done OR modification was * successful. * * Locking: The internal device_opp and opp structures are RCU protected. @@ -1158,7 +1158,7 @@ unlock: * mutex locking or synchronize_rcu() blocking calls cannot be used. * * Return: -EINVAL for bad pointers, -ENOMEM if no memory available for the - * copy operation, returns 0 if no modifcation was done OR modification was + * copy operation, returns 0 if no modification was done OR modification was * successful. */ int dev_pm_opp_enable(struct device *dev, unsigned long freq) @@ -1184,7 +1184,7 @@ EXPORT_SYMBOL_GPL(dev_pm_opp_enable); * mutex locking or synchronize_rcu() blocking calls cannot be used. * * Return: -EINVAL for bad pointers, -ENOMEM if no memory available for the - * copy operation, returns 0 if no modifcation was done OR modification was + * copy operation, returns 0 if no modification was done OR modification was * successful. */ int dev_pm_opp_disable(struct device *dev, unsigned long freq) From 15b94fa32a422f4b97dc34e4b7060ec83d10bee5 Mon Sep 17 00:00:00 2001 From: Lv Zheng Date: Thu, 24 Sep 2015 14:54:40 +0800 Subject: [PATCH 476/647] ACPI / EC: Fix a memory leak issue in acpi_ec_query() When query handler is not found, "result" is actually stil 0, and "struct acpi_ec_query" is not NULL, so the deletion code of "struct acpi_ec_query" at the end of the function cannot be invoked. As a consequence, memory leak can be observed. The issue is introduced by this commit: Commit: 02b771b64b73226052d6e731a0987db3b47281e9 Subject: ACPI / EC: Fix an issue caused by the serialized _Qxx This patch fixes such memory leakage. Fixes: 02b771b64b73 (ACPI / EC: Fix an issue caused by the serialized _Qxx evaluations) Signed-off-by: Lv Zheng Signed-off-by: Rafael J. Wysocki --- drivers/acpi/ec.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index 2614a839c60d..42c66b64c12c 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -1044,8 +1044,10 @@ static int acpi_ec_query(struct acpi_ec *ec, u8 *data) goto err_exit; mutex_lock(&ec->mutex); + result = -ENODATA; list_for_each_entry(handler, &ec->list, node) { if (value == handler->query_bit) { + result = 0; q->handler = acpi_ec_get_query_handler(handler); ec_dbg_evt("Query(0x%02x) scheduled", q->handler->query_bit); From 5ebc76035303016ec41bb752bec156ea9fde7c34 Mon Sep 17 00:00:00 2001 From: Jiang Liu Date: Thu, 17 Sep 2015 14:02:45 +0800 Subject: [PATCH 477/647] ACPI, PCI, irq: Do not share PCI IRQ with ISA IRQ Avoid IRQs occupied by ISA IRQs when allocating IRQs for PCI link devices, otherwise it may cause interrupt storm due to incompatible pin attributes. This issue was triggered on a KVM virtual machine, which 1) uses IRQ9 for SCI in high level mode. 2) defines an PCI interrupt link device (LNKS) with IRQ9 as the only possible irq. 3) has an PCI device referring to link device LNKS. So it causes interrupt storm when enabling the PCI device because PCI IRQ works in low level mode. Signed-off-by: Jiang Liu Acked-by: Bjorn Helgaas Signed-off-by: Rafael J. Wysocki --- drivers/acpi/pci_irq.c | 1 + drivers/acpi/pci_link.c | 13 +++++++++++++ include/linux/acpi.h | 1 + 3 files changed, 15 insertions(+) diff --git a/drivers/acpi/pci_irq.c b/drivers/acpi/pci_irq.c index 6da0f9beab19..c9336751e5e3 100644 --- a/drivers/acpi/pci_irq.c +++ b/drivers/acpi/pci_irq.c @@ -372,6 +372,7 @@ static int acpi_isa_register_gsi(struct pci_dev *dev) /* Interrupt Line values above 0xF are forbidden */ if (dev->irq > 0 && (dev->irq <= 0xF) && + acpi_isa_irq_available(dev->irq) && (acpi_isa_irq_to_gsi(dev->irq, &dev_gsi) == 0)) { dev_warn(&dev->dev, "PCI INT %c: no GSI - using ISA IRQ %d\n", pin_name(dev->pin), dev->irq); diff --git a/drivers/acpi/pci_link.c b/drivers/acpi/pci_link.c index 3b4ea98e3ea0..246e50d22120 100644 --- a/drivers/acpi/pci_link.c +++ b/drivers/acpi/pci_link.c @@ -553,6 +553,13 @@ static int acpi_pci_link_allocate(struct acpi_pci_link *link) irq = link->irq.possible[i]; } } + if (acpi_irq_penalty[irq] >= PIRQ_PENALTY_ISA_ALWAYS) { + printk(KERN_ERR PREFIX "No IRQ available for %s [%s]. " + "Try pci=noacpi or acpi=off\n", + acpi_device_name(link->device), + acpi_device_bid(link->device)); + return -ENODEV; + } /* Attempt to enable the link device at this IRQ. */ if (acpi_pci_link_set(link, irq)) { @@ -821,6 +828,12 @@ void acpi_penalize_isa_irq(int irq, int active) } } +bool acpi_isa_irq_available(int irq) +{ + return irq >= 0 && (irq >= ARRAY_SIZE(acpi_irq_penalty) || + acpi_irq_penalty[irq] < PIRQ_PENALTY_ISA_ALWAYS); +} + /* * Penalize IRQ used by ACPI SCI. If ACPI SCI pin attributes conflict with * PCI IRQ attributes, mark ACPI SCI as ISA_ALWAYS so it won't be use for diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 7235c4851460..43856d19cf4d 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -217,6 +217,7 @@ struct pci_dev; int acpi_pci_irq_enable (struct pci_dev *dev); void acpi_penalize_isa_irq(int irq, int active); +bool acpi_isa_irq_available(int irq); void acpi_penalize_sci_irq(int irq, int trigger, int polarity); void acpi_pci_irq_disable (struct pci_dev *dev); From d323efc786910bcc0c8f8b9f97780c70544ac4df Mon Sep 17 00:00:00 2001 From: Jiang Liu Date: Thu, 17 Sep 2015 14:02:46 +0800 Subject: [PATCH 478/647] ACPI / PCI: Remove duplicated penalty on SCI IRQ Now we have dedicated interface acpi_penalize_sci_irq() to penalize ISA IRQ used by ACPI SCI, so remove duplicated code to penalize ACPI SCI in acpi_irq_penalty_init(). Signed-off-by: Jiang Liu Acked-by: Bjorn Helgaas Signed-off-by: Rafael J. Wysocki --- drivers/acpi/pci_link.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/acpi/pci_link.c b/drivers/acpi/pci_link.c index 246e50d22120..7c8408b946ca 100644 --- a/drivers/acpi/pci_link.c +++ b/drivers/acpi/pci_link.c @@ -498,8 +498,7 @@ int __init acpi_irq_penalty_init(void) PIRQ_PENALTY_PCI_POSSIBLE; } } - /* Add a penalty for the SCI */ - acpi_irq_penalty[acpi_gbl_FADT.sci_interrupt] += PIRQ_PENALTY_PCI_USING; + return 0; } From 2866196f294954ce9fa226825c8c1eaa64c7da8a Mon Sep 17 00:00:00 2001 From: Doug Ledford Date: Fri, 25 Sep 2015 22:30:24 -0400 Subject: [PATCH 479/647] IB/ipoib: increase the max mcast backlog queue When performing sendonly joins, we queue the packets that trigger the join until the join completes. This may take on the order of hundreds of milliseconds. It is easy to have many more than three packets come in during that time. Expand the maximum queue depth in order to try and prevent dropped packets during the time it takes to join the multicast group. Signed-off-by: Doug Ledford --- drivers/infiniband/ulp/ipoib/ipoib.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index 4ff4e5290354..4cd5428a2399 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -80,7 +80,7 @@ enum { IPOIB_NUM_WC = 4, IPOIB_MAX_PATH_REC_QUEUE = 3, - IPOIB_MAX_MCAST_QUEUE = 3, + IPOIB_MAX_MCAST_QUEUE = 64, IPOIB_FLAG_OPER_UP = 0, IPOIB_FLAG_INITIALIZED = 1, From b4731977243c1ae9cfd8b3971ffc15425b7d7545 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Fri, 25 Sep 2015 20:45:27 -0700 Subject: [PATCH 480/647] MAINTAINERS: remove amd5536udc USB gadget driver maintainer Thomas can no longer work on the driver, so he asked me to mark the MAINTAINER entry as "Orphan" with the hope that someone else would someday pick it up. Cc: Thomas Dahlmann Signed-off-by: Greg Kroah-Hartman --- MAINTAINERS | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 7ba7ab749c85..28d5ed5abe9b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -615,9 +615,8 @@ F: Documentation/hwmon/fam15h_power F: drivers/hwmon/fam15h_power.c AMD GEODE CS5536 USB DEVICE CONTROLLER DRIVER -M: Thomas Dahlmann L: linux-geode@lists.infradead.org (moderated for non-subscribers) -S: Supported +S: Orphan F: drivers/usb/gadget/udc/amd5536udc.* AMD GEODE PROCESSOR/CHIPSET SUPPORT From 59f069789c98678710ed30a4be0daa3546ec82c7 Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 25 Sep 2015 11:56:56 +0100 Subject: [PATCH 481/647] net: update docbook comment for __mdiobus_register() Update the docbook comment for __mdiobus_register() to include the new module owner argument. This resolves a warning found by the 0-day builder. Signed-off-by: Russell King Signed-off-by: David S. Miller --- drivers/net/phy/mdio_bus.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index c340e412b38f..12f44c53cc8e 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -236,11 +236,14 @@ static inline void of_mdiobus_link_phydev(struct mii_bus *mdio, #endif /** - * mdiobus_register - bring up all the PHYs on a given bus and attach them to bus + * __mdiobus_register - bring up all the PHYs on a given bus and attach them to bus * @bus: target mii_bus + * @owner: module containing bus accessor functions * * Description: Called by a bus driver to bring up all the PHYs - * on a given bus, and attach them to the bus. + * on a given bus, and attach them to the bus. Drivers should use + * mdiobus_register() rather than __mdiobus_register() unless they + * need to pass a specific owner module. * * Returns 0 on success or < 0 on error. */ From bdb06cbf77cb01911694cc9076ffa8196b7b9b61 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 24 Sep 2015 15:31:29 -0600 Subject: [PATCH 482/647] net: Fix panic in icmp_route_lookup Andrey reported a panic: [ 7249.865507] BUG: unable to handle kernel pointer dereference at 000000b4 [ 7249.865559] IP: [] icmp_route_lookup+0xaa/0x320 [ 7249.865598] *pdpt = 0000000030f7f001 *pde = 0000000000000000 [ 7249.865637] Oops: 0000 [#1] ... [ 7249.866811] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.3.0-999-generic #201509220155 [ 7249.866876] Hardware name: MSI MS-7250/MS-7250, BIOS 080014 08/02/2006 [ 7249.866916] task: c1a5ab00 ti: c1a52000 task.ti: c1a52000 [ 7249.866949] EIP: 0060:[] EFLAGS: 00210246 CPU: 0 [ 7249.866981] EIP is at icmp_route_lookup+0xaa/0x320 [ 7249.867012] EAX: 00000000 EBX: f483ba48 ECX: 00000000 EDX: f2e18a00 [ 7249.867045] ESI: 000000c0 EDI: f483ba70 EBP: f483b9ec ESP: f483b974 [ 7249.867077] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 [ 7249.867108] CR0: 8005003b CR2: 000000b4 CR3: 36ee07c0 CR4: 000006f0 [ 7249.867141] Stack: [ 7249.867165] 320310ee 00000000 00000042 320310ee 00000000 c1aeca00 f3920240 f0c69180 [ 7249.867268] f483ba04 f855058b a89b66cd f483ba44 f8962f4b 00000000 e659266c f483ba54 [ 7249.867361] 8004753c f483ba5c f8962f4b f2031140 000003c1 ffbd8fa0 c16b0e00 00000064 [ 7249.867448] Call Trace: [ 7249.867494] [] ? e1000_xmit_frame+0x87b/0xdc0 [e1000e] [ 7249.867534] [] ? tcp_in_window+0xeb/0xb10 [nf_conntrack] [ 7249.867576] [] ? tcp_in_window+0xeb/0xb10 [nf_conntrack] [ 7249.867615] [] ? icmp_send+0xa0/0x380 [ 7249.867648] [] icmp_send+0x2cf/0x380 [ 7249.867681] [] nf_send_unreach+0xa6/0xc0 [nf_reject_ipv4] [ 7249.867714] [] reject_tg+0x7a/0x9f [ipt_REJECT] [ 7249.867746] [] ipt_do_table+0x317/0x70c [ip_tables] [ 7249.867780] [] ? __nf_conntrack_find_get+0x166/0x3b0 [nf_conntrack] [ 7249.867838] [] ? nf_conntrack_in+0x398/0x600 [nf_conntrack] [ 7249.867889] [] iptable_filter_hook+0x35/0x80 [iptable_filter] [ 7249.867933] [] nf_iterate+0x71/0x80 [ 7249.867970] [] nf_hook_slow+0x65/0xc0 [ 7249.868002] [] __ip_local_out_sk+0xc1/0xd0 [ 7249.868034] [] ? ip_forward_options+0x1a0/0x1a0 [ 7249.868066] [] ip_local_out_sk+0x16/0x30 [ 7249.868097] [] ip_send_skb+0x14/0x80 [ 7249.868129] [] ip_push_pending_frames+0x34/0x40 [ 7249.868163] [] ip_send_unicast_reply+0x282/0x310 [ 7249.868196] [] tcp_v4_send_reset+0x1b3/0x380 [ 7249.868227] [] tcp_v4_rcv+0x323/0x990 [ 7249.868257] [] ? nf_iterate+0x71/0x80 [ 7249.868289] [] ip_local_deliver_finish+0x8b/0x230 [ 7249.868322] [] ip_local_deliver+0x4c/0xa0 [ 7249.868353] [] ? ip_rcv_finish+0x390/0x390 [ 7249.868384] [] ip_rcv_finish+0x7c/0x390 [ 7249.868415] [] ip_rcv+0x2e0/0x420 ... Prior to the VRF change the oif was not set in the flow struct, so the VRF support should really have only added the vrf_master_ifindex lookup. Fixes: 613d09b30f8b ("net: Use VRF device index for lookups on TX") Cc: Andrey Melnikov Signed-off-by: David Ahern Signed-off-by: David S. Miller --- net/ipv4/icmp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 79fe05befcae..e5eb8ac4089d 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -427,7 +427,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) fl4.flowi4_mark = mark; fl4.flowi4_tos = RT_TOS(ip_hdr(skb)->tos); fl4.flowi4_proto = IPPROTO_ICMP; - fl4.flowi4_oif = vrf_master_ifindex(skb->dev) ? : skb->dev->ifindex; + fl4.flowi4_oif = vrf_master_ifindex(skb->dev); security_skb_classify_flow(skb, flowi4_to_flowi(&fl4)); rt = ip_route_output_key(net, &fl4); if (IS_ERR(rt)) @@ -461,7 +461,7 @@ static struct rtable *icmp_route_lookup(struct net *net, fl4->flowi4_proto = IPPROTO_ICMP; fl4->fl4_icmp_type = type; fl4->fl4_icmp_code = code; - fl4->flowi4_oif = vrf_master_ifindex(skb_in->dev) ? : skb_in->dev->ifindex; + fl4->flowi4_oif = vrf_master_ifindex(skb_in->dev); security_skb_classify_flow(skb_in, flowi4_to_flowi(fl4)); rt = __ip_route_output_key(net, fl4); From 756357b8e4b072fd5ee86421f794e071a348802b Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 25 Sep 2015 21:12:39 -0400 Subject: [PATCH 483/647] tools/power turbostat: IVB Xeon: fix --debug regression Staring in Linux-4.3-rc1, commit 6fb3143b561c ("tools/power turbostat: dump CONFIG_TDP") touches MSR 0x648, which is not supported on IVB-Xeon. This results in "turbostat --debug" exiting on those systems: turbostat: /dev/cpu/2/msr offset 0x648 read failed: Input/output error Remove IVB-Xeon from the list of machines supporting with that MSR. Signed-off-by: Len Brown --- tools/power/x86/turbostat/turbostat.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index 9655cb49c7cb..e05d30336613 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -1926,8 +1926,6 @@ int has_config_tdp(unsigned int family, unsigned int model) switch (model) { case 0x3A: /* IVB */ - case 0x3E: /* IVB Xeon */ - case 0x3C: /* HSW */ case 0x3F: /* HSX */ case 0x45: /* HSW */ From b2b34dfe4d9aa4c468fc363b3b666974783ed1f9 Mon Sep 17 00:00:00 2001 From: Hubert Chrzaniuk Date: Mon, 14 Sep 2015 13:31:00 +0200 Subject: [PATCH 484/647] tools/power turbostat: KNL workaround for %Busy and Avg_MHz KNL increments APERF and MPERF every 1024 clocks. This is compliant with the architecture specification, which requires that only the ratio of APERF/MPERF need be valid. However, turbostat takes advantage of the fact that these two MSRs increment every un-halted clock at the actual and base frequency: AVG_MHz = APERF_delta/measurement_interval %Busy = MPERF_delta/TSC_delta This quirk is needed for these calculations to also work on KNL, which would otherwise show a value 1024x smaller than expected. Signed-off-by: Hubert Chrzaniuk Signed-off-by: Len Brown --- tools/power/x86/turbostat/turbostat.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index e05d30336613..d333c819fa1f 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -71,6 +71,7 @@ unsigned int extra_msr_offset32; unsigned int extra_msr_offset64; unsigned int extra_delta_offset32; unsigned int extra_delta_offset64; +unsigned int aperf_mperf_multiplier = 1; int do_smi; double bclk; unsigned int show_pkg; @@ -984,6 +985,8 @@ int get_counters(struct thread_data *t, struct core_data *c, struct pkg_data *p) return -3; if (get_msr(cpu, MSR_IA32_MPERF, &t->mperf)) return -4; + t->aperf = t->aperf * aperf_mperf_multiplier; + t->mperf = t->mperf * aperf_mperf_multiplier; } if (do_smi) { @@ -2541,6 +2544,13 @@ int is_knl(unsigned int family, unsigned int model) return 0; } +unsigned int get_aperf_mperf_multiplier(unsigned int family, unsigned int model) +{ + if (is_knl(family, model)) + return 1024; + return 1; +} + #define SLM_BCLK_FREQS 5 double slm_freq_table[SLM_BCLK_FREQS] = { 83.3, 100.0, 133.3, 116.7, 80.0}; @@ -2742,6 +2752,9 @@ void process_cpuid() } } + if (has_aperf) + aperf_mperf_multiplier = get_aperf_mperf_multiplier(family, model); + do_nhm_platform_info = do_nhm_cstates = do_smi = probe_nhm_msrs(family, model); do_snb_cstates = has_snb_msrs(family, model); do_pc2 = do_snb_cstates && (pkg_cstate_limit >= PCL__2); From a2b7b74945dbfe5d734eafe8aa52f9f1f8bc6931 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Sat, 26 Sep 2015 00:12:38 -0400 Subject: [PATCH 485/647] tools/power turbostat: SKL: Adjust for TSC difference from base frequency On a Skylake with 1500MHz base frequency, the TSC runs at 1512MHz. This is because the TSC is no longer in the n*100 MHz BCLK domain, but is now in the m*24MHz crystal clock domain. (24 MHz * 63 = 1512 MHz) This adds error to several calculations in turbostat, unless the TSC sample sizes are adjusted for this difference. Note that calculations in the time domain are immune from this issue, as the timing sub-system has already calibrated the TSC against a known wall clock. AVG_MHz = APERF_delta/measurement_interval need no adjustment. APERF_delta is in the BCLK domain, and measurement_interval is in the time domain. TSC_MHz = TSC_delta/measurement_interval needs no adjustment -- as we really do want to report the actual measured TSC delta here, and measurement_interval is in the accurate time domain. %Busy = MPERF_delta/TSC_delta needs adjustment to use TSC_BCLK_DOMAIN_delta. TSC_BCLK_DOMAIN_delta = TSC_delta * base_hz / tsc_hz Bzy_MHz = TSC_delta/APERF_delta/MPERF_delta/measurement_interval need adjustment as above. No other metrics in turbostat need to be adjusted. Before: CPU Avg_MHz %Busy Bzy_MHz TSC_MHz - 550 24.84 2216 1512 0 2191 98.73 2219 1514 2 0 0.01 2130 1512 1 9 0.43 2016 1512 3 2 0.08 2016 1512 After: CPU Avg_MHz %Busy Bzy_MHz TSC_MHz - 550 25.05 2198 1512 0 2190 99.62 2199 1512 2 0 0.01 2152 1512 1 9 0.46 2000 1512 3 2 0.10 2000 1512 Note that in this example, the "Before" Bzy_MHz was reported as exceeding the 2200 max turbo rate. Also, even a pinned spin loop would not be reported as over 99% busy. Signed-off-by: Len Brown --- tools/power/x86/turbostat/turbostat.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index d333c819fa1f..31d756b4ea78 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -74,6 +74,8 @@ unsigned int extra_delta_offset64; unsigned int aperf_mperf_multiplier = 1; int do_smi; double bclk; +double base_hz; +double tsc_tweak = 1.0; unsigned int show_pkg; unsigned int show_core; unsigned int show_cpu; @@ -503,7 +505,7 @@ int format_counters(struct thread_data *t, struct core_data *c, /* %Busy */ if (has_aperf) { if (!skip_c0) - outp += sprintf(outp, "%8.2f", 100.0 * t->mperf/t->tsc); + outp += sprintf(outp, "%8.2f", 100.0 * t->mperf/t->tsc/tsc_tweak); else outp += sprintf(outp, "********"); } @@ -511,7 +513,7 @@ int format_counters(struct thread_data *t, struct core_data *c, /* Bzy_MHz */ if (has_aperf) outp += sprintf(outp, "%8.0f", - 1.0 * t->tsc / units * t->aperf / t->mperf / interval_float); + 1.0 * t->tsc * tsc_tweak / units * t->aperf / t->mperf / interval_float); /* TSC_MHz */ outp += sprintf(outp, "%8.0f", 1.0 * t->tsc/units/interval_float); @@ -1152,6 +1154,19 @@ int slv_pkg_cstate_limits[16] = {PCL__0, PCL__1, PCLRSV, PCLRSV, PCL__4, PCLRSV, int amt_pkg_cstate_limits[16] = {PCL__0, PCL__1, PCL__2, PCLRSV, PCLRSV, PCLRSV, PCL__6, PCL__7, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV}; int phi_pkg_cstate_limits[16] = {PCL__0, PCL__2, PCL_6N, PCL_6R, PCLRSV, PCLRSV, PCLRSV, PCLUNL, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV}; + +static void +calculate_tsc_tweak() +{ + unsigned long long msr; + unsigned int base_ratio; + + get_msr(base_cpu, MSR_NHM_PLATFORM_INFO, &msr); + base_ratio = (msr >> 8) & 0xFF; + base_hz = base_ratio * bclk * 1000000; + tsc_tweak = base_hz / tsc_hz; +} + static void dump_nhm_platform_info(void) { @@ -2773,6 +2788,9 @@ void process_cpuid() if (debug) dump_cstate_pstate_config_info(); + if (has_skl_msrs(family, model)) + calculate_tsc_tweak(); + return; } From af71b980c0d20586cc331b617c47094a8ec7e1db Mon Sep 17 00:00:00 2001 From: Len Brown Date: Sat, 26 Sep 2015 09:49:55 -0400 Subject: [PATCH 486/647] tools/power turbosat: update version number Signed-off-by: Len Brown --- tools/power/x86/turbostat/turbostat.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index 31d756b4ea78..bde0ef1a63df 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -3119,7 +3119,7 @@ int get_and_dump_counters(void) } void print_version() { - fprintf(stderr, "turbostat version 4.7 17-June, 2015" + fprintf(stderr, "turbostat version 4.8 26-Sep, 2015" " - Len Brown \n"); } From f26bf06beae70175eda91e893190784bd1bcc7c0 Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Thu, 24 Sep 2015 16:00:15 +0200 Subject: [PATCH 487/647] net: hisilicon: fix handling platform_get_irq result The function can return negative value. The problem has been detected using proposed semantic patch scripts/coccinelle/tests/assign_signed_to_unsigned.cocci [1]. [1]: http://permalink.gmane.org/gmane.linux.kernel/2046107 Signed-off-by: Andrzej Hajda Signed-off-by: David S. Miller --- drivers/net/ethernet/hisilicon/hip04_eth.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/hisilicon/hip04_eth.c b/drivers/net/ethernet/hisilicon/hip04_eth.c index cc2d8b4b18e3..253f8ed0537a 100644 --- a/drivers/net/ethernet/hisilicon/hip04_eth.c +++ b/drivers/net/ethernet/hisilicon/hip04_eth.c @@ -816,7 +816,7 @@ static int hip04_mac_probe(struct platform_device *pdev) struct net_device *ndev; struct hip04_priv *priv; struct resource *res; - unsigned int irq; + int irq; int ret; ndev = alloc_etherdev(sizeof(struct hip04_priv)); From 72521ea07c0af37b8cb21228368128191c3f1a58 Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Thu, 24 Sep 2015 16:00:24 +0200 Subject: [PATCH 488/647] r8169: fix handling rtl_readphy result The function can return negative value. The problem has been detected using proposed semantic patch scripts/coccinelle/tests/assign_signed_to_unsigned.cocci [1]. [1]: http://permalink.gmane.org/gmane.linux.kernel/2046107 Signed-off-by: Andrzej Hajda Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/r8169.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c index 2b32e0c5a0b4..b4f21232019a 100644 --- a/drivers/net/ethernet/realtek/r8169.c +++ b/drivers/net/ethernet/realtek/r8169.c @@ -6081,7 +6081,7 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp) { void __iomem *ioaddr = tp->mmio_addr; struct pci_dev *pdev = tp->pci_dev; - u16 rg_saw_cnt; + int rg_saw_cnt; u32 data; static const struct ephy_info e_info_8168h_1[] = { { 0x1e, 0x0800, 0x0001 }, From 9ffecb10283508260936b96022d4ee43a7798b4c Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 27 Sep 2015 07:50:08 -0400 Subject: [PATCH 489/647] Linux 4.3-rc3 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 84f4b31e3c6e..1d341eba143d 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 4 PATCHLEVEL = 3 SUBLEVEL = 0 -EXTRAVERSION = -rc2 +EXTRAVERSION = -rc3 NAME = Hurr durr I'ma sheep # *DOCUMENTATION* From 7573b94e08aeb5b814e2f277210bdcdf21a83869 Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Tue, 22 Sep 2015 11:29:09 -0700 Subject: [PATCH 490/647] MIPS: CM: Provide a function to map from CPU to VP ID. The VP ID of a given CPU may not match up with the CPU number used by Linux. For example, if the width of the VP part of the VP ID is wider than log2(number of VPs per core) and the system has multiple cores then this will be the case. Alternatively, if a pre-r6 system implements the MT ASE with multiple VPEs per core and Linux is built without support for the MT ASE then the numbers won't match up either. Provide a function to convert from CPU number to VP ID. Signed-off-by: Paul Burton Acked-by: Thomas Gleixner Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Cc: James Hogan Cc: Markos Chandras Patchwork: https://patchwork.linux-mips.org/patch/11211/ Signed-off-by: Ralf Baechle --- arch/mips/include/asm/mips-cm.h | 39 +++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+) diff --git a/arch/mips/include/asm/mips-cm.h b/arch/mips/include/asm/mips-cm.h index d75b75e78ebb..1f1927ab4269 100644 --- a/arch/mips/include/asm/mips-cm.h +++ b/arch/mips/include/asm/mips-cm.h @@ -194,6 +194,7 @@ BUILD_CM_RW(reg3_mask, MIPS_CM_GCB_OFS + 0xc8) BUILD_CM_R_(gic_status, MIPS_CM_GCB_OFS + 0xd0) BUILD_CM_R_(cpc_status, MIPS_CM_GCB_OFS + 0xf0) BUILD_CM_RW(l2_config, MIPS_CM_GCB_OFS + 0x130) +BUILD_CM_RW(sys_config2, MIPS_CM_GCB_OFS + 0x150) /* Core Local & Core Other register accessor functions */ BUILD_CM_Cx_RW(reset_release, 0x00) @@ -316,6 +317,10 @@ BUILD_CM_Cx_R_(tcid_8_priority, 0x80) #define CM_GCR_L2_CONFIG_ASSOC_SHF 0 #define CM_GCR_L2_CONFIG_ASSOC_MSK (_ULCAST_(0xff) << 0) +/* GCR_SYS_CONFIG2 register fields */ +#define CM_GCR_SYS_CONFIG2_MAXVPW_SHF 0 +#define CM_GCR_SYS_CONFIG2_MAXVPW_MSK (_ULCAST_(0xf) << 0) + /* GCR_Cx_COHERENCE register fields */ #define CM_GCR_Cx_COHERENCE_COHDOMAINEN_SHF 0 #define CM_GCR_Cx_COHERENCE_COHDOMAINEN_MSK (_ULCAST_(0xff) << 0) @@ -405,4 +410,38 @@ static inline int mips_cm_revision(void) return read_gcr_rev(); } +/** + * mips_cm_max_vp_width() - return the width in bits of VP indices + * + * Return: the width, in bits, of VP indices in fields that combine core & VP + * indices. + */ +static inline unsigned int mips_cm_max_vp_width(void) +{ + extern int smp_num_siblings; + + if (mips_cm_revision() >= CM_REV_CM3) + return read_gcr_sys_config2() & CM_GCR_SYS_CONFIG2_MAXVPW_MSK; + + return smp_num_siblings; +} + +/** + * mips_cm_vp_id() - calculate the hardware VP ID for a CPU + * @cpu: the CPU whose VP ID to calculate + * + * Hardware such as the GIC uses identifiers for VPs which may not match the + * CPU numbers used by Linux. This function calculates the hardware VP + * identifier corresponding to a given CPU. + * + * Return: the VP ID for the CPU. + */ +static inline unsigned int mips_cm_vp_id(unsigned int cpu) +{ + unsigned int core = cpu_data[cpu].core; + unsigned int vp = cpu_vpe_id(&cpu_data[cpu]); + + return (core * mips_cm_max_vp_width()) + vp; +} + #endif /* __MIPS_ASM_MIPS_CM_H__ */ From ab41f6c8620a6e65df4ee19d284c97efdd3d9b63 Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Tue, 22 Sep 2015 11:29:10 -0700 Subject: [PATCH 491/647] irqchip: mips-gic: Convert CPU numbers to VP IDs. Make use of the mips_cm_vp_id function to convert from Linux CPU numbers to the VP IDs used by hardware, which are not identical in all systems. Without doing so we map interrupts to incorrect VP(E)s. Signed-off-by: Paul Burton Acked-by: Thomas Gleixner Cc: linux-mips@linux-mips.org Cc: Paul Burton Cc: Marc Zyngier Cc: Jason Cooper Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/11212/ Signed-off-by: Ralf Baechle --- drivers/irqchip/irq-mips-gic.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/irqchip/irq-mips-gic.c b/drivers/irqchip/irq-mips-gic.c index af2f16bb8a94..842a53d3f4ad 100644 --- a/drivers/irqchip/irq-mips-gic.c +++ b/drivers/irqchip/irq-mips-gic.c @@ -426,7 +426,7 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *cpumask, spin_lock_irqsave(&gic_lock, flags); /* Re-route this IRQ */ - gic_map_to_vpe(irq, cpumask_first(&tmp)); + gic_map_to_vpe(irq, mips_cm_vp_id(cpumask_first(&tmp))); /* Update the pcpu_masks */ for (i = 0; i < NR_CPUS; i++) @@ -599,7 +599,7 @@ static __init void gic_ipi_init_one(unsigned int intr, int cpu, GIC_SHARED_TO_HWIRQ(intr)); int i; - gic_map_to_vpe(intr, cpu); + gic_map_to_vpe(intr, mips_cm_vp_id(cpu)); for (i = 0; i < NR_CPUS; i++) clear_bit(intr, pcpu_masks[i].pcpu_mask); set_bit(intr, pcpu_masks[cpu].pcpu_mask); From d77d5ac9c9b5abf45aeb6e12930fab832e5c81d1 Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Tue, 22 Sep 2015 11:29:11 -0700 Subject: [PATCH 492/647] irqchip: mips-gic: Fix pending & mask reads for MIPS64 with 32b GIC. gic_handle_shared_int reads the GIC interrupt pending & mask registers directly into a bitmap, which is defined as an array of unsigned longs. The GIC pending registers may be 32 bits wide if the CM is older than CM3, regardless of the bit width of the CPU, but for MIPS64 kernels the unsigned longs in the bitmap will be 64 bits wide. In this case we need to perform 2 x 32 bit reads per 64 bit unsigned long in order to avoid missing interrupts. Signed-off-by: Paul Burton Acked-by: Thomas Gleixner Cc: linux-mips@linux-mips.org Cc: Marc Zyngier Cc: Jason Cooper Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/11213/ Signed-off-by: Ralf Baechle --- drivers/irqchip/irq-mips-gic.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/irqchip/irq-mips-gic.c b/drivers/irqchip/irq-mips-gic.c index 842a53d3f4ad..aeaa061f0dbf 100644 --- a/drivers/irqchip/irq-mips-gic.c +++ b/drivers/irqchip/irq-mips-gic.c @@ -320,6 +320,14 @@ static void gic_handle_shared_int(bool chained) intrmask[i] = gic_read(intrmask_reg); pending_reg += gic_reg_step; intrmask_reg += gic_reg_step; + + if (!config_enabled(CONFIG_64BIT) || mips_cm_is64) + continue; + + pending[i] |= (u64)gic_read(pending_reg) << 32; + intrmask[i] |= (u64)gic_read(intrmask_reg) << 32; + pending_reg += gic_reg_step; + intrmask_reg += gic_reg_step; } bitmap_and(pending, pending, intrmask, gic_shared_intrs); From def3ab5d0a0fe53026c2495b054dcc46cf923dac Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Fri, 25 Sep 2015 08:59:36 -0700 Subject: [PATCH 493/647] MIPS: mm: compile maar_init unconditionally maar_init was previously only compiled when CONFIG_NEED_MULTIPLE_NODES was not set, which has been fine since it is only called from the standard implementation of mem_init which has the same condition. In preparation for calling it from the SMP startup code on secondary CPUs, move maar_init outside of the #ifndef such that it is always compiled. Signed-off-by: Paul Burton Cc: Markos Chandras Cc: linux-mips@linux-mips.org Cc: Steven J. Hill Cc: David Hildenbrand Cc: linux-kernel@vger.kernel.org Cc: Ingo Molnar Patchwork: https://patchwork.linux-mips.org/patch/11237/ Signed-off-by: Ralf Baechle --- arch/mips/mm/init.c | 126 ++++++++++++++++++++++---------------------- 1 file changed, 63 insertions(+), 63 deletions(-) diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c index 66d0f49c5bec..074ac5459026 100644 --- a/arch/mips/mm/init.c +++ b/arch/mips/mm/init.c @@ -252,6 +252,69 @@ void __init fixrange_init(unsigned long start, unsigned long end, #endif } +unsigned __weak platform_maar_init(unsigned num_pairs) +{ + struct maar_config cfg[BOOT_MEM_MAP_MAX]; + unsigned i, num_configured, num_cfg = 0; + phys_addr_t skip; + + for (i = 0; i < boot_mem_map.nr_map; i++) { + switch (boot_mem_map.map[i].type) { + case BOOT_MEM_RAM: + case BOOT_MEM_INIT_RAM: + break; + default: + continue; + } + + skip = 0x10000 - (boot_mem_map.map[i].addr & 0xffff); + + cfg[num_cfg].lower = boot_mem_map.map[i].addr; + cfg[num_cfg].lower += skip; + + cfg[num_cfg].upper = cfg[num_cfg].lower; + cfg[num_cfg].upper += boot_mem_map.map[i].size - 1; + cfg[num_cfg].upper -= skip; + + cfg[num_cfg].attrs = MIPS_MAAR_S; + num_cfg++; + } + + num_configured = maar_config(cfg, num_cfg, num_pairs); + if (num_configured < num_cfg) + pr_warn("Not enough MAAR pairs (%u) for all bootmem regions (%u)\n", + num_pairs, num_cfg); + + return num_configured; +} + +static void maar_init(void) +{ + unsigned num_maars, used, i; + + if (!cpu_has_maar) + return; + + /* Detect the number of MAARs */ + write_c0_maari(~0); + back_to_back_c0_hazard(); + num_maars = read_c0_maari() + 1; + + /* MAARs should be in pairs */ + WARN_ON(num_maars % 2); + + /* Configure the required MAARs */ + used = platform_maar_init(num_maars / 2); + + /* Disable any further MAARs */ + for (i = (used * 2); i < num_maars; i++) { + write_c0_maari(i); + back_to_back_c0_hazard(); + write_c0_maar(0); + back_to_back_c0_hazard(); + } +} + #ifndef CONFIG_NEED_MULTIPLE_NODES int page_is_ram(unsigned long pagenr) { @@ -334,69 +397,6 @@ static inline void mem_init_free_highmem(void) #endif } -unsigned __weak platform_maar_init(unsigned num_pairs) -{ - struct maar_config cfg[BOOT_MEM_MAP_MAX]; - unsigned i, num_configured, num_cfg = 0; - phys_addr_t skip; - - for (i = 0; i < boot_mem_map.nr_map; i++) { - switch (boot_mem_map.map[i].type) { - case BOOT_MEM_RAM: - case BOOT_MEM_INIT_RAM: - break; - default: - continue; - } - - skip = 0x10000 - (boot_mem_map.map[i].addr & 0xffff); - - cfg[num_cfg].lower = boot_mem_map.map[i].addr; - cfg[num_cfg].lower += skip; - - cfg[num_cfg].upper = cfg[num_cfg].lower; - cfg[num_cfg].upper += boot_mem_map.map[i].size - 1; - cfg[num_cfg].upper -= skip; - - cfg[num_cfg].attrs = MIPS_MAAR_S; - num_cfg++; - } - - num_configured = maar_config(cfg, num_cfg, num_pairs); - if (num_configured < num_cfg) - pr_warn("Not enough MAAR pairs (%u) for all bootmem regions (%u)\n", - num_pairs, num_cfg); - - return num_configured; -} - -static void maar_init(void) -{ - unsigned num_maars, used, i; - - if (!cpu_has_maar) - return; - - /* Detect the number of MAARs */ - write_c0_maari(~0); - back_to_back_c0_hazard(); - num_maars = read_c0_maari() + 1; - - /* MAARs should be in pairs */ - WARN_ON(num_maars % 2); - - /* Configure the required MAARs */ - used = platform_maar_init(num_maars / 2); - - /* Disable any further MAARs */ - for (i = (used * 2); i < num_maars; i++) { - write_c0_maari(i); - back_to_back_c0_hazard(); - write_c0_maar(0); - back_to_back_c0_hazard(); - } -} - void __init mem_init(void) { #ifdef CONFIG_HIGHMEM From 651ca7f4dab77f07fdac9cfb68bcab6bd2b7f827 Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Fri, 25 Sep 2015 08:59:37 -0700 Subject: [PATCH 494/647] MIPS: print MAAR configuration during boot Verifying that the MAAR configuration is as expected is useful when debugging the performance of a system. Print out the memory regions configured via MAAR along with their attributes. Signed-off-by: Paul Burton Cc: linux-mips@linux-mips.org Cc: Steven J. Hill Cc: David Hildenbrand Cc: linux-kernel@vger.kernel.org Cc: Peter Zijlstra (Intel) Patchwork: https://patchwork.linux-mips.org/patch/11238/ Signed-off-by: Ralf Baechle --- arch/mips/mm/init.c | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c index 074ac5459026..023c164b9eb6 100644 --- a/arch/mips/mm/init.c +++ b/arch/mips/mm/init.c @@ -291,6 +291,7 @@ unsigned __weak platform_maar_init(unsigned num_pairs) static void maar_init(void) { unsigned num_maars, used, i; + phys_addr_t lower, upper, attr; if (!cpu_has_maar) return; @@ -313,6 +314,34 @@ static void maar_init(void) write_c0_maar(0); back_to_back_c0_hazard(); } + + pr_info("MAAR configuration:\n"); + for (i = 0; i < num_maars; i += 2) { + write_c0_maari(i); + back_to_back_c0_hazard(); + upper = read_c0_maar(); + + write_c0_maari(i + 1); + back_to_back_c0_hazard(); + lower = read_c0_maar(); + + attr = lower & upper; + lower = (lower & MIPS_MAAR_ADDR) << 4; + upper = ((upper & MIPS_MAAR_ADDR) << 4) | 0xffff; + + pr_info(" [%d]: ", i / 2); + if (!(attr & MIPS_MAAR_V)) { + pr_cont("disabled\n"); + continue; + } + + pr_cont("%pa-%pa", &lower, &upper); + + if (attr & MIPS_MAAR_S) + pr_cont(" speculate"); + + pr_cont("\n"); + } } #ifndef CONFIG_NEED_MULTIPLE_NODES From e060f6ed281669b6d2f22d8dafd664b532386918 Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Fri, 25 Sep 2015 08:59:38 -0700 Subject: [PATCH 495/647] MIPS: Initialise MAARs on secondary CPUs MAARs should be initialised on each CPU (or rather, core) in the system in order to achieve consistent behaviour & performance. Previously they have only been initialised on the boot CPU which leads to performance problems if tasks are later scheduled on a secondary CPU, particularly if those tasks make use of unaligned vector accesses where some CPUs don't handle any cases in hardware for non-speculative memory regions. Fix this by recording the MAAR configuration from the boot CPU and applying it to secondary CPUs as part of their bringup. Reported-by: Doug Gilmore Signed-off-by: Paul Burton Cc: linux-mips@linux-mips.org Cc: Rusty Russell Cc: Steven J. Hill Cc: Andrew Bresticker Cc: Bjorn Helgaas Cc: David Hildenbrand Cc: linux-kernel@vger.kernel.org Cc: Aaro Koskinen Cc: James Hogan Cc: Ingo Molnar Cc: Markos Chandras Cc: Hemmo Nieminen Cc: Alex Smith Cc: Peter Zijlstra (Intel) Patchwork: https://patchwork.linux-mips.org/patch/11239/ Signed-off-by: Ralf Baechle --- arch/mips/include/asm/maar.h | 9 +++++++++ arch/mips/kernel/smp.c | 2 ++ arch/mips/mm/init.c | 28 +++++++++++++++++++++++++--- 3 files changed, 36 insertions(+), 3 deletions(-) diff --git a/arch/mips/include/asm/maar.h b/arch/mips/include/asm/maar.h index b02891f9caaf..21d9607c80d7 100644 --- a/arch/mips/include/asm/maar.h +++ b/arch/mips/include/asm/maar.h @@ -65,6 +65,15 @@ static inline void write_maar_pair(unsigned idx, phys_addr_t lower, back_to_back_c0_hazard(); } +/** + * maar_init() - initialise MAARs + * + * Performs initialisation of MAARs for the current CPU, making use of the + * platforms implementation of platform_maar_init where necessary and + * duplicating the setup it provides on secondary CPUs. + */ +extern void maar_init(void); + /** * struct maar_config - MAAR configuration data * @lower: The lowest address that the MAAR pair will affect. Must be diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c index a31896c33716..bd4385a8e6e8 100644 --- a/arch/mips/kernel/smp.c +++ b/arch/mips/kernel/smp.c @@ -42,6 +42,7 @@ #include #include #include +#include cpumask_t cpu_callin_map; /* Bitmask of started secondaries */ @@ -157,6 +158,7 @@ asmlinkage void start_secondary(void) mips_clockevent_init(); mp_ops->init_secondary(); cpu_report(); + maar_init(); /* * XXX parity protection should be folded in here when it's converted diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c index 023c164b9eb6..8770e619185e 100644 --- a/arch/mips/mm/init.c +++ b/arch/mips/mm/init.c @@ -44,6 +44,7 @@ #include #include #include +#include /* * We have up to 8 empty zeroed pages so we can map one of the right colour @@ -288,10 +289,14 @@ unsigned __weak platform_maar_init(unsigned num_pairs) return num_configured; } -static void maar_init(void) +void maar_init(void) { unsigned num_maars, used, i; phys_addr_t lower, upper, attr; + static struct { + struct maar_config cfgs[3]; + unsigned used; + } recorded = { { { 0 } }, 0 }; if (!cpu_has_maar) return; @@ -304,8 +309,14 @@ static void maar_init(void) /* MAARs should be in pairs */ WARN_ON(num_maars % 2); - /* Configure the required MAARs */ - used = platform_maar_init(num_maars / 2); + /* Set MAARs using values we recorded already */ + if (recorded.used) { + used = maar_config(recorded.cfgs, recorded.used, num_maars / 2); + BUG_ON(used != recorded.used); + } else { + /* Configure the required MAARs */ + used = platform_maar_init(num_maars / 2); + } /* Disable any further MAARs */ for (i = (used * 2); i < num_maars; i++) { @@ -315,6 +326,9 @@ static void maar_init(void) back_to_back_c0_hazard(); } + if (recorded.used) + return; + pr_info("MAAR configuration:\n"); for (i = 0; i < num_maars; i += 2) { write_c0_maari(i); @@ -341,6 +355,14 @@ static void maar_init(void) pr_cont(" speculate"); pr_cont("\n"); + + /* Record the setup for use on secondary CPUs */ + if (used <= ARRAY_SIZE(recorded.cfgs)) { + recorded.cfgs[recorded.used].lower = lower; + recorded.cfgs[recorded.used].upper = upper; + recorded.cfgs[recorded.used].attrs = attr; + recorded.used++; + } } } From 1a5e251996e1b602f2ddc9261ee9de0ca1875bfa Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Sun, 27 Sep 2015 17:13:55 -0700 Subject: [PATCH 496/647] Input: serio - fix blocking of parport If parkbd_allocate_serio() fails to allocate memory we are releasing the parport but we missed unregistering the device. As a result this device with exclusive access to that parport remains registered. And no other device will be able to use that parport even though this driver has failed to load. Signed-off-by: Sudip Mukherjee Signed-off-by: Dmitry Torokhov --- drivers/input/serio/parkbd.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/serio/parkbd.c b/drivers/input/serio/parkbd.c index 26b45936f9fd..1e8cd6f1fe9e 100644 --- a/drivers/input/serio/parkbd.c +++ b/drivers/input/serio/parkbd.c @@ -194,6 +194,7 @@ static int __init parkbd_init(void) parkbd_port = parkbd_allocate_serio(); if (!parkbd_port) { parport_release(parkbd_dev); + parport_unregister_device(parkbd_dev); return -ENOMEM; } From 5b3f33eb408ad06acf59b09fe0550bf3756e320a Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 6 Sep 2015 11:59:27 +0200 Subject: [PATCH 497/647] m68k: Wire up direct socket calls Signed-off-by: Geert Uytterhoeven Acked-by: Greg Ungerer --- arch/m68k/include/asm/unistd.h | 2 +- arch/m68k/include/uapi/asm/unistd.h | 17 +++++++++++++++++ arch/m68k/kernel/syscalltable.S | 18 +++++++++++++++++- 3 files changed, 35 insertions(+), 2 deletions(-) diff --git a/arch/m68k/include/asm/unistd.h b/arch/m68k/include/asm/unistd.h index 244e0dbe45db..7599cb94f9a0 100644 --- a/arch/m68k/include/asm/unistd.h +++ b/arch/m68k/include/asm/unistd.h @@ -4,7 +4,7 @@ #include -#define NR_syscalls 356 +#define NR_syscalls 373 #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_OLD_STAT diff --git a/arch/m68k/include/uapi/asm/unistd.h b/arch/m68k/include/uapi/asm/unistd.h index 61fb6cb9d2ae..b29ef18adb4a 100644 --- a/arch/m68k/include/uapi/asm/unistd.h +++ b/arch/m68k/include/uapi/asm/unistd.h @@ -361,5 +361,22 @@ #define __NR_memfd_create 353 #define __NR_bpf 354 #define __NR_execveat 355 +#define __NR_socket 356 +#define __NR_socketpair 357 +#define __NR_bind 358 +#define __NR_connect 359 +#define __NR_listen 360 +#define __NR_accept4 361 +#define __NR_getsockopt 362 +#define __NR_setsockopt 363 +#define __NR_getsockname 364 +#define __NR_getpeername 365 +#define __NR_sendto 366 +#define __NR_sendmsg 367 +#define __NR_recvfrom 368 +#define __NR_recvmsg 369 +#define __NR_shutdown 370 +#define __NR_recvmmsg 371 +#define __NR_sendmmsg 372 #endif /* _UAPI_ASM_M68K_UNISTD_H_ */ diff --git a/arch/m68k/kernel/syscalltable.S b/arch/m68k/kernel/syscalltable.S index a0ec4303f2c8..e0fe52b62a3e 100644 --- a/arch/m68k/kernel/syscalltable.S +++ b/arch/m68k/kernel/syscalltable.S @@ -376,4 +376,20 @@ ENTRY(sys_call_table) .long sys_memfd_create .long sys_bpf .long sys_execveat /* 355 */ - + .long sys_socket + .long sys_socketpair + .long sys_bind + .long sys_connect + .long sys_listen /* 360 */ + .long sys_accept4 + .long sys_getsockopt + .long sys_setsockopt + .long sys_getsockname + .long sys_getpeername /* 365 */ + .long sys_sendto + .long sys_sendmsg + .long sys_recvfrom + .long sys_recvmsg + .long sys_shutdown /* 370 */ + .long sys_recvmmsg + .long sys_sendmmsg From b92858f2be96f49c483436e851a4667543ab5768 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 6 Sep 2015 12:01:40 +0200 Subject: [PATCH 498/647] m68k: Wire up userfaultfd $ ./userfaultfd 10 99 nr_pages: 2560, nr_pages_per_cpu: 2560 bounces: 98, mode: racing, userfaults: 1121 bounces: 97, mode: rnd, userfaults: 977 bounces: 96, mode:, userfaults: 1119 bounces: 95, mode: rnd racing ver poll, userfaults: 1040 bounces: 94, mode: racing ver poll, userfaults: 1022 bounces: 93, mode: rnd ver poll, userfaults: 946 bounces: 92, mode: ver poll, userfaults: 1115 bounces: 91, mode: rnd racing poll, userfaults: 977 bounces: 90, mode: racing poll, userfaults: 899 bounces: 89, mode: rnd poll, userfaults: 881 bounces: 88, mode: poll, userfaults: 1069 bounces: 87, mode: rnd racing ver, userfaults: 1114 bounces: 86, mode: racing ver, userfaults: 1109 bounces: 85, mode: rnd ver, userfaults: 1165 bounces: 84, mode: ver, userfaults: 1107 bounces: 83, mode: rnd racing, userfaults: 1134 bounces: 82, mode: racing, userfaults: 1105 bounces: 81, mode: rnd, userfaults: 1323 bounces: 80, mode:, userfaults: 1103 bounces: 79, mode: rnd racing ver poll, userfaults: 909 bounces: 78, mode: racing ver poll, userfaults: 1095 bounces: 77, mode: rnd ver poll, userfaults: 951 bounces: 76, mode: ver poll, userfaults: 1099 bounces: 75, mode: rnd racing poll, userfaults: 1035 bounces: 74, mode: racing poll, userfaults: 1097 bounces: 73, mode: rnd poll, userfaults: 1159 bounces: 72, mode: poll, userfaults: 1042 bounces: 71, mode: rnd racing ver, userfaults: 848 bounces: 70, mode: racing ver, userfaults: 1093 bounces: 69, mode: rnd ver, userfaults: 892 bounces: 68, mode: ver, userfaults: 1091 bounces: 67, mode: rnd racing, userfaults: 1219 bounces: 66, mode: racing, userfaults: 1089 bounces: 65, mode: rnd, userfaults: 988 bounces: 64, mode:, userfaults: 1087 bounces: 63, mode: rnd racing ver poll, userfaults: 882 bounces: 62, mode: racing ver poll, userfaults: 984 bounces: 61, mode: rnd ver poll, userfaults: 701 bounces: 60, mode: ver poll, userfaults: 1071 bounces: 59, mode: rnd racing poll, userfaults: 1137 bounces: 58, mode: racing poll, userfaults: 1032 bounces: 57, mode: rnd poll, userfaults: 911 bounces: 56, mode: poll, userfaults: 1079 bounces: 55, mode: rnd racing ver, userfaults: 1106 bounces: 54, mode: racing ver, userfaults: 1077 bounces: 53, mode: rnd ver, userfaults: 886 bounces: 52, mode: ver, userfaults: 1075 bounces: 51, mode: rnd racing, userfaults: 1101 bounces: 50, mode: racing, userfaults: 1073 bounces: 49, mode: rnd, userfaults: 1070 bounces: 48, mode:, userfaults: 1071 bounces: 47, mode: rnd racing ver poll, userfaults: 1077 bounces: 46, mode: racing ver poll, userfaults: 910 bounces: 45, mode: rnd ver poll, userfaults: 1063 bounces: 44, mode: ver poll, userfaults: 1028 bounces: 43, mode: rnd racing poll, userfaults: 1043 bounces: 42, mode: racing poll, userfaults: 1065 bounces: 41, mode: rnd poll, userfaults: 912 bounces: 40, mode: poll, userfaults: 1063 bounces: 39, mode: rnd racing ver, userfaults: 880 bounces: 38, mode: racing ver, userfaults: 1061 bounces: 37, mode: rnd ver, userfaults: 1144 bounces: 36, mode: ver, userfaults: 1059 bounces: 35, mode: rnd racing, userfaults: 967 bounces: 34, mode: racing, userfaults: 1057 bounces: 33, mode: rnd, userfaults: 1076 bounces: 32, mode:, userfaults: 1055 bounces: 31, mode: rnd racing ver poll, userfaults: 997 bounces: 30, mode: racing ver poll, userfaults: 1053 bounces: 29, mode: rnd ver poll, userfaults: 968 bounces: 28, mode: ver poll, userfaults: 978 bounces: 27, mode: rnd racing poll, userfaults: 1008 bounces: 26, mode: racing poll, userfaults: 1049 bounces: 25, mode: rnd poll, userfaults: 900 bounces: 24, mode: poll, userfaults: 1047 bounces: 23, mode: rnd racing ver, userfaults: 988 bounces: 22, mode: racing ver, userfaults: 1045 bounces: 21, mode: rnd ver, userfaults: 1027 bounces: 20, mode: ver, userfaults: 1043 bounces: 19, mode: rnd racing, userfaults: 1017 bounces: 18, mode: racing, userfaults: 1041 bounces: 17, mode: rnd, userfaults: 979 bounces: 16, mode:, userfaults: 1039 bounces: 15, mode: rnd racing ver poll, userfaults: 1134 bounces: 14, mode: racing ver poll, userfaults: 1037 bounces: 13, mode: rnd ver poll, userfaults: 1046 bounces: 12, mode: ver poll, userfaults: 1035 bounces: 11, mode: rnd racing poll, userfaults: 1060 bounces: 10, mode: racing poll, userfaults: 1033 bounces: 9, mode: rnd poll, userfaults: 1003 bounces: 8, mode: poll, userfaults: 929 bounces: 7, mode: rnd racing ver, userfaults: 964 bounces: 6, mode: racing ver, userfaults: 1029 bounces: 5, mode: rnd ver, userfaults: 1053 bounces: 4, mode: ver, userfaults: 1027 bounces: 3, mode: rnd racing, userfaults: 863 bounces: 2, mode: racing, userfaults: 1025 bounces: 1, mode: rnd, userfaults: 1043 bounces: 0, mode:, userfaults: 950 Signed-off-by: Geert Uytterhoeven Acked-by: Greg Ungerer --- arch/m68k/include/asm/unistd.h | 2 +- arch/m68k/include/uapi/asm/unistd.h | 1 + arch/m68k/kernel/syscalltable.S | 1 + 3 files changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/m68k/include/asm/unistd.h b/arch/m68k/include/asm/unistd.h index 7599cb94f9a0..d25d5a5c83cb 100644 --- a/arch/m68k/include/asm/unistd.h +++ b/arch/m68k/include/asm/unistd.h @@ -4,7 +4,7 @@ #include -#define NR_syscalls 373 +#define NR_syscalls 374 #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_OLD_STAT diff --git a/arch/m68k/include/uapi/asm/unistd.h b/arch/m68k/include/uapi/asm/unistd.h index b29ef18adb4a..0eebb28488ec 100644 --- a/arch/m68k/include/uapi/asm/unistd.h +++ b/arch/m68k/include/uapi/asm/unistd.h @@ -378,5 +378,6 @@ #define __NR_shutdown 370 #define __NR_recvmmsg 371 #define __NR_sendmmsg 372 +#define __NR_userfaultfd 373 #endif /* _UAPI_ASM_M68K_UNISTD_H_ */ diff --git a/arch/m68k/kernel/syscalltable.S b/arch/m68k/kernel/syscalltable.S index e0fe52b62a3e..7b5a315f6df2 100644 --- a/arch/m68k/kernel/syscalltable.S +++ b/arch/m68k/kernel/syscalltable.S @@ -393,3 +393,4 @@ ENTRY(sys_call_table) .long sys_shutdown /* 370 */ .long sys_recvmmsg .long sys_sendmmsg + .long sys_userfaultfd From 7f843dab134b7ce8804b67cca2271267d3a0213d Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 13 Sep 2015 11:49:13 +0200 Subject: [PATCH 499/647] m68k: Wire up membarrier $ ./membarrier_test membarrier MEMBARRIER_CMD_QUERY syscall available. membarrier: MEMBARRIER_CMD_SHARED success. membarrier: tests done! $ Signed-off-by: Geert Uytterhoeven Acked-by: Greg Ungerer --- arch/m68k/include/asm/unistd.h | 2 +- arch/m68k/include/uapi/asm/unistd.h | 1 + arch/m68k/kernel/syscalltable.S | 1 + 3 files changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/m68k/include/asm/unistd.h b/arch/m68k/include/asm/unistd.h index d25d5a5c83cb..0793a7f17417 100644 --- a/arch/m68k/include/asm/unistd.h +++ b/arch/m68k/include/asm/unistd.h @@ -4,7 +4,7 @@ #include -#define NR_syscalls 374 +#define NR_syscalls 375 #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_OLD_STAT diff --git a/arch/m68k/include/uapi/asm/unistd.h b/arch/m68k/include/uapi/asm/unistd.h index 0eebb28488ec..5e6fae6c275f 100644 --- a/arch/m68k/include/uapi/asm/unistd.h +++ b/arch/m68k/include/uapi/asm/unistd.h @@ -379,5 +379,6 @@ #define __NR_recvmmsg 371 #define __NR_sendmmsg 372 #define __NR_userfaultfd 373 +#define __NR_membarrier 374 #endif /* _UAPI_ASM_M68K_UNISTD_H_ */ diff --git a/arch/m68k/kernel/syscalltable.S b/arch/m68k/kernel/syscalltable.S index 7b5a315f6df2..5dd0e80042f5 100644 --- a/arch/m68k/kernel/syscalltable.S +++ b/arch/m68k/kernel/syscalltable.S @@ -394,3 +394,4 @@ ENTRY(sys_call_table) .long sys_recvmmsg .long sys_sendmmsg .long sys_userfaultfd + .long sys_membarrier From 8474ba74193d302e8340dddd1e16c85cc4b98caf Mon Sep 17 00:00:00 2001 From: Andreas Schwab Date: Wed, 23 Sep 2015 23:12:09 +0200 Subject: [PATCH 500/647] m68k: Define asmlinkage_protect Make sure the compiler does not modify arguments of syscall functions. This can happen if the compiler generates a tailcall to another function. For example, without asmlinkage_protect sys_openat is compiled into this function: sys_openat: clr.l %d0 move.w 18(%sp),%d0 move.l %d0,16(%sp) jbra do_sys_open Note how the fourth argument is modified in place, modifying the register %d4 that gets restored from this stack slot when the function returns to user-space. The caller may expect the register to be unmodified across system calls. Signed-off-by: Andreas Schwab Signed-off-by: Geert Uytterhoeven Cc: stable@vger.kernel.org --- arch/m68k/include/asm/linkage.h | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) diff --git a/arch/m68k/include/asm/linkage.h b/arch/m68k/include/asm/linkage.h index 5a822bb790f7..066e74f666ae 100644 --- a/arch/m68k/include/asm/linkage.h +++ b/arch/m68k/include/asm/linkage.h @@ -4,4 +4,34 @@ #define __ALIGN .align 4 #define __ALIGN_STR ".align 4" +/* + * Make sure the compiler doesn't do anything stupid with the + * arguments on the stack - they are owned by the *caller*, not + * the callee. This just fools gcc into not spilling into them, + * and keeps it from doing tailcall recursion and/or using the + * stack slots for temporaries, since they are live and "used" + * all the way to the end of the function. + */ +#define asmlinkage_protect(n, ret, args...) \ + __asmlinkage_protect##n(ret, ##args) +#define __asmlinkage_protect_n(ret, args...) \ + __asm__ __volatile__ ("" : "=r" (ret) : "0" (ret), ##args) +#define __asmlinkage_protect0(ret) \ + __asmlinkage_protect_n(ret) +#define __asmlinkage_protect1(ret, arg1) \ + __asmlinkage_protect_n(ret, "m" (arg1)) +#define __asmlinkage_protect2(ret, arg1, arg2) \ + __asmlinkage_protect_n(ret, "m" (arg1), "m" (arg2)) +#define __asmlinkage_protect3(ret, arg1, arg2, arg3) \ + __asmlinkage_protect_n(ret, "m" (arg1), "m" (arg2), "m" (arg3)) +#define __asmlinkage_protect4(ret, arg1, arg2, arg3, arg4) \ + __asmlinkage_protect_n(ret, "m" (arg1), "m" (arg2), "m" (arg3), \ + "m" (arg4)) +#define __asmlinkage_protect5(ret, arg1, arg2, arg3, arg4, arg5) \ + __asmlinkage_protect_n(ret, "m" (arg1), "m" (arg2), "m" (arg3), \ + "m" (arg4), "m" (arg5)) +#define __asmlinkage_protect6(ret, arg1, arg2, arg3, arg4, arg5, arg6) \ + __asmlinkage_protect_n(ret, "m" (arg1), "m" (arg2), "m" (arg3), \ + "m" (arg4), "m" (arg5), "m" (arg6)) + #endif From 95bc06ef049b808a067327bd8490b608b47e3870 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 13 Sep 2015 11:24:02 +0200 Subject: [PATCH 501/647] m68k/defconfig: Update defconfigs for v4.3-rc1 Signed-off-by: Geert Uytterhoeven --- arch/m68k/configs/amiga_defconfig | 9 ++++++++- arch/m68k/configs/apollo_defconfig | 9 ++++++++- arch/m68k/configs/atari_defconfig | 9 ++++++++- arch/m68k/configs/bvme6000_defconfig | 9 ++++++++- arch/m68k/configs/hp300_defconfig | 9 ++++++++- arch/m68k/configs/mac_defconfig | 9 ++++++++- arch/m68k/configs/multi_defconfig | 9 ++++++++- arch/m68k/configs/mvme147_defconfig | 9 ++++++++- arch/m68k/configs/mvme16x_defconfig | 9 ++++++++- arch/m68k/configs/q40_defconfig | 9 ++++++++- arch/m68k/configs/sun3_defconfig | 9 ++++++++- arch/m68k/configs/sun3x_defconfig | 9 ++++++++- 12 files changed, 96 insertions(+), 12 deletions(-) diff --git a/arch/m68k/configs/amiga_defconfig b/arch/m68k/configs/amiga_defconfig index 0b6b40d37b95..5b4ec541ba7c 100644 --- a/arch/m68k/configs/amiga_defconfig +++ b/arch/m68k/configs/amiga_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -57,7 +58,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -67,10 +67,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -179,6 +181,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -206,6 +209,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -271,6 +275,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -370,6 +375,7 @@ CONFIG_ZORRO8390=y # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_SMSC is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -537,6 +543,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/apollo_defconfig b/arch/m68k/configs/apollo_defconfig index eeb3a8991fc4..6e5198e2c124 100644 --- a/arch/m68k/configs/apollo_defconfig +++ b/arch/m68k/configs/apollo_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -55,7 +56,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -65,10 +65,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -177,6 +179,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -204,6 +207,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -269,6 +273,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -344,6 +349,7 @@ CONFIG_VETH=m # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -495,6 +501,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/atari_defconfig b/arch/m68k/configs/atari_defconfig index 3a7006654ce9..f75600b0ca23 100644 --- a/arch/m68k/configs/atari_defconfig +++ b/arch/m68k/configs/atari_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -55,7 +56,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -65,10 +65,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -177,6 +179,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -204,6 +207,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -269,6 +273,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -355,6 +360,7 @@ CONFIG_NE2000=y # CONFIG_NET_VENDOR_SEEQ is not set CONFIG_SMC91X=y # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -517,6 +523,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/bvme6000_defconfig b/arch/m68k/configs/bvme6000_defconfig index 0586b323a673..a42d91c389a6 100644 --- a/arch/m68k/configs/bvme6000_defconfig +++ b/arch/m68k/configs/bvme6000_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -53,7 +54,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -63,10 +63,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -175,6 +177,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -202,6 +205,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -267,6 +271,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -343,6 +348,7 @@ CONFIG_BVME6000_NET=y # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -488,6 +494,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/hp300_defconfig b/arch/m68k/configs/hp300_defconfig index ad1dbce07aa4..77f4a11083e9 100644 --- a/arch/m68k/configs/hp300_defconfig +++ b/arch/m68k/configs/hp300_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -55,7 +56,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -65,10 +65,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -177,6 +179,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -204,6 +207,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -269,6 +273,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -345,6 +350,7 @@ CONFIG_HPLANCE=y # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -497,6 +503,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/mac_defconfig b/arch/m68k/configs/mac_defconfig index b44acacaecf4..5a329f77329b 100644 --- a/arch/m68k/configs/mac_defconfig +++ b/arch/m68k/configs/mac_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -54,7 +55,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -64,10 +64,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -176,6 +178,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -203,6 +206,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -271,6 +275,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -364,6 +369,7 @@ CONFIG_MAC8390=y # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_SMSC is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -519,6 +525,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/multi_defconfig b/arch/m68k/configs/multi_defconfig index 8afca3753db1..83c80d2030ec 100644 --- a/arch/m68k/configs/multi_defconfig +++ b/arch/m68k/configs/multi_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -64,7 +65,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -74,10 +74,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -186,6 +188,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -213,6 +216,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -281,6 +285,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -410,6 +415,7 @@ CONFIG_ZORRO8390=y # CONFIG_NET_VENDOR_SEEQ is not set CONFIG_SMC91X=y # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PLIP=m @@ -599,6 +605,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/mvme147_defconfig b/arch/m68k/configs/mvme147_defconfig index ef00875994d9..6cb42c3bf5a2 100644 --- a/arch/m68k/configs/mvme147_defconfig +++ b/arch/m68k/configs/mvme147_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -52,7 +53,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -62,10 +62,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -174,6 +176,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -201,6 +204,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -266,6 +270,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -343,6 +348,7 @@ CONFIG_MVME147_NET=y # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -488,6 +494,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/mvme16x_defconfig b/arch/m68k/configs/mvme16x_defconfig index 387c2bd90ff1..c7508c30330c 100644 --- a/arch/m68k/configs/mvme16x_defconfig +++ b/arch/m68k/configs/mvme16x_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -53,7 +54,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -63,10 +63,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -175,6 +177,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -202,6 +205,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -267,6 +271,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -343,6 +348,7 @@ CONFIG_MVME16x_NET=y # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -488,6 +494,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/q40_defconfig b/arch/m68k/configs/q40_defconfig index 35355c1bc714..64b71664a303 100644 --- a/arch/m68k/configs/q40_defconfig +++ b/arch/m68k/configs/q40_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -53,7 +54,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -63,10 +63,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -175,6 +177,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -202,6 +205,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -267,6 +271,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -354,6 +359,7 @@ CONFIG_NE2000=y # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_SMSC is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PLIP=m @@ -510,6 +516,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/sun3_defconfig b/arch/m68k/configs/sun3_defconfig index 8442d267b877..9a4cab78a2ea 100644 --- a/arch/m68k/configs/sun3_defconfig +++ b/arch/m68k/configs/sun3_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -50,7 +51,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -60,10 +60,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -172,6 +174,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -199,6 +202,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -264,6 +268,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -341,6 +346,7 @@ CONFIG_SUN3_82586=y # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set # CONFIG_NET_VENDOR_SUN is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -489,6 +495,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m CONFIG_CRYPTO_MANAGER=y diff --git a/arch/m68k/configs/sun3x_defconfig b/arch/m68k/configs/sun3x_defconfig index 0e1b542e1555..1a2eaac13dbd 100644 --- a/arch/m68k/configs/sun3x_defconfig +++ b/arch/m68k/configs/sun3x_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -50,7 +51,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -60,10 +60,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -172,6 +174,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -199,6 +202,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -264,6 +268,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -341,6 +346,7 @@ CONFIG_SUN3LANCE=y # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -489,6 +495,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m From bc5f2ab11ca6dda4a4826e7e78d5365d7c3e1569 Mon Sep 17 00:00:00 2001 From: Rodrigo Vivi Date: Wed, 23 Sep 2015 11:32:36 -0700 Subject: [PATCH 502/647] drm/i915/skl: Don't call intel_prepare_ddi when encoder list isn't yet initialized. In case something goes wrong with power well initialization we were calling intel_prepare_ddi during boot while encoder list isnt't initilized. [ 9.618747] i915 0000:00:02.0: Invalid ROM contents [ 9.631446] [drm] failed to find VBIOS tables [ 9.720036] BUG: unable to handle kernel NULL pointer dereference at 00000000 00000058 [ 9.721986] IP: [] ddi_get_encoder_port+0x82/0x190 [i915] [ 9.723736] PGD 0 [ 9.724286] Oops: 0000 [#1] PREEMPT SMP [ 9.725386] Modules linked in: intel_powerclamp snd_hda_intel(+) coretemp crc 32c_intel snd_hda_codec snd_hda_core serio_raw snd_pcm snd_timer i915(+) parport _pc parport pinctrl_sunrisepoint pinctrl_intel nfsd nfs_acl [ 9.730635] CPU: 0 PID: 497 Comm: systemd-udevd Not tainted 4.3.0-rc2-eywa-10 967-g72de2cfd-dirty #2 [ 9.732785] Hardware name: Intel Corporation Cannonlake Client platform/Skyla ke DT DDR4 RVP8, BIOS CNLSE2R1.R00.X021.B00.1508040310 08/04/2015 [ 9.735785] task: ffff88008a704700 ti: ffff88016a1ac000 task.ti: ffff88016a1a c000 [ 9.737584] RIP: 0010:[] [] ddi_get_enco der_port+0x82/0x190 [i915] [ 9.739934] RSP: 0000:ffff88016a1af710 EFLAGS: 00010296 [ 9.741184] RAX: 000000000000004e RBX: ffff88008a9edc98 RCX: 0000000000000001 [ 9.742934] RDX: 000000000000004e RSI: ffffffff81fc1e82 RDI: 00000000ffffffff [ 9.744634] RBP: ffff88016a1af730 R08: 0000000000000000 R09: 0000000000000578 [ 9.746333] R10: 0000000000001065 R11: 0000000000000578 R12: fffffffffffffff8 [ 9.748033] R13: ffff88016a1af7a8 R14: ffff88016a1af794 R15: 0000000000000000 [ 9.749733] FS: 00007eff2e1e07c0(0000) GS:ffff88016fc00000(0000) knlGS:00000 00000000000 [ 9.751683] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 9.753083] CR2: 0000000000000058 CR3: 000000016922b000 CR4: 00000000003406f0 [ 9.754782] Stack: [ 9.755332] ffff88008a9edc98 ffff88008a9ed800 ffffffffa01d07b0 00000000fffb9 09e [ 9.757232] ffff88016a1af7d8 ffffffffa0154ea7 0000000000000246 ffff88016a370 080 [ 9.759182] ffff88016a370080 ffff88008a9ed800 0000000000000246 ffff88008a9ed c98 [ 9.761132] Call Trace: [ 9.761782] [] intel_prepare_ddi+0x67/0x860 [i915] [ 9.763332] [] ? _raw_spin_unlock_irqrestore+0x26/0x40 [ 9.765031] [] ? gen9_read32+0x141/0x360 [i915] [ 9.766531] [] skl_set_power_well+0x431/0xa80 [i915] [ 9.768181] [] skl_power_well_enable+0x13/0x20 [i915] [ 9.769781] [] intel_power_well_enable+0x28/0x50 [i915] [ 9.771481] [] intel_display_power_get+0x92/0xc0 [i915] [ 9.773180] [] intel_display_set_init_power+0x3b/0x40 [i91 5] [ 9.774980] [] intel_power_domains_init_hw+0x120/0x520 [i9 15] [ 9.776780] [] i915_driver_load+0xb21/0xf40 [i915] So let's protect this case. My first attempt was to remove the intel_prepare_ddi, but Daniel had pointed out this is really needed to restore those registers values. And Imre pointed out that this case was without the flag protection and this was actually where things were going bad. So I've just checked and this indeed solves my issue. The regressing intel_prepare_ddi call was added in commit 1d2b9526a790d55b7ae870934a74937081f62de2 Author: Damien Lespiau Date: Fri Mar 6 18:50:53 2015 +0000 drm/i915/skl: Restore the DDI translation tables when enabling PW1 Cc: Imre Deak Cc: Daniel Vetter Signed-off-by: Rodrigo Vivi Reviewed-by: Imre Deak [Jani: regression reference] Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_runtime_pm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index af7fdb3bd663..7401cf90b0db 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -246,7 +246,8 @@ static void skl_power_well_post_enable(struct drm_i915_private *dev_priv, } if (power_well->data == SKL_DISP_PW_1) { - intel_prepare_ddi(dev); + if (!dev_priv->power_domains.initializing) + intel_prepare_ddi(dev); gen8_irq_power_well_post_enable(dev_priv, 1 << PIPE_A); } } From c73318f43d3967e3ce810665d9c74a7d238d24d1 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Thu, 3 Sep 2015 13:06:09 +0200 Subject: [PATCH 503/647] watchdog: Fix module autoload for OF platform driver These platform drivers have a OF device ID table but the OF module alias information is not created so module autoloading won't work. Signed-off-by: Luis de Bethencourt Reviewed-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/gef_wdt.c | 1 + drivers/watchdog/mena21_wdt.c | 1 + drivers/watchdog/moxart_wdt.c | 1 + 3 files changed, 3 insertions(+) diff --git a/drivers/watchdog/gef_wdt.c b/drivers/watchdog/gef_wdt.c index cc1bdfc2ff71..006e2348022c 100644 --- a/drivers/watchdog/gef_wdt.c +++ b/drivers/watchdog/gef_wdt.c @@ -303,6 +303,7 @@ static const struct of_device_id gef_wdt_ids[] = { }, {}, }; +MODULE_DEVICE_TABLE(of, gef_wdt_ids); static struct platform_driver gef_wdt_driver = { .driver = { diff --git a/drivers/watchdog/mena21_wdt.c b/drivers/watchdog/mena21_wdt.c index 69013007dc47..098fa9c34d6d 100644 --- a/drivers/watchdog/mena21_wdt.c +++ b/drivers/watchdog/mena21_wdt.c @@ -253,6 +253,7 @@ static const struct of_device_id a21_wdt_ids[] = { { .compatible = "men,a021-wdt" }, { }, }; +MODULE_DEVICE_TABLE(of, a21_wdt_ids); static struct platform_driver a21_wdt_driver = { .probe = a21_wdt_probe, diff --git a/drivers/watchdog/moxart_wdt.c b/drivers/watchdog/moxart_wdt.c index 2789da2c0515..60b0605bd7e6 100644 --- a/drivers/watchdog/moxart_wdt.c +++ b/drivers/watchdog/moxart_wdt.c @@ -168,6 +168,7 @@ static const struct of_device_id moxart_watchdog_match[] = { { .compatible = "moxa,moxart-watchdog" }, { }, }; +MODULE_DEVICE_TABLE(of, moxart_watchdog_match); static struct platform_driver moxart_wdt_driver = { .probe = moxart_wdt_probe, From 898e6861ff7cfc9f539b57859a27fbd1fe4298ae Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Noralf=20Tr=C3=B8nnes?= Date: Wed, 17 Jun 2015 16:04:04 +0200 Subject: [PATCH 504/647] watchdog: bcm2835: Fix poweroff behaviour MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently poweroff/halt results in a reboot on the Raspberry Pi. The firmware uses the RSTS register to know which partiton to boot from. The partiton value is spread into bits 0, 2, 4, 6, 8, 10. Partiton 63 is a special partition used by the firmware to indicate halt. The firmware made this change in 19 Aug 2013 and was matched by the downstream commit: Changes for new NOOBS multi partition booting from gsh Signed-off-by: Noralf Trønnes Tested-by: Stephen Warren Acked-by: Stephen Warren Reviewed-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/bcm2835_wdt.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/watchdog/bcm2835_wdt.c b/drivers/watchdog/bcm2835_wdt.c index 66c3e656a616..8a5ce5b5a0b6 100644 --- a/drivers/watchdog/bcm2835_wdt.c +++ b/drivers/watchdog/bcm2835_wdt.c @@ -36,6 +36,13 @@ #define PM_RSTC_WRCFG_FULL_RESET 0x00000020 #define PM_RSTC_RESET 0x00000102 +/* + * The Raspberry Pi firmware uses the RSTS register to know which partiton + * to boot from. The partiton value is spread into bits 0, 2, 4, 6, 8, 10. + * Partiton 63 is a special partition used by the firmware to indicate halt. + */ +#define PM_RSTS_RASPBERRYPI_HALT 0x555 + #define SECS_TO_WDOG_TICKS(x) ((x) << 16) #define WDOG_TICKS_TO_SECS(x) ((x) >> 16) @@ -151,8 +158,7 @@ static void bcm2835_power_off(void) * hard reset. */ val = readl_relaxed(wdt->base + PM_RSTS); - val &= PM_RSTC_WRCFG_CLR; - val |= PM_PASSWORD | PM_RSTS_HADWRH_SET; + val |= PM_PASSWORD | PM_RSTS_RASPBERRYPI_HALT; writel_relaxed(val, wdt->base + PM_RSTS); /* Continue with normal reset mechanism */ From 3cef072d3bbd4344823545e50d8cb240a6d4635d Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Fri, 11 Sep 2015 06:28:08 -0700 Subject: [PATCH 505/647] watchdog: iTCO: Fix dependencies on I2C If I2C is built as module, the iTCO watchdog driver must be built as module as well. I2C_I801 must only be selected if I2C is configured. This fixes the following build errors, seen if I2C=m and ITCO_WDT=y. i2c-i801.c:(.text+0x2bf055): undefined reference to `i2c_del_adapter' i2c-i801.c:(.text+0x2c13e0): undefined reference to `i2c_add_adapter' i2c-i801.c:(.text+0x2c17bd): undefined reference to `i2c_new_device' Fixes: 2a7a0e9bf7b3 ("watchdog: iTCO_wdt: Add support for TCO on Intel Sunrisepoint") Reviewed-by: Matt Fleming Cc: Lee Jones Signed-off-by: Guenter Roeck Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig index c68edc16aa54..79e1aa1b0959 100644 --- a/drivers/watchdog/Kconfig +++ b/drivers/watchdog/Kconfig @@ -817,8 +817,9 @@ config ITCO_WDT tristate "Intel TCO Timer/Watchdog" depends on (X86 || IA64) && PCI select WATCHDOG_CORE + depends on I2C || I2C=n select LPC_ICH if !EXPERT - select I2C_I801 if !EXPERT + select I2C_I801 if !EXPERT && I2C ---help--- Hardware driver for the intel TCO timer based watchdog devices. These drivers are included in the Intel 82801 I/O Controller From 9bac175d8ed0b1dd3d3611c0713666b724eeace3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Radim=20Kr=C4=8Dm=C3=A1=C5=99?= Date: Fri, 18 Sep 2015 17:54:30 +0200 Subject: [PATCH 506/647] Revert "KVM: x86: zero kvmclock_offset when vcpu0 initializes kvmclock system MSR" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Shifting pvclock_vcpu_time_info.system_time on write to KVM system time MSR is a change of ABI. Probably only 2.6.16 based SLES 10 breaks due to its custom enhancements to kvmclock, but KVM never declared the MSR only for one-shot initialization. (Doc says that only one write is needed.) This reverts commit b7e60c5aedd2b63f16ef06fde4f81ca032211bc5. And adds a note to the definition of PVCLOCK_COUNTS_FROM_ZERO. Signed-off-by: Radim Krčmář Acked-by: Marcelo Tosatti Signed-off-by: Paolo Bonzini --- arch/x86/include/asm/pvclock-abi.h | 1 + arch/x86/kvm/x86.c | 4 ---- 2 files changed, 1 insertion(+), 4 deletions(-) diff --git a/arch/x86/include/asm/pvclock-abi.h b/arch/x86/include/asm/pvclock-abi.h index 655e07a48f6c..67f08230103a 100644 --- a/arch/x86/include/asm/pvclock-abi.h +++ b/arch/x86/include/asm/pvclock-abi.h @@ -41,6 +41,7 @@ struct pvclock_wall_clock { #define PVCLOCK_TSC_STABLE_BIT (1 << 0) #define PVCLOCK_GUEST_STOPPED (1 << 1) +/* PVCLOCK_COUNTS_FROM_ZERO broke ABI and can't be used anymore. */ #define PVCLOCK_COUNTS_FROM_ZERO (1 << 2) #endif /* __ASSEMBLY__ */ #endif /* _ASM_X86_PVCLOCK_ABI_H */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 991466bf8dee..92511d4b7236 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1708,8 +1708,6 @@ static int kvm_guest_time_update(struct kvm_vcpu *v) vcpu->pvclock_set_guest_stopped_request = false; } - pvclock_flags |= PVCLOCK_COUNTS_FROM_ZERO; - /* If the host uses TSC clocksource, then it is stable */ if (use_master_clock) pvclock_flags |= PVCLOCK_TSC_STABLE_BIT; @@ -2007,8 +2005,6 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info) &vcpu->requests); ka->boot_vcpu_runs_old_kvmclock = tmp; - - ka->kvmclock_offset = -get_kernel_ns(); } vcpu->arch.time = data; From dfc53c5e73f8b73abf920241e45eab87335ae742 Mon Sep 17 00:00:00 2001 From: Michel Thierry Date: Mon, 28 Sep 2015 13:25:12 +0100 Subject: [PATCH 507/647] drm/i915: Consider HW CSB write pointer before resetting the sw read pointer A previous commit resets the Context Status Buffer (CSB) read pointer in ring init commit c0a03a2e4c4e ("drm/i915: Reset CSB read pointer in ring init") This is generally correct, but this pointer is not reset after suspend/resume in some platforms (cht). In this case, the driver should read the register value instead of resetting the sw read counter to 0. Otherwise we process old events, leading to unwanted pre-emptions or something worse. But in other platforms (bdw) and also during GPU reset or power up, the CSBWP is reset to 0x7 (an invalid number), and in this case the read pointer should be set to 5 (the interrupt code will increment this counter one more time, and will start reading from CSB[0]). v2: When the CSB registers are reset, the read pointer needs to be set to 5, otherwise the first write (CSB[0]) won't be read (Mika). Replace magic numbers with GEN8_CSB_ENTRIES (6) and GEN8_CSB_PTR_MASK (0x07). Cc: Mika Kuoppala Cc: stable@vger.kernel.org # v4.0+ Signed-off-by: Lei Shen Signed-off-by: Deepak S Signed-off-by: Michel Thierry Reviewed-by: Mika Kuoppala Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_lrc.c | 39 ++++++++++++++++++++++++++------ drivers/gpu/drm/i915/intel_lrc.h | 2 ++ 2 files changed, 34 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 72e0edd7bbde..7412caedcf7f 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -484,18 +484,18 @@ void intel_lrc_irq_handler(struct intel_engine_cs *ring) status_pointer = I915_READ(RING_CONTEXT_STATUS_PTR(ring)); read_pointer = ring->next_context_status_buffer; - write_pointer = status_pointer & 0x07; + write_pointer = status_pointer & GEN8_CSB_PTR_MASK; if (read_pointer > write_pointer) - write_pointer += 6; + write_pointer += GEN8_CSB_ENTRIES; spin_lock(&ring->execlist_lock); while (read_pointer < write_pointer) { read_pointer++; status = I915_READ(RING_CONTEXT_STATUS_BUF(ring) + - (read_pointer % 6) * 8); + (read_pointer % GEN8_CSB_ENTRIES) * 8); status_id = I915_READ(RING_CONTEXT_STATUS_BUF(ring) + - (read_pointer % 6) * 8 + 4); + (read_pointer % GEN8_CSB_ENTRIES) * 8 + 4); if (status & GEN8_CTX_STATUS_IDLE_ACTIVE) continue; @@ -521,10 +521,12 @@ void intel_lrc_irq_handler(struct intel_engine_cs *ring) spin_unlock(&ring->execlist_lock); WARN(submit_contexts > 2, "More than two context complete events?\n"); - ring->next_context_status_buffer = write_pointer % 6; + ring->next_context_status_buffer = write_pointer % GEN8_CSB_ENTRIES; I915_WRITE(RING_CONTEXT_STATUS_PTR(ring), - _MASKED_FIELD(0x07 << 8, ((u32)ring->next_context_status_buffer & 0x07) << 8)); + _MASKED_FIELD(GEN8_CSB_PTR_MASK << 8, + ((u32)ring->next_context_status_buffer & + GEN8_CSB_PTR_MASK) << 8)); } static int execlists_context_queue(struct drm_i915_gem_request *request) @@ -1422,6 +1424,7 @@ static int gen8_init_common_ring(struct intel_engine_cs *ring) { struct drm_device *dev = ring->dev; struct drm_i915_private *dev_priv = dev->dev_private; + u8 next_context_status_buffer_hw; I915_WRITE_IMR(ring, ~(ring->irq_enable_mask | ring->irq_keep_mask)); I915_WRITE(RING_HWSTAM(ring->mmio_base), 0xffffffff); @@ -1436,7 +1439,29 @@ static int gen8_init_common_ring(struct intel_engine_cs *ring) _MASKED_BIT_DISABLE(GFX_REPLAY_MODE) | _MASKED_BIT_ENABLE(GFX_RUN_LIST_ENABLE)); POSTING_READ(RING_MODE_GEN7(ring)); - ring->next_context_status_buffer = 0; + + /* + * Instead of resetting the Context Status Buffer (CSB) read pointer to + * zero, we need to read the write pointer from hardware and use its + * value because "this register is power context save restored". + * Effectively, these states have been observed: + * + * | Suspend-to-idle (freeze) | Suspend-to-RAM (mem) | + * BDW | CSB regs not reset | CSB regs reset | + * CHT | CSB regs not reset | CSB regs not reset | + */ + next_context_status_buffer_hw = (I915_READ(RING_CONTEXT_STATUS_PTR(ring)) + & GEN8_CSB_PTR_MASK); + + /* + * When the CSB registers are reset (also after power-up / gpu reset), + * CSB write pointer is set to all 1's, which is not valid, use '5' in + * this special case, so the first element read is CSB[0]. + */ + if (next_context_status_buffer_hw == GEN8_CSB_PTR_MASK) + next_context_status_buffer_hw = (GEN8_CSB_ENTRIES - 1); + + ring->next_context_status_buffer = next_context_status_buffer_hw; DRM_DEBUG_DRIVER("Execlists enabled for %s\n", ring->name); memset(&ring->hangcheck, 0, sizeof(ring->hangcheck)); diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h index 64f89f9982a2..3c63bb32ad81 100644 --- a/drivers/gpu/drm/i915/intel_lrc.h +++ b/drivers/gpu/drm/i915/intel_lrc.h @@ -25,6 +25,8 @@ #define _INTEL_LRC_H_ #define GEN8_LR_CONTEXT_ALIGN 4096 +#define GEN8_CSB_ENTRIES 6 +#define GEN8_CSB_PTR_MASK 0x07 /* Execlists regs */ #define RING_ELSP(ring) ((ring)->mmio_base+0x230) From 3a48d13d76c0088a988a2e4f5b4d94872bdf58f3 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Mon, 7 Sep 2015 20:06:57 +0530 Subject: [PATCH 508/647] tile: fix build failure When building with allmodconfig the build was failing with the error: arch/tile/kernel/usb.c:70:1: warning: data definition has no type or storage class [enabled by default] arch/tile/kernel/usb.c:70:1: error: type defaults to 'int' in declaration of 'arch_initcall' [-Werror=implicit-int] arch/tile/kernel/usb.c:70:1: warning: parameter names (without types) in function declaration [enabled by default] arch/tile/kernel/usb.c:63:19: warning: 'tilegx_usb_init' defined but not used [-Wunused-function] Include linux/module.h to resolve the build failure. Signed-off-by: Sudip Mukherjee Signed-off-by: Chris Metcalf --- arch/tile/kernel/usb.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/tile/kernel/usb.c b/arch/tile/kernel/usb.c index f0da5a237e94..9f1e05e12255 100644 --- a/arch/tile/kernel/usb.c +++ b/arch/tile/kernel/usb.c @@ -22,6 +22,7 @@ #include #include #include +#include #include static u64 ehci_dmamask = DMA_BIT_MASK(32); From d79bdc7f004404204a6ac07785f8d6717070ecdb Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Mon, 28 Sep 2015 15:59:22 -0700 Subject: [PATCH 509/647] Input: omap4-keypad - fix memory leak If omap4_keypad_parse_dt() fails we returned the error code but we missed releasing keypad_data. Signed-off-by: Sudip Mukherjee Signed-off-by: Dmitry Torokhov --- drivers/input/keyboard/omap4-keypad.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/keyboard/omap4-keypad.c b/drivers/input/keyboard/omap4-keypad.c index b052afec9a11..6639b2b8528a 100644 --- a/drivers/input/keyboard/omap4-keypad.c +++ b/drivers/input/keyboard/omap4-keypad.c @@ -266,7 +266,7 @@ static int omap4_keypad_probe(struct platform_device *pdev) error = omap4_keypad_parse_dt(&pdev->dev, keypad_data); if (error) - return error; + goto err_free_keypad; res = request_mem_region(res->start, resource_size(res), pdev->name); if (!res) { From d3b367bc26ea2e07a83fe73f0ccbddd729cb1f9a Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 28 Sep 2015 16:38:07 -0700 Subject: [PATCH 510/647] Input: zhenhua - ensure we have BITREVERSE It uses bitrev8(), so it must ensure that lib/bitrev.o gets included in vmlinux. Signed-off-by: Andrew Morton Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/joystick/Kconfig b/drivers/input/joystick/Kconfig index 56eb471b5576..4215b5382092 100644 --- a/drivers/input/joystick/Kconfig +++ b/drivers/input/joystick/Kconfig @@ -196,6 +196,7 @@ config JOYSTICK_TWIDJOY config JOYSTICK_ZHENHUA tristate "5-byte Zhenhua RC transmitter" select SERIO + select BITREVERSE help Say Y here if you have a Zhen Hua PPM-4CH transmitter which is supplied with a ready to fly micro electric indoor helicopters From 22ef28b43f2c70edf5618918a49cbda84795c0a5 Mon Sep 17 00:00:00 2001 From: duson Date: Mon, 28 Sep 2015 17:17:01 -0700 Subject: [PATCH 511/647] Input: elan_i2c - add all valid ic type for i2c/smbus Signed-off-by: Duson Lin Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/elan_i2c_core.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/input/mouse/elan_i2c_core.c b/drivers/input/mouse/elan_i2c_core.c index b51062622050..5e1665bbaa0b 100644 --- a/drivers/input/mouse/elan_i2c_core.c +++ b/drivers/input/mouse/elan_i2c_core.c @@ -98,16 +98,25 @@ static int elan_get_fwinfo(u8 iap_version, u16 *validpage_count, u16 *signature_address) { switch (iap_version) { + case 0x00: + case 0x06: case 0x08: *validpage_count = 512; break; - case 0x09: case 0x03: + case 0x07: + case 0x09: + case 0x0A: + case 0x0B: + case 0x0C: *validpage_count = 768; break; case 0x0D: *validpage_count = 896; break; + case 0x0E: + *validpage_count = 640; + break; default: /* unknown ic type clear value */ *validpage_count = 0; From c4bbac3913c0d649898a0d767728a585869a7d7d Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Mon, 28 Sep 2015 11:21:48 -0700 Subject: [PATCH 512/647] i40e: fix VLAN inside VXLAN Previously to this patch, the hardware was removing VLAN tags from the inner header of VXLAN packets. The hardware configuration can be changed to leave the packet alone since that is what the linux stack expects for this type of VLAN in VXLAN packet. Signed-off-by: Jesse Brandeburg Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher Signed-off-by: David S. Miller --- drivers/net/ethernet/intel/i40e/i40e_main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 851c1a159be8..2fdf978ae6a5 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -2672,7 +2672,8 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring) rx_ctx.lrxqthresh = 2; rx_ctx.crcstrip = 1; rx_ctx.l2tsel = 1; - rx_ctx.showiv = 1; + /* this controls whether VLAN is stripped from inner headers */ + rx_ctx.showiv = 0; #ifdef I40E_FCOE rx_ctx.fc_ena = (vsi->type == I40E_VSI_FCOE); #endif From 43ae93a93e8c95c5e6389dc8e11704712b1ab2e9 Mon Sep 17 00:00:00 2001 From: Mitch Williams Date: Mon, 28 Sep 2015 17:31:26 -0700 Subject: [PATCH 513/647] i40e/i40evf: check for stopped admin queue It's possible that while we are waiting for the spinlock, another entity (that owns the spinlock) has shut down the admin queue. If we then attempt to use the queue, we will panic. Add a check for this condition on the receive side. This matches an existing check on the send queue side. Signed-off-by: Mitch Williams Acked-by: Jesse Brandeburg Signed-off-by: Jeff Kirsher Signed-off-by: David S. Miller --- drivers/net/ethernet/intel/i40e/i40e_adminq.c | 9 +++++++++ drivers/net/ethernet/intel/i40evf/i40e_adminq.c | 9 +++++++++ 2 files changed, 18 insertions(+) diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.c b/drivers/net/ethernet/intel/i40e/i40e_adminq.c index 3e0d20037675..62488a67149d 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_adminq.c +++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.c @@ -946,6 +946,13 @@ i40e_status i40e_clean_arq_element(struct i40e_hw *hw, /* take the lock before we start messing with the ring */ mutex_lock(&hw->aq.arq_mutex); + if (hw->aq.arq.count == 0) { + i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE, + "AQRX: Admin queue not initialized.\n"); + ret_code = I40E_ERR_QUEUE_EMPTY; + goto clean_arq_element_err; + } + /* set next_to_use to head */ ntu = (rd32(hw, hw->aq.arq.head) & I40E_PF_ARQH_ARQH_MASK); if (ntu == ntc) { @@ -1007,6 +1014,8 @@ clean_arq_element_out: /* Set pending if needed, unlock and return */ if (pending != NULL) *pending = (ntc > ntu ? hw->aq.arq.count : 0) + (ntu - ntc); + +clean_arq_element_err: mutex_unlock(&hw->aq.arq_mutex); if (i40e_is_nvm_update_op(&e->desc)) { diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq.c b/drivers/net/ethernet/intel/i40evf/i40e_adminq.c index f08450b90774..929d47152bf2 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_adminq.c +++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq.c @@ -887,6 +887,13 @@ i40e_status i40evf_clean_arq_element(struct i40e_hw *hw, /* take the lock before we start messing with the ring */ mutex_lock(&hw->aq.arq_mutex); + if (hw->aq.arq.count == 0) { + i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE, + "AQRX: Admin queue not initialized.\n"); + ret_code = I40E_ERR_QUEUE_EMPTY; + goto clean_arq_element_err; + } + /* set next_to_use to head */ ntu = (rd32(hw, hw->aq.arq.head) & I40E_VF_ARQH1_ARQH_MASK); if (ntu == ntc) { @@ -948,6 +955,8 @@ clean_arq_element_out: /* Set pending if needed, unlock and return */ if (pending != NULL) *pending = (ntc > ntu ? hw->aq.arq.count : 0) + (ntu - ntc); + +clean_arq_element_err: mutex_unlock(&hw->aq.arq_mutex); return ret_code; From f05940e61845951517eda02a28ccc091888aaab9 Mon Sep 17 00:00:00 2001 From: Karl Heiss Date: Thu, 24 Sep 2015 12:15:06 -0400 Subject: [PATCH 514/647] sctp: Whitespace fix Fix indentation in sctp_generate_heartbeat_event. Signed-off-by: Karl Heiss Signed-off-by: David S. Miller --- net/sctp/sm_sideeffect.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 35df1266bf07..f554b9a96e07 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -388,8 +388,8 @@ void sctp_generate_heartbeat_event(unsigned long data) asoc->state, asoc->ep, asoc, transport, GFP_ATOMIC); - if (error) - asoc->base.sk->sk_err = -error; + if (error) + asoc->base.sk->sk_err = -error; out_unlock: bh_unlock_sock(asoc->base.sk); From 635682a14427d241bab7bbdeebb48a7d7b91638e Mon Sep 17 00:00:00 2001 From: Karl Heiss Date: Thu, 24 Sep 2015 12:15:07 -0400 Subject: [PATCH 515/647] sctp: Prevent soft lockup when sctp_accept() is called during a timeout event A case can occur when sctp_accept() is called by the user during a heartbeat timeout event after the 4-way handshake. Since sctp_assoc_migrate() changes both assoc->base.sk and assoc->ep, the bh_sock_lock in sctp_generate_heartbeat_event() will be taken with the listening socket but released with the new association socket. The result is a deadlock on any future attempts to take the listening socket lock. Note that this race can occur with other SCTP timeouts that take the bh_lock_sock() in the event sctp_accept() is called. BUG: soft lockup - CPU#9 stuck for 67s! [swapper:0] ... RIP: 0010:[] [] _spin_lock+0x1e/0x30 RSP: 0018:ffff880028323b20 EFLAGS: 00000206 RAX: 0000000000000002 RBX: ffff880028323b20 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffff880028323be0 RDI: ffff8804632c4b48 RBP: ffffffff8100bb93 R08: 0000000000000000 R09: 0000000000000000 R10: ffff880610662280 R11: 0000000000000100 R12: ffff880028323aa0 R13: ffff8804383c3880 R14: ffff880028323a90 R15: ffffffff81534225 FS: 0000000000000000(0000) GS:ffff880028320000(0000) knlGS:0000000000000000 CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b CR2: 00000000006df528 CR3: 0000000001a85000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process swapper (pid: 0, threadinfo ffff880616b70000, task ffff880616b6cab0) Stack: ffff880028323c40 ffffffffa01c2582 ffff880614cfb020 0000000000000000 0100000000000000 00000014383a6c44 ffff8804383c3880 ffff880614e93c00 ffff880614e93c00 0000000000000000 ffff8804632c4b00 ffff8804383c38b8 Call Trace: [] ? sctp_rcv+0x492/0xa10 [sctp] [] ? nf_iterate+0x69/0xb0 [] ? ip_local_deliver_finish+0x0/0x2d0 [] ? nf_hook_slow+0x76/0x120 [] ? ip_local_deliver_finish+0x0/0x2d0 [] ? ip_local_deliver_finish+0xdd/0x2d0 [] ? ip_local_deliver+0x98/0xa0 [] ? ip_rcv_finish+0x12d/0x440 [] ? ip_rcv+0x275/0x350 [] ? __netif_receive_skb+0x4ab/0x750 ... With lockdep debugging: ===================================== [ BUG: bad unlock balance detected! ] ------------------------------------- CslRx/12087 is trying to release lock (slock-AF_INET) at: [] sctp_generate_timeout_event+0x40/0xe0 [sctp] but there are no more locks to release! other info that might help us debug this: 2 locks held by CslRx/12087: #0: (&asoc->timers[i]){+.-...}, at: [] run_timer_softirq+0x16f/0x3e0 #1: (slock-AF_INET){+.-...}, at: [] sctp_generate_timeout_event+0x23/0xe0 [sctp] Ensure the socket taken is also the same one that is released by saving a copy of the socket before entering the timeout event critical section. Signed-off-by: Karl Heiss Signed-off-by: David S. Miller --- net/sctp/sm_sideeffect.c | 42 ++++++++++++++++++++++------------------ 1 file changed, 23 insertions(+), 19 deletions(-) diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index f554b9a96e07..6098d4c42fa9 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -244,12 +244,13 @@ void sctp_generate_t3_rtx_event(unsigned long peer) int error; struct sctp_transport *transport = (struct sctp_transport *) peer; struct sctp_association *asoc = transport->asoc; - struct net *net = sock_net(asoc->base.sk); + struct sock *sk = asoc->base.sk; + struct net *net = sock_net(sk); /* Check whether a task is in the sock. */ - bh_lock_sock(asoc->base.sk); - if (sock_owned_by_user(asoc->base.sk)) { + bh_lock_sock(sk); + if (sock_owned_by_user(sk)) { pr_debug("%s: sock is busy\n", __func__); /* Try again later. */ @@ -272,10 +273,10 @@ void sctp_generate_t3_rtx_event(unsigned long peer) transport, GFP_ATOMIC); if (error) - asoc->base.sk->sk_err = -error; + sk->sk_err = -error; out_unlock: - bh_unlock_sock(asoc->base.sk); + bh_unlock_sock(sk); sctp_transport_put(transport); } @@ -285,11 +286,12 @@ out_unlock: static void sctp_generate_timeout_event(struct sctp_association *asoc, sctp_event_timeout_t timeout_type) { - struct net *net = sock_net(asoc->base.sk); + struct sock *sk = asoc->base.sk; + struct net *net = sock_net(sk); int error = 0; - bh_lock_sock(asoc->base.sk); - if (sock_owned_by_user(asoc->base.sk)) { + bh_lock_sock(sk); + if (sock_owned_by_user(sk)) { pr_debug("%s: sock is busy: timer %d\n", __func__, timeout_type); @@ -312,10 +314,10 @@ static void sctp_generate_timeout_event(struct sctp_association *asoc, (void *)timeout_type, GFP_ATOMIC); if (error) - asoc->base.sk->sk_err = -error; + sk->sk_err = -error; out_unlock: - bh_unlock_sock(asoc->base.sk); + bh_unlock_sock(sk); sctp_association_put(asoc); } @@ -365,10 +367,11 @@ void sctp_generate_heartbeat_event(unsigned long data) int error = 0; struct sctp_transport *transport = (struct sctp_transport *) data; struct sctp_association *asoc = transport->asoc; - struct net *net = sock_net(asoc->base.sk); + struct sock *sk = asoc->base.sk; + struct net *net = sock_net(sk); - bh_lock_sock(asoc->base.sk); - if (sock_owned_by_user(asoc->base.sk)) { + bh_lock_sock(sk); + if (sock_owned_by_user(sk)) { pr_debug("%s: sock is busy\n", __func__); /* Try again later. */ @@ -389,10 +392,10 @@ void sctp_generate_heartbeat_event(unsigned long data) transport, GFP_ATOMIC); if (error) - asoc->base.sk->sk_err = -error; + sk->sk_err = -error; out_unlock: - bh_unlock_sock(asoc->base.sk); + bh_unlock_sock(sk); sctp_transport_put(transport); } @@ -403,10 +406,11 @@ void sctp_generate_proto_unreach_event(unsigned long data) { struct sctp_transport *transport = (struct sctp_transport *) data; struct sctp_association *asoc = transport->asoc; - struct net *net = sock_net(asoc->base.sk); + struct sock *sk = asoc->base.sk; + struct net *net = sock_net(sk); - bh_lock_sock(asoc->base.sk); - if (sock_owned_by_user(asoc->base.sk)) { + bh_lock_sock(sk); + if (sock_owned_by_user(sk)) { pr_debug("%s: sock is busy\n", __func__); /* Try again later. */ @@ -427,7 +431,7 @@ void sctp_generate_proto_unreach_event(unsigned long data) asoc->state, asoc->ep, asoc, transport, GFP_ATOMIC); out_unlock: - bh_unlock_sock(asoc->base.sk); + bh_unlock_sock(sk); sctp_association_put(asoc); } From 661dfc65f7981481ba2e31aaa702371e82336e56 Mon Sep 17 00:00:00 2001 From: Ivan Mikhaylov Date: Fri, 25 Sep 2015 11:52:27 +0400 Subject: [PATCH 516/647] net/ibm/emac: bump version numbers for correct work with ethtool The size of the MAC register dump used to be the size specified by the reg property in the device tree. Userland has no good way of finding out that size, and it was not specified consistently for each MAC type, so ethtool would end up printing junk at the end of the register dump if the device tree didn't match the size it assumed. Using the new version numbers indicates unambiguously that the size of the MAC register dump is dependent only on the MAC type. Fixes: 5369c71f7ca2 ("net/ibm/emac: fix size of emac dump memory areas") Signed-off-by: Ivan Mikhaylov Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/emac/core.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/ibm/emac/core.h b/drivers/net/ethernet/ibm/emac/core.h index 28df37420da9..ac02c675c59c 100644 --- a/drivers/net/ethernet/ibm/emac/core.h +++ b/drivers/net/ethernet/ibm/emac/core.h @@ -460,8 +460,8 @@ struct emac_ethtool_regs_subhdr { u32 index; }; -#define EMAC_ETHTOOL_REGS_VER 0 -#define EMAC4_ETHTOOL_REGS_VER 1 -#define EMAC4SYNC_ETHTOOL_REGS_VER 2 +#define EMAC_ETHTOOL_REGS_VER 3 +#define EMAC4_ETHTOOL_REGS_VER 4 +#define EMAC4SYNC_ETHTOOL_REGS_VER 5 #endif /* __IBM_NEWEMAC_CORE_H */ From 06a15f51cf3618e32a73871ee6a547ef7fd902b5 Mon Sep 17 00:00:00 2001 From: Alexander Couzens Date: Mon, 28 Sep 2015 11:32:42 +0200 Subject: [PATCH 517/647] l2tp: protect tunnel->del_work by ref_count There is a small chance that tunnel_free() is called before tunnel->del_work scheduled resulting in a zero pointer dereference. Signed-off-by: Alexander Couzens Acked-by: James Chapman Signed-off-by: David S. Miller --- net/l2tp/l2tp_core.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index f6b090df3930..afca2eb4dfa7 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -1319,7 +1319,7 @@ static void l2tp_tunnel_del_work(struct work_struct *work) tunnel = container_of(work, struct l2tp_tunnel, del_work); sk = l2tp_tunnel_sock_lookup(tunnel); if (!sk) - return; + goto out; sock = sk->sk_socket; @@ -1341,6 +1341,8 @@ static void l2tp_tunnel_del_work(struct work_struct *work) } l2tp_tunnel_sock_put(sk); +out: + l2tp_tunnel_dec_refcount(tunnel); } /* Create a socket for the tunnel, if one isn't set up by @@ -1636,8 +1638,13 @@ EXPORT_SYMBOL_GPL(l2tp_tunnel_create); */ int l2tp_tunnel_delete(struct l2tp_tunnel *tunnel) { + l2tp_tunnel_inc_refcount(tunnel); l2tp_tunnel_closeall(tunnel); - return (false == queue_work(l2tp_wq, &tunnel->del_work)); + if (false == queue_work(l2tp_wq, &tunnel->del_work)) { + l2tp_tunnel_dec_refcount(tunnel); + return 1; + } + return 0; } EXPORT_SYMBOL_GPL(l2tp_tunnel_delete); From 2103d6b818fcdae15ffa04cf385f770e6c3892c3 Mon Sep 17 00:00:00 2001 From: Denys Vlasenko Date: Mon, 28 Sep 2015 14:34:04 +0200 Subject: [PATCH 518/647] net: sctp: Don't use 64 kilobyte lookup table for four elements Seemingly innocuous sctp_trans_state_to_prio_map[] array is way bigger than it looks, since "[SCTP_UNKNOWN] = 2" expands into "[0xffff] = 2" ! This patch replaces it with switch() statement. Signed-off-by: Denys Vlasenko CC: Vlad Yasevich CC: Neil Horman CC: Marcelo Ricardo Leitner CC: linux-sctp@vger.kernel.org CC: netdev@vger.kernel.org CC: linux-kernel@vger.kernel.org Acked-by: Marcelo Ricardo Leitner Acked-by: Neil Horman Signed-off-by: David S. Miller --- net/sctp/associola.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 197c3f59ecbf..b00f1f9611d6 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -1208,20 +1208,22 @@ void sctp_assoc_update(struct sctp_association *asoc, * within this document. * * Our basic strategy is to round-robin transports in priorities - * according to sctp_state_prio_map[] e.g., if no such + * according to sctp_trans_score() e.g., if no such * transport with state SCTP_ACTIVE exists, round-robin through * SCTP_UNKNOWN, etc. You get the picture. */ -static const u8 sctp_trans_state_to_prio_map[] = { - [SCTP_ACTIVE] = 3, /* best case */ - [SCTP_UNKNOWN] = 2, - [SCTP_PF] = 1, - [SCTP_INACTIVE] = 0, /* worst case */ -}; - static u8 sctp_trans_score(const struct sctp_transport *trans) { - return sctp_trans_state_to_prio_map[trans->state]; + switch (trans->state) { + case SCTP_ACTIVE: + return 3; /* best case */ + case SCTP_UNKNOWN: + return 2; + case SCTP_PF: + return 1; + default: /* case SCTP_INACTIVE */ + return 0; /* worst case */ + } } static struct sctp_transport *sctp_trans_elect_tie(struct sctp_transport *trans1, From 43934ece2ea72c1dd279c0b0478c1a036d5d77ee Mon Sep 17 00:00:00 2001 From: Ulf Hansson Date: Mon, 14 Sep 2015 12:18:55 +0200 Subject: [PATCH 519/647] mmc: core: Don't return an error for CD/WP GPIOs when GPIOLIB is unset When CONFIG_GPIOLIB is unset, its stubs will return -ENOSYS. That means when the mmc core parses DT for CD/WP GPIOs via mmc_of_parse(), -ENOSYS becomes propagated to the caller. Typically this means that the mmc host driver fails to probe. As the CD/WP GPIOs are already treated as optional, let's extend that to cover the case when CONFIG_GPIOLIB is unset. Reported-by: Michal Simek Fixes: 16b23787fc70 ("mmc: sdhci-of-arasan: Call OF parsing for MMC") Signed-off-by: Ulf Hansson Tested-by: Michal Simek Acked-by: Venu Byravarasu --- drivers/mmc/core/host.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c index abd933b7029b..5466f25f0281 100644 --- a/drivers/mmc/core/host.c +++ b/drivers/mmc/core/host.c @@ -457,7 +457,7 @@ int mmc_of_parse(struct mmc_host *host) 0, &cd_gpio_invert); if (!ret) dev_info(host->parent, "Got CD GPIO\n"); - else if (ret != -ENOENT) + else if (ret != -ENOENT && ret != -ENOSYS) return ret; /* @@ -481,7 +481,7 @@ int mmc_of_parse(struct mmc_host *host) ret = mmc_gpiod_request_ro(host, "wp", 0, false, 0, &ro_gpio_invert); if (!ret) dev_info(host->parent, "Got WP GPIO\n"); - else if (ret != -ENOENT) + else if (ret != -ENOENT && ret != -ENOSYS) return ret; if (of_property_read_bool(np, "disable-wp")) From 51424b2860670ec20e1dd5177fe70ab4b6fd7a5b Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Wed, 23 Sep 2015 22:06:48 +0200 Subject: [PATCH 520/647] mmc: sunxi: Fix clk-delay settings In recent allwinner kernel sources the mmc clk-delay settings have been slightly tweaked, and for sun9i they are completely different then what we are using. This commit brings us in sync with what allwinner does, fixing problems accessing sdcards on some A33 devices (and likely others). For pre sun9i hardware this makes the following changes: -At 400Khz change the sample delay from 7 to 0 (introduced in A31 sdk) -At 50 Mhz change the sample delay from 5 to 4 (introduced in A23 sdk) This also drops the clk-delay calculation for clocks > 50 MHz, we do not need this as we've: mmc->f_max = 50000000, and the delays in the old code were not correct (at 100 MHz the delay must be a multiple of 60, at 200 MHz a multiple of 120). Signed-off-by: Hans de Goede Signed-off-by: Ulf Hansson --- drivers/mmc/host/sunxi-mmc.c | 53 ++++++++++++++++++++++++++---------- 1 file changed, 39 insertions(+), 14 deletions(-) diff --git a/drivers/mmc/host/sunxi-mmc.c b/drivers/mmc/host/sunxi-mmc.c index a7b7a6771598..b981b8552e43 100644 --- a/drivers/mmc/host/sunxi-mmc.c +++ b/drivers/mmc/host/sunxi-mmc.c @@ -210,6 +210,16 @@ #define SDXC_IDMAC_DES0_CES BIT(30) /* card error summary */ #define SDXC_IDMAC_DES0_OWN BIT(31) /* 1-idma owns it, 0-host owns it */ +#define SDXC_CLK_400K 0 +#define SDXC_CLK_25M 1 +#define SDXC_CLK_50M 2 +#define SDXC_CLK_50M_DDR 3 + +struct sunxi_mmc_clk_delay { + u32 output; + u32 sample; +}; + struct sunxi_idma_des { u32 config; u32 buf_size; @@ -229,6 +239,7 @@ struct sunxi_mmc_host { struct clk *clk_mmc; struct clk *clk_sample; struct clk *clk_output; + const struct sunxi_mmc_clk_delay *clk_delays; /* irq */ spinlock_t lock; @@ -654,25 +665,19 @@ static int sunxi_mmc_clk_set_rate(struct sunxi_mmc_host *host, /* determine delays */ if (rate <= 400000) { - oclk_dly = 180; - sclk_dly = 42; + oclk_dly = host->clk_delays[SDXC_CLK_400K].output; + sclk_dly = host->clk_delays[SDXC_CLK_400K].sample; } else if (rate <= 25000000) { - oclk_dly = 180; - sclk_dly = 75; + oclk_dly = host->clk_delays[SDXC_CLK_25M].output; + sclk_dly = host->clk_delays[SDXC_CLK_25M].sample; } else if (rate <= 50000000) { if (ios->timing == MMC_TIMING_UHS_DDR50) { - oclk_dly = 60; - sclk_dly = 120; + oclk_dly = host->clk_delays[SDXC_CLK_50M_DDR].output; + sclk_dly = host->clk_delays[SDXC_CLK_50M_DDR].sample; } else { - oclk_dly = 90; - sclk_dly = 150; + oclk_dly = host->clk_delays[SDXC_CLK_50M].output; + sclk_dly = host->clk_delays[SDXC_CLK_50M].sample; } - } else if (rate <= 100000000) { - oclk_dly = 6; - sclk_dly = 24; - } else if (rate <= 200000000) { - oclk_dly = 3; - sclk_dly = 12; } else { return -EINVAL; } @@ -871,6 +876,7 @@ static void sunxi_mmc_request(struct mmc_host *mmc, struct mmc_request *mrq) static const struct of_device_id sunxi_mmc_of_match[] = { { .compatible = "allwinner,sun4i-a10-mmc", }, { .compatible = "allwinner,sun5i-a13-mmc", }, + { .compatible = "allwinner,sun9i-a80-mmc", }, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, sunxi_mmc_of_match); @@ -884,6 +890,20 @@ static struct mmc_host_ops sunxi_mmc_ops = { .hw_reset = sunxi_mmc_hw_reset, }; +static const struct sunxi_mmc_clk_delay sunxi_mmc_clk_delays[] = { + [SDXC_CLK_400K] = { .output = 180, .sample = 180 }, + [SDXC_CLK_25M] = { .output = 180, .sample = 75 }, + [SDXC_CLK_50M] = { .output = 90, .sample = 120 }, + [SDXC_CLK_50M_DDR] = { .output = 60, .sample = 120 }, +}; + +static const struct sunxi_mmc_clk_delay sun9i_mmc_clk_delays[] = { + [SDXC_CLK_400K] = { .output = 180, .sample = 180 }, + [SDXC_CLK_25M] = { .output = 180, .sample = 75 }, + [SDXC_CLK_50M] = { .output = 150, .sample = 120 }, + [SDXC_CLK_50M_DDR] = { .output = 90, .sample = 120 }, +}; + static int sunxi_mmc_resource_request(struct sunxi_mmc_host *host, struct platform_device *pdev) { @@ -895,6 +915,11 @@ static int sunxi_mmc_resource_request(struct sunxi_mmc_host *host, else host->idma_des_size_bits = 16; + if (of_device_is_compatible(np, "allwinner,sun9i-a80-mmc")) + host->clk_delays = sun9i_mmc_clk_delays; + else + host->clk_delays = sunxi_mmc_clk_delays; + ret = mmc_regulator_get_supply(host->mmc); if (ret) { if (ret != -EPROBE_DEFER) From fd546ee6a7dc4b71ebc6d1205bf72ea3c1c7030a Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Sat, 26 Sep 2015 21:41:01 +0200 Subject: [PATCH 521/647] mmc: pxamci: fix card detect with slot-gpio API Move pxamci to mmc slot-gpio API to fix interrupt request. It fixes the case where the card detection is on a gpio expander, on I2C for example on zylonite board. In this case, the card detect netsted interrupt is called from a threaded interrupt. The request_irq() fails, because a hard irq cannot be a nested interrupt from a threaded interrupt (set __setup_irq()). This was tested on zylonite and mioa701 boards. Signed-off-by: Robert Jarzmik Cc: Petr Cvek Signed-off-by: Ulf Hansson --- drivers/mmc/host/pxamci.c | 66 +++++++++++++-------------------------- 1 file changed, 22 insertions(+), 44 deletions(-) diff --git a/drivers/mmc/host/pxamci.c b/drivers/mmc/host/pxamci.c index 1420f29628c7..8cadd74e8407 100644 --- a/drivers/mmc/host/pxamci.c +++ b/drivers/mmc/host/pxamci.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include #include @@ -454,12 +455,8 @@ static int pxamci_get_ro(struct mmc_host *mmc) { struct pxamci_host *host = mmc_priv(mmc); - if (host->pdata && gpio_is_valid(host->pdata->gpio_card_ro)) { - if (host->pdata->gpio_card_ro_invert) - return !gpio_get_value(host->pdata->gpio_card_ro); - else - return gpio_get_value(host->pdata->gpio_card_ro); - } + if (host->pdata && gpio_is_valid(host->pdata->gpio_card_ro)) + return mmc_gpio_get_ro(mmc); if (host->pdata && host->pdata->get_ro) return !!host->pdata->get_ro(mmc_dev(mmc)); /* @@ -551,6 +548,7 @@ static void pxamci_enable_sdio_irq(struct mmc_host *host, int enable) static const struct mmc_host_ops pxamci_ops = { .request = pxamci_request, + .get_cd = mmc_gpio_get_cd, .get_ro = pxamci_get_ro, .set_ios = pxamci_set_ios, .enable_sdio_irq = pxamci_enable_sdio_irq, @@ -790,37 +788,31 @@ static int pxamci_probe(struct platform_device *pdev) gpio_power = host->pdata->gpio_power; } if (gpio_is_valid(gpio_power)) { - ret = gpio_request(gpio_power, "mmc card power"); + ret = devm_gpio_request(&pdev->dev, gpio_power, + "mmc card power"); if (ret) { - dev_err(&pdev->dev, "Failed requesting gpio_power %d\n", gpio_power); + dev_err(&pdev->dev, "Failed requesting gpio_power %d\n", + gpio_power); goto out; } gpio_direction_output(gpio_power, host->pdata->gpio_power_invert); } - if (gpio_is_valid(gpio_ro)) { - ret = gpio_request(gpio_ro, "mmc card read only"); - if (ret) { - dev_err(&pdev->dev, "Failed requesting gpio_ro %d\n", gpio_ro); - goto err_gpio_ro; - } - gpio_direction_input(gpio_ro); + if (gpio_is_valid(gpio_ro)) + ret = mmc_gpio_request_ro(mmc, gpio_ro); + if (ret) { + dev_err(&pdev->dev, "Failed requesting gpio_ro %d\n", gpio_ro); + goto out; + } else { + mmc->caps |= host->pdata->gpio_card_ro_invert ? + MMC_CAP2_RO_ACTIVE_HIGH : 0; } - if (gpio_is_valid(gpio_cd)) { - ret = gpio_request(gpio_cd, "mmc card detect"); - if (ret) { - dev_err(&pdev->dev, "Failed requesting gpio_cd %d\n", gpio_cd); - goto err_gpio_cd; - } - gpio_direction_input(gpio_cd); - ret = request_irq(gpio_to_irq(gpio_cd), pxamci_detect_irq, - IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, - "mmc card detect", mmc); - if (ret) { - dev_err(&pdev->dev, "failed to request card detect IRQ\n"); - goto err_request_irq; - } + if (gpio_is_valid(gpio_cd)) + ret = mmc_gpio_request_cd(mmc, gpio_cd, 0); + if (ret) { + dev_err(&pdev->dev, "Failed requesting gpio_cd %d\n", gpio_cd); + goto out; } if (host->pdata && host->pdata->init) @@ -835,13 +827,7 @@ static int pxamci_probe(struct platform_device *pdev) return 0; -err_request_irq: - gpio_free(gpio_cd); -err_gpio_cd: - gpio_free(gpio_ro); -err_gpio_ro: - gpio_free(gpio_power); - out: +out: if (host) { if (host->dma_chan_rx) dma_release_channel(host->dma_chan_rx); @@ -873,14 +859,6 @@ static int pxamci_remove(struct platform_device *pdev) gpio_ro = host->pdata->gpio_card_ro; gpio_power = host->pdata->gpio_power; } - if (gpio_is_valid(gpio_cd)) { - free_irq(gpio_to_irq(gpio_cd), mmc); - gpio_free(gpio_cd); - } - if (gpio_is_valid(gpio_ro)) - gpio_free(gpio_ro); - if (gpio_is_valid(gpio_power)) - gpio_free(gpio_power); if (host->vcc) regulator_put(host->vcc); From cf6f54e3f133229f02a90c04fe0ff9dd9d3264b4 Mon Sep 17 00:00:00 2001 From: Richard Weinberger Date: Wed, 8 Jul 2015 11:46:36 +0200 Subject: [PATCH 522/647] UBIFS: Kill unneeded locking in ubifs_init_security Fixes the following lockdep splat: [ 1.244527] ============================================= [ 1.245193] [ INFO: possible recursive locking detected ] [ 1.245193] 4.2.0-rc1+ #37 Not tainted [ 1.245193] --------------------------------------------- [ 1.245193] cp/742 is trying to acquire lock: [ 1.245193] (&sb->s_type->i_mutex_key#9){+.+.+.}, at: [] ubifs_init_security+0x29/0xb0 [ 1.245193] [ 1.245193] but task is already holding lock: [ 1.245193] (&sb->s_type->i_mutex_key#9){+.+.+.}, at: [] path_openat+0x3af/0x1280 [ 1.245193] [ 1.245193] other info that might help us debug this: [ 1.245193] Possible unsafe locking scenario: [ 1.245193] [ 1.245193] CPU0 [ 1.245193] ---- [ 1.245193] lock(&sb->s_type->i_mutex_key#9); [ 1.245193] lock(&sb->s_type->i_mutex_key#9); [ 1.245193] [ 1.245193] *** DEADLOCK *** [ 1.245193] [ 1.245193] May be due to missing lock nesting notation [ 1.245193] [ 1.245193] 2 locks held by cp/742: [ 1.245193] #0: (sb_writers#5){.+.+.+}, at: [] mnt_want_write+0x1f/0x50 [ 1.245193] #1: (&sb->s_type->i_mutex_key#9){+.+.+.}, at: [] path_openat+0x3af/0x1280 [ 1.245193] [ 1.245193] stack backtrace: [ 1.245193] CPU: 2 PID: 742 Comm: cp Not tainted 4.2.0-rc1+ #37 [ 1.245193] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140816_022509-build35 04/01/2014 [ 1.245193] ffffffff8252d530 ffff88007b023a38 ffffffff814f6f49 ffffffff810b56c5 [ 1.245193] ffff88007c30cc80 ffff88007b023af8 ffffffff810a150d ffff88007b023a68 [ 1.245193] 000000008101302a ffff880000000000 00000008f447e23f ffffffff8252d500 [ 1.245193] Call Trace: [ 1.245193] [] dump_stack+0x4c/0x65 [ 1.245193] [] ? console_unlock+0x1c5/0x510 [ 1.245193] [] __lock_acquire+0x1a6d/0x1ea0 [ 1.245193] [] ? __lock_is_held+0x58/0x80 [ 1.245193] [] lock_acquire+0xd3/0x270 [ 1.245193] [] ? ubifs_init_security+0x29/0xb0 [ 1.245193] [] mutex_lock_nested+0x6b/0x3a0 [ 1.245193] [] ? ubifs_init_security+0x29/0xb0 [ 1.245193] [] ? ubifs_init_security+0x29/0xb0 [ 1.245193] [] ubifs_init_security+0x29/0xb0 [ 1.245193] [] ubifs_create+0xa6/0x1f0 [ 1.245193] [] ? path_openat+0x3af/0x1280 [ 1.245193] [] vfs_create+0x95/0xc0 [ 1.245193] [] path_openat+0x7cc/0x1280 [ 1.245193] [] ? __lock_acquire+0x543/0x1ea0 [ 1.245193] [] ? sched_clock_cpu+0x90/0xc0 [ 1.245193] [] ? calc_global_load_tick+0x60/0x90 [ 1.245193] [] ? sched_clock_cpu+0x90/0xc0 [ 1.245193] [] ? __alloc_fd+0xaf/0x180 [ 1.245193] [] do_filp_open+0x75/0xd0 [ 1.245193] [] ? _raw_spin_unlock+0x26/0x40 [ 1.245193] [] ? __alloc_fd+0xaf/0x180 [ 1.245193] [] do_sys_open+0x129/0x200 [ 1.245193] [] SyS_open+0x19/0x20 [ 1.245193] [] entry_SYSCALL_64_fastpath+0x12/0x6f While the lockdep splat is a false positive, becuase path_openat holds i_mutex of the parent directory and ubifs_init_security() tries to acquire i_mutex of a new inode, it reveals that taking i_mutex in ubifs_init_security() is in vain because it is only being called in the inode allocation path and therefore nobody else can see the inode yet. Cc: stable@vger.kernel.org # 3.20- Reported-and-tested-by: Boris Brezillon Reviewed-and-tested-by: Dongsheng Yang Signed-off-by: Richard Weinberger Signed-off-by: dedekind1@gmail.com --- fs/ubifs/xattr.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c index 96f3448b6eb4..fd65b3f1923c 100644 --- a/fs/ubifs/xattr.c +++ b/fs/ubifs/xattr.c @@ -652,11 +652,8 @@ int ubifs_init_security(struct inode *dentry, struct inode *inode, { int err; - mutex_lock(&inode->i_mutex); err = security_inode_init_security(inode, dentry, qstr, &init_xattrs, 0); - mutex_unlock(&inode->i_mutex); - if (err) { struct ubifs_info *c = dentry->i_sb->s_fs_info; ubifs_err(c, "cannot initialize security for inode %lu, error %d", From 281fda27673f833a01d516658a64d22a32c8e072 Mon Sep 17 00:00:00 2001 From: Richard Weinberger Date: Tue, 22 Sep 2015 23:58:07 +0200 Subject: [PATCH 523/647] UBI: Validate data_size Make sure that data_size is less than LEB size. Otherwise a handcrafted UBI image is able to trigger an out of bounds memory access in ubi_compare_lebs(). Cc: stable@vger.kernel.org Signed-off-by: Richard Weinberger Reviewed-by: David Gstir --- drivers/mtd/ubi/io.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c index 5bbd1f094f4e..1fc23e48fe8e 100644 --- a/drivers/mtd/ubi/io.c +++ b/drivers/mtd/ubi/io.c @@ -926,6 +926,11 @@ static int validate_vid_hdr(const struct ubi_device *ubi, goto bad; } + if (data_size > ubi->leb_size) { + ubi_err(ubi, "bad data_size"); + goto bad; + } + if (vol_type == UBI_VID_STATIC) { /* * Although from high-level point of view static volumes may From 7c7feb2ebfc9c0552c51f0c050db1d1a004faac5 Mon Sep 17 00:00:00 2001 From: shengyong Date: Mon, 28 Sep 2015 17:57:19 +0000 Subject: [PATCH 524/647] UBI: return ENOSPC if no enough space available UBI: attaching mtd1 to ubi0 UBI: scanning is finished UBI error: init_volumes: not enough PEBs, required 706, available 686 UBI error: ubi_wl_init: no enough physical eraseblocks (-20, need 1) UBI error: ubi_attach_mtd_dev: failed to attach mtd1, error -12 <= NOT ENOMEM UBI error: ubi_init: cannot attach mtd1 If available PEBs are not enough when initializing volumes, return -ENOSPC directly. If available PEBs are not enough when initializing WL, return -ENOSPC instead of -ENOMEM. Cc: stable@vger.kernel.org Signed-off-by: Sheng Yong Signed-off-by: Richard Weinberger Reviewed-by: David Gstir --- drivers/mtd/ubi/vtbl.c | 1 + drivers/mtd/ubi/wl.c | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 80bdd5b88bac..d85c19762160 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -649,6 +649,7 @@ static int init_volumes(struct ubi_device *ubi, if (ubi->corr_peb_count) ubi_err(ubi, "%d PEBs are corrupted and not used", ubi->corr_peb_count); + return -ENOSPC; } ubi->rsvd_pebs += reserved_pebs; ubi->avail_pebs -= reserved_pebs; diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index 275d9fb6fe5c..eb4489f9082f 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -1601,6 +1601,7 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai) if (ubi->corr_peb_count) ubi_err(ubi, "%d PEBs are corrupted and not used", ubi->corr_peb_count); + err = -ENOSPC; goto out_free; } ubi->avail_pebs -= reserved_pebs; From ccc42592d436d021d17f86729d24806f30dbad25 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Sun, 20 Sep 2015 07:00:10 -0700 Subject: [PATCH 525/647] clocksource/drivers/rockchip: Fix bad NO_IRQ usage The current code assumes the 'irq_of_parse_and_map' will return NO_IRQ in case of failure. Unfortunately, the NO_IRQ is not consistent across the different architectures and we must not rely on it. NO_IRQ is equal to '-1' on ARM and 'irq_of_parse_and_map' returns '0' in case of an error. Hence, the latter won't be detected and will lead to a crash. Fix this by just checking 'irq' is different from zero. Signed-off-by: Daniel Lezcano --- drivers/clocksource/rockchip_timer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/clocksource/rockchip_timer.c b/drivers/clocksource/rockchip_timer.c index bb2c2b050964..d3c1742ded1a 100644 --- a/drivers/clocksource/rockchip_timer.c +++ b/drivers/clocksource/rockchip_timer.c @@ -148,7 +148,7 @@ static void __init rk_timer_init(struct device_node *np) bc_timer.freq = clk_get_rate(timer_clk); irq = irq_of_parse_and_map(np, 0); - if (irq == NO_IRQ) { + if (!irq) { pr_err("Failed to map interrupts for '%s'\n", TIMER_NAME); return; } From bdf7344e14d826d0df438a55fc51146d179e198d Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Sun, 20 Sep 2015 07:00:35 -0700 Subject: [PATCH 526/647] clocksource/drivers/keystone: Fix bad NO_IRQ usage The current code assumes the 'irq_of_parse_and_map' will return NO_IRQ in case of failure. Unfortunately, the NO_IRQ is not consistent across the different architectures and we must not rely on it. NO_IRQ is equal to '-1' on ARM and 'irq_of_parse_and_map' returns '0' in case of an error. Hence, the latter won't be detected and will lead to a crash. Fix this by just checking 'irq' is different from zero. Signed-off-by: Daniel Lezcano --- drivers/clocksource/timer-keystone.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/clocksource/timer-keystone.c b/drivers/clocksource/timer-keystone.c index edacf3902e10..1cea08cf603e 100644 --- a/drivers/clocksource/timer-keystone.c +++ b/drivers/clocksource/timer-keystone.c @@ -152,7 +152,7 @@ static void __init keystone_timer_init(struct device_node *np) int irq, error; irq = irq_of_parse_and_map(np, 0); - if (irq == NO_IRQ) { + if (!irq) { pr_err("%s: failed to map interrupts\n", __func__); return; } From 1356aae08338f1c19ce1c67bf8c543a267688fc3 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 27 Sep 2015 02:09:19 +0900 Subject: [PATCH 527/647] blk-mq: avoid setting hctx->tags->cpumask before allocation When unmapped hw queue is remapped after CPU topology is changed, hctx->tags->cpumask has to be set after hctx->tags is setup in blk_mq_map_swqueue(), otherwise it causes null pointer dereference. Fixes: f26cdc8536 ("blk-mq: Shared tag enhancements") Signed-off-by: Akinobu Mita Cc: Keith Busch Cc: Ming Lei Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index f2d67b4047a0..2fd7283ec62b 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -1811,7 +1811,6 @@ static void blk_mq_map_swqueue(struct request_queue *q) hctx = q->mq_ops->map_queue(q, i); cpumask_set_cpu(i, hctx->cpumask); - cpumask_set_cpu(i, hctx->tags->cpumask); ctx->index_hw = hctx->nr_ctx; hctx->ctxs[hctx->nr_ctx++] = ctx; } @@ -1851,6 +1850,14 @@ static void blk_mq_map_swqueue(struct request_queue *q) hctx->next_cpu = cpumask_first(hctx->cpumask); hctx->next_cpu_batch = BLK_MQ_CPU_WORK_BATCH; } + + queue_for_each_ctx(q, ctx, i) { + if (!cpu_online(i)) + continue; + + hctx = q->mq_ops->map_queue(q, i); + cpumask_set_cpu(i, hctx->tags->cpumask); + } } static void blk_mq_update_tag_set_depth(struct blk_mq_tag_set *set) From 4593fdbe7a2f44d5e64c627c715dd0bcec9bdf14 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 27 Sep 2015 02:09:20 +0900 Subject: [PATCH 528/647] blk-mq: fix sysfs registration/unregistration race There is a race between cpu hotplug handling and adding/deleting gendisk for blk-mq, where both are trying to register and unregister the same sysfs entries. null_add_dev --> blk_mq_init_queue --> blk_mq_init_allocated_queue --> add to 'all_q_list' (*) --> add_disk --> blk_register_queue --> blk_mq_register_disk (++) null_del_dev --> del_gendisk --> blk_unregister_queue --> blk_mq_unregister_disk (--) --> blk_cleanup_queue --> blk_mq_free_queue --> del from 'all_q_list' (*) blk_mq_queue_reinit --> blk_mq_sysfs_unregister (-) --> blk_mq_sysfs_register (+) While the request queue is added to 'all_q_list' (*), blk_mq_queue_reinit() can be called for the queue anytime by CPU hotplug callback. But blk_mq_sysfs_unregister (-) and blk_mq_sysfs_register (+) in blk_mq_queue_reinit must not be called before blk_mq_register_disk (++) and after blk_mq_unregister_disk (--) is finished. Because '/sys/block/*/mq/' is not exists. There has already been BLK_MQ_F_SYSFS_UP flag in hctx->flags which can be used to track these sysfs stuff, but it is only fixing this issue partially. In order to fix it completely, we just need per-queue flag instead of per-hctx flag with appropriate locking. So this introduces q->mq_sysfs_init_done which is properly protected with all_q_mutex. Also, we need to ensure that blk_mq_map_swqueue() is called with all_q_mutex is held. Since hctx->nr_ctx is reset temporarily and updated in blk_mq_map_swqueue(), so we should avoid blk_mq_register_hctx() seeing the temporary hctx->nr_ctx value in CPU hotplug handling or adding/deleting gendisk . Signed-off-by: Akinobu Mita Reviewed-by: Ming Lei Cc: Ming Lei Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq-sysfs.c | 30 ++++++++++++++++++++++-------- block/blk-mq.c | 6 +++--- include/linux/blk-mq.h | 1 - include/linux/blkdev.h | 2 ++ 4 files changed, 27 insertions(+), 12 deletions(-) diff --git a/block/blk-mq-sysfs.c b/block/blk-mq-sysfs.c index 279c5d674edf..189f5ae6522a 100644 --- a/block/blk-mq-sysfs.c +++ b/block/blk-mq-sysfs.c @@ -343,7 +343,7 @@ static void blk_mq_unregister_hctx(struct blk_mq_hw_ctx *hctx) struct blk_mq_ctx *ctx; int i; - if (!hctx->nr_ctx || !(hctx->flags & BLK_MQ_F_SYSFS_UP)) + if (!hctx->nr_ctx) return; hctx_for_each_ctx(hctx, ctx, i) @@ -358,7 +358,7 @@ static int blk_mq_register_hctx(struct blk_mq_hw_ctx *hctx) struct blk_mq_ctx *ctx; int i, ret; - if (!hctx->nr_ctx || !(hctx->flags & BLK_MQ_F_SYSFS_UP)) + if (!hctx->nr_ctx) return 0; ret = kobject_add(&hctx->kobj, &q->mq_kobj, "%u", hctx->queue_num); @@ -381,6 +381,8 @@ void blk_mq_unregister_disk(struct gendisk *disk) struct blk_mq_ctx *ctx; int i, j; + blk_mq_disable_hotplug(); + queue_for_each_hw_ctx(q, hctx, i) { blk_mq_unregister_hctx(hctx); @@ -395,6 +397,9 @@ void blk_mq_unregister_disk(struct gendisk *disk) kobject_put(&q->mq_kobj); kobject_put(&disk_to_dev(disk)->kobj); + + q->mq_sysfs_init_done = false; + blk_mq_enable_hotplug(); } static void blk_mq_sysfs_init(struct request_queue *q) @@ -425,27 +430,30 @@ int blk_mq_register_disk(struct gendisk *disk) struct blk_mq_hw_ctx *hctx; int ret, i; + blk_mq_disable_hotplug(); + blk_mq_sysfs_init(q); ret = kobject_add(&q->mq_kobj, kobject_get(&dev->kobj), "%s", "mq"); if (ret < 0) - return ret; + goto out; kobject_uevent(&q->mq_kobj, KOBJ_ADD); queue_for_each_hw_ctx(q, hctx, i) { - hctx->flags |= BLK_MQ_F_SYSFS_UP; ret = blk_mq_register_hctx(hctx); if (ret) break; } - if (ret) { + if (ret) blk_mq_unregister_disk(disk); - return ret; - } + else + q->mq_sysfs_init_done = true; +out: + blk_mq_enable_hotplug(); - return 0; + return ret; } EXPORT_SYMBOL_GPL(blk_mq_register_disk); @@ -454,6 +462,9 @@ void blk_mq_sysfs_unregister(struct request_queue *q) struct blk_mq_hw_ctx *hctx; int i; + if (!q->mq_sysfs_init_done) + return; + queue_for_each_hw_ctx(q, hctx, i) blk_mq_unregister_hctx(hctx); } @@ -463,6 +474,9 @@ int blk_mq_sysfs_register(struct request_queue *q) struct blk_mq_hw_ctx *hctx; int i, ret = 0; + if (!q->mq_sysfs_init_done) + return ret; + queue_for_each_hw_ctx(q, hctx, i) { ret = blk_mq_register_hctx(hctx); if (ret) diff --git a/block/blk-mq.c b/block/blk-mq.c index 2fd7283ec62b..0262131ac5f2 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -2035,13 +2035,13 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, goto err_hctxs; mutex_lock(&all_q_mutex); + list_add_tail(&q->all_q_node, &all_q_list); - mutex_unlock(&all_q_mutex); - blk_mq_add_queue_tag_set(set, q); - blk_mq_map_swqueue(q); + mutex_unlock(&all_q_mutex); + return q; err_hctxs: diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 37d1602c4f7a..b80ba4572a31 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -145,7 +145,6 @@ enum { BLK_MQ_F_SHOULD_MERGE = 1 << 0, BLK_MQ_F_TAG_SHARED = 1 << 1, BLK_MQ_F_SG_MERGE = 1 << 2, - BLK_MQ_F_SYSFS_UP = 1 << 3, BLK_MQ_F_DEFER_ISSUE = 1 << 4, BLK_MQ_F_ALLOC_POLICY_START_BIT = 8, BLK_MQ_F_ALLOC_POLICY_BITS = 1, diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 99da9ebc7377..19c2e947d4d1 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -456,6 +456,8 @@ struct request_queue { struct blk_mq_tag_set *tag_set; struct list_head tag_set_list; struct bio_set *bio_split; + + bool mq_sysfs_init_done; }; #define QUEUE_FLAG_QUEUED 1 /* uses generic tag queueing */ From a723bab3d7529133f71fc8a5e96f86e3639a0d13 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 27 Sep 2015 02:09:21 +0900 Subject: [PATCH 529/647] blk-mq: Fix use after of free q->mq_map CPU hotplug handling for blk-mq (blk_mq_queue_reinit) updates q->mq_map by blk_mq_update_queue_map() for all request queues in all_q_list. On the other hand, q->mq_map is released before deleting the queue from all_q_list. So if CPU hotplug event occurs in the window, invalid memory access can happen. Fix it by releasing q->mq_map in blk_mq_release() to make it happen latter than removal from all_q_list. Signed-off-by: Akinobu Mita Suggested-by: Ming Lei Reviewed-by: Ming Lei Cc: Ming Lei Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index 0262131ac5f2..92648d8d6a4a 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -1925,6 +1925,9 @@ void blk_mq_release(struct request_queue *q) kfree(hctx); } + kfree(q->mq_map); + q->mq_map = NULL; + kfree(q->queue_hw_ctx); /* ctx kobj stays in queue_ctx */ @@ -2070,11 +2073,6 @@ void blk_mq_free_queue(struct request_queue *q) blk_mq_free_hw_queues(q, set); percpu_ref_exit(&q->mq_usage_counter); - - kfree(q->mq_map); - - q->mq_map = NULL; - mutex_lock(&all_q_mutex); list_del_init(&q->all_q_node); mutex_unlock(&all_q_mutex); From 0e6263682014d480b8d7b8c10287f4536066b54f Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 27 Sep 2015 02:09:22 +0900 Subject: [PATCH 530/647] blk-mq: fix q->mq_usage_counter access race CPU hotplug handling for blk-mq (blk_mq_queue_reinit) accesses q->mq_usage_counter while freezing all request queues in all_q_list. On the other hand, q->mq_usage_counter is deinitialized in blk_mq_free_queue() before deleting the queue from all_q_list. So if CPU hotplug event occurs in the window, percpu_ref_kill() is called with q->mq_usage_counter which has already been marked dead, and it triggers warning. Fix it by deleting the queue from all_q_list earlier than destroying q->mq_usage_counter. Signed-off-by: Akinobu Mita Reviewed-by: Ming Lei Cc: Ming Lei Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index 92648d8d6a4a..3a39184e82e5 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -2067,15 +2067,16 @@ void blk_mq_free_queue(struct request_queue *q) { struct blk_mq_tag_set *set = q->tag_set; + mutex_lock(&all_q_mutex); + list_del_init(&q->all_q_node); + mutex_unlock(&all_q_mutex); + blk_mq_del_queue_tag_set(q); blk_mq_exit_hw_queues(q, set, set->nr_hw_queues); blk_mq_free_hw_queues(q, set); percpu_ref_exit(&q->mq_usage_counter); - mutex_lock(&all_q_mutex); - list_del_init(&q->all_q_node); - mutex_unlock(&all_q_mutex); } /* Basically redo blk_mq_init_queue with queue frozen */ From 5778322e67ed34dc9f391a4a5cbcbb856071ceba Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 27 Sep 2015 02:09:23 +0900 Subject: [PATCH 531/647] blk-mq: avoid inserting requests before establishing new mapping Notifier callbacks for CPU_ONLINE action can be run on the other CPU than the CPU which was just onlined. So it is possible for the process running on the just onlined CPU to insert request and run hw queue before establishing new mapping which is done by blk_mq_queue_reinit_notify(). This can cause a problem when the CPU has just been onlined first time since the request queue was initialized. At this time ctx->index_hw for the CPU, which is the index in hctx->ctxs[] for this ctx, is still zero before blk_mq_queue_reinit_notify() is called by notifier callbacks for CPU_ONLINE action. For example, there is a single hw queue (hctx) and two CPU queues (ctx0 for CPU0, and ctx1 for CPU1). Now CPU1 is just onlined and a request is inserted into ctx1->rq_list and set bit0 in pending bitmap as ctx1->index_hw is still zero. And then while running hw queue, flush_busy_ctxs() finds bit0 is set in pending bitmap and tries to retrieve requests in hctx->ctxs[0]->rq_list. But htx->ctxs[0] is a pointer to ctx0, so the request in ctx1->rq_list is ignored. Fix it by ensuring that new mapping is established before onlined cpu starts running. Signed-off-by: Akinobu Mita Reviewed-by: Ming Lei Cc: Jens Axboe Cc: Ming Lei Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq-cpumap.c | 9 ++++--- block/blk-mq.c | 59 +++++++++++++++++++++++++++++++++---------- block/blk-mq.h | 3 ++- 3 files changed, 52 insertions(+), 19 deletions(-) diff --git a/block/blk-mq-cpumap.c b/block/blk-mq-cpumap.c index 1e28ddb656b8..8764c241e5bb 100644 --- a/block/blk-mq-cpumap.c +++ b/block/blk-mq-cpumap.c @@ -31,7 +31,8 @@ static int get_first_sibling(unsigned int cpu) return cpu; } -int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues) +int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues, + const struct cpumask *online_mask) { unsigned int i, nr_cpus, nr_uniq_cpus, queue, first_sibling; cpumask_var_t cpus; @@ -41,7 +42,7 @@ int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues) cpumask_clear(cpus); nr_cpus = nr_uniq_cpus = 0; - for_each_online_cpu(i) { + for_each_cpu(i, online_mask) { nr_cpus++; first_sibling = get_first_sibling(i); if (!cpumask_test_cpu(first_sibling, cpus)) @@ -51,7 +52,7 @@ int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues) queue = 0; for_each_possible_cpu(i) { - if (!cpu_online(i)) { + if (!cpumask_test_cpu(i, online_mask)) { map[i] = 0; continue; } @@ -95,7 +96,7 @@ unsigned int *blk_mq_make_queue_map(struct blk_mq_tag_set *set) if (!map) return NULL; - if (!blk_mq_update_queue_map(map, set->nr_hw_queues)) + if (!blk_mq_update_queue_map(map, set->nr_hw_queues, cpu_online_mask)) return map; kfree(map); diff --git a/block/blk-mq.c b/block/blk-mq.c index 3a39184e82e5..a5dbd069c9da 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -1789,7 +1789,8 @@ static void blk_mq_init_cpu_queues(struct request_queue *q, } } -static void blk_mq_map_swqueue(struct request_queue *q) +static void blk_mq_map_swqueue(struct request_queue *q, + const struct cpumask *online_mask) { unsigned int i; struct blk_mq_hw_ctx *hctx; @@ -1806,7 +1807,7 @@ static void blk_mq_map_swqueue(struct request_queue *q) */ queue_for_each_ctx(q, ctx, i) { /* If the cpu isn't online, the cpu is mapped to first hctx */ - if (!cpu_online(i)) + if (!cpumask_test_cpu(i, online_mask)) continue; hctx = q->mq_ops->map_queue(q, i); @@ -1852,7 +1853,7 @@ static void blk_mq_map_swqueue(struct request_queue *q) } queue_for_each_ctx(q, ctx, i) { - if (!cpu_online(i)) + if (!cpumask_test_cpu(i, online_mask)) continue; hctx = q->mq_ops->map_queue(q, i); @@ -2037,13 +2038,15 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, if (blk_mq_init_hw_queues(q, set)) goto err_hctxs; + get_online_cpus(); mutex_lock(&all_q_mutex); list_add_tail(&q->all_q_node, &all_q_list); blk_mq_add_queue_tag_set(set, q); - blk_mq_map_swqueue(q); + blk_mq_map_swqueue(q, cpu_online_mask); mutex_unlock(&all_q_mutex); + put_online_cpus(); return q; @@ -2080,13 +2083,14 @@ void blk_mq_free_queue(struct request_queue *q) } /* Basically redo blk_mq_init_queue with queue frozen */ -static void blk_mq_queue_reinit(struct request_queue *q) +static void blk_mq_queue_reinit(struct request_queue *q, + const struct cpumask *online_mask) { WARN_ON_ONCE(!atomic_read(&q->mq_freeze_depth)); blk_mq_sysfs_unregister(q); - blk_mq_update_queue_map(q->mq_map, q->nr_hw_queues); + blk_mq_update_queue_map(q->mq_map, q->nr_hw_queues, online_mask); /* * redo blk_mq_init_cpu_queues and blk_mq_init_hw_queues. FIXME: maybe @@ -2094,7 +2098,7 @@ static void blk_mq_queue_reinit(struct request_queue *q) * involves free and re-allocate memory, worthy doing?) */ - blk_mq_map_swqueue(q); + blk_mq_map_swqueue(q, online_mask); blk_mq_sysfs_register(q); } @@ -2103,16 +2107,43 @@ static int blk_mq_queue_reinit_notify(struct notifier_block *nb, unsigned long action, void *hcpu) { struct request_queue *q; + int cpu = (unsigned long)hcpu; + /* + * New online cpumask which is going to be set in this hotplug event. + * Declare this cpumasks as global as cpu-hotplug operation is invoked + * one-by-one and dynamically allocating this could result in a failure. + */ + static struct cpumask online_new; /* - * Before new mappings are established, hotadded cpu might already - * start handling requests. This doesn't break anything as we map - * offline CPUs to first hardware queue. We will re-init the queue - * below to get optimal settings. + * Before hotadded cpu starts handling requests, new mappings must + * be established. Otherwise, these requests in hw queue might + * never be dispatched. + * + * For example, there is a single hw queue (hctx) and two CPU queues + * (ctx0 for CPU0, and ctx1 for CPU1). + * + * Now CPU1 is just onlined and a request is inserted into + * ctx1->rq_list and set bit0 in pending bitmap as ctx1->index_hw is + * still zero. + * + * And then while running hw queue, flush_busy_ctxs() finds bit0 is + * set in pending bitmap and tries to retrieve requests in + * hctx->ctxs[0]->rq_list. But htx->ctxs[0] is a pointer to ctx0, + * so the request in ctx1->rq_list is ignored. */ - if (action != CPU_DEAD && action != CPU_DEAD_FROZEN && - action != CPU_ONLINE && action != CPU_ONLINE_FROZEN) + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_DEAD: + case CPU_UP_CANCELED: + cpumask_copy(&online_new, cpu_online_mask); + break; + case CPU_UP_PREPARE: + cpumask_copy(&online_new, cpu_online_mask); + cpumask_set_cpu(cpu, &online_new); + break; + default: return NOTIFY_OK; + } mutex_lock(&all_q_mutex); @@ -2136,7 +2167,7 @@ static int blk_mq_queue_reinit_notify(struct notifier_block *nb, } list_for_each_entry(q, &all_q_list, all_q_node) - blk_mq_queue_reinit(q); + blk_mq_queue_reinit(q, &online_new); list_for_each_entry(q, &all_q_list, all_q_node) blk_mq_unfreeze_queue(q); diff --git a/block/blk-mq.h b/block/blk-mq.h index 6a48c4c0d8a2..f4fea7964910 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -51,7 +51,8 @@ void blk_mq_disable_hotplug(void); * CPU -> queue mappings */ extern unsigned int *blk_mq_make_queue_map(struct blk_mq_tag_set *set); -extern int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues); +extern int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues, + const struct cpumask *online_mask); extern int blk_mq_hw_queue_to_node(unsigned int *map, unsigned int); /* From 60de074ba1e8f327db19bc33d8530131ac01695d Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 27 Sep 2015 02:09:25 +0900 Subject: [PATCH 532/647] blk-mq: fix deadlock when reading cpu_list CPU hotplug handling for blk-mq (blk_mq_queue_reinit) acquires all_q_mutex in blk_mq_queue_reinit_notify() and then removes sysfs entries by blk_mq_sysfs_unregister(). Removing sysfs entry needs to be blocked until the active reference of the kernfs_node to be zero. On the other hand, reading blk_mq_hw_sysfs_cpu sysfs entry (e.g. /sys/block/nullb0/mq/0/cpu_list) acquires all_q_mutex in blk_mq_hw_sysfs_cpus_show(). If these happen at the same time, a deadlock can happen. Because one can wait for the active reference to be zero with holding all_q_mutex, and the other tries to acquire all_q_mutex with holding the active reference. The reason that all_q_mutex is acquired in blk_mq_hw_sysfs_cpus_show() is to avoid reading an imcomplete hctx->cpumask. Since reading sysfs entry for blk-mq needs to acquire q->sysfs_lock, we can avoid deadlock and reading an imcomplete hctx->cpumask by protecting q->sysfs_lock while hctx->cpumask is being updated. Signed-off-by: Akinobu Mita Reviewed-by: Ming Lei Cc: Ming Lei Cc: Wanpeng Li Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq-sysfs.c | 4 ---- block/blk-mq.c | 7 +++++++ 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/block/blk-mq-sysfs.c b/block/blk-mq-sysfs.c index 189f5ae6522a..788fffd9b409 100644 --- a/block/blk-mq-sysfs.c +++ b/block/blk-mq-sysfs.c @@ -229,8 +229,6 @@ static ssize_t blk_mq_hw_sysfs_cpus_show(struct blk_mq_hw_ctx *hctx, char *page) unsigned int i, first = 1; ssize_t ret = 0; - blk_mq_disable_hotplug(); - for_each_cpu(i, hctx->cpumask) { if (first) ret += sprintf(ret + page, "%u", i); @@ -240,8 +238,6 @@ static ssize_t blk_mq_hw_sysfs_cpus_show(struct blk_mq_hw_ctx *hctx, char *page) first = 0; } - blk_mq_enable_hotplug(); - ret += sprintf(ret + page, "\n"); return ret; } diff --git a/block/blk-mq.c b/block/blk-mq.c index a5dbd069c9da..31c0c6259c4c 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -1797,6 +1797,11 @@ static void blk_mq_map_swqueue(struct request_queue *q, struct blk_mq_ctx *ctx; struct blk_mq_tag_set *set = q->tag_set; + /* + * Avoid others reading imcomplete hctx->cpumask through sysfs + */ + mutex_lock(&q->sysfs_lock); + queue_for_each_hw_ctx(q, hctx, i) { cpumask_clear(hctx->cpumask); hctx->nr_ctx = 0; @@ -1816,6 +1821,8 @@ static void blk_mq_map_swqueue(struct request_queue *q, hctx->ctxs[hctx->nr_ctx++] = ctx; } + mutex_unlock(&q->sysfs_lock); + queue_for_each_hw_ctx(q, hctx, i) { struct blk_mq_ctxmap *map = &hctx->ctx_map; From 54313503f9a3c34a9e45aad7654976b7f50b9ae7 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 25 Sep 2015 14:36:55 +0300 Subject: [PATCH 533/647] drm/amdgpu: signedness bug in amdgpu_cs_parser_init() The "i" variable should be signed or it leads to a crash in the error handling code. Fixes: 1d263474c441 ('drm/amdgpu: unwind properly in amdgpu_cs_parser_init()') Signed-off-by: Dan Carpenter Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 749420f1ea6f..cb3c274edb0a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -156,7 +156,8 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) uint64_t *chunk_array_user; uint64_t *chunk_array; struct amdgpu_fpriv *fpriv = p->filp->driver_priv; - unsigned size, i; + unsigned size; + int i; int ret; if (cs->in.num_chunks == 0) From 4281f46ef839050d2ef60348f661eb463c21cc2e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Mon, 28 Sep 2015 18:16:31 +0900 Subject: [PATCH 534/647] drm/radeon: Restore LCD backlight level on resume (>= R5xx) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Instead of only enabling the backlight (which seems to set it to max brightness), just re-set the current backlight level, which also takes care of enabling the backlight if necessary. Only the radeon_atom_encoder_dpms_dig part tested on a Kaveri laptop, the radeon_atom_encoder_dpms_avivo part is only compile tested. Cc: stable@vger.kernel.org Signed-off-by: Michel Dänzer Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/atombios_encoders.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index c3872598b85a..65adb9c72377 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -1624,8 +1624,9 @@ radeon_atom_encoder_dpms_avivo(struct drm_encoder *encoder, int mode) } else atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { - args.ucAction = ATOM_LCD_BLON; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; + + atombios_set_backlight_level(radeon_encoder, dig->backlight_level); } break; case DRM_MODE_DPMS_STANDBY: @@ -1706,8 +1707,7 @@ radeon_atom_encoder_dpms_dig(struct drm_encoder *encoder, int mode) atombios_dig_encoder_setup(encoder, ATOM_ENCODER_CMD_DP_VIDEO_ON, 0); } if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) - atombios_dig_transmitter_setup(encoder, - ATOM_TRANSMITTER_ACTION_LCD_BLON, 0, 0); + atombios_set_backlight_level(radeon_encoder, dig->backlight_level); if (ext_encoder) atombios_external_encoder_setup(encoder, ext_encoder, ATOM_ENABLE); break; From 74b3112e95073b351e3b0b9799795bc76f8415fa Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 29 Sep 2015 13:53:30 -0400 Subject: [PATCH 535/647] drm/amdgpu: Restore LCD backlight level on resume Instead of only enabling the backlight (which seems to set it to max brightness), just re-set the current backlight level, which also takes care of enabling the backlight if necessary. Port of radeon commit: drm/radeon: Restore LCD backlight level on resume (>= R5xx) Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/atombios_encoders.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c index cd6edc40c9cd..1e0bba29e167 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c @@ -1279,8 +1279,7 @@ amdgpu_atombios_encoder_setup_dig(struct drm_encoder *encoder, int action) amdgpu_atombios_encoder_setup_dig_encoder(encoder, ATOM_ENCODER_CMD_DP_VIDEO_ON, 0); } if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) - amdgpu_atombios_encoder_setup_dig_transmitter(encoder, - ATOM_TRANSMITTER_ACTION_LCD_BLON, 0, 0); + amdgpu_atombios_encoder_set_backlight_level(amdgpu_encoder, dig->backlight_level); if (ext_encoder) amdgpu_atombios_encoder_setup_external_encoder(encoder, ext_encoder, ATOM_ENABLE); } else { From 4613012db1d911f80897f9446a49de817b2c4c47 Mon Sep 17 00:00:00 2001 From: Aaron Conole Date: Sat, 26 Sep 2015 18:50:42 -0400 Subject: [PATCH 536/647] af_unix: Convert the unix_sk macro to an inline function for type safety As suggested by Eric Dumazet this change replaces the #define with a static inline function to enjoy complaints by the compiler when misusing the API. Signed-off-by: Aaron Conole Signed-off-by: David S. Miller --- include/net/af_unix.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 4a167b30a12f..cb1b9bbda332 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -63,7 +63,11 @@ struct unix_sock { #define UNIX_GC_MAYBE_CYCLE 1 struct socket_wq peer_wq; }; -#define unix_sk(__sk) ((struct unix_sock *)__sk) + +static inline struct unix_sock *unix_sk(struct sock *sk) +{ + return (struct unix_sock *)sk; +} #define peer_wait peer_wq.wait From 9f389e35674f5b086edd70ed524ca0f287259725 Mon Sep 17 00:00:00 2001 From: Aaron Conole Date: Sat, 26 Sep 2015 18:50:43 -0400 Subject: [PATCH 537/647] af_unix: return data from multiple SKBs on recv() with MSG_PEEK flag AF_UNIX sockets now return multiple skbs from recv() when MSG_PEEK flag is set. This is referenced in kernel bugzilla #12323 @ https://bugzilla.kernel.org/show_bug.cgi?id=12323 As described both in the BZ and lkml thread @ http://lkml.org/lkml/2008/1/8/444 calling recv() with MSG_PEEK on an AF_UNIX socket only reads a single skb, where the desired effect is to return as much skb data has been queued, until hitting the recv buffer size (whichever comes first). The modified MSG_PEEK path will now move to the next skb in the tree and jump to the again: label, rather than following the natural loop structure. This requires duplicating some of the loop head actions. This was tested using the python socketpair python code attached to the bugzilla issue. Signed-off-by: Aaron Conole Signed-off-by: David S. Miller --- net/unix/af_unix.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 03ee4d359f6a..ef31b40ad550 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2179,8 +2179,21 @@ unlock: if (UNIXCB(skb).fp) scm.fp = scm_fp_dup(UNIXCB(skb).fp); - sk_peek_offset_fwd(sk, chunk); + if (skip) { + sk_peek_offset_fwd(sk, chunk); + skip -= chunk; + } + if (UNIXCB(skb).fp) + break; + + last = skb; + last_len = skb->len; + unix_state_lock(sk); + skb = skb_peek_next(skb, &sk->sk_receive_queue); + if (skb) + goto again; + unix_state_unlock(sk); break; } } while (size); From 4c52b1da538800a30f030eeb697366c23daf2ef3 Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Mon, 28 Sep 2015 10:49:48 +0200 Subject: [PATCH 538/647] bna: fix error handling Several functions can return negative value in case of error, so their return type should be fixed as well as type of variables to which this value is assigned. The problem has been detected using proposed semantic patch scripts/coccinelle/tests/assign_signed_to_unsigned.cocci [1]. [1]: http://permalink.gmane.org/gmane.linux.kernel/2046107 Signed-off-by: Andrzej Hajda Signed-off-by: David S. Miller --- drivers/net/ethernet/brocade/bna/bfa_ioc.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/brocade/bna/bfa_ioc.c b/drivers/net/ethernet/brocade/bna/bfa_ioc.c index b7a0f7879de2..9e59663a6ead 100644 --- a/drivers/net/ethernet/brocade/bna/bfa_ioc.c +++ b/drivers/net/ethernet/brocade/bna/bfa_ioc.c @@ -1543,7 +1543,7 @@ bfa_flash_cmd_act_check(void __iomem *pci_bar) } /* Flush FLI data fifo. */ -static u32 +static int bfa_flash_fifo_flush(void __iomem *pci_bar) { u32 i; @@ -1573,11 +1573,11 @@ bfa_flash_fifo_flush(void __iomem *pci_bar) } /* Read flash status. */ -static u32 +static int bfa_flash_status_read(void __iomem *pci_bar) { union bfa_flash_dev_status_reg dev_status; - u32 status; + int status; u32 ret_status; int i; @@ -1611,11 +1611,11 @@ bfa_flash_status_read(void __iomem *pci_bar) } /* Start flash read operation. */ -static u32 +static int bfa_flash_read_start(void __iomem *pci_bar, u32 offset, u32 len, char *buf) { - u32 status; + int status; /* len must be mutiple of 4 and not exceeding fifo size */ if (len == 0 || len > BFA_FLASH_FIFO_SIZE || (len & 0x03) != 0) @@ -1703,7 +1703,8 @@ static enum bfa_status bfa_flash_raw_read(void __iomem *pci_bar, u32 offset, char *buf, u32 len) { - u32 n, status; + u32 n; + int status; u32 off, l, s, residue, fifo_sz; residue = len; From 75c261b51ba19f0791de608f0acfb94956f78c76 Mon Sep 17 00:00:00 2001 From: Alexander Stein Date: Mon, 28 Sep 2015 15:05:33 +0200 Subject: [PATCH 539/647] net sysfs: Print link speed as signed integer Otherwise 4294967295 (MBit/s) (-1) will be printed when there is no link. Documentation/ABI/testing/sysfs-class-net does not state if this shall be signed or unsigned. Also remove the now unused variable fmt_udec. Signed-off-by: Alexander Stein Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 805a95a48107..830f8a7c1cb1 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -31,7 +31,6 @@ static const char fmt_hex[] = "%#x\n"; static const char fmt_long_hex[] = "%#lx\n"; static const char fmt_dec[] = "%d\n"; -static const char fmt_udec[] = "%u\n"; static const char fmt_ulong[] = "%lu\n"; static const char fmt_u64[] = "%llu\n"; @@ -202,7 +201,7 @@ static ssize_t speed_show(struct device *dev, if (netif_running(netdev)) { struct ethtool_cmd cmd; if (!__ethtool_get_settings(netdev, &cmd)) - ret = sprintf(buf, fmt_udec, ethtool_cmd_speed(&cmd)); + ret = sprintf(buf, fmt_dec, ethtool_cmd_speed(&cmd)); } rtnl_unlock(); return ret; From 741a11d9e4103a8e1c590ef1280143fe654e4e33 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Mon, 28 Sep 2015 10:12:13 -0700 Subject: [PATCH 540/647] net: ipv6: Add RT6_LOOKUP_F_IFACE flag if oif is set Wolfgang reported that IPv6 stack is ignoring oif in output route lookups: With ipv6, ip -6 route get always returns the specific route. $ ip -6 r 2001:db8:e2::1 dev enp2s0 proto kernel metric 256 2001:db8:e2::/64 dev enp2s0 metric 1024 2001:db8:e3::1 dev enp3s0 proto kernel metric 256 2001:db8:e3::/64 dev enp3s0 metric 1024 fe80::/64 dev enp3s0 proto kernel metric 256 default via 2001:db8:e3::255 dev enp3s0 metric 1024 $ ip -6 r get 2001:db8:e2::100 2001:db8:e2::100 from :: dev enp2s0 src 2001:db8:e3::1 metric 0 cache $ ip -6 r get 2001:db8:e2::100 oif enp3s0 2001:db8:e2::100 from :: dev enp2s0 src 2001:db8:e3::1 metric 0 cache The stack does consider the oif but a mismatch in rt6_device_match is not considered fatal because RT6_LOOKUP_F_IFACE is not set in the flags. Cc: Wolfgang Nothdurft Signed-off-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/route.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index f204089e854c..cb32ce250db0 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1193,7 +1193,8 @@ struct dst_entry *ip6_route_output(struct net *net, const struct sock *sk, fl6->flowi6_iif = LOOPBACK_IFINDEX; - if ((sk && sk->sk_bound_dev_if) || rt6_need_strict(&fl6->daddr)) + if ((sk && sk->sk_bound_dev_if) || rt6_need_strict(&fl6->daddr) || + fl6->flowi6_oif) flags |= RT6_LOOKUP_F_IFACE; if (!ipv6_addr_any(&fl6->saddr)) From 218c1f76b8b25d6dc9d01443f071cb618e206b0c Mon Sep 17 00:00:00 2001 From: Dmitry Vyukov Date: Tue, 29 Sep 2015 15:54:58 -0700 Subject: [PATCH 541/647] Input: psmouse - fix data race in __ps2_command The data race happens on ps2dev->cmdcnt and ps2dev->cmdbuf contents. __ps2_command reads that data concurrently with the interrupt handler. As the result, for example, if a response arrives just after the timeout, __ps2_command can copy out garbage from ps2dev->cmdbuf but then see that ps2dev->cmdcnt is 0 and return success. Stop the interrupt handler with serio_pause_rx() before reading the results. The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov Signed-off-by: Dmitry Torokhov --- drivers/input/serio/libps2.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c index 75516996db20..316f2c897101 100644 --- a/drivers/input/serio/libps2.c +++ b/drivers/input/serio/libps2.c @@ -212,12 +212,17 @@ int __ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command) * time before the ACK arrives. */ if (ps2_sendbyte(ps2dev, command & 0xff, - command == PS2_CMD_RESET_BAT ? 1000 : 200)) - goto out; + command == PS2_CMD_RESET_BAT ? 1000 : 200)) { + serio_pause_rx(ps2dev->serio); + goto out_reset_flags; + } - for (i = 0; i < send; i++) - if (ps2_sendbyte(ps2dev, param[i], 200)) - goto out; + for (i = 0; i < send; i++) { + if (ps2_sendbyte(ps2dev, param[i], 200)) { + serio_pause_rx(ps2dev->serio); + goto out_reset_flags; + } + } /* * The reset command takes a long time to execute. @@ -234,17 +239,18 @@ int __ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command) !(ps2dev->flags & PS2_FLAG_CMD), timeout); } + serio_pause_rx(ps2dev->serio); + if (param) for (i = 0; i < receive; i++) param[i] = ps2dev->cmdbuf[(receive - 1) - i]; if (ps2dev->cmdcnt && (command != PS2_CMD_RESET_BAT || ps2dev->cmdcnt != 1)) - goto out; + goto out_reset_flags; rc = 0; - out: - serio_pause_rx(ps2dev->serio); + out_reset_flags: ps2dev->flags = 0; serio_continue_rx(ps2dev->serio); From 31b33dfb0a144469dd805514c9e63f4993729a48 Mon Sep 17 00:00:00 2001 From: Pravin B Shelar Date: Mon, 28 Sep 2015 17:24:25 -0700 Subject: [PATCH 542/647] skbuff: Fix skb checksum partial check. Earlier patch 6ae459bda tried to detect void ckecksum partial skb by comparing pull length to checksum offset. But it does not work for all cases since checksum-offset depends on updates to skb->data. Following patch fixes it by validating checksum start offset after skb-data pointer is updated. Negative value of checksum offset start means there is no need to checksum. Fixes: 6ae459bda ("skbuff: Fix skb checksum flag on skb pull") Reported-by: Andrew Vagin Signed-off-by: Pravin B Shelar Signed-off-by: David S. Miller --- include/linux/skbuff.h | 2 +- net/core/skbuff.c | 9 +++++---- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 2b0a30a6e31c..4398411236f1 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -2708,7 +2708,7 @@ static inline void skb_postpull_rcsum(struct sk_buff *skb, if (skb->ip_summed == CHECKSUM_COMPLETE) skb->csum = csum_sub(skb->csum, csum_partial(start, len, 0)); else if (skb->ip_summed == CHECKSUM_PARTIAL && - skb_checksum_start_offset(skb) <= len) + skb_checksum_start_offset(skb) < 0) skb->ip_summed = CHECKSUM_NONE; } diff --git a/net/core/skbuff.c b/net/core/skbuff.c index dad4dd37e2aa..fab4599ba8b2 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2958,11 +2958,12 @@ EXPORT_SYMBOL_GPL(skb_append_pagefrags); */ unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len) { + unsigned char *data = skb->data; + BUG_ON(len > skb->len); - skb->len -= len; - BUG_ON(skb->len < skb->data_len); - skb_postpull_rcsum(skb, skb->data, len); - return skb->data += len; + __skb_pull(skb, len); + skb_postpull_rcsum(skb, data, len); + return skb->data; } EXPORT_SYMBOL_GPL(skb_pull_rcsum); From c047a1f918af75e572a19ba0581c3e3e202ed698 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Tue, 29 Sep 2015 01:50:56 +0200 Subject: [PATCH 543/647] dsa: mv88e6xxx: Enable forwarding for unknown to the CPU port Frames destined to an unknown address must be forwarded to the CPU port. Otherwise incoming ARP, dhcp leases, etc, do not work. Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/dsa/mv88e6xxx.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index f8baa897d1a0..1f7dd927cc5e 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -2051,6 +2051,8 @@ static int mv88e6xxx_setup_port(struct dsa_switch *ds, int port) reg |= PORT_CONTROL_FRAME_ETHER_TYPE_DSA; else reg |= PORT_CONTROL_FRAME_MODE_DSA; + reg |= PORT_CONTROL_FORWARD_UNKNOWN | + PORT_CONTROL_FORWARD_UNKNOWN_MC; } if (mv88e6xxx_6352_family(ds) || mv88e6xxx_6351_family(ds) || From 23860f103b53e764a3cbbf615d08f88362a28295 Mon Sep 17 00:00:00 2001 From: Robb Manes Date: Tue, 29 Sep 2015 11:03:37 -0400 Subject: [PATCH 544/647] net/mlx4: Handle return codes in mlx4_qp_attach_common Both new_steering_entry() and existing_steering_entry() return values based on their success or failure, but currently they fall through silently. This can make troubleshooting difficult, as we were unable to tell which one of these two functions returned errors or specifically what code was returned. This patch remedies that situation by passing the return codes to err, which is returned by mlx4_qp_attach_common() itself. This also addresses a leak in the call to mlx4_bitmap_free() as well. Signed-off-by: Robb Manes Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/mcg.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c index bd9ea0d01aae..1d4e2e054647 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mcg.c +++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c @@ -1184,10 +1184,11 @@ out: if (prot == MLX4_PROT_ETH) { /* manage the steering entry for promisc mode */ if (new_entry) - new_steering_entry(dev, port, steer, index, qp->qpn); + err = new_steering_entry(dev, port, steer, + index, qp->qpn); else - existing_steering_entry(dev, port, steer, - index, qp->qpn); + err = existing_steering_entry(dev, port, steer, + index, qp->qpn); } if (err && link && index != -1) { if (index < dev->caps.num_mgms) From 9ae6d4935e3df35a23bbbe531c6b9ff314e7fd0f Mon Sep 17 00:00:00 2001 From: Thomas Huth Date: Tue, 29 Sep 2015 17:45:28 +0200 Subject: [PATCH 545/647] testptp: Silence compiler warnings on ppc64 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When compiling Documentation/ptp/testptp.c the following compiler warnings are printed out: Documentation/ptp/testptp.c: In function ‘main’: Documentation/ptp/testptp.c:367:11: warning: format ‘%lld’ expects argument of type ‘long long int’, but argument 3 has type ‘__s64’ [-Wformat=] event.t.sec, event.t.nsec); ^ Documentation/ptp/testptp.c:505:5: warning: format ‘%lld’ expects argument of type ‘long long int’, but argument 2 has type ‘__s64’ [-Wformat=] (pct+2*i)->sec, (pct+2*i)->nsec); ^ Documentation/ptp/testptp.c:507:5: warning: format ‘%lld’ expects argument of type ‘long long int’, but argument 2 has type ‘__s64’ [-Wformat=] (pct+2*i+1)->sec, (pct+2*i+1)->nsec); ^ Documentation/ptp/testptp.c:509:5: warning: format ‘%lld’ expects argument of type ‘long long int’, but argument 2 has type ‘__s64’ [-Wformat=] (pct+2*i+2)->sec, (pct+2*i+2)->nsec); This happens because __s64 is by default defined as "long" on ppc64, not as "long long". However, to fix these warnings, it's possible to define the __SANE_USERSPACE_TYPES__ so that __s64 gets defined to "long long" on ppc64, too. Signed-off-by: Thomas Huth Acked-by: Richard Cochran Signed-off-by: David S. Miller --- Documentation/ptp/testptp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/ptp/testptp.c b/Documentation/ptp/testptp.c index 2bc8abc57fa0..6c6247aaa7b9 100644 --- a/Documentation/ptp/testptp.c +++ b/Documentation/ptp/testptp.c @@ -18,6 +18,7 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ #define _GNU_SOURCE +#define __SANE_USERSPACE_TYPES__ /* For PPC64, to get LL64 types */ #include #include #include From 57a47532c4312159935c98b7f1cf0e62296b9171 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Tue, 29 Sep 2015 14:17:54 -0400 Subject: [PATCH 546/647] net: dsa: fix preparation of a port STP update Because of the default 0 value of ret in dsa_slave_port_attr_set, a driver may return -EOPNOTSUPP from the commit phase of a STP state, which triggers a WARN() from switchdev. This happened on a 6185 switch which does not support hardware bridging. Fixes: 3563606258cf ("switchdev: convert STP update to switchdev attr set") Reported-by: Andrew Lunn Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- net/dsa/slave.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/net/dsa/slave.c b/net/dsa/slave.c index cce97385f743..7d91f4612ac0 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -458,12 +458,17 @@ static int dsa_slave_stp_update(struct net_device *dev, u8 state) static int dsa_slave_port_attr_set(struct net_device *dev, struct switchdev_attr *attr) { - int ret = 0; + struct dsa_slave_priv *p = netdev_priv(dev); + struct dsa_switch *ds = p->parent; + int ret; switch (attr->id) { case SWITCHDEV_ATTR_PORT_STP_STATE: - if (attr->trans == SWITCHDEV_TRANS_COMMIT) - ret = dsa_slave_stp_update(dev, attr->u.stp_state); + if (attr->trans == SWITCHDEV_TRANS_PREPARE) + ret = ds->drv->port_stp_update ? 0 : -EOPNOTSUPP; + else + ret = ds->drv->port_stp_update(ds, p->port, + attr->u.stp_state); break; default: ret = -EOPNOTSUPP; From b84f78782052ee4516903e5d0566a5eee365b771 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Tue, 29 Sep 2015 19:07:07 -0700 Subject: [PATCH 547/647] net: Initialize flow flags in input path The fib_table_lookup tracepoint found 2 places where the flowi4_flags is not initialized. Signed-off-by: David Ahern Signed-off-by: David S. Miller --- net/ipv4/fib_frontend.c | 1 + net/ipv4/route.c | 1 + 2 files changed, 2 insertions(+) diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 6fcbd215cdbc..690bcbc59f26 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -340,6 +340,7 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst, fl4.flowi4_tos = tos; fl4.flowi4_scope = RT_SCOPE_UNIVERSE; fl4.flowi4_tun_key.tun_id = 0; + fl4.flowi4_flags = 0; no_addr = idev->ifa_list == NULL; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index c6ad99ad0ffb..c81deb85acb4 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1737,6 +1737,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, fl4.flowi4_mark = skb->mark; fl4.flowi4_tos = tos; fl4.flowi4_scope = RT_SCOPE_UNIVERSE; + fl4.flowi4_flags = 0; fl4.daddr = daddr; fl4.saddr = saddr; err = fib_lookup(net, &fl4, &res, 0); From 1e034743e918d195d339af340ae933727c072bce Mon Sep 17 00:00:00 2001 From: Vitaly Kuznetsov Date: Wed, 23 Sep 2015 12:02:57 +0200 Subject: [PATCH 548/647] x86/hyperv: Fix the build in the !CONFIG_KEXEC_CORE case Recent changes in the Hyper-V driver: b4370df2b1f5 ("Drivers: hv: vmbus: add special crash handler") broke the build when CONFIG_KEXEC_CORE is not set: arch/x86/built-in.o: In function `hv_machine_crash_shutdown': arch/x86/kernel/cpu/mshyperv.c:112: undefined reference to `native_machine_crash_shutdown' Decorate all kexec related code with #ifdef CONFIG_KEXEC_CORE. Reported-by: Jim Davis Reported-by: Stephen Hemminger Signed-off-by: Vitaly Kuznetsov Signed-off-by: Thomas Gleixner Cc: devel@linuxdriverproject.org Cc: K. Y. Srinivasan Cc: Haiyang Zhang Cc: Greg Kroah-Hartman Cc: Peter Zijlstra Cc: Linus Torvalds Link: http://lkml.kernel.org/r/1443002577-25370-1-git-send-email-vkuznets@redhat.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/mshyperv.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c index 381c8b9b3a33..20e242ea1bc4 100644 --- a/arch/x86/kernel/cpu/mshyperv.c +++ b/arch/x86/kernel/cpu/mshyperv.c @@ -34,11 +34,10 @@ struct ms_hyperv_info ms_hyperv; EXPORT_SYMBOL_GPL(ms_hyperv); -static void (*hv_kexec_handler)(void); -static void (*hv_crash_handler)(struct pt_regs *regs); - #if IS_ENABLED(CONFIG_HYPERV) static void (*vmbus_handler)(void); +static void (*hv_kexec_handler)(void); +static void (*hv_crash_handler)(struct pt_regs *regs); void hyperv_vector_handler(struct pt_regs *regs) { @@ -96,8 +95,8 @@ void hv_remove_crash_handler(void) hv_crash_handler = NULL; } EXPORT_SYMBOL_GPL(hv_remove_crash_handler); -#endif +#ifdef CONFIG_KEXEC_CORE static void hv_machine_shutdown(void) { if (kexec_in_progress && hv_kexec_handler) @@ -111,7 +110,8 @@ static void hv_machine_crash_shutdown(struct pt_regs *regs) hv_crash_handler(regs); native_machine_crash_shutdown(regs); } - +#endif /* CONFIG_KEXEC_CORE */ +#endif /* CONFIG_HYPERV */ static uint32_t __init ms_hyperv_platform(void) { @@ -186,8 +186,10 @@ static void __init ms_hyperv_init_platform(void) no_timer_check = 1; #endif +#if IS_ENABLED(CONFIG_HYPERV) && defined(CONFIG_KEXEC_CORE) machine_ops.shutdown = hv_machine_shutdown; machine_ops.crash_shutdown = hv_machine_crash_shutdown; +#endif mark_tsc_unstable("running on Hyper-V"); } From 40482e64b0b84388561b00b880eeca7000f72d38 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Emilio=20L=C3=B3pez?= Date: Sun, 13 Sep 2015 17:15:53 -0300 Subject: [PATCH 549/647] dmaengine: sun4i: fix unsafe list iteration MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently, sun4i_dma_free_contract iterates over lists and frees memory as it goes through them, causing reads to recently freed memory to be performed. Fix this by using the safe version of the iterator, so freed memory is not referenced at all. Reported-by: Dan Carpenter Signed-off-by: Emilio López Acked-by: Maxime Ripard Signed-off-by: Vinod Koul --- drivers/dma/sun4i-dma.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/dma/sun4i-dma.c b/drivers/dma/sun4i-dma.c index a1a500d96ff2..1661d518224a 100644 --- a/drivers/dma/sun4i-dma.c +++ b/drivers/dma/sun4i-dma.c @@ -599,13 +599,13 @@ get_next_cyclic_promise(struct sun4i_dma_contract *contract) static void sun4i_dma_free_contract(struct virt_dma_desc *vd) { struct sun4i_dma_contract *contract = to_sun4i_dma_contract(vd); - struct sun4i_dma_promise *promise; + struct sun4i_dma_promise *promise, *tmp; /* Free all the demands and completed demands */ - list_for_each_entry(promise, &contract->demands, list) + list_for_each_entry_safe(promise, tmp, &contract->demands, list) kfree(promise); - list_for_each_entry(promise, &contract->completed_demands, list) + list_for_each_entry_safe(promise, tmp, &contract->completed_demands, list) kfree(promise); kfree(contract); From 4ac86a6dcec1c3878de9747bf5a2aa4455be69e3 Mon Sep 17 00:00:00 2001 From: Andrey Ryabinin Date: Tue, 29 Sep 2015 19:40:14 +0300 Subject: [PATCH 550/647] x86, efi, kasan: Fix build failure on !KASAN && KMEMCHECK=y kernels MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With KMEMCHECK=y, KASAN=n we get this build failure: arch/x86/platform/efi/efi.c:673:3: error: implicit declaration of function ‘memcpy’ [-Werror=implicit-function-declaration] arch/x86/platform/efi/efi_64.c:139:2: error: implicit declaration of function ‘memcpy’ [-Werror=implicit-function-declaration] arch/x86/include/asm/desc.h:121:2: error: implicit declaration of function ‘memcpy’ [-Werror=implicit-function-declaration] Don't #undef memcpy if KASAN=n. Reported-by: Ingo Molnar Reported-by: Sedat Dilek Signed-off-by: Andrey Ryabinin Cc: Linus Torvalds Cc: Matt Fleming Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Peter Zijlstra Cc: Thomas Gleixner Fixes: 769a8089c1fd ("x86, efi, kasan: #undef memset/memcpy/memmove per arch") Link: http://lkml.kernel.org/r/1443544814-20122-1-git-send-email-ryabinin.a.a@gmail.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/efi.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index ab5f1d447ef9..ae68be92f755 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -86,6 +86,7 @@ extern u64 asmlinkage efi_call(void *fp, ...); extern void __iomem *__init efi_ioremap(unsigned long addr, unsigned long size, u32 type, u64 attribute); +#ifdef CONFIG_KASAN /* * CONFIG_KASAN may redefine memset to __memset. __memset function is present * only in kernel binary. Since the EFI stub linked into a separate binary it @@ -95,6 +96,7 @@ extern void __iomem *__init efi_ioremap(unsigned long addr, unsigned long size, #undef memcpy #undef memset #undef memmove +#endif #endif /* CONFIG_X86_32 */ From 5e0fb1f9eb754eed8432392bfdc100ef295676cd Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Wed, 16 Sep 2015 14:16:45 +0900 Subject: [PATCH 551/647] drm/exynos: fix layering violation of address There is no guarantee that DMA addresses are the same as physical addresses, but dma_to_pfn() knows how to convert a dma_addr_t to a PFN which can then be converted to a struct page. Suggested-by: Russell King Signed-off-by: Joonyoung Shim Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_gem.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index f12fbc36b120..a4fb78e43034 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -77,7 +77,8 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem_obj *obj) start_addr = obj->dma_addr; while (i < nr_pages) { - obj->pages[i] = phys_to_page(start_addr); + obj->pages[i] = pfn_to_page(dma_to_pfn(dev->dev, + start_addr)); start_addr += PAGE_SIZE; i++; } From 7c93537a47e15e04064158ccdbe28f88ee61170c Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Wed, 16 Sep 2015 14:14:54 +0900 Subject: [PATCH 552/647] drm/exynos: fix missed calling of drm_prime_gem_destroy() When obj->import_attach is existed, code calling drm_prime_gem_destroy() was removed from commit 67e93c808b48 ("drm/exynos: stop copying sg table"), and it's a fault. The drm_prime_gem_destroy() is cleanup function which GEM drivers need to call when they use drm_gem_prime_import() to import dma-bufs, so exynos-drm driver using drm_gem_prime_import() needs calling drm_prime_gem_destroy(). Signed-off-by: Joonyoung Shim Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_gem.c | 9 +++++---- drivers/gpu/drm/exynos/exynos_drm_gem.h | 2 ++ 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index a4fb78e43034..ce05f1365712 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -157,11 +157,10 @@ void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj) * once dmabuf's refcount becomes 0. */ if (obj->import_attach) - goto out; + drm_prime_gem_destroy(obj, exynos_gem_obj->sgt); + else + exynos_drm_free_buf(exynos_gem_obj); - exynos_drm_free_buf(exynos_gem_obj); - -out: drm_gem_free_mmap_offset(obj); /* release file pointer to gem object. */ @@ -589,6 +588,8 @@ exynos_drm_gem_prime_import_sg_table(struct drm_device *dev, if (ret < 0) goto err_free_large; + exynos_gem_obj->sgt = sgt; + if (sgt->nents == 1) { /* always physically continuous memory if sgt->nents is 1. */ exynos_gem_obj->flags |= EXYNOS_BO_CONTIG; diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h index cd62f8410d1e..97bb48b6e862 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.h +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h @@ -39,6 +39,7 @@ * - this address could be physical address without IOMMU and * device address with IOMMU. * @pages: Array of backing pages. + * @sgt: Imported sg_table. * * P.S. this object would be transferred to user as kms_bo.handle so * user can access the buffer through kms_bo.handle. @@ -52,6 +53,7 @@ struct exynos_drm_gem_obj { dma_addr_t dma_addr; struct dma_attrs dma_attrs; struct page **pages; + struct sg_table *sgt; }; struct page **exynos_gem_get_pages(struct drm_gem_object *obj, gfp_t gfpmask); From 0269b1a17f8f887c02ae37affef884e1ffda3df6 Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Wed, 16 Sep 2015 14:29:32 +0900 Subject: [PATCH 553/647] drm/exynos: remove unnecessary NULL assignment They will be freed right or was freed already, so NULL assignment is unnecessary. Signed-off-by: Joonyoung Shim Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_gem.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index ce05f1365712..85aeae91b50c 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -118,8 +118,6 @@ static void exynos_drm_free_buf(struct exynos_drm_gem_obj *obj) } else dma_free_attrs(dev->dev, obj->size, obj->pages, (dma_addr_t)obj->dma_addr, &obj->dma_attrs); - - obj->dma_addr = (dma_addr_t)NULL; } static int exynos_drm_gem_handle_create(struct drm_gem_object *obj, @@ -167,7 +165,6 @@ void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj) drm_gem_object_release(obj); kfree(exynos_gem_obj); - exynos_gem_obj = NULL; } unsigned long exynos_drm_gem_get_size(struct drm_device *dev, From b319dc6a61518cc02ae21afb383901236009aab1 Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Wed, 16 Sep 2015 14:29:33 +0900 Subject: [PATCH 554/647] drm/exynos: staticize exynos_drm_gem_init() The exynos_drm_gem_init() is used only in exynos_drm_gem.c file. Make it static and don't export it. Signed-off-by: Joonyoung Shim Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_gem.c | 3 +-- drivers/gpu/drm/exynos/exynos_drm_gem.h | 4 ---- 2 files changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index 85aeae91b50c..5638a7a88c7b 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -187,8 +187,7 @@ unsigned long exynos_drm_gem_get_size(struct drm_device *dev, return exynos_gem_obj->size; } - -struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev, +static struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev, unsigned long size) { struct exynos_drm_gem_obj *exynos_gem_obj; diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h index 97bb48b6e862..b62d1007c0e0 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.h +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h @@ -61,10 +61,6 @@ struct page **exynos_gem_get_pages(struct drm_gem_object *obj, gfp_t gfpmask); /* destroy a buffer with gem object */ void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj); -/* create a private gem object and initialize it. */ -struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev, - unsigned long size); - /* create a new buffer with gem object */ struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev, unsigned int flags, From 333e8e58b0bed137e940ac92c2af3f3aa4603504 Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Wed, 16 Sep 2015 14:29:34 +0900 Subject: [PATCH 555/647] drm/exynos: cleanup function calling written twice By if statment, some function callings are written twice. It needs several line feed by indentation in if statment. Make to one function calling from outside if statment. Signed-off-by: Joonyoung Shim Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_gem.c | 56 ++++++++++--------------- 1 file changed, 23 insertions(+), 33 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index 5638a7a88c7b..10fc61408f22 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -56,24 +56,25 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem_obj *obj) nr_pages = obj->size >> PAGE_SHIFT; if (!is_drm_iommu_supported(dev)) { - dma_addr_t start_addr; - unsigned int i = 0; - obj->pages = drm_calloc_large(nr_pages, sizeof(struct page *)); if (!obj->pages) { DRM_ERROR("failed to allocate pages.\n"); return -ENOMEM; } + } - obj->cookie = dma_alloc_attrs(dev->dev, - obj->size, - &obj->dma_addr, GFP_KERNEL, - &obj->dma_attrs); - if (!obj->cookie) { - DRM_ERROR("failed to allocate buffer.\n"); + obj->cookie = dma_alloc_attrs(dev->dev, obj->size, &obj->dma_addr, + GFP_KERNEL, &obj->dma_attrs); + if (!obj->cookie) { + DRM_ERROR("failed to allocate buffer.\n"); + if (obj->pages) drm_free_large(obj->pages); - return -ENOMEM; - } + return -ENOMEM; + } + + if (obj->pages) { + dma_addr_t start_addr; + unsigned int i = 0; start_addr = obj->dma_addr; while (i < nr_pages) { @@ -83,13 +84,7 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem_obj *obj) i++; } } else { - obj->pages = dma_alloc_attrs(dev->dev, obj->size, - &obj->dma_addr, GFP_KERNEL, - &obj->dma_attrs); - if (!obj->pages) { - DRM_ERROR("failed to allocate buffer.\n"); - return -ENOMEM; - } + obj->pages = obj->cookie; } DRM_DEBUG_KMS("dma_addr(0x%lx), size(0x%lx)\n", @@ -111,13 +106,11 @@ static void exynos_drm_free_buf(struct exynos_drm_gem_obj *obj) DRM_DEBUG_KMS("dma_addr(0x%lx), size(0x%lx)\n", (unsigned long)obj->dma_addr, obj->size); - if (!is_drm_iommu_supported(dev)) { - dma_free_attrs(dev->dev, obj->size, obj->cookie, - (dma_addr_t)obj->dma_addr, &obj->dma_attrs); + dma_free_attrs(dev->dev, obj->size, obj->cookie, + (dma_addr_t)obj->dma_addr, &obj->dma_attrs); + + if (!is_drm_iommu_supported(dev)) drm_free_large(obj->pages); - } else - dma_free_attrs(dev->dev, obj->size, obj->pages, - (dma_addr_t)obj->dma_addr, &obj->dma_attrs); } static int exynos_drm_gem_handle_create(struct drm_gem_object *obj, @@ -398,6 +391,7 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv, struct drm_mode_create_dumb *args) { struct exynos_drm_gem_obj *exynos_gem_obj; + unsigned int flags; int ret; /* @@ -409,16 +403,12 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv, args->pitch = args->width * ((args->bpp + 7) / 8); args->size = args->pitch * args->height; - if (is_drm_iommu_supported(dev)) { - exynos_gem_obj = exynos_drm_gem_create(dev, - EXYNOS_BO_NONCONTIG | EXYNOS_BO_WC, - args->size); - } else { - exynos_gem_obj = exynos_drm_gem_create(dev, - EXYNOS_BO_CONTIG | EXYNOS_BO_WC, - args->size); - } + if (is_drm_iommu_supported(dev)) + flags = EXYNOS_BO_NONCONTIG | EXYNOS_BO_WC; + else + flags = EXYNOS_BO_CONTIG | EXYNOS_BO_WC; + exynos_gem_obj = exynos_drm_gem_create(dev, flags, args->size); if (IS_ERR(exynos_gem_obj)) { dev_warn(dev->dev, "FB allocation failed.\n"); return PTR_ERR(exynos_gem_obj); From b4cfd4ddfb555c8cda7b0aa55c0e37522f8e01e4 Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Wed, 16 Sep 2015 14:29:35 +0900 Subject: [PATCH 556/647] drm/exynos: cleanup line feed in exynos_drm_gem_get_ioctl The beginning of statement in function is next line of a brace. Signed-off-by: Joonyoung Shim Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_gem.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index 10fc61408f22..3df1e0e81014 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -331,7 +331,8 @@ int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem_obj *exynos_gem_obj, int exynos_drm_gem_get_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) -{ struct exynos_drm_gem_obj *exynos_gem_obj; +{ + struct exynos_drm_gem_obj *exynos_gem_obj; struct drm_exynos_gem_info *args = data; struct drm_gem_object *obj; From b9e71911f34616276399da3dc1abc47797be2396 Mon Sep 17 00:00:00 2001 From: Daniel Kurtz Date: Wed, 9 Sep 2015 12:55:25 -0300 Subject: [PATCH 557/647] drm/exynos: Remove useless EXPORT_SYMBOL_GPLs All the user of these functions are inside exynos-drm driver and you don't need to export the symbols for that case. Signed-off-by: Daniel Kurtz Signed-off-by: Gustavo Padovan Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_core.c | 6 ------ drivers/gpu/drm/exynos/exynos_drm_g2d.c | 3 --- 2 files changed, 9 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_core.c b/drivers/gpu/drm/exynos/exynos_drm_core.c index c68a6a2a9b57..7f55ba6771c6 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_core.c +++ b/drivers/gpu/drm/exynos/exynos_drm_core.c @@ -28,7 +28,6 @@ int exynos_drm_subdrv_register(struct exynos_drm_subdrv *subdrv) return 0; } -EXPORT_SYMBOL_GPL(exynos_drm_subdrv_register); int exynos_drm_subdrv_unregister(struct exynos_drm_subdrv *subdrv) { @@ -39,7 +38,6 @@ int exynos_drm_subdrv_unregister(struct exynos_drm_subdrv *subdrv) return 0; } -EXPORT_SYMBOL_GPL(exynos_drm_subdrv_unregister); int exynos_drm_device_subdrv_probe(struct drm_device *dev) { @@ -69,7 +67,6 @@ int exynos_drm_device_subdrv_probe(struct drm_device *dev) return 0; } -EXPORT_SYMBOL_GPL(exynos_drm_device_subdrv_probe); int exynos_drm_device_subdrv_remove(struct drm_device *dev) { @@ -87,7 +84,6 @@ int exynos_drm_device_subdrv_remove(struct drm_device *dev) return 0; } -EXPORT_SYMBOL_GPL(exynos_drm_device_subdrv_remove); int exynos_drm_subdrv_open(struct drm_device *dev, struct drm_file *file) { @@ -111,7 +107,6 @@ err: } return ret; } -EXPORT_SYMBOL_GPL(exynos_drm_subdrv_open); void exynos_drm_subdrv_close(struct drm_device *dev, struct drm_file *file) { @@ -122,4 +117,3 @@ void exynos_drm_subdrv_close(struct drm_device *dev, struct drm_file *file) subdrv->close(dev, subdrv->dev, file); } } -EXPORT_SYMBOL_GPL(exynos_drm_subdrv_close); diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c index 3734c34aed16..c17efdb238a6 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c @@ -1059,7 +1059,6 @@ int exynos_g2d_get_ver_ioctl(struct drm_device *drm_dev, void *data, return 0; } -EXPORT_SYMBOL_GPL(exynos_g2d_get_ver_ioctl); int exynos_g2d_set_cmdlist_ioctl(struct drm_device *drm_dev, void *data, struct drm_file *file) @@ -1230,7 +1229,6 @@ err: g2d_put_cmdlist(g2d, node); return ret; } -EXPORT_SYMBOL_GPL(exynos_g2d_set_cmdlist_ioctl); int exynos_g2d_exec_ioctl(struct drm_device *drm_dev, void *data, struct drm_file *file) @@ -1293,7 +1291,6 @@ int exynos_g2d_exec_ioctl(struct drm_device *drm_dev, void *data, out: return 0; } -EXPORT_SYMBOL_GPL(exynos_g2d_exec_ioctl); static int g2d_subdrv_probe(struct drm_device *drm_dev, struct device *dev) { From 8755556fbbdd1a6af21a3c4d9ce8f451e999e457 Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Fri, 25 Sep 2015 18:10:32 +0900 Subject: [PATCH 558/647] drm/exynos: remove call to drm_gem_free_mmap_offset() The drm_gem_object_release() function already performs this cleanup, so there is no reason to do it explicitly. Signed-off-by: Joonyoung Shim Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_gem.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index 3df1e0e81014..5c94863a4ca6 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -152,8 +152,6 @@ void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj) else exynos_drm_free_buf(exynos_gem_obj); - drm_gem_free_mmap_offset(obj); - /* release file pointer to gem object. */ drm_gem_object_release(obj); @@ -530,7 +528,6 @@ int exynos_drm_gem_mmap(struct file *filp, struct vm_area_struct *vma) err_close_vm: drm_gem_vm_close(vma); - drm_gem_free_mmap_offset(obj); return ret; } From 48cf53f4343ae12ddc1c60dbe116161ecf7a2885 Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Tue, 28 Jul 2015 17:53:23 +0900 Subject: [PATCH 559/647] drm/exynos: create a fake mmap offset with gem creation Don't create a fake mmap offset in exynos_drm_gem_dumb_map_offset. If not, it will call drm_gem_create_mmap_offset whenever user requests DRM_IOCTL_MODE_MAP_DUMB ioctl. Signed-off-by: Joonyoung Shim Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_gem.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index 5c94863a4ca6..048ad13e5d82 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -199,6 +199,13 @@ static struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev, return ERR_PTR(ret); } + ret = drm_gem_create_mmap_offset(obj); + if (ret < 0) { + drm_gem_object_release(obj); + kfree(exynos_gem_obj); + return ERR_PTR(ret); + } + DRM_DEBUG_KMS("created file object = 0x%x\n", (unsigned int)obj->filp); return exynos_gem_obj; @@ -445,14 +452,9 @@ int exynos_drm_gem_dumb_map_offset(struct drm_file *file_priv, goto unlock; } - ret = drm_gem_create_mmap_offset(obj); - if (ret) - goto out; - *offset = drm_vma_node_offset_addr(&obj->vma_node); DRM_DEBUG_KMS("offset = 0x%lx\n", (unsigned long)*offset); -out: drm_gem_object_unreference(obj); unlock: mutex_unlock(&dev->struct_mutex); From 7082947eff3b1ddbb5d320331c846a5a2fc88261 Mon Sep 17 00:00:00 2001 From: Thierry Reding Date: Thu, 24 Sep 2015 12:14:13 +0200 Subject: [PATCH 560/647] drm/exynos: Suspend/resume is unused if !PM Protect the suspend and resume callbacks with an #ifdef CONFIG_PM_SLEEP guard to avoid "defined but not used" warnings. Signed-off-by: Thierry Reding Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_drv.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index 831d2e4cacf9..ae9e6b2d3758 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -304,6 +304,7 @@ int exynos_atomic_commit(struct drm_device *dev, struct drm_atomic_state *state, return 0; } +#ifdef CONFIG_PM_SLEEP static int exynos_drm_suspend(struct drm_device *dev, pm_message_t state) { struct drm_connector *connector; @@ -340,6 +341,7 @@ static int exynos_drm_resume(struct drm_device *dev) return 0; } +#endif static int exynos_drm_open(struct drm_device *dev, struct drm_file *file) { From 641a2fef39369c27df887e82cd63aee3de93f511 Mon Sep 17 00:00:00 2001 From: Thierry Reding Date: Thu, 24 Sep 2015 12:14:14 +0200 Subject: [PATCH 561/647] drm/exynos: fimc: Clock control is unused if !PM Protect the fimc_clk_ctrl() function with an #ifdef CONFIG_PM guard to avoid "defined but not used" warnings. Signed-off-by: Thierry Reding Reviewed-by: Andrzej Hajda Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_fimc.c | 36 ++++++++++++------------ 1 file changed, 18 insertions(+), 18 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index 2a652359af64..dd3a5e6d58c8 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -1206,23 +1206,6 @@ static struct exynos_drm_ipp_ops fimc_dst_ops = { .set_addr = fimc_dst_set_addr, }; -static int fimc_clk_ctrl(struct fimc_context *ctx, bool enable) -{ - DRM_DEBUG_KMS("enable[%d]\n", enable); - - if (enable) { - clk_prepare_enable(ctx->clocks[FIMC_CLK_GATE]); - clk_prepare_enable(ctx->clocks[FIMC_CLK_WB_A]); - ctx->suspended = false; - } else { - clk_disable_unprepare(ctx->clocks[FIMC_CLK_GATE]); - clk_disable_unprepare(ctx->clocks[FIMC_CLK_WB_A]); - ctx->suspended = true; - } - - return 0; -} - static irqreturn_t fimc_irq_handler(int irq, void *dev_id) { struct fimc_context *ctx = dev_id; @@ -1780,6 +1763,24 @@ static int fimc_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM +static int fimc_clk_ctrl(struct fimc_context *ctx, bool enable) +{ + DRM_DEBUG_KMS("enable[%d]\n", enable); + + if (enable) { + clk_prepare_enable(ctx->clocks[FIMC_CLK_GATE]); + clk_prepare_enable(ctx->clocks[FIMC_CLK_WB_A]); + ctx->suspended = false; + } else { + clk_disable_unprepare(ctx->clocks[FIMC_CLK_GATE]); + clk_disable_unprepare(ctx->clocks[FIMC_CLK_WB_A]); + ctx->suspended = true; + } + + return 0; +} + #ifdef CONFIG_PM_SLEEP static int fimc_suspend(struct device *dev) { @@ -1806,7 +1807,6 @@ static int fimc_resume(struct device *dev) } #endif -#ifdef CONFIG_PM static int fimc_runtime_suspend(struct device *dev) { struct fimc_context *ctx = get_fimc_context(dev); From 85a82038b2744f87cb297f93ddecd04a8c2979bd Mon Sep 17 00:00:00 2001 From: Thierry Reding Date: Thu, 24 Sep 2015 12:14:15 +0200 Subject: [PATCH 562/647] drm/exynos: rotator: Clock control is unused if !PM Protect the rotator_clk_crtl() function with an #ifdef CONFIG_PM guard to avoid "defined but not used" warnings. Signed-off-by: Thierry Reding Reviewed-by: Andrzej Hajda Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_rotator.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c index 425e70625388..2f5c118f4c8e 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c +++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c @@ -786,6 +786,7 @@ static int rotator_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM static int rotator_clk_crtl(struct rot_context *rot, bool enable) { if (enable) { @@ -822,7 +823,6 @@ static int rotator_resume(struct device *dev) } #endif -#ifdef CONFIG_PM static int rotator_runtime_suspend(struct device *dev) { struct rot_context *rot = dev_get_drvdata(dev); From 214fc4e423ff38b41b60db2209cf49b4e9a7209b Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Thu, 24 Sep 2015 12:03:35 +0300 Subject: [PATCH 563/647] dmaengine: fix balance of privatecnt dma_release_channel() decrements privatecnt counter and almost all dma_get* function increments it with the exception of dma_get_slave_channel(). In most cases this does not cause issue since normally the channel is not requested and released, but if a driver requests DMA channel via dma_get_slave_channel() and releases the channel the privatecnt will be unbalanced and this will prevent for example getting channel for memcpy. Signed-off-by: Peter Ujfalusi Signed-off-by: Vinod Koul --- drivers/dma/dmaengine.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index 3ff284c8e3d5..09479d4be4db 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -554,10 +554,18 @@ struct dma_chan *dma_get_slave_channel(struct dma_chan *chan) mutex_lock(&dma_list_mutex); if (chan->client_count == 0) { + struct dma_device *device = chan->device; + + dma_cap_set(DMA_PRIVATE, device->cap_mask); + device->privatecnt++; err = dma_chan_get(chan); - if (err) + if (err) { pr_debug("%s: failed to get %s: (%d)\n", __func__, dma_chan_name(chan), err); + chan = NULL; + if (--device->privatecnt == 0) + dma_cap_clear(DMA_PRIVATE, device->cap_mask); + } } else chan = NULL; From ee08b59d47d859ed0a11ab331a3fbc5ab3b56100 Mon Sep 17 00:00:00 2001 From: Rameshwar Prasad Sahu Date: Wed, 16 Sep 2015 13:33:23 +0530 Subject: [PATCH 564/647] dmaengine: xgene-dma: Fix overwritting DMA tx ring This patch fixes an over flow issue with the TX ring descriptor. Each descriptor is 32B in size and an operation requires 2 of these descriptors. Signed-off-by: Rameshwar Prasad Sahu Signed-off-by: Vinod Koul --- drivers/dma/xgene-dma.c | 37 +++++++++++-------------------------- 1 file changed, 11 insertions(+), 26 deletions(-) diff --git a/drivers/dma/xgene-dma.c b/drivers/dma/xgene-dma.c index 4c5dddaad2d5..8d57b1b12e41 100644 --- a/drivers/dma/xgene-dma.c +++ b/drivers/dma/xgene-dma.c @@ -59,7 +59,6 @@ #define XGENE_DMA_RING_MEM_RAM_SHUTDOWN 0xD070 #define XGENE_DMA_RING_BLK_MEM_RDY 0xD074 #define XGENE_DMA_RING_BLK_MEM_RDY_VAL 0xFFFFFFFF -#define XGENE_DMA_RING_DESC_CNT(v) (((v) & 0x0001FFFE) >> 1) #define XGENE_DMA_RING_ID_GET(owner, num) (((owner) << 6) | (num)) #define XGENE_DMA_RING_DST_ID(v) ((1 << 10) | (v)) #define XGENE_DMA_RING_CMD_OFFSET 0x2C @@ -379,14 +378,6 @@ static u8 xgene_dma_encode_xor_flyby(u32 src_cnt) return flyby_type[src_cnt]; } -static u32 xgene_dma_ring_desc_cnt(struct xgene_dma_ring *ring) -{ - u32 __iomem *cmd_base = ring->cmd_base; - u32 ring_state = ioread32(&cmd_base[1]); - - return XGENE_DMA_RING_DESC_CNT(ring_state); -} - static void xgene_dma_set_src_buffer(__le64 *ext8, size_t *len, dma_addr_t *paddr) { @@ -659,15 +650,12 @@ static void xgene_dma_clean_running_descriptor(struct xgene_dma_chan *chan, dma_pool_free(chan->desc_pool, desc, desc->tx.phys); } -static int xgene_chan_xfer_request(struct xgene_dma_ring *ring, - struct xgene_dma_desc_sw *desc_sw) +static void xgene_chan_xfer_request(struct xgene_dma_chan *chan, + struct xgene_dma_desc_sw *desc_sw) { + struct xgene_dma_ring *ring = &chan->tx_ring; struct xgene_dma_desc_hw *desc_hw; - /* Check if can push more descriptor to hw for execution */ - if (xgene_dma_ring_desc_cnt(ring) > (ring->slots - 2)) - return -EBUSY; - /* Get hw descriptor from DMA tx ring */ desc_hw = &ring->desc_hw[ring->head]; @@ -694,11 +682,13 @@ static int xgene_chan_xfer_request(struct xgene_dma_ring *ring, memcpy(desc_hw, &desc_sw->desc2, sizeof(*desc_hw)); } + /* Increment the pending transaction count */ + chan->pending += ((desc_sw->flags & + XGENE_DMA_FLAG_64B_DESC) ? 2 : 1); + /* Notify the hw that we have descriptor ready for execution */ iowrite32((desc_sw->flags & XGENE_DMA_FLAG_64B_DESC) ? 2 : 1, ring->cmd); - - return 0; } /** @@ -710,7 +700,6 @@ static int xgene_chan_xfer_request(struct xgene_dma_ring *ring, static void xgene_chan_xfer_ld_pending(struct xgene_dma_chan *chan) { struct xgene_dma_desc_sw *desc_sw, *_desc_sw; - int ret; /* * If the list of pending descriptors is empty, then we @@ -735,18 +724,13 @@ static void xgene_chan_xfer_ld_pending(struct xgene_dma_chan *chan) if (chan->pending >= chan->max_outstanding) return; - ret = xgene_chan_xfer_request(&chan->tx_ring, desc_sw); - if (ret) - return; + xgene_chan_xfer_request(chan, desc_sw); /* * Delete this element from ld pending queue and append it to * ld running queue */ list_move_tail(&desc_sw->node, &chan->ld_running); - - /* Increment the pending transaction count */ - chan->pending++; } } @@ -821,7 +805,8 @@ static void xgene_dma_cleanup_descriptors(struct xgene_dma_chan *chan) * Decrement the pending transaction count * as we have processed one */ - chan->pending--; + chan->pending -= ((desc_sw->flags & + XGENE_DMA_FLAG_64B_DESC) ? 2 : 1); /* * Delete this node from ld running queue and append it to @@ -1485,7 +1470,7 @@ static int xgene_dma_create_chan_rings(struct xgene_dma_chan *chan) tx_ring->id, tx_ring->num, tx_ring->desc_vaddr); /* Set the max outstanding request possible to this channel */ - chan->max_outstanding = rx_ring->slots; + chan->max_outstanding = tx_ring->slots; return ret; } From 50bbfbffa5c894def440ce8157dfe53e60960d35 Mon Sep 17 00:00:00 2001 From: Gustavo Padovan Date: Tue, 29 Sep 2015 17:59:17 -0300 Subject: [PATCH 565/647] drm/exynos: remove fimd_mode_fixup() The only thing mode_fixup was doing was set the adjusted_mode->vrefresh to 60, but it already has the value of 60 when the fimd_mode_fixup() is called. That means this call is actually pointless and can be removed. Signed-off-by: Gustavo Padovan Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_fimd.c | 12 ------------ 1 file changed, 12 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index 750a9e6b9e8d..03ad549e7698 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -41,7 +41,6 @@ * CPU Interface. */ -#define FIMD_DEFAULT_FRAMERATE 60 #define MIN_FB_WIDTH_FOR_16WORD_BURST 128 /* position control register for hardware window 0, 2 ~ 4.*/ @@ -377,16 +376,6 @@ static u32 fimd_calc_clkdiv(struct fimd_context *ctx, return (clkdiv < 0x100) ? clkdiv : 0xff; } -static bool fimd_mode_fixup(struct exynos_drm_crtc *crtc, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) -{ - if (adjusted_mode->vrefresh == 0) - adjusted_mode->vrefresh = FIMD_DEFAULT_FRAMERATE; - - return true; -} - static void fimd_commit(struct exynos_drm_crtc *crtc) { struct fimd_context *ctx = crtc->ctx; @@ -888,7 +877,6 @@ static void fimd_dp_clock_enable(struct exynos_drm_crtc *crtc, bool enable) static const struct exynos_drm_crtc_ops fimd_crtc_ops = { .enable = fimd_enable, .disable = fimd_disable, - .mode_fixup = fimd_mode_fixup, .commit = fimd_commit, .enable_vblank = fimd_enable_vblank, .disable_vblank = fimd_disable_vblank, From 208d7ee3c89e5e94627aa33331b15dd4df814707 Mon Sep 17 00:00:00 2001 From: Gustavo Padovan Date: Tue, 29 Sep 2015 17:59:18 -0300 Subject: [PATCH 566/647] drm/exynos: remove decon_mode_fixup() The only thing mode_fixup was doing was set the adjusted_mode->vrefresh to 60, but it already has the value of 60 when the decon_mode_fixup() is called. That means this call is actually pointless and can be removed. Signed-off-by: Gustavo Padovan Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos7_drm_decon.c | 12 ------------ 1 file changed, 12 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c index cbdb78ef3bac..e6cbaca821a4 100644 --- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c @@ -37,7 +37,6 @@ * DECON stands for Display and Enhancement controller. */ -#define DECON_DEFAULT_FRAMERATE 60 #define MIN_FB_WIDTH_FOR_16WORD_BURST 128 #define WINDOWS_NR 2 @@ -165,16 +164,6 @@ static u32 decon_calc_clkdiv(struct decon_context *ctx, return (clkdiv < 0x100) ? clkdiv : 0xff; } -static bool decon_mode_fixup(struct exynos_drm_crtc *crtc, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) -{ - if (adjusted_mode->vrefresh == 0) - adjusted_mode->vrefresh = DECON_DEFAULT_FRAMERATE; - - return true; -} - static void decon_commit(struct exynos_drm_crtc *crtc) { struct decon_context *ctx = crtc->ctx; @@ -637,7 +626,6 @@ static void decon_disable(struct exynos_drm_crtc *crtc) static const struct exynos_drm_crtc_ops decon_crtc_ops = { .enable = decon_enable, .disable = decon_disable, - .mode_fixup = decon_mode_fixup, .commit = decon_commit, .enable_vblank = decon_enable_vblank, .disable_vblank = decon_disable_vblank, From 10d97d5869f9a50eb3142c7ee562ecc5b5b33f5b Mon Sep 17 00:00:00 2001 From: Gustavo Padovan Date: Tue, 29 Sep 2015 17:59:19 -0300 Subject: [PATCH 567/647] drm/exynos: remove unused mode_fixup() code CRTC's mode_fixup() isn't used anymore in exynos, remove it. Signed-off-by: Gustavo Padovan Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_crtc.c | 15 --------------- drivers/gpu/drm/exynos/exynos_drm_drv.h | 4 ---- 2 files changed, 19 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c index 0872aa2f450f..ed28823d3b35 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c @@ -41,20 +41,6 @@ static void exynos_drm_crtc_disable(struct drm_crtc *crtc) exynos_crtc->ops->disable(exynos_crtc); } -static bool -exynos_drm_crtc_mode_fixup(struct drm_crtc *crtc, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) -{ - struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc); - - if (exynos_crtc->ops->mode_fixup) - return exynos_crtc->ops->mode_fixup(exynos_crtc, mode, - adjusted_mode); - - return true; -} - static void exynos_drm_crtc_mode_set_nofb(struct drm_crtc *crtc) { @@ -99,7 +85,6 @@ static void exynos_crtc_atomic_flush(struct drm_crtc *crtc, static struct drm_crtc_helper_funcs exynos_crtc_helper_funcs = { .enable = exynos_drm_crtc_enable, .disable = exynos_drm_crtc_disable, - .mode_fixup = exynos_drm_crtc_mode_fixup, .mode_set_nofb = exynos_drm_crtc_mode_set_nofb, .atomic_begin = exynos_crtc_atomic_begin, .atomic_flush = exynos_crtc_atomic_flush, diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index b7ba21dfb696..6c717ba672db 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -82,7 +82,6 @@ struct exynos_drm_plane { * * @enable: enable the device * @disable: disable the device - * @mode_fixup: fix mode data before applying it * @commit: set current hw specific display mode to hw. * @enable_vblank: specific driver callback for enabling vblank interrupt. * @disable_vblank: specific driver callback for disabling vblank interrupt. @@ -103,9 +102,6 @@ struct exynos_drm_crtc; struct exynos_drm_crtc_ops { void (*enable)(struct exynos_drm_crtc *crtc); void (*disable)(struct exynos_drm_crtc *crtc); - bool (*mode_fixup)(struct exynos_drm_crtc *crtc, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode); void (*commit)(struct exynos_drm_crtc *crtc); int (*enable_vblank)(struct exynos_drm_crtc *crtc); void (*disable_vblank)(struct exynos_drm_crtc *crtc); From 6bea0f6d1c47b07be88dfd93f013ae05fcb3d8bf Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 28 Sep 2015 18:57:03 +0300 Subject: [PATCH 568/647] dmaengine: dw: properly read DWC_PARAMS register In case we have less than maximum allowed channels (8) and autoconfiguration is enabled the DWC_PARAMS read is wrong because it uses different arithmetic to what is needed for channel priority setup. Re-do the caclulations properly. This now works on AVR32 board well. Fixes: fed2574b3c9f (dw_dmac: introduce software emulation of LLP transfers) Cc: yitian.bu@tangramtek.com Signed-off-by: Andy Shevchenko Signed-off-by: Vinod Koul --- drivers/dma/dw/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/dma/dw/core.c b/drivers/dma/dw/core.c index cf1c87fa1edd..bedce038c6e2 100644 --- a/drivers/dma/dw/core.c +++ b/drivers/dma/dw/core.c @@ -1591,7 +1591,6 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata) INIT_LIST_HEAD(&dw->dma.channels); for (i = 0; i < nr_channels; i++) { struct dw_dma_chan *dwc = &dw->chan[i]; - int r = nr_channels - i - 1; dwc->chan.device = &dw->dma; dma_cookie_init(&dwc->chan); @@ -1603,7 +1602,7 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata) /* 7 is highest priority & 0 is lowest. */ if (pdata->chan_priority == CHAN_PRIORITY_ASCENDING) - dwc->priority = r; + dwc->priority = nr_channels - i - 1; else dwc->priority = i; @@ -1622,6 +1621,7 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata) /* Hardware configuration */ if (autocfg) { unsigned int dwc_params; + unsigned int r = DW_DMA_MAX_NR_CHANNELS - i - 1; void __iomem *addr = chip->regs + r * sizeof(u32); dwc_params = dma_read_byaddr(addr, DWC_PARAMS); From 96fc7a9cee671f10e14aaca44833696a71f1ebdb Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 30 Sep 2015 14:24:31 +0200 Subject: [PATCH 569/647] MIPS: Wire up userfaultfd and membarrier syscalls. Signed-off-by: Ralf Baechle --- arch/mips/include/uapi/asm/unistd.h | 18 ++++++++++++------ arch/mips/kernel/scall32-o32.S | 2 ++ arch/mips/kernel/scall64-64.S | 2 ++ arch/mips/kernel/scall64-n32.S | 2 ++ arch/mips/kernel/scall64-o32.S | 2 ++ 5 files changed, 20 insertions(+), 6 deletions(-) diff --git a/arch/mips/include/uapi/asm/unistd.h b/arch/mips/include/uapi/asm/unistd.h index c03088f9f514..cfabadb135d9 100644 --- a/arch/mips/include/uapi/asm/unistd.h +++ b/arch/mips/include/uapi/asm/unistd.h @@ -377,16 +377,18 @@ #define __NR_memfd_create (__NR_Linux + 354) #define __NR_bpf (__NR_Linux + 355) #define __NR_execveat (__NR_Linux + 356) +#define __NR_userfaultfd (__NR_Linux + 357) +#define __NR_membarrier (__NR_Linux + 358) /* * Offset of the last Linux o32 flavoured syscall */ -#define __NR_Linux_syscalls 356 +#define __NR_Linux_syscalls 358 #endif /* _MIPS_SIM == _MIPS_SIM_ABI32 */ #define __NR_O32_Linux 4000 -#define __NR_O32_Linux_syscalls 356 +#define __NR_O32_Linux_syscalls 358 #if _MIPS_SIM == _MIPS_SIM_ABI64 @@ -711,16 +713,18 @@ #define __NR_memfd_create (__NR_Linux + 314) #define __NR_bpf (__NR_Linux + 315) #define __NR_execveat (__NR_Linux + 316) +#define __NR_userfaultfd (__NR_Linux + 317) +#define __NR_membarrier (__NR_Linux + 318) /* * Offset of the last Linux 64-bit flavoured syscall */ -#define __NR_Linux_syscalls 316 +#define __NR_Linux_syscalls 318 #endif /* _MIPS_SIM == _MIPS_SIM_ABI64 */ #define __NR_64_Linux 5000 -#define __NR_64_Linux_syscalls 316 +#define __NR_64_Linux_syscalls 318 #if _MIPS_SIM == _MIPS_SIM_NABI32 @@ -1049,15 +1053,17 @@ #define __NR_memfd_create (__NR_Linux + 318) #define __NR_bpf (__NR_Linux + 319) #define __NR_execveat (__NR_Linux + 320) +#define __NR_userfaultfd (__NR_Linux + 321) +#define __NR_membarrier (__NR_Linux + 322) /* * Offset of the last N32 flavoured syscall */ -#define __NR_Linux_syscalls 320 +#define __NR_Linux_syscalls 322 #endif /* _MIPS_SIM == _MIPS_SIM_NABI32 */ #define __NR_N32_Linux 6000 -#define __NR_N32_Linux_syscalls 320 +#define __NR_N32_Linux_syscalls 322 #endif /* _UAPI_ASM_UNISTD_H */ diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S index 4cc13508d967..d06e30c1e4cd 100644 --- a/arch/mips/kernel/scall32-o32.S +++ b/arch/mips/kernel/scall32-o32.S @@ -599,3 +599,5 @@ EXPORT(sys_call_table) PTR sys_memfd_create PTR sys_bpf /* 4355 */ PTR sys_execveat + PTR sys_userfaultfd + PTR sys_membarrier diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S index a6f6b762c47a..428c2cd2326c 100644 --- a/arch/mips/kernel/scall64-64.S +++ b/arch/mips/kernel/scall64-64.S @@ -436,4 +436,6 @@ EXPORT(sys_call_table) PTR sys_memfd_create PTR sys_bpf /* 5315 */ PTR sys_execveat + PTR sys_userfaultfd + PTR sys_membarrier .size sys_call_table,.-sys_call_table diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 4b2010654c46..3868cf4c83df 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -429,4 +429,6 @@ EXPORT(sysn32_call_table) PTR sys_memfd_create PTR sys_bpf PTR compat_sys_execveat /* 6320 */ + PTR sys_userfaultfd + PTR sys_membarrier .size sysn32_call_table,.-sysn32_call_table diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index f543ff4feef9..6622980d82ea 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -584,4 +584,6 @@ EXPORT(sys32_call_table) PTR sys_memfd_create PTR sys_bpf /* 4355 */ PTR compat_sys_execveat + PTR sys_userfaultfd + PTR sys_membarrier .size sys32_call_table,.-sys32_call_table From 575f9c8604e0b4c7b36fb41fc5fd280a3c336906 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Tue, 29 Sep 2015 07:49:56 -0700 Subject: [PATCH 570/647] drm/vmwgfx: Fix a command submission hang regression When we're out of command buffer space, we turn on the command buffer processed irq without re-checking for finished command buffers afterwards. This might lead to a missed irq and the command submission process waiting forever for space. Fix this by rerunning the command buffer submission handler whenever we're out of command space. This ensures both that we don't needlessly turn on the irq, and that if we decide to turn on the irq, we recheck for finished command buffers before going to sleep. Reported-and-tested-by: Bryan Li Signed-off-by: Thomas Hellstrom Reviewed-by: Sinclair Yeh --- drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c index 5ae8f921da2a..8a76821177a6 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c @@ -681,6 +681,14 @@ static bool vmw_cmdbuf_try_alloc(struct vmw_cmdbuf_man *man, 0, 0, DRM_MM_SEARCH_DEFAULT, DRM_MM_CREATE_DEFAULT); + if (ret) { + (void) vmw_cmdbuf_man_process(man); + ret = drm_mm_insert_node_generic(&man->mm, info->node, + info->page_size, 0, 0, + DRM_MM_SEARCH_DEFAULT, + DRM_MM_CREATE_DEFAULT); + } + spin_unlock_bh(&man->lock); info->done = !ret; From 031277d4d33d33f0174fbb569ca8f68238175617 Mon Sep 17 00:00:00 2001 From: Chaotian Jing Date: Wed, 30 Sep 2015 17:37:18 +0800 Subject: [PATCH 571/647] mmc: core: fix dead loop of mmc_retune When get a CRC error, start the mmc_retune, it will issue CMD19/CMD21 to do tune, assume there were 10 clock phase need to try, phase 0 to phase 6 is ok, phase 7 to phase 9 is NG, we try it from 0 to 9, so the last CMD19/CMD21 will get CRC error, host->need_retune was set and cause mmc_retune was called, then dead loop of mmc_retune Signed-off-by: Chaotian Jing Acked-by: Adrian Hunter Fixes: bd11e8bd03ca ("mmc: core: Flag re-tuning is needed on CRC errors") Signed-off-by: Ulf Hansson --- drivers/mmc/core/core.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index 0520064dc33b..a3eb20bdcd97 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -134,9 +134,11 @@ void mmc_request_done(struct mmc_host *host, struct mmc_request *mrq) int err = cmd->error; /* Flag re-tuning needed on CRC errors */ - if (err == -EILSEQ || (mrq->sbc && mrq->sbc->error == -EILSEQ) || + if ((cmd->opcode != MMC_SEND_TUNING_BLOCK && + cmd->opcode != MMC_SEND_TUNING_BLOCK_HS200) && + (err == -EILSEQ || (mrq->sbc && mrq->sbc->error == -EILSEQ) || (mrq->data && mrq->data->error == -EILSEQ) || - (mrq->stop && mrq->stop->error == -EILSEQ)) + (mrq->stop && mrq->stop->error == -EILSEQ))) mmc_retune_needed(host); if (err && cmd->retries && mmc_host_is_spi(host)) { From 4ad640e99e5e5514d623210bc937e665ffd8f43f Mon Sep 17 00:00:00 2001 From: Egbert Eich Date: Wed, 23 Sep 2015 16:13:00 +0200 Subject: [PATCH 572/647] drm: Add a non-locking version of drm_kms_helper_poll_enable(), v2 drm_kms_helper_poll_enable() was converted to lock the mode_config mutex in commit 8c4ccc4ab6f64e859d4ff8d7c02c2ed2e956e07f ("drm/probe-helper: Grab mode_config.mutex in poll_init/enable"). This disregarded the cases where this function is called from a context where this mutex is already locked. Add a non-locking version as well. Changes since v1: - use function name suffix '_locked' for the function that is to be called from a locked context. Signed-off-by: Egbert Eich Reviewed-by: Daniel Vetter Signed-off-by: Jani Nikula --- drivers/gpu/drm/drm_probe_helper.c | 19 ++++++++++++++++--- include/drm/drm_crtc_helper.h | 1 + 2 files changed, 17 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c index d734780b31c0..a18164f2f6d2 100644 --- a/drivers/gpu/drm/drm_probe_helper.c +++ b/drivers/gpu/drm/drm_probe_helper.c @@ -94,7 +94,18 @@ static int drm_helper_probe_add_cmdline_mode(struct drm_connector *connector) } #define DRM_OUTPUT_POLL_PERIOD (10*HZ) -static void __drm_kms_helper_poll_enable(struct drm_device *dev) +/** + * drm_kms_helper_poll_enable_locked - re-enable output polling. + * @dev: drm_device + * + * This function re-enables the output polling work without + * locking the mode_config mutex. + * + * This is like drm_kms_helper_poll_enable() however it is to be + * called from a context where the mode_config mutex is locked + * already. + */ +void drm_kms_helper_poll_enable_locked(struct drm_device *dev) { bool poll = false; struct drm_connector *connector; @@ -113,6 +124,8 @@ static void __drm_kms_helper_poll_enable(struct drm_device *dev) if (poll) schedule_delayed_work(&dev->mode_config.output_poll_work, DRM_OUTPUT_POLL_PERIOD); } +EXPORT_SYMBOL(drm_kms_helper_poll_enable_locked); + static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connector *connector, uint32_t maxX, uint32_t maxY, bool merge_type_bits) @@ -174,7 +187,7 @@ static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connect /* Re-enable polling in case the global poll config changed. */ if (drm_kms_helper_poll != dev->mode_config.poll_running) - __drm_kms_helper_poll_enable(dev); + drm_kms_helper_poll_enable_locked(dev); dev->mode_config.poll_running = drm_kms_helper_poll; @@ -428,7 +441,7 @@ EXPORT_SYMBOL(drm_kms_helper_poll_disable); void drm_kms_helper_poll_enable(struct drm_device *dev) { mutex_lock(&dev->mode_config.mutex); - __drm_kms_helper_poll_enable(dev); + drm_kms_helper_poll_enable_locked(dev); mutex_unlock(&dev->mode_config.mutex); } EXPORT_SYMBOL(drm_kms_helper_poll_enable); diff --git a/include/drm/drm_crtc_helper.h b/include/drm/drm_crtc_helper.h index 2a747a91fded..3febb4b9fce9 100644 --- a/include/drm/drm_crtc_helper.h +++ b/include/drm/drm_crtc_helper.h @@ -240,5 +240,6 @@ extern void drm_kms_helper_hotplug_event(struct drm_device *dev); extern void drm_kms_helper_poll_disable(struct drm_device *dev); extern void drm_kms_helper_poll_enable(struct drm_device *dev); +extern void drm_kms_helper_poll_enable_locked(struct drm_device *dev); #endif From b94be972538f4523f09243af7c726158f0bfae33 Mon Sep 17 00:00:00 2001 From: Egbert Eich Date: Wed, 23 Sep 2015 16:13:01 +0200 Subject: [PATCH 573/647] drm/i915: Call non-locking version of drm_kms_helper_poll_enable(), v2 drm_kms_helper_poll_enable() is called from a context in intel_hpd_irq_storm_disable() where the the mode_config mutex is already locked. When this function was converted to lock this mutex in commit 8c4ccc4ab6f6 ("drm/probe-helper: Grab mode_config.mutex in poll_init/enable") a deadlock occurred. Call the newly implemented non-locking version of this function. Changes since v1: - use function name suffix '_locked' for the function that is to be called from a locked context. Signed-off-by: Egbert Eich Reviewed-by: Daniel Vetter Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_hotplug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_hotplug.c b/drivers/gpu/drm/i915/intel_hotplug.c index 53c0173a39fe..b17785719598 100644 --- a/drivers/gpu/drm/i915/intel_hotplug.c +++ b/drivers/gpu/drm/i915/intel_hotplug.c @@ -180,7 +180,7 @@ static void intel_hpd_irq_storm_disable(struct drm_i915_private *dev_priv) /* Enable polling and queue hotplug re-enabling. */ if (hpd_disabled) { - drm_kms_helper_poll_enable(dev); + drm_kms_helper_poll_enable_locked(dev); mod_delayed_work(system_wq, &dev_priv->hotplug.reenable_work, msecs_to_jiffies(HPD_STORM_REENABLE_DELAY)); } From 53960059d56ecef67d4ddd546731623641a3d2d1 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Fri, 27 Mar 2015 08:33:43 +0000 Subject: [PATCH 574/647] MIPS: dma-default: Fix 32-bit fall back to GFP_DMA If there is a DMA zone (usually 24bit = 16MB I believe), but no DMA32 zone, as is the case for some 32-bit kernels, then massage_gfp_flags() will cause DMA memory allocated for devices with a 32..63-bit coherent_dma_mask to fall back to using __GFP_DMA, even though there may only be 32-bits of physical address available anyway. Correct that case to compare against a mask the size of phys_addr_t instead of always using a 64-bit mask. Signed-off-by: James Hogan Fixes: a2e715a86c6d ("MIPS: DMA: Fix computation of DMA flags from device's coherent_dma_mask.") Cc: Ralf Baechle Cc: linux-mips@linux-mips.org Cc: # 2.6.36+ Patchwork: https://patchwork.linux-mips.org/patch/9610/ Signed-off-by: Ralf Baechle --- arch/mips/mm/dma-default.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/mm/dma-default.c b/arch/mips/mm/dma-default.c index a914dc1cb6d1..d8117be729a2 100644 --- a/arch/mips/mm/dma-default.c +++ b/arch/mips/mm/dma-default.c @@ -100,7 +100,7 @@ static gfp_t massage_gfp_flags(const struct device *dev, gfp_t gfp) else #endif #if defined(CONFIG_ZONE_DMA) && !defined(CONFIG_ZONE_DMA32) - if (dev->coherent_dma_mask < DMA_BIT_MASK(64)) + if (dev->coherent_dma_mask < DMA_BIT_MASK(sizeof(phys_addr_t) * 8)) dma_flag = __GFP_DMA; else #endif From 1e5fb282f8eda889776ee83f9214d5df9edaa26d Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Wed, 5 Aug 2015 15:42:36 -0700 Subject: [PATCH 575/647] MIPS: CPS: Stop dangling delay slot from has_mt. The has_mt macro ended with a branch, leaving its callers with a delay slot that would be executed if Config3.MT is not set. However it would not be executed if Config3 (or earlier Config registers) don't exist which makes it somewhat inconsistent at best. Fill the delay slot in the macro & fix the mips_cps_boot_vpes caller appropriately. Signed-off-by: Paul Burton Cc: Markos Chandras Cc: James Hogan Cc: # 3.16+ Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/10865/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/cps-vec.S | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/mips/kernel/cps-vec.S b/arch/mips/kernel/cps-vec.S index 9f71c06aebf6..fa159aab252c 100644 --- a/arch/mips/kernel/cps-vec.S +++ b/arch/mips/kernel/cps-vec.S @@ -39,6 +39,7 @@ mfc0 \dest, CP0_CONFIG, 3 andi \dest, \dest, MIPS_CONF3_MT beqz \dest, \nomt + nop .endm .section .text.cps-vec @@ -226,7 +227,6 @@ LEAF(mips_cps_core_init) #ifdef CONFIG_MIPS_MT /* Check that the core implements the MT ASE */ has_mt t0, 3f - nop .set push .set mips64r2 @@ -310,8 +310,8 @@ LEAF(mips_cps_boot_vpes) PTR_ADDU t0, t0, t1 /* Calculate this VPEs ID. If the core doesn't support MT use 0 */ + li t9, 0 has_mt ta2, 1f - li t9, 0 /* Find the number of VPEs present in the core */ mfc0 t1, CP0_MVPCONF0 From a5b0f6db0e6cf6224e50f6585e9c8f0c2d38a8f8 Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Wed, 5 Aug 2015 15:42:37 -0700 Subject: [PATCH 576/647] MIPS: CPS: Don't include MT code in non-MT kernels. The MT-specific code in mips_cps_boot_vpes can safely be omitted from kernels which don't support MT, with the default VPE==0 case being used as it would be after the has_mt (Config3.MT) check failed at runtime. Discarding the code entirely will save us a few bytes & allow cleaner handling of MT ASE instructions by later patches. Signed-off-by: Paul Burton Cc: Markos Chandras Cc: James Hogan Cc: # 3.16+ Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/10866/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/cps-vec.S | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/mips/kernel/cps-vec.S b/arch/mips/kernel/cps-vec.S index fa159aab252c..57642f56dffb 100644 --- a/arch/mips/kernel/cps-vec.S +++ b/arch/mips/kernel/cps-vec.S @@ -311,6 +311,7 @@ LEAF(mips_cps_boot_vpes) /* Calculate this VPEs ID. If the core doesn't support MT use 0 */ li t9, 0 +#ifdef CONFIG_MIPS_MT has_mt ta2, 1f /* Find the number of VPEs present in the core */ @@ -330,6 +331,7 @@ LEAF(mips_cps_boot_vpes) /* Retrieve the VPE ID from EBase.CPUNum */ mfc0 t9, $15, 1 and t9, t9, t1 +#endif 1: /* Calculate a pointer to this VPEs struct vpe_boot_config */ li t1, VPEBOOTCFG_SIZE From 7a63076d9a31a6c2073da45021eeb4f89d2a8b56 Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Wed, 5 Aug 2015 15:42:38 -0700 Subject: [PATCH 577/647] MIPS: CPS: #ifdef on CONFIG_MIPS_MT_SMP rather than CONFIG_MIPS_MT The CONFIG_MIPS_MT symbol can be selected by CONFIG_MIPS_VPE_LOADER in addition to CONFIG_MIPS_MT_SMP. We only want MT code in the CPS SMP boot vector if we're using MT for SMP. Thus switch the config symbol we ifdef against to CONFIG_MIPS_MT_SMP. Signed-off-by: Paul Burton Cc: Markos Chandras Cc: James Hogan Cc: # 3.16+ Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/10867/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/cps-vec.S | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/mips/kernel/cps-vec.S b/arch/mips/kernel/cps-vec.S index 57642f56dffb..209ded16806b 100644 --- a/arch/mips/kernel/cps-vec.S +++ b/arch/mips/kernel/cps-vec.S @@ -224,7 +224,7 @@ LEAF(excep_ejtag) END(excep_ejtag) LEAF(mips_cps_core_init) -#ifdef CONFIG_MIPS_MT +#ifdef CONFIG_MIPS_MT_SMP /* Check that the core implements the MT ASE */ has_mt t0, 3f @@ -311,7 +311,7 @@ LEAF(mips_cps_boot_vpes) /* Calculate this VPEs ID. If the core doesn't support MT use 0 */ li t9, 0 -#ifdef CONFIG_MIPS_MT +#ifdef CONFIG_MIPS_MT_SMP has_mt ta2, 1f /* Find the number of VPEs present in the core */ @@ -339,7 +339,7 @@ LEAF(mips_cps_boot_vpes) PTR_L ta3, COREBOOTCFG_VPECONFIG(t0) PTR_ADDU v0, v0, ta3 -#ifdef CONFIG_MIPS_MT +#ifdef CONFIG_MIPS_MT_SMP /* If the core doesn't support MT then return */ bnez ta2, 1f @@ -453,7 +453,7 @@ LEAF(mips_cps_boot_vpes) 2: .set pop -#endif /* CONFIG_MIPS_MT */ +#endif /* CONFIG_MIPS_MT_SMP */ /* Return */ jr ra From 2392eec65c493c3f49a1f23f4af713e3c68cf6f5 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 29 Sep 2015 10:35:45 -0400 Subject: [PATCH 578/647] drm/amdgpu/cgs: remove import_gpu_mem It was added for completeness, but we don't have any users for it yet. Daniel noted that it may be racy. Remove it. Change-Id: I5f5546f8911a4f294008a62dc86a73f3face38d1 Reviewed-by: Daniel Vetter Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c | 39 ------------------------- drivers/gpu/drm/amd/include/cgs_linux.h | 17 ----------- 2 files changed, 56 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c index 1c3fc99c5465..8e995148f56e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c @@ -208,44 +208,6 @@ static int amdgpu_cgs_alloc_gpu_mem(void *cgs_device, return ret; } -static int amdgpu_cgs_import_gpu_mem(void *cgs_device, int dmabuf_fd, - cgs_handle_t *handle) -{ - CGS_FUNC_ADEV; - int r; - uint32_t dma_handle; - struct drm_gem_object *obj; - struct amdgpu_bo *bo; - struct drm_device *dev = adev->ddev; - struct drm_file *file_priv = NULL, *priv; - - mutex_lock(&dev->struct_mutex); - list_for_each_entry(priv, &dev->filelist, lhead) { - rcu_read_lock(); - if (priv->pid == get_pid(task_pid(current))) - file_priv = priv; - rcu_read_unlock(); - if (file_priv) - break; - } - mutex_unlock(&dev->struct_mutex); - r = dev->driver->prime_fd_to_handle(dev, - file_priv, dmabuf_fd, - &dma_handle); - spin_lock(&file_priv->table_lock); - - /* Check if we currently have a reference on the object */ - obj = idr_find(&file_priv->object_idr, dma_handle); - if (obj == NULL) { - spin_unlock(&file_priv->table_lock); - return -EINVAL; - } - spin_unlock(&file_priv->table_lock); - bo = gem_to_amdgpu_bo(obj); - *handle = (cgs_handle_t)bo; - return 0; -} - static int amdgpu_cgs_free_gpu_mem(void *cgs_device, cgs_handle_t handle) { struct amdgpu_bo *obj = (struct amdgpu_bo *)handle; @@ -810,7 +772,6 @@ static const struct cgs_ops amdgpu_cgs_ops = { }; static const struct cgs_os_ops amdgpu_cgs_os_ops = { - amdgpu_cgs_import_gpu_mem, amdgpu_cgs_add_irq_source, amdgpu_cgs_irq_get, amdgpu_cgs_irq_put diff --git a/drivers/gpu/drm/amd/include/cgs_linux.h b/drivers/gpu/drm/amd/include/cgs_linux.h index 488642f08267..3b47ae313e36 100644 --- a/drivers/gpu/drm/amd/include/cgs_linux.h +++ b/drivers/gpu/drm/amd/include/cgs_linux.h @@ -26,19 +26,6 @@ #include "cgs_common.h" -/** - * cgs_import_gpu_mem() - Import dmabuf handle - * @cgs_device: opaque device handle - * @dmabuf_fd: DMABuf file descriptor - * @handle: memory handle (output) - * - * Must be called in the process context that dmabuf_fd belongs to. - * - * Return: 0 on success, -errno otherwise - */ -typedef int (*cgs_import_gpu_mem_t)(void *cgs_device, int dmabuf_fd, - cgs_handle_t *handle); - /** * cgs_irq_source_set_func() - Callback for enabling/disabling interrupt sources * @private_data: private data provided to cgs_add_irq_source @@ -114,16 +101,12 @@ typedef int (*cgs_irq_get_t)(void *cgs_device, unsigned src_id, unsigned type); typedef int (*cgs_irq_put_t)(void *cgs_device, unsigned src_id, unsigned type); struct cgs_os_ops { - cgs_import_gpu_mem_t import_gpu_mem; - /* IRQ handling */ cgs_add_irq_source_t add_irq_source; cgs_irq_get_t irq_get; cgs_irq_put_t irq_put; }; -#define cgs_import_gpu_mem(dev,dmabuf_fd,handle) \ - CGS_OS_CALL(import_gpu_mem,dev,dmabuf_fd,handle) #define cgs_add_irq_source(dev,src_id,num_types,set,handler,private_data) \ CGS_OS_CALL(add_irq_source,dev,src_id,num_types,set,handler, \ private_data) From 3e4e3805643445b71b7ee1b84892e43d004e24e2 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Thu, 10 Sep 2015 15:00:39 +0200 Subject: [PATCH 579/647] drm/amdgpu: only print meaningful VM faults MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Port of radeon commit 9b7d786b900baf7c0d1a7e211570aef1cb27590f. Signed-off-by: Christian König Reviewed-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c | 8 ++++++-- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 8 ++++++-- 2 files changed, 12 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index 774528ab8704..fab5471d25d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -1262,6 +1262,12 @@ static int gmc_v7_0_process_interrupt(struct amdgpu_device *adev, addr = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_ADDR); status = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_STATUS); mc_client = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_MCCLIENT); + /* reset addr and status */ + WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1); + + if (!addr && !status) + return 0; + dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n", entry->src_id, entry->src_data); dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", @@ -1269,8 +1275,6 @@ static int gmc_v7_0_process_interrupt(struct amdgpu_device *adev, dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n", status); gmc_v7_0_vm_decode_fault(adev, status, addr, mc_client); - /* reset addr and status */ - WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 9a07742620d0..7bc9e9fcf3d2 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -1262,6 +1262,12 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev, addr = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_ADDR); status = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_STATUS); mc_client = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_MCCLIENT); + /* reset addr and status */ + WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1); + + if (!addr && !status) + return 0; + dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n", entry->src_id, entry->src_data); dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", @@ -1269,8 +1275,6 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev, dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n", status); gmc_v8_0_vm_decode_fault(adev, status, addr, mc_client); - /* reset addr and status */ - WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1); return 0; } From b9a532277938798b53178d5a66af6e2915cb27cf Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 30 Sep 2015 12:48:40 -0400 Subject: [PATCH 580/647] Initialize msg/shm IPC objects before doing ipc_addid() As reported by Dmitry Vyukov, we really shouldn't do ipc_addid() before having initialized the IPC object state. Yes, we initialize the IPC object in a locked state, but with all the lockless RCU lookup work, that IPC object lock no longer means that the state cannot be seen. We already did this for the IPC semaphore code (see commit e8577d1f0329: "ipc/sem.c: fully initialize sem_array before making it visible") but we clearly forgot about msg and shm. Reported-by: Dmitry Vyukov Cc: Manfred Spraul Cc: Davidlohr Bueso Cc: stable@vger.kernel.org Signed-off-by: Linus Torvalds --- ipc/msg.c | 14 +++++++------- ipc/shm.c | 13 +++++++------ ipc/util.c | 8 ++++---- 3 files changed, 18 insertions(+), 17 deletions(-) diff --git a/ipc/msg.c b/ipc/msg.c index 66c4f567eb73..1471db9a7e61 100644 --- a/ipc/msg.c +++ b/ipc/msg.c @@ -137,13 +137,6 @@ static int newque(struct ipc_namespace *ns, struct ipc_params *params) return retval; } - /* ipc_addid() locks msq upon success. */ - id = ipc_addid(&msg_ids(ns), &msq->q_perm, ns->msg_ctlmni); - if (id < 0) { - ipc_rcu_putref(msq, msg_rcu_free); - return id; - } - msq->q_stime = msq->q_rtime = 0; msq->q_ctime = get_seconds(); msq->q_cbytes = msq->q_qnum = 0; @@ -153,6 +146,13 @@ static int newque(struct ipc_namespace *ns, struct ipc_params *params) INIT_LIST_HEAD(&msq->q_receivers); INIT_LIST_HEAD(&msq->q_senders); + /* ipc_addid() locks msq upon success. */ + id = ipc_addid(&msg_ids(ns), &msq->q_perm, ns->msg_ctlmni); + if (id < 0) { + ipc_rcu_putref(msq, msg_rcu_free); + return id; + } + ipc_unlock_object(&msq->q_perm); rcu_read_unlock(); diff --git a/ipc/shm.c b/ipc/shm.c index 222131e8e38f..41787276e141 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -551,12 +551,6 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) if (IS_ERR(file)) goto no_file; - id = ipc_addid(&shm_ids(ns), &shp->shm_perm, ns->shm_ctlmni); - if (id < 0) { - error = id; - goto no_id; - } - shp->shm_cprid = task_tgid_vnr(current); shp->shm_lprid = 0; shp->shm_atim = shp->shm_dtim = 0; @@ -565,6 +559,13 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) shp->shm_nattch = 0; shp->shm_file = file; shp->shm_creator = current; + + id = ipc_addid(&shm_ids(ns), &shp->shm_perm, ns->shm_ctlmni); + if (id < 0) { + error = id; + goto no_id; + } + list_add(&shp->shm_clist, ¤t->sysvshm.shm_clist); /* diff --git a/ipc/util.c b/ipc/util.c index be4230020a1f..0f401d94b7c6 100644 --- a/ipc/util.c +++ b/ipc/util.c @@ -237,6 +237,10 @@ int ipc_addid(struct ipc_ids *ids, struct kern_ipc_perm *new, int size) rcu_read_lock(); spin_lock(&new->lock); + current_euid_egid(&euid, &egid); + new->cuid = new->uid = euid; + new->gid = new->cgid = egid; + id = idr_alloc(&ids->ipcs_idr, new, (next_id < 0) ? 0 : ipcid_to_idx(next_id), 0, GFP_NOWAIT); @@ -249,10 +253,6 @@ int ipc_addid(struct ipc_ids *ids, struct kern_ipc_perm *new, int size) ids->in_use++; - current_euid_egid(&euid, &egid); - new->cuid = new->uid = euid; - new->gid = new->cgid = egid; - if (next_id < 0) { new->seq = ids->seq++; if (ids->seq > IPCID_SEQ_MAX) From eddd3826a1a0190e5235703d1e666affa4d13b96 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 30 Sep 2015 08:38:22 +0000 Subject: [PATCH 581/647] x86/process: Add proper bound checks in 64bit get_wchan() Dmitry Vyukov reported the following using trinity and the memory error detector AddressSanitizer (https://code.google.com/p/address-sanitizer/wiki/AddressSanitizerForKernel). [ 124.575597] ERROR: AddressSanitizer: heap-buffer-overflow on address ffff88002e280000 [ 124.576801] ffff88002e280000 is located 131938492886538 bytes to the left of 28857600-byte region [ffffffff81282e0a, ffffffff82e0830a) [ 124.578633] Accessed by thread T10915: [ 124.579295] inlined in describe_heap_address ./arch/x86/mm/asan/report.c:164 [ 124.579295] #0 ffffffff810dd277 in asan_report_error ./arch/x86/mm/asan/report.c:278 [ 124.580137] #1 ffffffff810dc6a0 in asan_check_region ./arch/x86/mm/asan/asan.c:37 [ 124.581050] #2 ffffffff810dd423 in __tsan_read8 ??:0 [ 124.581893] #3 ffffffff8107c093 in get_wchan ./arch/x86/kernel/process_64.c:444 The address checks in the 64bit implementation of get_wchan() are wrong in several ways: - The lower bound of the stack is not the start of the stack page. It's the start of the stack page plus sizeof (struct thread_info) - The upper bound must be: top_of_stack - TOP_OF_KERNEL_STACK_PADDING - 2 * sizeof(unsigned long). The 2 * sizeof(unsigned long) is required because the stack pointer points at the frame pointer. The layout on the stack is: ... IP FP ... IP FP. So we need to make sure that both IP and FP are in the bounds. Fix the bound checks and get rid of the mix of numeric constants, u64 and unsigned long. Making all unsigned long allows us to use the same function for 32bit as well. Use READ_ONCE() when accessing the stack. This does not prevent a concurrent wakeup of the task and the stack changing, but at least it avoids TOCTOU. Also check task state at the end of the loop. Again that does not prevent concurrent changes, but it avoids walking for nothing. Add proper comments while at it. Reported-by: Dmitry Vyukov Reported-by: Sasha Levin Based-on-patch-from: Wolfram Gloger Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Dmitry Vyukov Cc: Andrey Ryabinin Cc: Andy Lutomirski Cc: Andrey Konovalov Cc: Kostya Serebryany Cc: Alexander Potapenko Cc: kasan-dev Cc: Denys Vlasenko Cc: Andi Kleen Cc: Wolfram Gloger Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/20150930083302.694788319@linutronix.de Signed-off-by: Thomas Gleixner --- arch/x86/kernel/process_64.c | 52 +++++++++++++++++++++++++++++------- 1 file changed, 42 insertions(+), 10 deletions(-) diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 3c1bbcf12924..f1fd0889e8af 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -499,27 +499,59 @@ void set_personality_ia32(bool x32) } EXPORT_SYMBOL_GPL(set_personality_ia32); +/* + * Called from fs/proc with a reference on @p to find the function + * which called into schedule(). This needs to be done carefully + * because the task might wake up and we might look at a stack + * changing under us. + */ unsigned long get_wchan(struct task_struct *p) { - unsigned long stack; - u64 fp, ip; + unsigned long start, bottom, top, sp, fp, ip; int count = 0; if (!p || p == current || p->state == TASK_RUNNING) return 0; - stack = (unsigned long)task_stack_page(p); - if (p->thread.sp < stack || p->thread.sp >= stack+THREAD_SIZE) + + start = (unsigned long)task_stack_page(p); + if (!start) return 0; - fp = *(u64 *)(p->thread.sp); + + /* + * Layout of the stack page: + * + * ----------- topmax = start + THREAD_SIZE - sizeof(unsigned long) + * PADDING + * ----------- top = topmax - TOP_OF_KERNEL_STACK_PADDING + * stack + * ----------- bottom = start + sizeof(thread_info) + * thread_info + * ----------- start + * + * The tasks stack pointer points at the location where the + * framepointer is stored. The data on the stack is: + * ... IP FP ... IP FP + * + * We need to read FP and IP, so we need to adjust the upper + * bound by another unsigned long. + */ + top = start + THREAD_SIZE - TOP_OF_KERNEL_STACK_PADDING; + top -= 2 * sizeof(unsigned long); + bottom = start + sizeof(struct thread_info); + + sp = READ_ONCE(p->thread.sp); + if (sp < bottom || sp > top) + return 0; + + fp = READ_ONCE(*(unsigned long *)sp); do { - if (fp < (unsigned long)stack || - fp >= (unsigned long)stack+THREAD_SIZE) + if (fp < bottom || fp > top) return 0; - ip = *(u64 *)(fp+8); + ip = READ_ONCE(*(unsigned long *)(fp + sizeof(unsigned long))); if (!in_sched_functions(ip)) return ip; - fp = *(u64 *)fp; - } while (count++ < 16); + fp = READ_ONCE(*(unsigned long *)fp); + } while (count++ < 16 && p->state != TASK_RUNNING); return 0; } From 7ba78053aacb89998a052843e3c56983c31d57f0 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 30 Sep 2015 08:38:23 +0000 Subject: [PATCH 582/647] x86/process: Unify 32bit and 64bit implementations of get_wchan() The stack layout and the functionality is identical. Use the 64bit version for all of x86. Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Dmitry Vyukov Cc: Andrey Ryabinin Cc: Andy Lutomirski Cc: Andrey Konovalov Cc: Kostya Serebryany Cc: Alexander Potapenko Cc: kasan-dev Cc: Denys Vlasenko Cc: Andi Kleen Cc: Sasha Levin Cc: Wolfram Gloger Link: http://lkml.kernel.org/r/20150930083302.779694618@linutronix.de Signed-off-by: Thomas Gleixner --- arch/x86/kernel/process.c | 55 +++++++++++++++++++++++++++++++++++ arch/x86/kernel/process_32.c | 28 ------------------ arch/x86/kernel/process_64.c | 56 ------------------------------------ 3 files changed, 55 insertions(+), 84 deletions(-) diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 6d0e62ae8516..39e585a554b7 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -506,3 +506,58 @@ unsigned long arch_randomize_brk(struct mm_struct *mm) return randomize_range(mm->brk, range_end, 0) ? : mm->brk; } +/* + * Called from fs/proc with a reference on @p to find the function + * which called into schedule(). This needs to be done carefully + * because the task might wake up and we might look at a stack + * changing under us. + */ +unsigned long get_wchan(struct task_struct *p) +{ + unsigned long start, bottom, top, sp, fp, ip; + int count = 0; + + if (!p || p == current || p->state == TASK_RUNNING) + return 0; + + start = (unsigned long)task_stack_page(p); + if (!start) + return 0; + + /* + * Layout of the stack page: + * + * ----------- topmax = start + THREAD_SIZE - sizeof(unsigned long) + * PADDING + * ----------- top = topmax - TOP_OF_KERNEL_STACK_PADDING + * stack + * ----------- bottom = start + sizeof(thread_info) + * thread_info + * ----------- start + * + * The tasks stack pointer points at the location where the + * framepointer is stored. The data on the stack is: + * ... IP FP ... IP FP + * + * We need to read FP and IP, so we need to adjust the upper + * bound by another unsigned long. + */ + top = start + THREAD_SIZE - TOP_OF_KERNEL_STACK_PADDING; + top -= 2 * sizeof(unsigned long); + bottom = start + sizeof(struct thread_info); + + sp = READ_ONCE(p->thread.sp); + if (sp < bottom || sp > top) + return 0; + + fp = READ_ONCE(*(unsigned long *)sp); + do { + if (fp < bottom || fp > top) + return 0; + ip = READ_ONCE(*(unsigned long *)(fp + sizeof(unsigned long))); + if (!in_sched_functions(ip)) + return ip; + fp = READ_ONCE(*(unsigned long *)fp); + } while (count++ < 16 && p->state != TASK_RUNNING); + return 0; +} diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index c13df2c735f8..737527b40e5b 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -324,31 +324,3 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) return prev_p; } - -#define top_esp (THREAD_SIZE - sizeof(unsigned long)) -#define top_ebp (THREAD_SIZE - 2*sizeof(unsigned long)) - -unsigned long get_wchan(struct task_struct *p) -{ - unsigned long bp, sp, ip; - unsigned long stack_page; - int count = 0; - if (!p || p == current || p->state == TASK_RUNNING) - return 0; - stack_page = (unsigned long)task_stack_page(p); - sp = p->thread.sp; - if (!stack_page || sp < stack_page || sp > top_esp+stack_page) - return 0; - /* include/asm-i386/system.h:switch_to() pushes bp last. */ - bp = *(unsigned long *) sp; - do { - if (bp < stack_page || bp > top_ebp+stack_page) - return 0; - ip = *(unsigned long *) (bp+4); - if (!in_sched_functions(ip)) - return ip; - bp = *(unsigned long *) bp; - } while (count++ < 16); - return 0; -} - diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index f1fd0889e8af..b35921a670b2 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -499,62 +499,6 @@ void set_personality_ia32(bool x32) } EXPORT_SYMBOL_GPL(set_personality_ia32); -/* - * Called from fs/proc with a reference on @p to find the function - * which called into schedule(). This needs to be done carefully - * because the task might wake up and we might look at a stack - * changing under us. - */ -unsigned long get_wchan(struct task_struct *p) -{ - unsigned long start, bottom, top, sp, fp, ip; - int count = 0; - - if (!p || p == current || p->state == TASK_RUNNING) - return 0; - - start = (unsigned long)task_stack_page(p); - if (!start) - return 0; - - /* - * Layout of the stack page: - * - * ----------- topmax = start + THREAD_SIZE - sizeof(unsigned long) - * PADDING - * ----------- top = topmax - TOP_OF_KERNEL_STACK_PADDING - * stack - * ----------- bottom = start + sizeof(thread_info) - * thread_info - * ----------- start - * - * The tasks stack pointer points at the location where the - * framepointer is stored. The data on the stack is: - * ... IP FP ... IP FP - * - * We need to read FP and IP, so we need to adjust the upper - * bound by another unsigned long. - */ - top = start + THREAD_SIZE - TOP_OF_KERNEL_STACK_PADDING; - top -= 2 * sizeof(unsigned long); - bottom = start + sizeof(struct thread_info); - - sp = READ_ONCE(p->thread.sp); - if (sp < bottom || sp > top) - return 0; - - fp = READ_ONCE(*(unsigned long *)sp); - do { - if (fp < bottom || fp > top) - return 0; - ip = READ_ONCE(*(unsigned long *)(fp + sizeof(unsigned long))); - if (!in_sched_functions(ip)) - return ip; - fp = READ_ONCE(*(unsigned long *)fp); - } while (count++ < 16 && p->state != TASK_RUNNING); - return 0; -} - long do_arch_prctl(struct task_struct *task, int code, unsigned long addr) { int ret = 0; From a1cf09031e641d3cceaca4a4dd20ef6a785bc9b3 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Tue, 15 Sep 2015 15:36:00 +0200 Subject: [PATCH 583/647] dmaengine: at_xdmac: change block increment addressing mode The addressing mode we were using was not only incrementing the address at each microblock, but also at each data boundary, which was severely slowing the transfer, without any benefit since we were not using the data stride. Switch to the micro block increment only in order to get back to an acceptable performance level. Signed-off-by: Maxime Ripard Signed-off-by: Ludovic Desroches Fixes: 6007ccb57744 ("dmaengine: xdmac: Add interleaved transfer support") Cc: stable@vger.kernel.org #4.2 Signed-off-by: Vinod Koul --- drivers/dma/at_xdmac.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c index a165b4bfd330..ffea6027b631 100644 --- a/drivers/dma/at_xdmac.c +++ b/drivers/dma/at_xdmac.c @@ -875,14 +875,14 @@ at_xdmac_interleaved_queue_desc(struct dma_chan *chan, if (xt->src_inc) { if (xt->src_sgl) - chan_cc |= AT_XDMAC_CC_SAM_UBS_DS_AM; + chan_cc |= AT_XDMAC_CC_SAM_UBS_AM; else chan_cc |= AT_XDMAC_CC_SAM_INCREMENTED_AM; } if (xt->dst_inc) { if (xt->dst_sgl) - chan_cc |= AT_XDMAC_CC_DAM_UBS_DS_AM; + chan_cc |= AT_XDMAC_CC_DAM_UBS_AM; else chan_cc |= AT_XDMAC_CC_DAM_INCREMENTED_AM; } From 0be2136b67067617b36c70e525d7534108361e36 Mon Sep 17 00:00:00 2001 From: Ludovic Desroches Date: Tue, 15 Sep 2015 15:39:11 +0200 Subject: [PATCH 584/647] dmaengine: at_xdmac: clean used descriptor When putting back a descriptor to the free descs list, some fields are not set to 0, it can cause bugs if someone uses it without having this in mind. Descriptor are not put back one by one so it is easier to clean descriptors when we request them. Signed-off-by: Ludovic Desroches Cc: stable@vger.kernel.org #4.2 Signed-off-by: Vinod Koul --- drivers/dma/at_xdmac.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c index ffea6027b631..dd24375b76dd 100644 --- a/drivers/dma/at_xdmac.c +++ b/drivers/dma/at_xdmac.c @@ -455,6 +455,15 @@ static struct at_xdmac_desc *at_xdmac_alloc_desc(struct dma_chan *chan, return desc; } +void at_xdmac_init_used_desc(struct at_xdmac_desc *desc) +{ + memset(&desc->lld, 0, sizeof(desc->lld)); + INIT_LIST_HEAD(&desc->descs_list); + desc->direction = DMA_TRANS_NONE; + desc->xfer_size = 0; + desc->active_xfer = false; +} + /* Call must be protected by lock. */ static struct at_xdmac_desc *at_xdmac_get_desc(struct at_xdmac_chan *atchan) { @@ -466,7 +475,7 @@ static struct at_xdmac_desc *at_xdmac_get_desc(struct at_xdmac_chan *atchan) desc = list_first_entry(&atchan->free_descs_list, struct at_xdmac_desc, desc_node); list_del(&desc->desc_node); - desc->active_xfer = false; + at_xdmac_init_used_desc(desc); } return desc; From aa3ee5f569fda51e54c224c0df60e187e9c5e582 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Sat, 26 Sep 2015 17:15:47 +0800 Subject: [PATCH 585/647] dmaengine: zxdma: Fix off-by-one for testing valid pchan request The valid pchan range is 0 ~ d->dma_requests - 1. Signed-off-by: Axel Lin Reviewed-by: Jun Nie Signed-off-by: Vinod Koul --- drivers/dma/zx296702_dma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/dma/zx296702_dma.c b/drivers/dma/zx296702_dma.c index 39915a6b7986..c017fcd8e07c 100644 --- a/drivers/dma/zx296702_dma.c +++ b/drivers/dma/zx296702_dma.c @@ -739,7 +739,7 @@ static struct dma_chan *zx_of_dma_simple_xlate(struct of_phandle_args *dma_spec, struct dma_chan *chan; struct zx_dma_chan *c; - if (request > d->dma_requests) + if (request >= d->dma_requests) return NULL; chan = dma_get_any_slave_channel(&d->slave); From e87ffbdf06971a80ad2a11217200bdd936195af1 Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Wed, 30 Sep 2015 19:42:14 +0200 Subject: [PATCH 586/647] dmaengine: pxa_dma: fix the no-requestor case A very small number of devices don't use the flow control offered by requestor lines. In these specific cases, the pxa dma driver should be aware of that and not try to use a requestor line. Signed-off-by: Robert Jarzmik Signed-off-by: Vinod Koul --- drivers/dma/pxa_dma.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) diff --git a/drivers/dma/pxa_dma.c b/drivers/dma/pxa_dma.c index c6723ecd5848..064a8266a3b7 100644 --- a/drivers/dma/pxa_dma.c +++ b/drivers/dma/pxa_dma.c @@ -473,8 +473,10 @@ static void pxad_free_phy(struct pxad_chan *chan) return; /* clear the channel mapping in DRCMR */ - reg = pxad_drcmr(chan->drcmr); - writel_relaxed(0, chan->phy->base + reg); + if (chan->drcmr <= DRCMR_CHLNUM) { + reg = pxad_drcmr(chan->drcmr); + writel_relaxed(0, chan->phy->base + reg); + } spin_lock_irqsave(&pdev->phy_lock, flags); for (i = 0; i < 32; i++) @@ -516,8 +518,10 @@ static void phy_enable(struct pxad_phy *phy, bool misaligned) "%s(); phy=%p(%d) misaligned=%d\n", __func__, phy, phy->idx, misaligned); - reg = pxad_drcmr(phy->vchan->drcmr); - writel_relaxed(DRCMR_MAPVLD | phy->idx, phy->base + reg); + if (phy->vchan->drcmr <= DRCMR_CHLNUM) { + reg = pxad_drcmr(phy->vchan->drcmr); + writel_relaxed(DRCMR_MAPVLD | phy->idx, phy->base + reg); + } dalgn = phy_readl_relaxed(phy, DALGN); if (misaligned) @@ -911,14 +915,18 @@ static void pxad_get_config(struct pxad_chan *chan, width = chan->cfg.src_addr_width; dev_addr = chan->cfg.src_addr; *dev_src = dev_addr; - *dcmd |= PXA_DCMD_INCTRGADDR | PXA_DCMD_FLOWSRC; + *dcmd |= PXA_DCMD_INCTRGADDR; + if (chan->drcmr <= DRCMR_CHLNUM) + *dcmd |= PXA_DCMD_FLOWSRC; } if (dir == DMA_MEM_TO_DEV) { maxburst = chan->cfg.dst_maxburst; width = chan->cfg.dst_addr_width; dev_addr = chan->cfg.dst_addr; *dev_dst = dev_addr; - *dcmd |= PXA_DCMD_INCSRCADDR | PXA_DCMD_FLOWTRG; + *dcmd |= PXA_DCMD_INCSRCADDR; + if (chan->drcmr <= DRCMR_CHLNUM) + *dcmd |= PXA_DCMD_FLOWTRG; } if (dir == DMA_MEM_TO_MEM) *dcmd |= PXA_DCMD_BURST32 | PXA_DCMD_INCTRGADDR | From 7b09a1bba4091a9d208481d7831682a1f3061ab9 Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Wed, 30 Sep 2015 19:42:15 +0200 Subject: [PATCH 587/647] dmaengine: pxa_dma: fix residue corner case A very tiny temporal window exists in the residue calculation where : - upon entering residue calculation, the transfer is ongoing - when reading the current transfer pointer, it just changed to the "finisher/linker" descriptor In this case, the residue returned is the whole transfer length instead of 0. Fix it. This appears almost in one extreme case, where the driver is used by older clients which inquire for residue in interrupt context, such as the smsc91x ethernet driver, in a tight loop : interrupt_handler() dmaengine_submit() do { dmaengine_tx_status() } while (residue > 0 || status != DMA_ERROR) Signed-off-by: Robert Jarzmik Signed-off-by: Vinod Koul --- drivers/dma/pxa_dma.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/dma/pxa_dma.c b/drivers/dma/pxa_dma.c index 064a8266a3b7..fc4156afa070 100644 --- a/drivers/dma/pxa_dma.c +++ b/drivers/dma/pxa_dma.c @@ -1186,6 +1186,16 @@ static unsigned int pxad_residue(struct pxad_chan *chan, else curr = phy_readl_relaxed(chan->phy, DTADR); + /* + * curr has to be actually read before checking descriptor + * completion, so that a curr inside a status updater + * descriptor implies the following test returns true, and + * preventing reordering of curr load and the test. + */ + rmb(); + if (is_desc_completed(vd)) + goto out; + for (i = 0; i < sw_desc->nb_desc - 1; i++) { hw_desc = sw_desc->hw_desc[i]; if (sw_desc->hw_desc[0]->dcmd & PXA_DCMD_INCSRCADDR) From 8d0d94015e96b8853c4f7f06eac3f269e1b3d866 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Fabiano=20Fid=C3=AAncio?= Date: Thu, 24 Sep 2015 15:18:34 +0200 Subject: [PATCH 588/647] drm/qxl: recreate the primary surface when the bo is not primary MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When disabling/enabling a crtc the primary area must be updated independently of which crtc has been disabled/enabled. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1264735 Signed-off-by: Fabiano Fidêncio Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/qxl/qxl_display.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c index dd845f82cc24..4649bd2ed340 100644 --- a/drivers/gpu/drm/qxl/qxl_display.c +++ b/drivers/gpu/drm/qxl/qxl_display.c @@ -618,7 +618,7 @@ static int qxl_crtc_mode_set(struct drm_crtc *crtc, adjusted_mode->hdisplay, adjusted_mode->vdisplay); - if (qcrtc->index == 0) + if (bo->is_primary == false) recreate_primary = true; if (bo->surf.stride * bo->surf.height > qdev->vram_size) { From 148ba09c465cc54d8e68f041bf9a30332b315c39 Mon Sep 17 00:00:00 2001 From: Inki Dae Date: Wed, 30 Sep 2015 20:12:29 +0900 Subject: [PATCH 589/647] drm/exynos: dp: remove suspend/resume functions This patch removes unnecessary pm suspend/resume functions. All kms sub drivers will be controlled by top of Exynos drm driver and connector dpms so these sub drivers shouldn't have their own pm interfaces. Signed-off-by: Inki Dae Reviewed-by: Gustavo Padovan --- drivers/gpu/drm/exynos/exynos_dp_core.c | 23 ----------------------- 1 file changed, 23 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.c b/drivers/gpu/drm/exynos/exynos_dp_core.c index d66ade0efac8..124fb9a56f02 100644 --- a/drivers/gpu/drm/exynos/exynos_dp_core.c +++ b/drivers/gpu/drm/exynos/exynos_dp_core.c @@ -1383,28 +1383,6 @@ static int exynos_dp_remove(struct platform_device *pdev) return 0; } -#ifdef CONFIG_PM_SLEEP -static int exynos_dp_suspend(struct device *dev) -{ - struct exynos_dp_device *dp = dev_get_drvdata(dev); - - exynos_dp_disable(&dp->encoder); - return 0; -} - -static int exynos_dp_resume(struct device *dev) -{ - struct exynos_dp_device *dp = dev_get_drvdata(dev); - - exynos_dp_enable(&dp->encoder); - return 0; -} -#endif - -static const struct dev_pm_ops exynos_dp_pm_ops = { - SET_SYSTEM_SLEEP_PM_OPS(exynos_dp_suspend, exynos_dp_resume) -}; - static const struct of_device_id exynos_dp_match[] = { { .compatible = "samsung,exynos5-dp" }, {}, @@ -1417,7 +1395,6 @@ struct platform_driver dp_driver = { .driver = { .name = "exynos-dp", .owner = THIS_MODULE, - .pm = &exynos_dp_pm_ops, .of_match_table = exynos_dp_match, }, }; From 3c79fb8c9424a24bf812b9a8cb4f19b781052b0b Mon Sep 17 00:00:00 2001 From: Gustavo Padovan Date: Wed, 30 Sep 2015 18:40:54 -0300 Subject: [PATCH 590/647] drm/exynos: fimd: actually disable dp clock MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit fimd_dp_clock_enable() was setting the always to enabled, this patch fix this to actually use the value that is set to 'val'. Reported-by: Emilio López Signed-off-by: Gustavo Padovan Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_fimd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index 03ad549e7698..3d1aba67758b 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -871,7 +871,7 @@ static void fimd_dp_clock_enable(struct exynos_drm_crtc *crtc, bool enable) return; val = enable ? DP_MIE_CLK_DP_ENABLE : DP_MIE_CLK_DISABLE; - writel(DP_MIE_CLK_DP_ENABLE, ctx->regs + DP_MIE_CLKCON); + writel(val, ctx->regs + DP_MIE_CLKCON); } static const struct exynos_drm_crtc_ops fimd_crtc_ops = { From cda374253f862bd0f43edda6935a48294fe8ed3e Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Thu, 1 Oct 2015 14:21:40 +0900 Subject: [PATCH 591/647] drm/exynos: Staticize local function in exynos_drm_gem.c The exynos_drm_gem_mmap_buffer() is not used outside so make it static. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Inki Dae --- drivers/gpu/drm/exynos/exynos_drm_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index 048ad13e5d82..407afedb6003 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -307,7 +307,7 @@ void exynos_drm_gem_put_dma_addr(struct drm_device *dev, drm_gem_object_unreference_unlocked(obj); } -int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem_obj *exynos_gem_obj, +static int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem_obj *exynos_gem_obj, struct vm_area_struct *vma) { struct drm_device *drm_dev = exynos_gem_obj->base.dev; From 5b235dc2647e4977b17b5c41d959d0f455831c3f Mon Sep 17 00:00:00 2001 From: Alban Bedel Date: Fri, 4 Sep 2015 14:29:16 +0200 Subject: [PATCH 592/647] MIPS: Fix the build on jz4740 after removing the custom gpio.h Somehow the wrong version of the patch to remove the use of custom gpio.h on mips has been merged. This patch add the missing fixes for a build error on jz4740 because linux/gpio.h doesn't provide any machine specfics definitions anymore. Signed-off-by: Alban Bedel Cc: Paul Burton Cc: Lars-Peter Clausen Cc: Brian Norris Cc: Thomas Gleixner Cc: Linus Walleij Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/11089/ Signed-off-by: Ralf Baechle --- arch/mips/jz4740/board-qi_lb60.c | 1 + arch/mips/jz4740/gpio.c | 1 + 2 files changed, 2 insertions(+) diff --git a/arch/mips/jz4740/board-qi_lb60.c b/arch/mips/jz4740/board-qi_lb60.c index 4e62bf85d0b0..459cb017306c 100644 --- a/arch/mips/jz4740/board-qi_lb60.c +++ b/arch/mips/jz4740/board-qi_lb60.c @@ -26,6 +26,7 @@ #include #include +#include #include #include #include diff --git a/arch/mips/jz4740/gpio.c b/arch/mips/jz4740/gpio.c index a74e181058b0..8c6d76c9b2d6 100644 --- a/arch/mips/jz4740/gpio.c +++ b/arch/mips/jz4740/gpio.c @@ -28,6 +28,7 @@ #include #include +#include #define JZ4740_GPIO_BASE_A (32*0) #define JZ4740_GPIO_BASE_B (32*1) From f4829a9b7a61e159367350008a608b062c4f6840 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sun, 27 Sep 2015 21:01:50 +0200 Subject: [PATCH 593/647] blk-mq: fix racy updates of rq->errors blk_mq_complete_request may be a no-op if the request has already been completed by others means (e.g. a timeout or cancellation), but currently drivers have to set rq->errors before calling blk_mq_complete_request, which might leave us with the wrong error value. Add an error parameter to blk_mq_complete_request so that we can defer setting rq->errors until we known we won the race to complete the request. Signed-off-by: Christoph Hellwig Reviewed-by: Sagi Grimberg Signed-off-by: Jens Axboe --- block/blk-mq.c | 12 ++++++------ drivers/block/loop.c | 11 +++++------ drivers/block/null_blk.c | 2 +- drivers/block/nvme-core.c | 16 +++++++--------- drivers/block/virtio_blk.c | 2 +- drivers/block/xen-blkfront.c | 19 ++++++++++--------- drivers/scsi/scsi_lib.c | 2 +- include/linux/blk-mq.h | 2 +- 8 files changed, 32 insertions(+), 34 deletions(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index 31c0c6259c4c..2306330530e8 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -393,14 +393,16 @@ void __blk_mq_complete_request(struct request *rq) * Ends all I/O on a request. It does not handle partial completions. * The actual completion happens out-of-order, through a IPI handler. **/ -void blk_mq_complete_request(struct request *rq) +void blk_mq_complete_request(struct request *rq, int error) { struct request_queue *q = rq->q; if (unlikely(blk_should_fake_timeout(q))) return; - if (!blk_mark_rq_complete(rq)) + if (!blk_mark_rq_complete(rq)) { + rq->errors = error; __blk_mq_complete_request(rq); + } } EXPORT_SYMBOL(blk_mq_complete_request); @@ -616,10 +618,8 @@ static void blk_mq_check_expired(struct blk_mq_hw_ctx *hctx, * If a request wasn't started before the queue was * marked dying, kill it here or it'll go unnoticed. */ - if (unlikely(blk_queue_dying(rq->q))) { - rq->errors = -EIO; - blk_mq_complete_request(rq); - } + if (unlikely(blk_queue_dying(rq->q))) + blk_mq_complete_request(rq, -EIO); return; } if (rq->cmd_flags & REQ_NO_TIMEOUT) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index f9889b6bc02c..674f800a3b57 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1486,17 +1486,16 @@ static void loop_handle_cmd(struct loop_cmd *cmd) { const bool write = cmd->rq->cmd_flags & REQ_WRITE; struct loop_device *lo = cmd->rq->q->queuedata; - int ret = -EIO; + int ret = 0; - if (write && (lo->lo_flags & LO_FLAGS_READ_ONLY)) + if (write && (lo->lo_flags & LO_FLAGS_READ_ONLY)) { + ret = -EIO; goto failed; + } ret = do_req_filebacked(lo, cmd->rq); - failed: - if (ret) - cmd->rq->errors = -EIO; - blk_mq_complete_request(cmd->rq); + blk_mq_complete_request(cmd->rq, ret ? -EIO : 0); } static void loop_queue_write_work(struct work_struct *work) diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c index a295b98c6bae..1c9e4fe5aa44 100644 --- a/drivers/block/null_blk.c +++ b/drivers/block/null_blk.c @@ -289,7 +289,7 @@ static inline void null_handle_cmd(struct nullb_cmd *cmd) case NULL_IRQ_SOFTIRQ: switch (queue_mode) { case NULL_Q_MQ: - blk_mq_complete_request(cmd->rq); + blk_mq_complete_request(cmd->rq, cmd->rq->errors); break; case NULL_Q_RQ: blk_complete_request(cmd->rq); diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 30758bdf69ea..6f04771f1019 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -618,16 +618,15 @@ static void req_completion(struct nvme_queue *nvmeq, void *ctx, spin_unlock_irqrestore(req->q->queue_lock, flags); return; } + if (req->cmd_type == REQ_TYPE_DRV_PRIV) { if (cmd_rq->ctx == CMD_CTX_CANCELLED) - req->errors = -EINTR; - else - req->errors = status; + status = -EINTR; } else { - req->errors = nvme_error_status(status); + status = nvme_error_status(status); } - } else - req->errors = 0; + } + if (req->cmd_type == REQ_TYPE_DRV_PRIV) { u32 result = le32_to_cpup(&cqe->result); req->special = (void *)(uintptr_t)result; @@ -650,7 +649,7 @@ static void req_completion(struct nvme_queue *nvmeq, void *ctx, } nvme_free_iod(nvmeq->dev, iod); - blk_mq_complete_request(req); + blk_mq_complete_request(req, status); } /* length is in bytes. gfp flags indicates whether we may sleep. */ @@ -863,8 +862,7 @@ static int nvme_queue_rq(struct blk_mq_hw_ctx *hctx, if (ns && ns->ms && !blk_integrity_rq(req)) { if (!(ns->pi_type && ns->ms == 8) && req->cmd_type != REQ_TYPE_DRV_PRIV) { - req->errors = -EFAULT; - blk_mq_complete_request(req); + blk_mq_complete_request(req, -EFAULT); return BLK_MQ_RQ_QUEUE_OK; } } diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index e93899cc6f60..6ca35495a5be 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -144,7 +144,7 @@ static void virtblk_done(struct virtqueue *vq) do { virtqueue_disable_cb(vq); while ((vbr = virtqueue_get_buf(vblk->vqs[qid].vq, &len)) != NULL) { - blk_mq_complete_request(vbr->req); + blk_mq_complete_request(vbr->req, vbr->req->errors); req_done = true; } if (unlikely(virtqueue_is_broken(vq))) diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 0823a96902f8..611170896b8c 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -1142,6 +1142,7 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) RING_IDX i, rp; unsigned long flags; struct blkfront_info *info = (struct blkfront_info *)dev_id; + int error; spin_lock_irqsave(&info->io_lock, flags); @@ -1182,37 +1183,37 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) continue; } - req->errors = (bret->status == BLKIF_RSP_OKAY) ? 0 : -EIO; + error = (bret->status == BLKIF_RSP_OKAY) ? 0 : -EIO; switch (bret->operation) { case BLKIF_OP_DISCARD: if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) { struct request_queue *rq = info->rq; printk(KERN_WARNING "blkfront: %s: %s op failed\n", info->gd->disk_name, op_name(bret->operation)); - req->errors = -EOPNOTSUPP; + error = -EOPNOTSUPP; info->feature_discard = 0; info->feature_secdiscard = 0; queue_flag_clear(QUEUE_FLAG_DISCARD, rq); queue_flag_clear(QUEUE_FLAG_SECDISCARD, rq); } - blk_mq_complete_request(req); + blk_mq_complete_request(req, error); break; case BLKIF_OP_FLUSH_DISKCACHE: case BLKIF_OP_WRITE_BARRIER: if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) { printk(KERN_WARNING "blkfront: %s: %s op failed\n", info->gd->disk_name, op_name(bret->operation)); - req->errors = -EOPNOTSUPP; + error = -EOPNOTSUPP; } if (unlikely(bret->status == BLKIF_RSP_ERROR && info->shadow[id].req.u.rw.nr_segments == 0)) { printk(KERN_WARNING "blkfront: %s: empty %s op failed\n", info->gd->disk_name, op_name(bret->operation)); - req->errors = -EOPNOTSUPP; + error = -EOPNOTSUPP; } - if (unlikely(req->errors)) { - if (req->errors == -EOPNOTSUPP) - req->errors = 0; + if (unlikely(error)) { + if (error == -EOPNOTSUPP) + error = 0; info->feature_flush = 0; xlvbd_flush(info); } @@ -1223,7 +1224,7 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) dev_dbg(&info->xbdev->dev, "Bad return from blkdev data " "request: %x\n", bret->status); - blk_mq_complete_request(req); + blk_mq_complete_request(req, error); break; default: BUG(); diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index cbfc5990052b..126a48c6431e 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1957,7 +1957,7 @@ static int scsi_mq_prep_fn(struct request *req) static void scsi_mq_done(struct scsi_cmnd *cmd) { trace_scsi_dispatch_cmd_done(cmd); - blk_mq_complete_request(cmd->request); + blk_mq_complete_request(cmd->request, cmd->request->errors); } static int scsi_queue_rq(struct blk_mq_hw_ctx *hctx, diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index b80ba4572a31..c1b5c867ff07 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -214,7 +214,7 @@ void blk_mq_add_to_requeue_list(struct request *rq, bool at_head); void blk_mq_cancel_requeue_work(struct request_queue *q); void blk_mq_kick_requeue_list(struct request_queue *q); void blk_mq_abort_requeue_list(struct request_queue *q); -void blk_mq_complete_request(struct request *rq); +void blk_mq_complete_request(struct request *rq, int error); void blk_mq_stop_hw_queue(struct blk_mq_hw_ctx *hctx); void blk_mq_start_hw_queue(struct blk_mq_hw_ctx *hctx); From 0bf6cd5b9531bcc29c0a5e504b6ce2984c6fd8d8 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sun, 27 Sep 2015 21:01:51 +0200 Subject: [PATCH 594/647] blk-mq: factor out a helper to iterate all tags for a request_queue And replace the blk_mq_tag_busy_iter with it - the driver use has been replaced with a new helper a while ago, and internal to the block we only need the new version. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq-tag.c | 27 ++++++++++++++++++++------- block/blk-mq-tag.h | 2 ++ block/blk-mq.c | 14 +++----------- include/linux/blk-mq.h | 2 -- 4 files changed, 25 insertions(+), 20 deletions(-) diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index 9115c6d59948..ed96474d75cb 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -471,17 +471,30 @@ void blk_mq_all_tag_busy_iter(struct blk_mq_tags *tags, busy_tag_iter_fn *fn, } EXPORT_SYMBOL(blk_mq_all_tag_busy_iter); -void blk_mq_tag_busy_iter(struct blk_mq_hw_ctx *hctx, busy_iter_fn *fn, +void blk_mq_queue_tag_busy_iter(struct request_queue *q, busy_iter_fn *fn, void *priv) { - struct blk_mq_tags *tags = hctx->tags; + struct blk_mq_hw_ctx *hctx; + int i; + + + queue_for_each_hw_ctx(q, hctx, i) { + struct blk_mq_tags *tags = hctx->tags; + + /* + * If not software queues are currently mapped to this + * hardware queue, there's nothing to check + */ + if (!blk_mq_hw_queue_mapped(hctx)) + continue; + + if (tags->nr_reserved_tags) + bt_for_each(hctx, &tags->breserved_tags, 0, fn, priv, true); + bt_for_each(hctx, &tags->bitmap_tags, tags->nr_reserved_tags, fn, priv, + false); + } - if (tags->nr_reserved_tags) - bt_for_each(hctx, &tags->breserved_tags, 0, fn, priv, true); - bt_for_each(hctx, &tags->bitmap_tags, tags->nr_reserved_tags, fn, priv, - false); } -EXPORT_SYMBOL(blk_mq_tag_busy_iter); static unsigned int bt_unused_tags(struct blk_mq_bitmap_tags *bt) { diff --git a/block/blk-mq-tag.h b/block/blk-mq-tag.h index 9eb2cf4f01cb..d468a79f2c4a 100644 --- a/block/blk-mq-tag.h +++ b/block/blk-mq-tag.h @@ -58,6 +58,8 @@ extern ssize_t blk_mq_tag_sysfs_show(struct blk_mq_tags *tags, char *page); extern void blk_mq_tag_init_last_tag(struct blk_mq_tags *tags, unsigned int *last_tag); extern int blk_mq_tag_update_depth(struct blk_mq_tags *tags, unsigned int depth); extern void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool); +void blk_mq_queue_tag_busy_iter(struct request_queue *q, busy_iter_fn *fn, + void *priv); enum { BLK_MQ_TAG_CACHE_MIN = 1, diff --git a/block/blk-mq.c b/block/blk-mq.c index 2306330530e8..7785ae96267a 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -641,24 +641,16 @@ static void blk_mq_rq_timer(unsigned long priv) .next = 0, .next_set = 0, }; - struct blk_mq_hw_ctx *hctx; int i; - queue_for_each_hw_ctx(q, hctx, i) { - /* - * If not software queues are currently mapped to this - * hardware queue, there's nothing to check - */ - if (!blk_mq_hw_queue_mapped(hctx)) - continue; - - blk_mq_tag_busy_iter(hctx, blk_mq_check_expired, &data); - } + blk_mq_queue_tag_busy_iter(q, blk_mq_check_expired, &data); if (data.next_set) { data.next = blk_rq_timeout(round_jiffies_up(data.next)); mod_timer(&q->timeout, data.next); } else { + struct blk_mq_hw_ctx *hctx; + queue_for_each_hw_ctx(q, hctx, i) { /* the hctx may be unmapped, so check it here */ if (blk_mq_hw_queue_mapped(hctx)) diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index c1b5c867ff07..5e7d43ab61c0 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -223,8 +223,6 @@ void blk_mq_start_hw_queues(struct request_queue *q); void blk_mq_start_stopped_hw_queues(struct request_queue *q, bool async); void blk_mq_run_hw_queues(struct request_queue *q, bool async); void blk_mq_delay_queue(struct blk_mq_hw_ctx *hctx, unsigned long msecs); -void blk_mq_tag_busy_iter(struct blk_mq_hw_ctx *hctx, busy_iter_fn *fn, - void *priv); void blk_mq_all_tag_busy_iter(struct blk_mq_tags *tags, busy_tag_iter_fn *fn, void *priv); void blk_mq_freeze_queue(struct request_queue *q); From 95c2b17534654829db428f11bcf4297c059a2a7e Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Sat, 26 Sep 2015 12:23:56 +0100 Subject: [PATCH 595/647] genirq: Fix race in register_irq_proc() Per-IRQ directories in procfs are created only when a handler is first added to the irqdesc, not when the irqdesc is created. In the case of a shared IRQ, multiple tasks can race to create a directory. This race condition seems to have been present forever, but is easier to hit with async probing. Signed-off-by: Ben Hutchings Link: http://lkml.kernel.org/r/1443266636.2004.2.camel@decadent.org.uk Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org --- kernel/irq/proc.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index e3a8c9577ba6..a50ddc9417ff 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -12,6 +12,7 @@ #include #include #include +#include #include "internals.h" @@ -323,18 +324,29 @@ void register_handler_proc(unsigned int irq, struct irqaction *action) void register_irq_proc(unsigned int irq, struct irq_desc *desc) { + static DEFINE_MUTEX(register_lock); char name [MAX_NAMELEN]; - if (!root_irq_dir || (desc->irq_data.chip == &no_irq_chip) || desc->dir) + if (!root_irq_dir || (desc->irq_data.chip == &no_irq_chip)) return; + /* + * irq directories are registered only when a handler is + * added, not when the descriptor is created, so multiple + * tasks might try to register at the same time. + */ + mutex_lock(®ister_lock); + + if (desc->dir) + goto out_unlock; + memset(name, 0, MAX_NAMELEN); sprintf(name, "%d", irq); /* create /proc/irq/1234 */ desc->dir = proc_mkdir(name, root_irq_dir); if (!desc->dir) - return; + goto out_unlock; #ifdef CONFIG_SMP /* create /proc/irq//smp_affinity */ @@ -355,6 +367,9 @@ void register_irq_proc(unsigned int irq, struct irq_desc *desc) proc_create_data("spurious", 0444, desc->dir, &irq_spurious_proc_fops, (void *)(long)irq); + +out_unlock: + mutex_unlock(®ister_lock); } void unregister_irq_proc(unsigned int irq, struct irq_desc *desc) From a5caa209ba9c29c6421292e7879d2387a2ef39c9 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Fri, 25 Sep 2015 23:02:18 +0100 Subject: [PATCH 596/647] x86/efi: Fix boot crash by mapping EFI memmap entries bottom-up at runtime, instead of top-down Beginning with UEFI v2.5 EFI_PROPERTIES_TABLE was introduced that signals that the firmware PE/COFF loader supports splitting code and data sections of PE/COFF images into separate EFI memory map entries. This allows the kernel to map those regions with strict memory protections, e.g. EFI_MEMORY_RO for code, EFI_MEMORY_XP for data, etc. Unfortunately, an unwritten requirement of this new feature is that the regions need to be mapped with the same offsets relative to each other as observed in the EFI memory map. If this is not done crashes like this may occur, BUG: unable to handle kernel paging request at fffffffefe6086dd IP: [] 0xfffffffefe6086dd Call Trace: [] efi_call+0x7e/0x100 [] ? virt_efi_set_variable+0x61/0x90 [] efi_delete_dummy_variable+0x63/0x70 [] efi_enter_virtual_mode+0x383/0x392 [] start_kernel+0x38a/0x417 [] x86_64_start_reservations+0x2a/0x2c [] x86_64_start_kernel+0xeb/0xef Here 0xfffffffefe6086dd refers to an address the firmware expects to be mapped but which the OS never claimed was mapped. The issue is that included in these regions are relative addresses to other regions which were emitted by the firmware toolchain before the "splitting" of sections occurred at runtime. Needless to say, we don't satisfy this unwritten requirement on x86_64 and instead map the EFI memory map entries in reverse order. The above crash is almost certainly triggerable with any kernel newer than v3.13 because that's when we rewrote the EFI runtime region mapping code, in commit d2f7cbe7b26a ("x86/efi: Runtime services virtual mapping"). For kernel versions before v3.13 things may work by pure luck depending on the fragmentation of the kernel virtual address space at the time we map the EFI regions. Instead of mapping the EFI memory map entries in reverse order, where entry N has a higher virtual address than entry N+1, map them in the same order as they appear in the EFI memory map to preserve this relative offset between regions. This patch has been kept as small as possible with the intention that it should be applied aggressively to stable and distribution kernels. It is very much a bugfix rather than support for a new feature, since when EFI_PROPERTIES_TABLE is enabled we must map things as outlined above to even boot - we have no way of asking the firmware not to split the code/data regions. In fact, this patch doesn't even make use of the more strict memory protections available in UEFI v2.5. That will come later. Suggested-by: Ard Biesheuvel Reported-by: Ard Biesheuvel Signed-off-by: Matt Fleming Cc: Cc: Borislav Petkov Cc: Chun-Yi Cc: Dave Young Cc: H. Peter Anvin Cc: James Bottomley Cc: Lee, Chun-Yi Cc: Leif Lindholm Cc: Linus Torvalds Cc: Matthew Garrett Cc: Mike Galbraith Cc: Peter Jones Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Link: http://lkml.kernel.org/r/1443218539-7610-2-git-send-email-matt@codeblueprint.co.uk Signed-off-by: Ingo Molnar --- arch/x86/platform/efi/efi.c | 67 ++++++++++++++++++++++++++++++++++++- 1 file changed, 66 insertions(+), 1 deletion(-) diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index 1db84c0758b7..6a28ded74211 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -704,6 +704,70 @@ out: return ret; } +/* + * Iterate the EFI memory map in reverse order because the regions + * will be mapped top-down. The end result is the same as if we had + * mapped things forward, but doesn't require us to change the + * existing implementation of efi_map_region(). + */ +static inline void *efi_map_next_entry_reverse(void *entry) +{ + /* Initial call */ + if (!entry) + return memmap.map_end - memmap.desc_size; + + entry -= memmap.desc_size; + if (entry < memmap.map) + return NULL; + + return entry; +} + +/* + * efi_map_next_entry - Return the next EFI memory map descriptor + * @entry: Previous EFI memory map descriptor + * + * This is a helper function to iterate over the EFI memory map, which + * we do in different orders depending on the current configuration. + * + * To begin traversing the memory map @entry must be %NULL. + * + * Returns %NULL when we reach the end of the memory map. + */ +static void *efi_map_next_entry(void *entry) +{ + if (!efi_enabled(EFI_OLD_MEMMAP) && efi_enabled(EFI_64BIT)) { + /* + * Starting in UEFI v2.5 the EFI_PROPERTIES_TABLE + * config table feature requires us to map all entries + * in the same order as they appear in the EFI memory + * map. That is to say, entry N must have a lower + * virtual address than entry N+1. This is because the + * firmware toolchain leaves relative references in + * the code/data sections, which are split and become + * separate EFI memory regions. Mapping things + * out-of-order leads to the firmware accessing + * unmapped addresses. + * + * Since we need to map things this way whether or not + * the kernel actually makes use of + * EFI_PROPERTIES_TABLE, let's just switch to this + * scheme by default for 64-bit. + */ + return efi_map_next_entry_reverse(entry); + } + + /* Initial call */ + if (!entry) + return memmap.map; + + entry += memmap.desc_size; + if (entry >= memmap.map_end) + return NULL; + + return entry; +} + /* * Map the efi memory ranges of the runtime services and update new_mmap with * virtual addresses. @@ -714,7 +778,8 @@ static void * __init efi_map_regions(int *count, int *pg_shift) unsigned long left = 0; efi_memory_desc_t *md; - for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { + p = NULL; + while ((p = efi_map_next_entry(p))) { md = p; if (!(md->attribute & EFI_MEMORY_RUNTIME)) { #ifdef CONFIG_X86_64 From 0ce3cc008ec04258b6a6314b09f1a6012810881a Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Fri, 25 Sep 2015 23:02:19 +0100 Subject: [PATCH 597/647] arm64/efi: Fix boot crash by not padding between EFI_MEMORY_RUNTIME regions The new Properties Table feature introduced in UEFIv2.5 may split memory regions that cover PE/COFF memory images into separate code and data regions. Since these regions only differ in the type (runtime code vs runtime data) and the permission bits, but not in the memory type attributes (UC/WC/WT/WB), the spec does not require them to be aligned to 64 KB. Since the relative offset of PE/COFF .text and .data segments cannot be changed on the fly, this means that we can no longer pad out those regions to be mappable using 64 KB pages. Unfortunately, there is no annotation in the UEFI memory map that identifies data regions that were split off from a code region, so we must apply this logic to all adjacent runtime regions whose attributes only differ in the permission bits. So instead of rounding each memory region to 64 KB alignment at both ends, only round down regions that are not directly preceded by another runtime region with the same type attributes. Since the UEFI spec does not mandate that the memory map be sorted, this means we also need to sort it first. Note that this change will result in all EFI_MEMORY_RUNTIME regions whose start addresses are not aligned to the OS page size to be mapped with executable permissions (i.e., on kernels compiled with 64 KB pages). However, since these mappings are only active during the time that UEFI Runtime Services are being invoked, the window for abuse is rather small. Tested-by: Mark Salter Tested-by: Mark Rutland [UEFI 2.4 only] Signed-off-by: Ard Biesheuvel Signed-off-by: Matt Fleming Reviewed-by: Mark Salter Reviewed-by: Mark Rutland Cc: # v4.0+ Cc: Catalin Marinas Cc: Leif Lindholm Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Will Deacon Cc: linux-kernel@vger.kernel.org Link: http://lkml.kernel.org/r/1443218539-7610-3-git-send-email-matt@codeblueprint.co.uk Signed-off-by: Ingo Molnar --- arch/arm64/kernel/efi.c | 3 +- drivers/firmware/efi/libstub/arm-stub.c | 86 +++++++++++++++++++++---- 2 files changed, 74 insertions(+), 15 deletions(-) diff --git a/arch/arm64/kernel/efi.c b/arch/arm64/kernel/efi.c index e8ca6eaedd02..13671a9cf016 100644 --- a/arch/arm64/kernel/efi.c +++ b/arch/arm64/kernel/efi.c @@ -258,7 +258,8 @@ static bool __init efi_virtmap_init(void) */ if (!is_normal_ram(md)) prot = __pgprot(PROT_DEVICE_nGnRE); - else if (md->type == EFI_RUNTIME_SERVICES_CODE) + else if (md->type == EFI_RUNTIME_SERVICES_CODE || + !PAGE_ALIGNED(md->phys_addr)) prot = PAGE_KERNEL_EXEC; else prot = PAGE_KERNEL; diff --git a/drivers/firmware/efi/libstub/arm-stub.c b/drivers/firmware/efi/libstub/arm-stub.c index e29560e6b40b..950c87f5d279 100644 --- a/drivers/firmware/efi/libstub/arm-stub.c +++ b/drivers/firmware/efi/libstub/arm-stub.c @@ -13,6 +13,7 @@ */ #include +#include #include #include "efistub.h" @@ -305,6 +306,44 @@ fail: */ #define EFI_RT_VIRTUAL_BASE 0x40000000 +static int cmp_mem_desc(const void *l, const void *r) +{ + const efi_memory_desc_t *left = l, *right = r; + + return (left->phys_addr > right->phys_addr) ? 1 : -1; +} + +/* + * Returns whether region @left ends exactly where region @right starts, + * or false if either argument is NULL. + */ +static bool regions_are_adjacent(efi_memory_desc_t *left, + efi_memory_desc_t *right) +{ + u64 left_end; + + if (left == NULL || right == NULL) + return false; + + left_end = left->phys_addr + left->num_pages * EFI_PAGE_SIZE; + + return left_end == right->phys_addr; +} + +/* + * Returns whether region @left and region @right have compatible memory type + * mapping attributes, and are both EFI_MEMORY_RUNTIME regions. + */ +static bool regions_have_compatible_memory_type_attrs(efi_memory_desc_t *left, + efi_memory_desc_t *right) +{ + static const u64 mem_type_mask = EFI_MEMORY_WB | EFI_MEMORY_WT | + EFI_MEMORY_WC | EFI_MEMORY_UC | + EFI_MEMORY_RUNTIME; + + return ((left->attribute ^ right->attribute) & mem_type_mask) == 0; +} + /* * efi_get_virtmap() - create a virtual mapping for the EFI memory map * @@ -317,33 +356,52 @@ void efi_get_virtmap(efi_memory_desc_t *memory_map, unsigned long map_size, int *count) { u64 efi_virt_base = EFI_RT_VIRTUAL_BASE; - efi_memory_desc_t *out = runtime_map; + efi_memory_desc_t *in, *prev = NULL, *out = runtime_map; int l; - for (l = 0; l < map_size; l += desc_size) { - efi_memory_desc_t *in = (void *)memory_map + l; + /* + * To work around potential issues with the Properties Table feature + * introduced in UEFI 2.5, which may split PE/COFF executable images + * in memory into several RuntimeServicesCode and RuntimeServicesData + * regions, we need to preserve the relative offsets between adjacent + * EFI_MEMORY_RUNTIME regions with the same memory type attributes. + * The easiest way to find adjacent regions is to sort the memory map + * before traversing it. + */ + sort(memory_map, map_size / desc_size, desc_size, cmp_mem_desc, NULL); + + for (l = 0; l < map_size; l += desc_size, prev = in) { u64 paddr, size; + in = (void *)memory_map + l; if (!(in->attribute & EFI_MEMORY_RUNTIME)) continue; + paddr = in->phys_addr; + size = in->num_pages * EFI_PAGE_SIZE; + /* * Make the mapping compatible with 64k pages: this allows * a 4k page size kernel to kexec a 64k page size kernel and * vice versa. */ - paddr = round_down(in->phys_addr, SZ_64K); - size = round_up(in->num_pages * EFI_PAGE_SIZE + - in->phys_addr - paddr, SZ_64K); + if (!regions_are_adjacent(prev, in) || + !regions_have_compatible_memory_type_attrs(prev, in)) { - /* - * Avoid wasting memory on PTEs by choosing a virtual base that - * is compatible with section mappings if this region has the - * appropriate size and physical alignment. (Sections are 2 MB - * on 4k granule kernels) - */ - if (IS_ALIGNED(in->phys_addr, SZ_2M) && size >= SZ_2M) - efi_virt_base = round_up(efi_virt_base, SZ_2M); + paddr = round_down(in->phys_addr, SZ_64K); + size += in->phys_addr - paddr; + + /* + * Avoid wasting memory on PTEs by choosing a virtual + * base that is compatible with section mappings if this + * region has the appropriate size and physical + * alignment. (Sections are 2 MB on 4k granule kernels) + */ + if (IS_ALIGNED(in->phys_addr, SZ_2M) && size >= SZ_2M) + efi_virt_base = round_up(efi_virt_base, SZ_2M); + else + efi_virt_base = round_up(efi_virt_base, SZ_64K); + } in->virt_addr = efi_virt_base + in->phys_addr - paddr; efi_virt_base += size; From 606decd67049217684e3cb5a54104d51ddd4ef35 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Thu, 1 Oct 2015 13:12:47 +0200 Subject: [PATCH 598/647] Revert "KVM: x86: apply guest MTRR virtualization on host reserved pages" This reverts commit fd717f11015f673487ffc826e59b2bad69d20fe5. It was reported to cause Machine Check Exceptions (bug 104091). Reported-by: harn-solo@gmx.de Cc: stable@vger.kernel.org # 4.2+ Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm.c | 7 +++++-- arch/x86/kvm/vmx.c | 11 ++++++++--- 2 files changed, 13 insertions(+), 5 deletions(-) diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 94b7d15db3fc..9e88078f4146 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1167,11 +1167,14 @@ static u64 svm_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) u8 mtrr; /* - * 1. MMIO: trust guest MTRR, so same as item 3. + * 1. MMIO: always map as UC * 2. No passthrough: always map as WB, and force guest PAT to WB as well * 3. Passthrough: can't guarantee the result, try to trust guest. */ - if (!is_mmio && !kvm_arch_has_assigned_device(vcpu->kvm)) + if (is_mmio) + return _PAGE_NOCACHE; + + if (!kvm_arch_has_assigned_device(vcpu->kvm)) return 0; if (!kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED) && diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 64076740251e..06ef4908ba61 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -8617,17 +8617,22 @@ static u64 vmx_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) u64 ipat = 0; /* For VT-d and EPT combination - * 1. MMIO: guest may want to apply WC, trust it. + * 1. MMIO: always map as UC * 2. EPT with VT-d: * a. VT-d without snooping control feature: can't guarantee the - * result, try to trust guest. So the same as item 1. + * result, try to trust guest. * b. VT-d with snooping control feature: snooping control feature of * VT-d engine can guarantee the cache correctness. Just set it * to WB to keep consistent with host. So the same as item 3. * 3. EPT without VT-d: always map as WB and set IPAT=1 to keep * consistent with host MTRR */ - if (!is_mmio && !kvm_arch_has_noncoherent_dma(vcpu->kvm)) { + if (is_mmio) { + cache = MTRR_TYPE_UNCACHABLE; + goto exit; + } + + if (!kvm_arch_has_noncoherent_dma(vcpu->kvm)) { ipat = VMX_EPT_IPAT_BIT; cache = MTRR_TYPE_WRBACK; goto exit; From 625422f60c55bbc368b8568ff925770b36bfc189 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Thu, 1 Oct 2015 13:28:15 +0200 Subject: [PATCH 599/647] Revert "KVM: SVM: Sync g_pat with guest-written PAT value" This reverts commit e098223b789b4a618dacd79e5e0dad4a9d5018d1, which has a dependency on other commits being reverted. Cc: stable@vger.kernel.org # 4.2+ Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 9e88078f4146..22601206ffd7 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -3354,16 +3354,6 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) case MSR_VM_IGNNE: vcpu_unimpl(vcpu, "unimplemented wrmsr: 0x%x data 0x%llx\n", ecx, data); break; - case MSR_IA32_CR_PAT: - if (npt_enabled) { - if (!kvm_mtrr_valid(vcpu, MSR_IA32_CR_PAT, data)) - return 1; - vcpu->arch.pat = data; - svm_set_guest_pat(svm, &svm->vmcb->save.g_pat); - mark_dirty(svm->vmcb, VMCB_NPT); - break; - } - /* fall through */ default: return kvm_set_msr_common(vcpu, msr); } From bcf166a9942c3aabd8b752a7c38a49f57c54cfb8 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Thu, 1 Oct 2015 13:19:55 +0200 Subject: [PATCH 600/647] Revert "KVM: svm: handle KVM_X86_QUIRK_CD_NW_CLEARED in svm_get_mt_mask" This reverts commit 5492830370171b6a4ede8a3bfba687a8d0f25fa5. It builds on the commit that is being reverted next. Cc: stable@vger.kernel.org # 4.2+ Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 22601206ffd7..7c242b7007ac 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1177,10 +1177,6 @@ static u64 svm_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) if (!kvm_arch_has_assigned_device(vcpu->kvm)) return 0; - if (!kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED) && - kvm_read_cr0(vcpu) & X86_CR0_CD) - return _PAGE_NOCACHE; - mtrr = kvm_mtrr_get_guest_memory_type(vcpu, gfn); return mtrr2protval[mtrr]; } @@ -1676,10 +1672,13 @@ static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) if (!vcpu->fpu_active) cr0 |= X86_CR0_TS; - - /* These are emulated via page tables. */ - cr0 &= ~(X86_CR0_CD | X86_CR0_NW); - + /* + * re-enable caching here because the QEMU bios + * does not do it - this results in some delay at + * reboot + */ + if (kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED)) + cr0 &= ~(X86_CR0_CD | X86_CR0_NW); svm->vmcb->save.cr0 = cr0; mark_dirty(svm->vmcb, VMCB_CR); update_cr0_intercept(svm); From fc07e76ac7ffa3afd621a1c3858a503386a14281 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Thu, 1 Oct 2015 13:20:22 +0200 Subject: [PATCH 601/647] Revert "KVM: SVM: use NPT page attributes" This reverts commit 3c2e7f7de3240216042b61073803b61b9b3cfb22. Initializing the mapping from MTRR to PAT values was reported to fail nondeterministically, and it also caused extremely slow boot (due to caching getting disabled---bug 103321) with assigned devices. Reported-by: Markus Trippelsdorf Reported-by: Sebastian Schuette Cc: stable@vger.kernel.org # 4.2+ Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm.c | 101 +++------------------------------------------ 1 file changed, 5 insertions(+), 96 deletions(-) diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 7c242b7007ac..99b3c5f40bf7 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -866,64 +866,6 @@ static void svm_disable_lbrv(struct vcpu_svm *svm) set_msr_interception(msrpm, MSR_IA32_LASTINTTOIP, 0, 0); } -#define MTRR_TYPE_UC_MINUS 7 -#define MTRR2PROTVAL_INVALID 0xff - -static u8 mtrr2protval[8]; - -static u8 fallback_mtrr_type(int mtrr) -{ - /* - * WT and WP aren't always available in the host PAT. Treat - * them as UC and UC- respectively. Everything else should be - * there. - */ - switch (mtrr) - { - case MTRR_TYPE_WRTHROUGH: - return MTRR_TYPE_UNCACHABLE; - case MTRR_TYPE_WRPROT: - return MTRR_TYPE_UC_MINUS; - default: - BUG(); - } -} - -static void build_mtrr2protval(void) -{ - int i; - u64 pat; - - for (i = 0; i < 8; i++) - mtrr2protval[i] = MTRR2PROTVAL_INVALID; - - /* Ignore the invalid MTRR types. */ - mtrr2protval[2] = 0; - mtrr2protval[3] = 0; - - /* - * Use host PAT value to figure out the mapping from guest MTRR - * values to nested page table PAT/PCD/PWT values. We do not - * want to change the host PAT value every time we enter the - * guest. - */ - rdmsrl(MSR_IA32_CR_PAT, pat); - for (i = 0; i < 8; i++) { - u8 mtrr = pat >> (8 * i); - - if (mtrr2protval[mtrr] == MTRR2PROTVAL_INVALID) - mtrr2protval[mtrr] = __cm_idx2pte(i); - } - - for (i = 0; i < 8; i++) { - if (mtrr2protval[i] == MTRR2PROTVAL_INVALID) { - u8 fallback = fallback_mtrr_type(i); - mtrr2protval[i] = mtrr2protval[fallback]; - BUG_ON(mtrr2protval[i] == MTRR2PROTVAL_INVALID); - } - } -} - static __init int svm_hardware_setup(void) { int cpu; @@ -990,7 +932,6 @@ static __init int svm_hardware_setup(void) } else kvm_disable_tdp(); - build_mtrr2protval(); return 0; err: @@ -1145,42 +1086,6 @@ static u64 svm_compute_tsc_offset(struct kvm_vcpu *vcpu, u64 target_tsc) return target_tsc - tsc; } -static void svm_set_guest_pat(struct vcpu_svm *svm, u64 *g_pat) -{ - struct kvm_vcpu *vcpu = &svm->vcpu; - - /* Unlike Intel, AMD takes the guest's CR0.CD into account. - * - * AMD does not have IPAT. To emulate it for the case of guests - * with no assigned devices, just set everything to WB. If guests - * have assigned devices, however, we cannot force WB for RAM - * pages only, so use the guest PAT directly. - */ - if (!kvm_arch_has_assigned_device(vcpu->kvm)) - *g_pat = 0x0606060606060606; - else - *g_pat = vcpu->arch.pat; -} - -static u64 svm_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) -{ - u8 mtrr; - - /* - * 1. MMIO: always map as UC - * 2. No passthrough: always map as WB, and force guest PAT to WB as well - * 3. Passthrough: can't guarantee the result, try to trust guest. - */ - if (is_mmio) - return _PAGE_NOCACHE; - - if (!kvm_arch_has_assigned_device(vcpu->kvm)) - return 0; - - mtrr = kvm_mtrr_get_guest_memory_type(vcpu, gfn); - return mtrr2protval[mtrr]; -} - static void init_vmcb(struct vcpu_svm *svm, bool init_event) { struct vmcb_control_area *control = &svm->vmcb->control; @@ -1277,7 +1182,6 @@ static void init_vmcb(struct vcpu_svm *svm, bool init_event) clr_cr_intercept(svm, INTERCEPT_CR3_READ); clr_cr_intercept(svm, INTERCEPT_CR3_WRITE); save->g_pat = svm->vcpu.arch.pat; - svm_set_guest_pat(svm, &save->g_pat); save->cr3 = 0; save->cr4 = 0; } @@ -4187,6 +4091,11 @@ static bool svm_has_high_real_mode_segbase(void) return true; } +static u64 svm_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) +{ + return 0; +} + static void svm_cpuid_update(struct kvm_vcpu *vcpu) { } From 038161dea1eaeee80341134e2675b24656a71b59 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Dirk=20M=C3=BCller?= Date: Thu, 1 Oct 2015 13:46:01 +0200 Subject: [PATCH 602/647] Update KVM homepage Url The old one appears to be a generic catch all page, which is unhelpful. Signed-off-by: Dirk Mueller Signed-off-by: Paolo Bonzini --- MAINTAINERS | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 9f6685f6c5a9..797236befd27 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5957,7 +5957,7 @@ F: virt/kvm/ KERNEL VIRTUAL MACHINE (KVM) FOR AMD-V M: Joerg Roedel L: kvm@vger.kernel.org -W: http://kvm.qumranet.com +W: http://www.linux-kvm.org/ S: Maintained F: arch/x86/include/asm/svm.h F: arch/x86/kvm/svm.c @@ -5965,7 +5965,7 @@ F: arch/x86/kvm/svm.c KERNEL VIRTUAL MACHINE (KVM) FOR POWERPC M: Alexander Graf L: kvm-ppc@vger.kernel.org -W: http://kvm.qumranet.com +W: http://www.linux-kvm.org/ T: git git://github.com/agraf/linux-2.6.git S: Supported F: arch/powerpc/include/asm/kvm* From d2922422c48df93f3edff7d872ee4f3191fefb08 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Dirk=20M=C3=BCller?= Date: Thu, 1 Oct 2015 13:43:42 +0200 Subject: [PATCH 603/647] Use WARN_ON_ONCE for missing X86_FEATURE_NRIPS The cpu feature flags are not ever going to change, so warning everytime can cause a lot of kernel log spam (in our case more than 10GB/hour). The warning seems to only occur when nested virtualization is enabled, so it's probably triggered by a KVM bug. This is a sensible and safe change anyway, and the KVM bug fix might not be suitable for stable releases anyway. Cc: stable@vger.kernel.org Signed-off-by: Dirk Mueller Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 99b3c5f40bf7..2f9ed1ff0632 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -514,7 +514,7 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu) struct vcpu_svm *svm = to_svm(vcpu); if (svm->vmcb->control.next_rip != 0) { - WARN_ON(!static_cpu_has(X86_FEATURE_NRIPS)); + WARN_ON_ONCE(!static_cpu_has(X86_FEATURE_NRIPS)); svm->next_rip = svm->vmcb->control.next_rip; } From 1e16a8f11669c98a0adf5fb5f8522aebc1f69f71 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 1 Oct 2015 15:45:44 +0200 Subject: [PATCH 604/647] MIPS: BPF: Do all exports of symbols with FEXPORT(). FEXPORT also marks the symbol as code using .type symbol, @function. Without objdump -d will output only a hexdump for code following the affected symbols. Signed-off-by: Ralf Baechle --- arch/mips/net/bpf_jit_asm.S | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/arch/mips/net/bpf_jit_asm.S b/arch/mips/net/bpf_jit_asm.S index dabf4179cd7e..eabb5e0889ee 100644 --- a/arch/mips/net/bpf_jit_asm.S +++ b/arch/mips/net/bpf_jit_asm.S @@ -57,8 +57,7 @@ LEAF(sk_load_word) is_offset_negative(word) - .globl sk_load_word_positive -sk_load_word_positive: +FEXPORT(sk_load_word_positive) is_offset_in_header(4, word) /* Offset within header boundaries */ PTR_ADDU t1, $r_skb_data, offset @@ -85,8 +84,7 @@ sk_load_word_positive: LEAF(sk_load_half) is_offset_negative(half) - .globl sk_load_half_positive -sk_load_half_positive: +FEXPORT(sk_load_half_positive) is_offset_in_header(2, half) /* Offset within header boundaries */ PTR_ADDU t1, $r_skb_data, offset @@ -109,8 +107,7 @@ sk_load_half_positive: LEAF(sk_load_byte) is_offset_negative(byte) - .globl sk_load_byte_positive -sk_load_byte_positive: +FEXPORT(sk_load_byte_positive) is_offset_in_header(1, byte) /* Offset within header boundaries */ PTR_ADDU t1, $r_skb_data, offset From 1a541b4e3cd6f5795022514114854b3e1345f24e Mon Sep 17 00:00:00 2001 From: Steve Capper Date: Thu, 1 Oct 2015 13:06:07 +0100 Subject: [PATCH 605/647] arm64: Fix THP protection change logic 6910fa1 ("arm64: enable PTE type bit in the mask for pte_modify") fixes a problem whereby a large block of PROT_NONE mapped memory is incorrectly mapped as block descriptors when mprotect is called. Unfortunately, a subtle bug was introduced by this fix to the THP logic. If one mmaps a large block of memory, then faults it such that it is collapsed into THPs; resulting calls to mprotect on this area of memory will lead to incorrect table descriptors being written instead of block descriptors. This is because pmd_modify calls pte_modify which is now allowed to modify the type of the page table entry. This patch reverts commit 6910fa16dbe142f6a0fd0fd7c249f9883ff7fc8a, and fixes the problem it was trying to address by adjusting PAGE_NONE to represent a table entry. Thus no change in pte type is required when moving from PROT_NONE to a different protection. Fixes: 6910fa16dbe1 ("arm64: enable PTE type bit in the mask for pte_modify") Cc: # 4.0+ Cc: Feng Kan Reported-by: Ganapatrao Kulkarni Tested-by: Ganapatrao Kulkarni Reviewed-by: Catalin Marinas Signed-off-by: Steve Capper Signed-off-by: Catalin Marinas --- arch/arm64/include/asm/pgtable.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h index b0329be95cb1..26b066690593 100644 --- a/arch/arm64/include/asm/pgtable.h +++ b/arch/arm64/include/asm/pgtable.h @@ -79,7 +79,7 @@ extern void __pgd_error(const char *file, int line, unsigned long val); #define PAGE_S2 __pgprot(PROT_DEFAULT | PTE_S2_MEMATTR(MT_S2_NORMAL) | PTE_S2_RDONLY) #define PAGE_S2_DEVICE __pgprot(PROT_DEFAULT | PTE_S2_MEMATTR(MT_S2_DEVICE_nGnRE) | PTE_S2_RDONLY | PTE_UXN) -#define PAGE_NONE __pgprot(((_PAGE_DEFAULT) & ~PTE_TYPE_MASK) | PTE_PROT_NONE | PTE_PXN | PTE_UXN) +#define PAGE_NONE __pgprot(((_PAGE_DEFAULT) & ~PTE_VALID) | PTE_PROT_NONE | PTE_PXN | PTE_UXN) #define PAGE_SHARED __pgprot(_PAGE_DEFAULT | PTE_USER | PTE_NG | PTE_PXN | PTE_UXN | PTE_WRITE) #define PAGE_SHARED_EXEC __pgprot(_PAGE_DEFAULT | PTE_USER | PTE_NG | PTE_PXN | PTE_WRITE) #define PAGE_COPY __pgprot(_PAGE_DEFAULT | PTE_USER | PTE_NG | PTE_PXN | PTE_UXN) @@ -496,7 +496,7 @@ static inline pud_t *pud_offset(pgd_t *pgd, unsigned long addr) static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) { const pteval_t mask = PTE_USER | PTE_PXN | PTE_UXN | PTE_RDONLY | - PTE_PROT_NONE | PTE_WRITE | PTE_TYPE_MASK; + PTE_PROT_NONE | PTE_VALID | PTE_WRITE; /* preserve the hardware dirty information */ if (pte_hw_dirty(pte)) pte = pte_mkdirty(pte); From a523841ee4e506fa1f05ff3a85b1e6d8176a3d4d Mon Sep 17 00:00:00 2001 From: Andrey Ryabinin Date: Thu, 1 Oct 2015 15:36:48 -0700 Subject: [PATCH 606/647] arch/x86/include/asm/efi.h: fix build failure With KMEMCHECK=y, KASAN=n: arch/x86/platform/efi/efi.c:673:3: error: implicit declaration of function `memcpy' [-Werror=implicit-function-declaration] arch/x86/platform/efi/efi_64.c:139:2: error: implicit declaration of function `memcpy' [-Werror=implicit-function-declaration] arch/x86/include/asm/desc.h:121:2: error: implicit declaration of function `memcpy' [-Werror=implicit-function-declaration] Don't #undef memcpy if KASAN=n. Fixes: 769a8089c1fd ("x86, efi, kasan: #undef memset/memcpy/memmove per arch") Signed-off-by: Andrey Ryabinin Reported-by: Ingo Molnar Reported-by: Sedat Dilek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/include/asm/efi.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index ab5f1d447ef9..ae68be92f755 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -86,6 +86,7 @@ extern u64 asmlinkage efi_call(void *fp, ...); extern void __iomem *__init efi_ioremap(unsigned long addr, unsigned long size, u32 type, u64 attribute); +#ifdef CONFIG_KASAN /* * CONFIG_KASAN may redefine memset to __memset. __memset function is present * only in kernel binary. Since the EFI stub linked into a separate binary it @@ -95,6 +96,7 @@ extern void __iomem *__init efi_ioremap(unsigned long addr, unsigned long size, #undef memcpy #undef memset #undef memmove +#endif #endif /* CONFIG_X86_32 */ From 9ff42d10c3b3e26d9555878f31b9a2e5c24efa57 Mon Sep 17 00:00:00 2001 From: Andre Przywara Date: Thu, 1 Oct 2015 15:36:51 -0700 Subject: [PATCH 607/647] userfaultfd: remove kernel header include from uapi header As include/uapi/linux/userfaultfd.h is a user visible header file, it should not include kernel-exclusive header files. So trying to build the userfaultfd test program from the selftests directory fails, since it contains a reference to linux/compiler.h. As it turns out, that header is not really needed there, so we can simply remove it to fix that issue. Signed-off-by: Andre Przywara Cc: Andrea Arcangeli Cc: Shuah Khan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/uapi/linux/userfaultfd.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/include/uapi/linux/userfaultfd.h b/include/uapi/linux/userfaultfd.h index df0e09bb7dd5..9057d7af3ae1 100644 --- a/include/uapi/linux/userfaultfd.h +++ b/include/uapi/linux/userfaultfd.h @@ -11,8 +11,6 @@ #include -#include - #define UFFD_API ((__u64)0xAA) /* * After implementing the respective features it will become: From 03a2d2a3eafe4015412cf4e9675ca0e2d9204074 Mon Sep 17 00:00:00 2001 From: Joonsoo Kim Date: Thu, 1 Oct 2015 15:36:54 -0700 Subject: [PATCH 608/647] mm/slab: fix unexpected index mapping result of kmalloc_size(INDEX_NODE+1) Commit description is copied from the original post of this bug: http://comments.gmane.org/gmane.linux.kernel.mm/135349 Kernels after v3.9 use kmalloc_size(INDEX_NODE + 1) to get the next larger cache size than the size index INDEX_NODE mapping. In kernels 3.9 and earlier we used malloc_sizes[INDEX_L3 + 1].cs_size. However, sometimes we can't get the right output we expected via kmalloc_size(INDEX_NODE + 1), causing a BUG(). The mapping table in the latest kernel is like: index = {0, 1, 2 , 3, 4, 5, 6, n} size = {0, 96, 192, 8, 16, 32, 64, 2^n} The mapping table before 3.10 is like this: index = {0 , 1 , 2, 3, 4 , 5 , 6, n} size = {32, 64, 96, 128, 192, 256, 512, 2^(n+3)} The problem on my mips64 machine is as follows: (1) When configured DEBUG_SLAB && DEBUG_PAGEALLOC && DEBUG_LOCK_ALLOC && DEBUG_SPINLOCK, the sizeof(struct kmem_cache_node) will be "150", and the macro INDEX_NODE turns out to be "2": #define INDEX_NODE kmalloc_index(sizeof(struct kmem_cache_node)) (2) Then the result of kmalloc_size(INDEX_NODE + 1) is 8. (3) Then "if(size >= kmalloc_size(INDEX_NODE + 1)" will lead to "size = PAGE_SIZE". (4) Then "if ((size >= (PAGE_SIZE >> 3))" test will be satisfied and "flags |= CFLGS_OFF_SLAB" will be covered. (5) if (flags & CFLGS_OFF_SLAB)" test will be satisfied and will go to "cachep->slabp_cache = kmalloc_slab(slab_size, 0u)", and the result here may be NULL while kernel bootup. (6) Finally,"BUG_ON(ZERO_OR_NULL_PTR(cachep->slabp_cache));" causes the BUG info as the following shows (may be only mips64 has this problem): This patch fixes the problem of kmalloc_size(INDEX_NODE + 1) and removes the BUG by adding 'size >= 256' check to guarantee that all necessary small sized slabs are initialized regardless sequence of slab size in mapping table. Fixes: e33660165c90 ("slab: Use common kmalloc_index/kmalloc_size...") Signed-off-by: Joonsoo Kim Reported-by: Liuhailong Acked-by: Christoph Lameter Cc: Pekka Enberg Cc: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/slab.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/mm/slab.c b/mm/slab.c index c77ebe6cc87c..4fcc5dd8d5a6 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2190,9 +2190,16 @@ __kmem_cache_create (struct kmem_cache *cachep, unsigned long flags) size += BYTES_PER_WORD; } #if FORCED_DEBUG && defined(CONFIG_DEBUG_PAGEALLOC) - if (size >= kmalloc_size(INDEX_NODE + 1) - && cachep->object_size > cache_line_size() - && ALIGN(size, cachep->align) < PAGE_SIZE) { + /* + * To activate debug pagealloc, off-slab management is necessary + * requirement. In early phase of initialization, small sized slab + * doesn't get initialized so it would not be possible. So, we need + * to check size >= 256. It guarantees that all necessary small + * sized slab is initialized in current slab initialization sequence. + */ + if (!slab_early_init && size >= kmalloc_size(INDEX_NODE) && + size >= 256 && cachep->object_size > cache_line_size() && + ALIGN(size, cachep->align) < PAGE_SIZE) { cachep->obj_offset += PAGE_SIZE - ALIGN(size, cachep->align); size = PAGE_SIZE; } From 2f84a8990ebbe235c59716896e017c6b2ca1200f Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 1 Oct 2015 15:36:57 -0700 Subject: [PATCH 609/647] mm: hugetlbfs: skip shared VMAs when unmapping private pages to satisfy a fault SunDong reported the following on https://bugzilla.kernel.org/show_bug.cgi?id=103841 I think I find a linux bug, I have the test cases is constructed. I can stable recurring problems in fedora22(4.0.4) kernel version, arch for x86_64. I construct transparent huge page, when the parent and child process with MAP_SHARE, MAP_PRIVATE way to access the same huge page area, it has the opportunity to lead to huge page copy on write failure, and then it will munmap the child corresponding mmap area, but then the child mmap area with VM_MAYSHARE attributes, child process munmap this area can trigger VM_BUG_ON in set_vma_resv_flags functions (vma - > vm_flags & VM_MAYSHARE). There were a number of problems with the report (e.g. it's hugetlbfs that triggers this, not transparent huge pages) but it was fundamentally correct in that a VM_BUG_ON in set_vma_resv_flags() can be triggered that looks like this vma ffff8804651fd0d0 start 00007fc474e00000 end 00007fc475e00000 next ffff8804651fd018 prev ffff8804651fd188 mm ffff88046b1b1800 prot 8000000000000027 anon_vma (null) vm_ops ffffffff8182a7a0 pgoff 0 file ffff88106bdb9800 private_data (null) flags: 0x84400fb(read|write|shared|mayread|maywrite|mayexec|mayshare|dontexpand|hugetlb) ------------ kernel BUG at mm/hugetlb.c:462! SMP Modules linked in: xt_pkttype xt_LOG xt_limit [..] CPU: 38 PID: 26839 Comm: map Not tainted 4.0.4-default #1 Hardware name: Dell Inc. PowerEdge R810/0TT6JF, BIOS 2.7.4 04/26/2012 set_vma_resv_flags+0x2d/0x30 The VM_BUG_ON is correct because private and shared mappings have different reservation accounting but the warning clearly shows that the VMA is shared. When a private COW fails to allocate a new page then only the process that created the VMA gets the page -- all the children unmap the page. If the children access that data in the future then they get killed. The problem is that the same file is mapped shared and private. During the COW, the allocation fails, the VMAs are traversed to unmap the other private pages but a shared VMA is found and the bug is triggered. This patch identifies such VMAs and skips them. Signed-off-by: Mel Gorman Reported-by: SunDong Reviewed-by: Michal Hocko Cc: Andrea Arcangeli Cc: Hugh Dickins Cc: Naoya Horiguchi Cc: David Rientjes Reviewed-by: Naoya Horiguchi Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 999fb0aef8f1..9cc773483624 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -3201,6 +3201,14 @@ static void unmap_ref_private(struct mm_struct *mm, struct vm_area_struct *vma, if (iter_vma == vma) continue; + /* + * Shared VMAs have their own reserves and do not affect + * MAP_PRIVATE accounting but it is possible that a shared + * VMA is using the same page so check and skip such VMAs. + */ + if (iter_vma->vm_flags & VM_MAYSHARE) + continue; + /* * Unmap the page from other VMAs without their own reserves. * They get marked to be SIGKILLed if they fault in these From 8346c416d17bf5b4ea1508662959bb62e73fd6a5 Mon Sep 17 00:00:00 2001 From: Ross Zwisler Date: Thu, 1 Oct 2015 15:36:59 -0700 Subject: [PATCH 610/647] dax: fix NULL pointer in __dax_pmd_fault() Commit 46c043ede471 ("mm: take i_mmap_lock in unmap_mapping_range() for DAX") moved some code in __dax_pmd_fault() that was responsible for zeroing newly allocated PMD pages. The new location didn't properly set up 'kaddr', so when run this code resulted in a NULL pointer BUG. Fix this by getting the correct 'kaddr' via bdev_direct_access(). Signed-off-by: Ross Zwisler Reported-by: Dan Williams Reviewed-by: Dan Williams Cc: Alexander Viro Cc: Matthew Wilcox Cc: "Kirill A. Shutemov" Cc: Dave Chinner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dax.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/fs/dax.c b/fs/dax.c index 7ae6df7ea1d2..bcfb14bfc1e4 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -569,8 +569,20 @@ int __dax_pmd_fault(struct vm_area_struct *vma, unsigned long address, if (!buffer_size_valid(&bh) || bh.b_size < PMD_SIZE) goto fallback; + sector = bh.b_blocknr << (blkbits - 9); + if (buffer_unwritten(&bh) || buffer_new(&bh)) { int i; + + length = bdev_direct_access(bh.b_bdev, sector, &kaddr, &pfn, + bh.b_size); + if (length < 0) { + result = VM_FAULT_SIGBUS; + goto out; + } + if ((length < PMD_SIZE) || (pfn & PG_PMD_COLOUR)) + goto fallback; + for (i = 0; i < PTRS_PER_PMD; i++) clear_pmem(kaddr + i * PAGE_SIZE, PAGE_SIZE); wmb_pmem(); @@ -623,7 +635,6 @@ int __dax_pmd_fault(struct vm_area_struct *vma, unsigned long address, result = VM_FAULT_NOPAGE; spin_unlock(ptl); } else { - sector = bh.b_blocknr << (blkbits - 9); length = bdev_direct_access(bh.b_bdev, sector, &kaddr, &pfn, bh.b_size); if (length < 0) { From 0610c25daa3e76e38ad5a8fae683a89ff9f71798 Mon Sep 17 00:00:00 2001 From: Greg Thelen Date: Thu, 1 Oct 2015 15:37:02 -0700 Subject: [PATCH 611/647] memcg: fix dirty page migration The problem starts with a file backed dirty page which is charged to a memcg. Then page migration is used to move oldpage to newpage. Migration: - copies the oldpage's data to newpage - clears oldpage.PG_dirty - sets newpage.PG_dirty - uncharges oldpage from memcg - charges newpage to memcg Clearing oldpage.PG_dirty decrements the charged memcg's dirty page count. However, because newpage is not yet charged, setting newpage.PG_dirty does not increment the memcg's dirty page count. After migration completes newpage.PG_dirty is eventually cleared, often in account_page_cleaned(). At this time newpage is charged to a memcg so the memcg's dirty page count is decremented which causes underflow because the count was not previously incremented by migration. This underflow causes balance_dirty_pages() to see a very large unsigned number of dirty memcg pages which leads to aggressive throttling of buffered writes by processes in non root memcg. This issue: - can harm performance of non root memcg buffered writes. - can report too small (even negative) values in memory.stat[(total_)dirty] counters of all memcg, including the root. To avoid polluting migrate.c with #ifdef CONFIG_MEMCG checks, introduce page_memcg() and set_page_memcg() helpers. Test: 0) setup and enter limited memcg mkdir /sys/fs/cgroup/test echo 1G > /sys/fs/cgroup/test/memory.limit_in_bytes echo $$ > /sys/fs/cgroup/test/cgroup.procs 1) buffered writes baseline dd if=/dev/zero of=/data/tmp/foo bs=1M count=1k sync grep ^dirty /sys/fs/cgroup/test/memory.stat 2) buffered writes with compaction antagonist to induce migration yes 1 > /proc/sys/vm/compact_memory & rm -rf /data/tmp/foo dd if=/dev/zero of=/data/tmp/foo bs=1M count=1k kill % sync grep ^dirty /sys/fs/cgroup/test/memory.stat 3) buffered writes without antagonist, should match baseline rm -rf /data/tmp/foo dd if=/dev/zero of=/data/tmp/foo bs=1M count=1k sync grep ^dirty /sys/fs/cgroup/test/memory.stat (speed, dirty residue) unpatched patched 1) 841 MB/s 0 dirty pages 886 MB/s 0 dirty pages 2) 611 MB/s -33427456 dirty pages 793 MB/s 0 dirty pages 3) 114 MB/s -33427456 dirty pages 891 MB/s 0 dirty pages Notice that unpatched baseline performance (1) fell after migration (3): 841 -> 114 MB/s. In the patched kernel, post migration performance matches baseline. Fixes: c4843a7593a9 ("memcg: add per cgroup dirty page accounting") Signed-off-by: Greg Thelen Reported-by: Dave Hansen Acked-by: Michal Hocko Acked-by: Johannes Weiner Cc: [4.2+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 21 +++++++++++++++++++++ mm/migrate.c | 12 +++++++++++- 2 files changed, 32 insertions(+), 1 deletion(-) diff --git a/include/linux/mm.h b/include/linux/mm.h index 91c08f6f0dc9..80001de019ba 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -905,6 +905,27 @@ static inline void set_page_links(struct page *page, enum zone_type zone, #endif } +#ifdef CONFIG_MEMCG +static inline struct mem_cgroup *page_memcg(struct page *page) +{ + return page->mem_cgroup; +} + +static inline void set_page_memcg(struct page *page, struct mem_cgroup *memcg) +{ + page->mem_cgroup = memcg; +} +#else +static inline struct mem_cgroup *page_memcg(struct page *page) +{ + return NULL; +} + +static inline void set_page_memcg(struct page *page, struct mem_cgroup *memcg) +{ +} +#endif + /* * Some inline functions in vmstat.h depend on page_zone() */ diff --git a/mm/migrate.c b/mm/migrate.c index 7452a00bbb50..842ecd7aaf7f 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -740,6 +740,15 @@ static int move_to_new_page(struct page *newpage, struct page *page, if (PageSwapBacked(page)) SetPageSwapBacked(newpage); + /* + * Indirectly called below, migrate_page_copy() copies PG_dirty and thus + * needs newpage's memcg set to transfer memcg dirty page accounting. + * So perform memcg migration in two steps: + * 1. set newpage->mem_cgroup (here) + * 2. clear page->mem_cgroup (below) + */ + set_page_memcg(newpage, page_memcg(page)); + mapping = page_mapping(page); if (!mapping) rc = migrate_page(mapping, newpage, page, mode); @@ -756,9 +765,10 @@ static int move_to_new_page(struct page *newpage, struct page *page, rc = fallback_migrate_page(mapping, newpage, page, mode); if (rc != MIGRATEPAGE_SUCCESS) { + set_page_memcg(newpage, NULL); newpage->mapping = NULL; } else { - mem_cgroup_migrate(page, newpage, false); + set_page_memcg(page, NULL); if (page_was_mapped) remove_migration_ptes(page, newpage); page->mapping = NULL; From 484ebb3b8c8b27dd2171696462a3116edb9ff801 Mon Sep 17 00:00:00 2001 From: Greg Thelen Date: Thu, 1 Oct 2015 15:37:05 -0700 Subject: [PATCH 612/647] memcg: make mem_cgroup_read_stat() unsigned mem_cgroup_read_stat() returns a page count by summing per cpu page counters. The summing is racy wrt. updates, so a transient negative sum is possible. Callers don't want negative values: - mem_cgroup_wb_stats() doesn't want negative nr_dirty or nr_writeback. This could confuse dirty throttling. - oom reports and memory.stat shouldn't show confusing negative usage. - tree_usage() already avoids negatives. Avoid returning negative page counts from mem_cgroup_read_stat() and convert it to unsigned. [akpm@linux-foundation.org: fix old typo while we're in there] Signed-off-by: Greg Thelen Cc: Johannes Weiner Acked-by: Michal Hocko Cc: [4.2+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 6ddaeba34e09..03cc0a742ff1 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -644,12 +644,14 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz) } /* + * Return page count for single (non recursive) @memcg. + * * Implementation Note: reading percpu statistics for memcg. * * Both of vmstat[] and percpu_counter has threshold and do periodic * synchronization to implement "quick" read. There are trade-off between * reading cost and precision of value. Then, we may have a chance to implement - * a periodic synchronizion of counter in memcg's counter. + * a periodic synchronization of counter in memcg's counter. * * But this _read() function is used for user interface now. The user accounts * memory usage by memory cgroup and he _always_ requires exact value because @@ -659,17 +661,24 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz) * * If there are kernel internal actions which can make use of some not-exact * value, and reading all cpu value can be performance bottleneck in some - * common workload, threashold and synchonization as vmstat[] should be + * common workload, threshold and synchronization as vmstat[] should be * implemented. */ -static long mem_cgroup_read_stat(struct mem_cgroup *memcg, - enum mem_cgroup_stat_index idx) +static unsigned long +mem_cgroup_read_stat(struct mem_cgroup *memcg, enum mem_cgroup_stat_index idx) { long val = 0; int cpu; + /* Per-cpu values can be negative, use a signed accumulator */ for_each_possible_cpu(cpu) val += per_cpu(memcg->stat->count[idx], cpu); + /* + * Summing races with updates, so val may be negative. Avoid exposing + * transient negative values. + */ + if (val < 0) + val = 0; return val; } @@ -1254,7 +1263,7 @@ void mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p) for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { if (i == MEM_CGROUP_STAT_SWAP && !do_swap_account) continue; - pr_cont(" %s:%ldKB", mem_cgroup_stat_names[i], + pr_cont(" %s:%luKB", mem_cgroup_stat_names[i], K(mem_cgroup_read_stat(iter, i))); } @@ -2819,14 +2828,11 @@ static unsigned long tree_stat(struct mem_cgroup *memcg, enum mem_cgroup_stat_index idx) { struct mem_cgroup *iter; - long val = 0; + unsigned long val = 0; - /* Per-cpu values can be negative, use a signed accumulator */ for_each_mem_cgroup_tree(iter, memcg) val += mem_cgroup_read_stat(iter, idx); - if (val < 0) /* race ? */ - val = 0; return val; } @@ -3169,7 +3175,7 @@ static int memcg_stat_show(struct seq_file *m, void *v) for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { if (i == MEM_CGROUP_STAT_SWAP && !do_swap_account) continue; - seq_printf(m, "%s %ld\n", mem_cgroup_stat_names[i], + seq_printf(m, "%s %lu\n", mem_cgroup_stat_names[i], mem_cgroup_read_stat(memcg, i) * PAGE_SIZE); } @@ -3194,13 +3200,13 @@ static int memcg_stat_show(struct seq_file *m, void *v) (u64)memsw * PAGE_SIZE); for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { - long long val = 0; + unsigned long long val = 0; if (i == MEM_CGROUP_STAT_SWAP && !do_swap_account) continue; for_each_mem_cgroup_tree(mi, memcg) val += mem_cgroup_read_stat(mi, i) * PAGE_SIZE; - seq_printf(m, "total_%s %lld\n", mem_cgroup_stat_names[i], val); + seq_printf(m, "total_%s %llu\n", mem_cgroup_stat_names[i], val); } for (i = 0; i < MEM_CGROUP_EVENTS_NSTATS; i++) { From 09a59a9d57a9d6f49510c93304d6e105deb83b93 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 1 Oct 2015 15:37:08 -0700 Subject: [PATCH 613/647] drivers/input/joystick/Kconfig: zhenhua.c needs BITREVERSE It uses bitrev8(), so it must ensure that lib/bitrev.o gets included in vmlinux. Cc: Fengguang Wu Cc: yalin wang Cc: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/input/joystick/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/joystick/Kconfig b/drivers/input/joystick/Kconfig index 56eb471b5576..4215b5382092 100644 --- a/drivers/input/joystick/Kconfig +++ b/drivers/input/joystick/Kconfig @@ -196,6 +196,7 @@ config JOYSTICK_TWIDJOY config JOYSTICK_ZHENHUA tristate "5-byte Zhenhua RC transmitter" select SERIO + select BITREVERSE help Say Y here if you have a Zhen Hua PPM-4CH transmitter which is supplied with a ready to fly micro electric indoor helicopters From 54aea4542980a3ed580426a81c5af799df4d610d Mon Sep 17 00:00:00 2001 From: Petr Mladek Date: Thu, 1 Oct 2015 15:37:11 -0700 Subject: [PATCH 614/647] kprobes: use _do_fork() in samples to make them work again Commit 3033f14ab78c ("clone: support passing tls argument via C rather than pt_regs magic") introduced _do_fork() that allowed to pass @tls parameter. The old do_fork() is defined only for architectures that are not ready to use this way and do not define HAVE_COPY_THREAD_TLS. Let's use _do_fork() in the kprobe examples to make them work again on all architectures. Signed-off-by: Petr Mladek Cc: Ingo Molnar Cc: Masami Hiramatsu Cc: Andy Lutomirski Cc: Peter Zijlstra Cc: "H. Peter Anvin" Cc: Thomas Gleixner Cc: Thiago Macieira Cc: Jiri Kosina Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- samples/kprobes/jprobe_example.c | 14 +++++++------- samples/kprobes/kprobe_example.c | 6 +++--- samples/kprobes/kretprobe_example.c | 4 ++-- 3 files changed, 12 insertions(+), 12 deletions(-) diff --git a/samples/kprobes/jprobe_example.c b/samples/kprobes/jprobe_example.c index 9119ac6a8270..c285a3b8a9f1 100644 --- a/samples/kprobes/jprobe_example.c +++ b/samples/kprobes/jprobe_example.c @@ -1,13 +1,13 @@ /* * Here's a sample kernel module showing the use of jprobes to dump - * the arguments of do_fork(). + * the arguments of _do_fork(). * * For more information on theory of operation of jprobes, see * Documentation/kprobes.txt * * Build and insert the kernel module as done in the kprobe example. * You will see the trace data in /var/log/messages and on the - * console whenever do_fork() is invoked to create a new process. + * console whenever _do_fork() is invoked to create a new process. * (Some messages may be suppressed if syslogd is configured to * eliminate duplicate messages.) */ @@ -17,13 +17,13 @@ #include /* - * Jumper probe for do_fork. + * Jumper probe for _do_fork. * Mirror principle enables access to arguments of the probed routine * from the probe handler. */ -/* Proxy routine having the same arguments as actual do_fork() routine */ -static long jdo_fork(unsigned long clone_flags, unsigned long stack_start, +/* Proxy routine having the same arguments as actual _do_fork() routine */ +static long j_do_fork(unsigned long clone_flags, unsigned long stack_start, unsigned long stack_size, int __user *parent_tidptr, int __user *child_tidptr) { @@ -36,9 +36,9 @@ static long jdo_fork(unsigned long clone_flags, unsigned long stack_start, } static struct jprobe my_jprobe = { - .entry = jdo_fork, + .entry = j_do_fork, .kp = { - .symbol_name = "do_fork", + .symbol_name = "_do_fork", }, }; diff --git a/samples/kprobes/kprobe_example.c b/samples/kprobes/kprobe_example.c index 366db1a9fb65..727eb21c9c56 100644 --- a/samples/kprobes/kprobe_example.c +++ b/samples/kprobes/kprobe_example.c @@ -1,13 +1,13 @@ /* * NOTE: This example is works on x86 and powerpc. * Here's a sample kernel module showing the use of kprobes to dump a - * stack trace and selected registers when do_fork() is called. + * stack trace and selected registers when _do_fork() is called. * * For more information on theory of operation of kprobes, see * Documentation/kprobes.txt * * You will see the trace data in /var/log/messages and on the console - * whenever do_fork() is invoked to create a new process. + * whenever _do_fork() is invoked to create a new process. */ #include @@ -16,7 +16,7 @@ /* For each probe you need to allocate a kprobe structure */ static struct kprobe kp = { - .symbol_name = "do_fork", + .symbol_name = "_do_fork", }; /* kprobe pre_handler: called just before the probed instruction is executed */ diff --git a/samples/kprobes/kretprobe_example.c b/samples/kprobes/kretprobe_example.c index 1041b6731598..ebb1d1aed547 100644 --- a/samples/kprobes/kretprobe_example.c +++ b/samples/kprobes/kretprobe_example.c @@ -7,7 +7,7 @@ * * usage: insmod kretprobe_example.ko func= * - * If no func_name is specified, do_fork is instrumented + * If no func_name is specified, _do_fork is instrumented * * For more information on theory of operation of kretprobes, see * Documentation/kprobes.txt @@ -25,7 +25,7 @@ #include #include -static char func_name[NAME_MAX] = "do_fork"; +static char func_name[NAME_MAX] = "_do_fork"; module_param_string(func, func_name, NAME_MAX, S_IRUGO); MODULE_PARM_DESC(func, "Function to kretprobe; this module will report the" " function's execution time"); From ef510194cefe0cd369ef73419cd65b0a5bb4fb5b Mon Sep 17 00:00:00 2001 From: Greg Thelen Date: Thu, 1 Oct 2015 15:37:13 -0700 Subject: [PATCH 615/647] memcg: remove pcp_counter_lock Commit 733a572e66d2 ("memcg: make mem_cgroup_read_{stat|event}() iterate possible cpus instead of online") removed the last use of the per memcg pcp_counter_lock but forgot to remove the variable. Kill the vestigial variable. Signed-off-by: Greg Thelen Acked-by: Michal Hocko Acked-by: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 1 - mm/memcontrol.c | 1 - 2 files changed, 2 deletions(-) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index ad800e62cb7a..6452ff4c463f 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -242,7 +242,6 @@ struct mem_cgroup { * percpu counter. */ struct mem_cgroup_stat_cpu __percpu *stat; - spinlock_t pcp_counter_lock; #if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_INET) struct cg_proto tcp_mem; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 03cc0a742ff1..1fedbde68f59 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -4185,7 +4185,6 @@ static struct mem_cgroup *mem_cgroup_alloc(void) if (memcg_wb_domain_init(memcg, GFP_KERNEL)) goto out_free_stat; - spin_lock_init(&memcg->pcp_counter_lock); return memcg; out_free_stat: From 44241628bb207ec211bebd156aaf69470d90c209 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 1 Oct 2015 15:37:16 -0700 Subject: [PATCH 616/647] thermal: avoid division by zero in power allocator During boot I get a div by zero Oops regression starting in v4.3-rc3. Signed-off-by: Andrea Arcangeli Reviewed-by: Javi Merino Cc: Zhang Rui Cc: Eduardo Valentin Cc: Daniel Kurtz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/thermal/power_allocator.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/thermal/power_allocator.c b/drivers/thermal/power_allocator.c index 7ff96270c933..e570ff084add 100644 --- a/drivers/thermal/power_allocator.c +++ b/drivers/thermal/power_allocator.c @@ -144,6 +144,16 @@ static void estimate_pid_constants(struct thermal_zone_device *tz, switch_on_temp = 0; temperature_threshold = control_temp - switch_on_temp; + /* + * estimate_pid_constants() tries to find appropriate default + * values for thermal zones that don't provide them. If a + * system integrator has configured a thermal zone with two + * passive trip points at the same temperature, that person + * hasn't put any effort to set up the thermal zone properly + * so just give up. + */ + if (!temperature_threshold) + return; if (!tz->tzp->k_po || force) tz->tzp->k_po = int_to_frac(sustainable_power) / From 676bd99178cd962ed24ffdad222b7069d330a969 Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Thu, 1 Oct 2015 15:37:19 -0700 Subject: [PATCH 617/647] dmapool: fix overflow condition in pool_find_page() If a DMA pool lies at the very top of the dma_addr_t range (as may happen with an IOMMU involved), the calculated end address of the pool wraps around to zero, and page lookup always fails. Tweak the relevant calculation to be overflow-proof. Signed-off-by: Robin Murphy Cc: Arnd Bergmann Cc: Marek Szyprowski Cc: Sumit Semwal Cc: Sakari Ailus Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/dmapool.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/dmapool.c b/mm/dmapool.c index 71a8998cd03a..312a716fa14c 100644 --- a/mm/dmapool.c +++ b/mm/dmapool.c @@ -394,7 +394,7 @@ static struct dma_page *pool_find_page(struct dma_pool *pool, dma_addr_t dma) list_for_each_entry(page, &pool->page_list, page_list) { if (dma < page->dma) continue; - if (dma < (page->dma + pool->allocation)) + if ((dma - page->dma) < pool->allocation) return page; } return NULL; From 03f9abb28c3e10a1d0e08df8bca69e0606b23ea6 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 30 Sep 2015 14:47:37 -0400 Subject: [PATCH 618/647] drm: handle cursor_set2 in restore_fbdev_mode MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If a driver uses the cursor_set2 crtc callback rather than cursor_set, use that. This fixes the fbdev helper for drivers that use cursor_set2. Signed-off-by: Alex Deucher Reviewed-by: Christian König Reviewed-by: Michel Dänzer Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_fb_helper.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 418d299f3b12..ca08c472311b 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -345,7 +345,11 @@ static bool restore_fbdev_mode(struct drm_fb_helper *fb_helper) struct drm_crtc *crtc = mode_set->crtc; int ret; - if (crtc->funcs->cursor_set) { + if (crtc->funcs->cursor_set2) { + ret = crtc->funcs->cursor_set2(crtc, NULL, 0, 0, 0, 0, 0); + if (ret) + error = true; + } else if (crtc->funcs->cursor_set) { ret = crtc->funcs->cursor_set(crtc, NULL, 0, 0, 0); if (ret) error = true; From 0c6dadbe79c54818ed4c268297b663f0c4d1ef98 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 30 Sep 2015 14:47:38 -0400 Subject: [PATCH 619/647] drm/radeon: drop radeon_fb_helper_set_par MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It was just a wrapper around drm_fb_helper_set_par that called cursor_set2 in addition. Now that the core handles this, drop this radeon specific version. Signed-off-by: Alex Deucher Reviewed-by: Christian König Reviewed-by: Michel Dänzer Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_fb.c | 32 +----------------------------- 1 file changed, 1 insertion(+), 31 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c index 7214858ffcea..1aa657fe31cb 100644 --- a/drivers/gpu/drm/radeon/radeon_fb.c +++ b/drivers/gpu/drm/radeon/radeon_fb.c @@ -48,40 +48,10 @@ struct radeon_fbdev { struct radeon_device *rdev; }; -/** - * radeon_fb_helper_set_par - Hide cursor on CRTCs used by fbdev. - * - * @info: fbdev info - * - * This function hides the cursor on all CRTCs used by fbdev. - */ -static int radeon_fb_helper_set_par(struct fb_info *info) -{ - int ret; - - ret = drm_fb_helper_set_par(info); - - /* XXX: with universal plane support fbdev will automatically disable - * all non-primary planes (including the cursor) - */ - if (ret == 0) { - struct drm_fb_helper *fb_helper = info->par; - int i; - - for (i = 0; i < fb_helper->crtc_count; i++) { - struct drm_crtc *crtc = fb_helper->crtc_info[i].mode_set.crtc; - - radeon_crtc_cursor_set2(crtc, NULL, 0, 0, 0, 0, 0); - } - } - - return ret; -} - static struct fb_ops radeonfb_ops = { .owner = THIS_MODULE, .fb_check_var = drm_fb_helper_check_var, - .fb_set_par = radeon_fb_helper_set_par, + .fb_set_par = drm_fb_helper_set_par, .fb_fillrect = drm_fb_helper_cfb_fillrect, .fb_copyarea = drm_fb_helper_cfb_copyarea, .fb_imageblit = drm_fb_helper_cfb_imageblit, From 1c960876be7cffd2798a9e2be090e0a5afaee895 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Wed, 16 Sep 2015 11:04:49 +1000 Subject: [PATCH 620/647] drm/dp/mst: don't pass port into the path builder function This is unnecessary and it makes it easier to see what is needed from port. also add blank line to make things nicer. Reviewed-by: Daniel Vetter Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_dp_mst_topology.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index e23df5fd3836..87112d389fd4 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -1027,8 +1027,8 @@ static void drm_dp_check_port_guid(struct drm_dp_mst_branch *mstb, } } -static void build_mst_prop_path(struct drm_dp_mst_port *port, - struct drm_dp_mst_branch *mstb, +static void build_mst_prop_path(const struct drm_dp_mst_branch *mstb, + int pnum, char *proppath, size_t proppath_size) { @@ -1041,7 +1041,7 @@ static void build_mst_prop_path(struct drm_dp_mst_port *port, snprintf(temp, sizeof(temp), "-%d", port_num); strlcat(proppath, temp, proppath_size); } - snprintf(temp, sizeof(temp), "-%d", port->port_num); + snprintf(temp, sizeof(temp), "-%d", pnum); strlcat(proppath, temp, proppath_size); } @@ -1113,7 +1113,8 @@ static void drm_dp_add_port(struct drm_dp_mst_branch *mstb, if (created && !port->input) { char proppath[255]; - build_mst_prop_path(port, mstb, proppath, sizeof(proppath)); + + build_mst_prop_path(mstb, port->port_num, proppath, sizeof(proppath)); port->connector = (*mstb->mgr->cbs->add_connector)(mstb->mgr, port, proppath); if (port->port_num >= 8) { From df4839fdc9b3c922586b945f062f38cbbda022bb Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Wed, 16 Sep 2015 10:37:28 +1000 Subject: [PATCH 621/647] drm/dp/mst: fixup handling hotplug on port removal. output ports should always have a connector, unless in the rare case connector allocation fails in the driver. In this case we only need to teardown the pdt, and free the struct, and there is no need to send a hotplug msg. In the case were we add the port to the destroy list we need to send a hotplug if we destroy any connectors, so userspace knows to reprobe stuff. this patch also handles port->connector allocation failing which should be a rare event, but makes the code consistent. Reviewed-by: Daniel Vetter Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_dp_mst_topology.c | 36 +++++++++++++++++++-------- 1 file changed, 26 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index 87112d389fd4..d11052c2f84a 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -863,29 +863,33 @@ static void drm_dp_destroy_port(struct kref *kref) { struct drm_dp_mst_port *port = container_of(kref, struct drm_dp_mst_port, kref); struct drm_dp_mst_topology_mgr *mgr = port->mgr; + if (!port->input) { port->vcpi.num_slots = 0; kfree(port->cached_edid); - /* we can't destroy the connector here, as - we might be holding the mode_config.mutex - from an EDID retrieval */ + /* + * The only time we don't have a connector + * on an output port is if the connector init + * fails. + */ if (port->connector) { + /* we can't destroy the connector here, as + * we might be holding the mode_config.mutex + * from an EDID retrieval */ + mutex_lock(&mgr->destroy_connector_lock); list_add(&port->next, &mgr->destroy_connector_list); mutex_unlock(&mgr->destroy_connector_lock); schedule_work(&mgr->destroy_connector_work); return; } + /* no need to clean up vcpi + * as if we have no connector we never setup a vcpi */ drm_dp_port_teardown_pdt(port, port->pdt); - - if (!port->input && port->vcpi.vcpi > 0) - drm_dp_mst_put_payload_id(mgr, port->vcpi.vcpi); } kfree(port); - - (*mgr->cbs->hotplug)(mgr); } static void drm_dp_put_port(struct drm_dp_mst_port *port) @@ -1116,12 +1120,21 @@ static void drm_dp_add_port(struct drm_dp_mst_branch *mstb, build_mst_prop_path(mstb, port->port_num, proppath, sizeof(proppath)); port->connector = (*mstb->mgr->cbs->add_connector)(mstb->mgr, port, proppath); - + if (!port->connector) { + /* remove it from the port list */ + mutex_lock(&mstb->mgr->lock); + list_del(&port->next); + mutex_unlock(&mstb->mgr->lock); + /* drop port list reference */ + drm_dp_put_port(port); + goto out; + } if (port->port_num >= 8) { port->cached_edid = drm_get_edid(port->connector, &port->aux.ddc); } } +out: /* put reference to this port */ drm_dp_put_port(port); } @@ -2672,7 +2685,7 @@ static void drm_dp_destroy_connector_work(struct work_struct *work) { struct drm_dp_mst_topology_mgr *mgr = container_of(work, struct drm_dp_mst_topology_mgr, destroy_connector_work); struct drm_dp_mst_port *port; - + bool send_hotplug = false; /* * Not a regular list traverse as we have to drop the destroy * connector lock before destroying the connector, to avoid AB->BA @@ -2695,7 +2708,10 @@ static void drm_dp_destroy_connector_work(struct work_struct *work) if (!port->input && port->vcpi.vcpi > 0) drm_dp_mst_put_payload_id(mgr, port->vcpi.vcpi); kfree(port); + send_hotplug = true; } + if (send_hotplug) + (*mgr->cbs->hotplug)(mgr); } /** From 68d8c9fc91a0f63a2a10ccf7adf56f69125c11c1 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Sun, 6 Sep 2015 18:53:00 +1000 Subject: [PATCH 622/647] drm/dp/mst: update the link_address_sent before sending the link address (v3) Update the state before sending the msg to close it. v2: reset value if return indicates we haven't send the msg. v3: just clean the code up. Pointed out by Adam J Richter on Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91481 Reviewed-by: Daniel Vetter Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_dp_mst_topology.c | 25 ++++++++++++------------- 1 file changed, 12 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index d11052c2f84a..a2f5caaee242 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -53,8 +53,8 @@ static int drm_dp_send_dpcd_write(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port, int offset, int size, u8 *bytes); -static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, - struct drm_dp_mst_branch *mstb); +static void drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_branch *mstb); static int drm_dp_send_enum_path_resources(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_branch *mstb, struct drm_dp_mst_port *port); @@ -1109,10 +1109,8 @@ static void drm_dp_add_port(struct drm_dp_mst_branch *mstb, drm_dp_port_teardown_pdt(port, old_pdt); ret = drm_dp_port_setup_pdt(port); - if (ret == true) { + if (ret == true) drm_dp_send_link_address(mstb->mgr, port->mstb); - port->mstb->link_address_sent = true; - } } if (created && !port->input) { @@ -1216,10 +1214,9 @@ static void drm_dp_check_and_send_link_address(struct drm_dp_mst_topology_mgr *m { struct drm_dp_mst_port *port; struct drm_dp_mst_branch *mstb_child; - if (!mstb->link_address_sent) { + if (!mstb->link_address_sent) drm_dp_send_link_address(mgr, mstb); - mstb->link_address_sent = true; - } + list_for_each_entry(port, &mstb->ports, next) { if (port->input) continue; @@ -1472,8 +1469,8 @@ static void drm_dp_queue_down_tx(struct drm_dp_mst_topology_mgr *mgr, mutex_unlock(&mgr->qlock); } -static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, - struct drm_dp_mst_branch *mstb) +static void drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_branch *mstb) { int len; struct drm_dp_sideband_msg_tx *txmsg; @@ -1481,11 +1478,12 @@ static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, txmsg = kzalloc(sizeof(*txmsg), GFP_KERNEL); if (!txmsg) - return -ENOMEM; + return; txmsg->dst = mstb; len = build_link_address(txmsg); + mstb->link_address_sent = true; drm_dp_queue_down_tx(mgr, txmsg); ret = drm_dp_mst_wait_tx_reply(mstb, txmsg); @@ -1513,11 +1511,12 @@ static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, } (*mgr->cbs->hotplug)(mgr); } - } else + } else { + mstb->link_address_sent = false; DRM_DEBUG_KMS("link address failed %d\n", ret); + } kfree(txmsg); - return 0; } static int drm_dp_send_enum_path_resources(struct drm_dp_mst_topology_mgr *mgr, From d9515c5ec1a20c77d83471e634ad9bb12deb0eac Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Wed, 16 Sep 2015 17:55:23 +1000 Subject: [PATCH 623/647] drm/dp/mst: split connector registration into two parts (v2) In order to cache the EDID properly for tiled displays, we need to retrieve it before we register the connector with userspace, otherwise userspace can call get resources and try and get the edid before we've even cached it. This fixes some problems when hotplugging mst monitors, with X/mutter running. As mutter seems to get 0 modes for one of the monitors in the tile. v2: fix warning in radeon handle tile setting in cached path rather than get edid path. Reviewed-by: Daniel Vetter Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_dp_mst_topology.c | 8 +++++--- drivers/gpu/drm/i915/intel_dp_mst.c | 9 ++++++++- drivers/gpu/drm/radeon/radeon_dp_mst.c | 11 +++++++++-- include/drm/drm_dp_mst_helper.h | 1 + 4 files changed, 23 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index a2f5caaee242..36454a146ae9 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -1129,7 +1129,9 @@ static void drm_dp_add_port(struct drm_dp_mst_branch *mstb, } if (port->port_num >= 8) { port->cached_edid = drm_get_edid(port->connector, &port->aux.ddc); + drm_mode_connector_set_tile_property(port->connector); } + (*mstb->mgr->cbs->register_connector)(port->connector); } out: @@ -2276,10 +2278,10 @@ struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_ if (port->cached_edid) edid = drm_edid_duplicate(port->cached_edid); - else + else { edid = drm_get_edid(connector, &port->aux.ddc); - - drm_mode_connector_set_tile_property(connector); + drm_mode_connector_set_tile_property(connector); + } drm_dp_put_port(port); return edid; } diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c index 3e4be5a3becd..6ade06888432 100644 --- a/drivers/gpu/drm/i915/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/intel_dp_mst.c @@ -462,11 +462,17 @@ static struct drm_connector *intel_dp_add_mst_connector(struct drm_dp_mst_topolo drm_object_attach_property(&connector->base, dev->mode_config.tile_property, 0); drm_mode_connector_set_path_property(connector, pathprop); + return connector; +} + +static void intel_dp_register_mst_connector(struct drm_connector *connector) +{ + struct intel_connector *intel_connector = to_intel_connector(connector); + struct drm_device *dev = connector->dev; drm_modeset_lock_all(dev); intel_connector_add_to_fbdev(intel_connector); drm_modeset_unlock_all(dev); drm_connector_register(&intel_connector->base); - return connector; } static void intel_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, @@ -512,6 +518,7 @@ static void intel_dp_mst_hotplug(struct drm_dp_mst_topology_mgr *mgr) static struct drm_dp_mst_topology_cbs mst_cbs = { .add_connector = intel_dp_add_mst_connector, + .register_connector = intel_dp_register_mst_connector, .destroy_connector = intel_dp_destroy_mst_connector, .hotplug = intel_dp_mst_hotplug, }; diff --git a/drivers/gpu/drm/radeon/radeon_dp_mst.c b/drivers/gpu/drm/radeon/radeon_dp_mst.c index 5e09c061847f..6cddae44fa6e 100644 --- a/drivers/gpu/drm/radeon/radeon_dp_mst.c +++ b/drivers/gpu/drm/radeon/radeon_dp_mst.c @@ -265,7 +265,6 @@ static struct drm_connector *radeon_dp_add_mst_connector(struct drm_dp_mst_topol { struct radeon_connector *master = container_of(mgr, struct radeon_connector, mst_mgr); struct drm_device *dev = master->base.dev; - struct radeon_device *rdev = dev->dev_private; struct radeon_connector *radeon_connector; struct drm_connector *connector; @@ -286,12 +285,19 @@ static struct drm_connector *radeon_dp_add_mst_connector(struct drm_dp_mst_topol drm_object_attach_property(&connector->base, dev->mode_config.path_property, 0); drm_mode_connector_set_path_property(connector, pathprop); + return connector; +} + +static void radeon_dp_register_mst_connector(struct drm_connector *connector) +{ + struct drm_device *dev = connector->dev; + struct radeon_device *rdev = dev->dev_private; + drm_modeset_lock_all(dev); radeon_fb_add_connector(rdev, connector); drm_modeset_unlock_all(dev); drm_connector_register(connector); - return connector; } static void radeon_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, @@ -324,6 +330,7 @@ static void radeon_dp_mst_hotplug(struct drm_dp_mst_topology_mgr *mgr) struct drm_dp_mst_topology_cbs mst_cbs = { .add_connector = radeon_dp_add_mst_connector, + .register_connector = radeon_dp_register_mst_connector, .destroy_connector = radeon_dp_destroy_mst_connector, .hotplug = radeon_dp_mst_hotplug, }; diff --git a/include/drm/drm_dp_mst_helper.h b/include/drm/drm_dp_mst_helper.h index 86d0b25ed054..0f408b002d98 100644 --- a/include/drm/drm_dp_mst_helper.h +++ b/include/drm/drm_dp_mst_helper.h @@ -374,6 +374,7 @@ struct drm_dp_mst_topology_mgr; struct drm_dp_mst_topology_cbs { /* create a connector for a port */ struct drm_connector *(*add_connector)(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port, const char *path); + void (*register_connector)(struct drm_connector *connector); void (*destroy_connector)(struct drm_dp_mst_topology_mgr *mgr, struct drm_connector *connector); void (*hotplug)(struct drm_dp_mst_topology_mgr *mgr); From 274d83524895fe41ca8debae4eec60ede7252bb5 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Wed, 30 Sep 2015 10:39:42 +1000 Subject: [PATCH 624/647] drm/dp/mst: drop cancel work sync in the mstb destroy path (v2) Since 9eb1e57f564d4e6e10991402726cc83fe0b9172f drm/dp/mst: make sure mst_primary mstb is valid in work function we validate the mstb structs in the work function, and doing that takes a reference. So we should never get here with the work function running using the mstb device, only if the work function hasn't run yet or is running for another mstb. So we don't need to sync the work here, this was causing lockdep spew as below. [ +0.000160] ============================================= [ +0.000001] [ INFO: possible recursive locking detected ] [ +0.000002] 3.10.0-320.el7.rhel72.stable.backport.3.x86_64.debug #1 Tainted: G W ------------ [ +0.000001] --------------------------------------------- [ +0.000001] kworker/4:2/1262 is trying to acquire lock: [ +0.000001] ((&mgr->work)){+.+.+.}, at: [] flush_work+0x5/0x2e0 [ +0.000007] but task is already holding lock: [ +0.000001] ((&mgr->work)){+.+.+.}, at: [] process_one_work+0x1b4/0x710 [ +0.000004] other info that might help us debug this: [ +0.000001] Possible unsafe locking scenario: [ +0.000002] CPU0 [ +0.000000] ---- [ +0.000001] lock((&mgr->work)); [ +0.000002] lock((&mgr->work)); [ +0.000001] *** DEADLOCK *** [ +0.000001] May be due to missing lock nesting notation [ +0.000002] 2 locks held by kworker/4:2/1262: [ +0.000001] #0: (events_long){.+.+.+}, at: [] process_one_work+0x1b4/0x710 [ +0.000004] #1: ((&mgr->work)){+.+.+.}, at: [] process_one_work+0x1b4/0x710 [ +0.000003] stack backtrace: [ +0.000003] CPU: 4 PID: 1262 Comm: kworker/4:2 Tainted: G W ------------ 3.10.0-320.el7.rhel72.stable.backport.3.x86_64.debug #1 [ +0.000001] Hardware name: LENOVO 20EGS0R600/20EGS0R600, BIOS GNET71WW (2.19 ) 02/05/2015 [ +0.000008] Workqueue: events_long drm_dp_mst_link_probe_work [drm_kms_helper] [ +0.000001] ffffffff82c26c90 00000000a527b914 ffff88046399bae8 ffffffff816fe04d [ +0.000004] ffff88046399bb58 ffffffff8110f47f ffff880461438000 0001009b840fc003 [ +0.000002] ffff880461438a98 0000000000000000 0000000804dc26e1 ffffffff824a2c00 [ +0.000003] Call Trace: [ +0.000004] [] dump_stack+0x19/0x1b [ +0.000004] [] __lock_acquire+0x115f/0x1250 [ +0.000002] [] lock_acquire+0x99/0x1e0 [ +0.000002] [] ? flush_work+0x5/0x2e0 [ +0.000002] [] flush_work+0x4e/0x2e0 [ +0.000002] [] ? flush_work+0x5/0x2e0 [ +0.000004] [] ? native_sched_clock+0x35/0x80 [ +0.000002] [] ? sched_clock+0x9/0x10 [ +0.000002] [] ? local_clock+0x25/0x30 [ +0.000002] [] ? mark_held_locks+0xb9/0x140 [ +0.000003] [] ? __cancel_work_timer+0x95/0x160 [ +0.000002] [] __cancel_work_timer+0xa8/0x160 [ +0.000002] [] cancel_work_sync+0x10/0x20 [ +0.000007] [] drm_dp_destroy_mst_branch_device+0x27/0x120 [drm_kms_helper] [ +0.000006] [] drm_dp_mst_link_probe_work+0x78/0xa0 [drm_kms_helper] [ +0.000002] [] process_one_work+0x220/0x710 [ +0.000002] [] ? process_one_work+0x1b4/0x710 [ +0.000005] [] worker_thread+0x11b/0x3a0 [ +0.000003] [] ? process_one_work+0x710/0x710 [ +0.000002] [] kthread+0xed/0x100 [ +0.000003] [] ? insert_kthread_work+0x80/0x80 [ +0.000003] [] ret_from_fork+0x58/0x90 v2: add flush_work. Reviewed-by: Daniel Vetter Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_dp_mst_topology.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index 36454a146ae9..3084b3a7aeb3 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -804,8 +804,6 @@ static void drm_dp_destroy_mst_branch_device(struct kref *kref) struct drm_dp_mst_port *port, *tmp; bool wake_tx = false; - cancel_work_sync(&mstb->mgr->work); - /* * destroy all ports - don't need lock * as there are no more references to the mst branch @@ -1993,6 +1991,8 @@ void drm_dp_mst_topology_mgr_suspend(struct drm_dp_mst_topology_mgr *mgr) drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL, DP_MST_EN | DP_UPSTREAM_IS_SRC); mutex_unlock(&mgr->lock); + flush_work(&mgr->work); + flush_work(&mgr->destroy_connector_work); } EXPORT_SYMBOL(drm_dp_mst_topology_mgr_suspend); @@ -2765,6 +2765,7 @@ EXPORT_SYMBOL(drm_dp_mst_topology_mgr_init); */ void drm_dp_mst_topology_mgr_destroy(struct drm_dp_mst_topology_mgr *mgr) { + flush_work(&mgr->work); flush_work(&mgr->destroy_connector_work); mutex_lock(&mgr->payload_lock); kfree(mgr->payloads); From ccf03d6995fa4b784f5b987726ba98f4859bf326 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Thu, 1 Oct 2015 16:28:25 +1000 Subject: [PATCH 625/647] drm/dp/mst: add some defines for logical/physical ports This just removes the magic number. Acked-by: Daniel Vetter Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_dp_mst_topology.c | 2 +- include/drm/drm_dp_helper.h | 4 ++++ 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index 3084b3a7aeb3..bf27a07dbce3 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -1125,7 +1125,7 @@ static void drm_dp_add_port(struct drm_dp_mst_branch *mstb, drm_dp_put_port(port); goto out; } - if (port->port_num >= 8) { + if (port->port_num >= DP_MST_LOGICAL_PORT_0) { port->cached_edid = drm_get_edid(port->connector, &port->aux.ddc); drm_mode_connector_set_tile_property(port->connector); } diff --git a/include/drm/drm_dp_helper.h b/include/drm/drm_dp_helper.h index 499e9f625aef..0212d139a480 100644 --- a/include/drm/drm_dp_helper.h +++ b/include/drm/drm_dp_helper.h @@ -568,6 +568,10 @@ #define MODE_I2C_READ 4 #define MODE_I2C_STOP 8 +/* DP 1.2 MST PORTs - Section 2.5.1 v1.2a spec */ +#define DP_MST_PHYSICAL_PORT_0 0 +#define DP_MST_LOGICAL_PORT_0 8 + #define DP_LINK_STATUS_SIZE 6 bool drm_dp_channel_eq_ok(const u8 link_status[DP_LINK_STATUS_SIZE], int lane_count); From e3c41e37b0f4b18cbd4dac76cbeece5a7558b909 Mon Sep 17 00:00:00 2001 From: "Lee, Chun-Yi" Date: Tue, 29 Sep 2015 20:58:57 +0800 Subject: [PATCH 626/647] x86/kexec: Fix kexec crash in syscall kexec_file_load() The original bug is a page fault crash that sometimes happens on big machines when preparing ELF headers: BUG: unable to handle kernel paging request at ffffc90613fc9000 IP: [] prepare_elf64_ram_headers_callback+0x165/0x260 The bug is caused by us under-counting the number of memory ranges and subsequently not allocating enough ELF header space for them. The bug is typically masked on smaller systems, because the ELF header allocation is rounded up to the next page. This patch modifies the code in fill_up_crash_elf_data() by using walk_system_ram_res() instead of walk_system_ram_range() to correctly count the max number of crash memory ranges. That's because the walk_system_ram_range() filters out small memory regions that reside in the same page, but walk_system_ram_res() does not. Here's how I found the bug: After tracing prepare_elf64_headers() and prepare_elf64_ram_headers_callback(), the code uses walk_system_ram_res() to fill-in crash memory regions information to the program header, so it counts those small memory regions that reside in a page area. But, when the kernel was using walk_system_ram_range() in fill_up_crash_elf_data() to count the number of crash memory regions, it filters out small regions. I printed those small memory regions, for example: kexec: Get nr_ram ranges. vaddr=0xffff880077592258 paddr=0x77592258, sz=0xdc0 Based on the code in walk_system_ram_range(), this memory region will be filtered out: pfn = (0x77592258 + 0x1000 - 1) >> 12 = 0x77593 end_pfn = (0x77592258 + 0xfc0 -1 + 1) >> 12 = 0x77593 end_pfn - pfn = 0x77593 - 0x77593 = 0 <=== if (end_pfn > pfn) is FALSE So, the max_nr_ranges that's counted by the kernel doesn't include small memory regions - causing us to under-allocate the required space. That causes the page fault crash that happens in a later code path when preparing ELF headers. This bug is not easy to reproduce on small machines that have few CPUs, because the allocated page aligned ELF buffer has more free space to cover those small memory regions' PT_LOAD headers. Signed-off-by: Lee, Chun-Yi Cc: Andy Lutomirski Cc: Baoquan He Cc: Jiang Liu Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Stephen Rothwell Cc: Takashi Iwai Cc: Thomas Gleixner Cc: Viresh Kumar Cc: Vivek Goyal Cc: kexec@lists.infradead.org Cc: linux-kernel@vger.kernel.org Cc: Link: http://lkml.kernel.org/r/1443531537-29436-1-git-send-email-jlee@suse.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/crash.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index e068d6683dba..74ca2fe7a0b3 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -185,10 +185,9 @@ void native_machine_crash_shutdown(struct pt_regs *regs) } #ifdef CONFIG_KEXEC_FILE -static int get_nr_ram_ranges_callback(unsigned long start_pfn, - unsigned long nr_pfn, void *arg) +static int get_nr_ram_ranges_callback(u64 start, u64 end, void *arg) { - int *nr_ranges = arg; + unsigned int *nr_ranges = arg; (*nr_ranges)++; return 0; @@ -214,7 +213,7 @@ static void fill_up_crash_elf_data(struct crash_elf_data *ced, ced->image = image; - walk_system_ram_range(0, -1, &nr_ranges, + walk_system_ram_res(0, -1, &nr_ranges, get_nr_ram_ranges_callback); ced->max_nr_ranges = nr_ranges; From ab76f7b4ab2397ffdd2f1eb07c55697d19991d10 Mon Sep 17 00:00:00 2001 From: Stephen Smalley Date: Thu, 1 Oct 2015 09:04:22 -0400 Subject: [PATCH 627/647] x86/mm: Set NX on gap between __ex_table and rodata Unused space between the end of __ex_table and the start of rodata can be left W+x in the kernel page tables. Extend the setting of the NX bit to cover this gap by starting from text_end rather than rodata_start. Before: ---[ High Kernel Mapping ]--- 0xffffffff80000000-0xffffffff81000000 16M pmd 0xffffffff81000000-0xffffffff81600000 6M ro PSE GLB x pmd 0xffffffff81600000-0xffffffff81754000 1360K ro GLB x pte 0xffffffff81754000-0xffffffff81800000 688K RW GLB x pte 0xffffffff81800000-0xffffffff81a00000 2M ro PSE GLB NX pmd 0xffffffff81a00000-0xffffffff81b3b000 1260K ro GLB NX pte 0xffffffff81b3b000-0xffffffff82000000 4884K RW GLB NX pte 0xffffffff82000000-0xffffffff82200000 2M RW PSE GLB NX pmd 0xffffffff82200000-0xffffffffa0000000 478M pmd After: ---[ High Kernel Mapping ]--- 0xffffffff80000000-0xffffffff81000000 16M pmd 0xffffffff81000000-0xffffffff81600000 6M ro PSE GLB x pmd 0xffffffff81600000-0xffffffff81754000 1360K ro GLB x pte 0xffffffff81754000-0xffffffff81800000 688K RW GLB NX pte 0xffffffff81800000-0xffffffff81a00000 2M ro PSE GLB NX pmd 0xffffffff81a00000-0xffffffff81b3b000 1260K ro GLB NX pte 0xffffffff81b3b000-0xffffffff82000000 4884K RW GLB NX pte 0xffffffff82000000-0xffffffff82200000 2M RW PSE GLB NX pmd 0xffffffff82200000-0xffffffffa0000000 478M pmd Signed-off-by: Stephen Smalley Acked-by: Kees Cook Cc: Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Link: http://lkml.kernel.org/r/1443704662-3138-1-git-send-email-sds@tycho.nsa.gov Signed-off-by: Ingo Molnar --- arch/x86/mm/init_64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 30564e2752d3..df48430c279b 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -1132,7 +1132,7 @@ void mark_rodata_ro(void) * has been zapped already via cleanup_highmem(). */ all_end = roundup((unsigned long)_brk_end, PMD_SIZE); - set_memory_nx(rodata_start, (all_end - rodata_start) >> PAGE_SHIFT); + set_memory_nx(text_end, (all_end - text_end) >> PAGE_SHIFT); rodata_test(); From 88724bfa68be792c1487d759e87568c36ac1a1cc Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 24 Sep 2015 14:00:51 +1000 Subject: [PATCH 628/647] md: wait for pending superblock updates before switching to read-only If a superblock update is pending, wait for it to complete before letting md_set_readonly() switch to readonly. Otherwise we might lose important information about a device having failed. For external arrays, waiting for superblock updates can wait on user-space, so in that case, just return an error. Reported-and-tested-by: Shaohua Li Signed-off-by: NeilBrown --- drivers/md/md.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/md/md.c b/drivers/md/md.c index 4f5ecbe94ccb..b8247bc32cf6 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -5409,9 +5409,13 @@ static int md_set_readonly(struct mddev *mddev, struct block_device *bdev) * which will now never happen */ wake_up_process(mddev->sync_thread->tsk); + if (mddev->external && test_bit(MD_CHANGE_PENDING, &mddev->flags)) + return -EBUSY; mddev_unlock(mddev); wait_event(resync_wait, !test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)); + wait_event(mddev->sb_wait, + !test_bit(MD_CHANGE_PENDING, &mddev->flags)); mddev_lock_nointr(mddev); mutex_lock(&mddev->open_mutex); From ebda780bce8d58ec0abab157397c9e099c41a05f Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Fri, 18 Sep 2015 10:20:13 -0700 Subject: [PATCH 629/647] raid5: update analysis state for failed stripe handle_failed_stripe() makes the stripe fail, eg, all IO will return with a failure, but it doesn't update stripe_head_state. Later handle_stripe() has special handling for raid6 for handle_stripe_fill(). That check before handle_stripe_fill() doesn't skip the failed stripe and we get a kernel crash in need_this_block. This patch clear the analysis state to make sure no functions wrongly called after handle_failed_stripe() Signed-off-by: Shaohua Li Signed-off-by: NeilBrown --- drivers/md/raid5.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 15ef2c641b2b..2105e5f1fd31 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3150,6 +3150,8 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh, spin_unlock_irq(&sh->stripe_lock); if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags)) wake_up(&conf->wait_for_overlap); + if (bi) + s->to_read--; while (bi && bi->bi_iter.bi_sector < sh->dev[i].sector + STRIPE_SECTORS) { struct bio *nextbi = @@ -3169,6 +3171,8 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh, */ clear_bit(R5_LOCKED, &sh->dev[i].flags); } + s->to_write = 0; + s->written = 0; if (test_and_clear_bit(STRIPE_FULL_WRITE, &sh->state)) if (atomic_dec_and_test(&conf->pending_full_writes)) From 36707bb2e7c6730d79d6cdc6d1475d3d7e94c518 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 24 Sep 2015 15:25:36 +1000 Subject: [PATCH 630/647] md/raid5: don't index beyond end of array in need_this_block(). When need_this_block probably shouldn't be called when there are more than 2 failed devices, we really don't want it to try indexing beyond the end of the failed_num[] of fdev[] arrays. So limit the loops to at most 2 iterations. Reported-by: Shaohua Li Signed-off-by: NeilBrown --- drivers/md/raid5.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 2105e5f1fd31..d66ff723df45 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3304,7 +3304,7 @@ static int need_this_block(struct stripe_head *sh, struct stripe_head_state *s, */ return 0; - for (i = 0; i < s->failed; i++) { + for (i = 0; i < s->failed && i < 2; i++) { if (fdev[i]->towrite && !test_bit(R5_UPTODATE, &fdev[i]->flags) && !test_bit(R5_OVERWRITE, &fdev[i]->flags)) @@ -3328,7 +3328,7 @@ static int need_this_block(struct stripe_head *sh, struct stripe_head_state *s, sh->sector < sh->raid_conf->mddev->recovery_cp) /* reconstruct-write isn't being forced */ return 0; - for (i = 0; i < s->failed; i++) { + for (i = 0; i < s->failed && i < 2; i++) { if (s->failed_num[i] != sh->pd_idx && s->failed_num[i] != sh->qd_idx && !test_bit(R5_UPTODATE, &fdev[i]->flags) && From 66eefe5de11db1e0d8f2edc3880d50e7c36a9d43 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 24 Sep 2015 15:47:47 +1000 Subject: [PATCH 631/647] md/raid0: apply base queue limits *before* disk_stack_limits Calling e.g. blk_queue_max_hw_sectors() after calls to disk_stack_limits() discards the settings determined by disk_stack_limits(). So we need to make those calls first. Fixes: 199dc6ed5179 ("md/raid0: update queue parameter in a safer location.") Cc: stable@vger.kernel.org (v2.6.35+ - please apply with 199dc6ed5179). Reported-by: Jes Sorensen Signed-off-by: NeilBrown --- drivers/md/raid0.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c index 63e619b2f44e..f8e5db0cb5aa 100644 --- a/drivers/md/raid0.c +++ b/drivers/md/raid0.c @@ -376,12 +376,6 @@ static int raid0_run(struct mddev *mddev) struct md_rdev *rdev; bool discard_supported = false; - rdev_for_each(rdev, mddev) { - disk_stack_limits(mddev->gendisk, rdev->bdev, - rdev->data_offset << 9); - if (blk_queue_discard(bdev_get_queue(rdev->bdev))) - discard_supported = true; - } blk_queue_max_hw_sectors(mddev->queue, mddev->chunk_sectors); blk_queue_max_write_same_sectors(mddev->queue, mddev->chunk_sectors); blk_queue_max_discard_sectors(mddev->queue, mddev->chunk_sectors); @@ -390,6 +384,12 @@ static int raid0_run(struct mddev *mddev) blk_queue_io_opt(mddev->queue, (mddev->chunk_sectors << 9) * mddev->raid_disks); + rdev_for_each(rdev, mddev) { + disk_stack_limits(mddev->gendisk, rdev->bdev, + rdev->data_offset << 9); + if (blk_queue_discard(bdev_get_queue(rdev->bdev))) + discard_supported = true; + } if (!discard_supported) queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD, mddev->queue); else From d4929add83ad4660b1824a9282ab5dd4d60140fa Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Fri, 18 Sep 2015 10:20:12 -0700 Subject: [PATCH 632/647] md: clear CHANGE_PENDING in readonly array If faulty disks of an array are more than allowed degraded number, the array enters error handling. It will be marked as read-only with MD_CHANGE_PENDING/RECOVERY_NEEDED set. But currently recovery doesn't clear CHANGE_PENDING bit for read-only array. If MD_CHANGE_PENDING is set for a raid5 array, all returned IO will be hold on a list till the bit is clear. But recovery nevery clears this bit, the IO is always in pending state and nevery finish. This has bad effects like upper layer can't get an IO error and the array can't be stopped. Fixes: c3cce6cda162 ("md/raid5: ensure device failure recorded before write request returns.") Signed-off-by: Shaohua Li Signed-off-by: NeilBrown --- drivers/md/md.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/md/md.c b/drivers/md/md.c index b8247bc32cf6..c702de18207a 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -8164,6 +8164,7 @@ void md_check_recovery(struct mddev *mddev) md_reap_sync_thread(mddev); clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery); clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery); + clear_bit(MD_CHANGE_PENDING, &mddev->flags); goto unlock; } From 644df1a85fc4b0c7a16800f55717261546f4e651 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 13 Sep 2015 14:15:10 +0200 Subject: [PATCH 633/647] md: drop null test before destroy functions Remove unneeded NULL test. The semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // @@ expression x; @@ -if (x != NULL) \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x); // Signed-off-by: Julia Lawall Signed-off-by: NeilBrown --- drivers/md/multipath.c | 3 +-- drivers/md/raid1.c | 6 ++---- drivers/md/raid10.c | 9 +++------ drivers/md/raid5.c | 3 +-- 4 files changed, 7 insertions(+), 14 deletions(-) diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index d222522c52e0..d132f06afdd1 100644 --- a/drivers/md/multipath.c +++ b/drivers/md/multipath.c @@ -470,8 +470,7 @@ static int multipath_run (struct mddev *mddev) return 0; out_free_conf: - if (conf->pool) - mempool_destroy(conf->pool); + mempool_destroy(conf->pool); kfree(conf->multipaths); kfree(conf); mddev->private = NULL; diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 4517f06c41ba..5f4f5536f66f 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -2843,8 +2843,7 @@ static struct r1conf *setup_conf(struct mddev *mddev) abort: if (conf) { - if (conf->r1bio_pool) - mempool_destroy(conf->r1bio_pool); + mempool_destroy(conf->r1bio_pool); kfree(conf->mirrors); safe_put_page(conf->tmppage); kfree(conf->poolinfo); @@ -2946,8 +2945,7 @@ static void raid1_free(struct mddev *mddev, void *priv) { struct r1conf *conf = priv; - if (conf->r1bio_pool) - mempool_destroy(conf->r1bio_pool); + mempool_destroy(conf->r1bio_pool); kfree(conf->mirrors); safe_put_page(conf->tmppage); kfree(conf->poolinfo); diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 0fc33eb88855..7c99a4037715 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -3486,8 +3486,7 @@ static struct r10conf *setup_conf(struct mddev *mddev) printk(KERN_ERR "md/raid10:%s: couldn't allocate memory.\n", mdname(mddev)); if (conf) { - if (conf->r10bio_pool) - mempool_destroy(conf->r10bio_pool); + mempool_destroy(conf->r10bio_pool); kfree(conf->mirrors); safe_put_page(conf->tmppage); kfree(conf); @@ -3682,8 +3681,7 @@ static int run(struct mddev *mddev) out_free_conf: md_unregister_thread(&mddev->thread); - if (conf->r10bio_pool) - mempool_destroy(conf->r10bio_pool); + mempool_destroy(conf->r10bio_pool); safe_put_page(conf->tmppage); kfree(conf->mirrors); kfree(conf); @@ -3696,8 +3694,7 @@ static void raid10_free(struct mddev *mddev, void *priv) { struct r10conf *conf = priv; - if (conf->r10bio_pool) - mempool_destroy(conf->r10bio_pool); + mempool_destroy(conf->r10bio_pool); safe_put_page(conf->tmppage); kfree(conf->mirrors); kfree(conf->mirrors_old); diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index d66ff723df45..49bb8d3ff9be 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -2271,8 +2271,7 @@ static void shrink_stripes(struct r5conf *conf) drop_one_stripe(conf)) ; - if (conf->slab_cache) - kmem_cache_destroy(conf->slab_cache); + kmem_cache_destroy(conf->slab_cache); conf->slab_cache = NULL; } From e8ff8bf09ff49733534ff3cee91bde030186055f Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Wed, 16 Sep 2015 10:20:05 -0400 Subject: [PATCH 634/647] md/raid1: Avoid raid1 resync getting stuck close_sync() needs to set conf->next_resync to a large, but safe value below MaxSector and use it to determine whether or not to set start_next_window in wait_barrier() Solution suggested by Neil Brown. Reported-by: Nate Dailey Tested-by: Xiao Ni Signed-off-by: Jes Sorensen Signed-off-by: NeilBrown --- drivers/md/raid1.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 5f4f5536f66f..049df6c4a8cc 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -881,8 +881,7 @@ static sector_t wait_barrier(struct r1conf *conf, struct bio *bio) } if (bio && bio_data_dir(bio) == WRITE) { - if (bio->bi_iter.bi_sector >= - conf->mddev->curr_resync_completed) { + if (bio->bi_iter.bi_sector >= conf->next_resync) { if (conf->start_next_window == MaxSector) conf->start_next_window = conf->next_resync + @@ -1516,7 +1515,7 @@ static void close_sync(struct r1conf *conf) conf->r1buf_pool = NULL; spin_lock_irq(&conf->resync_lock); - conf->next_resync = 0; + conf->next_resync = MaxSector - 2 * NEXT_NORMALIO_DISTANCE; conf->start_next_window = MaxSector; conf->current_window_requests += conf->next_window_requests; From da6fb7a9e5bd6f04f7e15070f630bdf1ea502841 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 1 Oct 2015 16:03:38 +1000 Subject: [PATCH 635/647] md/bitmap: don't pass -1 to bitmap_storage_alloc. Passing -1 to bitmap_storage_alloc() causes page->index to be set to -1, which is quite problematic. So only pass ->cluster_slot if mddev_is_clustered(). Fixes: b97e92574c0b ("Use separate bitmaps for each nodes in the cluster") Cc: stable@vger.kernel.org (v4.1+) Signed-off-by: NeilBrown --- drivers/md/bitmap.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index e51de52eeb94..48b5890c28e3 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -1997,7 +1997,8 @@ int bitmap_resize(struct bitmap *bitmap, sector_t blocks, if (bitmap->mddev->bitmap_info.offset || bitmap->mddev->bitmap_info.file) ret = bitmap_storage_alloc(&store, chunks, !bitmap->mddev->bitmap_info.external, - bitmap->cluster_slot); + mddev_is_clustered(bitmap->mddev) + ? bitmap->cluster_slot : 0); if (ret) goto err; From f4b4aae1828855db761bf998ce37d3062b1d6446 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 1 Oct 2015 01:40:43 +0100 Subject: [PATCH 636/647] x86/headers/uapi: Fix __BITS_PER_LONG value for x32 builds On x32, gcc predefines __x86_64__ but long is only 32-bit. Use __ILP32__ to distinguish x32. Fixes this compiler error in perf: tools/include/asm-generic/bitops/__ffs.h: In function '__ffs': tools/include/asm-generic/bitops/__ffs.h:19:8: error: right shift count >= width of type [-Werror=shift-count-overflow] word >>= 32; ^ This isn't sufficient to build perf for x32, though. Signed-off-by: Ben Hutchings Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Link: http://lkml.kernel.org/r/1443660043.2730.15.camel@decadent.org.uk Signed-off-by: Ingo Molnar --- arch/x86/include/uapi/asm/bitsperlong.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/include/uapi/asm/bitsperlong.h b/arch/x86/include/uapi/asm/bitsperlong.h index b0ae1c4dc791..217909b4d6f5 100644 --- a/arch/x86/include/uapi/asm/bitsperlong.h +++ b/arch/x86/include/uapi/asm/bitsperlong.h @@ -1,7 +1,7 @@ #ifndef __ASM_X86_BITSPERLONG_H #define __ASM_X86_BITSPERLONG_H -#ifdef __x86_64__ +#if defined(__x86_64__) && !defined(__ILP32__) # define __BITS_PER_LONG 64 #else # define __BITS_PER_LONG 32 From 0c5d187828588dd1b36cb93b4481a8db467ef3e8 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 2 Oct 2015 09:48:57 +0200 Subject: [PATCH 637/647] MIPS: BPF: Fix load delay slots. The entire bpf_jit_asm.S is written in noreorder mode because "we know better" according to a comment. This also prevented the assembler from throwing in the required NOPs for MIPS I processors which have no load-use interlock, thus the load's consumer might end up using the old value of the register from prior to the load. Fixed by putting the assembler in reorder mode for just the affected load instructions. This is not enough for gas to actually try to be clever by looking at the next instruction and inserting a nop only when needed but as the comment said "we know better", so getting gas to unconditionally emit a NOP is just right in this case and prevents adding further ifdefery. Signed-off-by: Ralf Baechle --- arch/mips/net/bpf_jit_asm.S | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/mips/net/bpf_jit_asm.S b/arch/mips/net/bpf_jit_asm.S index eabb5e0889ee..5d2e0c8d29c0 100644 --- a/arch/mips/net/bpf_jit_asm.S +++ b/arch/mips/net/bpf_jit_asm.S @@ -61,7 +61,9 @@ FEXPORT(sk_load_word_positive) is_offset_in_header(4, word) /* Offset within header boundaries */ PTR_ADDU t1, $r_skb_data, offset + .set reorder lw $r_A, 0(t1) + .set noreorder #ifdef CONFIG_CPU_LITTLE_ENDIAN # if defined(__mips_isa_rev) && (__mips_isa_rev >= 2) wsbh t0, $r_A @@ -88,7 +90,9 @@ FEXPORT(sk_load_half_positive) is_offset_in_header(2, half) /* Offset within header boundaries */ PTR_ADDU t1, $r_skb_data, offset + .set reorder lh $r_A, 0(t1) + .set noreorder #ifdef CONFIG_CPU_LITTLE_ENDIAN # if defined(__mips_isa_rev) && (__mips_isa_rev >= 2) wsbh t0, $r_A From ee556d00cf20012e889344a0adbbf809ab5015a3 Mon Sep 17 00:00:00 2001 From: Li Bin Date: Wed, 30 Sep 2015 10:49:55 +0800 Subject: [PATCH 638/647] arm64: ftrace: fix function_graph tracer panic When function graph tracer is enabled, the following operation will trigger panic: mount -t debugfs nodev /sys/kernel echo next_tgid > /sys/kernel/tracing/set_ftrace_filter echo function_graph > /sys/kernel/tracing/current_tracer ls /proc/ ------------[ cut here ]------------ [ 198.501417] Unable to handle kernel paging request at virtual address cb88537fdc8ba316 [ 198.506126] pgd = ffffffc008f79000 [ 198.509363] [cb88537fdc8ba316] *pgd=00000000488c6003, *pud=00000000488c6003, *pmd=0000000000000000 [ 198.517726] Internal error: Oops: 94000005 [#1] SMP [ 198.518798] Modules linked in: [ 198.520582] CPU: 1 PID: 1388 Comm: ls Tainted: G [ 198.521800] Hardware name: linux,dummy-virt (DT) [ 198.522852] task: ffffffc0fa9e8000 ti: ffffffc0f9ab0000 task.ti: ffffffc0f9ab0000 [ 198.524306] PC is at next_tgid+0x30/0x100 [ 198.525205] LR is at return_to_handler+0x0/0x20 [ 198.526090] pc : [] lr : [] pstate: 60000145 [ 198.527392] sp : ffffffc0f9ab3d40 [ 198.528084] x29: ffffffc0f9ab3d40 x28: ffffffc0f9ab0000 [ 198.529406] x27: ffffffc000d6a000 x26: ffffffc000b786e8 [ 198.530659] x25: ffffffc0002a1900 x24: ffffffc0faf16c00 [ 198.531942] x23: ffffffc0f9ab3ea0 x22: 0000000000000002 [ 198.533202] x21: ffffffc000d85050 x20: 0000000000000002 [ 198.534446] x19: 0000000000000002 x18: 0000000000000000 [ 198.535719] x17: 000000000049fa08 x16: ffffffc000242efc [ 198.537030] x15: 0000007fa472b54c x14: ffffffffff000000 [ 198.538347] x13: ffffffc0fada84a0 x12: 0000000000000001 [ 198.539634] x11: ffffffc0f9ab3d70 x10: ffffffc0f9ab3d70 [ 198.540915] x9 : ffffffc0000907c0 x8 : ffffffc0f9ab3d40 [ 198.542215] x7 : 0000002e330f08f0 x6 : 0000000000000015 [ 198.543508] x5 : 0000000000000f08 x4 : ffffffc0f9835ec0 [ 198.544792] x3 : cb88537fdc8ba316 x2 : cb88537fdc8ba306 [ 198.546108] x1 : 0000000000000002 x0 : ffffffc000d85050 [ 198.547432] [ 198.547920] Process ls (pid: 1388, stack limit = 0xffffffc0f9ab0020) [ 198.549170] Stack: (0xffffffc0f9ab3d40 to 0xffffffc0f9ab4000) [ 198.582568] Call trace: [ 198.583313] [] next_tgid+0x30/0x100 [ 198.584359] [] ftrace_graph_caller+0x6c/0x70 [ 198.585503] [] ftrace_graph_caller+0x6c/0x70 [ 198.586574] [] ftrace_graph_caller+0x6c/0x70 [ 198.587660] [] ftrace_graph_caller+0x6c/0x70 [ 198.588896] Code: aa0003f5 2a0103f4 b4000102 91004043 (885f7c60) [ 198.591092] ---[ end trace 6a346f8f20949ac8 ]--- This is because when using function graph tracer, if the traced function return value is in multi regs ([x0-x7]), return_to_handler may corrupt them. So in return_to_handler, the parameter regs should be protected properly. Cc: # 3.18+ Signed-off-by: Li Bin Acked-by: AKASHI Takahiro Signed-off-by: Catalin Marinas --- arch/arm64/kernel/entry-ftrace.S | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kernel/entry-ftrace.S b/arch/arm64/kernel/entry-ftrace.S index 08cafc518b9a..0f03a8fe2314 100644 --- a/arch/arm64/kernel/entry-ftrace.S +++ b/arch/arm64/kernel/entry-ftrace.S @@ -178,6 +178,24 @@ ENTRY(ftrace_stub) ENDPROC(ftrace_stub) #ifdef CONFIG_FUNCTION_GRAPH_TRACER + /* save return value regs*/ + .macro save_return_regs + sub sp, sp, #64 + stp x0, x1, [sp] + stp x2, x3, [sp, #16] + stp x4, x5, [sp, #32] + stp x6, x7, [sp, #48] + .endm + + /* restore return value regs*/ + .macro restore_return_regs + ldp x0, x1, [sp] + ldp x2, x3, [sp, #16] + ldp x4, x5, [sp, #32] + ldp x6, x7, [sp, #48] + add sp, sp, #64 + .endm + /* * void ftrace_graph_caller(void) * @@ -204,11 +222,11 @@ ENDPROC(ftrace_graph_caller) * only when CONFIG_HAVE_FUNCTION_GRAPH_FP_TEST is enabled. */ ENTRY(return_to_handler) - str x0, [sp, #-16]! + save_return_regs mov x0, x29 // parent's fp bl ftrace_return_to_handler// addr = ftrace_return_to_hander(fp); mov x30, x0 // restore the original return address - ldr x0, [sp], #16 + restore_return_regs ret END(return_to_handler) #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ From 0fa24340f7c88d2814547d8d24d5e3a1803009cc Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Mon, 21 Sep 2015 10:07:41 -0700 Subject: [PATCH 639/647] MIPS: Fix octeon FP context switch handling Commit 1a3d59579b9f ("MIPS: Tidy up FPU context switching") removed FP context saving from the asm-written resume function in favour of reusing existing code to perform the same task. However it only removed the FP context saving code from the r4k_switch.S implementation of resume. Octeon uses its own implementation in octeon_switch.S, so remove FP context saving there too in order to prevent attempting to save context twice. That formerly led to an exception from the second save as follows because the FPU had already been disabled by the first save: do_cpu invoked from kernel context![#1]: CPU: 0 PID: 2 Comm: kthreadd Not tainted 4.3.0-rc2-dirty #2 task: 800000041f84a008 ti: 800000041f864000 task.ti: 800000041f864000 $ 0 : 0000000000000000 0000000010008ce1 0000000000100000 ffffffffbfffffff $ 4 : 800000041f84a008 800000041f84ac08 800000041f84c000 0000000000000004 $ 8 : 0000000000000001 0000000000000000 0000000000000000 0000000000000001 $12 : 0000000010008ce3 0000000000119c60 0000000000000036 800000041f864000 $16 : 800000041f84ac08 800000000792ce80 800000041f84a008 ffffffff81758b00 $20 : 0000000000000000 ffffffff8175ae50 0000000000000000 ffffffff8176c740 $24 : 0000000000000006 ffffffff81170300 $28 : 800000041f864000 800000041f867d90 0000000000000000 ffffffff815f3fa0 Hi : 0000000000fa8257 Lo : ffffffffe15cfc00 epc : ffffffff8112821c resume+0x9c/0x200 ra : ffffffff815f3fa0 __schedule+0x3f0/0x7d8 Status: 10008ce2 KX SX UX KERNEL EXL Cause : 1080002c (ExcCode 0b) PrId : 000d0601 (Cavium Octeon+) Modules linked in: Process kthreadd (pid: 2, threadinfo=800000041f864000, task=800000041f84a008, tls=0000000000000000) Stack : ffffffff81604218 ffffffff815f7e08 800000041f84a008 ffffffff811681b0 800000041f84a008 ffffffff817e9878 0000000000000000 ffffffff81770000 ffffffff81768340 ffffffff81161398 0000000000000001 0000000000000000 0000000000000000 ffffffff815f4424 0000000000000000 ffffffff81161d68 ffffffff81161be8 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 ffffffff8111e16c 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 0000000000000000 ... Call Trace: [] resume+0x9c/0x200 [] __schedule+0x3f0/0x7d8 [] schedule+0x34/0x98 [] kthreadd+0x180/0x198 [] ret_from_kernel_thread+0x14/0x1c Tested using cavium_octeon_defconfig on an EdgeRouter Lite. Fixes: 1a3d59579b9f ("MIPS: Tidy up FPU context switching") Reported-by: Aaro Koskinen Signed-off-by: Paul Burton Cc: linux-mips@linux-mips.org Cc: Aleksey Makarov Cc: linux-kernel@vger.kernel.org Cc: Chandrakala Chavva Cc: David Daney Cc: Leonid Rosenboim Patchwork: https://patchwork.linux-mips.org/patch/11166/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/octeon_switch.S | 26 +------------------------- 1 file changed, 1 insertion(+), 25 deletions(-) diff --git a/arch/mips/kernel/octeon_switch.S b/arch/mips/kernel/octeon_switch.S index 423ae83af1fb..3375745b9198 100644 --- a/arch/mips/kernel/octeon_switch.S +++ b/arch/mips/kernel/octeon_switch.S @@ -18,7 +18,7 @@ .set pop /* * task_struct *resume(task_struct *prev, task_struct *next, - * struct thread_info *next_ti, int usedfpu) + * struct thread_info *next_ti) */ .align 7 LEAF(resume) @@ -28,30 +28,6 @@ cpu_save_nonscratch a0 LONG_S ra, THREAD_REG31(a0) - /* - * check if we need to save FPU registers - */ - .set push - .set noreorder - beqz a3, 1f - PTR_L t3, TASK_THREAD_INFO(a0) - .set pop - - /* - * clear saved user stack CU1 bit - */ - LONG_L t0, ST_OFF(t3) - li t1, ~ST0_CU1 - and t0, t0, t1 - LONG_S t0, ST_OFF(t3) - - .set push - .set arch=mips64r2 - fpu_save_double a0 t0 t1 # c0_status passed in t0 - # clobbers t1 - .set pop -1: - #if CONFIG_CAVIUM_OCTEON_CVMSEG_SIZE > 0 /* Check if we need to store CVMSEG state */ dmfc0 t0, $11,7 /* CvmMemCtl */ From 085c2f25d36ef4a69bb1dab933daee0692426f15 Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Mon, 21 Sep 2015 10:07:42 -0700 Subject: [PATCH 640/647] MIPS: Fix R2300 FP context switch handling Commit 1a3d59579b9f ("MIPS: Tidy up FPU context switching") removed FP context saving from the asm-written resume function in favour of reusing existing code to perform the same task. However it only removed the FP context saving code from the r4k_switch.S implementation of resume. Remove it from the r2300_switch.S implementation too in order to prevent attempting to save the FP context twice, which would likely lead to an exception from the second save because the FPU had already been disabled by the first save. This patch has only been build tested, using rbtx49xx_defconfig. Fixes: 1a3d59579b9f ("MIPS: Tidy up FPU context switching") Signed-off-by: Paul Burton Cc: linux-mips@linux-mips.org Cc: Maciej W. Rozycki Cc: linux-kernel@vger.kernel.org Cc: Manuel Lauss Patchwork: https://patchwork.linux-mips.org/patch/11167/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/r2300_switch.S | 28 +--------------------------- 1 file changed, 1 insertion(+), 27 deletions(-) diff --git a/arch/mips/kernel/r2300_switch.S b/arch/mips/kernel/r2300_switch.S index 5087a4b72e6b..ac27ef7d4d0e 100644 --- a/arch/mips/kernel/r2300_switch.S +++ b/arch/mips/kernel/r2300_switch.S @@ -30,19 +30,9 @@ */ #define ST_OFF (_THREAD_SIZE - 32 - PT_SIZE + PT_STATUS) -/* - * FPU context is saved iff the process has used it's FPU in the current - * time slice as indicated by TIF_USEDFPU. In any case, the CU1 bit for user - * space STATUS register should be 0, so that a process *always* starts its - * userland with FPU disabled after each context switch. - * - * FPU will be enabled as soon as the process accesses FPU again, through - * do_cpu() trap. - */ - /* * task_struct *resume(task_struct *prev, task_struct *next, - * struct thread_info *next_ti, int usedfpu) + * struct thread_info *next_ti) */ LEAF(resume) mfc0 t1, CP0_STATUS @@ -50,22 +40,6 @@ LEAF(resume) cpu_save_nonscratch a0 sw ra, THREAD_REG31(a0) - beqz a3, 1f - - PTR_L t3, TASK_THREAD_INFO(a0) - - /* - * clear saved user stack CU1 bit - */ - lw t0, ST_OFF(t3) - li t1, ~ST0_CU1 - and t0, t0, t1 - sw t0, ST_OFF(t3) - - fpu_save_single a0, t0 # clobbers t0 - -1: - #if defined(CONFIG_CC_STACKPROTECTOR) && !defined(CONFIG_SMP) PTR_LA t8, __stack_chk_guard LONG_L t9, TASK_STACK_CANARY(a1) From 66803dd9198cb57a4b7ed4a6846a63ab1d59a2e0 Mon Sep 17 00:00:00 2001 From: Matt Bennett Date: Wed, 30 Sep 2015 17:40:42 +1300 Subject: [PATCH 641/647] MIPS: Octeon: Fix kernel panic on startup from memory corruption During development it was found that a number of builds would panic during the kernel init process, more specifically in 'delayed_fput()'. The panic showed the kernel trying to access a memory address of '0xb7fdc00' while traversing the 'delayed_fput_list' structure. Comparing this memory address to the value of the pointer used on builds that did not panic confirmed that the pointer on crashing builds must have been corrupted at some stage earlier in the init process. By traversing the list earlier and earlier in the code it was found that 'plat_mem_setup()' was responsible for corrupting the list. Specifically the line: memory = cvmx_bootmem_phy_alloc(mem_alloc_size, __pa_symbol(&__init_end), -1, 0x100000, CVMX_BOOTMEM_FLAG_NO_LOCKING); Which would eventually call: cvmx_bootmem_phy_set_size(new_ent_addr, cvmx_bootmem_phy_get_size (ent_addr) - (desired_min_addr - ent_addr)); Where 'new_ent_addr'=0x4800000 (the address of 'delayed_fput_list') and the second argument (size)=0xb7fdc00 (the address causing the kernel panic). The job of this part of 'plat_mem_setup()' is to allocate chunks of memory for the kernel to use. At the start of each chunk of memory the size of the chunk is written, hence the value 0xb7fdc00 is written onto memory at 0x4800000, therefore the kernel panics when it goes back to access 'delayed_fput_list' later on in the initialisation process. On builds that were not crashing it was found that the compiler had placed 'delayed_fput_list' at 0x4800008, meaning it wasn't corrupted (but something else in memory was overwritten). As can be seen in the first function call above the code begins to allocate chunks of memory beginning from the symbol '__init_end'. The MIPS linker script (vmlinux.lds.S) however defines the .bss section to begin after '__init_end'. Therefore memory within the .bss section is allocated to the kernel to use (System.map shows 'delayed_fput_list' and other kernel structures to be in .bss). To stop the kernel panic (and the .bss section being corrupted) memory should begin being allocated from the symbol '_end'. Signed-off-by: Matt Bennett Acked-by: David Daney Cc: linux-mips@linux-mips.org Cc: aleksey.makarov@auriga.com Patchwork: https://patchwork.linux-mips.org/patch/11251/ Signed-off-by: Ralf Baechle --- arch/mips/cavium-octeon/setup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index 89a628455bc2..bd634259eab9 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -933,7 +933,7 @@ void __init plat_mem_setup(void) while ((boot_mem_map.nr_map < BOOT_MEM_MAP_MAX) && (total < MAX_MEMORY)) { memory = cvmx_bootmem_phy_alloc(mem_alloc_size, - __pa_symbol(&__init_end), -1, + __pa_symbol(&_end), -1, 0x100000, CVMX_BOOTMEM_FLAG_NO_LOCKING); if (memory >= 0) { From 62d78461447198b49383f20301aaa15fe97dfa4f Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Fri, 2 Oct 2015 10:31:32 -0700 Subject: [PATCH 642/647] Revert "Input: synaptics - fix handling of disabling gesture mode" This reverts commit e51e38494a8ecc18650efb0c840600637891de2c: we actually do want the device to work in extended W mode, as this is the mode that allows us receiving multiple contact information. Cc: stable@vger.kernel.org --- drivers/input/mouse/synaptics.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index 994ae7886156..6025eb430c0a 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -519,18 +519,14 @@ static int synaptics_set_mode(struct psmouse *psmouse) struct synaptics_data *priv = psmouse->private; priv->mode = 0; - - if (priv->absolute_mode) { + if (priv->absolute_mode) priv->mode |= SYN_BIT_ABSOLUTE_MODE; - if (SYN_CAP_EXTENDED(priv->capabilities)) - priv->mode |= SYN_BIT_W_MODE; - } - - if (!SYN_MODE_WMODE(priv->mode) && priv->disable_gesture) + if (priv->disable_gesture) priv->mode |= SYN_BIT_DISABLE_GESTURE; - if (psmouse->rate >= 80) priv->mode |= SYN_BIT_HIGH_RATE; + if (SYN_CAP_EXTENDED(priv->capabilities)) + priv->mode |= SYN_BIT_W_MODE; if (synaptics_mode_cmd(psmouse, priv->mode)) return -1; From c8415b9470727f70afce8607d4fe521789aa6c1c Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Fri, 2 Oct 2015 16:44:05 +0100 Subject: [PATCH 643/647] irqchip/gic-v3-its: Silence warning when its_lpi_alloc_chunks gets inlined More agressive inlining in recent versions of GCC have uncovered a new set of warnings: drivers/irqchip/irq-gic-v3-its.c: In function its_msi_prepare: drivers/irqchip/irq-gic-v3-its.c:1148:26: warning: lpi_base may be used uninitialized in this function [-Wmaybe-uninitialized] dev->event_map.lpi_base = lpi_base; ^ drivers/irqchip/irq-gic-v3-its.c:1116:6: note: lpi_base was declared here int lpi_base; ^ drivers/irqchip/irq-gic-v3-its.c:1149:25: warning: nr_lpis may be used uninitialized in this function [-Wmaybe-uninitialized] dev->event_map.nr_lpis = nr_lpis; ^ drivers/irqchip/irq-gic-v3-its.c:1117:6: note: nr_lpis was declared here int nr_lpis; ^ The warning is fairly benign (there is no code path that could actually use uninitialized variables), but let's silence it anyway by zeroing the variables on the error path. Reported-by: Alex Shi Tested-by: Ard Biesheuvel Signed-off-by: Marc Zyngier Cc: linux-arm-kernel@lists.infradead.org Cc: David Daney Cc: Jason Cooper Link: http://lkml.kernel.org/r/1443800646-8074-2-git-send-email-marc.zyngier@arm.com Signed-off-by: Thomas Gleixner --- drivers/irqchip/irq-gic-v3-its.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c index ac7ae2b3cb83..25ceae9f7348 100644 --- a/drivers/irqchip/irq-gic-v3-its.c +++ b/drivers/irqchip/irq-gic-v3-its.c @@ -719,6 +719,9 @@ static unsigned long *its_lpi_alloc_chunks(int nr_irqs, int *base, int *nr_ids) out: spin_unlock(&lpi_lock); + if (!bitmap) + *base = *nr_ids = 0; + return bitmap; } From 791c76d58465a248cbd1ee422c8075cb90fa615f Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Fri, 2 Oct 2015 16:44:06 +0100 Subject: [PATCH 644/647] irqchip/gic-v3-its: Count additional LPIs for the aliased devices When configuring the interrupt mapping for a new device, we iterate over all the possible aliases to account for their maximum MSI allocation. This was introduced by e8137f4f5088 ("irqchip: gicv3-its: Iterate over PCI aliases to generate ITS configuration"). Turns out that the code doing that is a bit braindead, and repeatedly accounts for the same device over and over. Fix this by counting the actual alias that is passed to us by the core code. Signed-off-by: Marc Zyngier Cc: linux-arm-kernel@lists.infradead.org Cc: Alex Shi Cc: Ard Biesheuvel Cc: David Daney Cc: Jason Cooper Link: http://lkml.kernel.org/r/1443800646-8074-3-git-send-email-marc.zyngier@arm.com Signed-off-by: Thomas Gleixner --- drivers/irqchip/irq-gic-v3-its-pci-msi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/irqchip/irq-gic-v3-its-pci-msi.c b/drivers/irqchip/irq-gic-v3-its-pci-msi.c index cf351c637464..a7c8c9ffbafd 100644 --- a/drivers/irqchip/irq-gic-v3-its-pci-msi.c +++ b/drivers/irqchip/irq-gic-v3-its-pci-msi.c @@ -62,7 +62,7 @@ static int its_get_pci_alias(struct pci_dev *pdev, u16 alias, void *data) dev_alias->dev_id = alias; if (pdev != dev_alias->pdev) - dev_alias->count += its_pci_msi_vec_count(dev_alias->pdev); + dev_alias->count += its_pci_msi_vec_count(pdev); return 0; } From 67dfae0cd72fec5cd158b6e5fb1647b7dbe0834c Mon Sep 17 00:00:00 2001 From: John Stultz Date: Mon, 14 Sep 2015 18:05:20 -0700 Subject: [PATCH 645/647] clocksource: Fix abs() usage w/ 64bit values This patch fixes one cases where abs() was being used with 64-bit nanosecond values, where the result may be capped at 32-bits. This potentially could cause watchdog false negatives on 32-bit systems, so this patch addresses the issue by using abs64(). Signed-off-by: John Stultz Cc: Prarit Bhargava Cc: Richard Cochran Cc: Ingo Molnar Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1442279124-7309-2-git-send-email-john.stultz@linaro.org Signed-off-by: Thomas Gleixner --- kernel/time/clocksource.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 841b72f720e8..3a38775b50c2 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -217,7 +217,7 @@ static void clocksource_watchdog(unsigned long data) continue; /* Check the deviation from the watchdog clocksource. */ - if ((abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD)) { + if (abs64(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD) { pr_warn("timekeeping watchdog: Marking clocksource '%s' as unstable because the skew is too large:\n", cs->name); pr_warn(" '%s' wd_now: %llx wd_last: %llx mask: %llx\n", From d218af78492a36a4ae607c08fedfb59258440314 Mon Sep 17 00:00:00 2001 From: Markos Chandras Date: Fri, 25 Sep 2015 08:17:42 +0100 Subject: [PATCH 646/647] MIPS: scall: Always run the seccomp syscall filters The MIPS syscall handler code used to return -ENOSYS on invalid syscalls. Whilst this is expected, it caused problems for seccomp filters because the said filters never had the change to run since the code returned -ENOSYS before triggering them. This caused problems on the chromium testsuite for filters looking for invalid syscalls. This has now changed and the seccomp filters are always run even if the syscall is invalid. We return -ENOSYS once we return from the seccomp filters. Moreover, similar codepaths have been merged in the process which simplifies somewhat the overall syscall code. Signed-off-by: Markos Chandras Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/11236/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/scall32-o32.S | 39 ++++++++++++++-------------------- arch/mips/kernel/scall64-64.S | 38 ++++++++++++++------------------- arch/mips/kernel/scall64-n32.S | 19 +++++------------ arch/mips/kernel/scall64-o32.S | 19 +++++------------ 4 files changed, 42 insertions(+), 73 deletions(-) diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S index d06e30c1e4cd..65a74e4f0f45 100644 --- a/arch/mips/kernel/scall32-o32.S +++ b/arch/mips/kernel/scall32-o32.S @@ -36,16 +36,8 @@ NESTED(handle_sys, PT_SIZE, sp) lw t1, PT_EPC(sp) # skip syscall on return subu v0, v0, __NR_O32_Linux # check syscall number - sltiu t0, v0, __NR_O32_Linux_syscalls + 1 addiu t1, 4 # skip to next instruction sw t1, PT_EPC(sp) - beqz t0, illegal_syscall - - sll t0, v0, 2 - la t1, sys_call_table - addu t1, t0 - lw t2, (t1) # syscall routine - beqz t2, illegal_syscall sw a3, PT_R26(sp) # save a3 for syscall restarting @@ -96,6 +88,16 @@ loads_done: li t1, _TIF_WORK_SYSCALL_ENTRY and t0, t1 bnez t0, syscall_trace_entry # -> yes +syscall_common: + sltiu t0, v0, __NR_O32_Linux_syscalls + 1 + beqz t0, illegal_syscall + + sll t0, v0, 2 + la t1, sys_call_table + addu t1, t0 + lw t2, (t1) # syscall routine + + beqz t2, illegal_syscall jalr t2 # Do The Real Thing (TM) @@ -116,7 +118,7 @@ o32_syscall_exit: syscall_trace_entry: SAVE_STATIC - move s0, t2 + move s0, v0 move a0, sp /* @@ -129,27 +131,18 @@ syscall_trace_entry: 1: jal syscall_trace_enter - bltz v0, 2f # seccomp failed? Skip syscall + bltz v0, 1f # seccomp failed? Skip syscall + + move v0, s0 # restore syscall - move t0, s0 RESTORE_STATIC lw a0, PT_R4(sp) # Restore argument registers lw a1, PT_R5(sp) lw a2, PT_R6(sp) lw a3, PT_R7(sp) - jalr t0 + j syscall_common - li t0, -EMAXERRNO - 1 # error? - sltu t0, t0, v0 - sw t0, PT_R7(sp) # set error flag - beqz t0, 1f - - lw t1, PT_R2(sp) # syscall number - negu v0 # error - sw t1, PT_R0(sp) # save it for syscall restarting -1: sw v0, PT_R2(sp) # result - -2: j syscall_exit +1: j syscall_exit /* ------------------------------------------------------------------------ */ diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S index 428c2cd2326c..e732981cf99f 100644 --- a/arch/mips/kernel/scall64-64.S +++ b/arch/mips/kernel/scall64-64.S @@ -39,18 +39,11 @@ NESTED(handle_sys64, PT_SIZE, sp) .set at #endif - dsubu t0, v0, __NR_64_Linux # check syscall number - sltiu t0, t0, __NR_64_Linux_syscalls + 1 #if !defined(CONFIG_MIPS32_O32) && !defined(CONFIG_MIPS32_N32) ld t1, PT_EPC(sp) # skip syscall on return daddiu t1, 4 # skip to next instruction sd t1, PT_EPC(sp) #endif - beqz t0, illegal_syscall - - dsll t0, v0, 3 # offset into table - ld t2, (sys_call_table - (__NR_64_Linux * 8))(t0) - # syscall routine sd a3, PT_R26(sp) # save a3 for syscall restarting @@ -59,6 +52,17 @@ NESTED(handle_sys64, PT_SIZE, sp) and t0, t1, t0 bnez t0, syscall_trace_entry +syscall_common: + dsubu t2, v0, __NR_64_Linux + sltiu t0, t2, __NR_64_Linux_syscalls + 1 + beqz t0, illegal_syscall + + dsll t0, t2, 3 # offset into table + dla t2, sys_call_table + daddu t0, t2, t0 + ld t2, (t0) # syscall routine + beqz t2, illegal_syscall + jalr t2 # Do The Real Thing (TM) li t0, -EMAXERRNO - 1 # error? @@ -78,14 +82,14 @@ n64_syscall_exit: syscall_trace_entry: SAVE_STATIC - move s0, t2 + move s0, v0 move a0, sp move a1, v0 jal syscall_trace_enter - bltz v0, 2f # seccomp failed? Skip syscall + bltz v0, 1f # seccomp failed? Skip syscall - move t0, s0 + move v0, s0 RESTORE_STATIC ld a0, PT_R4(sp) # Restore argument registers ld a1, PT_R5(sp) @@ -93,19 +97,9 @@ syscall_trace_entry: ld a3, PT_R7(sp) ld a4, PT_R8(sp) ld a5, PT_R9(sp) - jalr t0 + j syscall_common - li t0, -EMAXERRNO - 1 # error? - sltu t0, t0, v0 - sd t0, PT_R7(sp) # set error flag - beqz t0, 1f - - ld t1, PT_R2(sp) # syscall number - dnegu v0 # error - sd t1, PT_R0(sp) # save it for syscall restarting -1: sd v0, PT_R2(sp) # result - -2: j syscall_exit +1: j syscall_exit illegal_syscall: /* This also isn't a 64-bit syscall, throw an error. */ diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 3868cf4c83df..c79484397584 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -52,6 +52,7 @@ NESTED(handle_sysn32, PT_SIZE, sp) and t0, t1, t0 bnez t0, n32_syscall_trace_entry +syscall_common: jalr t2 # Do The Real Thing (TM) li t0, -EMAXERRNO - 1 # error? @@ -75,9 +76,9 @@ n32_syscall_trace_entry: move a1, v0 jal syscall_trace_enter - bltz v0, 2f # seccomp failed? Skip syscall + bltz v0, 1f # seccomp failed? Skip syscall - move t0, s0 + move t2, s0 RESTORE_STATIC ld a0, PT_R4(sp) # Restore argument registers ld a1, PT_R5(sp) @@ -85,19 +86,9 @@ n32_syscall_trace_entry: ld a3, PT_R7(sp) ld a4, PT_R8(sp) ld a5, PT_R9(sp) - jalr t0 + j syscall_common - li t0, -EMAXERRNO - 1 # error? - sltu t0, t0, v0 - sd t0, PT_R7(sp) # set error flag - beqz t0, 1f - - ld t1, PT_R2(sp) # syscall number - dnegu v0 # error - sd t1, PT_R0(sp) # save it for syscall restarting -1: sd v0, PT_R2(sp) # result - -2: j syscall_exit +1: j syscall_exit not_n32_scall: /* This is not an n32 compatibility syscall, pass it on to diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index 6622980d82ea..6369cfd390c6 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -87,6 +87,7 @@ loads_done: and t0, t1, t0 bnez t0, trace_a_syscall +syscall_common: jalr t2 # Do The Real Thing (TM) li t0, -EMAXERRNO - 1 # error? @@ -130,9 +131,9 @@ trace_a_syscall: 1: jal syscall_trace_enter - bltz v0, 2f # seccomp failed? Skip syscall + bltz v0, 1f # seccomp failed? Skip syscall - move t0, s0 + move t2, s0 RESTORE_STATIC ld a0, PT_R4(sp) # Restore argument registers ld a1, PT_R5(sp) @@ -142,19 +143,9 @@ trace_a_syscall: ld a5, PT_R9(sp) ld a6, PT_R10(sp) ld a7, PT_R11(sp) # For indirect syscalls - jalr t0 + j syscall_common - li t0, -EMAXERRNO - 1 # error? - sltu t0, t0, v0 - sd t0, PT_R7(sp) # set error flag - beqz t0, 1f - - ld t1, PT_R2(sp) # syscall number - dnegu v0 # error - sd t1, PT_R0(sp) # save it for syscall restarting -1: sd v0, PT_R2(sp) # result - -2: j syscall_exit +1: j syscall_exit /* ------------------------------------------------------------------------ */ From 049e6dde7e57f0054fdc49102e7ef4830c698b46 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 4 Oct 2015 16:57:17 +0100 Subject: [PATCH 647/647] Linux 4.3-rc4 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 1d341eba143d..fd46821e428d 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 4 PATCHLEVEL = 3 SUBLEVEL = 0 -EXTRAVERSION = -rc3 +EXTRAVERSION = -rc4 NAME = Hurr durr I'ma sheep # *DOCUMENTATION*