From 6c87df37dcb9c6c33923707fa5191e0a65874d60 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 27 Oct 2008 22:38:27 +0300 Subject: [PATCH 01/25] proc: revert /proc/uptime to ->read_proc hook Turned out some VMware userspace does pread(2) on /proc/uptime, but seqfiles currently don't allow pread() resulting in -ESPIPE. Seqfiles in theory can do pread(), but this can be a long story, so revert to ->read_proc until then. http://bugzilla.kernel.org/show_bug.cgi?id=11856 Signed-off-by: Alexey Dobriyan --- fs/proc/uptime.c | 38 ++++++++++++++++++++------------------ 1 file changed, 20 insertions(+), 18 deletions(-) diff --git a/fs/proc/uptime.c b/fs/proc/uptime.c index 0c10a0b3f146..df26aa88fa47 100644 --- a/fs/proc/uptime.c +++ b/fs/proc/uptime.c @@ -1,43 +1,45 @@ -#include #include #include #include -#include #include #include -static int uptime_proc_show(struct seq_file *m, void *v) +static int proc_calc_metrics(char *page, char **start, off_t off, + int count, int *eof, int len) +{ + if (len <= off + count) + *eof = 1; + *start = page + off; + len -= off; + if (len > count) + len = count; + if (len < 0) + len = 0; + return len; +} + +static int uptime_read_proc(char *page, char **start, off_t off, int count, + int *eof, void *data) { struct timespec uptime; struct timespec idle; + int len; cputime_t idletime = cputime_add(init_task.utime, init_task.stime); do_posix_clock_monotonic_gettime(&uptime); monotonic_to_bootbased(&uptime); cputime_to_timespec(idletime, &idle); - seq_printf(m, "%lu.%02lu %lu.%02lu\n", + len = sprintf(page, "%lu.%02lu %lu.%02lu\n", (unsigned long) uptime.tv_sec, (uptime.tv_nsec / (NSEC_PER_SEC / 100)), (unsigned long) idle.tv_sec, (idle.tv_nsec / (NSEC_PER_SEC / 100))); - return 0; + return proc_calc_metrics(page, start, off, count, eof, len); } -static int uptime_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, uptime_proc_show, NULL); -} - -static const struct file_operations uptime_proc_fops = { - .open = uptime_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static int __init proc_uptime_init(void) { - proc_create("uptime", 0, NULL, &uptime_proc_fops); + create_proc_read_entry("uptime", 0, NULL, uptime_read_proc, NULL); return 0; } module_init(proc_uptime_init); From 65e082c9a33a6e9f24e9a713a7d38d11206d3c3d Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 24 Oct 2008 17:18:10 -0400 Subject: [PATCH 02/25] build fix: CONFIG_DRM_I915=y && CONFIG_ACPI=n MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit drivers/gpu/drm/i915/i915_opregion.c:340: error: implicit declaration of function ‘register_acpi_notifier’ drivers/gpu/drm/i915/i915_opregion.c:361: error: implicit declaration of function ‘unregister_acpi_notifier’ Signed-off-by: Len Brown Signed-off-by: Eric Anholt Signed-off-by: Dave Airlie --- drivers/gpu/drm/i915/Makefile | 3 ++- drivers/gpu/drm/i915/i915_drv.h | 7 +++++++ 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 5ba78e4fd2b5..d8fb5d8ee7ea 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -3,13 +3,14 @@ # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher. ccflags-y := -Iinclude/drm -i915-y := i915_drv.o i915_dma.o i915_irq.o i915_mem.o i915_opregion.o \ +i915-y := i915_drv.o i915_dma.o i915_irq.o i915_mem.o \ i915_suspend.o \ i915_gem.o \ i915_gem_debug.o \ i915_gem_proc.o \ i915_gem_tiling.o +i915-$(CONFIG_ACPI) += i915_opregion.o i915-$(CONFIG_COMPAT) += i915_ioc32.o obj-$(CONFIG_DRM_I915) += i915.o diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index f20ffe17df71..901e80cf5813 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -539,11 +539,18 @@ extern int i915_restore_state(struct drm_device *dev); extern int i915_save_state(struct drm_device *dev); extern int i915_restore_state(struct drm_device *dev); +#ifdef CONFIG_ACPI /* i915_opregion.c */ extern int intel_opregion_init(struct drm_device *dev); extern void intel_opregion_free(struct drm_device *dev); extern void opregion_asle_intr(struct drm_device *dev); extern void opregion_enable_asle(struct drm_device *dev); +#else +static inline int intel_opregion_init(struct drm_device *dev) { return 0; } +static inline void intel_opregion_free(struct drm_device *dev) { return; } +static inline void opregion_asle_intr(struct drm_device *dev) { return; } +static inline void opregion_enable_asle(struct drm_device *dev) { return; } +#endif /** * Lock test for when it's just for synchronization of ring access. From 4e270e9b8a9d246290f3901f1fb6c5efdb734ddf Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 28 Oct 2008 07:48:34 +1000 Subject: [PATCH 03/25] drm/radeon: fixup further bus mastering confusion. rs400/480 are like previous chips not like rs6xx chips. Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_cp.c | 15 ++++++++------- drivers/gpu/drm/radeon/radeon_drv.h | 12 ++++++------ 2 files changed, 14 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_cp.c b/drivers/gpu/drm/radeon/radeon_cp.c index 59a2132a8f57..073894824e6b 100644 --- a/drivers/gpu/drm/radeon/radeon_cp.c +++ b/drivers/gpu/drm/radeon/radeon_cp.c @@ -653,15 +653,16 @@ static void radeon_cp_init_ring_buffer(struct drm_device * dev, RADEON_WRITE(RADEON_SCRATCH_UMSK, 0x7); /* Turn on bus mastering */ - if (((dev_priv->flags & RADEON_FAMILY_MASK) == CHIP_RS400) || - ((dev_priv->flags & RADEON_FAMILY_MASK) == CHIP_RS690) || + if (((dev_priv->flags & RADEON_FAMILY_MASK) == CHIP_RS690) || ((dev_priv->flags & RADEON_FAMILY_MASK) == CHIP_RS740)) { - /* rs400, rs690/rs740 */ - tmp = RADEON_READ(RADEON_BUS_CNTL) & ~RS400_BUS_MASTER_DIS; + /* rs600/rs690/rs740 */ + tmp = RADEON_READ(RADEON_BUS_CNTL) & ~RS600_BUS_MASTER_DIS; RADEON_WRITE(RADEON_BUS_CNTL, tmp); - } else if (!(((dev_priv->flags & RADEON_FAMILY_MASK) == CHIP_RV380) || - ((dev_priv->flags & RADEON_FAMILY_MASK) >= CHIP_R423))) { - /* r1xx, r2xx, r300, r(v)350, r420/r481, rs480 */ + } else if (((dev_priv->flags & RADEON_FAMILY_MASK) <= CHIP_RV350) || + ((dev_priv->flags & RADEON_FAMILY_MASK) == CHIP_R420) || + ((dev_priv->flags & RADEON_FAMILY_MASK) == CHIP_RS400) || + ((dev_priv->flags & RADEON_FAMILY_MASK) == CHIP_RS480)) { + /* r1xx, r2xx, r300, r(v)350, r420/r481, rs400/rs480 */ tmp = RADEON_READ(RADEON_BUS_CNTL) & ~RADEON_BUS_MASTER_DIS; RADEON_WRITE(RADEON_BUS_CNTL, tmp); } /* PCIE cards appears to not need this */ diff --git a/drivers/gpu/drm/radeon/radeon_drv.h b/drivers/gpu/drm/radeon/radeon_drv.h index 4dbb813910c3..02f5575ba395 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.h +++ b/drivers/gpu/drm/radeon/radeon_drv.h @@ -447,12 +447,12 @@ extern int r300_do_cp_cmdbuf(struct drm_device *dev, * handling, not bus mastering itself. */ #define RADEON_BUS_CNTL 0x0030 -/* r1xx, r2xx, r300, r(v)350, r420/r481, rs480 */ +/* r1xx, r2xx, r300, r(v)350, r420/r481, rs400/rs480 */ # define RADEON_BUS_MASTER_DIS (1 << 6) -/* rs400, rs690/rs740 */ -# define RS400_BUS_MASTER_DIS (1 << 14) -# define RS400_MSI_REARM (1 << 20) -/* see RS480_MSI_REARM in AIC_CNTL for rs480 */ +/* rs600/rs690/rs740 */ +# define RS600_BUS_MASTER_DIS (1 << 14) +# define RS600_MSI_REARM (1 << 20) +/* see RS400_MSI_REARM in AIC_CNTL for rs480 */ #define RADEON_BUS_CNTL1 0x0034 # define RADEON_PMI_BM_DIS (1 << 2) @@ -937,7 +937,7 @@ extern int r300_do_cp_cmdbuf(struct drm_device *dev, #define RADEON_AIC_CNTL 0x01d0 # define RADEON_PCIGART_TRANSLATE_EN (1 << 0) -# define RS480_MSI_REARM (1 << 3) +# define RS400_MSI_REARM (1 << 3) #define RADEON_AIC_STAT 0x01d4 #define RADEON_AIC_PT_BASE 0x01d8 #define RADEON_AIC_LO_ADDR 0x01dc From fd9409343521eac22b6ed51686128a643c7c976b Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Thu, 30 Oct 2008 19:37:09 -0700 Subject: [PATCH 04/25] x86: add iomap_atomic*()/iounmap_atomic() on 32-bit using fixmaps Impact: introduce new APIs, separate kmap code from CONFIG_HIGHMEM This takes the code used for CONFIG_HIGHMEM memory mappings except that it's designed for dynamic IO resource mapping. These fixmaps are available even with CONFIG_HIGHMEM turned off. Signed-off-by: Keith Packard Signed-off-by: Eric Anholt Signed-off-by: Ingo Molnar --- arch/x86/include/asm/fixmap.h | 4 +++ arch/x86/include/asm/fixmap_32.h | 4 --- arch/x86/include/asm/highmem.h | 5 +-- arch/x86/mm/Makefile | 2 +- arch/x86/mm/init_32.c | 3 +- arch/x86/mm/iomap_32.c | 59 ++++++++++++++++++++++++++++++++ include/asm-x86/iomap.h | 30 ++++++++++++++++ 7 files changed, 96 insertions(+), 11 deletions(-) create mode 100644 arch/x86/mm/iomap_32.c create mode 100644 include/asm-x86/iomap.h diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h index 8668a94f850e..23696d44a0af 100644 --- a/arch/x86/include/asm/fixmap.h +++ b/arch/x86/include/asm/fixmap.h @@ -9,6 +9,10 @@ extern int fixmaps_set; +extern pte_t *kmap_pte; +extern pgprot_t kmap_prot; +extern pte_t *pkmap_page_table; + void __native_set_fixmap(enum fixed_addresses idx, pte_t pte); void native_set_fixmap(enum fixed_addresses idx, unsigned long phys, pgprot_t flags); diff --git a/arch/x86/include/asm/fixmap_32.h b/arch/x86/include/asm/fixmap_32.h index 09f29ab5c139..c7115c1d7217 100644 --- a/arch/x86/include/asm/fixmap_32.h +++ b/arch/x86/include/asm/fixmap_32.h @@ -28,10 +28,8 @@ extern unsigned long __FIXADDR_TOP; #include #include #include -#ifdef CONFIG_HIGHMEM #include #include -#endif /* * Here we define all the compile-time 'special' virtual @@ -75,10 +73,8 @@ enum fixed_addresses { #ifdef CONFIG_X86_CYCLONE_TIMER FIX_CYCLONE_TIMER, /*cyclone timer register*/ #endif -#ifdef CONFIG_HIGHMEM FIX_KMAP_BEGIN, /* reserved pte's for temporary kernel mappings */ FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1, -#endif #ifdef CONFIG_PCI_MMCONFIG FIX_PCIE_MCFG, #endif diff --git a/arch/x86/include/asm/highmem.h b/arch/x86/include/asm/highmem.h index a3b3b7c3027b..bf9276bea660 100644 --- a/arch/x86/include/asm/highmem.h +++ b/arch/x86/include/asm/highmem.h @@ -25,14 +25,11 @@ #include #include #include +#include /* declarations for highmem.c */ extern unsigned long highstart_pfn, highend_pfn; -extern pte_t *kmap_pte; -extern pgprot_t kmap_prot; -extern pte_t *pkmap_page_table; - /* * Right now we initialize only a single pte table. It can be extended * easily, subsequent pte tables have to be allocated in one physical diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index 59f89b434b45..fea4565ff576 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -1,7 +1,7 @@ obj-y := init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \ pat.o pgtable.o gup.o -obj-$(CONFIG_X86_32) += pgtable_32.o +obj-$(CONFIG_X86_32) += pgtable_32.o iomap_32.o obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o obj-$(CONFIG_X86_PTDUMP) += dump_pagetables.o diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 8396868e82c5..c483f4242079 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -334,7 +334,6 @@ int devmem_is_allowed(unsigned long pagenr) return 0; } -#ifdef CONFIG_HIGHMEM pte_t *kmap_pte; pgprot_t kmap_prot; @@ -357,6 +356,7 @@ static void __init kmap_init(void) kmap_prot = PAGE_KERNEL; } +#ifdef CONFIG_HIGHMEM static void __init permanent_kmaps_init(pgd_t *pgd_base) { unsigned long vaddr; @@ -436,7 +436,6 @@ static void __init set_highmem_pages_init(void) #endif /* !CONFIG_NUMA */ #else -# define kmap_init() do { } while (0) # define permanent_kmaps_init(pgd_base) do { } while (0) # define set_highmem_pages_init() do { } while (0) #endif /* CONFIG_HIGHMEM */ diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c new file mode 100644 index 000000000000..d0151d8ce452 --- /dev/null +++ b/arch/x86/mm/iomap_32.c @@ -0,0 +1,59 @@ +/* + * Copyright © 2008 Ingo Molnar + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * 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. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA. + */ + +#include +#include + +/* Map 'pfn' using fixed map 'type' and protections 'prot' + */ +void * +iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot) +{ + enum fixed_addresses idx; + unsigned long vaddr; + + pagefault_disable(); + + idx = type + KM_TYPE_NR*smp_processor_id(); + vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); + set_pte(kmap_pte-idx, pfn_pte(pfn, prot)); + arch_flush_lazy_mmu_mode(); + + return (void*) vaddr; +} +EXPORT_SYMBOL_GPL(iomap_atomic_prot_pfn); + +void +iounmap_atomic(void *kvaddr, enum km_type type) +{ + unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK; + enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id(); + + /* + * Force other mappings to Oops if they'll try to access this pte + * without first remap it. Keeping stale mappings around is a bad idea + * also, in case the page changes cacheability attributes or becomes + * a protected page in a hypervisor. + */ + if (vaddr == __fix_to_virt(FIX_KMAP_BEGIN+idx)) + kpte_clear_flush(kmap_pte-idx, vaddr); + + arch_flush_lazy_mmu_mode(); + pagefault_enable(); +} +EXPORT_SYMBOL_GPL(iounmap_atomic); diff --git a/include/asm-x86/iomap.h b/include/asm-x86/iomap.h new file mode 100644 index 000000000000..c1f06289b14b --- /dev/null +++ b/include/asm-x86/iomap.h @@ -0,0 +1,30 @@ +/* + * Copyright © 2008 Ingo Molnar + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * 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. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA. + */ + +#include +#include +#include +#include +#include +#include + +void * +iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot); + +void +iounmap_atomic(void *kvaddr, enum km_type type); From 9663f2e6a6cf3f82b06d8fb699b11b80f92553ba Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Thu, 30 Oct 2008 19:38:18 -0700 Subject: [PATCH 05/25] resources: add io-mapping functions to dynamically map large device apertures Impact: add new generic io_map_*() APIs Graphics devices have large PCI apertures which would consume a significant fraction of a 32-bit address space if mapped during driver initialization. Using ioremap at runtime is impractical as it is too slow. This new set of interfaces uses atomic mappings on 32-bit processors and a large static mapping on 64-bit processors to provide reasonable 32-bit performance and optimal 64-bit performance. The current implementation sits atop the io_map_atomic fixmap-based mechanism for 32-bit processors. This includes some editorial suggestions from Randy Dunlap for Documentation/io-mapping.txt Signed-off-by: Keith Packard Signed-off-by: Eric Anholt Signed-off-by: Ingo Molnar --- Documentation/io-mapping.txt | 76 ++++++++++++++++++++++ include/linux/io-mapping.h | 118 +++++++++++++++++++++++++++++++++++ 2 files changed, 194 insertions(+) create mode 100644 Documentation/io-mapping.txt create mode 100644 include/linux/io-mapping.h diff --git a/Documentation/io-mapping.txt b/Documentation/io-mapping.txt new file mode 100644 index 000000000000..cd2f726becc8 --- /dev/null +++ b/Documentation/io-mapping.txt @@ -0,0 +1,76 @@ +The io_mapping functions in linux/io-mapping.h provide an abstraction for +efficiently mapping small regions of an I/O device to the CPU. The initial +usage is to support the large graphics aperture on 32-bit processors where +ioremap_wc cannot be used to statically map the entire aperture to the CPU +as it would consume too much of the kernel address space. + +A mapping object is created during driver initialization using + + struct io_mapping *io_mapping_create_wc(unsigned long base, + unsigned long size) + + 'base' is the bus address of the region to be made + mappable, while 'size' indicates how large a mapping region to + enable. Both are in bytes. + + This _wc variant provides a mapping which may only be used + with the io_mapping_map_atomic_wc or io_mapping_map_wc. + +With this mapping object, individual pages can be mapped either atomically +or not, depending on the necessary scheduling environment. Of course, atomic +maps are more efficient: + + void *io_mapping_map_atomic_wc(struct io_mapping *mapping, + unsigned long offset) + + 'offset' is the offset within the defined mapping region. + Accessing addresses beyond the region specified in the + creation function yields undefined results. Using an offset + which is not page aligned yields an undefined result. The + return value points to a single page in CPU address space. + + This _wc variant returns a write-combining map to the + page and may only be used with mappings created by + io_mapping_create_wc + + Note that the task may not sleep while holding this page + mapped. + + void io_mapping_unmap_atomic(void *vaddr) + + 'vaddr' must be the the value returned by the last + io_mapping_map_atomic_wc call. This unmaps the specified + page and allows the task to sleep once again. + +If you need to sleep while holding the lock, you can use the non-atomic +variant, although they may be significantly slower. + + void *io_mapping_map_wc(struct io_mapping *mapping, + unsigned long offset) + + This works like io_mapping_map_atomic_wc except it allows + the task to sleep while holding the page mapped. + + void io_mapping_unmap(void *vaddr) + + This works like io_mapping_unmap_atomic, except it is used + for pages mapped with io_mapping_map_wc. + +At driver close time, the io_mapping object must be freed: + + void io_mapping_free(struct io_mapping *mapping) + +Current Implementation: + +The initial implementation of these functions uses existing mapping +mechanisms and so provides only an abstraction layer and no new +functionality. + +On 64-bit processors, io_mapping_create_wc calls ioremap_wc for the whole +range, creating a permanent kernel-visible mapping to the resource. The +map_atomic and map functions add the requested offset to the base of the +virtual address returned by ioremap_wc. + +On 32-bit processors, io_mapping_map_atomic_wc uses io_map_atomic_prot_pfn, +which uses the fixmaps to get us a mapping to a page using an atomic fashion. +For io_mapping_map_wc, ioremap_wc() is used to get a mapping of the region. diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h new file mode 100644 index 000000000000..1b566993db6e --- /dev/null +++ b/include/linux/io-mapping.h @@ -0,0 +1,118 @@ +/* + * Copyright © 2008 Keith Packard + * + * This file is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License + * 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. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA. + */ + +#ifndef _LINUX_IO_MAPPING_H +#define _LINUX_IO_MAPPING_H + +#include +#include +#include +#include + +/* + * The io_mapping mechanism provides an abstraction for mapping + * individual pages from an io device to the CPU in an efficient fashion. + * + * See Documentation/io_mapping.txt + */ + +/* this struct isn't actually defined anywhere */ +struct io_mapping; + +#ifdef CONFIG_X86_64 + +/* Create the io_mapping object*/ +static inline struct io_mapping * +io_mapping_create_wc(unsigned long base, unsigned long size) +{ + return (struct io_mapping *) ioremap_wc(base, size); +} + +static inline void +io_mapping_free(struct io_mapping *mapping) +{ + iounmap(mapping); +} + +/* Atomic map/unmap */ +static inline void * +io_mapping_map_atomic_wc(struct io_mapping *mapping, unsigned long offset) +{ + return ((char *) mapping) + offset; +} + +static inline void +io_mapping_unmap_atomic(void *vaddr) +{ +} + +/* Non-atomic map/unmap */ +static inline void * +io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) +{ + return ((char *) mapping) + offset; +} + +static inline void +io_mapping_unmap(void *vaddr) +{ +} + +#endif /* CONFIG_X86_64 */ + +#ifdef CONFIG_X86_32 +static inline struct io_mapping * +io_mapping_create_wc(unsigned long base, unsigned long size) +{ + return (struct io_mapping *) base; +} + +static inline void +io_mapping_free(struct io_mapping *mapping) +{ +} + +/* Atomic map/unmap */ +static inline void * +io_mapping_map_atomic_wc(struct io_mapping *mapping, unsigned long offset) +{ + offset += (unsigned long) mapping; + return iomap_atomic_prot_pfn(offset >> PAGE_SHIFT, KM_USER0, + __pgprot(__PAGE_KERNEL_WC)); +} + +static inline void +io_mapping_unmap_atomic(void *vaddr) +{ + iounmap_atomic(vaddr, KM_USER0); +} + +static inline void * +io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) +{ + offset += (unsigned long) mapping; + return ioremap_wc(offset, PAGE_SIZE); +} + +static inline void +io_mapping_unmap(void *vaddr) +{ + iounmap(vaddr); +} +#endif /* CONFIG_X86_32 */ + +#endif /* _LINUX_IO_MAPPING_H */ From 0839ccb8ac6a9e2d5e175a4ae9c82b5c574d510d Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Thu, 30 Oct 2008 19:38:48 -0700 Subject: [PATCH 06/25] i915: use io-mapping interfaces instead of a variety of mapping kludges Impact: optimize/clean-up the IO mapping implementation of the i915 DRM driver Switch the i915 device aperture mapping to the io-mapping interface, taking advantage of the cleaner API to extend it across all of the mapping uses, including both pwrite and relocation updates. This dramatically improves performance on 64-bit kernels which were using the same slow path as 32-bit non-HIGHMEM kernels prior to this patch. Signed-off-by: Keith Packard Signed-off-by: Eric Anholt Signed-off-by: Ingo Molnar --- drivers/gpu/drm/i915/i915_drv.h | 3 + drivers/gpu/drm/i915/i915_gem.c | 174 +++++++++++++++----------------- 2 files changed, 83 insertions(+), 94 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index f20ffe17df71..126b2f13258c 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -31,6 +31,7 @@ #define _I915_DRV_H_ #include "i915_reg.h" +#include /* General customization: */ @@ -246,6 +247,8 @@ typedef struct drm_i915_private { struct { struct drm_mm gtt_space; + struct io_mapping *gtt_mapping; + /** * List of objects currently involved in rendering from the * ringbuffer. diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 17ae330ff269..61183b95b108 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -171,35 +171,50 @@ i915_gem_pread_ioctl(struct drm_device *dev, void *data, return 0; } -/* - * Try to write quickly with an atomic kmap. Return true on success. - * - * If this fails (which includes a partial write), we'll redo the whole - * thing with the slow version. - * - * This is a workaround for the low performance of iounmap (approximate - * 10% cpu cost on normal 3D workloads). kmap_atomic on HIGHMEM kernels - * happens to let us map card memory without taking IPIs. When the vmap - * rework lands we should be able to dump this hack. +/* This is the fast write path which cannot handle + * page faults in the source data */ -static inline int fast_user_write(unsigned long pfn, char __user *user_data, - int l, int o) -{ -#ifdef CONFIG_HIGHMEM - unsigned long unwritten; - char *vaddr_atomic; - vaddr_atomic = kmap_atomic_pfn(pfn, KM_USER0); -#if WATCH_PWRITE - DRM_INFO("pwrite i %d o %d l %d pfn %ld vaddr %p\n", - i, o, l, pfn, vaddr_atomic); -#endif - unwritten = __copy_from_user_inatomic_nocache(vaddr_atomic + o, user_data, l); - kunmap_atomic(vaddr_atomic, KM_USER0); - return !unwritten; -#else +static inline int +fast_user_write(struct io_mapping *mapping, + loff_t page_base, int page_offset, + char __user *user_data, + int length) +{ + char *vaddr_atomic; + unsigned long unwritten; + + vaddr_atomic = io_mapping_map_atomic_wc(mapping, page_base); + unwritten = __copy_from_user_inatomic_nocache(vaddr_atomic + page_offset, + user_data, length); + io_mapping_unmap_atomic(vaddr_atomic); + if (unwritten) + return -EFAULT; + return 0; +} + +/* Here's the write path which can sleep for + * page faults + */ + +static inline int +slow_user_write(struct io_mapping *mapping, + loff_t page_base, int page_offset, + char __user *user_data, + int length) +{ + char __iomem *vaddr; + unsigned long unwritten; + + vaddr = io_mapping_map_wc(mapping, page_base); + if (vaddr == NULL) + return -EFAULT; + unwritten = __copy_from_user(vaddr + page_offset, + user_data, length); + io_mapping_unmap(vaddr); + if (unwritten) + return -EFAULT; return 0; -#endif } static int @@ -208,10 +223,12 @@ i915_gem_gtt_pwrite(struct drm_device *dev, struct drm_gem_object *obj, struct drm_file *file_priv) { struct drm_i915_gem_object *obj_priv = obj->driver_private; + drm_i915_private_t *dev_priv = dev->dev_private; ssize_t remain; - loff_t offset; + loff_t offset, page_base; char __user *user_data; - int ret = 0; + int page_offset, page_length; + int ret; user_data = (char __user *) (uintptr_t) args->data_ptr; remain = args->size; @@ -235,57 +252,37 @@ i915_gem_gtt_pwrite(struct drm_device *dev, struct drm_gem_object *obj, obj_priv->dirty = 1; while (remain > 0) { - unsigned long pfn; - int i, o, l; - /* Operation in this page * - * i = page number - * o = offset within page - * l = bytes to copy + * page_base = page offset within aperture + * page_offset = offset within page + * page_length = bytes to copy for this page */ - i = offset >> PAGE_SHIFT; - o = offset & (PAGE_SIZE-1); - l = remain; - if ((o + l) > PAGE_SIZE) - l = PAGE_SIZE - o; + page_base = (offset & ~(PAGE_SIZE-1)); + page_offset = offset & (PAGE_SIZE-1); + page_length = remain; + if ((page_offset + remain) > PAGE_SIZE) + page_length = PAGE_SIZE - page_offset; - pfn = (dev->agp->base >> PAGE_SHIFT) + i; + ret = fast_user_write (dev_priv->mm.gtt_mapping, page_base, + page_offset, user_data, page_length); - if (!fast_user_write(pfn, user_data, l, o)) { - unsigned long unwritten; - char __iomem *vaddr; - - vaddr = ioremap_wc(pfn << PAGE_SHIFT, PAGE_SIZE); -#if WATCH_PWRITE - DRM_INFO("pwrite slow i %d o %d l %d " - "pfn %ld vaddr %p\n", - i, o, l, pfn, vaddr); -#endif - if (vaddr == NULL) { - ret = -EFAULT; + /* If we get a fault while copying data, then (presumably) our + * source page isn't available. In this case, use the + * non-atomic function + */ + if (ret) { + ret = slow_user_write (dev_priv->mm.gtt_mapping, + page_base, page_offset, + user_data, page_length); + if (ret) goto fail; - } - unwritten = __copy_from_user(vaddr + o, user_data, l); -#if WATCH_PWRITE - DRM_INFO("unwritten %ld\n", unwritten); -#endif - iounmap(vaddr); - if (unwritten) { - ret = -EFAULT; - goto fail; - } } - remain -= l; - user_data += l; - offset += l; + remain -= page_length; + user_data += page_length; + offset += page_length; } -#if WATCH_PWRITE && 1 - i915_gem_clflush_object(obj); - i915_gem_dump_object(obj, args->offset + args->size, __func__, ~0); - i915_gem_clflush_object(obj); -#endif fail: i915_gem_object_unpin(obj); @@ -1503,12 +1500,12 @@ i915_gem_object_pin_and_relocate(struct drm_gem_object *obj, struct drm_i915_gem_exec_object *entry) { struct drm_device *dev = obj->dev; + drm_i915_private_t *dev_priv = dev->dev_private; struct drm_i915_gem_relocation_entry reloc; struct drm_i915_gem_relocation_entry __user *relocs; struct drm_i915_gem_object *obj_priv = obj->driver_private; int i, ret; - uint32_t last_reloc_offset = -1; - void __iomem *reloc_page = NULL; + void __iomem *reloc_page; /* Choose the GTT offset for our buffer and put it there. */ ret = i915_gem_object_pin(obj, (uint32_t) entry->alignment); @@ -1631,26 +1628,11 @@ i915_gem_object_pin_and_relocate(struct drm_gem_object *obj, * perform. */ reloc_offset = obj_priv->gtt_offset + reloc.offset; - if (reloc_page == NULL || - (last_reloc_offset & ~(PAGE_SIZE - 1)) != - (reloc_offset & ~(PAGE_SIZE - 1))) { - if (reloc_page != NULL) - iounmap(reloc_page); - - reloc_page = ioremap_wc(dev->agp->base + - (reloc_offset & - ~(PAGE_SIZE - 1)), - PAGE_SIZE); - last_reloc_offset = reloc_offset; - if (reloc_page == NULL) { - drm_gem_object_unreference(target_obj); - i915_gem_object_unpin(obj); - return -ENOMEM; - } - } - + reloc_page = io_mapping_map_atomic_wc(dev_priv->mm.gtt_mapping, + (reloc_offset & + ~(PAGE_SIZE - 1))); reloc_entry = (uint32_t __iomem *)(reloc_page + - (reloc_offset & (PAGE_SIZE - 1))); + (reloc_offset & (PAGE_SIZE - 1))); reloc_val = target_obj_priv->gtt_offset + reloc.delta; #if WATCH_BUF @@ -1659,6 +1641,7 @@ i915_gem_object_pin_and_relocate(struct drm_gem_object *obj, readl(reloc_entry), reloc_val); #endif writel(reloc_val, reloc_entry); + io_mapping_unmap_atomic(reloc_page); /* Write the updated presumed offset for this entry back out * to the user. @@ -1674,9 +1657,6 @@ i915_gem_object_pin_and_relocate(struct drm_gem_object *obj, drm_gem_object_unreference(target_obj); } - if (reloc_page != NULL) - iounmap(reloc_page); - #if WATCH_BUF if (0) i915_gem_dump_object(obj, 128, __func__, ~0); @@ -2518,6 +2498,10 @@ i915_gem_entervt_ioctl(struct drm_device *dev, void *data, if (ret != 0) return ret; + dev_priv->mm.gtt_mapping = io_mapping_create_wc(dev->agp->base, + dev->agp->agp_info.aper_size + * 1024 * 1024); + mutex_lock(&dev->struct_mutex); BUG_ON(!list_empty(&dev_priv->mm.active_list)); BUG_ON(!list_empty(&dev_priv->mm.flushing_list)); @@ -2535,11 +2519,13 @@ int i915_gem_leavevt_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) { + drm_i915_private_t *dev_priv = dev->dev_private; int ret; ret = i915_gem_idle(dev); drm_irq_uninstall(dev); + io_mapping_free(dev_priv->mm.gtt_mapping); return ret; } From 4605b718e8b8f0dd3d811ddf90f630fd0835b7bf Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 31 Oct 2008 14:18:24 +0100 Subject: [PATCH 07/25] ALSA: hda - Disable broken mic auto-muting in Realtek codes The recent addition of automatic mic-muting is broken in some cases. The code assumes that the pin nids <= 0x18, but the digital pins can be less than 0x18. Also, it assumes the front-mic being the internal mic, but it depends on the hardware implementation actually. Instead of complex case-fixes, better to disable the code as now. Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 4eceab9bd109..a80d57cbc352 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -829,6 +829,7 @@ static void alc_sku_automute(struct hda_codec *codec) spec->jack_present ? 0 : PIN_OUT); } +#if 0 /* it's broken in some acses -- temporarily disabled */ static void alc_mic_automute(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; @@ -849,6 +850,9 @@ static void alc_mic_automute(struct hda_codec *codec) snd_hda_codec_amp_stereo(codec, 0x0b, HDA_INPUT, capsrc_idx_fmic, HDA_AMP_MUTE, present ? HDA_AMP_MUTE : 0); } +#else +#define alc_mic_automute(codec) /* NOP */ +#endif /* disabled */ /* unsolicited event for HP jack sensing */ static void alc_sku_unsol_event(struct hda_codec *codec, unsigned int res) @@ -1058,12 +1062,14 @@ do_sku: AC_VERB_SET_UNSOLICITED_ENABLE, AC_USRSP_EN | ALC880_HP_EVENT); +#if 0 /* it's broken in some acses -- temporarily disabled */ if (spec->autocfg.input_pins[AUTO_PIN_MIC] && spec->autocfg.input_pins[AUTO_PIN_FRONT_MIC]) snd_hda_codec_write(codec, spec->autocfg.input_pins[AUTO_PIN_MIC], 0, AC_VERB_SET_UNSOLICITED_ENABLE, AC_USRSP_EN | ALC880_MIC_EVENT); +#endif /* disabled */ spec->unsol_event = alc_sku_unsol_event; } From ee956e090e114ede6542c76a7465ed6ce118ad52 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 31 Oct 2008 17:16:31 +0100 Subject: [PATCH 08/25] ALSA: hda - Add digital-mic for ALC269 auto-probe mode The digital mic wasn't detected properly for ALC269 auto-probing mode because of its widget number. Fixed now. Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index a80d57cbc352..d327a371595c 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -12244,8 +12244,26 @@ static int alc269_auto_create_multi_out_ctls(struct alc_spec *spec, return 0; } -#define alc269_auto_create_analog_input_ctls \ - alc880_auto_create_analog_input_ctls +static int alc269_auto_create_analog_input_ctls(struct alc_spec *spec, + const struct auto_pin_cfg *cfg) +{ + int err; + + err = alc880_auto_create_analog_input_ctls(spec, cfg); + if (err < 0) + return err; + /* digital-mic input pin is excluded in alc880_auto_create..() + * because it's under 0x18 + */ + if (cfg->input_pins[AUTO_PIN_MIC] == 0x12 || + cfg->input_pins[AUTO_PIN_FRONT_MIC] == 0x12) { + struct hda_input_mux *imux = &spec->private_imux; + imux->items[imux->num_items].label = "Int Mic"; + imux->items[imux->num_items].index = 0x05; + imux->num_items++; + } + return 0; +} #ifdef CONFIG_SND_HDA_POWER_SAVE #define alc269_loopbacks alc880_loopbacks From 5a125c3c79167e78ba44efef03af7090ef28eeaf Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Wed, 22 Oct 2008 21:40:13 -0700 Subject: [PATCH 09/25] i915: Add GEM ioctl to get available aperture size. This will let userland know when to submit its batchbuffers, before they get too big to fit in the aperture. Signed-off-by: Eric Anholt Signed-off-by: Dave Airlie --- drivers/gpu/drm/i915/i915_dma.c | 1 + drivers/gpu/drm/i915/i915_drv.h | 2 ++ drivers/gpu/drm/i915/i915_gem.c | 22 ++++++++++++++++++++++ include/drm/i915_drm.h | 13 +++++++++++++ 4 files changed, 38 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 01de536e0211..256e22963ae4 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -960,6 +960,7 @@ struct drm_ioctl_desc i915_ioctls[] = { DRM_IOCTL_DEF(DRM_I915_GEM_SW_FINISH, i915_gem_sw_finish_ioctl, 0), DRM_IOCTL_DEF(DRM_I915_GEM_SET_TILING, i915_gem_set_tiling, 0), DRM_IOCTL_DEF(DRM_I915_GEM_GET_TILING, i915_gem_get_tiling, 0), + DRM_IOCTL_DEF(DRM_I915_GEM_GET_APERTURE, i915_gem_get_aperture_ioctl, 0), }; int i915_max_ioctl = DRM_ARRAY_SIZE(i915_ioctls); diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 901e80cf5813..cc8a9f3f7a60 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -502,6 +502,8 @@ int i915_gem_set_tiling(struct drm_device *dev, void *data, struct drm_file *file_priv); int i915_gem_get_tiling(struct drm_device *dev, void *data, struct drm_file *file_priv); +int i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); void i915_gem_load(struct drm_device *dev); int i915_gem_proc_init(struct drm_minor *minor); void i915_gem_proc_cleanup(struct drm_minor *minor); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 17ae330ff269..c1733ac4a7f5 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -79,6 +79,28 @@ i915_gem_init_ioctl(struct drm_device *dev, void *data, return 0; } +int +i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + drm_i915_private_t *dev_priv = dev->dev_private; + struct drm_i915_gem_get_aperture *args = data; + struct drm_i915_gem_object *obj_priv; + + if (!(dev->driver->driver_features & DRIVER_GEM)) + return -ENODEV; + + args->aper_size = dev->gtt_total; + args->aper_available_size = args->aper_size; + + list_for_each_entry(obj_priv, &dev_priv->mm.active_list, list) { + if (obj_priv->pin_count > 0) + args->aper_available_size -= obj_priv->obj->size; + } + + return 0; +} + /** * Creates a new mm object and returns a handle to it. diff --git a/include/drm/i915_drm.h b/include/drm/i915_drm.h index eb4b35031a55..152b34da927c 100644 --- a/include/drm/i915_drm.h +++ b/include/drm/i915_drm.h @@ -159,6 +159,7 @@ typedef struct _drm_i915_sarea { #define DRM_I915_GEM_SW_FINISH 0x20 #define DRM_I915_GEM_SET_TILING 0x21 #define DRM_I915_GEM_GET_TILING 0x22 +#define DRM_I915_GEM_GET_APERTURE 0x23 #define DRM_IOCTL_I915_INIT DRM_IOW( DRM_COMMAND_BASE + DRM_I915_INIT, drm_i915_init_t) #define DRM_IOCTL_I915_FLUSH DRM_IO ( DRM_COMMAND_BASE + DRM_I915_FLUSH) @@ -190,6 +191,7 @@ typedef struct _drm_i915_sarea { #define DRM_IOCTL_I915_GEM_SW_FINISH DRM_IOW (DRM_COMMAND_BASE + DRM_I915_GEM_SW_FINISH, struct drm_i915_gem_sw_finish) #define DRM_IOCTL_I915_GEM_SET_TILING DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_GEM_SET_TILING, struct drm_i915_gem_set_tiling) #define DRM_IOCTL_I915_GEM_GET_TILING DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_GEM_GET_TILING, struct drm_i915_gem_get_tiling) +#define DRM_IOCTL_I915_GEM_GET_APERTURE DRM_IOR (DRM_COMMAND_BASE + DRM_I915_GEM_GET_APERTURE, struct drm_i915_gem_get_aperture) /* Allow drivers to submit batchbuffers directly to hardware, relying * on the security mechanisms provided by hardware. @@ -600,4 +602,15 @@ struct drm_i915_gem_get_tiling { uint32_t swizzle_mode; }; +struct drm_i915_gem_get_aperture { + /** Total size of the aperture used by i915_gem_execbuffer, in bytes */ + uint64_t aper_size; + + /** + * Available space in the aperture used by i915_gem_execbuffer, in + * bytes + */ + uint64_t aper_available_size; +}; + #endif /* _I915_DRM_H_ */ From 219df32faec97349516c29f33008fea59a46e99a Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 3 Nov 2008 08:17:05 +0100 Subject: [PATCH 10/25] ALSA: rawmidi - Add open check in rawmidi callbacks The drivers (e.g. mtpav) may call rawmidi functions in irq handlers even though the streams are not opened. This results in Oops or panic. This patch adds the rawmidi state check before actually operating the rawmidi buffers. Tested-by: Ingo Molnar Signed-off-by: Takashi Iwai --- sound/core/rawmidi.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/sound/core/rawmidi.c b/sound/core/rawmidi.c index c4995c9f5730..39672f68ce5d 100644 --- a/sound/core/rawmidi.c +++ b/sound/core/rawmidi.c @@ -148,6 +148,8 @@ static int snd_rawmidi_runtime_free(struct snd_rawmidi_substream *substream) static inline void snd_rawmidi_output_trigger(struct snd_rawmidi_substream *substream,int up) { + if (!substream->opened) + return; if (up) { tasklet_hi_schedule(&substream->runtime->tasklet); } else { @@ -158,6 +160,8 @@ static inline void snd_rawmidi_output_trigger(struct snd_rawmidi_substream *subs static void snd_rawmidi_input_trigger(struct snd_rawmidi_substream *substream, int up) { + if (!substream->opened) + return; substream->ops->trigger(substream, up); if (!up && substream->runtime->event) tasklet_kill(&substream->runtime->tasklet); @@ -857,6 +861,8 @@ int snd_rawmidi_receive(struct snd_rawmidi_substream *substream, int result = 0, count1; struct snd_rawmidi_runtime *runtime = substream->runtime; + if (!substream->opened) + return -EBADFD; if (runtime->buffer == NULL) { snd_printd("snd_rawmidi_receive: input is not active!!!\n"); return -EINVAL; @@ -1126,6 +1132,8 @@ int snd_rawmidi_transmit_ack(struct snd_rawmidi_substream *substream, int count) int snd_rawmidi_transmit(struct snd_rawmidi_substream *substream, unsigned char *buffer, int count) { + if (!substream->opened) + return -EBADFD; count = snd_rawmidi_transmit_peek(substream, buffer, count); if (count < 0) return count; From b02555c3845f02924b8224ff1fd9a44f2c144dbb Mon Sep 17 00:00:00 2001 From: Zoltan Devai Date: Mon, 3 Nov 2008 00:30:28 +0100 Subject: [PATCH 11/25] ALSA: Fix PIT lockup on some chipsets when using the PC-Speaker Fix PIT lockup on some chipsets when using the PC-Speaker. Signed-off-by: Zoltan Devai Signed-off-by: Takashi Iwai --- sound/drivers/pcsp/pcsp_input.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/drivers/pcsp/pcsp_input.c b/sound/drivers/pcsp/pcsp_input.c index cd9b83e7f7d1..0444cdeb4bec 100644 --- a/sound/drivers/pcsp/pcsp_input.c +++ b/sound/drivers/pcsp/pcsp_input.c @@ -24,13 +24,13 @@ static void pcspkr_do_sound(unsigned int count) spin_lock_irqsave(&i8253_lock, flags); if (count) { - /* enable counter 2 */ - outb_p(inb_p(0x61) | 3, 0x61); /* set command for counter 2, 2 byte write */ outb_p(0xB6, 0x43); /* select desired HZ */ outb_p(count & 0xff, 0x42); outb((count >> 8) & 0xff, 0x42); + /* enable counter 2 */ + outb_p(inb_p(0x61) | 3, 0x61); } else { /* disable counter 2 */ outb(inb_p(0x61) & 0xFC, 0x61); From bb072bf0980abbe4fc905df91857a8ad34ca673c Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Sun, 2 Nov 2008 03:50:35 +0100 Subject: [PATCH 12/25] sound: struct device - replace bus_id with dev_name(), dev_set_name() [stripped sound/isa/* changes, replaced with the next patch -- tiwai] Signed-off-by: Kay Sievers Signed-off-by: Takashi Iwai --- sound/aoa/soundbus/core.c | 2 +- sound/drivers/ml403-ac97cr.c | 4 ++-- sound/soc/soc-core.c | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/sound/aoa/soundbus/core.c b/sound/aoa/soundbus/core.c index f84f3e505788..fa8ab2815a98 100644 --- a/sound/aoa/soundbus/core.c +++ b/sound/aoa/soundbus/core.c @@ -176,7 +176,7 @@ int soundbus_add_one(struct soundbus_dev *dev) return -EINVAL; } - snprintf(dev->ofdev.dev.bus_id, BUS_ID_SIZE, "soundbus:%x", ++devcount); + dev_set_name(&dev->ofdev.dev, "soundbus:%x", ++devcount); dev->ofdev.dev.bus = &soundbus_bus_type; return of_device_register(&dev->ofdev); } diff --git a/sound/drivers/ml403-ac97cr.c b/sound/drivers/ml403-ac97cr.c index ecdbeb6d3603..7783843ca9ae 100644 --- a/sound/drivers/ml403-ac97cr.c +++ b/sound/drivers/ml403-ac97cr.c @@ -1153,7 +1153,7 @@ snd_ml403_ac97cr_create(struct snd_card *card, struct platform_device *pfdev, /* get irq */ irq = platform_get_irq(pfdev, 0); if (request_irq(irq, snd_ml403_ac97cr_irq, IRQF_DISABLED, - pfdev->dev.bus_id, (void *)ml403_ac97cr)) { + dev_name(&pfdev->dev), (void *)ml403_ac97cr)) { snd_printk(KERN_ERR SND_ML403_AC97CR_DRIVER ": " "unable to grab IRQ %d\n", irq); @@ -1166,7 +1166,7 @@ snd_ml403_ac97cr_create(struct snd_card *card, struct platform_device *pfdev, ml403_ac97cr->irq); irq = platform_get_irq(pfdev, 1); if (request_irq(irq, snd_ml403_ac97cr_irq, IRQF_DISABLED, - pfdev->dev.bus_id, (void *)ml403_ac97cr)) { + dev_name(&pfdev->dev), (void *)ml403_ac97cr)) { snd_printk(KERN_ERR SND_ML403_AC97CR_DRIVER ": " "unable to grab IRQ %d\n", irq); diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index 462e635dfc74..0af3bbde5bbe 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -95,8 +95,8 @@ static int soc_ac97_dev_register(struct snd_soc_codec *codec) codec->ac97->dev.parent = NULL; codec->ac97->dev.release = soc_ac97_device_release; - snprintf(codec->ac97->dev.bus_id, BUS_ID_SIZE, "%d-%d:%s", - codec->card->number, 0, codec->name); + dev_set_name(&codec->ac97->dev, "%d-%d:%s", + codec->card->number, 0, codec->name); err = device_register(&codec->ac97->dev); if (err < 0) { snd_printk(KERN_ERR "Can't register ac97 bus\n"); From 0418ff0c8e48aae4e5e0cbcd3e19a057bea2e55a Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 3 Nov 2008 08:51:33 +0100 Subject: [PATCH 13/25] ALSA: remove direct access of dev->bus_id in sound/isa/* Removed the direct accesses of dev->bus_id in sound/isa/* by replacement with dev_err() or dev_warn() functions. Signed-off-by: Takashi Iwai --- sound/isa/ad1848/ad1848.c | 6 +++--- sound/isa/adlib.c | 12 ++++++------ sound/isa/cs423x/cs4231.c | 8 ++++---- sound/isa/cs423x/cs4236.c | 8 ++++---- sound/isa/es1688/es1688.c | 9 +++------ sound/isa/gus/gusclassic.c | 13 +++++-------- sound/isa/gus/gusextreme.c | 19 +++++++------------ sound/isa/sb/sb8.c | 4 ++-- 8 files changed, 34 insertions(+), 45 deletions(-) diff --git a/sound/isa/ad1848/ad1848.c b/sound/isa/ad1848/ad1848.c index b68d20edc20f..223a6c038819 100644 --- a/sound/isa/ad1848/ad1848.c +++ b/sound/isa/ad1848/ad1848.c @@ -70,15 +70,15 @@ static int __devinit snd_ad1848_match(struct device *dev, unsigned int n) return 0; if (port[n] == SNDRV_AUTO_PORT) { - snd_printk(KERN_ERR "%s: please specify port\n", dev->bus_id); + dev_err(dev, "please specify port\n"); return 0; } if (irq[n] == SNDRV_AUTO_IRQ) { - snd_printk(KERN_ERR "%s: please specify irq\n", dev->bus_id); + dev_err(dev, "please specify irq\n"); return 0; } if (dma1[n] == SNDRV_AUTO_DMA) { - snd_printk(KERN_ERR "%s: please specify dma1\n", dev->bus_id); + dev_err(dev, "please specify dma1\n"); return 0; } return 1; diff --git a/sound/isa/adlib.c b/sound/isa/adlib.c index efa8c80d05b6..374b7177e111 100644 --- a/sound/isa/adlib.c +++ b/sound/isa/adlib.c @@ -36,7 +36,7 @@ static int __devinit snd_adlib_match(struct device *dev, unsigned int n) return 0; if (port[n] == SNDRV_AUTO_PORT) { - snd_printk(KERN_ERR "%s: please specify port\n", dev->bus_id); + dev_err(dev, "please specify port\n"); return 0; } return 1; @@ -55,13 +55,13 @@ static int __devinit snd_adlib_probe(struct device *dev, unsigned int n) card = snd_card_new(index[n], id[n], THIS_MODULE, 0); if (!card) { - snd_printk(KERN_ERR "%s: could not create card\n", dev->bus_id); + dev_err(dev, "could not create card\n"); return -EINVAL; } card->private_data = request_region(port[n], 4, CRD_NAME); if (!card->private_data) { - snd_printk(KERN_ERR "%s: could not grab ports\n", dev->bus_id); + dev_err(dev, "could not grab ports\n"); error = -EBUSY; goto out; } @@ -73,13 +73,13 @@ static int __devinit snd_adlib_probe(struct device *dev, unsigned int n) error = snd_opl3_create(card, port[n], port[n] + 2, OPL3_HW_AUTO, 1, &opl3); if (error < 0) { - snd_printk(KERN_ERR "%s: could not create OPL\n", dev->bus_id); + dev_err(dev, "could not create OPL\n"); goto out; } error = snd_opl3_hwdep_new(opl3, 0, 0, NULL); if (error < 0) { - snd_printk(KERN_ERR "%s: could not create FM\n", dev->bus_id); + dev_err(dev, "could not create FM\n"); goto out; } @@ -87,7 +87,7 @@ static int __devinit snd_adlib_probe(struct device *dev, unsigned int n) error = snd_card_register(card); if (error < 0) { - snd_printk(KERN_ERR "%s: could not register card\n", dev->bus_id); + dev_err(dev, "could not register card\n"); goto out; } diff --git a/sound/isa/cs423x/cs4231.c b/sound/isa/cs423x/cs4231.c index ddd289120aa8..f019d449e2d6 100644 --- a/sound/isa/cs423x/cs4231.c +++ b/sound/isa/cs423x/cs4231.c @@ -74,15 +74,15 @@ static int __devinit snd_cs4231_match(struct device *dev, unsigned int n) return 0; if (port[n] == SNDRV_AUTO_PORT) { - snd_printk(KERN_ERR "%s: please specify port\n", dev->bus_id); + dev_err(dev, "please specify port\n"); return 0; } if (irq[n] == SNDRV_AUTO_IRQ) { - snd_printk(KERN_ERR "%s: please specify irq\n", dev->bus_id); + dev_err(dev, "please specify irq\n"); return 0; } if (dma1[n] == SNDRV_AUTO_DMA) { - snd_printk(KERN_ERR "%s: please specify dma1\n", dev->bus_id); + dev_err(dev, "please specify dma1\n"); return 0; } return 1; @@ -133,7 +133,7 @@ static int __devinit snd_cs4231_probe(struct device *dev, unsigned int n) mpu_port[n], 0, mpu_irq[n], mpu_irq[n] >= 0 ? IRQF_DISABLED : 0, NULL) < 0) - printk(KERN_WARNING "%s: MPU401 not detected\n", dev->bus_id); + dev_warn(dev, "MPU401 not detected\n"); } snd_card_set_dev(card, dev); diff --git a/sound/isa/cs423x/cs4236.c b/sound/isa/cs423x/cs4236.c index 91f9c15d3e30..019c9401663e 100644 --- a/sound/isa/cs423x/cs4236.c +++ b/sound/isa/cs423x/cs4236.c @@ -488,19 +488,19 @@ static int __devinit snd_cs423x_isa_match(struct device *pdev, return 0; if (port[dev] == SNDRV_AUTO_PORT) { - snd_printk(KERN_ERR "%s: please specify port\n", pdev->bus_id); + dev_err(pdev, "please specify port\n"); return 0; } if (cport[dev] == SNDRV_AUTO_PORT) { - snd_printk(KERN_ERR "%s: please specify cport\n", pdev->bus_id); + dev_err(pdev, "please specify cport\n"); return 0; } if (irq[dev] == SNDRV_AUTO_IRQ) { - snd_printk(KERN_ERR "%s: please specify irq\n", pdev->bus_id); + dev_err(pdev, "please specify irq\n"); return 0; } if (dma1[dev] == SNDRV_AUTO_DMA) { - snd_printk(KERN_ERR "%s: please specify dma1\n", pdev->bus_id); + dev_err(pdev, "please specify dma1\n"); return 0; } return 1; diff --git a/sound/isa/es1688/es1688.c b/sound/isa/es1688/es1688.c index f88639ea64b2..b46377139cf8 100644 --- a/sound/isa/es1688/es1688.c +++ b/sound/isa/es1688/es1688.c @@ -88,16 +88,14 @@ static int __devinit snd_es1688_legacy_create(struct snd_card *card, if (irq[n] == SNDRV_AUTO_IRQ) { irq[n] = snd_legacy_find_free_irq(possible_irqs); if (irq[n] < 0) { - snd_printk(KERN_ERR "%s: unable to find a free IRQ\n", - dev->bus_id); + dev_err(dev, "unable to find a free IRQ\n"); return -EBUSY; } } if (dma8[n] == SNDRV_AUTO_DMA) { dma8[n] = snd_legacy_find_free_dma(possible_dmas); if (dma8[n] < 0) { - snd_printk(KERN_ERR "%s: unable to find a free DMA\n", - dev->bus_id); + dev_err(dev, "unable to find a free DMA\n"); return -EBUSY; } } @@ -147,8 +145,7 @@ static int __devinit snd_es1688_probe(struct device *dev, unsigned int n) if (snd_opl3_create(card, chip->port, chip->port + 2, OPL3_HW_OPL3, 0, &opl3) < 0) - printk(KERN_WARNING "%s: opl3 not detected at 0x%lx\n", - dev->bus_id, chip->port); + dev_warn(dev, "opl3 not detected at 0x%lx\n", chip->port); else { error = snd_opl3_hwdep_new(opl3, 0, 1, NULL); if (error < 0) diff --git a/sound/isa/gus/gusclassic.c b/sound/isa/gus/gusclassic.c index 8f914b37bf89..426532a4d730 100644 --- a/sound/isa/gus/gusclassic.c +++ b/sound/isa/gus/gusclassic.c @@ -90,24 +90,21 @@ static int __devinit snd_gusclassic_create(struct snd_card *card, if (irq[n] == SNDRV_AUTO_IRQ) { irq[n] = snd_legacy_find_free_irq(possible_irqs); if (irq[n] < 0) { - snd_printk(KERN_ERR "%s: unable to find a free IRQ\n", - dev->bus_id); + dev_err(dev, "unable to find a free IRQ\n"); return -EBUSY; } } if (dma1[n] == SNDRV_AUTO_DMA) { dma1[n] = snd_legacy_find_free_dma(possible_dmas); if (dma1[n] < 0) { - snd_printk(KERN_ERR "%s: unable to find a free DMA1\n", - dev->bus_id); + dev_err(dev, "unable to find a free DMA1\n"); return -EBUSY; } } if (dma2[n] == SNDRV_AUTO_DMA) { dma2[n] = snd_legacy_find_free_dma(possible_dmas); if (dma2[n] < 0) { - snd_printk(KERN_ERR "%s: unable to find a free DMA2\n", - dev->bus_id); + dev_err(dev, "unable to find a free DMA2\n"); return -EBUSY; } } @@ -174,8 +171,8 @@ static int __devinit snd_gusclassic_probe(struct device *dev, unsigned int n) error = -ENODEV; if (gus->max_flag || gus->ess_flag) { - snd_printk(KERN_ERR "%s: GUS Classic or ACE soundcard was " - "not detected at 0x%lx\n", dev->bus_id, gus->gf1.port); + dev_err(dev, "GUS Classic or ACE soundcard was " + "not detected at 0x%lx\n", gus->gf1.port); goto out; } diff --git a/sound/isa/gus/gusextreme.c b/sound/isa/gus/gusextreme.c index da13185eb0a0..7ad4c3b41a84 100644 --- a/sound/isa/gus/gusextreme.c +++ b/sound/isa/gus/gusextreme.c @@ -106,16 +106,14 @@ static int __devinit snd_gusextreme_es1688_create(struct snd_card *card, if (irq[n] == SNDRV_AUTO_IRQ) { irq[n] = snd_legacy_find_free_irq(possible_irqs); if (irq[n] < 0) { - snd_printk(KERN_ERR "%s: unable to find a free IRQ " - "for ES1688\n", dev->bus_id); + dev_err(dev, "unable to find a free IRQ for ES1688\n"); return -EBUSY; } } if (dma8[n] == SNDRV_AUTO_DMA) { dma8[n] = snd_legacy_find_free_dma(possible_dmas); if (dma8[n] < 0) { - snd_printk(KERN_ERR "%s: unable to find a free DMA " - "for ES1688\n", dev->bus_id); + dev_err(dev, "unable to find a free DMA for ES1688\n"); return -EBUSY; } } @@ -143,16 +141,14 @@ static int __devinit snd_gusextreme_gus_card_create(struct snd_card *card, if (gf1_irq[n] == SNDRV_AUTO_IRQ) { gf1_irq[n] = snd_legacy_find_free_irq(possible_irqs); if (gf1_irq[n] < 0) { - snd_printk(KERN_ERR "%s: unable to find a free IRQ " - "for GF1\n", dev->bus_id); + dev_err(dev, "unable to find a free IRQ for GF1\n"); return -EBUSY; } } if (dma1[n] == SNDRV_AUTO_DMA) { dma1[n] = snd_legacy_find_free_dma(possible_dmas); if (dma1[n] < 0) { - snd_printk(KERN_ERR "%s: unable to find a free DMA " - "for GF1\n", dev->bus_id); + dev_err(dev, "unable to find a free DMA for GF1\n"); return -EBUSY; } } @@ -278,8 +274,8 @@ static int __devinit snd_gusextreme_probe(struct device *dev, unsigned int n) error = -ENODEV; if (!gus->ess_flag) { - snd_printk(KERN_ERR "%s: GUS Extreme soundcard was not " - "detected at 0x%lx\n", dev->bus_id, gus->gf1.port); + dev_err(dev, "GUS Extreme soundcard was not " + "detected at 0x%lx\n", gus->gf1.port); goto out; } gus->codec_flag = 1; @@ -310,8 +306,7 @@ static int __devinit snd_gusextreme_probe(struct device *dev, unsigned int n) if (snd_opl3_create(card, es1688->port, es1688->port + 2, OPL3_HW_OPL3, 0, &opl3) < 0) - printk(KERN_ERR "%s: opl3 not detected at 0x%lx\n", - dev->bus_id, es1688->port); + dev_warn(dev, "opl3 not detected at 0x%lx\n", es1688->port); else { error = snd_opl3_hwdep_new(opl3, 0, 2, NULL); if (error < 0) diff --git a/sound/isa/sb/sb8.c b/sound/isa/sb/sb8.c index 336a34277907..667eccc676a4 100644 --- a/sound/isa/sb/sb8.c +++ b/sound/isa/sb/sb8.c @@ -85,11 +85,11 @@ static int __devinit snd_sb8_match(struct device *pdev, unsigned int dev) if (!enable[dev]) return 0; if (irq[dev] == SNDRV_AUTO_IRQ) { - snd_printk(KERN_ERR "%s: please specify irq\n", pdev->bus_id); + dev_err(pdev, "please specify irq\n"); return 0; } if (dma8[dev] == SNDRV_AUTO_DMA) { - snd_printk(KERN_ERR "%s: please specify dma8\n", pdev->bus_id); + dev_err(pdev, "please specify dma8\n"); return 0; } return 1; From 69e50282b726bab75c8050c4836dc89b7eb7bf1a Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 3 Nov 2008 10:07:43 +0100 Subject: [PATCH 14/25] ALSA: hda - Add a quirk for another Acer Aspire (1025:0090) Added a quirk for another Acer Aspier laptop (1025:0090) with ALC883 codec. Reported in Novell bnc#426935: https://bugzilla.novell.com/show_bug.cgi?id=426935 Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index d327a371595c..a4666c96a44f 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -8414,6 +8414,7 @@ static const char *alc883_models[ALC883_MODEL_LAST] = { static struct snd_pci_quirk alc883_cfg_tbl[] = { SND_PCI_QUIRK(0x1019, 0x6668, "ECS", ALC883_3ST_6ch_DIG), SND_PCI_QUIRK(0x1025, 0x006c, "Acer Aspire 9810", ALC883_ACER_ASPIRE), + SND_PCI_QUIRK(0x1025, 0x0090, "Acer Aspire", ALC883_ACER_ASPIRE), SND_PCI_QUIRK(0x1025, 0x0110, "Acer Aspire", ALC883_ACER_ASPIRE), SND_PCI_QUIRK(0x1025, 0x0112, "Acer Aspire 9303", ALC883_ACER_ASPIRE), SND_PCI_QUIRK(0x1025, 0x0121, "Acer Aspire 5920G", ALC883_ACER_ASPIRE), From 55e03a68d2489d116a5c5e8111ecef3f69831ed6 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 3 Nov 2008 10:21:36 +0100 Subject: [PATCH 15/25] ALSA: emu10k1 - Add more invert_shared_spdif flag to Audigy models Reported in Novell bnc#440862: https://bugzilla.novell.com/show_bug.cgi?id=440862 Signed-off-by: Takashi Iwai --- sound/pci/emu10k1/emu10k1_main.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/sound/pci/emu10k1/emu10k1_main.c b/sound/pci/emu10k1/emu10k1_main.c index 2f283ea6ad9a..de5ee8f097f6 100644 --- a/sound/pci/emu10k1/emu10k1_main.c +++ b/sound/pci/emu10k1/emu10k1_main.c @@ -1464,6 +1464,7 @@ static struct snd_emu_chip_details emu_chip_details[] = { .ca0151_chip = 1, .spk71 = 1, .spdif_bug = 1, + .invert_shared_spdif = 1, /* digital/analog switch swapped */ .ac97_chip = 1} , {.vendor = 0x1102, .device = 0x0004, .subsystem = 0x20021102, .driver = "Audigy2", .name = "Audigy 2 ZS [SB0350]", @@ -1473,6 +1474,7 @@ static struct snd_emu_chip_details emu_chip_details[] = { .ca0151_chip = 1, .spk71 = 1, .spdif_bug = 1, + .invert_shared_spdif = 1, /* digital/analog switch swapped */ .ac97_chip = 1} , {.vendor = 0x1102, .device = 0x0004, .subsystem = 0x20011102, .driver = "Audigy2", .name = "Audigy 2 ZS [2001]", @@ -1482,6 +1484,7 @@ static struct snd_emu_chip_details emu_chip_details[] = { .ca0151_chip = 1, .spk71 = 1, .spdif_bug = 1, + .invert_shared_spdif = 1, /* digital/analog switch swapped */ .ac97_chip = 1} , /* Audigy 2 */ /* Tested by James@superbug.co.uk 3rd July 2005 */ From b3aa557722b3d5858f14ca559e03461c24125aaf Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 31 Oct 2008 15:44:07 -0400 Subject: [PATCH 16/25] ftrace: use kretprobe trampoline name to test in output Impact: ia64+tracing build fix When a function is kprobed, the return address is set to the kprobe_trampoline, or something similar. This caused the output of the trace to look confusing when the parent seemed to be this "kprobe_trampoline" function. To fix this, Abhishek Sagar added a test of the instruction pointer of the parent to see if it matched the kprobe_trampoline. If it did, the output would print a "[unknown/kretprobe'd]" instead. Unfortunately, not all archs do this the same way, and the trampoline function may not be exported, which causes failures in builds. This patch will compare the name instead of the pointer to see if it matches. This prevents us from depending on a function from being exported, and should work on all archs. The worst that can happen is that an arch might use a different name and then we go back to the confusing output. At least the arch will still build. Reported-by: Abhishek Sagar Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar Tested-by: Abhishek Sagar Acked-by: Abhishek Sagar --- kernel/trace/trace.c | 39 +++++++++++++++++++++------------------ 1 file changed, 21 insertions(+), 18 deletions(-) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 85bee775a03e..9f3b478f9171 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1088,17 +1088,20 @@ static void s_stop(struct seq_file *m, void *p) mutex_unlock(&trace_types_lock); } -#define KRETPROBE_MSG "[unknown/kretprobe'd]" - #ifdef CONFIG_KRETPROBES -static inline int kretprobed(unsigned long addr) +static inline const char *kretprobed(const char *name) { - return addr == (unsigned long)kretprobe_trampoline; + static const char tramp_name[] = "kretprobe_trampoline"; + int size = sizeof(tramp_name); + + if (strncmp(tramp_name, name, size) == 0) + return "[unknown/kretprobe'd]"; + return name; } #else -static inline int kretprobed(unsigned long addr) +static inline const char *kretprobed(const char *name) { - return 0; + return name; } #endif /* CONFIG_KRETPROBES */ @@ -1107,10 +1110,13 @@ seq_print_sym_short(struct trace_seq *s, const char *fmt, unsigned long address) { #ifdef CONFIG_KALLSYMS char str[KSYM_SYMBOL_LEN]; + const char *name; kallsyms_lookup(address, NULL, NULL, NULL, str); - return trace_seq_printf(s, fmt, str); + name = kretprobed(str); + + return trace_seq_printf(s, fmt, name); #endif return 1; } @@ -1121,9 +1127,12 @@ seq_print_sym_offset(struct trace_seq *s, const char *fmt, { #ifdef CONFIG_KALLSYMS char str[KSYM_SYMBOL_LEN]; + const char *name; sprint_symbol(str, address); - return trace_seq_printf(s, fmt, str); + name = kretprobed(str); + + return trace_seq_printf(s, fmt, name); #endif return 1; } @@ -1377,10 +1386,7 @@ print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu) seq_print_ip_sym(s, field->ip, sym_flags); trace_seq_puts(s, " ("); - if (kretprobed(field->parent_ip)) - trace_seq_puts(s, KRETPROBE_MSG); - else - seq_print_ip_sym(s, field->parent_ip, sym_flags); + seq_print_ip_sym(s, field->parent_ip, sym_flags); trace_seq_puts(s, ")\n"); break; } @@ -1496,12 +1502,9 @@ static enum print_line_t print_trace_fmt(struct trace_iterator *iter) ret = trace_seq_printf(s, " <-"); if (!ret) return TRACE_TYPE_PARTIAL_LINE; - if (kretprobed(field->parent_ip)) - ret = trace_seq_puts(s, KRETPROBE_MSG); - else - ret = seq_print_ip_sym(s, - field->parent_ip, - sym_flags); + ret = seq_print_ip_sym(s, + field->parent_ip, + sym_flags); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } From 7e5e26a3d8ac4bcadb380073dc9604c07a9a6198 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 31 Oct 2008 09:36:38 -0400 Subject: [PATCH 17/25] ftrace: fix hardirq header for non ftrace archs Impact: build fix for non-ftrace architectures Not all archs implement ftrace, and therefore do not have an asm/ftrace.h. This patch corrects the problem. The ftrace_nmi_enter/exit now must be defined for all archs that implement dynamic ftrace. Currently, only x86 does. Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- arch/arm/include/asm/ftrace.h | 5 ----- arch/powerpc/include/asm/ftrace.h | 5 ----- arch/sh/include/asm/ftrace.h | 5 ----- arch/sparc/include/asm/ftrace.h | 5 ----- arch/x86/include/asm/ftrace.h | 16 ---------------- include/linux/ftrace.h | 5 ++++- include/linux/hardirq.h | 2 +- 7 files changed, 5 insertions(+), 38 deletions(-) diff --git a/arch/arm/include/asm/ftrace.h b/arch/arm/include/asm/ftrace.h index 3f3a1d1508ea..39c8bc1a006a 100644 --- a/arch/arm/include/asm/ftrace.h +++ b/arch/arm/include/asm/ftrace.h @@ -1,11 +1,6 @@ #ifndef _ASM_ARM_FTRACE #define _ASM_ARM_FTRACE -#ifndef __ASSEMBLY__ -static inline void ftrace_nmi_enter(void) { } -static inline void ftrace_nmi_exit(void) { } -#endif - #ifdef CONFIG_FUNCTION_TRACER #define MCOUNT_ADDR ((long)(mcount)) #define MCOUNT_INSN_SIZE 4 /* sizeof mcount call */ diff --git a/arch/powerpc/include/asm/ftrace.h b/arch/powerpc/include/asm/ftrace.h index 1cd72700fbc0..b298f7a631e6 100644 --- a/arch/powerpc/include/asm/ftrace.h +++ b/arch/powerpc/include/asm/ftrace.h @@ -1,11 +1,6 @@ #ifndef _ASM_POWERPC_FTRACE #define _ASM_POWERPC_FTRACE -#ifndef __ASSEMBLY__ -static inline void ftrace_nmi_enter(void) { } -static inline void ftrace_nmi_exit(void) { } -#endif - #ifdef CONFIG_FUNCTION_TRACER #define MCOUNT_ADDR ((long)(_mcount)) #define MCOUNT_INSN_SIZE 4 /* sizeof mcount call */ diff --git a/arch/sh/include/asm/ftrace.h b/arch/sh/include/asm/ftrace.h index 31ada0370cb6..3aed362c9463 100644 --- a/arch/sh/include/asm/ftrace.h +++ b/arch/sh/include/asm/ftrace.h @@ -1,11 +1,6 @@ #ifndef __ASM_SH_FTRACE_H #define __ASM_SH_FTRACE_H -#ifndef __ASSEMBLY__ -static inline void ftrace_nmi_enter(void) { } -static inline void ftrace_nmi_exit(void) { } -#endif - #ifndef __ASSEMBLY__ extern void mcount(void); #endif diff --git a/arch/sparc/include/asm/ftrace.h b/arch/sparc/include/asm/ftrace.h index 62055ac0496e..d27716cd38c1 100644 --- a/arch/sparc/include/asm/ftrace.h +++ b/arch/sparc/include/asm/ftrace.h @@ -1,11 +1,6 @@ #ifndef _ASM_SPARC64_FTRACE #define _ASM_SPARC64_FTRACE -#ifndef __ASSEMBLY__ -static inline void ftrace_nmi_enter(void) { } -static inline void ftrace_nmi_exit(void) { } -#endif - #ifdef CONFIG_MCOUNT #define MCOUNT_ADDR ((long)(_mcount)) #define MCOUNT_INSN_SIZE 4 /* sizeof mcount call */ diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h index a23468194b8c..f8173ed1c970 100644 --- a/arch/x86/include/asm/ftrace.h +++ b/arch/x86/include/asm/ftrace.h @@ -17,23 +17,7 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr) */ return addr - 1; } - -#ifdef CONFIG_DYNAMIC_FTRACE -extern void ftrace_nmi_enter(void); -extern void ftrace_nmi_exit(void); -#else -static inline void ftrace_nmi_enter(void) { } -static inline void ftrace_nmi_exit(void) { } -#endif #endif /* __ASSEMBLY__ */ - -#else /* CONFIG_FUNCTION_TRACER */ - -#ifndef __ASSEMBLY__ -static inline void ftrace_nmi_enter(void) { } -static inline void ftrace_nmi_exit(void) { } -#endif - #endif /* CONFIG_FUNCTION_TRACER */ #endif /* _ASM_X86_FTRACE_H */ diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index e46a7b34037c..0ad1b48aea69 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -44,7 +44,6 @@ static inline void ftrace_kill(void) { } #endif /* CONFIG_FUNCTION_TRACER */ #ifdef CONFIG_DYNAMIC_FTRACE - enum { FTRACE_FL_FREE = (1 << 0), FTRACE_FL_FAILED = (1 << 1), @@ -105,6 +104,8 @@ extern void ftrace_release(void *start, unsigned long size); extern void ftrace_disable_daemon(void); extern void ftrace_enable_daemon(void); +extern void ftrace_nmi_enter(void); +extern void ftrace_nmi_exit(void); #else # define skip_trace(ip) ({ 0; }) @@ -113,6 +114,8 @@ extern void ftrace_enable_daemon(void); # define ftrace_disable_daemon() do { } while (0) # define ftrace_enable_daemon() do { } while (0) static inline void ftrace_release(void *start, unsigned long size) { } +static inline void ftrace_nmi_enter(void) { } +static inline void ftrace_nmi_exit(void) { } #endif /* CONFIG_DYNAMIC_FTRACE */ /* totally disable ftrace - can not re-enable after this */ diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index 0087cb43becf..ffc16ab5a878 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -4,8 +4,8 @@ #include #include #include +#include #include -#include #include /* From 818e3dd30a4ff34fff6d90e87ae59c73f6a53691 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 31 Oct 2008 09:58:35 -0400 Subject: [PATCH 18/25] tracing, ring-buffer: add paranoid checks for loops While writing a new tracer, I had a bug where I caused the ring-buffer to recurse in a bad way. The bug was with the tracer I was writing and not the ring-buffer itself. But it took a long time to find the problem. This patch adds paranoid checks into the ring-buffer infrastructure that will catch bugs of this nature. Note: I put the bug back in the tracer and this patch showed the error nicely and prevented the lockup. Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/ring_buffer.c | 56 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 56 insertions(+) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index cedf4e268285..3f3380638646 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -1022,8 +1022,23 @@ rb_reserve_next_event(struct ring_buffer_per_cpu *cpu_buffer, struct ring_buffer_event *event; u64 ts, delta; int commit = 0; + int nr_loops = 0; again: + /* + * We allow for interrupts to reenter here and do a trace. + * If one does, it will cause this original code to loop + * back here. Even with heavy interrupts happening, this + * should only happen a few times in a row. If this happens + * 1000 times in a row, there must be either an interrupt + * storm or we have something buggy. + * Bail! + */ + if (unlikely(++nr_loops > 1000)) { + RB_WARN_ON(cpu_buffer, 1); + return NULL; + } + ts = ring_buffer_time_stamp(cpu_buffer->cpu); /* @@ -1532,10 +1547,23 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) { struct buffer_page *reader = NULL; unsigned long flags; + int nr_loops = 0; spin_lock_irqsave(&cpu_buffer->lock, flags); again: + /* + * This should normally only loop twice. But because the + * start of the reader inserts an empty page, it causes + * a case where we will loop three times. There should be no + * reason to loop four times (that I know of). + */ + if (unlikely(++nr_loops > 3)) { + RB_WARN_ON(cpu_buffer, 1); + reader = NULL; + goto out; + } + reader = cpu_buffer->reader_page; /* If there's more to read, return this page */ @@ -1665,6 +1693,7 @@ ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; struct buffer_page *reader; + int nr_loops = 0; if (!cpu_isset(cpu, buffer->cpumask)) return NULL; @@ -1672,6 +1701,19 @@ ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) cpu_buffer = buffer->buffers[cpu]; again: + /* + * We repeat when a timestamp is encountered. It is possible + * to get multiple timestamps from an interrupt entering just + * as one timestamp is about to be written. The max times + * that this can happen is the number of nested interrupts we + * can have. Nesting 10 deep of interrupts is clearly + * an anomaly. + */ + if (unlikely(++nr_loops > 10)) { + RB_WARN_ON(cpu_buffer, 1); + return NULL; + } + reader = rb_get_reader_page(cpu_buffer); if (!reader) return NULL; @@ -1722,6 +1764,7 @@ ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) struct ring_buffer *buffer; struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; + int nr_loops = 0; if (ring_buffer_iter_empty(iter)) return NULL; @@ -1730,6 +1773,19 @@ ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) buffer = cpu_buffer->buffer; again: + /* + * We repeat when a timestamp is encountered. It is possible + * to get multiple timestamps from an interrupt entering just + * as one timestamp is about to be written. The max times + * that this can happen is the number of nested interrupts we + * can have. Nesting 10 deep of interrupts is clearly + * an anomaly. + */ + if (unlikely(++nr_loops > 10)) { + RB_WARN_ON(cpu_buffer, 1); + return NULL; + } + if (rb_per_cpu_empty(cpu_buffer)) return NULL; From 6b3ab21ef1ac15db4b053ce0ba8eae0ef9361c8a Mon Sep 17 00:00:00 2001 From: Matthew Ranostay Date: Mon, 3 Nov 2008 08:12:43 -0500 Subject: [PATCH 19/25] ALSA: hda: make a STAC_DELL_EQ option Add support for explicitly enabling the EQ distortion hack for systems without software biquad support. Signed-off-by: Matthew Ranostay Cc: stable@kernel.org Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_sigmatel.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index df9b0bc7f878..e6085915d86d 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -69,6 +69,7 @@ enum { enum { STAC_92HD73XX_REF, STAC_DELL_M6, + STAC_DELL_EQ, STAC_92HD73XX_MODELS }; @@ -773,9 +774,7 @@ static struct hda_verb dell_eq_core_init[] = { }; static struct hda_verb dell_m6_core_init[] = { - /* set master volume to max value without distortion - * and direct control */ - { 0x1f, AC_VERB_SET_VOLUME_KNOB_CONTROL, 0xec}, + { 0x1f, AC_VERB_SET_VOLUME_KNOB_CONTROL, 0xff}, /* setup audio connections */ { 0x0d, AC_VERB_SET_CONNECT_SEL, 0x00}, { 0x0a, AC_VERB_SET_CONNECT_SEL, 0x01}, @@ -1600,11 +1599,13 @@ static unsigned int dell_m6_pin_configs[13] = { static unsigned int *stac92hd73xx_brd_tbl[STAC_92HD73XX_MODELS] = { [STAC_92HD73XX_REF] = ref92hd73xx_pin_configs, [STAC_DELL_M6] = dell_m6_pin_configs, + [STAC_DELL_EQ] = dell_m6_pin_configs, }; static const char *stac92hd73xx_models[STAC_92HD73XX_MODELS] = { [STAC_92HD73XX_REF] = "ref", [STAC_DELL_M6] = "dell-m6", + [STAC_DELL_EQ] = "dell-eq", }; static struct snd_pci_quirk stac92hd73xx_cfg_tbl[] = { @@ -4131,12 +4132,17 @@ again: sizeof(stac92hd73xx_dmux)); switch (spec->board_config) { - case STAC_DELL_M6: + case STAC_DELL_EQ: spec->init = dell_eq_core_init; + /* fallthru */ + case STAC_DELL_M6: spec->num_smuxes = 0; spec->mixer = &stac92hd73xx_6ch_mixer[DELL_M6_MIXER]; spec->amp_nids = &stac92hd73xx_amp_nids[DELL_M6_AMP]; spec->num_amps = 1; + + if (!spec->init) + spec->init = dell_m6_core_init; switch (codec->subsystem_id) { case 0x1028025e: /* Analog Mics */ case 0x1028025f: @@ -4146,8 +4152,6 @@ again: break; case 0x10280271: /* Digital Mics */ case 0x10280272: - spec->init = dell_m6_core_init; - /* fall-through */ case 0x10280254: case 0x10280255: stac92xx_set_config_reg(codec, 0x13, 0x90A60160); From 8d5c6603c408d91ecf543f244f10ccb8b500ad95 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Mon, 3 Nov 2008 18:21:44 +0100 Subject: [PATCH 20/25] io mapping: improve documentation Impact: add documentation Signed-off-by: Keith Packard Signed-off-by: Ingo Molnar --- Documentation/io-mapping.txt | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/Documentation/io-mapping.txt b/Documentation/io-mapping.txt index cd2f726becc8..473e43b2d588 100644 --- a/Documentation/io-mapping.txt +++ b/Documentation/io-mapping.txt @@ -71,6 +71,12 @@ range, creating a permanent kernel-visible mapping to the resource. The map_atomic and map functions add the requested offset to the base of the virtual address returned by ioremap_wc. -On 32-bit processors, io_mapping_map_atomic_wc uses io_map_atomic_prot_pfn, -which uses the fixmaps to get us a mapping to a page using an atomic fashion. -For io_mapping_map_wc, ioremap_wc() is used to get a mapping of the region. +On 32-bit processors with HIGHMEM defined, io_mapping_map_atomic_wc uses +kmap_atomic_pfn to map the specified page in an atomic fashion; +kmap_atomic_pfn isn't really supposed to be used with device pages, but it +provides an efficient mapping for this usage. + +On 32-bit processors without HIGHMEM defined, io_mapping_map_atomic_wc and +io_mapping_map_wc both use ioremap_wc, a terribly inefficient function which +performs an IPI to inform all processors about the new mapping. This results +in a significant performance penalty. From e5beae16901795223d677f15aa2fe192976278ee Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Mon, 3 Nov 2008 18:21:45 +0100 Subject: [PATCH 21/25] io mapping: clean up #ifdefs Impact: cleanup clean up ifdefs: change #ifdef CONFIG_X86_32/64 to CONFIG_HAVE_ATOMIC_IOMAP. flip around the #ifdef sections to clean up the structure. Signed-off-by: Keith Packard Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 4 ++ include/linux/io-mapping.h | 89 ++++++++++++++++++++------------------ 2 files changed, 52 insertions(+), 41 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 6f20718d3156..e60c59b81bdd 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1894,6 +1894,10 @@ config SYSVIPC_COMPAT endmenu +config HAVE_ATOMIC_IOMAP + def_bool y + depends on X86_32 + source "net/Kconfig" source "drivers/Kconfig" diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h index 1b566993db6e..82df31726a54 100644 --- a/include/linux/io-mapping.h +++ b/include/linux/io-mapping.h @@ -33,48 +33,15 @@ /* this struct isn't actually defined anywhere */ struct io_mapping; -#ifdef CONFIG_X86_64 +#ifdef CONFIG_HAVE_ATOMIC_IOMAP -/* Create the io_mapping object*/ -static inline struct io_mapping * -io_mapping_create_wc(unsigned long base, unsigned long size) -{ - return (struct io_mapping *) ioremap_wc(base, size); -} +/* + * For small address space machines, mapping large objects + * into the kernel virtual space isn't practical. Where + * available, use fixmap support to dynamically map pages + * of the object at run time. + */ -static inline void -io_mapping_free(struct io_mapping *mapping) -{ - iounmap(mapping); -} - -/* Atomic map/unmap */ -static inline void * -io_mapping_map_atomic_wc(struct io_mapping *mapping, unsigned long offset) -{ - return ((char *) mapping) + offset; -} - -static inline void -io_mapping_unmap_atomic(void *vaddr) -{ -} - -/* Non-atomic map/unmap */ -static inline void * -io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) -{ - return ((char *) mapping) + offset; -} - -static inline void -io_mapping_unmap(void *vaddr) -{ -} - -#endif /* CONFIG_X86_64 */ - -#ifdef CONFIG_X86_32 static inline struct io_mapping * io_mapping_create_wc(unsigned long base, unsigned long size) { @@ -113,6 +80,46 @@ io_mapping_unmap(void *vaddr) { iounmap(vaddr); } -#endif /* CONFIG_X86_32 */ + +#else + +/* Create the io_mapping object*/ +static inline struct io_mapping * +io_mapping_create_wc(unsigned long base, unsigned long size) +{ + return (struct io_mapping *) ioremap_wc(base, size); +} + +static inline void +io_mapping_free(struct io_mapping *mapping) +{ + iounmap(mapping); +} + +/* Atomic map/unmap */ +static inline void * +io_mapping_map_atomic_wc(struct io_mapping *mapping, unsigned long offset) +{ + return ((char *) mapping) + offset; +} + +static inline void +io_mapping_unmap_atomic(void *vaddr) +{ +} + +/* Non-atomic map/unmap */ +static inline void * +io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) +{ + return ((char *) mapping) + offset; +} + +static inline void +io_mapping_unmap(void *vaddr) +{ +} + +#endif /* HAVE_ATOMIC_IOMAP */ #endif /* _LINUX_IO_MAPPING_H */ From 9b803c0fc37dc49d06d99d63741e78274801c644 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 3 Nov 2008 15:15:08 -0500 Subject: [PATCH 22/25] ftrace: update txt document Impact: Documentation update only A lot of changes have gone into ftrace. This patch updates the ftrace.txt document. Signed-off-by: Steven Rostedt Acked-by: Randy Dunlap Signed-off-by: Ingo Molnar --- Documentation/ftrace.txt | 167 +++++++++++++++++---------------------- 1 file changed, 71 insertions(+), 96 deletions(-) diff --git a/Documentation/ftrace.txt b/Documentation/ftrace.txt index ea5a827395dd..0cb989cd5124 100644 --- a/Documentation/ftrace.txt +++ b/Documentation/ftrace.txt @@ -50,26 +50,26 @@ of ftrace. Here is a list of some of the key files: Note: all time values are in microseconds. - current_tracer : This is used to set or display the current tracer + current_tracer: This is used to set or display the current tracer that is configured. - available_tracers : This holds the different types of tracers that + available_tracers: This holds the different types of tracers that have been compiled into the kernel. The tracers listed here can be configured by echoing their name into current_tracer. - tracing_enabled : This sets or displays whether the current_tracer + tracing_enabled: This sets or displays whether the current_tracer is activated and tracing or not. Echo 0 into this file to disable the tracer or 1 to enable it. - trace : This file holds the output of the trace in a human readable + trace: This file holds the output of the trace in a human readable format (described below). - latency_trace : This file shows the same trace but the information + latency_trace: This file shows the same trace but the information is organized more to display possible latencies in the system (described below). - trace_pipe : The output is the same as the "trace" file but this + trace_pipe: The output is the same as the "trace" file but this file is meant to be streamed with live tracing. Reads from this file will block until new data is retrieved. Unlike the "trace" and "latency_trace" @@ -82,11 +82,11 @@ of ftrace. Here is a list of some of the key files: tracer is not adding more data, they will display the same information every time they are read. - iter_ctrl : This file lets the user control the amount of data + iter_ctrl: This file lets the user control the amount of data that is displayed in one of the above output files. - trace_max_latency : Some of the tracers record the max latency. + trace_max_latency: Some of the tracers record the max latency. For example, the time interrupts are disabled. This time is saved in this file. The max trace will also be stored, and displayed by either @@ -94,29 +94,26 @@ of ftrace. Here is a list of some of the key files: only be recorded if the latency is greater than the value in this file. (in microseconds) - trace_entries : This sets or displays the number of trace - entries each CPU buffer can hold. The tracer buffers - are the same size for each CPU. The displayed number - is the size of the CPU buffer and not total size. The + trace_entries: This sets or displays the number of bytes each CPU + buffer can hold. The tracer buffers are the same size + for each CPU. The displayed number is the size of the + CPU buffer and not total size of all buffers. The trace buffers are allocated in pages (blocks of memory that the kernel uses for allocation, usually 4 KB in size). - Since each entry is smaller than a page, if the last - allocated page has room for more entries than were - requested, the rest of the page is used to allocate - entries. + If the last page allocated has room for more bytes + than requested, the rest of the page will be used, + making the actual allocation bigger than requested. + (Note, the size may not be a multiple of the page size due + to buffer managment overhead.) This can only be updated when the current_tracer - is set to "none". + is set to "nop". - NOTE: It is planned on changing the allocated buffers - from being the number of possible CPUS to - the number of online CPUS. - - tracing_cpumask : This is a mask that lets the user only trace + tracing_cpumask: This is a mask that lets the user only trace on specified CPUS. The format is a hex string representing the CPUS. - set_ftrace_filter : When dynamic ftrace is configured in (see the + set_ftrace_filter: When dynamic ftrace is configured in (see the section below "dynamic ftrace"), the code is dynamically modified (code text rewrite) to disable calling of the function profiler (mcount). This lets tracing be configured @@ -130,14 +127,11 @@ of ftrace. Here is a list of some of the key files: be traced. If a function exists in both set_ftrace_filter and set_ftrace_notrace, the function will _not_ be traced. - available_filter_functions : When a function is encountered the first - time by the dynamic tracer, it is recorded and - later the call is converted into a nop. This file - lists the functions that have been recorded - by the dynamic tracer and these functions can - be used to set the ftrace filter by the above - "set_ftrace_filter" file. (See the section "dynamic ftrace" - below for more details). + available_filter_functions: This lists the functions that ftrace + has processed and can trace. These are the function + names that you can pass to "set_ftrace_filter" or + "set_ftrace_notrace". (See the section "dynamic ftrace" + below for more details.) The Tracers @@ -145,7 +139,7 @@ The Tracers Here is the list of current tracers that may be configured. - ftrace - function tracer that uses mcount to trace all functions. + function - function tracer that uses mcount to trace all functions. sched_switch - traces the context switches between tasks. @@ -166,8 +160,8 @@ Here is the list of current tracers that may be configured. the highest priority task to get scheduled after it has been woken up. - none - This is not a tracer. To remove all tracers from tracing - simply echo "none" into current_tracer. + nop - This is not a tracer. To remove all tracers from tracing + simply echo "nop" into current_tracer. Examples of using the tracer @@ -182,7 +176,7 @@ Output format: Here is an example of the output format of the file "trace" -------- -# tracer: ftrace +# tracer: function # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | @@ -192,7 +186,7 @@ Here is an example of the output format of the file "trace" -------- A header is printed with the tracer name that is represented by the trace. -In this case the tracer is "ftrace". Then a header showing the format. Task +In this case the tracer is "function". Then a header showing the format. Task name "bash", the task PID "4251", the CPU that it was running on "01", the timestamp in . format, the function name that was traced "path_put" and the parent function that called this function @@ -1003,22 +997,20 @@ is the stack for the hard interrupt. This hides the fact that NEED_RESCHED has been set. We do not see the 'N' until we switch back to the task's assigned stack. -ftrace ------- +function +-------- -ftrace is not only the name of the tracing infrastructure, but it -is also a name of one of the tracers. The tracer is the function -tracer. Enabling the function tracer can be done from the -debug file system. Make sure the ftrace_enabled is set otherwise -this tracer is a nop. +This tracer is the function tracer. Enabling the function tracer +can be done from the debug file system. Make sure the ftrace_enabled is +set; otherwise this tracer is a nop. # sysctl kernel.ftrace_enabled=1 - # echo ftrace > /debug/tracing/current_tracer + # echo function > /debug/tracing/current_tracer # echo 1 > /debug/tracing/tracing_enabled # usleep 1 # echo 0 > /debug/tracing/tracing_enabled # cat /debug/tracing/trace -# tracer: ftrace +# tracer: function # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | @@ -1040,10 +1032,10 @@ this tracer is a nop. [...] -Note: ftrace uses ring buffers to store the above entries. The newest data -may overwrite the oldest data. Sometimes using echo to stop the trace -is not sufficient because the tracing could have overwritten the data -that you wanted to record. For this reason, it is sometimes better to +Note: function tracer uses ring buffers to store the above entries. +The newest data may overwrite the oldest data. Sometimes using echo to +stop the trace is not sufficient because the tracing could have overwritten +the data that you wanted to record. For this reason, it is sometimes better to disable tracing directly from a program. This allows you to stop the tracing at the point that you hit the part that you are interested in. To disable the tracing directly from a C program, something like following @@ -1077,18 +1069,31 @@ every kernel function, produced by the -pg switch in gcc), starts of pointing to a simple return. (Enabling FTRACE will include the -pg switch in the compiling of the kernel.) -When dynamic ftrace is initialized, it calls kstop_machine to make -the machine act like a uniprocessor so that it can freely modify code -without worrying about other processors executing that same code. At -initialization, the mcount calls are changed to call a "record_ip" -function. After this, the first time a kernel function is called, -it has the calling address saved in a hash table. +At compile time every C file object is run through the +recordmcount.pl script (located in the scripts directory). This +script will process the C object using objdump to find all the +locations in the .text section that call mcount. (Note, only +the .text section is processed, since processing other sections +like .init.text may cause races due to those sections being freed). -Later on the ftraced kernel thread is awoken and will again call -kstop_machine if new functions have been recorded. The ftraced thread -will change all calls to mcount to "nop". Just calling mcount -and having mcount return has shown a 10% overhead. By converting -it to a nop, there is no measurable overhead to the system. +A new section called "__mcount_loc" is created that holds references +to all the mcount call sites in the .text section. This section is +compiled back into the original object. The final linker will add +all these references into a single table. + +On boot up, before SMP is initialized, the dynamic ftrace code +scans this table and updates all the locations into nops. It also +records the locations, which are added to the available_filter_functions +list. Modules are processed as they are loaded and before they are +executed. When a module is unloaded, it also removes its functions from +the ftrace function list. This is automatic in the module unload +code, and the module author does not need to worry about it. + +When tracing is enabled, kstop_machine is called to prevent races +with the CPUS executing code being modified (which can cause the +CPU to do undesireable things), and the nops are patched back +to calls. But this time, they do not call mcount (which is just +a function stub). They now call into the ftrace infrastructure. One special side-effect to the recording of the functions being traced is that we can now selectively choose which functions we @@ -1251,36 +1256,6 @@ Produces: We can see that there's no more lock or preempt tracing. -ftraced -------- - -As mentioned above, when dynamic ftrace is configured in, a kernel -thread wakes up once a second and checks to see if there are mcount -calls that need to be converted into nops. If there are not any, then -it simply goes back to sleep. But if there are some, it will call -kstop_machine to convert the calls to nops. - -There may be a case in which you do not want this added latency. -Perhaps you are doing some audio recording and this activity might -cause skips in the playback. There is an interface to disable -and enable the "ftraced" kernel thread. - - # echo 0 > /debug/tracing/ftraced_enabled - -This will disable the calling of kstop_machine to update the -mcount calls to nops. Remember that there is a large overhead -to calling mcount. Without this kernel thread, that overhead will -exist. - -If there are recorded calls to mcount, any write to the ftraced_enabled -file will cause the kstop_machine to run. This means that a -user can manually perform the updates when they want to by simply -echoing a '0' into the ftraced_enabled file. - -The updates are also done at the beginning of enabling a tracer -that uses ftrace function recording. - - trace_pipe ---------- @@ -1289,14 +1264,14 @@ on the tracing is different. Every read from trace_pipe is consumed. This means that subsequent reads will be different. The trace is live. - # echo ftrace > /debug/tracing/current_tracer + # echo function > /debug/tracing/current_tracer # cat /debug/tracing/trace_pipe > /tmp/trace.out & [1] 4153 # echo 1 > /debug/tracing/tracing_enabled # usleep 1 # echo 0 > /debug/tracing/tracing_enabled # cat /debug/tracing/trace -# tracer: ftrace +# tracer: function # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | @@ -1317,7 +1292,7 @@ is live. Note, reading the trace_pipe file will block until more input is added. By changing the tracer, trace_pipe will issue an EOF. We needed -to set the ftrace tracer _before_ cating the trace_pipe file. +to set the function tracer _before_ we "cat" the trace_pipe file. trace entries @@ -1334,10 +1309,10 @@ number of entries. 65620 Note, to modify this, you must have tracing completely disabled. To do that, -echo "none" into the current_tracer. If the current_tracer is not set -to "none", an EINVAL error will be returned. +echo "nop" into the current_tracer. If the current_tracer is not set +to "nop", an EINVAL error will be returned. - # echo none > /debug/tracing/current_tracer + # echo nop > /debug/tracing/current_tracer # echo 100000 > /debug/tracing/trace_entries # cat /debug/tracing/trace_entries 100045 From 42ec632e7b0185d3776aa5d23380cbdc963151f2 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 3 Nov 2008 15:18:56 -0500 Subject: [PATCH 23/25] ftrace: ftrace.txt version update Impact: Documentation update only Update the version that the ftrace document was written for. Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- Documentation/ftrace.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/ftrace.txt b/Documentation/ftrace.txt index 0cb989cd5124..9cc4d685dde5 100644 --- a/Documentation/ftrace.txt +++ b/Documentation/ftrace.txt @@ -8,7 +8,7 @@ Copyright 2008 Red Hat Inc. Reviewers: Elias Oltmanns, Randy Dunlap, Andrew Morton, John Kacur, and David Teigland. -Written for: 2.6.27-rc1 +Written for: 2.6.28-rc2 Introduction ------------ From 072ba49838b42c873c496d72c91bb237914cf3b6 Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Sun, 26 Oct 2008 15:26:57 -0700 Subject: [PATCH 24/25] ftrace: fix breakage in bin_fmt results In 777e208d40d0953efc6fb4ab58590da3f7d8f02d we changed from outputting field->cpu (a char) to iter->cpu (unsigned int), increasing the resulting structure size by 3 bytes. Signed-off-by: Eric Anholt Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 9f3b478f9171..974973e39e87 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1755,7 +1755,7 @@ static enum print_line_t print_bin_fmt(struct trace_iterator *iter) return TRACE_TYPE_HANDLED; SEQ_PUT_FIELD_RET(s, entry->pid); - SEQ_PUT_FIELD_RET(s, iter->cpu); + SEQ_PUT_FIELD_RET(s, entry->cpu); SEQ_PUT_FIELD_RET(s, iter->ts); switch (entry->type) { From 6a60dd121c5b6c2d827e99b38c1326f2600c3891 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 6 Nov 2008 15:55:21 -0500 Subject: [PATCH 25/25] ftrace: split out hardirq ftrace code into own header Impact: moving of function prototypes into own header file ftrace.h is too big of a file for hardirq.h, and some archs will fail to build because of the include dependencies not being met. This patch pulls out the required prototypes for hardirq.h into a smaller and safer ftrace_irq.h file. Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- include/linux/ftrace.h | 5 ----- include/linux/ftrace_irq.h | 13 +++++++++++++ include/linux/hardirq.h | 2 +- 3 files changed, 14 insertions(+), 6 deletions(-) create mode 100644 include/linux/ftrace_irq.h diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index 0ad1b48aea69..1b340e3fa249 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -104,9 +104,6 @@ extern void ftrace_release(void *start, unsigned long size); extern void ftrace_disable_daemon(void); extern void ftrace_enable_daemon(void); -extern void ftrace_nmi_enter(void); -extern void ftrace_nmi_exit(void); - #else # define skip_trace(ip) ({ 0; }) # define ftrace_force_update() ({ 0; }) @@ -114,8 +111,6 @@ extern void ftrace_nmi_exit(void); # define ftrace_disable_daemon() do { } while (0) # define ftrace_enable_daemon() do { } while (0) static inline void ftrace_release(void *start, unsigned long size) { } -static inline void ftrace_nmi_enter(void) { } -static inline void ftrace_nmi_exit(void) { } #endif /* CONFIG_DYNAMIC_FTRACE */ /* totally disable ftrace - can not re-enable after this */ diff --git a/include/linux/ftrace_irq.h b/include/linux/ftrace_irq.h new file mode 100644 index 000000000000..b1299d6729f2 --- /dev/null +++ b/include/linux/ftrace_irq.h @@ -0,0 +1,13 @@ +#ifndef _LINUX_FTRACE_IRQ_H +#define _LINUX_FTRACE_IRQ_H + + +#ifdef CONFIG_DYNAMIC_FTRACE +extern void ftrace_nmi_enter(void); +extern void ftrace_nmi_exit(void); +#else +static inline void ftrace_nmi_enter(void) { } +static inline void ftrace_nmi_exit(void) { } +#endif + +#endif /* _LINUX_FTRACE_IRQ_H */ diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index ffc16ab5a878..89a56d79e4c6 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -4,7 +4,7 @@ #include #include #include -#include +#include #include #include