From 1c5363153dc7ae694404e7732b4ce36eecc94ca7 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Mon, 12 Sep 2005 09:15:14 -0500 Subject: [PATCH 001/240] [SCSI] blacklist REPORT LUNS usage on transtec arrays They report being SCSI-3 but seem to give back rubbish to a REPORT_LUNS command. Force them to be sequentially scanned. Signed-off-by: James Bottomley --- drivers/scsi/scsi_devinfo.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/scsi/scsi_devinfo.c b/drivers/scsi/scsi_devinfo.c index 07b554affcf2..64fc9e21f35b 100644 --- a/drivers/scsi/scsi_devinfo.c +++ b/drivers/scsi/scsi_devinfo.c @@ -110,6 +110,7 @@ static struct { {"RELISYS", "Scorpio", NULL, BLIST_NOLUN}, /* responds to all lun */ {"SANKYO", "CP525", "6.64", BLIST_NOLUN}, /* causes failed REQ SENSE, extra reset */ {"TEXEL", "CD-ROM", "1.06", BLIST_NOLUN}, + {"transtec", "T5008", "0001", BLIST_NOREPORTLUN }, {"YAMAHA", "CDR100", "1.00", BLIST_NOLUN}, /* locks up */ {"YAMAHA", "CDR102", "1.00", BLIST_NOLUN}, /* locks up */ {"YAMAHA", "CRW8424S", "1.0", BLIST_NOLUN}, /* locks up */ From 154fb614df83086ceb18a2c19908154e78d4dc98 Mon Sep 17 00:00:00 2001 From: Dave C Boutcher Date: Tue, 13 Sep 2005 10:09:02 -0500 Subject: [PATCH 002/240] [SCSI] ibmvscsi compatibility fix Linda Xie ever so gently pointed out that she had a patch to preserve compatibility with older SLES targets, and I told her we didn't need to push it to mainline. This patch explicitly checks the version of the IBMVSCSI target and ensures that large scatterlists are not sent to older targets. Signed-off-by: Linda Xie Signed-off-by: Dave Boutcher Signed-off-by: James Bottomley --- drivers/scsi/ibmvscsi/ibmvscsi.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.c b/drivers/scsi/ibmvscsi/ibmvscsi.c index 5b14934ba861..ff25210b00ba 100644 --- a/drivers/scsi/ibmvscsi/ibmvscsi.c +++ b/drivers/scsi/ibmvscsi/ibmvscsi.c @@ -727,6 +727,16 @@ static void adapter_info_rsp(struct srp_event_struct *evt_struct) if (hostdata->madapter_info.port_max_txu[0]) hostdata->host->max_sectors = hostdata->madapter_info.port_max_txu[0] >> 9; + + if (hostdata->madapter_info.os_type == 3 && + strcmp(hostdata->madapter_info.srp_version, "1.6a") <= 0) { + printk("ibmvscsi: host (Ver. %s) doesn't support large" + "transfers\n", + hostdata->madapter_info.srp_version); + printk("ibmvscsi: limiting scatterlists to %d\n", + MAX_INDIRECT_BUFS); + hostdata->host->sg_tablesize = MAX_INDIRECT_BUFS; + } } } From 82f1b07b9ad88066c0fa867dd6b32ce43ae7ad22 Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Tue, 13 Sep 2005 08:50:39 -0700 Subject: [PATCH 003/240] [IA64] fix circular dependency on generation of asm-offsets.h Fix? One ugly hack is replaced by a different ugly hack. Signed-off-by: Tony Luck --- arch/ia64/Makefile | 12 +----------- arch/ia64/kernel/asm-offsets.c | 1 + include/asm-ia64/ptrace.h | 2 ++ include/asm-ia64/thread_info.h | 7 +++++++ 4 files changed, 11 insertions(+), 11 deletions(-) diff --git a/arch/ia64/Makefile b/arch/ia64/Makefile index 70f8ed2748d1..67932ad53082 100644 --- a/arch/ia64/Makefile +++ b/arch/ia64/Makefile @@ -82,17 +82,7 @@ unwcheck: vmlinux archclean: $(Q)$(MAKE) $(clean)=$(boot) -archprepare: include/asm-ia64/.offsets.h.stamp - -include/asm-ia64/.offsets.h.stamp: - mkdir -p include/asm-ia64 - [ -s include/asm-ia64/asm-offsets.h ] \ - || echo "#define IA64_TASK_SIZE 0" > include/asm-ia64/asm-offsets.h - touch $@ - - - -CLEAN_FILES += vmlinux.gz bootloader include/asm-ia64/.offsets.h.stamp +CLEAN_FILES += vmlinux.gz bootloader boot: lib/lib.a vmlinux $(Q)$(MAKE) $(build)=$(boot) $@ diff --git a/arch/ia64/kernel/asm-offsets.c b/arch/ia64/kernel/asm-offsets.c index f6a234289341..77225659e968 100644 --- a/arch/ia64/kernel/asm-offsets.c +++ b/arch/ia64/kernel/asm-offsets.c @@ -4,6 +4,7 @@ * to extract and format the required data. */ +#define ASM_OFFSETS_C 1 #include #include diff --git a/include/asm-ia64/ptrace.h b/include/asm-ia64/ptrace.h index fc544929ac34..a79d1a7ecc77 100644 --- a/include/asm-ia64/ptrace.h +++ b/include/asm-ia64/ptrace.h @@ -57,7 +57,9 @@ #include #include +#ifndef ASM_OFFSETS_C #include +#endif /* * Base-2 logarithm of number of pages to allocate per task structure diff --git a/include/asm-ia64/thread_info.h b/include/asm-ia64/thread_info.h index cf4a950a0f4f..171b2207bde4 100644 --- a/include/asm-ia64/thread_info.h +++ b/include/asm-ia64/thread_info.h @@ -5,7 +5,9 @@ #ifndef _ASM_IA64_THREAD_INFO_H #define _ASM_IA64_THREAD_INFO_H +#ifndef ASM_OFFSETS_C #include +#endif #include #include @@ -51,9 +53,14 @@ struct thread_info { }, \ } +#ifndef ASM_OFFSETS_C /* how to get the thread information struct from C */ #define current_thread_info() ((struct thread_info *) ((char *) current + IA64_TASK_SIZE)) #define alloc_thread_info(tsk) ((struct thread_info *) ((char *) (tsk) + IA64_TASK_SIZE)) +#else +#define current_thread_info() ((struct thread_info *) 0) +#define alloc_thread_info(tsk) ((struct thread_info *) 0) +#endif #define free_thread_info(ti) /* nothing */ #define __HAVE_ARCH_TASK_STRUCT_ALLOCATOR From a89f29f6ea8dba0b7c4ae5d1d0b43de6cb500ea6 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Tue, 30 Aug 2005 11:48:07 -0500 Subject: [PATCH 004/240] [SCSI] aic7xxx: move to dma_get_required_mask() and correct 39 bit assumptions This patch moves aic7xxx over to the dma_get_required_mask() API and dumps its open coded memory check. It also appears from this bug: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=167049 That 39 bit addressing doesn't work on older cards. I surmise that the AHC_LARGE_SCBS flag is the one that marks cards capable of using 39 bit addressing, so I also folded that check into the code. Signed-off-by: James Bottomley --- drivers/scsi/aic7xxx/aic7xxx_osm.c | 9 --------- drivers/scsi/aic7xxx/aic7xxx_osm.h | 2 -- drivers/scsi/aic7xxx/aic7xxx_osm_pci.c | 8 +++++--- 3 files changed, 5 insertions(+), 14 deletions(-) diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm.c b/drivers/scsi/aic7xxx/aic7xxx_osm.c index c932b3b94490..876d1de8480d 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_osm.c +++ b/drivers/scsi/aic7xxx/aic7xxx_osm.c @@ -1109,15 +1109,6 @@ ahc_linux_register_host(struct ahc_softc *ahc, struct scsi_host_template *templa return (0); } -uint64_t -ahc_linux_get_memsize(void) -{ - struct sysinfo si; - - si_meminfo(&si); - return ((uint64_t)si.totalram << PAGE_SHIFT); -} - /* * Place the SCSI bus into a known state by either resetting it, * or forcing transfer negotiations on the next command to any diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm.h b/drivers/scsi/aic7xxx/aic7xxx_osm.h index c52996269240..be9edbe26dbe 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_osm.h +++ b/drivers/scsi/aic7xxx/aic7xxx_osm.h @@ -494,8 +494,6 @@ ahc_insb(struct ahc_softc * ahc, long port, uint8_t *array, int count) int ahc_linux_register_host(struct ahc_softc *, struct scsi_host_template *); -uint64_t ahc_linux_get_memsize(void); - /*************************** Pretty Printing **********************************/ struct info_str { char *buffer; diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c b/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c index 0d44a6907dd2..3ce77ddc889e 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c +++ b/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c @@ -180,6 +180,7 @@ ahc_linux_pci_dev_probe(struct pci_dev *pdev, const struct pci_device_id *ent) struct ahc_pci_identity *entry; char *name; int error; + struct device *dev = &pdev->dev; pci = pdev; entry = ahc_find_pci_device(pci); @@ -209,11 +210,12 @@ ahc_linux_pci_dev_probe(struct pci_dev *pdev, const struct pci_device_id *ent) pci_set_master(pdev); if (sizeof(dma_addr_t) > 4 - && ahc_linux_get_memsize() > 0x80000000 - && pci_set_dma_mask(pdev, mask_39bit) == 0) { + && ahc->features & AHC_LARGE_SCBS + && dma_set_mask(dev, mask_39bit) == 0 + && dma_get_required_mask(dev) > DMA_32BIT_MASK) { ahc->flags |= AHC_39BIT_ADDRESSING; } else { - if (pci_set_dma_mask(pdev, DMA_32BIT_MASK)) { + if (dma_set_mask(dev, DMA_32BIT_MASK)) { printk(KERN_WARNING "aic7xxx: No suitable DMA available.\n"); return (-ENODEV); } From ad4162f3712ddf25e148cff1e7dc37eafdff3e51 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 14 Sep 2005 09:56:38 +0100 Subject: [PATCH 005/240] [ARM SMP] Add timer/watchdog defines for MPCore The timer/watchdog register definitions were missing from the mpcore watchdog patch. Add them. Signed-off-by: Russell King --- drivers/char/watchdog/mpcore_wdt.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/char/watchdog/mpcore_wdt.c b/drivers/char/watchdog/mpcore_wdt.c index c694eee1fb24..75ca84ed4adf 100644 --- a/drivers/char/watchdog/mpcore_wdt.c +++ b/drivers/char/watchdog/mpcore_wdt.c @@ -30,6 +30,8 @@ #include #include #include + +#include #include struct mpcore_wdt { From 9734c3fc89e5e2b5c132ed47fe096711eff2c092 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 13 Sep 2005 10:02:44 -0700 Subject: [PATCH 006/240] [PATCH] sk98lin: remove PCI id info for cards for conflicting devices Fix PCI device id issues with sk98lin driver. 1. DLINK 530-T card has no Vital Product Data (VPD) area so the sk98lin driver won't work. (skge does however) 2. Remove commented out Yukon2 stuff 3. Restrict Linksys card to revisions that don't conflict with r8169 version. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik --- drivers/net/sk98lin/skge.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/net/sk98lin/skge.c b/drivers/net/sk98lin/skge.c index 6ee4771addf1..2e72d79a143c 100644 --- a/drivers/net/sk98lin/skge.c +++ b/drivers/net/sk98lin/skge.c @@ -5216,17 +5216,15 @@ static struct pci_device_id skge_pci_tbl[] = { { PCI_VENDOR_ID_3COM, 0x80eb, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, { PCI_VENDOR_ID_SYSKONNECT, 0x4300, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, { PCI_VENDOR_ID_SYSKONNECT, 0x4320, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, - { PCI_VENDOR_ID_DLINK, 0x4c00, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, +/* DLink card does not have valid VPD so this driver gags + * { PCI_VENDOR_ID_DLINK, 0x4c00, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, + */ { PCI_VENDOR_ID_MARVELL, 0x4320, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, -#if 0 /* don't handle Yukon2 cards at the moment -- mlindner@syskonnect.de */ - { PCI_VENDOR_ID_MARVELL, 0x4360, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, - { PCI_VENDOR_ID_MARVELL, 0x4361, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, -#endif { PCI_VENDOR_ID_MARVELL, 0x5005, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, { PCI_VENDOR_ID_CNET, 0x434e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, - { PCI_VENDOR_ID_LINKSYS, 0x1032, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, + { PCI_VENDOR_ID_LINKSYS, 0x1032, PCI_ANY_ID, 0x0015, }, { PCI_VENDOR_ID_LINKSYS, 0x1064, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, - { 0, } + { 0 } }; MODULE_DEVICE_TABLE(pci, skge_pci_tbl); From 6b4d617d154a1cf51015f7d3db158835d2235768 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 12 Sep 2005 23:21:55 -0700 Subject: [PATCH 007/240] [PATCH] s2io warning fixes drivers/net/s2io.c: In function `init_shared_mem': drivers/net/s2io.c:431: warning: cast from pointer to integer of different size drivers/net/s2io.c: In function `free_shared_mem': drivers/net/s2io.c:662: warning: cast from pointer to integer of different size Cc: "David S. Miller" Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/net/s2io.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c index c829e6a2e8a6..dd451e099a4c 100644 --- a/drivers/net/s2io.c +++ b/drivers/net/s2io.c @@ -428,7 +428,7 @@ static int init_shared_mem(struct s2io_nic *nic) DBG_PRINT(INIT_DBG, "%s: Zero DMA address for TxDL. ", dev->name); DBG_PRINT(INIT_DBG, - "Virtual address %llx\n", (u64)tmp_v); + "Virtual address %p\n", tmp_v); tmp_v = pci_alloc_consistent(nic->pdev, PAGE_SIZE, &tmp_p); if (!tmp_v) { @@ -657,9 +657,10 @@ static void free_shared_mem(struct s2io_nic *nic) mac_control->zerodma_virt_addr, (dma_addr_t)0); DBG_PRINT(INIT_DBG, - "%s: Freeing TxDL with zero DMA addr. ", dev->name); - DBG_PRINT(INIT_DBG, "Virtual address %llx\n", - (u64)(mac_control->zerodma_virt_addr)); + "%s: Freeing TxDL with zero DMA addr. ", + dev->name); + DBG_PRINT(INIT_DBG, "Virtual address %p\n", + mac_control->zerodma_virt_addr); } kfree(mac_control->fifos[i].list_info); } From 3be034b68acab61d6878431593203fc1b10d10ab Mon Sep 17 00:00:00 2001 From: Keith Owens Date: Tue, 13 Sep 2005 15:05:13 +1000 Subject: [PATCH 008/240] [PATCH] Correct xircom_cb use of CONFIG_NET_POLL_CONTROLLER xircom_cb.c does #if CONFIG_NET_POLL_CONTROLLER instead of #ifdef, resulting in drivers/net/tulip/xircom_cb.c:120:5: warning: "CONFIG_NET_POLL_CONTROLLER" is not defined. Signed-off-by: Keith Owens Signed-off-by: Jeff Garzik --- drivers/net/tulip/xircom_cb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/tulip/xircom_cb.c b/drivers/net/tulip/xircom_cb.c index 26cc4f6378c7..60d1e05ab732 100644 --- a/drivers/net/tulip/xircom_cb.c +++ b/drivers/net/tulip/xircom_cb.c @@ -117,7 +117,7 @@ static int xircom_open(struct net_device *dev); static int xircom_close(struct net_device *dev); static void xircom_up(struct xircom_private *card); static struct net_device_stats *xircom_get_stats(struct net_device *dev); -#if CONFIG_NET_POLL_CONTROLLER +#ifdef CONFIG_NET_POLL_CONTROLLER static void xircom_poll_controller(struct net_device *dev); #endif From 39eb936c7ec8ef1dccb88d3bcfc1c5fa2410c472 Mon Sep 17 00:00:00 2001 From: Arnaud Patard Date: Tue, 13 Sep 2005 00:36:45 +0200 Subject: [PATCH 009/240] [PATCH] sata_sis: Fix typo in sata port2 initialisation This patch fixes a nasty typo I introduced in my previous patch (commit f2c853bca542f5ac0b036377637192a74f2091c2). The right offset of the second port in pure sata mode is 64 and not 0x64. Thanks to Martin Schuster for pointing this to me Signed-off-by: Arnaud Patard --- Signed-off-by: Jeff Garzik --- drivers/scsi/sata_sis.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/sata_sis.c b/drivers/scsi/sata_sis.c index a63f93186e41..237c1986f012 100644 --- a/drivers/scsi/sata_sis.c +++ b/drivers/scsi/sata_sis.c @@ -289,7 +289,7 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) if (ent->device != 0x182) { if ((pmr & SIS_PMR_COMBINED) == 0) { printk(KERN_INFO "sata_sis: Detected SiS 180/181 chipset in SATA mode\n"); - port2_start=0x64; + port2_start = 64; } else { printk(KERN_INFO "sata_sis: Detected SiS 180/181 chipset in combined mode\n"); From 668e4bc7229c1866f65f3bef6ab011f8c9034089 Mon Sep 17 00:00:00 2001 From: Uwe Koziolek Date: Sun, 11 Sep 2005 17:03:35 +0200 Subject: [PATCH 010/240] [PATCH] sata_sis: uninitialized variable There is an uninitialized variable issue in sata_sis.c Signed-off-by: Jeff Garzik --- drivers/scsi/sata_sis.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/sata_sis.c b/drivers/scsi/sata_sis.c index 237c1986f012..b227e51d12f4 100644 --- a/drivers/scsi/sata_sis.c +++ b/drivers/scsi/sata_sis.c @@ -161,7 +161,7 @@ static u32 sis_scr_cfg_read (struct ata_port *ap, unsigned int sc_reg) { struct pci_dev *pdev = to_pci_dev(ap->host_set->dev); unsigned int cfg_addr = get_scr_cfg_addr(ap->port_no, sc_reg, pdev->device); - u32 val, val2; + u32 val, val2 = 0; u8 pmr; if (sc_reg == SCR_ERROR) /* doesn't exist in PCI cfg space */ From 1141455d5e29e47004ad61e0fc385cb612d4e51c Mon Sep 17 00:00:00 2001 From: matthieu castet Date: Mon, 12 Sep 2005 23:31:39 +0200 Subject: [PATCH 011/240] [PATCH] airo : fix channel number in scan this patch display the correct channel number with iwlist scan Signed-off-by: Matthieu CASTET Signed-off-by: Jeff Garzik --- drivers/net/wireless/airo.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c index 2be65d308fbe..06998c2240d9 100644 --- a/drivers/net/wireless/airo.c +++ b/drivers/net/wireless/airo.c @@ -6852,7 +6852,10 @@ static inline char *airo_translate_scan(struct net_device *dev, /* Add frequency */ iwe.cmd = SIOCGIWFREQ; iwe.u.freq.m = le16_to_cpu(bss->dsChannel); - iwe.u.freq.m = frequency_list[iwe.u.freq.m] * 100000; + /* iwe.u.freq.m containt the channel (starting 1), our + * frequency_list array start at index 0... + */ + iwe.u.freq.m = frequency_list[iwe.u.freq.m - 1] * 100000; iwe.u.freq.e = 1; current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_FREQ_LEN); From bd061bf1ff37e186012c6e2522328b58e5d35ed8 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Mon, 12 Sep 2005 10:48:59 -0400 Subject: [PATCH 012/240] [PATCH] e1000: correct rx_dropped counting Do not count frames dropped by the hardware as part of rx_dropped. Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik --- drivers/net/e1000/e1000_main.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 7c8a0a22dcd5..ee687c902a20 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -2544,7 +2544,6 @@ e1000_update_stats(struct e1000_adapter *adapter) adapter->stats.crcerrs + adapter->stats.algnerrc + adapter->stats.rlec + adapter->stats.mpc + adapter->stats.cexterr; - adapter->net_stats.rx_dropped = adapter->stats.mpc; adapter->net_stats.rx_length_errors = adapter->stats.rlec; adapter->net_stats.rx_crc_errors = adapter->stats.crcerrs; adapter->net_stats.rx_frame_errors = adapter->stats.algnerrc; From ecf7130b087a9bd1b9d03dbf452630243210d22e Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Mon, 12 Sep 2005 10:48:59 -0400 Subject: [PATCH 013/240] [PATCH] e100: correct rx_dropped and add rx_missed_errors Do not count non-error frames dropped by the hardware as part of rx_dropped. Instead, count those frames dropped as rx_missed_errors. Also, do not count other error frames as part of rx_dropped. Finally, do not count oversized frames in rx_dropped (since they are counted as part of rx_length_errors). Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik --- drivers/net/e100.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/e100.c b/drivers/net/e100.c index 25cc20e415da..fbf1c06ec5c1 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -1387,13 +1387,13 @@ static void e100_update_stats(struct nic *nic) ns->collisions += nic->tx_collisions; ns->tx_errors += le32_to_cpu(s->tx_max_collisions) + le32_to_cpu(s->tx_lost_crs); - ns->rx_dropped += le32_to_cpu(s->rx_resource_errors); ns->rx_length_errors += le32_to_cpu(s->rx_short_frame_errors) + nic->rx_over_length_errors; ns->rx_crc_errors += le32_to_cpu(s->rx_crc_errors); ns->rx_frame_errors += le32_to_cpu(s->rx_alignment_errors); ns->rx_over_errors += le32_to_cpu(s->rx_overrun_errors); ns->rx_fifo_errors += le32_to_cpu(s->rx_overrun_errors); + ns->rx_missed_errors += le32_to_cpu(s->rx_resource_errors); ns->rx_errors += le32_to_cpu(s->rx_crc_errors) + le32_to_cpu(s->rx_alignment_errors) + le32_to_cpu(s->rx_short_frame_errors) + @@ -1727,12 +1727,10 @@ static inline int e100_rx_indicate(struct nic *nic, struct rx *rx, if(unlikely(!(rfd_status & cb_ok))) { /* Don't indicate if hardware indicates errors */ - nic->net_stats.rx_dropped++; dev_kfree_skb_any(skb); } else if(actual_size > ETH_DATA_LEN + VLAN_ETH_HLEN) { /* Don't indicate oversized frames */ nic->rx_over_length_errors++; - nic->net_stats.rx_dropped++; dev_kfree_skb_any(skb); } else { nic->net_stats.rx_packets++; From c535a9dd26d1a4c6dcbd486cbe181a9e71237af1 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Mon, 12 Sep 2005 10:49:00 -0400 Subject: [PATCH 014/240] [PATCH] ixgb: correct rx_dropped counting Do not count frames dropped by the hardware as part of rx_dropped. Signed-off-by: John W. Linville Signed-off-by: Jeff Garzik --- drivers/net/ixgb/ixgb_main.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c index 5c555373adbe..89d6d69be382 100644 --- a/drivers/net/ixgb/ixgb_main.c +++ b/drivers/net/ixgb/ixgb_main.c @@ -1616,8 +1616,6 @@ ixgb_update_stats(struct ixgb_adapter *adapter) adapter->stats.icbc + adapter->stats.ecbc + adapter->stats.mpc; - adapter->net_stats.rx_dropped = adapter->stats.mpc; - /* see above * adapter->net_stats.rx_length_errors = adapter->stats.rlec; */ From 46a60f2d718d56bba8695d6f1145eb40548d86f8 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 9 Sep 2005 12:54:56 -0700 Subject: [PATCH 015/240] [PATCH] skge: gmac register access errors in dual port Merge of four previous patches and the Kconfig fix * Remove debug printk's * whitespace cleanup and version number change * clear interrupts, reset phy, and reset hardware on shutdown * ignore 64bit counter overflow interrupts * fix a couple of places where second port could clobber state of first port. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik --- drivers/net/Kconfig | 2 +- drivers/net/skge.c | 98 +++++++++++++++++++++++++-------------------- drivers/net/skge.h | 2 +- 3 files changed, 56 insertions(+), 46 deletions(-) diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 54fff9c2e802..96f14ab1c1f5 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -1951,7 +1951,7 @@ config SKGE ---help--- This driver support the Marvell Yukon or SysKonnect SK-98xx/SK-95xx and related Gigabit Ethernet adapters. It is a new smaller driver - driver with better performance and more complete ethtool support. + with better performance and more complete ethtool support. It does not support the link failover and network management features that "portable" vendor supplied sk98lin driver does. diff --git a/drivers/net/skge.c b/drivers/net/skge.c index d7c98515fdfd..0208258e7826 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -42,7 +42,7 @@ #include "skge.h" #define DRV_NAME "skge" -#define DRV_VERSION "0.9" +#define DRV_VERSION "1.0" #define PFX DRV_NAME " " #define DEFAULT_TX_RING_SIZE 128 @@ -669,7 +669,7 @@ static void skge_led(struct skge_port *skge, enum led_mode mode) PHY_M_LED_BLINK_RT(BLINK_84MS) | PHY_M_LEDC_TX_CTRL | PHY_M_LEDC_DP_CTRL); - + gm_phy_write(hw, port, PHY_MARV_LED_OVER, PHY_M_LED_MO_RX(MO_LED_OFF) | (skge->speed == SPEED_100 ? @@ -876,7 +876,7 @@ static int skge_rx_fill(struct skge_port *skge) static void skge_link_up(struct skge_port *skge) { - skge_write8(skge->hw, SK_REG(skge->port, LNK_LED_REG), + skge_write8(skge->hw, SK_REG(skge->port, LNK_LED_REG), LED_BLK_OFF|LED_SYNC_OFF|LED_ON); netif_carrier_on(skge->netdev); @@ -987,6 +987,8 @@ static void genesis_reset(struct skge_hw *hw, int port) { const u8 zero[8] = { 0 }; + skge_write8(hw, SK_REG(port, GMAC_IRQ_MSK), 0); + /* reset the statistics module */ xm_write32(hw, port, XM_GP_PORT, XM_GP_RES_STAT); xm_write16(hw, port, XM_IMSK, 0xffff); /* disable XMAC IRQs */ @@ -1021,8 +1023,6 @@ static void bcom_check_link(struct skge_hw *hw, int port) (void) xm_phy_read(hw, port, PHY_BCOM_STAT); status = xm_phy_read(hw, port, PHY_BCOM_STAT); - pr_debug("bcom_check_link status=0x%x\n", status); - if ((status & PHY_ST_LSYNC) == 0) { u16 cmd = xm_read16(hw, port, XM_MMU_CMD); cmd &= ~(XM_MMU_ENA_RX | XM_MMU_ENA_TX); @@ -1106,8 +1106,6 @@ static void bcom_phy_init(struct skge_port *skge, int jumbo) { 0x17, 0x0013 }, { 0x15, 0x0A04 }, { 0x18, 0x0420 }, }; - pr_debug("bcom_phy_init\n"); - /* read Id from external PHY (all have the same address) */ id1 = xm_phy_read(hw, port, PHY_XMAC_ID1); @@ -1340,6 +1338,8 @@ static void genesis_stop(struct skge_port *skge) int port = skge->port; u32 reg; + genesis_reset(hw, port); + /* Clear Tx packet arbiter timeout IRQ */ skge_write16(hw, B3_PA_CTRL, port == 0 ? PA_CLR_TO_TX1 : PA_CLR_TO_TX2); @@ -1465,7 +1465,6 @@ static void genesis_link_up(struct skge_port *skge) u16 cmd; u32 mode, msk; - pr_debug("genesis_link_up\n"); cmd = xm_read16(hw, port, XM_MMU_CMD); /* @@ -1578,7 +1577,6 @@ static void yukon_init(struct skge_hw *hw, int port) struct skge_port *skge = netdev_priv(hw->dev[port]); u16 ctrl, ct1000, adv; - pr_debug("yukon_init\n"); if (skge->autoneg == AUTONEG_ENABLE) { u16 ectrl = gm_phy_read(hw, port, PHY_MARV_EXT_CTRL); @@ -1677,9 +1675,11 @@ static void yukon_mac_init(struct skge_hw *hw, int port) /* WA code for COMA mode -- set PHY reset */ if (hw->chip_id == CHIP_ID_YUKON_LITE && - hw->chip_rev >= CHIP_REV_YU_LITE_A3) - skge_write32(hw, B2_GP_IO, - (skge_read32(hw, B2_GP_IO) | GP_DIR_9 | GP_IO_9)); + hw->chip_rev >= CHIP_REV_YU_LITE_A3) { + reg = skge_read32(hw, B2_GP_IO); + reg |= GP_DIR_9 | GP_IO_9; + skge_write32(hw, B2_GP_IO, reg); + } /* hard reset */ skge_write32(hw, SK_REG(port, GPHY_CTRL), GPC_RST_SET); @@ -1687,10 +1687,12 @@ static void yukon_mac_init(struct skge_hw *hw, int port) /* WA code for COMA mode -- clear PHY reset */ if (hw->chip_id == CHIP_ID_YUKON_LITE && - hw->chip_rev >= CHIP_REV_YU_LITE_A3) - skge_write32(hw, B2_GP_IO, - (skge_read32(hw, B2_GP_IO) | GP_DIR_9) - & ~GP_IO_9); + hw->chip_rev >= CHIP_REV_YU_LITE_A3) { + reg = skge_read32(hw, B2_GP_IO); + reg |= GP_DIR_9; + reg &= ~GP_IO_9; + skge_write32(hw, B2_GP_IO, reg); + } /* Set hardware config mode */ reg = GPC_INT_POL_HI | GPC_DIS_FC | GPC_DIS_SLEEP | @@ -1729,7 +1731,7 @@ static void yukon_mac_init(struct skge_hw *hw, int port) } gma_write16(hw, port, GM_GP_CTRL, reg); - skge_read16(hw, GMAC_IRQ_SRC); + skge_read16(hw, SK_REG(port, GMAC_IRQ_SRC)); yukon_init(hw, port); @@ -1801,20 +1803,26 @@ static void yukon_stop(struct skge_port *skge) struct skge_hw *hw = skge->hw; int port = skge->port; - if (hw->chip_id == CHIP_ID_YUKON_LITE && - hw->chip_rev >= CHIP_REV_YU_LITE_A3) { - skge_write32(hw, B2_GP_IO, - skge_read32(hw, B2_GP_IO) | GP_DIR_9 | GP_IO_9); - } + skge_write8(hw, SK_REG(port, GMAC_IRQ_MSK), 0); + yukon_reset(hw, port); gma_write16(hw, port, GM_GP_CTRL, gma_read16(hw, port, GM_GP_CTRL) & ~(GM_GPCR_TX_ENA|GM_GPCR_RX_ENA)); gma_read16(hw, port, GM_GP_CTRL); + if (hw->chip_id == CHIP_ID_YUKON_LITE && + hw->chip_rev >= CHIP_REV_YU_LITE_A3) { + u32 io = skge_read32(hw, B2_GP_IO); + + io |= GP_DIR_9 | GP_IO_9; + skge_write32(hw, B2_GP_IO, io); + skge_read32(hw, B2_GP_IO); + } + /* set GPHY Control reset */ - skge_write32(hw, SK_REG(port, GPHY_CTRL), GPC_RST_SET); - skge_write32(hw, SK_REG(port, GMAC_CTRL), GMC_RST_SET); + skge_write8(hw, SK_REG(port, GPHY_CTRL), GPC_RST_SET); + skge_write8(hw, SK_REG(port, GMAC_CTRL), GMC_RST_SET); } static void yukon_get_stats(struct skge_port *skge, u64 *data) @@ -1873,10 +1881,8 @@ static void yukon_link_up(struct skge_port *skge) int port = skge->port; u16 reg; - pr_debug("yukon_link_up\n"); - /* Enable Transmit FIFO Underrun */ - skge_write8(hw, GMAC_IRQ_MSK, GMAC_DEF_MSK); + skge_write8(hw, SK_REG(port, GMAC_IRQ_MSK), GMAC_DEF_MSK); reg = gma_read16(hw, port, GM_GP_CTRL); if (skge->duplex == DUPLEX_FULL || skge->autoneg == AUTONEG_ENABLE) @@ -1896,7 +1902,6 @@ static void yukon_link_down(struct skge_port *skge) int port = skge->port; u16 ctrl; - pr_debug("yukon_link_down\n"); gm_phy_write(hw, port, PHY_MARV_INT_MASK, 0); ctrl = gma_read16(hw, port, GM_GP_CTRL); @@ -2112,7 +2117,6 @@ static int skge_up(struct net_device *dev) skge_write8(hw, Q_ADDR(rxqaddr[port], Q_CSR), CSR_START | CSR_IRQ_CL_F); skge_led(skge, LED_MODE_ON); - pr_debug("skge_up completed\n"); return 0; free_rx_ring: @@ -2135,15 +2139,20 @@ static int skge_down(struct net_device *dev) netif_stop_queue(dev); + skge_write8(skge->hw, SK_REG(skge->port, LNK_LED_REG), LED_OFF); + if (hw->chip_id == CHIP_ID_GENESIS) + genesis_stop(skge); + else + yukon_stop(skge); + + hw->intr_mask &= ~portirqmask[skge->port]; + skge_write32(hw, B0_IMSK, hw->intr_mask); + /* Stop transmitter */ skge_write8(hw, Q_ADDR(txqaddr[port], Q_CSR), CSR_STOP); skge_write32(hw, RB_ADDR(txqaddr[port], RB_CTRL), RB_RST_SET|RB_DIS_OP_MD); - if (hw->chip_id == CHIP_ID_GENESIS) - genesis_stop(skge); - else - yukon_stop(skge); /* Disable Force Sync bit and Enable Alloc bit */ skge_write8(hw, SK_REG(port, TXA_CTRL), @@ -2367,8 +2376,6 @@ static void genesis_set_multicast(struct net_device *dev) u32 mode; u8 filter[8]; - pr_debug("genesis_set_multicast flags=%x count=%d\n", dev->flags, dev->mc_count); - mode = xm_read32(hw, port, XM_MODE); mode |= XM_MD_ENA_HASH; if (dev->flags & IFF_PROMISC) @@ -2530,8 +2537,6 @@ static int skge_poll(struct net_device *dev, int *budget) unsigned int to_do = min(dev->quota, *budget); unsigned int work_done = 0; - pr_debug("skge_poll\n"); - for (e = ring->to_clean; work_done < to_do; e = e->next) { struct skge_rx_desc *rd = e->desc; struct sk_buff *skb; @@ -2672,9 +2677,9 @@ static void skge_error_irq(struct skge_hw *hw) if (hw->chip_id == CHIP_ID_GENESIS) { /* clear xmac errors */ if (hwstatus & (IS_NO_STAT_M1|IS_NO_TIST_M1)) - skge_write16(hw, SK_REG(0, RX_MFF_CTRL1), MFF_CLR_INSTAT); + skge_write16(hw, RX_MFF_CTRL1, MFF_CLR_INSTAT); if (hwstatus & (IS_NO_STAT_M2|IS_NO_TIST_M2)) - skge_write16(hw, SK_REG(0, RX_MFF_CTRL2), MFF_CLR_INSTAT); + skge_write16(hw, RX_MFF_CTRL2, MFF_CLR_INSTAT); } else { /* Timestamp (unused) overflow */ if (hwstatus & IS_IRQ_TIST_OV) @@ -3000,9 +3005,6 @@ static int skge_reset(struct skge_hw *hw) skge_write32(hw, B0_IMSK, hw->intr_mask); - if (hw->chip_id != CHIP_ID_GENESIS) - skge_write8(hw, GMAC_IRQ_MSK, 0); - spin_lock_bh(&hw->phy_lock); for (i = 0; i < hw->ports; i++) { if (hw->chip_id == CHIP_ID_GENESIS) @@ -3230,6 +3232,11 @@ static void __devexit skge_remove(struct pci_dev *pdev) dev0 = hw->dev[0]; unregister_netdev(dev0); + skge_write32(hw, B0_IMSK, 0); + skge_write16(hw, B0_LED, LED_STAT_OFF); + skge_pci_clear(hw); + skge_write8(hw, B0_CTST, CS_RST_SET); + tasklet_kill(&hw->ext_tasklet); free_irq(pdev->irq, hw); @@ -3238,7 +3245,7 @@ static void __devexit skge_remove(struct pci_dev *pdev) if (dev1) free_netdev(dev1); free_netdev(dev0); - skge_write16(hw, B0_LED, LED_STAT_OFF); + iounmap(hw->regs); kfree(hw); pci_set_drvdata(pdev, NULL); @@ -3257,7 +3264,10 @@ static int skge_suspend(struct pci_dev *pdev, pm_message_t state) struct skge_port *skge = netdev_priv(dev); if (netif_running(dev)) { netif_carrier_off(dev); - skge_down(dev); + if (skge->wol) + netif_stop_queue(dev); + else + skge_down(dev); } netif_device_detach(dev); wol |= skge->wol; diff --git a/drivers/net/skge.h b/drivers/net/skge.h index f1680beb8e68..efbf98c675d2 100644 --- a/drivers/net/skge.h +++ b/drivers/net/skge.h @@ -2008,7 +2008,7 @@ enum { GM_IS_RX_FF_OR = 1<<1, /* Receive FIFO Overrun */ GM_IS_RX_COMPL = 1<<0, /* Frame Reception Complete */ -#define GMAC_DEF_MSK (GM_IS_TX_CO_OV | GM_IS_RX_CO_OV | GM_IS_TX_FF_UR) +#define GMAC_DEF_MSK (GM_IS_RX_FF_OR | GM_IS_TX_FF_UR) /* GMAC_LINK_CTRL 16 bit GMAC Link Control Reg (YUKON only) */ /* Bits 15.. 2: reserved */ From e172577da02cde3916e75406b314e6f01c228a5c Mon Sep 17 00:00:00 2001 From: Frank Pavlic Date: Thu, 8 Sep 2005 09:50:06 +0200 Subject: [PATCH 016/240] [PATCH] s390: ctc driver fixes Jeff, sorry if I have flooded your inbox, I had some problems with the mail server here yesterday, but it seems to be fixed ... Ok patch 3-4 have no dependencies on patch 2 since only qeth driver is affected.Thus I have made a new patch 2 for ctc driver. Thank you . [patch 2/4] s390: ctc driver fixes From: Peter Tiedemann - race condition fixed - minor cleanup Signed-off-by: Peter Tiedemann Signed-off-by: Frank Pavlic diffstat: ctcmain.c | 41 ++++++++++++++++++++++------------------- 1 files changed, 22 insertions(+), 19 deletions(-) Signed-off-by: Jeff Garzik --- drivers/s390/net/ctcmain.c | 41 ++++++++++++++++++++------------------ 1 file changed, 22 insertions(+), 19 deletions(-) diff --git a/drivers/s390/net/ctcmain.c b/drivers/s390/net/ctcmain.c index 96ca863eaff2..0db4f57a6a95 100644 --- a/drivers/s390/net/ctcmain.c +++ b/drivers/s390/net/ctcmain.c @@ -1,5 +1,5 @@ /* - * $Id: ctcmain.c,v 1.74 2005/03/24 09:04:17 mschwide Exp $ + * $Id: ctcmain.c,v 1.78 2005/09/07 12:18:02 pavlic Exp $ * * CTC / ESCON network driver * @@ -37,10 +37,9 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. * - * RELEASE-TAG: CTC/ESCON network driver $Revision: 1.74 $ + * RELEASE-TAG: CTC/ESCON network driver $Revision: 1.78 $ * */ - #undef DEBUG #include #include @@ -135,7 +134,7 @@ static const char *dev_event_names[] = { "TX down", "Restart", }; - + /** * Events of the channel statemachine */ @@ -249,7 +248,7 @@ static void print_banner(void) { static int printed = 0; - char vbuf[] = "$Revision: 1.74 $"; + char vbuf[] = "$Revision: 1.78 $"; char *version = vbuf; if (printed) @@ -334,7 +333,7 @@ static const char *ch_state_names[] = { "Restarting", "Not operational", }; - + #ifdef DEBUG /** * Dump header and first 16 bytes of an sk_buff for debugging purposes. @@ -671,7 +670,7 @@ static void fsm_action_nop(fsm_instance * fi, int event, void *arg) { } - + /** * Actions for channel - statemachines. *****************************************************************************/ @@ -1514,7 +1513,6 @@ ch_action_reinit(fsm_instance *fi, int event, void *arg) fsm_addtimer(&privptr->restart_timer, 1000, DEV_EVENT_RESTART, dev); } - /** * The statemachine for a channel. */ @@ -1625,7 +1623,7 @@ static const fsm_node ch_fsm[] = { }; static const int CH_FSM_LEN = sizeof (ch_fsm) / sizeof (fsm_node); - + /** * Functions related to setup and device detection. *****************************************************************************/ @@ -1976,7 +1974,7 @@ ctc_irq_handler(struct ccw_device *cdev, unsigned long intparm, struct irb *irb) fsm_event(ch->fsm, CH_EVENT_IRQ, ch); } - + /** * Actions for interface - statemachine. *****************************************************************************/ @@ -2209,13 +2207,18 @@ transmit_skb(struct channel *ch, struct sk_buff *skb) int rc = 0; DBF_TEXT(trace, 5, __FUNCTION__); + /* we need to acquire the lock for testing the state + * otherwise we can have an IRQ changing the state to + * TXIDLE after the test but before acquiring the lock. + */ + spin_lock_irqsave(&ch->collect_lock, saveflags); if (fsm_getstate(ch->fsm) != CH_STATE_TXIDLE) { int l = skb->len + LL_HEADER_LENGTH; - spin_lock_irqsave(&ch->collect_lock, saveflags); - if (ch->collect_len + l > ch->max_bufsize - 2) - rc = -EBUSY; - else { + if (ch->collect_len + l > ch->max_bufsize - 2) { + spin_unlock_irqrestore(&ch->collect_lock, saveflags); + return -EBUSY; + } else { atomic_inc(&skb->users); header.length = l; header.type = skb->protocol; @@ -2231,7 +2234,7 @@ transmit_skb(struct channel *ch, struct sk_buff *skb) int ccw_idx; struct sk_buff *nskb; unsigned long hi; - + spin_unlock_irqrestore(&ch->collect_lock, saveflags); /** * Protect skb against beeing free'd by upper * layers. @@ -2256,6 +2259,7 @@ transmit_skb(struct channel *ch, struct sk_buff *skb) if (!nskb) { atomic_dec(&skb->users); skb_pull(skb, LL_HEADER_LENGTH + 2); + ctc_clear_busy(ch->netdev); return -ENOMEM; } else { memcpy(skb_put(nskb, skb->len), @@ -2281,6 +2285,7 @@ transmit_skb(struct channel *ch, struct sk_buff *skb) */ atomic_dec(&skb->users); skb_pull(skb, LL_HEADER_LENGTH + 2); + ctc_clear_busy(ch->netdev); return -EBUSY; } @@ -2327,9 +2332,10 @@ transmit_skb(struct channel *ch, struct sk_buff *skb) } } + ctc_clear_busy(ch->netdev); return rc; } - + /** * Interface API for upper network layers *****************************************************************************/ @@ -2421,7 +2427,6 @@ ctc_tx(struct sk_buff *skb, struct net_device * dev) dev->trans_start = jiffies; if (transmit_skb(privptr->channel[WRITE], skb) != 0) rc = 1; - ctc_clear_busy(dev); return rc; } @@ -2610,7 +2615,6 @@ stats_write(struct device *dev, struct device_attribute *attr, const char *buf, return count; } - static void ctc_netdev_unregister(struct net_device * dev) { @@ -2685,7 +2689,6 @@ ctc_proto_store(struct device *dev, struct device_attribute *attr, const char *b return count; } - static ssize_t ctc_type_show(struct device *dev, struct device_attribute *attr, char *buf) { From 9cb90de84b1d9c4686f12042a3696df38e0114c3 Mon Sep 17 00:00:00 2001 From: Frank Pavlic Date: Wed, 14 Sep 2005 18:03:26 +0200 Subject: [PATCH 017/240] [PATCH] s390: TSO related fixes in qeth driver Jeff, I'm sorry seems that they have not been sent out either ... ok here they come ... [patch 3/4] s390: TSO related fixes in qeth driver From: Frank Pavlic TSO related fixes : - changing value of large_send attribute while network traffic is running caused program check and thus device recovery. - Due to hardware restriction discard packet when it exceeds 60K otherwise qeth will cause program checks and thus traffic stall when trying to send such huge packets. Signed-off-by: Frank Pavlic diffstat: qeth.h | 4 ++-- qeth_main.c | 33 +++++++++++++++++++++------------ qeth_sys.c | 10 +++------- 3 files changed, 26 insertions(+), 21 deletions(-) Signed-off-by: Jeff Garzik --- drivers/s390/net/qeth.h | 4 ++-- drivers/s390/net/qeth_main.c | 33 +++++++++++++++++++++------------ drivers/s390/net/qeth_sys.c | 10 +++------- 3 files changed, 26 insertions(+), 21 deletions(-) diff --git a/drivers/s390/net/qeth.h b/drivers/s390/net/qeth.h index 3a0285669adf..5c67d5379740 100644 --- a/drivers/s390/net/qeth.h +++ b/drivers/s390/net/qeth.h @@ -24,7 +24,7 @@ #include "qeth_mpc.h" -#define VERSION_QETH_H "$Revision: 1.139 $" +#define VERSION_QETH_H "$Revision: 1.141 $" #ifdef CONFIG_QETH_IPV6 #define QETH_VERSION_IPV6 ":IPv6" @@ -1172,7 +1172,7 @@ extern int qeth_realloc_buffer_pool(struct qeth_card *, int); extern int -qeth_set_large_send(struct qeth_card *); +qeth_set_large_send(struct qeth_card *, enum qeth_large_send_types); extern void qeth_fill_header(struct qeth_card *, struct qeth_hdr *, diff --git a/drivers/s390/net/qeth_main.c b/drivers/s390/net/qeth_main.c index 79c74f3a11f5..6f784c3f43d5 100644 --- a/drivers/s390/net/qeth_main.c +++ b/drivers/s390/net/qeth_main.c @@ -1,6 +1,6 @@ /* * - * linux/drivers/s390/net/qeth_main.c ($Revision: 1.214 $) + * linux/drivers/s390/net/qeth_main.c ($Revision: 1.219 $) * * Linux on zSeries OSA Express and HiperSockets support * @@ -12,7 +12,7 @@ * Frank Pavlic (pavlic@de.ibm.com) and * Thomas Spatzier * - * $Revision: 1.214 $ $Date: 2005/05/04 20:19:18 $ + * $Revision: 1.219 $ $Date: 2005/05/04 20:19:18 $ * * 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 @@ -80,7 +80,7 @@ qeth_eyecatcher(void) #include "qeth_eddp.h" #include "qeth_tso.h" -#define VERSION_QETH_C "$Revision: 1.214 $" +#define VERSION_QETH_C "$Revision: 1.219 $" static const char *version = "qeth S/390 OSA-Express driver"; /** @@ -3795,12 +3795,16 @@ static inline int qeth_prepare_skb(struct qeth_card *card, struct sk_buff **skb, struct qeth_hdr **hdr, int ipv) { + int rc; #ifdef CONFIG_QETH_VLAN u16 *tag; #endif QETH_DBF_TEXT(trace, 6, "prepskb"); + rc = qeth_realloc_headroom(card, skb, sizeof(struct qeth_hdr)); + if (rc) + return rc; #ifdef CONFIG_QETH_VLAN if (card->vlangrp && vlan_tx_tag_present(*skb) && ((ipv == 6) || card->options.layer2) ) { @@ -4251,7 +4255,8 @@ out: } static inline int -qeth_get_elements_no(struct qeth_card *card, void *hdr, struct sk_buff *skb) +qeth_get_elements_no(struct qeth_card *card, void *hdr, + struct sk_buff *skb, int elems) { int elements_needed = 0; @@ -4261,9 +4266,10 @@ qeth_get_elements_no(struct qeth_card *card, void *hdr, struct sk_buff *skb) if (elements_needed == 0 ) elements_needed = 1 + (((((unsigned long) hdr) % PAGE_SIZE) + skb->len) >> PAGE_SHIFT); - if (elements_needed > QETH_MAX_BUFFER_ELEMENTS(card)){ + if ((elements_needed + elems) > QETH_MAX_BUFFER_ELEMENTS(card)){ PRINT_ERR("qeth_do_send_packet: invalid size of " - "IP packet. Discarded."); + "IP packet (Number=%d / Length=%d). Discarded.\n", + (elements_needed+elems), skb->len); return 0; } return elements_needed; @@ -4337,9 +4343,11 @@ qeth_send_packet(struct qeth_card *card, struct sk_buff *skb) return -EINVAL; } } else { - elements_needed += qeth_get_elements_no(card,(void*) hdr, skb); - if (!elements_needed) + int elems = qeth_get_elements_no(card,(void*) hdr, skb, + elements_needed); + if (!elems) return -EINVAL; + elements_needed += elems; } if (card->info.type != QETH_CARD_TYPE_IQD) @@ -7038,14 +7046,16 @@ qeth_setrouting_v6(struct qeth_card *card) } int -qeth_set_large_send(struct qeth_card *card) +qeth_set_large_send(struct qeth_card *card, enum qeth_large_send_types type) { int rc = 0; - if (card->dev == NULL) + if (card->dev == NULL) { + card->options.large_send = type; return 0; - + } netif_stop_queue(card->dev); + card->options.large_send = type; switch (card->options.large_send) { case QETH_LARGE_SEND_EDDP: card->dev->features |= NETIF_F_TSO | NETIF_F_SG; @@ -7066,7 +7076,6 @@ qeth_set_large_send(struct qeth_card *card) card->dev->features &= ~(NETIF_F_TSO | NETIF_F_SG); break; } - netif_wake_queue(card->dev); return rc; } diff --git a/drivers/s390/net/qeth_sys.c b/drivers/s390/net/qeth_sys.c index 98bedb0cb387..4bdc9046d98e 100644 --- a/drivers/s390/net/qeth_sys.c +++ b/drivers/s390/net/qeth_sys.c @@ -1,6 +1,6 @@ /* * - * linux/drivers/s390/net/qeth_sys.c ($Revision: 1.51 $) + * linux/drivers/s390/net/qeth_sys.c ($Revision: 1.53 $) * * Linux on zSeries OSA Express and HiperSockets support * This file contains code related to sysfs. @@ -20,7 +20,7 @@ #include "qeth_mpc.h" #include "qeth_fs.h" -const char *VERSION_QETH_SYS_C = "$Revision: 1.51 $"; +const char *VERSION_QETH_SYS_C = "$Revision: 1.53 $"; /*****************************************************************************/ /* */ @@ -771,9 +771,7 @@ qeth_dev_large_send_store(struct device *dev, struct device_attribute *attr, con if (!card) return -EINVAL; - tmp = strsep((char **) &buf, "\n"); - if (!strcmp(tmp, "no")){ type = QETH_LARGE_SEND_NO; } else if (!strcmp(tmp, "EDDP")) { @@ -786,10 +784,8 @@ qeth_dev_large_send_store(struct device *dev, struct device_attribute *attr, con } if (card->options.large_send == type) return count; - card->options.large_send = type; - if ((rc = qeth_set_large_send(card))) + if ((rc = qeth_set_large_send(card, type))) return rc; - return count; } From f3d242e8f2bf12cb45eb35b7d2e69483d1209b13 Mon Sep 17 00:00:00 2001 From: Frank Pavlic Date: Wed, 14 Sep 2005 18:05:31 +0200 Subject: [PATCH 018/240] [PATCH] s390: qeth driver fixes [patch 4/4] s390: qeth driver fixes . From: Frank Pavlic - Clear read channel first prior to using ccw_device_set_offline. - use QETH_DBF_TEXT instead of QETH_DBF_SPRINTF - invoke qeth_halt_channel and qeth_clear_channel for all channels, even if halt/clear for one of the channel fails. - enable qeth_arp_query function for GuestLAN devices Signed-off-by: Frank Pavlic diffstat: qeth.h | 2 - qeth_main.c | 106 +++++++++++++++++++++++++----------------------------------- qeth_sys.c | 11 +++--- 3 files changed, 53 insertions(+), 66 deletions(-) Signed-off-by: Jeff Garzik --- drivers/s390/net/qeth.h | 2 +- drivers/s390/net/qeth_main.c | 106 +++++++++++++++-------------------- drivers/s390/net/qeth_sys.c | 11 ++-- 3 files changed, 53 insertions(+), 66 deletions(-) diff --git a/drivers/s390/net/qeth.h b/drivers/s390/net/qeth.h index 5c67d5379740..2ad4797ce024 100644 --- a/drivers/s390/net/qeth.h +++ b/drivers/s390/net/qeth.h @@ -24,7 +24,7 @@ #include "qeth_mpc.h" -#define VERSION_QETH_H "$Revision: 1.141 $" +#define VERSION_QETH_H "$Revision: 1.142 $" #ifdef CONFIG_QETH_IPV6 #define QETH_VERSION_IPV6 ":IPv6" diff --git a/drivers/s390/net/qeth_main.c b/drivers/s390/net/qeth_main.c index 6f784c3f43d5..86582cf1e19e 100644 --- a/drivers/s390/net/qeth_main.c +++ b/drivers/s390/net/qeth_main.c @@ -1,6 +1,6 @@ /* * - * linux/drivers/s390/net/qeth_main.c ($Revision: 1.219 $) + * linux/drivers/s390/net/qeth_main.c ($Revision: 1.224 $) * * Linux on zSeries OSA Express and HiperSockets support * @@ -12,7 +12,7 @@ * Frank Pavlic (pavlic@de.ibm.com) and * Thomas Spatzier * - * $Revision: 1.219 $ $Date: 2005/05/04 20:19:18 $ + * $Revision: 1.224 $ $Date: 2005/05/04 20:19:18 $ * * 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 @@ -29,14 +29,6 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -/*** - * eye catcher; just for debugging purposes - */ -void volatile -qeth_eyecatcher(void) -{ - return; -} #include #include @@ -80,7 +72,7 @@ qeth_eyecatcher(void) #include "qeth_eddp.h" #include "qeth_tso.h" -#define VERSION_QETH_C "$Revision: 1.219 $" +#define VERSION_QETH_C "$Revision: 1.224 $" static const char *version = "qeth S/390 OSA-Express driver"; /** @@ -2759,11 +2751,9 @@ qeth_flush_buffers(struct qeth_qdio_out_q *queue, int under_int, queue->card->perf_stats.outbound_do_qdio_start_time; #endif if (rc){ - QETH_DBF_SPRINTF(trace, 0, "qeth_flush_buffers: do_QDIO " - "returned error (%i) on device %s.", - rc, CARD_DDEV_ID(queue->card)); QETH_DBF_TEXT(trace, 2, "flushbuf"); QETH_DBF_TEXT_(trace, 2, " err%d", rc); + QETH_DBF_TEXT_(trace, 2, "%s", CARD_DDEV_ID(queue->card)); queue->card->stats.tx_errors += count; /* this must not happen under normal circumstances. if it * happens something is really wrong -> recover */ @@ -2909,11 +2899,8 @@ qeth_qdio_output_handler(struct ccw_device * ccwdev, unsigned int status, QETH_DBF_TEXT(trace, 6, "qdouhdl"); if (status & QDIO_STATUS_LOOK_FOR_ERROR) { if (status & QDIO_STATUS_ACTIVATE_CHECK_CONDITION){ - QETH_DBF_SPRINTF(trace, 2, "On device %s: " - "received active check " - "condition (0x%08x).", - CARD_BUS_ID(card), status); - QETH_DBF_TEXT(trace, 2, "chkcond"); + QETH_DBF_TEXT(trace, 2, "achkcond"); + QETH_DBF_TEXT_(trace, 2, "%s", CARD_BUS_ID(card)); QETH_DBF_TEXT_(trace, 2, "%08x", status); netif_stop_queue(card->dev); qeth_schedule_recovery(card); @@ -3356,26 +3343,32 @@ qeth_halt_channel(struct qeth_channel *channel) static int qeth_halt_channels(struct qeth_card *card) { - int rc = 0; + int rc1 = 0, rc2=0, rc3 = 0; QETH_DBF_TEXT(trace,3,"haltchs"); - if ((rc = qeth_halt_channel(&card->read))) - return rc; - if ((rc = qeth_halt_channel(&card->write))) - return rc; - return qeth_halt_channel(&card->data); + rc1 = qeth_halt_channel(&card->read); + rc2 = qeth_halt_channel(&card->write); + rc3 = qeth_halt_channel(&card->data); + if (rc1) + return rc1; + if (rc2) + return rc2; + return rc3; } static int qeth_clear_channels(struct qeth_card *card) { - int rc = 0; + int rc1 = 0, rc2=0, rc3 = 0; QETH_DBF_TEXT(trace,3,"clearchs"); - if ((rc = qeth_clear_channel(&card->read))) - return rc; - if ((rc = qeth_clear_channel(&card->write))) - return rc; - return qeth_clear_channel(&card->data); + rc1 = qeth_clear_channel(&card->read); + rc2 = qeth_clear_channel(&card->write); + rc3 = qeth_clear_channel(&card->data); + if (rc1) + return rc1; + if (rc2) + return rc2; + return rc3; } static int @@ -3445,23 +3438,23 @@ qeth_mpc_initialize(struct qeth_card *card) } if ((rc = qeth_cm_enable(card))){ QETH_DBF_TEXT_(setup, 2, "2err%d", rc); - return rc; + goto out_qdio; } if ((rc = qeth_cm_setup(card))){ QETH_DBF_TEXT_(setup, 2, "3err%d", rc); - return rc; + goto out_qdio; } if ((rc = qeth_ulp_enable(card))){ QETH_DBF_TEXT_(setup, 2, "4err%d", rc); - return rc; + goto out_qdio; } if ((rc = qeth_ulp_setup(card))){ QETH_DBF_TEXT_(setup, 2, "5err%d", rc); - return rc; + goto out_qdio; } if ((rc = qeth_alloc_qdio_buffers(card))){ QETH_DBF_TEXT_(setup, 2, "5err%d", rc); - return rc; + goto out_qdio; } if ((rc = qeth_qdio_establish(card))){ QETH_DBF_TEXT_(setup, 2, "6err%d", rc); @@ -4281,7 +4274,7 @@ qeth_send_packet(struct qeth_card *card, struct sk_buff *skb) int ipv = 0; int cast_type; struct qeth_qdio_out_q *queue; - struct qeth_hdr *hdr; + struct qeth_hdr *hdr = NULL; int elements_needed = 0; enum qeth_large_send_types large_send = QETH_LARGE_SEND_NO; struct qeth_eddp_context *ctx = NULL; @@ -4512,7 +4505,11 @@ qeth_arp_set_no_entries(struct qeth_card *card, int no_entries) QETH_DBF_TEXT(trace,3,"arpstnoe"); - /* TODO: really not supported by GuestLAN? */ + /* + * currently GuestLAN only supports the ARP assist function + * IPA_CMD_ASS_ARP_QUERY_INFO, but not IPA_CMD_ASS_ARP_SET_NO_ENTRIES; + * thus we say EOPNOTSUPP for this ARP function + */ if (card->info.guestlan) return -EOPNOTSUPP; if (!qeth_is_supported(card,IPA_ARP_PROCESSING)) { @@ -4689,14 +4686,6 @@ qeth_arp_query(struct qeth_card *card, char *udata) QETH_DBF_TEXT(trace,3,"arpquery"); - /* - * currently GuestLAN does only deliver all zeros on query arp, - * even though arp processing is supported (according to IPA supp. - * funcs flags); since all zeros is no valueable information, - * we say EOPNOTSUPP for all ARP functions - */ - /*if (card->info.guestlan) - return -EOPNOTSUPP; */ if (!qeth_is_supported(card,/*IPA_QUERY_ARP_ADDR_INFO*/ IPA_ARP_PROCESSING)) { PRINT_WARN("ARP processing not supported " @@ -4902,10 +4891,9 @@ qeth_arp_add_entry(struct qeth_card *card, struct qeth_arp_cache_entry *entry) QETH_DBF_TEXT(trace,3,"arpadent"); /* - * currently GuestLAN does only deliver all zeros on query arp, - * even though arp processing is supported (according to IPA supp. - * funcs flags); since all zeros is no valueable information, - * we say EOPNOTSUPP for all ARP functions + * currently GuestLAN only supports the ARP assist function + * IPA_CMD_ASS_ARP_QUERY_INFO, but not IPA_CMD_ASS_ARP_ADD_ENTRY; + * thus we say EOPNOTSUPP for this ARP function */ if (card->info.guestlan) return -EOPNOTSUPP; @@ -4945,10 +4933,9 @@ qeth_arp_remove_entry(struct qeth_card *card, struct qeth_arp_cache_entry *entry QETH_DBF_TEXT(trace,3,"arprment"); /* - * currently GuestLAN does only deliver all zeros on query arp, - * even though arp processing is supported (according to IPA supp. - * funcs flags); since all zeros is no valueable information, - * we say EOPNOTSUPP for all ARP functions + * currently GuestLAN only supports the ARP assist function + * IPA_CMD_ASS_ARP_QUERY_INFO, but not IPA_CMD_ASS_ARP_REMOVE_ENTRY; + * thus we say EOPNOTSUPP for this ARP function */ if (card->info.guestlan) return -EOPNOTSUPP; @@ -4986,11 +4973,10 @@ qeth_arp_flush_cache(struct qeth_card *card) QETH_DBF_TEXT(trace,3,"arpflush"); /* - * currently GuestLAN does only deliver all zeros on query arp, - * even though arp processing is supported (according to IPA supp. - * funcs flags); since all zeros is no valueable information, - * we say EOPNOTSUPP for all ARP functions - */ + * currently GuestLAN only supports the ARP assist function + * IPA_CMD_ASS_ARP_QUERY_INFO, but not IPA_CMD_ASS_ARP_FLUSH_CACHE; + * thus we say EOPNOTSUPP for this ARP function + */ if (card->info.guestlan || (card->info.type == QETH_CARD_TYPE_IQD)) return -EOPNOTSUPP; if (!qeth_is_supported(card,IPA_ARP_PROCESSING)) { @@ -8266,7 +8252,6 @@ qeth_init(void) { int rc=0; - qeth_eyecatcher(); PRINT_INFO("loading %s (%s/%s/%s/%s/%s/%s/%s %s %s)\n", version, VERSION_QETH_C, VERSION_QETH_H, VERSION_QETH_MPC_H, VERSION_QETH_MPC_C, @@ -8347,7 +8332,6 @@ again: printk("qeth: removed\n"); } -EXPORT_SYMBOL(qeth_eyecatcher); module_init(qeth_init); module_exit(qeth_exit); MODULE_AUTHOR("Frank Pavlic "); diff --git a/drivers/s390/net/qeth_sys.c b/drivers/s390/net/qeth_sys.c index 4bdc9046d98e..dda105b73063 100644 --- a/drivers/s390/net/qeth_sys.c +++ b/drivers/s390/net/qeth_sys.c @@ -1,6 +1,6 @@ /* * - * linux/drivers/s390/net/qeth_sys.c ($Revision: 1.53 $) + * linux/drivers/s390/net/qeth_sys.c ($Revision: 1.54 $) * * Linux on zSeries OSA Express and HiperSockets support * This file contains code related to sysfs. @@ -20,7 +20,7 @@ #include "qeth_mpc.h" #include "qeth_fs.h" -const char *VERSION_QETH_SYS_C = "$Revision: 1.53 $"; +const char *VERSION_QETH_SYS_C = "$Revision: 1.54 $"; /*****************************************************************************/ /* */ @@ -722,10 +722,13 @@ qeth_dev_layer2_store(struct device *dev, struct device_attribute *attr, const c if (!card) return -EINVAL; + if (card->info.type == QETH_CARD_TYPE_IQD) { + PRINT_WARN("Layer2 on Hipersockets is not supported! \n"); + return -EPERM; + } if (((card->state != CARD_STATE_DOWN) && - (card->state != CARD_STATE_RECOVER)) || - (card->info.type != QETH_CARD_TYPE_OSAE)) + (card->state != CARD_STATE_RECOVER))) return -EPERM; i = simple_strtoul(buf, &tmp, 16); From 32a3658533c6f4c6bf370dd730213e802464ef9b Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Wed, 14 Sep 2005 09:52:42 -0400 Subject: [PATCH 019/240] [PATCH] pci: only call pci_restore_bars at boot Certain (SGI?) ia64 boxes object to having their PCI BARs restored unless absolutely necessary. This patch restricts calling pci_restore_bars from pci_set_power_state unless the current state is PCI_UNKNOWN, the actual (i.e. physical) state of the device is PCI_D3hot, and the device indicates that it will lose its configuration when transitioning to PCI_D0. Signed-off-by: John W. Linville Signed-off-by: Linus Torvalds --- drivers/pci/pci.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 992db89adce7..259d247b7551 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -309,17 +309,25 @@ pci_set_power_state(struct pci_dev *dev, pci_power_t state) pci_read_config_word(dev, pm + PCI_PM_CTRL, &pmcsr); - /* If we're in D3, force entire word to 0. + /* If we're (effectively) in D3, force entire word to 0. * This doesn't affect PME_Status, disables PME_En, and * sets PowerState to 0. */ - if (dev->current_state >= PCI_D3hot) { - if (!(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET)) + switch (dev->current_state) { + case PCI_UNKNOWN: /* Boot-up */ + if ((pmcsr & PCI_PM_CTRL_STATE_MASK) == PCI_D3hot + && !(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET)) need_restore = 1; + /* Fall-through: force to D0 */ + case PCI_D3hot: + case PCI_D3cold: + case PCI_POWER_ERROR: pmcsr = 0; - } else { + break; + default: pmcsr &= ~PCI_PM_CTRL_STATE_MASK; pmcsr |= state; + break; } /* enter specified state */ From fb085cf1d4294824571815d487daccc0609543f0 Mon Sep 17 00:00:00 2001 From: Alexander Nyberg Date: Wed, 14 Sep 2005 18:54:06 +0200 Subject: [PATCH 020/240] [PATCH] Fix fs/exec.c:788 (de_thread()) BUG_ON It turns out that the BUG_ON() in fs/exec.c: de_thread() is unreliable and can trigger due to the test itself being racy. de_thread() does while (atomic_read(&sig->count) > count) { } ..... ..... BUG_ON(!thread_group_empty(current)); but release_task does write_lock_irq(&tasklist_lock) __exit_signal (this is where atomic_dec(&sig->count) is run) __exit_sighand __unhash_process takes write lock on tasklist_lock remove itself out of PIDTYPE_TGID list write_unlock_irq(&tasklist_lock) so there's a clear (although small) window between the atomic_dec(&sig->count) and the actual PIDTYPE_TGID unhashing of the thread. And actually there is no need for all threads to have exited at this point, so we simply kill the BUG_ON. Big thanks to Marc Lehmann who provided the test-case. Fixes Bug 5170 (http://bugme.osdl.org/show_bug.cgi?id=5170) Signed-off-by: Alexander Nyberg Cc: Roland McGrath Cc: Andrew Morton Cc: Ingo Molnar Acked-by: Andi Kleen Signed-off-by: Linus Torvalds --- fs/exec.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/fs/exec.c b/fs/exec.c index 14dd03907ccb..6fae59a22b43 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -745,8 +745,8 @@ static inline int de_thread(struct task_struct *tsk) } /* - * Now there are really no other threads at all, - * so it's safe to stop telling them to kill themselves. + * There may be one thread left which is just exiting, + * but it's safe to stop telling the group to kill themselves. */ sig->flags = 0; @@ -785,7 +785,6 @@ no_thread_group: kmem_cache_free(sighand_cachep, oldsighand); } - BUG_ON(!thread_group_empty(current)); BUG_ON(!thread_group_leader(current)); return 0; } From 2fd4ef85e0db9ed75c98e13953257a967ea55e03 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 14 Sep 2005 06:13:02 +0100 Subject: [PATCH 021/240] [PATCH] error path in setup_arg_pages() misses vm_unacct_memory() Pavel Emelianov and Kirill Korotaev observe that fs and arch users of security_vm_enough_memory tend to forget to vm_unacct_memory when a failure occurs further down (typically in setup_arg_pages variants). These are all users of insert_vm_struct, and that reservation will only be unaccounted on exit if the vma is marked VM_ACCOUNT: which in some cases it is (hidden inside VM_STACK_FLAGS) and in some cases it isn't. So x86_64 32-bit and ppc64 vDSO ELFs have been leaking memory into Committed_AS each time they're run. But don't add VM_ACCOUNT to them, it's inappropriate to reserve against the very unlikely case that gdb be used to COW a vDSO page - we ought to do something about that in do_wp_page, but there are yet other inconsistencies to be resolved. The safe and economical way to fix this is to let insert_vm_struct do the security_vm_enough_memory check when it finds VM_ACCOUNT is set. And the MIPS irix_brk has been calling security_vm_enough_memory before calling do_brk which repeats it, doubly accounting and so also leaking. Remove that, and all the fs and arch calls to security_vm_enough_memory: give it a less misleading name later on. Signed-off-by: Hugh Dickins Signed-Off-By: Kirill Korotaev Signed-off-by: Linus Torvalds --- arch/ia64/ia32/binfmt_elf32.c | 6 ------ arch/mips/kernel/sysirix.c | 9 ++------- arch/ppc64/kernel/vdso.c | 15 +++++++++------ arch/x86_64/ia32/ia32_binfmt.c | 5 ----- arch/x86_64/ia32/syscall32.c | 6 +----- fs/exec.c | 5 ----- mm/mmap.c | 3 +++ 7 files changed, 15 insertions(+), 34 deletions(-) diff --git a/arch/ia64/ia32/binfmt_elf32.c b/arch/ia64/ia32/binfmt_elf32.c index 31de70b7c67f..a7280d9f6c16 100644 --- a/arch/ia64/ia32/binfmt_elf32.c +++ b/arch/ia64/ia32/binfmt_elf32.c @@ -216,12 +216,6 @@ ia32_setup_arg_pages (struct linux_binprm *bprm, int executable_stack) if (!mpnt) return -ENOMEM; - if (security_vm_enough_memory((IA32_STACK_TOP - (PAGE_MASK & (unsigned long) bprm->p)) - >> PAGE_SHIFT)) { - kmem_cache_free(vm_area_cachep, mpnt); - return -ENOMEM; - } - memset(mpnt, 0, sizeof(*mpnt)); down_write(¤t->mm->mmap_sem); diff --git a/arch/mips/kernel/sysirix.c b/arch/mips/kernel/sysirix.c index 4de155699c4f..7ae4af476974 100644 --- a/arch/mips/kernel/sysirix.c +++ b/arch/mips/kernel/sysirix.c @@ -581,18 +581,13 @@ asmlinkage int irix_brk(unsigned long brk) } /* - * Check if we have enough memory.. + * Ok, looks good - let it rip. */ - if (security_vm_enough_memory((newbrk-oldbrk) >> PAGE_SHIFT)) { + if (do_brk(oldbrk, newbrk-oldbrk) != oldbrk) { ret = -ENOMEM; goto out; } - - /* - * Ok, looks good - let it rip. - */ mm->brk = brk; - do_brk(oldbrk, newbrk-oldbrk); ret = 0; out: diff --git a/arch/ppc64/kernel/vdso.c b/arch/ppc64/kernel/vdso.c index 4777676365fe..efa985f05aca 100644 --- a/arch/ppc64/kernel/vdso.c +++ b/arch/ppc64/kernel/vdso.c @@ -224,10 +224,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int executable_stack) vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); if (vma == NULL) return -ENOMEM; - if (security_vm_enough_memory(vdso_pages)) { - kmem_cache_free(vm_area_cachep, vma); - return -ENOMEM; - } + memset(vma, 0, sizeof(*vma)); /* @@ -237,8 +234,10 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int executable_stack) */ vdso_base = get_unmapped_area(NULL, vdso_base, vdso_pages << PAGE_SHIFT, 0, 0); - if (vdso_base & ~PAGE_MASK) + if (vdso_base & ~PAGE_MASK) { + kmem_cache_free(vm_area_cachep, vma); return (int)vdso_base; + } current->thread.vdso_base = vdso_base; @@ -266,7 +265,11 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int executable_stack) vma->vm_ops = &vdso_vmops; down_write(&mm->mmap_sem); - insert_vm_struct(mm, vma); + if (insert_vm_struct(mm, vma)) { + up_write(&mm->mmap_sem); + kmem_cache_free(vm_area_cachep, vma); + return -ENOMEM; + } mm->total_vm += (vma->vm_end - vma->vm_start) >> PAGE_SHIFT; up_write(&mm->mmap_sem); diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c index c8131f342cfc..d9161e395978 100644 --- a/arch/x86_64/ia32/ia32_binfmt.c +++ b/arch/x86_64/ia32/ia32_binfmt.c @@ -353,11 +353,6 @@ int setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, int exec mpnt = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); if (!mpnt) return -ENOMEM; - - if (security_vm_enough_memory((IA32_STACK_TOP - (PAGE_MASK & (unsigned long) bprm->p))>>PAGE_SHIFT)) { - kmem_cache_free(vm_area_cachep, mpnt); - return -ENOMEM; - } memset(mpnt, 0, sizeof(*mpnt)); diff --git a/arch/x86_64/ia32/syscall32.c b/arch/x86_64/ia32/syscall32.c index adbc5f8089e9..3a01329473ab 100644 --- a/arch/x86_64/ia32/syscall32.c +++ b/arch/x86_64/ia32/syscall32.c @@ -52,17 +52,13 @@ int syscall32_setup_pages(struct linux_binprm *bprm, int exstack) vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); if (!vma) return -ENOMEM; - if (security_vm_enough_memory(npages)) { - kmem_cache_free(vm_area_cachep, vma); - return -ENOMEM; - } memset(vma, 0, sizeof(struct vm_area_struct)); /* Could randomize here */ vma->vm_start = VSYSCALL32_BASE; vma->vm_end = VSYSCALL32_END; /* MAYWRITE to allow gdb to COW and set breakpoints */ - vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYEXEC|VM_MAYWRITE; + vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYWRITE; vma->vm_flags |= mm->def_flags; vma->vm_page_prot = protection_map[vma->vm_flags & 7]; vma->vm_ops = &syscall32_vm_ops; diff --git a/fs/exec.c b/fs/exec.c index 6fae59a22b43..a04a575ad433 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -421,11 +421,6 @@ int setup_arg_pages(struct linux_binprm *bprm, if (!mpnt) return -ENOMEM; - if (security_vm_enough_memory(arg_size >> PAGE_SHIFT)) { - kmem_cache_free(vm_area_cachep, mpnt); - return -ENOMEM; - } - memset(mpnt, 0, sizeof(*mpnt)); down_write(&mm->mmap_sem); diff --git a/mm/mmap.c b/mm/mmap.c index 12334aecf8ad..8b8e05f07cdb 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1993,6 +1993,9 @@ int insert_vm_struct(struct mm_struct * mm, struct vm_area_struct * vma) __vma = find_vma_prepare(mm,vma->vm_start,&prev,&rb_link,&rb_parent); if (__vma && __vma->vm_start < vma->vm_end) return -ENOMEM; + if ((vma->vm_flags & VM_ACCOUNT) && + security_vm_enough_memory(vma_pages(vma))) + return -ENOMEM; vma_link(mm, vma, prev, rb_link, rb_parent); return 0; } From c7fb0b35ada6e0e691e70af5591a2006fbec85b5 Mon Sep 17 00:00:00 2001 From: Ivan Kokshaysky Date: Wed, 14 Sep 2005 23:05:30 +0400 Subject: [PATCH 022/240] [PATCH] yenta oops fix In some cases, especially on modern laptops with a lot of PCI and cardbus bridges, we're unable to assign correct secondary/subordinate bus numbers to all cardbus bridges due to BIOS limitations unless we are using "pci=assign-busses" boot option. So some cardbus controllers may not have attached subordinate pci_bus structure, and yenta driver must cope with it - just ignore such cardbus bridges. For example, see https://bugzilla.novell.com/show_bug.cgi?id=113778 Signed-off-by: Ivan Kokshaysky Signed-off-by: Linus Torvalds --- drivers/pcmcia/yenta_socket.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/pcmcia/yenta_socket.c b/drivers/pcmcia/yenta_socket.c index f0997c36c9b7..2e43911b4876 100644 --- a/drivers/pcmcia/yenta_socket.c +++ b/drivers/pcmcia/yenta_socket.c @@ -1045,7 +1045,18 @@ static int __devinit yenta_probe (struct pci_dev *dev, const struct pci_device_i { struct yenta_socket *socket; int ret; - + + /* + * If we failed to assign proper bus numbers for this cardbus + * controller during PCI probe, its subordinate pci_bus is NULL. + * Bail out if so. + */ + if (!dev->subordinate) { + printk(KERN_ERROR "Yenta: no bus associated with %s!\n", + pci_name(dev)); + return -ENODEV; + } + socket = kmalloc(sizeof(struct yenta_socket), GFP_KERNEL); if (!socket) return -ENOMEM; From c7e43c78ae4d8630c418ce3495787b995e61a580 Mon Sep 17 00:00:00 2001 From: Alok Kataria Date: Wed, 14 Sep 2005 12:17:53 -0700 Subject: [PATCH 023/240] [PATCH] Fix slab BUG_ON() triggered by change in array cache size With the new changes that we made in the initialization of the slab allocator, we first setup the cache from which array caches are allocated, and then the cache, from which kmem_list3's are allocated. Now if the array cache comes from a cache in which objsize > 32, (in this instance size-64) then, first size-64 cache will be allocated and then the size-128 (if this is the cache from which kmem_list3's are going to be allocated). So with these new changes, we are not guaranteed that we will be initializing the malloc_sizes array in a serialized order. Thus there is a bug in __find_general_cachep, as we are checking whether the first cache_sizes ptr is NULL. This is replaced by checking whether the array-cache cache is initialized. Attached is a patch which does that. Boots fine on a x86-64, with DEBUG_SPIN, DEBUG_SLAB, and preempt. Attached is a patch which does that. Boots fine on a x86-64, with DEBUG_SPIN, DEBUG_SLAB, and preempt.Thanks & Regards, Alok Signed-off-by: Alok N Kataria Signed-off-by: Shobhit Dayal Cc: Manfred Spraul Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/slab.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/slab.c b/mm/slab.c index 9e876d6dfad9..437d3388054b 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -659,7 +659,7 @@ static inline kmem_cache_t *__find_general_cachep(size_t size, * kmem_cache_create(), or __kmalloc(), before * the generic caches are initialized. */ - BUG_ON(csizep->cs_cachep == NULL); + BUG_ON(malloc_sizes[INDEX_AC].cs_cachep == NULL); #endif while (size > csizep->cs_size) csizep++; From 0b175a7e68c2f51555820efb0a01681e3419c1bc Mon Sep 17 00:00:00 2001 From: Dipankar Sarma Date: Thu, 15 Sep 2005 00:48:42 +0530 Subject: [PATCH 024/240] [PATCH] Fix the fdtable freeing in the case of vmalloced fdset/arrays Noted by David Miller: "The bug is that free_fd_array() takes a "num" argument, but when calling it from __free_fdtable() we're instead passing in the size in bytes (ie. "num * sizeof(struct file *)")." Yes it is a bug. I think I messed it up while merging newer changes with an older version where I was using size in bytes to optimize. Signed-off-by: Dipankar Sarma Signed-off-by: Linus Torvalds --- fs/file.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/fs/file.c b/fs/file.c index 2127a7b9dc3a..fd066b261c75 100644 --- a/fs/file.c +++ b/fs/file.c @@ -69,13 +69,9 @@ void free_fd_array(struct file **array, int num) static void __free_fdtable(struct fdtable *fdt) { - int fdset_size, fdarray_size; - - fdset_size = fdt->max_fdset / 8; - fdarray_size = fdt->max_fds * sizeof(struct file *); - free_fdset(fdt->open_fds, fdset_size); - free_fdset(fdt->close_on_exec, fdset_size); - free_fd_array(fdt->fd, fdarray_size); + free_fdset(fdt->open_fds, fdt->max_fdset); + free_fdset(fdt->close_on_exec, fdt->max_fdset); + free_fd_array(fdt->fd, fdt->max_fds); kfree(fdt); } From 5a23f34798f15838fe1b7f0fd7e5be388e08c2b2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 14 Sep 2005 13:05:17 -0700 Subject: [PATCH 025/240] Fix yenta error message when unable to find a bus assignment And mention 'pci=assign-busses' as a possible fix. Signed-off-by: Linus Torvalds --- drivers/pcmcia/yenta_socket.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pcmcia/yenta_socket.c b/drivers/pcmcia/yenta_socket.c index 2e43911b4876..ba4d78e5b121 100644 --- a/drivers/pcmcia/yenta_socket.c +++ b/drivers/pcmcia/yenta_socket.c @@ -1052,8 +1052,8 @@ static int __devinit yenta_probe (struct pci_dev *dev, const struct pci_device_i * Bail out if so. */ if (!dev->subordinate) { - printk(KERN_ERROR "Yenta: no bus associated with %s!\n", - pci_name(dev)); + printk(KERN_ERR "Yenta: no bus associated with %s! " + "(try 'pci=assign-busses')\n", pci_name(dev)); return -ENODEV; } From 186d330e682210100c671355580a8592e4a21692 Mon Sep 17 00:00:00 2001 From: Timothy Thelin Date: Tue, 13 Sep 2005 19:56:28 -0700 Subject: [PATCH 026/240] [SCSI] scsi: sd, sr, st, and scsi_lib all fail to copy cmd_len to new cmd This fixes an issue in scsi command initialization from a request where sd, sr, st, and scsi_lib all fail to copy the request's cmd_len to the scsi command's cmd_len field. Signed-off-by: Timothy Thelin Signed-off-by: James Bottomley --- drivers/scsi/scsi_lib.c | 1 + drivers/scsi/sd.c | 1 + drivers/scsi/sr.c | 1 + drivers/scsi/st.c | 1 + 4 files changed, 4 insertions(+) diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 863bb6495daa..f9026d08532a 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1268,6 +1268,7 @@ static int scsi_prep_fn(struct request_queue *q, struct request *req) } } else { memcpy(cmd->cmnd, req->cmd, sizeof(cmd->cmnd)); + cmd->cmd_len = req->cmd_len; if (rq_data_dir(req) == WRITE) cmd->sc_data_direction = DMA_TO_DEVICE; else if (req->data_len) diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index de564b386052..9a1dc0cea03c 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -235,6 +235,7 @@ static int sd_init_command(struct scsi_cmnd * SCpnt) return 0; memcpy(SCpnt->cmnd, rq->cmd, sizeof(SCpnt->cmnd)); + SCpnt->cmd_len = rq->cmd_len; if (rq_data_dir(rq) == WRITE) SCpnt->sc_data_direction = DMA_TO_DEVICE; else if (rq->data_len) diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c index ce63fc8312dc..561901b1cf11 100644 --- a/drivers/scsi/sr.c +++ b/drivers/scsi/sr.c @@ -326,6 +326,7 @@ static int sr_init_command(struct scsi_cmnd * SCpnt) return 0; memcpy(SCpnt->cmnd, rq->cmd, sizeof(SCpnt->cmnd)); + SCpnt->cmd_len = rq->cmd_len; if (!rq->data_len) SCpnt->sc_data_direction = DMA_NONE; else if (rq_data_dir(rq) == WRITE) diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c index a93308ae9736..d001c046551b 100644 --- a/drivers/scsi/st.c +++ b/drivers/scsi/st.c @@ -4206,6 +4206,7 @@ static int st_init_command(struct scsi_cmnd *SCpnt) return 0; memcpy(SCpnt->cmnd, rq->cmd, sizeof(SCpnt->cmnd)); + SCpnt->cmd_len = rq->cmd_len; if (rq_data_dir(rq) == WRITE) SCpnt->sc_data_direction = DMA_TO_DEVICE; From d39a942c3f4061d2218dcadf79ce10010dbcdb3c Mon Sep 17 00:00:00 2001 From: "Randy.Dunlap" Date: Tue, 13 Sep 2005 21:43:56 -0700 Subject: [PATCH 027/240] [SCSI] scsi: 2 drivers need MODULE_LICENSE() Modules need a license to prevent kernel tainting. Signed-off-by: Randy Dunlap Signed-off-by: James Bottomley --- drivers/scsi/fd_mcs.c | 2 ++ drivers/scsi/ibmmca.c | 2 ++ 2 files changed, 4 insertions(+) diff --git a/drivers/scsi/fd_mcs.c b/drivers/scsi/fd_mcs.c index fa652f8aa643..d59d449a9e4d 100644 --- a/drivers/scsi/fd_mcs.c +++ b/drivers/scsi/fd_mcs.c @@ -1360,3 +1360,5 @@ static Scsi_Host_Template driver_template = { .use_clustering = DISABLE_CLUSTERING, }; #include "scsi_module.c" + +MODULE_LICENSE("GPL"); diff --git a/drivers/scsi/ibmmca.c b/drivers/scsi/ibmmca.c index 6e54c7d9b33c..19392f651272 100644 --- a/drivers/scsi/ibmmca.c +++ b/drivers/scsi/ibmmca.c @@ -460,6 +460,8 @@ MODULE_PARM(adisplay, "1i"); MODULE_PARM(normal, "1i"); MODULE_PARM(ansi, "1i"); #endif + +MODULE_LICENSE("GPL"); #endif /*counter of concurrent disk read/writes, to turn on/off disk led */ static int disk_rw_in_progress = 0; From 59897dad98d63ac15e1e36fcc3a107c892b1826c Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Wed, 14 Sep 2005 12:57:42 -0400 Subject: [PATCH 028/240] [SCSI] fix sym scsi boot hang On Wed, 2005-09-14 at 18:06 +1000, Anton Blanchard wrote: > And in particular it looks like the scsi_unprep_request in > scsi_queue_insert is causing it. The following patch fixes the boot > problems on the vscsi machine: OK, my fault. Your fix is almost correct .. I was going to do this eventually, honest, because there's no need to unprep and reprep a command that comes in through scsi_queue_insert(). However, I decided to leave it in to exercise the scsi_unprep_request() path just to make sure it was working. What's happening, I think, is that we also use this path for retries. Since we kill and reget the command each time, the retries decrement is never seen, so we're retrying forever. Signed-off-by: James Bottomley --- drivers/scsi/scsi_lib.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index f9026d08532a..0c6924eecf59 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -140,14 +140,12 @@ static void scsi_unprep_request(struct request *req) * commands. * Notes: This could be called either from an interrupt context or a * normal process context. - * Notes: Upon return, cmd is a stale pointer. */ int scsi_queue_insert(struct scsi_cmnd *cmd, int reason) { struct Scsi_Host *host = cmd->device->host; struct scsi_device *device = cmd->device; struct request_queue *q = device->request_queue; - struct request *req = cmd->request; unsigned long flags; SCSI_LOG_MLQUEUE(1, @@ -188,9 +186,8 @@ int scsi_queue_insert(struct scsi_cmnd *cmd, int reason) * function. The SCSI request function detects the blocked condition * and plugs the queue appropriately. */ - scsi_unprep_request(req); spin_lock_irqsave(q->queue_lock, flags); - blk_requeue_request(q, req); + blk_requeue_request(q, cmd->request); spin_unlock_irqrestore(q->queue_lock, flags); scsi_run_queue(q); From 84d370b906b0eb00277a841216987285cf4114d4 Mon Sep 17 00:00:00 2001 From: Karsten Keil Date: Wed, 14 Sep 2005 14:19:13 -0700 Subject: [PATCH 029/240] [PATCH] i4l: Sedlbauer speed star II V 3.1 exist with various subversions the 4th id field should be not used Signed-off-by: Karsten Keil Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/isdn/hisax/sedlbauer_cs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/isdn/hisax/sedlbauer_cs.c b/drivers/isdn/hisax/sedlbauer_cs.c index c6b5bf7d2aca..dc334aab433e 100644 --- a/drivers/isdn/hisax/sedlbauer_cs.c +++ b/drivers/isdn/hisax/sedlbauer_cs.c @@ -611,7 +611,7 @@ static int sedlbauer_event(event_t event, int priority, } /* sedlbauer_event */ static struct pcmcia_device_id sedlbauer_ids[] = { - PCMCIA_DEVICE_PROD_ID1234("SEDLBAUER", "speed star II", "V 3.1", "(c) 93 - 98 cb ", 0x81fb79f5, 0xf3612e1d, 0x6b95c78a, 0x50d4149c), + PCMCIA_DEVICE_PROD_ID123("SEDLBAUER", "speed star II", "V 3.1", 0x81fb79f5, 0xf3612e1d, 0x6b95c78a), PCMCIA_DEVICE_PROD_ID123("SEDLBAUER", "ISDN-Adapter", "4D67", 0x81fb79f5, 0xe4e9bc12, 0x397b7e90), PCMCIA_DEVICE_PROD_ID123("SEDLBAUER", "ISDN-Adapter", "4D98", 0x81fb79f5, 0xe4e9bc12, 0x2e5c7fce), PCMCIA_DEVICE_PROD_ID123("SEDLBAUER", "ISDN-Adapter", " (C) 93-94 VK", 0x81fb79f5, 0xe4e9bc12, 0x8db143fe), From 8fd9808aec0198b25ff9f8bb19508d9492f86920 Mon Sep 17 00:00:00 2001 From: Pantelis Antoniou Date: Wed, 14 Sep 2005 14:19:14 -0700 Subject: [PATCH 030/240] [PATCH] ppc32 8xx: flush_tlb_range() declaration uses wrong pointer type On 8xx flush_tlb_range() declaration is using a "struct mm_struct *" pointer type while the function itself uses "struct vm_area_struct *". Signed-off-by: Marcelo Tosatti Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-ppc/tlbflush.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-ppc/tlbflush.h b/include/asm-ppc/tlbflush.h index 9850f53f54b0..9afee4ffc835 100644 --- a/include/asm-ppc/tlbflush.h +++ b/include/asm-ppc/tlbflush.h @@ -72,7 +72,7 @@ static inline void flush_tlb_page(struct vm_area_struct *vma, static inline void flush_tlb_page_nohash(struct vm_area_struct *vma, unsigned long vmaddr) { _tlbie(vmaddr); } -static inline void flush_tlb_range(struct mm_struct *mm, +static inline void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, unsigned long end) { __tlbia(); } static inline void flush_tlb_kernel_range(unsigned long start, From 0ed8e048c9e11e69ec951f94066105e80cdc59fd Mon Sep 17 00:00:00 2001 From: "Antonino A. Daplas" Date: Wed, 14 Sep 2005 14:19:15 -0700 Subject: [PATCH 031/240] [PATCH] nv_i2c oops fix The call to fb_firmware_edid may return NULL but this is not checked before trying to memcpy using this pointer. Signed-off-by: Antonino Daplas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/nvidia/nv_i2c.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/video/nvidia/nv_i2c.c b/drivers/video/nvidia/nv_i2c.c index ace484fa61ce..12f2884d3f0b 100644 --- a/drivers/video/nvidia/nv_i2c.c +++ b/drivers/video/nvidia/nv_i2c.c @@ -209,10 +209,13 @@ int nvidia_probe_i2c_connector(struct fb_info *info, int conn, u8 **out_edid) if (!edid && conn == 1) { /* try to get from firmware */ - edid = kmalloc(EDID_LENGTH, GFP_KERNEL); - if (edid) - memcpy(edid, fb_firmware_edid(info->device), - EDID_LENGTH); + const u8 *e = fb_firmware_edid(info->device); + + if (e != NULL) { + edid = kmalloc(EDID_LENGTH, GFP_KERNEL); + if (edid) + memcpy(edid, e, EDID_LENGTH); + } } if (out_edid) From 8b7fc4214b550fafe595330e28d7c2c72b8b62f6 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Wed, 14 Sep 2005 14:19:17 -0700 Subject: [PATCH 032/240] [PATCH] add PCI IDs so RME32 and RME96 drivers build While doing an allyesconfig build, I noticed that the commit commit 8cdfd2519c6c9a1e6057dc5970b2542b35895738 Author: Takashi Iwai Date: Wed Sep 7 14:08:11 2005 +0200 [ALSA] Remove superfluous PCI ID definitions broke the RME32 and RME96 drivers, since the PCI IDs they use seem to have changed names. Here's a patch to fix this -- compile tested only, since I have no idea what the hardware even is. Fix the build of the RME32 and RME96 drivers by having them use the PCI_DEVICE_ID_RME_xxx names defined in instead of the PCI_DEVICE_ID_xxx names that they used to define themselves. Also fix the typo in the id PCI_DEVICE_IDRME__DIGI96_8_PAD_OR_PST so the name is PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST. Signed-off-by: Roland Dreier Acked-by: Takashi Iwai Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pci_ids.h | 2 +- sound/pci/rme32.c | 44 +++++++++++++++--------------- sound/pci/rme96.c | 60 ++++++++++++++++++++--------------------- 3 files changed, 53 insertions(+), 53 deletions(-) diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index f6c1a142286a..72fe3385743c 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1355,7 +1355,7 @@ #define PCI_DEVICE_ID_RME_DIGI96 0x3fc0 #define PCI_DEVICE_ID_RME_DIGI96_8 0x3fc1 #define PCI_DEVICE_ID_RME_DIGI96_8_PRO 0x3fc2 -#define PCI_DEVICE_IDRME__DIGI96_8_PAD_OR_PST 0x3fc3 +#define PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST 0x3fc3 #define PCI_DEVICE_ID_XILINX_HAMMERFALL 0x3fc4 #define PCI_DEVICE_ID_XILINX_HAMMERFALL_DSP 0x3fc5 #define PCI_DEVICE_ID_XILINX_HAMMERFALL_DSP_MADI 0x3fc6 diff --git a/sound/pci/rme32.c b/sound/pci/rme32.c index 3daeecb9eb0e..cd313af6ebcf 100644 --- a/sound/pci/rme32.c +++ b/sound/pci/rme32.c @@ -228,11 +228,11 @@ typedef struct snd_rme32 { } rme32_t; static struct pci_device_id snd_rme32_ids[] = { - {PCI_VENDOR_ID_XILINX_RME, PCI_DEVICE_ID_DIGI32, + {PCI_VENDOR_ID_XILINX_RME, PCI_DEVICE_ID_RME_DIGI32, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0,}, - {PCI_VENDOR_ID_XILINX_RME, PCI_DEVICE_ID_DIGI32_8, + {PCI_VENDOR_ID_XILINX_RME, PCI_DEVICE_ID_RME_DIGI32_8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0,}, - {PCI_VENDOR_ID_XILINX_RME, PCI_DEVICE_ID_DIGI32_PRO, + {PCI_VENDOR_ID_XILINX_RME, PCI_DEVICE_ID_RME_DIGI32_PRO, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0,}, {0,} }; @@ -240,7 +240,7 @@ static struct pci_device_id snd_rme32_ids[] = { MODULE_DEVICE_TABLE(pci, snd_rme32_ids); #define RME32_ISWORKING(rme32) ((rme32)->wcreg & RME32_WCR_START) -#define RME32_PRO_WITH_8414(rme32) ((rme32)->pci->device == PCI_DEVICE_ID_DIGI32_PRO && (rme32)->rev == RME32_PRO_REVISION_WITH_8414) +#define RME32_PRO_WITH_8414(rme32) ((rme32)->pci->device == PCI_DEVICE_ID_RME_DIGI32_PRO && (rme32)->rev == RME32_PRO_REVISION_WITH_8414) static int snd_rme32_playback_prepare(snd_pcm_substream_t * substream); @@ -527,21 +527,21 @@ static int snd_rme32_playback_setrate(rme32_t * rme32, int rate) RME32_WCR_FREQ_1; break; case 64000: - if (rme32->pci->device != PCI_DEVICE_ID_DIGI32_PRO) + if (rme32->pci->device != PCI_DEVICE_ID_RME_DIGI32_PRO) return -EINVAL; rme32->wcreg |= RME32_WCR_DS_BM; rme32->wcreg = (rme32->wcreg | RME32_WCR_FREQ_0) & ~RME32_WCR_FREQ_1; break; case 88200: - if (rme32->pci->device != PCI_DEVICE_ID_DIGI32_PRO) + if (rme32->pci->device != PCI_DEVICE_ID_RME_DIGI32_PRO) return -EINVAL; rme32->wcreg |= RME32_WCR_DS_BM; rme32->wcreg = (rme32->wcreg | RME32_WCR_FREQ_1) & ~RME32_WCR_FREQ_0; break; case 96000: - if (rme32->pci->device != PCI_DEVICE_ID_DIGI32_PRO) + if (rme32->pci->device != PCI_DEVICE_ID_RME_DIGI32_PRO) return -EINVAL; rme32->wcreg |= RME32_WCR_DS_BM; rme32->wcreg = (rme32->wcreg | RME32_WCR_FREQ_0) | @@ -881,7 +881,7 @@ static int snd_rme32_playback_spdif_open(snd_pcm_substream_t * substream) runtime->hw = snd_rme32_spdif_fd_info; else runtime->hw = snd_rme32_spdif_info; - if (rme32->pci->device == PCI_DEVICE_ID_DIGI32_PRO) { + if (rme32->pci->device == PCI_DEVICE_ID_RME_DIGI32_PRO) { runtime->hw.rates |= SNDRV_PCM_RATE_64000 | SNDRV_PCM_RATE_88200 | SNDRV_PCM_RATE_96000; runtime->hw.rate_max = 96000; } @@ -1408,8 +1408,8 @@ static int __devinit snd_rme32_create(rme32_t * rme32) } /* set up ALSA pcm device for ADAT */ - if ((pci->device == PCI_DEVICE_ID_DIGI32) || - (pci->device == PCI_DEVICE_ID_DIGI32_PRO)) { + if ((pci->device == PCI_DEVICE_ID_RME_DIGI32) || + (pci->device == PCI_DEVICE_ID_RME_DIGI32_PRO)) { /* ADAT is not available on DIGI32 and DIGI32 Pro */ rme32->adat_pcm = NULL; } @@ -1639,11 +1639,11 @@ snd_rme32_info_inputtype_control(snd_kcontrol_t * kcontrol, uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; uinfo->count = 1; switch (rme32->pci->device) { - case PCI_DEVICE_ID_DIGI32: - case PCI_DEVICE_ID_DIGI32_8: + case PCI_DEVICE_ID_RME_DIGI32: + case PCI_DEVICE_ID_RME_DIGI32_8: uinfo->value.enumerated.items = 3; break; - case PCI_DEVICE_ID_DIGI32_PRO: + case PCI_DEVICE_ID_RME_DIGI32_PRO: uinfo->value.enumerated.items = 4; break; default: @@ -1670,11 +1670,11 @@ snd_rme32_get_inputtype_control(snd_kcontrol_t * kcontrol, ucontrol->value.enumerated.item[0] = snd_rme32_getinputtype(rme32); switch (rme32->pci->device) { - case PCI_DEVICE_ID_DIGI32: - case PCI_DEVICE_ID_DIGI32_8: + case PCI_DEVICE_ID_RME_DIGI32: + case PCI_DEVICE_ID_RME_DIGI32_8: items = 3; break; - case PCI_DEVICE_ID_DIGI32_PRO: + case PCI_DEVICE_ID_RME_DIGI32_PRO: items = 4; break; default: @@ -1697,11 +1697,11 @@ snd_rme32_put_inputtype_control(snd_kcontrol_t * kcontrol, int change, items = 3; switch (rme32->pci->device) { - case PCI_DEVICE_ID_DIGI32: - case PCI_DEVICE_ID_DIGI32_8: + case PCI_DEVICE_ID_RME_DIGI32: + case PCI_DEVICE_ID_RME_DIGI32_8: items = 3; break; - case PCI_DEVICE_ID_DIGI32_PRO: + case PCI_DEVICE_ID_RME_DIGI32_PRO: items = 4; break; default: @@ -1982,13 +1982,13 @@ snd_rme32_probe(struct pci_dev *pci, const struct pci_device_id *pci_id) strcpy(card->driver, "Digi32"); switch (rme32->pci->device) { - case PCI_DEVICE_ID_DIGI32: + case PCI_DEVICE_ID_RME_DIGI32: strcpy(card->shortname, "RME Digi32"); break; - case PCI_DEVICE_ID_DIGI32_8: + case PCI_DEVICE_ID_RME_DIGI32_8: strcpy(card->shortname, "RME Digi32/8"); break; - case PCI_DEVICE_ID_DIGI32_PRO: + case PCI_DEVICE_ID_RME_DIGI32_PRO: strcpy(card->shortname, "RME Digi32 PRO"); break; } diff --git a/sound/pci/rme96.c b/sound/pci/rme96.c index 9983b66dc564..c495cae78dbf 100644 --- a/sound/pci/rme96.c +++ b/sound/pci/rme96.c @@ -233,13 +233,13 @@ typedef struct snd_rme96 { } rme96_t; static struct pci_device_id snd_rme96_ids[] = { - { PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_DIGI96, + { PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_RME_DIGI96, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, - { PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_DIGI96_8, + { PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_RME_DIGI96_8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, - { PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_DIGI96_8_PRO, + { PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_RME_DIGI96_8_PRO, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, - { PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST, + { PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, { 0, } }; @@ -248,12 +248,12 @@ MODULE_DEVICE_TABLE(pci, snd_rme96_ids); #define RME96_ISPLAYING(rme96) ((rme96)->wcreg & RME96_WCR_START) #define RME96_ISRECORDING(rme96) ((rme96)->wcreg & RME96_WCR_START_2) -#define RME96_HAS_ANALOG_IN(rme96) ((rme96)->pci->device == PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST) -#define RME96_HAS_ANALOG_OUT(rme96) ((rme96)->pci->device == PCI_DEVICE_ID_DIGI96_8_PRO || \ - (rme96)->pci->device == PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST) +#define RME96_HAS_ANALOG_IN(rme96) ((rme96)->pci->device == PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST) +#define RME96_HAS_ANALOG_OUT(rme96) ((rme96)->pci->device == PCI_DEVICE_ID_RME_DIGI96_8_PRO || \ + (rme96)->pci->device == PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST) #define RME96_DAC_IS_1852(rme96) (RME96_HAS_ANALOG_OUT(rme96) && (rme96)->rev >= 4) -#define RME96_DAC_IS_1855(rme96) (((rme96)->pci->device == PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST && (rme96)->rev < 4) || \ - ((rme96)->pci->device == PCI_DEVICE_ID_DIGI96_8_PRO && (rme96)->rev == 2)) +#define RME96_DAC_IS_1855(rme96) (((rme96)->pci->device == PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST && (rme96)->rev < 4) || \ + ((rme96)->pci->device == PCI_DEVICE_ID_RME_DIGI96_8_PRO && (rme96)->rev == 2)) #define RME96_185X_MAX_OUT(rme96) ((1 << (RME96_DAC_IS_1852(rme96) ? RME96_AD1852_VOL_BITS : RME96_AD1855_VOL_BITS)) - 1) static int @@ -830,9 +830,9 @@ snd_rme96_setinputtype(rme96_t *rme96, RME96_WCR_INP_1; break; case RME96_INPUT_XLR: - if ((rme96->pci->device != PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST && - rme96->pci->device != PCI_DEVICE_ID_DIGI96_8_PRO) || - (rme96->pci->device == PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST && + if ((rme96->pci->device != PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST && + rme96->pci->device != PCI_DEVICE_ID_RME_DIGI96_8_PRO) || + (rme96->pci->device == PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST && rme96->rev > 4)) { /* Only Digi96/8 PRO and Digi96/8 PAD supports XLR */ @@ -1598,7 +1598,7 @@ snd_rme96_create(rme96_t *rme96) rme96->spdif_pcm->info_flags = 0; /* set up ALSA pcm device for ADAT */ - if (pci->device == PCI_DEVICE_ID_DIGI96) { + if (pci->device == PCI_DEVICE_ID_RME_DIGI96) { /* ADAT is not available on the base model */ rme96->adat_pcm = NULL; } else { @@ -1858,14 +1858,14 @@ snd_rme96_info_inputtype_control(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t * uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; uinfo->count = 1; switch (rme96->pci->device) { - case PCI_DEVICE_ID_DIGI96: - case PCI_DEVICE_ID_DIGI96_8: + case PCI_DEVICE_ID_RME_DIGI96: + case PCI_DEVICE_ID_RME_DIGI96_8: uinfo->value.enumerated.items = 3; break; - case PCI_DEVICE_ID_DIGI96_8_PRO: + case PCI_DEVICE_ID_RME_DIGI96_8_PRO: uinfo->value.enumerated.items = 4; break; - case PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST: + case PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST: if (rme96->rev > 4) { /* PST */ uinfo->value.enumerated.items = 4; @@ -1895,14 +1895,14 @@ snd_rme96_get_inputtype_control(snd_kcontrol_t * kcontrol, snd_ctl_elem_value_t ucontrol->value.enumerated.item[0] = snd_rme96_getinputtype(rme96); switch (rme96->pci->device) { - case PCI_DEVICE_ID_DIGI96: - case PCI_DEVICE_ID_DIGI96_8: + case PCI_DEVICE_ID_RME_DIGI96: + case PCI_DEVICE_ID_RME_DIGI96_8: items = 3; break; - case PCI_DEVICE_ID_DIGI96_8_PRO: + case PCI_DEVICE_ID_RME_DIGI96_8_PRO: items = 4; break; - case PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST: + case PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST: if (rme96->rev > 4) { /* for handling PST case, (INPUT_ANALOG is moved to INPUT_XLR */ if (ucontrol->value.enumerated.item[0] == RME96_INPUT_ANALOG) { @@ -1932,14 +1932,14 @@ snd_rme96_put_inputtype_control(snd_kcontrol_t * kcontrol, snd_ctl_elem_value_t int change, items = 3; switch (rme96->pci->device) { - case PCI_DEVICE_ID_DIGI96: - case PCI_DEVICE_ID_DIGI96_8: + case PCI_DEVICE_ID_RME_DIGI96: + case PCI_DEVICE_ID_RME_DIGI96_8: items = 3; break; - case PCI_DEVICE_ID_DIGI96_8_PRO: + case PCI_DEVICE_ID_RME_DIGI96_8_PRO: items = 4; break; - case PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST: + case PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST: if (rme96->rev > 4) { items = 4; } else { @@ -1953,7 +1953,7 @@ snd_rme96_put_inputtype_control(snd_kcontrol_t * kcontrol, snd_ctl_elem_value_t val = ucontrol->value.enumerated.item[0] % items; /* special case for PST */ - if (rme96->pci->device == PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST && rme96->rev > 4) { + if (rme96->pci->device == PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST && rme96->rev > 4) { if (val == RME96_INPUT_XLR) { val = RME96_INPUT_ANALOG; } @@ -2375,16 +2375,16 @@ snd_rme96_probe(struct pci_dev *pci, strcpy(card->driver, "Digi96"); switch (rme96->pci->device) { - case PCI_DEVICE_ID_DIGI96: + case PCI_DEVICE_ID_RME_DIGI96: strcpy(card->shortname, "RME Digi96"); break; - case PCI_DEVICE_ID_DIGI96_8: + case PCI_DEVICE_ID_RME_DIGI96_8: strcpy(card->shortname, "RME Digi96/8"); break; - case PCI_DEVICE_ID_DIGI96_8_PRO: + case PCI_DEVICE_ID_RME_DIGI96_8_PRO: strcpy(card->shortname, "RME Digi96/8 PRO"); break; - case PCI_DEVICE_ID_DIGI96_8_PAD_OR_PST: + case PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST: pci_read_config_byte(rme96->pci, 8, &val); if (val < 5) { strcpy(card->shortname, "RME Digi96/8 PAD"); From ef4cbee0b0d0f791bb593f99b702410f3c0efce6 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Wed, 14 Sep 2005 14:19:18 -0700 Subject: [PATCH 033/240] [PATCH] hvc_console: start kernel thread before registering tty Its possible that we can write to the hvc_console tty as soon it is registered. Recently this started happening due to (what looks like) a change to the hotplug code. Unfortunately at this stage we have not started the khvcd kernel thread and oops. The solution is to start the kernel thread before registering the tty. Signed-off-by: Anton Blanchard Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/hvc_console.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c index cddb789902db..f92177634677 100644 --- a/drivers/char/hvc_console.c +++ b/drivers/char/hvc_console.c @@ -839,9 +839,6 @@ int __init hvc_init(void) hvc_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(hvc_driver, &hvc_ops); - if (tty_register_driver(hvc_driver)) - panic("Couldn't register hvc console driver\n"); - /* Always start the kthread because there can be hotplug vty adapters * added later. */ hvc_task = kthread_run(khvcd, NULL, "khvcd"); @@ -851,6 +848,9 @@ int __init hvc_init(void) return -EIO; } + if (tty_register_driver(hvc_driver)) + panic("Couldn't register hvc console driver\n"); + return 0; } module_init(hvc_init); From 2d93486c6c110cf81db720359b4ec20de9c91450 Mon Sep 17 00:00:00 2001 From: Vincent Sanders Date: Wed, 14 Sep 2005 22:36:03 +0100 Subject: [PATCH 034/240] [ARM] 2907/1: GCC 4 serial driver compile fixes Patch from Vincent Sanders When building the ARM platforms several serial drivers fail to compile with GCC 4.01 due to extern/static ambiguity. Signed-off-by: Vincent Sanders Signed-off-by: Russell King --- drivers/serial/21285.c | 2 +- drivers/serial/amba-pl010.c | 2 +- drivers/serial/amba-pl011.c | 2 +- drivers/serial/clps711x.c | 2 +- drivers/serial/pxa.c | 4 ++-- drivers/serial/sa1100.c | 2 +- drivers/serial/serial_lh7a40x.c | 2 +- 7 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/serial/21285.c b/drivers/serial/21285.c index aec39fb261ca..b5cf39468d18 100644 --- a/drivers/serial/21285.c +++ b/drivers/serial/21285.c @@ -463,7 +463,7 @@ static int __init serial21285_console_setup(struct console *co, char *options) return uart_set_options(port, co, baud, parity, bits, flow); } -extern struct uart_driver serial21285_reg; +static struct uart_driver serial21285_reg; static struct console serial21285_console = { diff --git a/drivers/serial/amba-pl010.c b/drivers/serial/amba-pl010.c index 978e12437e61..679e678c7e6a 100644 --- a/drivers/serial/amba-pl010.c +++ b/drivers/serial/amba-pl010.c @@ -689,7 +689,7 @@ static int __init pl010_console_setup(struct console *co, char *options) return uart_set_options(port, co, baud, parity, bits, flow); } -extern struct uart_driver amba_reg; +static struct uart_driver amba_reg; static struct console amba_console = { .name = "ttyAM", .write = pl010_console_write, diff --git a/drivers/serial/amba-pl011.c b/drivers/serial/amba-pl011.c index 56071309744c..1ff629c74750 100644 --- a/drivers/serial/amba-pl011.c +++ b/drivers/serial/amba-pl011.c @@ -701,7 +701,7 @@ static int __init pl011_console_setup(struct console *co, char *options) return uart_set_options(&uap->port, co, baud, parity, bits, flow); } -extern struct uart_driver amba_reg; +static struct uart_driver amba_reg; static struct console amba_console = { .name = "ttyAMA", .write = pl011_console_write, diff --git a/drivers/serial/clps711x.c b/drivers/serial/clps711x.c index d822896b488c..78c1f36ad9b7 100644 --- a/drivers/serial/clps711x.c +++ b/drivers/serial/clps711x.c @@ -525,7 +525,7 @@ static int __init clps711xuart_console_setup(struct console *co, char *options) return uart_set_options(port, co, baud, parity, bits, flow); } -extern struct uart_driver clps711x_reg; +static struct uart_driver clps711x_reg; static struct console clps711x_console = { .name = "ttyCL", .write = clps711xuart_console_write, diff --git a/drivers/serial/pxa.c b/drivers/serial/pxa.c index eaa0af835290..672b359b07ce 100644 --- a/drivers/serial/pxa.c +++ b/drivers/serial/pxa.c @@ -589,8 +589,8 @@ serial_pxa_type(struct uart_port *port) #ifdef CONFIG_SERIAL_PXA_CONSOLE -extern struct uart_pxa_port serial_pxa_ports[]; -extern struct uart_driver serial_pxa_reg; +static struct uart_pxa_port serial_pxa_ports[]; +static struct uart_driver serial_pxa_reg; #define BOTH_EMPTY (UART_LSR_TEMT | UART_LSR_THRE) diff --git a/drivers/serial/sa1100.c b/drivers/serial/sa1100.c index 1225b14f6e9d..dd8aed242357 100644 --- a/drivers/serial/sa1100.c +++ b/drivers/serial/sa1100.c @@ -799,7 +799,7 @@ sa1100_console_setup(struct console *co, char *options) return uart_set_options(&sport->port, co, baud, parity, bits, flow); } -extern struct uart_driver sa1100_reg; +static struct uart_driver sa1100_reg; static struct console sa1100_console = { .name = "ttySA", .write = sa1100_console_write, diff --git a/drivers/serial/serial_lh7a40x.c b/drivers/serial/serial_lh7a40x.c index 8302376800c0..d01dbe5da3b9 100644 --- a/drivers/serial/serial_lh7a40x.c +++ b/drivers/serial/serial_lh7a40x.c @@ -632,7 +632,7 @@ static int __init lh7a40xuart_console_setup (struct console* co, char* options) return uart_set_options (port, co, baud, parity, bits, flow); } -extern struct uart_driver lh7a40x_reg; +static struct uart_driver lh7a40x_reg; static struct console lh7a40x_console = { .name = "ttyAM", .write = lh7a40xuart_console_write, From 2cef2d552f75b7feb7859d29a71388c020023872 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Wed, 14 Sep 2005 22:42:31 +0100 Subject: [PATCH 035/240] [ARM] 2910/1: missing Lubbock audio device declaration Patch from Nicolas Pitre This apparently fell in the crack somewhere. Add it back. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/mach-pxa/lubbock.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/arm/mach-pxa/lubbock.c b/arch/arm/mach-pxa/lubbock.c index 923f6eb774c0..1f38033921e9 100644 --- a/arch/arm/mach-pxa/lubbock.c +++ b/arch/arm/mach-pxa/lubbock.c @@ -146,6 +146,11 @@ static struct pxa2xx_udc_mach_info udc_info __initdata = { // no D+ pullup; lubbock can't connect/disconnect in software }; +static struct platform_device lub_audio_device = { + .name = "pxa2xx-ac97", + .id = -1, +}; + static struct resource sa1111_resources[] = { [0] = { .start = 0x10000000, @@ -195,6 +200,7 @@ static struct platform_device smc91x_device = { static struct platform_device *devices[] __initdata = { &sa1111_device, + &lub_audio_device, &smc91x_device, }; From da16e3245a4827d969943ca0493a4a6e34009342 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 14 Sep 2005 22:54:45 +0100 Subject: [PATCH 036/240] [I2C] Add a functionality method, and remove algorithm ids This allows i2c-pxa to finally build. Signed-off-by: Russell King --- drivers/i2c/busses/i2c-pxa.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c index fdf53ce04248..44b595d90a4a 100644 --- a/drivers/i2c/busses/i2c-pxa.c +++ b/drivers/i2c/busses/i2c-pxa.c @@ -914,19 +914,23 @@ static int i2c_pxa_xfer(struct i2c_adapter *adap, struct i2c_msg msgs[], int num return ret; } +static u32 i2c_pxa_functionality(struct i2c_adapter *adap) +{ + return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; +} + static struct i2c_algorithm i2c_pxa_algorithm = { - .name = "PXA-I2C-Algorithm", - .id = I2C_ALGO_PXA, .master_xfer = i2c_pxa_xfer, + .functionality = i2c_pxa_functionality, }; static struct pxa_i2c i2c_pxa = { .lock = SPIN_LOCK_UNLOCKED, .wait = __WAIT_QUEUE_HEAD_INITIALIZER(i2c_pxa.wait), .adap = { - .name = "pxa2xx-i2c", - .id = I2C_ALGO_PXA, + .owner = THIS_MODULE, .algo = &i2c_pxa_algorithm, + .name = "pxa2xx-i2c", .retries = 5, }, }; From f6af5da388a5d058cd9a5a47d030e3966b22e991 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 14 Sep 2005 23:10:48 +0100 Subject: [PATCH 037/240] [ARM SMP] Add timer/watchdog defines for MPCore Actually add the file this time. Signed-off-by: Russell King --- include/asm-arm/hardware/arm_twd.h | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) create mode 100644 include/asm-arm/hardware/arm_twd.h diff --git a/include/asm-arm/hardware/arm_twd.h b/include/asm-arm/hardware/arm_twd.h new file mode 100644 index 000000000000..131d5b40e072 --- /dev/null +++ b/include/asm-arm/hardware/arm_twd.h @@ -0,0 +1,16 @@ +#ifndef __ASM_HARDWARE_TWD_H +#define __ASM_HARDWARE_TWD_H + +#define TWD_TIMER_LOAD 0x00 +#define TWD_TIMER_COUNTER 0x04 +#define TWD_TIMER_CONTROL 0x08 +#define TWD_TIMER_INTSTAT 0x0C + +#define TWD_WDOG_LOAD 0x20 +#define TWD_WDOG_COUNTER 0x24 +#define TWD_WDOG_CONTROL 0x28 +#define TWD_WDOG_INTSTAT 0x2C +#define TWD_WDOG_RESETSTAT 0x30 +#define TWD_WDOG_DISABLE 0x34 + +#endif From 1619cca2921f6927f4240e03f413d4165c7002fc Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 14 Sep 2005 15:56:27 -0700 Subject: [PATCH 038/240] Partially revert "Fix time going twice as fast problem on ATI Xpress chipsets" Commit 66759a01adbfe8828dd063e32cf5ed3f46696181 introduced the fix for time ticking too fast on some boards by disabling one of the doubly connected timer pins on ATI boards. However, it ends up being _much_ too broad a brush, and that just makes some other ATI boards not work at all since they now have no timer source. So disable the automatic ATI southbridge detection, and just rely on people who see this problem disabling it by hand with the option "disable_timer_pin_1" on the kernel command line. Maybe somebody can figure out the proper tests at a later date. Acked-by: Peter Osterlund Signed-off-by: Linus Torvalds --- arch/i386/kernel/acpi/earlyquirk.c | 10 ---------- arch/x86_64/kernel/io_apic.c | 9 --------- 2 files changed, 19 deletions(-) diff --git a/arch/i386/kernel/acpi/earlyquirk.c b/arch/i386/kernel/acpi/earlyquirk.c index 1ae2aeeda18b..f1b9d2a46dab 100644 --- a/arch/i386/kernel/acpi/earlyquirk.c +++ b/arch/i386/kernel/acpi/earlyquirk.c @@ -7,7 +7,6 @@ #include #include #include -#include static int __init check_bridge(int vendor, int device) { @@ -16,15 +15,6 @@ static int __init check_bridge(int vendor, int device) if (vendor == PCI_VENDOR_ID_NVIDIA) { acpi_skip_timer_override = 1; } -#ifdef CONFIG_X86_LOCAL_APIC - /* - * ATI IXP chipsets get double timer interrupts. - * For now just do this for all ATI chipsets. - * FIXME: this needs to be checked for the non ACPI case too. - */ - if (vendor == PCI_VENDOR_ID_ATI) - disable_timer_pin_1 = 1; -#endif return 0; } diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 522944a000ad..c8eee20cd519 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -299,15 +299,6 @@ void __init check_ioapic(void) #endif /* RED-PEN skip them on mptables too? */ return; - case PCI_VENDOR_ID_ATI: - /* All timer interrupts on atiixp - are doubled. Disable one. */ - if (disable_timer_pin_1 == 0) { - disable_timer_pin_1 = 1; - printk(KERN_INFO - "ATI board detected. Disabling timer pin 1.\n"); - } - return; } /* No multi-function device? */ From 3c05d92ed49f644d1f5a960fa48637d63b946016 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 14 Sep 2005 20:50:35 -0700 Subject: [PATCH 039/240] [TCP]: Compute in_sacked properly when we split up a TSO frame. The problem is that the SACK fragmenting code may incorrectly call tcp_fragment() with a length larger than the skb->len. This happens when the skb on the transmit queue completely falls to the LHS of the SACK. And add a BUG() check to tcp_fragment() so we can spot this kind of error more quickly in the future. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 16 +++++++++------- net/ipv4/tcp_output.c | 2 ++ 2 files changed, 11 insertions(+), 7 deletions(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 29222b964951..a7537c7bbd06 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -979,14 +979,19 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ if (!before(TCP_SKB_CB(skb)->seq, end_seq)) break; + in_sack = !after(start_seq, TCP_SKB_CB(skb)->seq) && + !before(end_seq, TCP_SKB_CB(skb)->end_seq); + pcount = tcp_skb_pcount(skb); - if (pcount > 1 && - (after(start_seq, TCP_SKB_CB(skb)->seq) || - before(end_seq, TCP_SKB_CB(skb)->end_seq))) { + if (pcount > 1 && !in_sack && + after(TCP_SKB_CB(skb)->end_seq, start_seq)) { unsigned int pkt_len; - if (after(start_seq, TCP_SKB_CB(skb)->seq)) + in_sack = !after(start_seq, + TCP_SKB_CB(skb)->seq); + + if (!in_sack) pkt_len = (start_seq - TCP_SKB_CB(skb)->seq); else @@ -999,9 +1004,6 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ fack_count += pcount; - in_sack = !after(start_seq, TCP_SKB_CB(skb)->seq) && - !before(end_seq, TCP_SKB_CB(skb)->end_seq); - sacked = TCP_SKB_CB(skb)->sacked; /* Account D-SACK for retransmitted packet. */ diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index c10e4435e3b1..b018e31b6530 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -435,6 +435,8 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss int nsize, old_factor; u16 flags; + BUG_ON(len >= skb->len); + nsize = skb_headlen(skb) - len; if (nsize < 0) nsize = 0; From 20ae975dfd54de581287b2ca8a1ad97099ab0396 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 14 Sep 2005 20:52:37 -0700 Subject: [PATCH 040/240] [NETLINK]: Reserve a slot for NETLINK_GENERIC. As requested by Jamal. Signed-off-by: David S. Miller --- include/linux/netlink.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 7bbd25970c9e..bdebdc564506 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -20,6 +20,7 @@ #define NETLINK_IP6_FW 13 #define NETLINK_DNRTMSG 14 /* DECnet routing messages */ #define NETLINK_KOBJECT_UEVENT 15 /* Kernel messages to userspace */ +#define NETLINK_GENERIC 16 #define MAX_LINKS 32 From de9daad90ecb54f3c37c3f8967d581e20d927539 Mon Sep 17 00:00:00 2001 From: Denis Lukianov Date: Wed, 14 Sep 2005 20:53:42 -0700 Subject: [PATCH 041/240] [MCAST]: Fix MCAST_EXCLUDE line dupes This patch fixes line dupes at /ipv4/igmp.c and /ipv6/mcast.c in the 2.6 kernel, where MCAST_EXCLUDE is mistakenly used instead of MCAST_INCLUDE. Signed-off-by: Denis Lukianov Signed-off-by: David L Stevens Signed-off-by: David S. Miller --- net/ipv4/igmp.c | 2 +- net/ipv6/mcast.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 44607f4767b8..70c44e4c3ceb 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1603,7 +1603,7 @@ static void ip_mc_clear_src(struct ip_mc_list *pmc) } pmc->sources = NULL; pmc->sfmode = MCAST_EXCLUDE; - pmc->sfcount[MCAST_EXCLUDE] = 0; + pmc->sfcount[MCAST_INCLUDE] = 0; pmc->sfcount[MCAST_EXCLUDE] = 1; } diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 29fed6e58d0a..519899fb11d5 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1968,7 +1968,7 @@ static void ip6_mc_clear_src(struct ifmcaddr6 *pmc) } pmc->mca_sources = NULL; pmc->mca_sfmode = MCAST_EXCLUDE; - pmc->mca_sfcount[MCAST_EXCLUDE] = 0; + pmc->mca_sfcount[MCAST_INCLUDE] = 0; pmc->mca_sfcount[MCAST_EXCLUDE] = 1; } From 1c011bed5f49ce9e6193b18b226106a41ecfa95c Mon Sep 17 00:00:00 2001 From: Bart De Schuymer Date: Wed, 14 Sep 2005 20:55:16 -0700 Subject: [PATCH 042/240] [BRIDGE-NF]: Fix iptables redirect on bridge interface Here's a slightly altered patch, originally from Mark Glines who diagnosed and fixed the problem. Signed-off-by: Bart De Schuymer Signed-off-by: David S. Miller --- net/bridge/br_netfilter.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index 2d52fee63a8c..d8e36b775125 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -214,9 +214,11 @@ static int br_nf_pre_routing_finish(struct sk_buff *skb) .tos = RT_TOS(iph->tos)} }, .proto = 0}; if (!ip_route_output_key(&rt, &fl)) { - /* Bridged-and-DNAT'ed traffic doesn't - * require ip_forwarding. */ - if (((struct dst_entry *)rt)->dev == dev) { + /* - Bridged-and-DNAT'ed traffic doesn't + * require ip_forwarding. + * - Deal with redirected traffic. */ + if (((struct dst_entry *)rt)->dev == dev || + rt->rt_type == RTN_LOCAL) { skb->dst = (struct dst_entry *)rt; goto bridged_dnat; } From f5e229db9cdb27f83594712ca4bb98d9377eb6ed Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Wed, 14 Sep 2005 21:04:23 -0700 Subject: [PATCH 043/240] [IPVS]: Really invalidate persistent templates Agostino di Salle noticed that persistent templates are not invalidated due to buggy optimization. Signed-off-by: Julian Anastasov Signed-off-by: David S. Miller --- net/ipv4/ipvs/ip_vs_conn.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/ipvs/ip_vs_conn.c b/net/ipv4/ipvs/ip_vs_conn.c index e11952ea17af..5994521fddc5 100644 --- a/net/ipv4/ipvs/ip_vs_conn.c +++ b/net/ipv4/ipvs/ip_vs_conn.c @@ -467,7 +467,7 @@ int ip_vs_check_template(struct ip_vs_conn *ct) /* * Invalidate the connection template */ - if (ct->cport) { + if (ct->vport != 65535) { if (ip_vs_conn_unhash(ct)) { ct->dport = 65535; ct->vport = 65535; From 87375ab47cd0ba04124c6d3fd80db5c368f5dcb6 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Wed, 14 Sep 2005 21:08:51 -0700 Subject: [PATCH 044/240] [IPVS]: ip_vs_ftp breaks connections using persistence ip_vs_ftp when loaded can create NAT connections with unknown client port for passive FTP. For such expectations we lookup with cport=0 on incoming packet but it matches the format of the persistence templates causing packets to other persistent virtual servers to be forwarded to real server without creating connection. Later the reply packets are treated as foreign and not SNAT-ed. This patch changes the connection lookup for packets from clients: * introduce IP_VS_CONN_F_TEMPLATE connection flag to mark the connection as template * create new connection lookup function just for templates - ip_vs_ct_in_get * make sure ip_vs_conn_in_get hits only connections with IP_VS_CONN_F_NO_CPORT flag set when s_port is 0. By this way we avoid returning template when looking for cport=0 (ftp) Signed-off-by: Julian Anastasov Signed-off-by: David S. Miller --- include/net/ip_vs.h | 3 +++ net/ipv4/ipvs/ip_vs_conn.c | 41 +++++++++++++++++++++++++++++++++++--- net/ipv4/ipvs/ip_vs_core.c | 16 +++++++-------- net/ipv4/ipvs/ip_vs_sync.c | 20 +++++++++++++------ 4 files changed, 63 insertions(+), 17 deletions(-) diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index e426641c519f..06b4235aa016 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -84,6 +84,7 @@ #define IP_VS_CONN_F_IN_SEQ 0x0400 /* must do input seq adjust */ #define IP_VS_CONN_F_SEQ_MASK 0x0600 /* in/out sequence mask */ #define IP_VS_CONN_F_NO_CPORT 0x0800 /* no client port set yet */ +#define IP_VS_CONN_F_TEMPLATE 0x1000 /* template, not connection */ /* Move it to better place one day, for now keep it unique */ #define NFC_IPVS_PROPERTY 0x10000 @@ -739,6 +740,8 @@ enum { extern struct ip_vs_conn *ip_vs_conn_in_get (int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port); +extern struct ip_vs_conn *ip_vs_ct_in_get +(int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port); extern struct ip_vs_conn *ip_vs_conn_out_get (int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port); diff --git a/net/ipv4/ipvs/ip_vs_conn.c b/net/ipv4/ipvs/ip_vs_conn.c index 5994521fddc5..f828fa2eb7de 100644 --- a/net/ipv4/ipvs/ip_vs_conn.c +++ b/net/ipv4/ipvs/ip_vs_conn.c @@ -196,6 +196,7 @@ static inline struct ip_vs_conn *__ip_vs_conn_in_get list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { if (s_addr==cp->caddr && s_port==cp->cport && d_port==cp->vport && d_addr==cp->vaddr && + ((!s_port) ^ (!(cp->flags & IP_VS_CONN_F_NO_CPORT))) && protocol==cp->protocol) { /* HIT */ atomic_inc(&cp->refcnt); @@ -227,6 +228,40 @@ struct ip_vs_conn *ip_vs_conn_in_get return cp; } +/* Get reference to connection template */ +struct ip_vs_conn *ip_vs_ct_in_get +(int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port) +{ + unsigned hash; + struct ip_vs_conn *cp; + + hash = ip_vs_conn_hashkey(protocol, s_addr, s_port); + + ct_read_lock(hash); + + list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { + if (s_addr==cp->caddr && s_port==cp->cport && + d_port==cp->vport && d_addr==cp->vaddr && + cp->flags & IP_VS_CONN_F_TEMPLATE && + protocol==cp->protocol) { + /* HIT */ + atomic_inc(&cp->refcnt); + goto out; + } + } + cp = NULL; + + out: + ct_read_unlock(hash); + + IP_VS_DBG(7, "template lookup/in %s %u.%u.%u.%u:%d->%u.%u.%u.%u:%d %s\n", + ip_vs_proto_name(protocol), + NIPQUAD(s_addr), ntohs(s_port), + NIPQUAD(d_addr), ntohs(d_port), + cp?"hit":"not hit"); + + return cp; +} /* * Gets ip_vs_conn associated with supplied parameters in the ip_vs_conn_tab. @@ -367,7 +402,7 @@ ip_vs_bind_dest(struct ip_vs_conn *cp, struct ip_vs_dest *dest) atomic_read(&dest->refcnt)); /* Update the connection counters */ - if (cp->cport || (cp->flags & IP_VS_CONN_F_NO_CPORT)) { + if (!(cp->flags & IP_VS_CONN_F_TEMPLATE)) { /* It is a normal connection, so increase the inactive connection counter because it is in TCP SYNRECV state (inactive) or other protocol inacive state */ @@ -406,7 +441,7 @@ static inline void ip_vs_unbind_dest(struct ip_vs_conn *cp) atomic_read(&dest->refcnt)); /* Update the connection counters */ - if (cp->cport || (cp->flags & IP_VS_CONN_F_NO_CPORT)) { + if (!(cp->flags & IP_VS_CONN_F_TEMPLATE)) { /* It is a normal connection, so decrease the inactconns or activeconns counter */ if (cp->flags & IP_VS_CONN_F_INACTIVE) { @@ -776,7 +811,7 @@ void ip_vs_random_dropentry(void) ct_write_lock_bh(hash); list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { - if (!cp->cport && !(cp->flags & IP_VS_CONN_F_NO_CPORT)) + if (cp->flags & IP_VS_CONN_F_TEMPLATE) /* connection template */ continue; diff --git a/net/ipv4/ipvs/ip_vs_core.c b/net/ipv4/ipvs/ip_vs_core.c index 3ac7eeca04ac..981cc3244ef2 100644 --- a/net/ipv4/ipvs/ip_vs_core.c +++ b/net/ipv4/ipvs/ip_vs_core.c @@ -243,10 +243,10 @@ ip_vs_sched_persist(struct ip_vs_service *svc, if (ports[1] == svc->port) { /* Check if a template already exists */ if (svc->port != FTPPORT) - ct = ip_vs_conn_in_get(iph->protocol, snet, 0, + ct = ip_vs_ct_in_get(iph->protocol, snet, 0, iph->daddr, ports[1]); else - ct = ip_vs_conn_in_get(iph->protocol, snet, 0, + ct = ip_vs_ct_in_get(iph->protocol, snet, 0, iph->daddr, 0); if (!ct || !ip_vs_check_template(ct)) { @@ -272,14 +272,14 @@ ip_vs_sched_persist(struct ip_vs_service *svc, iph->daddr, ports[1], dest->addr, dest->port, - 0, + IP_VS_CONN_F_TEMPLATE, dest); else ct = ip_vs_conn_new(iph->protocol, snet, 0, iph->daddr, 0, dest->addr, 0, - 0, + IP_VS_CONN_F_TEMPLATE, dest); if (ct == NULL) return NULL; @@ -298,10 +298,10 @@ ip_vs_sched_persist(struct ip_vs_service *svc, * port zero template: */ if (svc->fwmark) - ct = ip_vs_conn_in_get(IPPROTO_IP, snet, 0, + ct = ip_vs_ct_in_get(IPPROTO_IP, snet, 0, htonl(svc->fwmark), 0); else - ct = ip_vs_conn_in_get(iph->protocol, snet, 0, + ct = ip_vs_ct_in_get(iph->protocol, snet, 0, iph->daddr, 0); if (!ct || !ip_vs_check_template(ct)) { @@ -326,14 +326,14 @@ ip_vs_sched_persist(struct ip_vs_service *svc, snet, 0, htonl(svc->fwmark), 0, dest->addr, 0, - 0, + IP_VS_CONN_F_TEMPLATE, dest); else ct = ip_vs_conn_new(iph->protocol, snet, 0, iph->daddr, 0, dest->addr, 0, - 0, + IP_VS_CONN_F_TEMPLATE, dest); if (ct == NULL) return NULL; diff --git a/net/ipv4/ipvs/ip_vs_sync.c b/net/ipv4/ipvs/ip_vs_sync.c index 574d1f509b46..2e5ced3d8062 100644 --- a/net/ipv4/ipvs/ip_vs_sync.c +++ b/net/ipv4/ipvs/ip_vs_sync.c @@ -297,16 +297,24 @@ static void ip_vs_process_message(const char *buffer, const size_t buflen) p = (char *)buffer + sizeof(struct ip_vs_sync_mesg); for (i=0; inr_conns; i++) { + unsigned flags; + s = (struct ip_vs_sync_conn *)p; - cp = ip_vs_conn_in_get(s->protocol, - s->caddr, s->cport, - s->vaddr, s->vport); + flags = ntohs(s->flags); + if (!(flags & IP_VS_CONN_F_TEMPLATE)) + cp = ip_vs_conn_in_get(s->protocol, + s->caddr, s->cport, + s->vaddr, s->vport); + else + cp = ip_vs_ct_in_get(s->protocol, + s->caddr, s->cport, + s->vaddr, s->vport); if (!cp) { cp = ip_vs_conn_new(s->protocol, s->caddr, s->cport, s->vaddr, s->vport, s->daddr, s->dport, - ntohs(s->flags), NULL); + flags, NULL); if (!cp) { IP_VS_ERR("ip_vs_conn_new failed\n"); return; @@ -315,11 +323,11 @@ static void ip_vs_process_message(const char *buffer, const size_t buflen) } else if (!cp->dest) { /* it is an entry created by the synchronization */ cp->state = ntohs(s->state); - cp->flags = ntohs(s->flags) | IP_VS_CONN_F_HASHED; + cp->flags = flags | IP_VS_CONN_F_HASHED; } /* Note that we don't touch its state and flags if it is a normal entry. */ - if (ntohs(s->flags) & IP_VS_CONN_F_SEQ_MASK) { + if (flags & IP_VS_CONN_F_SEQ_MASK) { opt = (struct ip_vs_sync_conn_options *)&s[1]; memcpy(&cp->in_seq, opt, sizeof(*opt)); p += FULL_CONN_SIZE; From 40796c5e8f2a93008e9034b3110a7e7b1fa0fba0 Mon Sep 17 00:00:00 2001 From: David L Stevens Date: Wed, 14 Sep 2005 21:10:20 -0700 Subject: [PATCH 045/240] [IPV6]: Fix per-socket multicast filtering in sk_reuse case per-socket multicast filters were not being applied to all sockets in the case of an exact-match bound address, due to an over-exuberant "return" in the look-up code. Fix below. IPv4 does not have this problem. Thanks to Hoerdt Mickael for reporting the bug. Signed-off-by: David L Stevens Signed-off-by: David S. Miller --- net/ipv6/udp.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 69b146843a20..2b9bf9bd177f 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -405,9 +405,8 @@ static struct sock *udp_v6_mcast_next(struct sock *sk, continue; if (!ipv6_addr_any(&np->rcv_saddr)) { - if (ipv6_addr_equal(&np->rcv_saddr, loc_addr)) - return s; - continue; + if (!ipv6_addr_equal(&np->rcv_saddr, loc_addr)) + continue; } if(!inet6_mc_check(s, loc_addr, rmt_addr)) continue; From 4a805e863d6b9466baf7084e1d6fdbe6e0628d8e Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 14 Sep 2005 21:40:00 -0700 Subject: [PATCH 046/240] [COMPAT]: Fixup compat_do_execve() Missing acct_update_integrals() and update_mem_hiwater() calls compared to it's native counterpart. Signed-off-by: David S. Miller --- fs/compat.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/compat.c b/fs/compat.c index ac3fb9ed8eea..a719e158e002 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -44,6 +44,8 @@ #include #include #include +#include +#include #include /* siocdevprivate_ioctl */ @@ -1487,6 +1489,8 @@ int compat_do_execve(char * filename, /* execve success */ security_bprm_free(bprm); + acct_update_integrals(current); + update_mem_hiwater(current); kfree(bprm); return retval; } From 4db2ce0199f04b6e99999f22e28ef9a0ae5f0d2f Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 14 Sep 2005 21:47:01 -0700 Subject: [PATCH 047/240] [LIB]: Consolidate _atomic_dec_and_lock() Several implementations were essentialy a common piece of C code using the cmpxchg() macro. Put the implementation in one spot that everyone can share, and convert sparc64 over to using this. Alpha is the lone arch-specific implementation, which codes up a special fast path for the common case in order to avoid GP reloading which a pure C version would require. Signed-off-by: David S. Miller --- arch/i386/Kconfig | 5 -- arch/i386/lib/Makefile | 1 - arch/i386/lib/dec_and_lock.c | 42 --------------- arch/ia64/Kconfig | 5 -- arch/ia64/lib/Makefile | 1 - arch/ia64/lib/dec_and_lock.c | 42 --------------- arch/m32r/Kconfig | 5 -- arch/mips/Kconfig | 4 -- arch/mips/lib/Makefile | 2 +- arch/mips/lib/dec_and_lock.c | 47 ----------------- arch/ppc/Kconfig | 4 -- arch/ppc/lib/Makefile | 2 +- arch/ppc/lib/dec_and_lock.c | 38 -------------- arch/ppc64/Kconfig | 4 -- arch/ppc64/lib/Makefile | 2 +- arch/ppc64/lib/dec_and_lock.c | 47 ----------------- arch/sparc64/Kconfig.debug | 8 --- arch/sparc64/kernel/sparc64_ksyms.c | 3 -- arch/sparc64/lib/Makefile | 2 - arch/sparc64/lib/dec_and_lock.S | 80 ----------------------------- arch/x86_64/Kconfig | 5 -- arch/x86_64/kernel/x8664_ksyms.c | 4 -- arch/x86_64/lib/Makefile | 2 - arch/x86_64/lib/dec_and_lock.c | 40 --------------- arch/xtensa/Kconfig | 4 -- lib/dec_and_lock.c | 35 +++++++++++++ 26 files changed, 38 insertions(+), 396 deletions(-) delete mode 100644 arch/i386/lib/dec_and_lock.c delete mode 100644 arch/ia64/lib/dec_and_lock.c delete mode 100644 arch/mips/lib/dec_and_lock.c delete mode 100644 arch/ppc/lib/dec_and_lock.c delete mode 100644 arch/ppc64/lib/dec_and_lock.c delete mode 100644 arch/sparc64/lib/dec_and_lock.S delete mode 100644 arch/x86_64/lib/dec_and_lock.c diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index b22f003eaa6d..d2703cda61ea 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -908,11 +908,6 @@ config IRQBALANCE The default yes will allow the kernel to do irq load balancing. Saying no will keep the kernel from doing irq load balancing. -config HAVE_DEC_LOCK - bool - depends on (SMP || PREEMPT) && X86_CMPXCHG - default y - # turning this on wastes a bunch of space. # Summit needs it only when NUMA is on config BOOT_IOREMAP diff --git a/arch/i386/lib/Makefile b/arch/i386/lib/Makefile index 7b1932d20f96..914933e9ec3d 100644 --- a/arch/i386/lib/Makefile +++ b/arch/i386/lib/Makefile @@ -7,4 +7,3 @@ lib-y = checksum.o delay.o usercopy.o getuser.o putuser.o memcpy.o strstr.o \ bitops.o lib-$(CONFIG_X86_USE_3DNOW) += mmx.o -lib-$(CONFIG_HAVE_DEC_LOCK) += dec_and_lock.o diff --git a/arch/i386/lib/dec_and_lock.c b/arch/i386/lib/dec_and_lock.c deleted file mode 100644 index 8b81b2524fa6..000000000000 --- a/arch/i386/lib/dec_and_lock.c +++ /dev/null @@ -1,42 +0,0 @@ -/* - * x86 version of "atomic_dec_and_lock()" using - * the atomic "cmpxchg" instruction. - * - * (For CPU's lacking cmpxchg, we use the slow - * generic version, and this one never even gets - * compiled). - */ - -#include -#include -#include - -int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock) -{ - int counter; - int newcount; - -repeat: - counter = atomic_read(atomic); - newcount = counter-1; - - if (!newcount) - goto slow_path; - - asm volatile("lock; cmpxchgl %1,%2" - :"=a" (newcount) - :"r" (newcount), "m" (atomic->counter), "0" (counter)); - - /* If the above failed, "eax" will have changed */ - if (newcount != counter) - goto repeat; - return 0; - -slow_path: - spin_lock(lock); - if (atomic_dec_and_test(atomic)) - return 1; - spin_unlock(lock); - return 0; -} -EXPORT_SYMBOL(_atomic_dec_and_lock); diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index ed25d66c8d50..945c15a0722b 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -298,11 +298,6 @@ config PREEMPT source "mm/Kconfig" -config HAVE_DEC_LOCK - bool - depends on (SMP || PREEMPT) - default y - config IA32_SUPPORT bool "Support for Linux/x86 binaries" help diff --git a/arch/ia64/lib/Makefile b/arch/ia64/lib/Makefile index 799407e7726f..cb1af597370b 100644 --- a/arch/ia64/lib/Makefile +++ b/arch/ia64/lib/Makefile @@ -15,7 +15,6 @@ lib-$(CONFIG_ITANIUM) += copy_page.o copy_user.o memcpy.o lib-$(CONFIG_MCKINLEY) += copy_page_mck.o memcpy_mck.o lib-$(CONFIG_PERFMON) += carta_random.o lib-$(CONFIG_MD_RAID5) += xor.o -lib-$(CONFIG_HAVE_DEC_LOCK) += dec_and_lock.o AFLAGS___divdi3.o = AFLAGS___udivdi3.o = -DUNSIGNED diff --git a/arch/ia64/lib/dec_and_lock.c b/arch/ia64/lib/dec_and_lock.c deleted file mode 100644 index c7ce92f968f1..000000000000 --- a/arch/ia64/lib/dec_and_lock.c +++ /dev/null @@ -1,42 +0,0 @@ -/* - * Copyright (C) 2003 Jerome Marchand, Bull S.A. - * Cleaned up by David Mosberger-Tang - * - * This file is released under the GPLv2, or at your option any later version. - * - * ia64 version of "atomic_dec_and_lock()" using the atomic "cmpxchg" instruction. This - * code is an adaptation of the x86 version of "atomic_dec_and_lock()". - */ - -#include -#include -#include -#include - -/* - * Decrement REFCOUNT and if the count reaches zero, acquire the spinlock. Both of these - * operations have to be done atomically, so that the count doesn't drop to zero without - * acquiring the spinlock first. - */ -int -_atomic_dec_and_lock (atomic_t *refcount, spinlock_t *lock) -{ - int old, new; - - do { - old = atomic_read(refcount); - new = old - 1; - - if (unlikely (old == 1)) { - /* oops, we may be decrementing to zero, do it the slow way... */ - spin_lock(lock); - if (atomic_dec_and_test(refcount)) - return 1; - spin_unlock(lock); - return 0; - } - } while (cmpxchg(&refcount->counter, old, new) != old); - return 0; -} - -EXPORT_SYMBOL(_atomic_dec_and_lock); diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig index 1ef3987ebc6a..4d100f3886e1 100644 --- a/arch/m32r/Kconfig +++ b/arch/m32r/Kconfig @@ -220,11 +220,6 @@ config PREEMPT Say Y here if you are building a kernel for a desktop, embedded or real-time system. Say N if you are unsure. -config HAVE_DEC_LOCK - bool - depends on (SMP || PREEMPT) - default n - config SMP bool "Symmetric multi-processing support" ---help--- diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 0eb71ac303af..4cd724c05700 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -1009,10 +1009,6 @@ config GENERIC_CALIBRATE_DELAY bool default y -config HAVE_DEC_LOCK - bool - default y - # # Select some configuration options automatically based on user selections. # diff --git a/arch/mips/lib/Makefile b/arch/mips/lib/Makefile index 21b92b9dd013..037303412909 100644 --- a/arch/mips/lib/Makefile +++ b/arch/mips/lib/Makefile @@ -2,7 +2,7 @@ # Makefile for MIPS-specific library files.. # -lib-y += csum_partial_copy.o dec_and_lock.o memcpy.o promlib.o \ +lib-y += csum_partial_copy.o memcpy.o promlib.o \ strlen_user.o strncpy_user.o strnlen_user.o obj-y += iomap.o diff --git a/arch/mips/lib/dec_and_lock.c b/arch/mips/lib/dec_and_lock.c deleted file mode 100644 index fd82c84a93b7..000000000000 --- a/arch/mips/lib/dec_and_lock.c +++ /dev/null @@ -1,47 +0,0 @@ -/* - * MIPS version of atomic_dec_and_lock() using cmpxchg - * - * 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. - */ - -#include -#include -#include -#include - -/* - * This is an implementation of the notion of "decrement a - * reference count, and return locked if it decremented to zero". - * - * This implementation can be used on any architecture that - * has a cmpxchg, and where atomic->value is an int holding - * the value of the atomic (i.e. the high bits aren't used - * for a lock or anything like that). - */ -int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock) -{ - int counter; - int newcount; - - for (;;) { - counter = atomic_read(atomic); - newcount = counter - 1; - if (!newcount) - break; /* do it the slow way */ - - newcount = cmpxchg(&atomic->counter, counter, newcount); - if (newcount == counter) - return 0; - } - - spin_lock(lock); - if (atomic_dec_and_test(atomic)) - return 1; - spin_unlock(lock); - return 0; -} - -EXPORT_SYMBOL(_atomic_dec_and_lock); diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig index 347ea284140b..776941c75672 100644 --- a/arch/ppc/Kconfig +++ b/arch/ppc/Kconfig @@ -26,10 +26,6 @@ config GENERIC_CALIBRATE_DELAY bool default y -config HAVE_DEC_LOCK - bool - default y - config PPC bool default y diff --git a/arch/ppc/lib/Makefile b/arch/ppc/lib/Makefile index f1e1fb4144f0..50358e4ea159 100644 --- a/arch/ppc/lib/Makefile +++ b/arch/ppc/lib/Makefile @@ -2,7 +2,7 @@ # Makefile for ppc-specific library files.. # -obj-y := checksum.o string.o strcase.o dec_and_lock.o div64.o +obj-y := checksum.o string.o strcase.o div64.o obj-$(CONFIG_8xx) += rheap.o obj-$(CONFIG_CPM2) += rheap.o diff --git a/arch/ppc/lib/dec_and_lock.c b/arch/ppc/lib/dec_and_lock.c deleted file mode 100644 index b18f0d9a00fc..000000000000 --- a/arch/ppc/lib/dec_and_lock.c +++ /dev/null @@ -1,38 +0,0 @@ -#include -#include -#include -#include - -/* - * This is an implementation of the notion of "decrement a - * reference count, and return locked if it decremented to zero". - * - * This implementation can be used on any architecture that - * has a cmpxchg, and where atomic->value is an int holding - * the value of the atomic (i.e. the high bits aren't used - * for a lock or anything like that). - */ -int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock) -{ - int counter; - int newcount; - - for (;;) { - counter = atomic_read(atomic); - newcount = counter - 1; - if (!newcount) - break; /* do it the slow way */ - - newcount = cmpxchg(&atomic->counter, counter, newcount); - if (newcount == counter) - return 0; - } - - spin_lock(lock); - if (atomic_dec_and_test(atomic)) - return 1; - spin_unlock(lock); - return 0; -} - -EXPORT_SYMBOL(_atomic_dec_and_lock); diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig index deca68ad644a..c658650af429 100644 --- a/arch/ppc64/Kconfig +++ b/arch/ppc64/Kconfig @@ -28,10 +28,6 @@ config GENERIC_ISA_DMA bool default y -config HAVE_DEC_LOCK - bool - default y - config EARLY_PRINTK bool default y diff --git a/arch/ppc64/lib/Makefile b/arch/ppc64/lib/Makefile index 76fbfa9f706f..0b6e967de948 100644 --- a/arch/ppc64/lib/Makefile +++ b/arch/ppc64/lib/Makefile @@ -2,7 +2,7 @@ # Makefile for ppc64-specific library files.. # -lib-y := checksum.o dec_and_lock.o string.o strcase.o +lib-y := checksum.o string.o strcase.o lib-y += copypage.o memcpy.o copyuser.o usercopy.o # Lock primitives are defined as no-ops in include/linux/spinlock.h diff --git a/arch/ppc64/lib/dec_and_lock.c b/arch/ppc64/lib/dec_and_lock.c deleted file mode 100644 index 7b9d4da5cf92..000000000000 --- a/arch/ppc64/lib/dec_and_lock.c +++ /dev/null @@ -1,47 +0,0 @@ -/* - * ppc64 version of atomic_dec_and_lock() using cmpxchg - * - * 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. - */ - -#include -#include -#include -#include - -/* - * This is an implementation of the notion of "decrement a - * reference count, and return locked if it decremented to zero". - * - * This implementation can be used on any architecture that - * has a cmpxchg, and where atomic->value is an int holding - * the value of the atomic (i.e. the high bits aren't used - * for a lock or anything like that). - */ -int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock) -{ - int counter; - int newcount; - - for (;;) { - counter = atomic_read(atomic); - newcount = counter - 1; - if (!newcount) - break; /* do it the slow way */ - - newcount = cmpxchg(&atomic->counter, counter, newcount); - if (newcount == counter) - return 0; - } - - spin_lock(lock); - if (atomic_dec_and_test(atomic)) - return 1; - spin_unlock(lock); - return 0; -} - -EXPORT_SYMBOL(_atomic_dec_and_lock); diff --git a/arch/sparc64/Kconfig.debug b/arch/sparc64/Kconfig.debug index cd8d39fb954d..af0e9411b83e 100644 --- a/arch/sparc64/Kconfig.debug +++ b/arch/sparc64/Kconfig.debug @@ -33,14 +33,6 @@ config DEBUG_BOOTMEM depends on DEBUG_KERNEL bool "Debug BOOTMEM initialization" -# We have a custom atomic_dec_and_lock() implementation but it's not -# compatible with spinlock debugging so we need to fall back on -# the generic version in that case. -config HAVE_DEC_LOCK - bool - depends on SMP && !DEBUG_SPINLOCK - default y - config MCOUNT bool depends on STACK_DEBUG diff --git a/arch/sparc64/kernel/sparc64_ksyms.c b/arch/sparc64/kernel/sparc64_ksyms.c index cbb5e59824e5..fb7a5370dbfc 100644 --- a/arch/sparc64/kernel/sparc64_ksyms.c +++ b/arch/sparc64/kernel/sparc64_ksyms.c @@ -163,9 +163,6 @@ EXPORT_SYMBOL(atomic64_add); EXPORT_SYMBOL(atomic64_add_ret); EXPORT_SYMBOL(atomic64_sub); EXPORT_SYMBOL(atomic64_sub_ret); -#ifdef CONFIG_SMP -EXPORT_SYMBOL(_atomic_dec_and_lock); -#endif /* Atomic bit operations. */ EXPORT_SYMBOL(test_and_set_bit); diff --git a/arch/sparc64/lib/Makefile b/arch/sparc64/lib/Makefile index d968aebe83b2..c295806500f7 100644 --- a/arch/sparc64/lib/Makefile +++ b/arch/sparc64/lib/Makefile @@ -14,6 +14,4 @@ lib-y := PeeCeeI.o copy_page.o clear_page.o strlen.o strncmp.o \ copy_in_user.o user_fixup.o memmove.o \ mcount.o ipcsum.o rwsem.o xor.o find_bit.o delay.o -lib-$(CONFIG_HAVE_DEC_LOCK) += dec_and_lock.o - obj-y += iomap.o diff --git a/arch/sparc64/lib/dec_and_lock.S b/arch/sparc64/lib/dec_and_lock.S deleted file mode 100644 index 8ee288dd0afc..000000000000 --- a/arch/sparc64/lib/dec_and_lock.S +++ /dev/null @@ -1,80 +0,0 @@ -/* $Id: dec_and_lock.S,v 1.5 2001/11/18 00:12:56 davem Exp $ - * dec_and_lock.S: Sparc64 version of "atomic_dec_and_lock()" - * using cas and ldstub instructions. - * - * Copyright (C) 2000 David S. Miller (davem@redhat.com) - */ -#include -#include - - .text - .align 64 - - /* CAS basically works like this: - * - * void CAS(MEM, REG1, REG2) - * { - * START_ATOMIC(); - * if (*(MEM) == REG1) { - * TMP = *(MEM); - * *(MEM) = REG2; - * REG2 = TMP; - * } else - * REG2 = *(MEM); - * END_ATOMIC(); - * } - */ - - .globl _atomic_dec_and_lock -_atomic_dec_and_lock: /* %o0 = counter, %o1 = lock */ -loop1: lduw [%o0], %g2 - subcc %g2, 1, %g7 - be,pn %icc, start_to_zero - nop -nzero: cas [%o0], %g2, %g7 - cmp %g2, %g7 - bne,pn %icc, loop1 - mov 0, %g1 - -out: - membar #StoreLoad | #StoreStore - retl - mov %g1, %o0 -start_to_zero: -#ifdef CONFIG_PREEMPT - ldsw [%g6 + TI_PRE_COUNT], %g3 - add %g3, 1, %g3 - stw %g3, [%g6 + TI_PRE_COUNT] -#endif -to_zero: - ldstub [%o1], %g3 - membar #StoreLoad | #StoreStore - brnz,pn %g3, spin_on_lock - nop -loop2: cas [%o0], %g2, %g7 /* ASSERT(g7 == 0) */ - cmp %g2, %g7 - - be,pt %icc, out - mov 1, %g1 - lduw [%o0], %g2 - subcc %g2, 1, %g7 - be,pn %icc, loop2 - nop - membar #StoreStore | #LoadStore - stb %g0, [%o1] -#ifdef CONFIG_PREEMPT - ldsw [%g6 + TI_PRE_COUNT], %g3 - sub %g3, 1, %g3 - stw %g3, [%g6 + TI_PRE_COUNT] -#endif - - b,pt %xcc, nzero - nop -spin_on_lock: - ldub [%o1], %g3 - membar #LoadLoad - brnz,pt %g3, spin_on_lock - nop - ba,pt %xcc, to_zero - nop - nop diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index e63323e03ea9..0969d570f3b5 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -277,11 +277,6 @@ source "mm/Kconfig" config HAVE_ARCH_EARLY_PFN_TO_NID def_bool y -config HAVE_DEC_LOCK - bool - depends on SMP - default y - config NR_CPUS int "Maximum number of CPUs (2-256)" range 2 256 diff --git a/arch/x86_64/kernel/x8664_ksyms.c b/arch/x86_64/kernel/x8664_ksyms.c index 68ec03070e5a..fd99ddd009bc 100644 --- a/arch/x86_64/kernel/x8664_ksyms.c +++ b/arch/x86_64/kernel/x8664_ksyms.c @@ -178,10 +178,6 @@ EXPORT_SYMBOL(rwsem_down_write_failed_thunk); EXPORT_SYMBOL(empty_zero_page); -#ifdef CONFIG_HAVE_DEC_LOCK -EXPORT_SYMBOL(_atomic_dec_and_lock); -#endif - EXPORT_SYMBOL(die_chain); EXPORT_SYMBOL(register_die_notifier); diff --git a/arch/x86_64/lib/Makefile b/arch/x86_64/lib/Makefile index 6b26a1c1e9ff..bba5db6cebd6 100644 --- a/arch/x86_64/lib/Makefile +++ b/arch/x86_64/lib/Makefile @@ -10,5 +10,3 @@ lib-y := csum-partial.o csum-copy.o csum-wrappers.o delay.o \ usercopy.o getuser.o putuser.o \ thunk.o clear_page.o copy_page.o bitstr.o bitops.o lib-y += memcpy.o memmove.o memset.o copy_user.o - -lib-$(CONFIG_HAVE_DEC_LOCK) += dec_and_lock.o diff --git a/arch/x86_64/lib/dec_and_lock.c b/arch/x86_64/lib/dec_and_lock.c deleted file mode 100644 index ab43394dc775..000000000000 --- a/arch/x86_64/lib/dec_and_lock.c +++ /dev/null @@ -1,40 +0,0 @@ -/* - * x86 version of "atomic_dec_and_lock()" using - * the atomic "cmpxchg" instruction. - * - * (For CPU's lacking cmpxchg, we use the slow - * generic version, and this one never even gets - * compiled). - */ - -#include -#include - -int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock) -{ - int counter; - int newcount; - -repeat: - counter = atomic_read(atomic); - newcount = counter-1; - - if (!newcount) - goto slow_path; - - asm volatile("lock; cmpxchgl %1,%2" - :"=a" (newcount) - :"r" (newcount), "m" (atomic->counter), "0" (counter)); - - /* If the above failed, "eax" will have changed */ - if (newcount != counter) - goto repeat; - return 0; - -slow_path: - spin_lock(lock); - if (atomic_dec_and_test(atomic)) - return 1; - spin_unlock(lock); - return 0; -} diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig index 2b6257bec4c3..7e841aa2a4aa 100644 --- a/arch/xtensa/Kconfig +++ b/arch/xtensa/Kconfig @@ -26,10 +26,6 @@ config RWSEM_XCHGADD_ALGORITHM bool default y -config HAVE_DEC_LOCK - bool - default y - config GENERIC_HARDIRQS bool default y diff --git a/lib/dec_and_lock.c b/lib/dec_and_lock.c index 2377af057d09..305a9663aee3 100644 --- a/lib/dec_and_lock.c +++ b/lib/dec_and_lock.c @@ -1,7 +1,41 @@ #include #include #include +#include +#ifdef __HAVE_ARCH_CMPXCHG +/* + * This is an implementation of the notion of "decrement a + * reference count, and return locked if it decremented to zero". + * + * This implementation can be used on any architecture that + * has a cmpxchg, and where atomic->value is an int holding + * the value of the atomic (i.e. the high bits aren't used + * for a lock or anything like that). + */ +int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock) +{ + int counter; + int newcount; + + for (;;) { + counter = atomic_read(atomic); + newcount = counter - 1; + if (!newcount) + break; /* do it the slow way */ + + newcount = cmpxchg(&atomic->counter, counter, newcount); + if (newcount == counter) + return 0; + } + + spin_lock(lock); + if (atomic_dec_and_test(atomic)) + return 1; + spin_unlock(lock); + return 0; +} +#else /* * This is an architecture-neutral, but slow, * implementation of the notion of "decrement @@ -33,5 +67,6 @@ int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock) spin_unlock(lock); return 0; } +#endif EXPORT_SYMBOL(_atomic_dec_and_lock); From fea2efe3bba15f0aa8f840fbe052699808187cb6 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 15 Sep 2005 12:30:11 +0100 Subject: [PATCH 048/240] [ARM] Remove PFN_TO_NID for !DISCONTIGMEM Platform classes need not define PFN_TO_NID when DISCONTIGMEM is not selected. Signed-off-by: Russell King --- include/asm-arm/arch-aaec2000/memory.h | 4 ---- include/asm-arm/arch-iop3xx/memory.h | 2 -- include/asm-arm/arch-lh7a40x/memory.h | 4 ---- include/asm-arm/arch-omap/memory.h | 1 - include/asm-arm/arch-pxa/memory.h | 4 ---- include/asm-arm/arch-sa1100/memory.h | 4 ---- 6 files changed, 19 deletions(-) diff --git a/include/asm-arm/arch-aaec2000/memory.h b/include/asm-arm/arch-aaec2000/memory.h index 681b6a6171a1..79c90813bc3e 100644 --- a/include/asm-arm/arch-aaec2000/memory.h +++ b/include/asm-arm/arch-aaec2000/memory.h @@ -64,10 +64,6 @@ #define NODE_MAX_MEM_SHIFT 26 #define NODE_MAX_MEM_SIZE (1 << NODE_MAX_MEM_SHIFT) -#else - -#define PFN_TO_NID(addr) (0) - #endif /* CONFIG_DISCONTIGMEM */ #endif /* __ASM_ARCH_MEMORY_H */ diff --git a/include/asm-arm/arch-iop3xx/memory.h b/include/asm-arm/arch-iop3xx/memory.h index dc4735cb0c10..45351f5cd904 100644 --- a/include/asm-arm/arch-iop3xx/memory.h +++ b/include/asm-arm/arch-iop3xx/memory.h @@ -36,6 +36,4 @@ #endif -#define PFN_TO_NID(addr) (0) - #endif diff --git a/include/asm-arm/arch-lh7a40x/memory.h b/include/asm-arm/arch-lh7a40x/memory.h index 7e2fea372663..c650e6feb9d5 100644 --- a/include/asm-arm/arch-lh7a40x/memory.h +++ b/include/asm-arm/arch-lh7a40x/memory.h @@ -85,10 +85,6 @@ (((unsigned long)(addr) & 0x01ffffff) >> PAGE_SHIFT) # endif -#else - -# define PFN_TO_NID(addr) (0) - #endif #endif diff --git a/include/asm-arm/arch-omap/memory.h b/include/asm-arm/arch-omap/memory.h index 84f81e315a25..ef32d61eec7a 100644 --- a/include/asm-arm/arch-omap/memory.h +++ b/include/asm-arm/arch-omap/memory.h @@ -86,6 +86,5 @@ #endif /* CONFIG_ARCH_OMAP1510 */ -#define PHYS_TO_NID(addr) (0) #endif diff --git a/include/asm-arm/arch-pxa/memory.h b/include/asm-arm/arch-pxa/memory.h index 217a80b820ff..58bad9748b5c 100644 --- a/include/asm-arm/arch-pxa/memory.h +++ b/include/asm-arm/arch-pxa/memory.h @@ -67,10 +67,6 @@ #define LOCAL_MAP_NR(addr) \ (((unsigned long)(addr) & 0x03ffffff) >> PAGE_SHIFT) -#else - -#define PFN_TO_NID(addr) (0) - #endif #endif diff --git a/include/asm-arm/arch-sa1100/memory.h b/include/asm-arm/arch-sa1100/memory.h index 32d3d5bde34d..8743ff5c1b23 100644 --- a/include/asm-arm/arch-sa1100/memory.h +++ b/include/asm-arm/arch-sa1100/memory.h @@ -99,10 +99,6 @@ __arch_adjust_zones(int node, unsigned long *size, unsigned long *holes) #define LOCAL_MAP_NR(addr) \ (((unsigned long)(addr) & 0x07ffffff) >> PAGE_SHIFT) -#else - -#define PFN_TO_NID(addr) (0) - #endif #endif From 917afce1000d978dfd3f07da5da9d864bc49c77e Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Thu, 15 Sep 2005 13:00:25 +0100 Subject: [PATCH 049/240] [ARM] 2911/1: ixp2000_reg_{read,write} accessors Patch from Lennert Buytenhek This patch: - changes the ixp2000_reg_write accessor to take a 'volatile void *' instead of a 'volatile unsigned long *', which then allows passing in a u32 * as first argument without being greeted with a warning; and - adds an ixp2000_reg_read accessor. We can then use these accessors in ixp2000 code to access on-chip peripherals, instead of directly dereferencing pointers. This is for use by the ixp2000 microengine driver which was recently announced on netdev. We can't use readl/writel on the ixp2000 since it is usually run in big-endian mode, and on big-endian platforms, readl/writel perform byteswapping. A future patch will remove the readback from ixp2000_reg_write, since it's not needed to prevent erratum #66, and add manual readbacks to the places that need them (writes are not synchronous since we map in device space using XCB=101 nowadays), such as interrupt disabling and GPIO manipulation. See also: http://lists.arm.linux.org.uk/pipermail/linux-arm-kernel/2005-February/027084.html Patch has been ACKed by Jeff Garzik. Signed-off-by: Lennert Buytenhek Signed-off-by: Russell King --- include/asm-arm/arch-ixp2000/platform.h | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/include/asm-arm/arch-ixp2000/platform.h b/include/asm-arm/arch-ixp2000/platform.h index c0caf3e3e6fd..6519498dbe25 100644 --- a/include/asm-arm/arch-ixp2000/platform.h +++ b/include/asm-arm/arch-ixp2000/platform.h @@ -31,20 +31,24 @@ #include /* Pickup local_irq_ functions */ -static inline void ixp2000_reg_write(volatile unsigned long *reg, unsigned long val) +static inline void ixp2000_reg_write(volatile void *reg, unsigned long val) { - volatile unsigned long dummy; + unsigned long dummy; unsigned long flags; local_irq_save(flags); - *reg = val; + *((volatile unsigned long *)reg) = val; barrier(); - dummy = *reg; + dummy = *((volatile unsigned long *)reg); local_irq_restore(flags); } #else -#define ixp2000_reg_write(reg, val) (*reg = val) +static inline void ixp2000_reg_write(volatile void *reg, unsigned long val) +{ + *((volatile unsigned long *)reg) = val; +} #endif /* IXDP2400 || IXDP2401 */ +#define ixp2000_reg_read(reg) (*((volatile unsigned long *)reg)) /* * Boards may multiplex different devices on the 2nd channel of From 44449bbf4b051ef7bbea648602f4e21658fe2354 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Thu, 15 Sep 2005 13:00:52 +0100 Subject: [PATCH 050/240] [ARM] 2909/1: remove IXP2000_PROD_ID Patch from Lennert Buytenhek The intel docs call it IXP2000_PRODUCT_ID, and we have a definition for IXP2000_PRODUCT_ID as well, so IXP2000_PROD_ID can go. It's only used in one place. Signed-off-by: Lennert Buytenhek Signed-off-by: Deepak Saxena Signed-off-by: Russell King --- include/asm-arm/arch-ixp2000/ixp2000-regs.h | 2 -- include/asm-arm/arch-ixp2000/platform.h | 2 +- 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/include/asm-arm/arch-ixp2000/ixp2000-regs.h b/include/asm-arm/arch-ixp2000/ixp2000-regs.h index 75623f81ef75..32aece069869 100644 --- a/include/asm-arm/arch-ixp2000/ixp2000-regs.h +++ b/include/asm-arm/arch-ixp2000/ixp2000-regs.h @@ -370,8 +370,6 @@ #define GLOBAL_REG_BASE (IXP2000_GLOBAL_REG_VIRT_BASE + 0x0a00) #define GLOBAL_REG(x) (volatile unsigned long*)(GLOBAL_REG_BASE | (x)) -#define IXP2000_PROD_ID GLOBAL_REG(0x00) - #define IXP2000_MAJ_PROD_TYPE_MASK 0x001F0000 #define IXP2000_MAJ_PROD_TYPE_IXP2000 0x00000000 #define IXP2000_MIN_PROD_TYPE_MASK 0x0000FF00 diff --git a/include/asm-arm/arch-ixp2000/platform.h b/include/asm-arm/arch-ixp2000/platform.h index 6519498dbe25..abdcf51bd283 100644 --- a/include/asm-arm/arch-ixp2000/platform.h +++ b/include/asm-arm/arch-ixp2000/platform.h @@ -88,7 +88,7 @@ void ixp2000_release_slowport(struct slowport_cfg *); */ static inline unsigned ixp2000_has_broken_slowport(void) { - unsigned long id = *IXP2000_PROD_ID; + unsigned long id = *IXP2000_PRODUCT_ID; unsigned long id_prod = id & (IXP2000_MAJ_PROD_TYPE_MASK | IXP2000_MIN_PROD_TYPE_MASK); return (((id_prod == From 2e4e8bbb6e2da073c5654a39a58eb0895e3ba4c7 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Thu, 15 Sep 2005 13:02:31 +0100 Subject: [PATCH 051/240] [ARM] 2904/1: update ixp2000 defconfigs to 2.6.13 Patch from Lennert Buytenhek Signed-off-by: Lennert Buytenhek Signed-off-by: Deepak Saxena Signed-off-by: Russell King --- arch/arm/configs/enp2611_defconfig | 180 ++++++++++++++-------------- arch/arm/configs/ixdp2400_defconfig | 180 ++++++++++++++-------------- arch/arm/configs/ixdp2401_defconfig | 180 ++++++++++++++-------------- arch/arm/configs/ixdp2800_defconfig | 180 ++++++++++++++-------------- arch/arm/configs/ixdp2801_defconfig | 180 ++++++++++++++-------------- 5 files changed, 455 insertions(+), 445 deletions(-) diff --git a/arch/arm/configs/enp2611_defconfig b/arch/arm/configs/enp2611_defconfig index f67ca01b4982..a8c001f77b1d 100644 --- a/arch/arm/configs/enp2611_defconfig +++ b/arch/arm/configs/enp2611_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.13-rc2 -# Thu Jul 7 16:41:21 2005 +# Linux kernel version: 2.6.13 +# Wed Sep 14 10:51:52 2005 # CONFIG_ARM=y CONFIG_MMU=y @@ -135,7 +135,6 @@ CONFIG_PCI_NAMES=y # # Kernel Features # -# CONFIG_SMP is not set # CONFIG_PREEMPT is not set # CONFIG_NO_IDLE_HZ is not set # CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set @@ -178,6 +177,68 @@ CONFIG_BINFMT_ELF=y # # CONFIG_PM is not set +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +CONFIG_PACKET=y +CONFIG_PACKET_MMAP=y +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +CONFIG_IP_PNP=y +CONFIG_IP_PNP_DHCP=y +CONFIG_IP_PNP_BOOTP=y +# CONFIG_IP_PNP_RARP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +CONFIG_SYN_COOKIES=y +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_TUNNEL is not set +# CONFIG_IP_TCPDIAG is not set +# CONFIG_IP_TCPDIAG_IPV6 is not set +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_BIC=y +# CONFIG_IPV6 is not set +# CONFIG_NETFILTER is not set + +# +# SCTP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_SCTP is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_NET_DIVERT is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +# CONFIG_NET_SCHED is not set +# CONFIG_NET_CLS_ROUTE is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set + # # Device Drivers # @@ -248,6 +309,7 @@ CONFIG_MTD_COMPLEX_MAPPINGS=y CONFIG_MTD_IXP2000=y # CONFIG_MTD_EDB7312 is not set # CONFIG_MTD_PCI is not set +# CONFIG_MTD_PLATRAM is not set # # Self-contained MTD device drivers @@ -334,72 +396,8 @@ CONFIG_IOSCHED_CFQ=y # CONFIG_I2O is not set # -# Networking support +# Network device support # -CONFIG_NET=y - -# -# Networking options -# -CONFIG_PACKET=y -CONFIG_PACKET_MMAP=y -CONFIG_UNIX=y -# CONFIG_NET_KEY is not set -CONFIG_INET=y -# CONFIG_IP_MULTICAST is not set -# CONFIG_IP_ADVANCED_ROUTER is not set -CONFIG_IP_FIB_HASH=y -CONFIG_IP_PNP=y -CONFIG_IP_PNP_DHCP=y -CONFIG_IP_PNP_BOOTP=y -# CONFIG_IP_PNP_RARP is not set -# CONFIG_NET_IPIP is not set -# CONFIG_NET_IPGRE is not set -# CONFIG_ARPD is not set -CONFIG_SYN_COOKIES=y -# CONFIG_INET_AH is not set -# CONFIG_INET_ESP is not set -# CONFIG_INET_IPCOMP is not set -# CONFIG_INET_TUNNEL is not set -# CONFIG_IP_TCPDIAG is not set -# CONFIG_IP_TCPDIAG_IPV6 is not set -# CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y -# CONFIG_IPV6 is not set -# CONFIG_NETFILTER is not set - -# -# SCTP Configuration (EXPERIMENTAL) -# -# CONFIG_IP_SCTP is not set -# CONFIG_ATM is not set -# CONFIG_BRIDGE is not set -# CONFIG_VLAN_8021Q is not set -# CONFIG_DECNET is not set -# CONFIG_LLC2 is not set -# CONFIG_IPX is not set -# CONFIG_ATALK is not set -# CONFIG_X25 is not set -# CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set -# CONFIG_ECONET is not set -# CONFIG_WAN_ROUTER is not set - -# -# QoS and/or fair queueing -# -# CONFIG_NET_SCHED is not set -# CONFIG_NET_CLS_ROUTE is not set - -# -# Network testing -# -# CONFIG_NET_PKTGEN is not set -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set -# CONFIG_HAMRADIO is not set -# CONFIG_IRDA is not set -# CONFIG_BT is not set CONFIG_NETDEVICES=y CONFIG_DUMMY=y # CONFIG_BONDING is not set @@ -509,6 +507,8 @@ CONFIG_DLCI_MAX=8 # CONFIG_SLIP is not set # CONFIG_SHAPER is not set # CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set # # ISDN subsystem @@ -649,11 +649,28 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_VIAPRO is not set # CONFIG_I2C_VOODOO3 is not set # CONFIG_I2C_PCA_ISA is not set +CONFIG_I2C_SENSOR=y # -# Hardware Sensors Chip support +# Miscellaneous I2C Chip support # -CONFIG_I2C_SENSOR=y +# CONFIG_SENSORS_DS1337 is not set +# CONFIG_SENSORS_DS1374 is not set +CONFIG_SENSORS_EEPROM=y +# CONFIG_SENSORS_PCF8574 is not set +# CONFIG_SENSORS_PCA9539 is not set +# CONFIG_SENSORS_PCF8591 is not set +# CONFIG_SENSORS_RTC8564 is not set +# CONFIG_SENSORS_MAX6875 is not set +# CONFIG_I2C_DEBUG_CORE is not set +# CONFIG_I2C_DEBUG_ALGO is not set +# CONFIG_I2C_DEBUG_BUS is not set +# CONFIG_I2C_DEBUG_CHIP is not set + +# +# Hardware Monitoring support +# +CONFIG_HWMON=y # CONFIG_SENSORS_ADM1021 is not set # CONFIG_SENSORS_ADM1025 is not set # CONFIG_SENSORS_ADM1026 is not set @@ -679,30 +696,15 @@ CONFIG_I2C_SENSOR=y # CONFIG_SENSORS_LM92 is not set # CONFIG_SENSORS_MAX1619 is not set # CONFIG_SENSORS_PC87360 is not set -# CONFIG_SENSORS_SMSC47B397 is not set # CONFIG_SENSORS_SIS5595 is not set # CONFIG_SENSORS_SMSC47M1 is not set +# CONFIG_SENSORS_SMSC47B397 is not set # CONFIG_SENSORS_VIA686A is not set # CONFIG_SENSORS_W83781D is not set # CONFIG_SENSORS_W83L785TS is not set # CONFIG_SENSORS_W83627HF is not set # CONFIG_SENSORS_W83627EHF is not set - -# -# Other I2C Chip support -# -# CONFIG_SENSORS_DS1337 is not set -# CONFIG_SENSORS_DS1374 is not set -CONFIG_SENSORS_EEPROM=y -# CONFIG_SENSORS_PCF8574 is not set -# CONFIG_SENSORS_PCA9539 is not set -# CONFIG_SENSORS_PCF8591 is not set -# CONFIG_SENSORS_RTC8564 is not set -# CONFIG_SENSORS_MAX6875 is not set -# CONFIG_I2C_DEBUG_CORE is not set -# CONFIG_I2C_DEBUG_ALGO is not set -# CONFIG_I2C_DEBUG_BUS is not set -# CONFIG_I2C_DEBUG_CHIP is not set +# CONFIG_HWMON_DEBUG_CHIP is not set # # Misc devices @@ -770,6 +772,7 @@ CONFIG_FS_POSIX_ACL=y # CONFIG_XFS_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set +CONFIG_INOTIFY=y # CONFIG_QUOTA is not set CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set @@ -812,8 +815,7 @@ CONFIG_RAMFS=y # CONFIG_JFFS_FS is not set CONFIG_JFFS2_FS=y CONFIG_JFFS2_FS_DEBUG=0 -# CONFIG_JFFS2_FS_NAND is not set -# CONFIG_JFFS2_FS_NOR_ECC is not set +CONFIG_JFFS2_FS_WRITEBUFFER=y # CONFIG_JFFS2_COMPRESSION_OPTIONS is not set CONFIG_JFFS2_ZLIB=y CONFIG_JFFS2_RTIME=y diff --git a/arch/arm/configs/ixdp2400_defconfig b/arch/arm/configs/ixdp2400_defconfig index 5c6c928215d0..5ea356efcb0e 100644 --- a/arch/arm/configs/ixdp2400_defconfig +++ b/arch/arm/configs/ixdp2400_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.13-rc2 -# Thu Jul 7 16:49:01 2005 +# Linux kernel version: 2.6.13 +# Wed Sep 14 10:52:01 2005 # CONFIG_ARM=y CONFIG_MMU=y @@ -136,7 +136,6 @@ CONFIG_PCI_NAMES=y # # Kernel Features # -# CONFIG_SMP is not set # CONFIG_PREEMPT is not set # CONFIG_NO_IDLE_HZ is not set # CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set @@ -179,6 +178,68 @@ CONFIG_BINFMT_ELF=y # # CONFIG_PM is not set +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +CONFIG_PACKET=y +CONFIG_PACKET_MMAP=y +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +CONFIG_IP_PNP=y +CONFIG_IP_PNP_DHCP=y +CONFIG_IP_PNP_BOOTP=y +# CONFIG_IP_PNP_RARP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +CONFIG_SYN_COOKIES=y +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_TUNNEL is not set +# CONFIG_IP_TCPDIAG is not set +# CONFIG_IP_TCPDIAG_IPV6 is not set +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_BIC=y +# CONFIG_IPV6 is not set +# CONFIG_NETFILTER is not set + +# +# SCTP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_SCTP is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_NET_DIVERT is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +# CONFIG_NET_SCHED is not set +# CONFIG_NET_CLS_ROUTE is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set + # # Device Drivers # @@ -249,6 +310,7 @@ CONFIG_MTD_COMPLEX_MAPPINGS=y CONFIG_MTD_IXP2000=y # CONFIG_MTD_EDB7312 is not set # CONFIG_MTD_PCI is not set +# CONFIG_MTD_PLATRAM is not set # # Self-contained MTD device drivers @@ -335,72 +397,8 @@ CONFIG_IOSCHED_CFQ=y # CONFIG_I2O is not set # -# Networking support +# Network device support # -CONFIG_NET=y - -# -# Networking options -# -CONFIG_PACKET=y -CONFIG_PACKET_MMAP=y -CONFIG_UNIX=y -# CONFIG_NET_KEY is not set -CONFIG_INET=y -# CONFIG_IP_MULTICAST is not set -# CONFIG_IP_ADVANCED_ROUTER is not set -CONFIG_IP_FIB_HASH=y -CONFIG_IP_PNP=y -CONFIG_IP_PNP_DHCP=y -CONFIG_IP_PNP_BOOTP=y -# CONFIG_IP_PNP_RARP is not set -# CONFIG_NET_IPIP is not set -# CONFIG_NET_IPGRE is not set -# CONFIG_ARPD is not set -CONFIG_SYN_COOKIES=y -# CONFIG_INET_AH is not set -# CONFIG_INET_ESP is not set -# CONFIG_INET_IPCOMP is not set -# CONFIG_INET_TUNNEL is not set -# CONFIG_IP_TCPDIAG is not set -# CONFIG_IP_TCPDIAG_IPV6 is not set -# CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y -# CONFIG_IPV6 is not set -# CONFIG_NETFILTER is not set - -# -# SCTP Configuration (EXPERIMENTAL) -# -# CONFIG_IP_SCTP is not set -# CONFIG_ATM is not set -# CONFIG_BRIDGE is not set -# CONFIG_VLAN_8021Q is not set -# CONFIG_DECNET is not set -# CONFIG_LLC2 is not set -# CONFIG_IPX is not set -# CONFIG_ATALK is not set -# CONFIG_X25 is not set -# CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set -# CONFIG_ECONET is not set -# CONFIG_WAN_ROUTER is not set - -# -# QoS and/or fair queueing -# -# CONFIG_NET_SCHED is not set -# CONFIG_NET_CLS_ROUTE is not set - -# -# Network testing -# -# CONFIG_NET_PKTGEN is not set -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set -# CONFIG_HAMRADIO is not set -# CONFIG_IRDA is not set -# CONFIG_BT is not set CONFIG_NETDEVICES=y CONFIG_DUMMY=y # CONFIG_BONDING is not set @@ -510,6 +508,8 @@ CONFIG_DLCI_MAX=8 # CONFIG_SLIP is not set # CONFIG_SHAPER is not set # CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set # # ISDN subsystem @@ -650,11 +650,28 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_VIAPRO is not set # CONFIG_I2C_VOODOO3 is not set # CONFIG_I2C_PCA_ISA is not set +CONFIG_I2C_SENSOR=y # -# Hardware Sensors Chip support +# Miscellaneous I2C Chip support # -CONFIG_I2C_SENSOR=y +# CONFIG_SENSORS_DS1337 is not set +# CONFIG_SENSORS_DS1374 is not set +CONFIG_SENSORS_EEPROM=y +# CONFIG_SENSORS_PCF8574 is not set +# CONFIG_SENSORS_PCA9539 is not set +# CONFIG_SENSORS_PCF8591 is not set +# CONFIG_SENSORS_RTC8564 is not set +# CONFIG_SENSORS_MAX6875 is not set +# CONFIG_I2C_DEBUG_CORE is not set +# CONFIG_I2C_DEBUG_ALGO is not set +# CONFIG_I2C_DEBUG_BUS is not set +# CONFIG_I2C_DEBUG_CHIP is not set + +# +# Hardware Monitoring support +# +CONFIG_HWMON=y # CONFIG_SENSORS_ADM1021 is not set # CONFIG_SENSORS_ADM1025 is not set # CONFIG_SENSORS_ADM1026 is not set @@ -680,30 +697,15 @@ CONFIG_I2C_SENSOR=y # CONFIG_SENSORS_LM92 is not set # CONFIG_SENSORS_MAX1619 is not set # CONFIG_SENSORS_PC87360 is not set -# CONFIG_SENSORS_SMSC47B397 is not set # CONFIG_SENSORS_SIS5595 is not set # CONFIG_SENSORS_SMSC47M1 is not set +# CONFIG_SENSORS_SMSC47B397 is not set # CONFIG_SENSORS_VIA686A is not set # CONFIG_SENSORS_W83781D is not set # CONFIG_SENSORS_W83L785TS is not set # CONFIG_SENSORS_W83627HF is not set # CONFIG_SENSORS_W83627EHF is not set - -# -# Other I2C Chip support -# -# CONFIG_SENSORS_DS1337 is not set -# CONFIG_SENSORS_DS1374 is not set -CONFIG_SENSORS_EEPROM=y -# CONFIG_SENSORS_PCF8574 is not set -# CONFIG_SENSORS_PCA9539 is not set -# CONFIG_SENSORS_PCF8591 is not set -# CONFIG_SENSORS_RTC8564 is not set -# CONFIG_SENSORS_MAX6875 is not set -# CONFIG_I2C_DEBUG_CORE is not set -# CONFIG_I2C_DEBUG_ALGO is not set -# CONFIG_I2C_DEBUG_BUS is not set -# CONFIG_I2C_DEBUG_CHIP is not set +# CONFIG_HWMON_DEBUG_CHIP is not set # # Misc devices @@ -771,6 +773,7 @@ CONFIG_FS_POSIX_ACL=y # CONFIG_XFS_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set +CONFIG_INOTIFY=y # CONFIG_QUOTA is not set CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set @@ -813,8 +816,7 @@ CONFIG_RAMFS=y # CONFIG_JFFS_FS is not set CONFIG_JFFS2_FS=y CONFIG_JFFS2_FS_DEBUG=0 -# CONFIG_JFFS2_FS_NAND is not set -# CONFIG_JFFS2_FS_NOR_ECC is not set +CONFIG_JFFS2_FS_WRITEBUFFER=y # CONFIG_JFFS2_COMPRESSION_OPTIONS is not set CONFIG_JFFS2_ZLIB=y CONFIG_JFFS2_RTIME=y diff --git a/arch/arm/configs/ixdp2401_defconfig b/arch/arm/configs/ixdp2401_defconfig index 6dc40f6be0ef..c542d7fc0c1b 100644 --- a/arch/arm/configs/ixdp2401_defconfig +++ b/arch/arm/configs/ixdp2401_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.13-rc2 -# Thu Jul 7 16:49:08 2005 +# Linux kernel version: 2.6.13 +# Wed Sep 14 10:52:10 2005 # CONFIG_ARM=y CONFIG_MMU=y @@ -136,7 +136,6 @@ CONFIG_PCI_NAMES=y # # Kernel Features # -# CONFIG_SMP is not set # CONFIG_PREEMPT is not set # CONFIG_NO_IDLE_HZ is not set # CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set @@ -179,6 +178,68 @@ CONFIG_BINFMT_ELF=y # # CONFIG_PM is not set +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +CONFIG_PACKET=y +CONFIG_PACKET_MMAP=y +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +CONFIG_IP_PNP=y +CONFIG_IP_PNP_DHCP=y +CONFIG_IP_PNP_BOOTP=y +# CONFIG_IP_PNP_RARP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +CONFIG_SYN_COOKIES=y +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_TUNNEL is not set +CONFIG_IP_TCPDIAG=y +# CONFIG_IP_TCPDIAG_IPV6 is not set +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_BIC=y +# CONFIG_IPV6 is not set +# CONFIG_NETFILTER is not set + +# +# SCTP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_SCTP is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_NET_DIVERT is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +# CONFIG_NET_SCHED is not set +# CONFIG_NET_CLS_ROUTE is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set + # # Device Drivers # @@ -249,6 +310,7 @@ CONFIG_MTD_COMPLEX_MAPPINGS=y CONFIG_MTD_IXP2000=y # CONFIG_MTD_EDB7312 is not set # CONFIG_MTD_PCI is not set +# CONFIG_MTD_PLATRAM is not set # # Self-contained MTD device drivers @@ -335,72 +397,8 @@ CONFIG_IOSCHED_CFQ=y # CONFIG_I2O is not set # -# Networking support +# Network device support # -CONFIG_NET=y - -# -# Networking options -# -CONFIG_PACKET=y -CONFIG_PACKET_MMAP=y -CONFIG_UNIX=y -# CONFIG_NET_KEY is not set -CONFIG_INET=y -# CONFIG_IP_MULTICAST is not set -# CONFIG_IP_ADVANCED_ROUTER is not set -CONFIG_IP_FIB_HASH=y -CONFIG_IP_PNP=y -CONFIG_IP_PNP_DHCP=y -CONFIG_IP_PNP_BOOTP=y -# CONFIG_IP_PNP_RARP is not set -# CONFIG_NET_IPIP is not set -# CONFIG_NET_IPGRE is not set -# CONFIG_ARPD is not set -CONFIG_SYN_COOKIES=y -# CONFIG_INET_AH is not set -# CONFIG_INET_ESP is not set -# CONFIG_INET_IPCOMP is not set -# CONFIG_INET_TUNNEL is not set -CONFIG_IP_TCPDIAG=y -# CONFIG_IP_TCPDIAG_IPV6 is not set -# CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y -# CONFIG_IPV6 is not set -# CONFIG_NETFILTER is not set - -# -# SCTP Configuration (EXPERIMENTAL) -# -# CONFIG_IP_SCTP is not set -# CONFIG_ATM is not set -# CONFIG_BRIDGE is not set -# CONFIG_VLAN_8021Q is not set -# CONFIG_DECNET is not set -# CONFIG_LLC2 is not set -# CONFIG_IPX is not set -# CONFIG_ATALK is not set -# CONFIG_X25 is not set -# CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set -# CONFIG_ECONET is not set -# CONFIG_WAN_ROUTER is not set - -# -# QoS and/or fair queueing -# -# CONFIG_NET_SCHED is not set -# CONFIG_NET_CLS_ROUTE is not set - -# -# Network testing -# -# CONFIG_NET_PKTGEN is not set -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set -# CONFIG_HAMRADIO is not set -# CONFIG_IRDA is not set -# CONFIG_BT is not set CONFIG_NETDEVICES=y CONFIG_DUMMY=y # CONFIG_BONDING is not set @@ -511,6 +509,8 @@ CONFIG_DLCI_MAX=8 # CONFIG_SLIP is not set # CONFIG_SHAPER is not set # CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set # # ISDN subsystem @@ -651,11 +651,28 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_VIAPRO is not set # CONFIG_I2C_VOODOO3 is not set # CONFIG_I2C_PCA_ISA is not set +CONFIG_I2C_SENSOR=y # -# Hardware Sensors Chip support +# Miscellaneous I2C Chip support # -CONFIG_I2C_SENSOR=y +# CONFIG_SENSORS_DS1337 is not set +# CONFIG_SENSORS_DS1374 is not set +CONFIG_SENSORS_EEPROM=y +# CONFIG_SENSORS_PCF8574 is not set +# CONFIG_SENSORS_PCA9539 is not set +# CONFIG_SENSORS_PCF8591 is not set +# CONFIG_SENSORS_RTC8564 is not set +# CONFIG_SENSORS_MAX6875 is not set +# CONFIG_I2C_DEBUG_CORE is not set +# CONFIG_I2C_DEBUG_ALGO is not set +# CONFIG_I2C_DEBUG_BUS is not set +# CONFIG_I2C_DEBUG_CHIP is not set + +# +# Hardware Monitoring support +# +CONFIG_HWMON=y # CONFIG_SENSORS_ADM1021 is not set # CONFIG_SENSORS_ADM1025 is not set # CONFIG_SENSORS_ADM1026 is not set @@ -681,30 +698,15 @@ CONFIG_I2C_SENSOR=y # CONFIG_SENSORS_LM92 is not set # CONFIG_SENSORS_MAX1619 is not set # CONFIG_SENSORS_PC87360 is not set -# CONFIG_SENSORS_SMSC47B397 is not set # CONFIG_SENSORS_SIS5595 is not set # CONFIG_SENSORS_SMSC47M1 is not set +# CONFIG_SENSORS_SMSC47B397 is not set # CONFIG_SENSORS_VIA686A is not set # CONFIG_SENSORS_W83781D is not set # CONFIG_SENSORS_W83L785TS is not set # CONFIG_SENSORS_W83627HF is not set # CONFIG_SENSORS_W83627EHF is not set - -# -# Other I2C Chip support -# -# CONFIG_SENSORS_DS1337 is not set -# CONFIG_SENSORS_DS1374 is not set -CONFIG_SENSORS_EEPROM=y -# CONFIG_SENSORS_PCF8574 is not set -# CONFIG_SENSORS_PCA9539 is not set -# CONFIG_SENSORS_PCF8591 is not set -# CONFIG_SENSORS_RTC8564 is not set -# CONFIG_SENSORS_MAX6875 is not set -# CONFIG_I2C_DEBUG_CORE is not set -# CONFIG_I2C_DEBUG_ALGO is not set -# CONFIG_I2C_DEBUG_BUS is not set -# CONFIG_I2C_DEBUG_CHIP is not set +# CONFIG_HWMON_DEBUG_CHIP is not set # # Misc devices @@ -772,6 +774,7 @@ CONFIG_FS_POSIX_ACL=y # CONFIG_XFS_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set +CONFIG_INOTIFY=y # CONFIG_QUOTA is not set CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set @@ -814,8 +817,7 @@ CONFIG_RAMFS=y # CONFIG_JFFS_FS is not set CONFIG_JFFS2_FS=y CONFIG_JFFS2_FS_DEBUG=0 -# CONFIG_JFFS2_FS_NAND is not set -# CONFIG_JFFS2_FS_NOR_ECC is not set +CONFIG_JFFS2_FS_WRITEBUFFER=y # CONFIG_JFFS2_COMPRESSION_OPTIONS is not set CONFIG_JFFS2_ZLIB=y CONFIG_JFFS2_RTIME=y diff --git a/arch/arm/configs/ixdp2800_defconfig b/arch/arm/configs/ixdp2800_defconfig index d2bb0b7153fe..f4d30e3d009e 100644 --- a/arch/arm/configs/ixdp2800_defconfig +++ b/arch/arm/configs/ixdp2800_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.13-rc2 -# Thu Jul 7 16:49:20 2005 +# Linux kernel version: 2.6.13 +# Wed Sep 14 10:52:23 2005 # CONFIG_ARM=y CONFIG_MMU=y @@ -136,7 +136,6 @@ CONFIG_PCI_NAMES=y # # Kernel Features # -# CONFIG_SMP is not set # CONFIG_PREEMPT is not set # CONFIG_NO_IDLE_HZ is not set # CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set @@ -179,6 +178,68 @@ CONFIG_BINFMT_ELF=y # # CONFIG_PM is not set +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +CONFIG_PACKET=y +CONFIG_PACKET_MMAP=y +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +CONFIG_IP_PNP=y +CONFIG_IP_PNP_DHCP=y +CONFIG_IP_PNP_BOOTP=y +# CONFIG_IP_PNP_RARP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +CONFIG_SYN_COOKIES=y +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_TUNNEL is not set +# CONFIG_IP_TCPDIAG is not set +# CONFIG_IP_TCPDIAG_IPV6 is not set +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_BIC=y +# CONFIG_IPV6 is not set +# CONFIG_NETFILTER is not set + +# +# SCTP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_SCTP is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_NET_DIVERT is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +# CONFIG_NET_SCHED is not set +# CONFIG_NET_CLS_ROUTE is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set + # # Device Drivers # @@ -249,6 +310,7 @@ CONFIG_MTD_COMPLEX_MAPPINGS=y CONFIG_MTD_IXP2000=y # CONFIG_MTD_EDB7312 is not set # CONFIG_MTD_PCI is not set +# CONFIG_MTD_PLATRAM is not set # # Self-contained MTD device drivers @@ -335,72 +397,8 @@ CONFIG_IOSCHED_CFQ=y # CONFIG_I2O is not set # -# Networking support +# Network device support # -CONFIG_NET=y - -# -# Networking options -# -CONFIG_PACKET=y -CONFIG_PACKET_MMAP=y -CONFIG_UNIX=y -# CONFIG_NET_KEY is not set -CONFIG_INET=y -# CONFIG_IP_MULTICAST is not set -# CONFIG_IP_ADVANCED_ROUTER is not set -CONFIG_IP_FIB_HASH=y -CONFIG_IP_PNP=y -CONFIG_IP_PNP_DHCP=y -CONFIG_IP_PNP_BOOTP=y -# CONFIG_IP_PNP_RARP is not set -# CONFIG_NET_IPIP is not set -# CONFIG_NET_IPGRE is not set -# CONFIG_ARPD is not set -CONFIG_SYN_COOKIES=y -# CONFIG_INET_AH is not set -# CONFIG_INET_ESP is not set -# CONFIG_INET_IPCOMP is not set -# CONFIG_INET_TUNNEL is not set -# CONFIG_IP_TCPDIAG is not set -# CONFIG_IP_TCPDIAG_IPV6 is not set -# CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y -# CONFIG_IPV6 is not set -# CONFIG_NETFILTER is not set - -# -# SCTP Configuration (EXPERIMENTAL) -# -# CONFIG_IP_SCTP is not set -# CONFIG_ATM is not set -# CONFIG_BRIDGE is not set -# CONFIG_VLAN_8021Q is not set -# CONFIG_DECNET is not set -# CONFIG_LLC2 is not set -# CONFIG_IPX is not set -# CONFIG_ATALK is not set -# CONFIG_X25 is not set -# CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set -# CONFIG_ECONET is not set -# CONFIG_WAN_ROUTER is not set - -# -# QoS and/or fair queueing -# -# CONFIG_NET_SCHED is not set -# CONFIG_NET_CLS_ROUTE is not set - -# -# Network testing -# -# CONFIG_NET_PKTGEN is not set -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set -# CONFIG_HAMRADIO is not set -# CONFIG_IRDA is not set -# CONFIG_BT is not set CONFIG_NETDEVICES=y CONFIG_DUMMY=y # CONFIG_BONDING is not set @@ -510,6 +508,8 @@ CONFIG_DLCI_MAX=8 # CONFIG_SLIP is not set # CONFIG_SHAPER is not set # CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set # # ISDN subsystem @@ -650,11 +650,28 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_VIAPRO is not set # CONFIG_I2C_VOODOO3 is not set # CONFIG_I2C_PCA_ISA is not set +CONFIG_I2C_SENSOR=y # -# Hardware Sensors Chip support +# Miscellaneous I2C Chip support # -CONFIG_I2C_SENSOR=y +# CONFIG_SENSORS_DS1337 is not set +# CONFIG_SENSORS_DS1374 is not set +CONFIG_SENSORS_EEPROM=y +# CONFIG_SENSORS_PCF8574 is not set +# CONFIG_SENSORS_PCA9539 is not set +# CONFIG_SENSORS_PCF8591 is not set +# CONFIG_SENSORS_RTC8564 is not set +# CONFIG_SENSORS_MAX6875 is not set +# CONFIG_I2C_DEBUG_CORE is not set +# CONFIG_I2C_DEBUG_ALGO is not set +# CONFIG_I2C_DEBUG_BUS is not set +# CONFIG_I2C_DEBUG_CHIP is not set + +# +# Hardware Monitoring support +# +CONFIG_HWMON=y # CONFIG_SENSORS_ADM1021 is not set # CONFIG_SENSORS_ADM1025 is not set # CONFIG_SENSORS_ADM1026 is not set @@ -680,30 +697,15 @@ CONFIG_I2C_SENSOR=y # CONFIG_SENSORS_LM92 is not set # CONFIG_SENSORS_MAX1619 is not set # CONFIG_SENSORS_PC87360 is not set -# CONFIG_SENSORS_SMSC47B397 is not set # CONFIG_SENSORS_SIS5595 is not set # CONFIG_SENSORS_SMSC47M1 is not set +# CONFIG_SENSORS_SMSC47B397 is not set # CONFIG_SENSORS_VIA686A is not set # CONFIG_SENSORS_W83781D is not set # CONFIG_SENSORS_W83L785TS is not set # CONFIG_SENSORS_W83627HF is not set # CONFIG_SENSORS_W83627EHF is not set - -# -# Other I2C Chip support -# -# CONFIG_SENSORS_DS1337 is not set -# CONFIG_SENSORS_DS1374 is not set -CONFIG_SENSORS_EEPROM=y -# CONFIG_SENSORS_PCF8574 is not set -# CONFIG_SENSORS_PCA9539 is not set -# CONFIG_SENSORS_PCF8591 is not set -# CONFIG_SENSORS_RTC8564 is not set -# CONFIG_SENSORS_MAX6875 is not set -# CONFIG_I2C_DEBUG_CORE is not set -# CONFIG_I2C_DEBUG_ALGO is not set -# CONFIG_I2C_DEBUG_BUS is not set -# CONFIG_I2C_DEBUG_CHIP is not set +# CONFIG_HWMON_DEBUG_CHIP is not set # # Misc devices @@ -771,6 +773,7 @@ CONFIG_FS_POSIX_ACL=y # CONFIG_XFS_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set +CONFIG_INOTIFY=y # CONFIG_QUOTA is not set CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set @@ -813,8 +816,7 @@ CONFIG_RAMFS=y # CONFIG_JFFS_FS is not set CONFIG_JFFS2_FS=y CONFIG_JFFS2_FS_DEBUG=0 -# CONFIG_JFFS2_FS_NAND is not set -# CONFIG_JFFS2_FS_NOR_ECC is not set +CONFIG_JFFS2_FS_WRITEBUFFER=y # CONFIG_JFFS2_COMPRESSION_OPTIONS is not set CONFIG_JFFS2_ZLIB=y CONFIG_JFFS2_RTIME=y diff --git a/arch/arm/configs/ixdp2801_defconfig b/arch/arm/configs/ixdp2801_defconfig index 2d6f960e3395..2612d90f969e 100644 --- a/arch/arm/configs/ixdp2801_defconfig +++ b/arch/arm/configs/ixdp2801_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.13-rc2 -# Thu Jul 7 16:49:13 2005 +# Linux kernel version: 2.6.13 +# Wed Sep 14 10:52:16 2005 # CONFIG_ARM=y CONFIG_MMU=y @@ -136,7 +136,6 @@ CONFIG_PCI_NAMES=y # # Kernel Features # -# CONFIG_SMP is not set # CONFIG_PREEMPT is not set # CONFIG_NO_IDLE_HZ is not set # CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set @@ -179,6 +178,68 @@ CONFIG_BINFMT_ELF=y # # CONFIG_PM is not set +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +CONFIG_PACKET=y +CONFIG_PACKET_MMAP=y +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +CONFIG_IP_PNP=y +CONFIG_IP_PNP_DHCP=y +CONFIG_IP_PNP_BOOTP=y +# CONFIG_IP_PNP_RARP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +CONFIG_SYN_COOKIES=y +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_TUNNEL is not set +# CONFIG_IP_TCPDIAG is not set +# CONFIG_IP_TCPDIAG_IPV6 is not set +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_BIC=y +# CONFIG_IPV6 is not set +# CONFIG_NETFILTER is not set + +# +# SCTP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_SCTP is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_NET_DIVERT is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +# CONFIG_NET_SCHED is not set +# CONFIG_NET_CLS_ROUTE is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set + # # Device Drivers # @@ -249,6 +310,7 @@ CONFIG_MTD_COMPLEX_MAPPINGS=y CONFIG_MTD_IXP2000=y # CONFIG_MTD_EDB7312 is not set # CONFIG_MTD_PCI is not set +# CONFIG_MTD_PLATRAM is not set # # Self-contained MTD device drivers @@ -335,72 +397,8 @@ CONFIG_IOSCHED_CFQ=y # CONFIG_I2O is not set # -# Networking support +# Network device support # -CONFIG_NET=y - -# -# Networking options -# -CONFIG_PACKET=y -CONFIG_PACKET_MMAP=y -CONFIG_UNIX=y -# CONFIG_NET_KEY is not set -CONFIG_INET=y -# CONFIG_IP_MULTICAST is not set -# CONFIG_IP_ADVANCED_ROUTER is not set -CONFIG_IP_FIB_HASH=y -CONFIG_IP_PNP=y -CONFIG_IP_PNP_DHCP=y -CONFIG_IP_PNP_BOOTP=y -# CONFIG_IP_PNP_RARP is not set -# CONFIG_NET_IPIP is not set -# CONFIG_NET_IPGRE is not set -# CONFIG_ARPD is not set -CONFIG_SYN_COOKIES=y -# CONFIG_INET_AH is not set -# CONFIG_INET_ESP is not set -# CONFIG_INET_IPCOMP is not set -# CONFIG_INET_TUNNEL is not set -# CONFIG_IP_TCPDIAG is not set -# CONFIG_IP_TCPDIAG_IPV6 is not set -# CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y -# CONFIG_IPV6 is not set -# CONFIG_NETFILTER is not set - -# -# SCTP Configuration (EXPERIMENTAL) -# -# CONFIG_IP_SCTP is not set -# CONFIG_ATM is not set -# CONFIG_BRIDGE is not set -# CONFIG_VLAN_8021Q is not set -# CONFIG_DECNET is not set -# CONFIG_LLC2 is not set -# CONFIG_IPX is not set -# CONFIG_ATALK is not set -# CONFIG_X25 is not set -# CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set -# CONFIG_ECONET is not set -# CONFIG_WAN_ROUTER is not set - -# -# QoS and/or fair queueing -# -# CONFIG_NET_SCHED is not set -# CONFIG_NET_CLS_ROUTE is not set - -# -# Network testing -# -# CONFIG_NET_PKTGEN is not set -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set -# CONFIG_HAMRADIO is not set -# CONFIG_IRDA is not set -# CONFIG_BT is not set CONFIG_NETDEVICES=y CONFIG_DUMMY=y # CONFIG_BONDING is not set @@ -511,6 +509,8 @@ CONFIG_DLCI_MAX=8 # CONFIG_SLIP is not set # CONFIG_SHAPER is not set # CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set # # ISDN subsystem @@ -651,11 +651,28 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_VIAPRO is not set # CONFIG_I2C_VOODOO3 is not set # CONFIG_I2C_PCA_ISA is not set +CONFIG_I2C_SENSOR=y # -# Hardware Sensors Chip support +# Miscellaneous I2C Chip support # -CONFIG_I2C_SENSOR=y +# CONFIG_SENSORS_DS1337 is not set +# CONFIG_SENSORS_DS1374 is not set +CONFIG_SENSORS_EEPROM=y +# CONFIG_SENSORS_PCF8574 is not set +# CONFIG_SENSORS_PCA9539 is not set +# CONFIG_SENSORS_PCF8591 is not set +# CONFIG_SENSORS_RTC8564 is not set +# CONFIG_SENSORS_MAX6875 is not set +# CONFIG_I2C_DEBUG_CORE is not set +# CONFIG_I2C_DEBUG_ALGO is not set +# CONFIG_I2C_DEBUG_BUS is not set +# CONFIG_I2C_DEBUG_CHIP is not set + +# +# Hardware Monitoring support +# +CONFIG_HWMON=y # CONFIG_SENSORS_ADM1021 is not set # CONFIG_SENSORS_ADM1025 is not set # CONFIG_SENSORS_ADM1026 is not set @@ -681,30 +698,15 @@ CONFIG_I2C_SENSOR=y # CONFIG_SENSORS_LM92 is not set # CONFIG_SENSORS_MAX1619 is not set # CONFIG_SENSORS_PC87360 is not set -# CONFIG_SENSORS_SMSC47B397 is not set # CONFIG_SENSORS_SIS5595 is not set # CONFIG_SENSORS_SMSC47M1 is not set +# CONFIG_SENSORS_SMSC47B397 is not set # CONFIG_SENSORS_VIA686A is not set # CONFIG_SENSORS_W83781D is not set # CONFIG_SENSORS_W83L785TS is not set # CONFIG_SENSORS_W83627HF is not set # CONFIG_SENSORS_W83627EHF is not set - -# -# Other I2C Chip support -# -# CONFIG_SENSORS_DS1337 is not set -# CONFIG_SENSORS_DS1374 is not set -CONFIG_SENSORS_EEPROM=y -# CONFIG_SENSORS_PCF8574 is not set -# CONFIG_SENSORS_PCA9539 is not set -# CONFIG_SENSORS_PCF8591 is not set -# CONFIG_SENSORS_RTC8564 is not set -# CONFIG_SENSORS_MAX6875 is not set -# CONFIG_I2C_DEBUG_CORE is not set -# CONFIG_I2C_DEBUG_ALGO is not set -# CONFIG_I2C_DEBUG_BUS is not set -# CONFIG_I2C_DEBUG_CHIP is not set +# CONFIG_HWMON_DEBUG_CHIP is not set # # Misc devices @@ -772,6 +774,7 @@ CONFIG_FS_POSIX_ACL=y # CONFIG_XFS_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set +CONFIG_INOTIFY=y # CONFIG_QUOTA is not set CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set @@ -814,8 +817,7 @@ CONFIG_RAMFS=y # CONFIG_JFFS_FS is not set CONFIG_JFFS2_FS=y CONFIG_JFFS2_FS_DEBUG=0 -# CONFIG_JFFS2_FS_NAND is not set -# CONFIG_JFFS2_FS_NOR_ECC is not set +CONFIG_JFFS2_FS_WRITEBUFFER=y # CONFIG_JFFS2_COMPRESSION_OPTIONS is not set CONFIG_JFFS2_ZLIB=y CONFIG_JFFS2_RTIME=y From b2b8c6944544319caa22ed8bcb3443153a1a3b4c Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Thu, 15 Sep 2005 13:02:32 +0100 Subject: [PATCH 052/240] [ARM] 2905/1: enable the ixp2000 i2c bus driver in ixp2000 defconfigs Patch from Lennert Buytenhek It's silly to have I2C enabled in all ixp2000 defconfigs but not to have the ixp2000 bus driver enabled in any of them. This patch enables CONFIG_I2C_IXP2000 for all in-tree ixp2000 boards. Signed-off-by: Lennert Buytenhek Signed-off-by: Deepak Saxena Signed-off-by: Russell King --- arch/arm/configs/enp2611_defconfig | 2 +- arch/arm/configs/ixdp2400_defconfig | 2 +- arch/arm/configs/ixdp2401_defconfig | 2 +- arch/arm/configs/ixdp2800_defconfig | 2 +- arch/arm/configs/ixdp2801_defconfig | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/arm/configs/enp2611_defconfig b/arch/arm/configs/enp2611_defconfig index a8c001f77b1d..30e6444f9aaa 100644 --- a/arch/arm/configs/enp2611_defconfig +++ b/arch/arm/configs/enp2611_defconfig @@ -635,7 +635,7 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_I810 is not set # CONFIG_I2C_PIIX4 is not set # CONFIG_I2C_ISA is not set -# CONFIG_I2C_IXP2000 is not set +CONFIG_I2C_IXP2000=y # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_PROSAVAGE is not set diff --git a/arch/arm/configs/ixdp2400_defconfig b/arch/arm/configs/ixdp2400_defconfig index 5ea356efcb0e..678720fa2e2e 100644 --- a/arch/arm/configs/ixdp2400_defconfig +++ b/arch/arm/configs/ixdp2400_defconfig @@ -636,7 +636,7 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_I810 is not set # CONFIG_I2C_PIIX4 is not set # CONFIG_I2C_ISA is not set -# CONFIG_I2C_IXP2000 is not set +CONFIG_I2C_IXP2000=y # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_PROSAVAGE is not set diff --git a/arch/arm/configs/ixdp2401_defconfig b/arch/arm/configs/ixdp2401_defconfig index c542d7fc0c1b..38c9a721d5c9 100644 --- a/arch/arm/configs/ixdp2401_defconfig +++ b/arch/arm/configs/ixdp2401_defconfig @@ -637,7 +637,7 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_I810 is not set # CONFIG_I2C_PIIX4 is not set # CONFIG_I2C_ISA is not set -# CONFIG_I2C_IXP2000 is not set +CONFIG_I2C_IXP2000=y # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_PROSAVAGE is not set diff --git a/arch/arm/configs/ixdp2800_defconfig b/arch/arm/configs/ixdp2800_defconfig index f4d30e3d009e..261e2343903b 100644 --- a/arch/arm/configs/ixdp2800_defconfig +++ b/arch/arm/configs/ixdp2800_defconfig @@ -636,7 +636,7 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_I810 is not set # CONFIG_I2C_PIIX4 is not set # CONFIG_I2C_ISA is not set -# CONFIG_I2C_IXP2000 is not set +CONFIG_I2C_IXP2000=y # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_PROSAVAGE is not set diff --git a/arch/arm/configs/ixdp2801_defconfig b/arch/arm/configs/ixdp2801_defconfig index 2612d90f969e..12ef23d1c016 100644 --- a/arch/arm/configs/ixdp2801_defconfig +++ b/arch/arm/configs/ixdp2801_defconfig @@ -637,7 +637,7 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_I2C_I810 is not set # CONFIG_I2C_PIIX4 is not set # CONFIG_I2C_ISA is not set -# CONFIG_I2C_IXP2000 is not set +CONFIG_I2C_IXP2000=y # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_PROSAVAGE is not set From c071760224d8418f65606bd67e755c6a5463ec53 Mon Sep 17 00:00:00 2001 From: Richard Purdie Date: Thu, 15 Sep 2005 14:52:00 +0100 Subject: [PATCH 053/240] [ARM] 2915/1: SA1100 Collie: Correct scoop device calls Patch from Richard Purdie This patch adds a missing parameter to the scoop calls made by collie.c Signed-off-by: Richard Purdie Signed-off-by: Russell King --- arch/arm/mach-sa1100/collie.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/arch/arm/mach-sa1100/collie.c b/arch/arm/mach-sa1100/collie.c index 8cb69113a57c..25d6a4e27533 100644 --- a/arch/arm/mach-sa1100/collie.c +++ b/arch/arm/mach-sa1100/collie.c @@ -111,12 +111,11 @@ static struct mtd_partition collie_partitions[] = { static void collie_set_vpp(int vpp) { - write_scoop_reg(SCOOP_GPCR, read_scoop_reg(SCOOP_GPCR) | COLLIE_SCP_VPEN); - if (vpp) { - write_scoop_reg(SCOOP_GPWR, read_scoop_reg(SCOOP_GPWR) | COLLIE_SCP_VPEN); - } else { - write_scoop_reg(SCOOP_GPWR, read_scoop_reg(SCOOP_GPWR) & ~COLLIE_SCP_VPEN); - } + write_scoop_reg(&colliescoop_device.dev, SCOOP_GPCR, read_scoop_reg(SCOOP_GPCR) | COLLIE_SCP_VPEN); + if (vpp) + write_scoop_reg(&colliescoop_device.dev, SCOOP_GPWR, read_scoop_reg(SCOOP_GPWR) | COLLIE_SCP_VPEN); + else + write_scoop_reg(&colliescoop_device.dev, SCOOP_GPWR, read_scoop_reg(SCOOP_GPWR) & ~COLLIE_SCP_VPEN); } static struct flash_platform_data collie_flash_data = { From fdce05bbfe171882adf53ce32e04da24276cb8cd Mon Sep 17 00:00:00 2001 From: Richard Purdie Date: Thu, 15 Sep 2005 14:53:21 +0100 Subject: [PATCH 054/240] [ARM] 2912/1: PXA Corgi: Cleanup some unneeded code Patch from Richard Purdie This patch cleans up the PXA Corgi platform code removing an unneeded static iomap, an unneeded function and some debug messages. Signed-off-by: Richard Purdie Signed-off-by: Russell King --- arch/arm/mach-pxa/corgi.c | 57 +++++++++++++-------------------------- 1 file changed, 18 insertions(+), 39 deletions(-) diff --git a/arch/arm/mach-pxa/corgi.c b/arch/arm/mach-pxa/corgi.c index 426c2bc517eb..be37586cb1b0 100644 --- a/arch/arm/mach-pxa/corgi.c +++ b/arch/arm/mach-pxa/corgi.c @@ -198,13 +198,10 @@ static void corgi_mci_setpower(struct device *dev, unsigned int vdd) { struct pxamci_platform_data* p_d = dev->platform_data; - if (( 1 << vdd) & p_d->ocr_mask) { - printk(KERN_DEBUG "%s: on\n", __FUNCTION__); + if (( 1 << vdd) & p_d->ocr_mask) GPSR1 = GPIO_bit(CORGI_GPIO_SD_PWR); - } else { - printk(KERN_DEBUG "%s: off\n", __FUNCTION__); + else GPCR1 = GPIO_bit(CORGI_GPIO_SD_PWR); - } } static int corgi_mci_get_ro(struct device *dev) @@ -259,6 +256,16 @@ static struct platform_device *devices[] __initdata = { static void __init corgi_init(void) { + /* setup sleep mode values */ + PWER = 0x00000002; + PFER = 0x00000000; + PRER = 0x00000002; + PGSR0 = 0x0158C000; + PGSR1 = 0x00FF0080; + PGSR2 = 0x0001C004; + /* Stop 3.6MHz and drive HIGH to PCMCIA and CS */ + PCFR |= PCFR_OPDE; + corgi_ssp_set_machinfo(&corgi_ssp_machinfo); pxa_gpio_mode(CORGI_GPIO_USB_PULLUP | GPIO_OUT); @@ -285,42 +292,14 @@ static void __init fixup_corgi(struct machine_desc *desc, mi->bank[0].size = (64*1024*1024); } -static void __init corgi_init_irq(void) -{ - pxa_init_irq(); -} - -static struct map_desc corgi_io_desc[] __initdata = { -/* virtual physical length */ -/* { 0xf1000000, 0x08000000, 0x01000000, MT_DEVICE },*/ /* LCDC (readable for Qt driver) */ -/* { 0xef700000, 0x10800000, 0x00001000, MT_DEVICE },*/ /* SCOOP */ - { 0xef800000, 0x00000000, 0x00800000, MT_DEVICE }, /* Boot Flash */ -}; - -static void __init corgi_map_io(void) -{ - pxa_map_io(); - iotable_init(corgi_io_desc,ARRAY_SIZE(corgi_io_desc)); - - /* setup sleep mode values */ - PWER = 0x00000002; - PFER = 0x00000000; - PRER = 0x00000002; - PGSR0 = 0x0158C000; - PGSR1 = 0x00FF0080; - PGSR2 = 0x0001C004; - /* Stop 3.6MHz and drive HIGH to PCMCIA and CS */ - PCFR |= PCFR_OPDE; -} - #ifdef CONFIG_MACH_CORGI MACHINE_START(CORGI, "SHARP Corgi") .phys_ram = 0xa0000000, .phys_io = 0x40000000, .io_pg_offst = (io_p2v(0x40000000) >> 18) & 0xfffc, .fixup = fixup_corgi, - .map_io = corgi_map_io, - .init_irq = corgi_init_irq, + .map_io = pxa_map_io, + .init_irq = pxa_init_irq, .init_machine = corgi_init, .timer = &pxa_timer, MACHINE_END @@ -332,8 +311,8 @@ MACHINE_START(SHEPHERD, "SHARP Shepherd") .phys_io = 0x40000000, .io_pg_offst = (io_p2v(0x40000000) >> 18) & 0xfffc, .fixup = fixup_corgi, - .map_io = corgi_map_io, - .init_irq = corgi_init_irq, + .map_io = pxa_map_io, + .init_irq = pxa_init_irq, .init_machine = corgi_init, .timer = &pxa_timer, MACHINE_END @@ -345,8 +324,8 @@ MACHINE_START(HUSKY, "SHARP Husky") .phys_io = 0x40000000, .io_pg_offst = (io_p2v(0x40000000) >> 18) & 0xfffc, .fixup = fixup_corgi, - .map_io = corgi_map_io, - .init_irq = corgi_init_irq, + .map_io = pxa_map_io, + .init_irq = pxa_init_irq, .init_machine = corgi_init, .timer = &pxa_timer, MACHINE_END From f29d245549aa38325c37716dbecea8d817c00274 Mon Sep 17 00:00:00 2001 From: Richard Purdie Date: Thu, 15 Sep 2005 14:53:22 +0100 Subject: [PATCH 055/240] [ARM] 2913/1: PXA Poodle: Cleanup some unneeded code Patch from Richard Purdie This patch cleans up the PXA Poodle platform code removing an unneeded static iomap. It also corrects errors in the platform header file and adds a missing GPIO define. Signed-off-by: Richard Purdie Signed-off-by: Russell King --- arch/arm/mach-pxa/poodle.c | 31 ++++++++++--------------------- include/asm-arm/arch-pxa/poodle.h | 25 +++++++++++++------------ 2 files changed, 23 insertions(+), 33 deletions(-) diff --git a/arch/arm/mach-pxa/poodle.c b/arch/arm/mach-pxa/poodle.c index 47cfb8bb8318..7c4a1cebfd6b 100644 --- a/arch/arm/mach-pxa/poodle.c +++ b/arch/arm/mach-pxa/poodle.c @@ -126,6 +126,15 @@ static void __init poodle_init(void) { int ret = 0; + /* setup sleep mode values */ + PWER = 0x00000002; + PFER = 0x00000000; + PRER = 0x00000002; + PGSR0 = 0x00008000; + PGSR1 = 0x003F0202; + PGSR2 = 0x0001C000; + PCFR |= PCFR_OPDE; + /* cpu initialize */ /* Pgsr Register */ PGSR0 = 0x0146dd80; @@ -171,32 +180,12 @@ static void __init fixup_poodle(struct machine_desc *desc, sharpsl_save_param(); } -static struct map_desc poodle_io_desc[] __initdata = { - /* virtual physical length */ - { 0xef800000, 0x00000000, 0x00800000, MT_DEVICE }, /* Boot Flash */ -}; - -static void __init poodle_map_io(void) -{ - pxa_map_io(); - iotable_init(poodle_io_desc, ARRAY_SIZE(poodle_io_desc)); - - /* setup sleep mode values */ - PWER = 0x00000002; - PFER = 0x00000000; - PRER = 0x00000002; - PGSR0 = 0x00008000; - PGSR1 = 0x003F0202; - PGSR2 = 0x0001C000; - PCFR |= PCFR_OPDE; -} - MACHINE_START(POODLE, "SHARP Poodle") .phys_ram = 0xa0000000, .phys_io = 0x40000000, .io_pg_offst = (io_p2v(0x40000000) >> 18) & 0xfffc, .fixup = fixup_poodle, - .map_io = poodle_map_io, + .map_io = pxa_map_io, .init_irq = pxa_init_irq, .timer = &pxa_timer, .init_machine = poodle_init, diff --git a/include/asm-arm/arch-pxa/poodle.h b/include/asm-arm/arch-pxa/poodle.h index 58bda9d571a5..6b5ac5144e70 100644 --- a/include/asm-arm/arch-pxa/poodle.h +++ b/include/asm-arm/arch-pxa/poodle.h @@ -37,24 +37,25 @@ #define POODLE_GPIO_nSD_DETECT (9) #define POODLE_GPIO_MAIN_BAT_LOW (13) #define POODLE_GPIO_BAT_COVER (13) +#define POODLE_GPIO_USB_PULLUP (20) #define POODLE_GPIO_ADC_TEMP_ON (21) #define POODLE_GPIO_BYPASS_ON (36) #define POODLE_GPIO_CHRG_ON (38) #define POODLE_GPIO_CHRG_FULL (16) /* PXA GPIOs */ -#define POODLE_IRQ_GPIO_ON_KEY IRQ_GPIO0 -#define POODLE_IRQ_GPIO_AC_IN IRQ_GPIO1 -#define POODLE_IRQ_GPIO_HP_IN IRQ_GPIO4 -#define POODLE_IRQ_GPIO_CO IRQ_GPIO16 -#define POODLE_IRQ_GPIO_TP_INT IRQ_GPIO5 -#define POODLE_IRQ_GPIO_WAKEUP IRQ_GPIO11 -#define POODLE_IRQ_GPIO_GA_INT IRQ_GPIO10 -#define POODLE_IRQ_GPIO_CF_IRQ IRQ_GPIO17 -#define POODLE_IRQ_GPIO_CF_CD IRQ_GPIO14 -#define POODLE_IRQ_GPIO_nSD_INT IRQ_GPIO8 -#define POODLE_IRQ_GPIO_nSD_DETECT IRQ_GPIO9 -#define POODLE_IRQ_GPIO_MAIN_BAT_LOW IRQ_GPIO13 +#define POODLE_IRQ_GPIO_ON_KEY IRQ_GPIO(0) +#define POODLE_IRQ_GPIO_AC_IN IRQ_GPIO(1) +#define POODLE_IRQ_GPIO_HP_IN IRQ_GPIO(4) +#define POODLE_IRQ_GPIO_CO IRQ_GPIO(16) +#define POODLE_IRQ_GPIO_TP_INT IRQ_GPIO(5) +#define POODLE_IRQ_GPIO_WAKEUP IRQ_GPIO(11) +#define POODLE_IRQ_GPIO_GA_INT IRQ_GPIO(10) +#define POODLE_IRQ_GPIO_CF_IRQ IRQ_GPIO(17) +#define POODLE_IRQ_GPIO_CF_CD IRQ_GPIO(14) +#define POODLE_IRQ_GPIO_nSD_INT IRQ_GPIO(8) +#define POODLE_IRQ_GPIO_nSD_DETECT IRQ_GPIO(9) +#define POODLE_IRQ_GPIO_MAIN_BAT_LOW IRQ_GPIO(13) /* SCOOP GPIOs */ #define POODLE_SCOOP_CHARGE_ON SCOOP_GPCR_PA11 From 13b9d47ed36e8019a99c798b84c341ef75868e02 Mon Sep 17 00:00:00 2001 From: Richard Purdie Date: Thu, 15 Sep 2005 14:53:22 +0100 Subject: [PATCH 056/240] [ARM] 2914/1: PXA Poodle: Add MMC and UDC support Patch from Richard Purdie This patch adds MMC and UDC support to the PXA Poodle platform. Signed-off-by: Richard Purdie Signed-off-by: Russell King --- arch/arm/mach-pxa/poodle.c | 82 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 82 insertions(+) diff --git a/arch/arm/mach-pxa/poodle.c b/arch/arm/mach-pxa/poodle.c index 7c4a1cebfd6b..f25638810017 100644 --- a/arch/arm/mach-pxa/poodle.c +++ b/arch/arm/mach-pxa/poodle.c @@ -30,6 +30,8 @@ #include #include +#include +#include #include #include @@ -93,6 +95,83 @@ static struct platform_device locomo_device = { .resource = locomo_resources, }; + +/* + * MMC/SD Device + * + * The card detect interrupt isn't debounced so we delay it by 250ms + * to give the card a chance to fully insert/eject. + */ +static struct pxamci_platform_data poodle_mci_platform_data; + +static int poodle_mci_init(struct device *dev, irqreturn_t (*poodle_detect_int)(int, void *, struct pt_regs *), void *data) +{ + int err; + + /* setup GPIO for PXA25x MMC controller */ + pxa_gpio_mode(GPIO6_MMCCLK_MD); + pxa_gpio_mode(GPIO8_MMCCS0_MD); + pxa_gpio_mode(POODLE_GPIO_nSD_DETECT | GPIO_IN); + pxa_gpio_mode(POODLE_GPIO_SD_PWR | GPIO_OUT); + + poodle_mci_platform_data.detect_delay = msecs_to_jiffies(250); + + err = request_irq(POODLE_IRQ_GPIO_nSD_DETECT, poodle_detect_int, SA_INTERRUPT, + "MMC card detect", data); + if (err) { + printk(KERN_ERR "poodle_mci_init: MMC/SD: can't request MMC card detect IRQ\n"); + return -1; + } + + set_irq_type(POODLE_IRQ_GPIO_nSD_DETECT, IRQT_BOTHEDGE); + + return 0; +} + +static void poodle_mci_setpower(struct device *dev, unsigned int vdd) +{ + struct pxamci_platform_data* p_d = dev->platform_data; + + if (( 1 << vdd) & p_d->ocr_mask) + GPSR1 = GPIO_bit(POODLE_GPIO_SD_PWR); + else + GPCR1 = GPIO_bit(POODLE_GPIO_SD_PWR); +} + +static void poodle_mci_exit(struct device *dev, void *data) +{ + free_irq(POODLE_IRQ_GPIO_nSD_DETECT, data); +} + +static struct pxamci_platform_data poodle_mci_platform_data = { + .ocr_mask = MMC_VDD_32_33|MMC_VDD_33_34, + .init = poodle_mci_init, + .setpower = poodle_mci_setpower, + .exit = poodle_mci_exit, +}; + + +/* + * USB Device Controller + */ +static void poodle_udc_command(int cmd) +{ + switch(cmd) { + case PXA2XX_UDC_CMD_CONNECT: + GPSR(POODLE_GPIO_USB_PULLUP) = GPIO_bit(POODLE_GPIO_USB_PULLUP); + break; + case PXA2XX_UDC_CMD_DISCONNECT: + GPCR(POODLE_GPIO_USB_PULLUP) = GPIO_bit(POODLE_GPIO_USB_PULLUP); + break; + } +} + +static struct pxa2xx_udc_mach_info udc_info __initdata = { + /* no connect GPIO; poodle can't tell connection status */ + .udc_command = poodle_udc_command, +}; + + /* PXAFB device */ static struct pxafb_mach_info poodle_fb_info __initdata = { .pixclock = 144700, @@ -164,6 +243,9 @@ static void __init poodle_init(void) GPSR2 = 0x00000000; set_pxa_fb_info(&poodle_fb_info); + pxa_gpio_mode(POODLE_GPIO_USB_PULLUP | GPIO_OUT); + pxa_set_udc_info(&udc_info); + pxa_set_mci_info(&poodle_mci_platform_data); scoop_num = 1; scoop_devs = &poodle_pcmcia_scoop[0]; From b5683557331b129658ab435391df527ef72d63dc Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Thu, 15 Sep 2005 08:59:36 -0500 Subject: [PATCH 057/240] [SCSI] atp870u: fix memory addressing bug From: Alan Cox The virt_to_bus() wasn't correctly taken out of this driver. It needs to be able to track both physical and virtual addresses for its prd table. Update the driver to do this with separate tracking entries. Signed-off-by: Alan Cox Signed-off-by: James Bottomley --- drivers/scsi/atp870u.c | 6 ++++-- drivers/scsi/atp870u.h | 5 +++-- 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/scsi/atp870u.c b/drivers/scsi/atp870u.c index e6153fe5842a..a8cfbef304b5 100644 --- a/drivers/scsi/atp870u.c +++ b/drivers/scsi/atp870u.c @@ -996,6 +996,7 @@ oktosend: #ifdef ED_DBGP printk("send_s870: prdaddr_2 0x%8x tmpcip %x target_id %d\n", dev->id[c][target_id].prdaddr,tmpcip,target_id); #endif + dev->id[c][target_id].prdaddr = dev->id[c][target_id].prd_bus; outl(dev->id[c][target_id].prdaddr, tmpcip); tmpcip = tmpcip - 2; outb(0x06, tmpcip); @@ -2572,7 +2573,7 @@ static void atp870u_free_tables(struct Scsi_Host *host) for (k = 0; k < 16; k++) { if (!atp_dev->id[j][k].prd_table) continue; - pci_free_consistent(atp_dev->pdev, 1024, atp_dev->id[j][k].prd_table, atp_dev->id[j][k].prdaddr); + pci_free_consistent(atp_dev->pdev, 1024, atp_dev->id[j][k].prd_table, atp_dev->id[j][k].prd_bus); atp_dev->id[j][k].prd_table = NULL; } } @@ -2584,12 +2585,13 @@ static int atp870u_init_tables(struct Scsi_Host *host) int c,k; for(c=0;c < 2;c++) { for(k=0;k<16;k++) { - atp_dev->id[c][k].prd_table = pci_alloc_consistent(atp_dev->pdev, 1024, &(atp_dev->id[c][k].prdaddr)); + atp_dev->id[c][k].prd_table = pci_alloc_consistent(atp_dev->pdev, 1024, &(atp_dev->id[c][k].prd_bus)); if (!atp_dev->id[c][k].prd_table) { printk("atp870u_init_tables fail\n"); atp870u_free_tables(host); return -ENOMEM; } + atp_dev->id[c][k].prdaddr = atp_dev->id[c][k].prd_bus; atp_dev->id[c][k].devsp=0x20; atp_dev->id[c][k].devtype = 0x7f; atp_dev->id[c][k].curr_req = NULL; diff --git a/drivers/scsi/atp870u.h b/drivers/scsi/atp870u.h index 89f43af39cf2..62bae64a01c1 100644 --- a/drivers/scsi/atp870u.h +++ b/drivers/scsi/atp870u.h @@ -54,8 +54,9 @@ struct atp_unit unsigned long tran_len; unsigned long last_len; unsigned char *prd_pos; - unsigned char *prd_table; - dma_addr_t prdaddr; + unsigned char *prd_table; /* Kernel address of PRD table */ + dma_addr_t prd_bus; /* Bus address of PRD */ + dma_addr_t prdaddr; /* Dynamically updated in driver */ struct scsi_cmnd *curr_req; } id[2][16]; struct Scsi_Host *host; From 1b3cb73f7306f97a68fa973dec9f3c3b68bd29cf Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 15 Sep 2005 15:17:59 +0100 Subject: [PATCH 058/240] [ARM] Tighten pfn_valid() test. Thomas Gleixner reported that mmaping and unmapping each physical page in turn eventually caused the kernel to oops. It appears that pfn_valid() in the discontigmem case was too simplistic for proper operation. Tighten the logic so we also check if the PFN is within the range of the selected memory node. Signed-off-by: Russell King --- include/asm-arm/memory.h | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/include/asm-arm/memory.h b/include/asm-arm/memory.h index e47bea7d1723..a8a933a775db 100644 --- a/include/asm-arm/memory.h +++ b/include/asm-arm/memory.h @@ -160,12 +160,25 @@ static inline __deprecated void *bus_to_virt(unsigned long x) #define page_to_pfn(page) \ (( (page) - page_zone(page)->zone_mem_map) \ + page_zone(page)->zone_start_pfn) + #define pfn_to_page(pfn) \ (PFN_TO_MAPBASE(pfn) + LOCAL_MAP_NR((pfn) << PAGE_SHIFT)) -#define pfn_valid(pfn) (PFN_TO_NID(pfn) < MAX_NUMNODES) + +#define pfn_valid(pfn) \ + ({ \ + unsigned int nid = PFN_TO_NID(pfn); \ + int valid = nid < MAX_NUMNODES; \ + if (valid) { \ + pg_data_t *node = NODE_DATA(nid); \ + valid = (pfn - node->node_start_pfn) < \ + node->node_spanned_pages; \ + } \ + valid; \ + }) #define virt_to_page(kaddr) \ (ADDR_TO_MAPBASE(kaddr) + LOCAL_MAP_NR(kaddr)) + #define virt_addr_valid(kaddr) (KVADDR_TO_NID(kaddr) < MAX_NUMNODES) /* From be201f7f4cd26b1bb994062803e0d87877e749ce Mon Sep 17 00:00:00 2001 From: Jimi Xenidis Date: Thu, 15 Sep 2005 08:42:19 -0400 Subject: [PATCH 059/240] [PATCH] ppc64: Fix recent regression As noted by Olof Johansson : "A recent patch changed the way the LPAR bit is checked during early boot. This resulted in a polarity change in a conditional branch without changing the branch, causing at least some legacy machines to not boot." This fixes it. Signed-off-by: Jimi Xenidis Signed-off-by: Linus Torvalds --- arch/ppc64/kernel/head.S | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/ppc64/kernel/head.S b/arch/ppc64/kernel/head.S index 58c314738c99..72c61041151a 100644 --- a/arch/ppc64/kernel/head.S +++ b/arch/ppc64/kernel/head.S @@ -1649,7 +1649,7 @@ _GLOBAL(__secondary_start) ld r3,0(r3) lwz r3,PLATFORM(r3) /* r3 = platform flags */ andi. r3,r3,PLATFORM_LPAR /* Test if bit 0 is set (LPAR bit) */ - bne 98f + beq 98f /* branch if result is 0 */ mfspr r3,PVR srwi r3,r3,16 cmpwi r3,0x37 /* SStar */ @@ -1813,7 +1813,7 @@ _STATIC(start_here_multiplatform) ld r3,0(r3) lwz r3,PLATFORM(r3) /* r3 = platform flags */ andi. r3,r3,PLATFORM_LPAR /* Test if bit 0 is set (LPAR bit) */ - bne 98f + beq 98f /* branch if result is 0 */ mfspr r3,PVR srwi r3,r3,16 cmpwi r3,0x37 /* SStar */ @@ -1834,7 +1834,7 @@ _STATIC(start_here_multiplatform) lwz r3,PLATFORM(r3) /* r3 = platform flags */ /* Test if bit 0 is set (LPAR bit) */ andi. r3,r3,PLATFORM_LPAR - bne 98f + bne 98f /* branch if result is !0 */ LOADADDR(r6,_SDR1) /* Only if NOT LPAR */ sub r6,r6,r26 ld r6,0(r6) /* get the value of _SDR1 */ From 7b6a186d65589901a73d460070504a5e02703c45 Mon Sep 17 00:00:00 2001 From: "Antonino A. Daplas" Date: Thu, 15 Sep 2005 20:58:57 +0800 Subject: [PATCH 060/240] [PATCH] savagefb: Fix load failure of the Twister chipset - The Twister chipsets are actually prosavages. Reclassify them as such and remove the S3_SAVAGE_TWISTER id. - Fix i2c code if fb_firmware_edid() returns NULL Signed-off-by: Antonino Daplas Signed-off-by: Linus Torvalds --- drivers/video/savage/savagefb-i2c.c | 11 +++++++---- drivers/video/savage/savagefb.h | 4 ---- drivers/video/savage/savagefb_driver.c | 11 +++++------ 3 files changed, 12 insertions(+), 14 deletions(-) diff --git a/drivers/video/savage/savagefb-i2c.c b/drivers/video/savage/savagefb-i2c.c index 959404ad68f4..3c98457783c4 100644 --- a/drivers/video/savage/savagefb-i2c.c +++ b/drivers/video/savage/savagefb-i2c.c @@ -274,10 +274,13 @@ int savagefb_probe_i2c_connector(struct fb_info *info, u8 **out_edid) if (!edid) { /* try to get from firmware */ - edid = kmalloc(EDID_LENGTH, GFP_KERNEL); - if (edid) - memcpy(edid, fb_firmware_edid(info->device), - EDID_LENGTH); + const u8 *e = fb_firmware_edid(info->device); + + if (e) { + edid = kmalloc(EDID_LENGTH, GFP_KERNEL); + if (edid) + memcpy(edid, e, EDID_LENGTH); + } } if (out_edid) diff --git a/drivers/video/savage/savagefb.h b/drivers/video/savage/savagefb.h index d6f94742c9f2..ea17f7e0482c 100644 --- a/drivers/video/savage/savagefb.h +++ b/drivers/video/savage/savagefb.h @@ -60,8 +60,6 @@ #define S3_SAVAGE_SERIES(chip) ((chip>=S3_SAVAGE3D) && (chip<=S3_SAVAGE2000)) -#define S3_MOBILE_TWISTER_SERIES(chip) ((chip==S3_TWISTER) || (chip == S3_PROSAVAGEDDR)) - /* Chip tags. These are used to group the adapters into * related families. */ @@ -74,8 +72,6 @@ typedef enum { S3_PROSAVAGE, S3_SUPERSAVAGE, S3_SAVAGE2000, - S3_PROSAVAGEDDR, - S3_TWISTER, S3_LAST } savage_chipset; diff --git a/drivers/video/savage/savagefb_driver.c b/drivers/video/savage/savagefb_driver.c index b5ca3ef8271f..7c285455c924 100644 --- a/drivers/video/savage/savagefb_driver.c +++ b/drivers/video/savage/savagefb_driver.c @@ -1773,8 +1773,7 @@ static int __devinit savage_init_hw (struct savagefb_par *par) } } - if (S3_SAVAGE_MOBILE_SERIES(par->chip) || - (S3_MOBILE_TWISTER_SERIES(par->chip) && !par->crtonly)) + if (S3_SAVAGE_MOBILE_SERIES(par->chip) && !par->crtonly) par->display_type = DISP_LCD; else if (dvi || (par->chip == S3_SAVAGE4 && par->dvi)) par->display_type = DISP_DFP; @@ -1783,7 +1782,7 @@ static int __devinit savage_init_hw (struct savagefb_par *par) /* Check LCD panel parrmation */ - if (par->chip == S3_SAVAGE_MX) { + if (par->display_type == DISP_LCD) { unsigned char cr6b = VGArCR( 0x6b ); int panelX = (VGArSEQ (0x61) + @@ -1922,15 +1921,15 @@ static int __devinit savage_init_fb_info (struct fb_info *info, snprintf (info->fix.id, 16, "ProSavageKM"); break; case FB_ACCEL_S3TWISTER_P: - par->chip = S3_TWISTER; + par->chip = S3_PROSAVAGE; snprintf (info->fix.id, 16, "TwisterP"); break; case FB_ACCEL_S3TWISTER_K: - par->chip = S3_TWISTER; + par->chip = S3_PROSAVAGE; snprintf (info->fix.id, 16, "TwisterK"); break; case FB_ACCEL_PROSAVAGE_DDR: - par->chip = S3_PROSAVAGEDDR; + par->chip = S3_PROSAVAGE; snprintf (info->fix.id, 16, "ProSavageDDR"); break; case FB_ACCEL_PROSAVAGE_DDRK: From 6d36ba629e0ef47a03d3703ee1d38143c25532a8 Mon Sep 17 00:00:00 2001 From: "Antonino A. Daplas" Date: Thu, 15 Sep 2005 21:14:56 +0800 Subject: [PATCH 061/240] [PATCH] vgacon: Fix sanity checking in vgacon_resize Reported by: walt "I routinely switch the console font during bootup to 8x8 so I can get 50 lines per screen. Until 09 Sept, just changing to the small font automatically gave me all 50 lines -- but now I'm only getting 25 lines even with the small font. The bottom half of the screen displays the text that already scrolled off the top." This bug is due to an erroneous check in the recently added hook, vgacon_resize(). It checks the new height against the original number of rows of the console. Because the original number of rows depends on both the scanline and the font height, check it instead against the scanline/fontheight. Signed-off-by: Antonino Daplas Signed-off-by: Linus Torvalds --- drivers/video/console/vgacon.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c index 0705cd741411..6ef6f7760e47 100644 --- a/drivers/video/console/vgacon.c +++ b/drivers/video/console/vgacon.c @@ -1020,7 +1020,9 @@ static int vgacon_font_get(struct vc_data *c, struct console_font *font) static int vgacon_resize(struct vc_data *c, unsigned int width, unsigned int height) { - if (width % 2 || width > ORIG_VIDEO_COLS || height > ORIG_VIDEO_LINES) + if (width % 2 || width > ORIG_VIDEO_COLS || + height > (ORIG_VIDEO_LINES * vga_default_font_height)/ + c->vc_font.height) return -EINVAL; if (CON_IS_VISIBLE(c) && !vga_is_gfx) /* who knows */ From 065d9cac98a5406ecd5a1368f8fd38f55739dee9 Mon Sep 17 00:00:00 2001 From: "Antonino A. Daplas" Date: Thu, 15 Sep 2005 21:34:33 +0800 Subject: [PATCH 062/240] [PATCH] vc: Use correct size on buffer copy in vc_resize In the unlikely case of the new screen width much wider then the old, use (old_row_size * new_rows) instead of new_screen_size to prevent a buffer overrun during the copy. Signed-off-by: Antonino Daplas Signed-off-by: Linus Torvalds --- drivers/char/vt.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/char/vt.c b/drivers/char/vt.c index 1e33cb032e07..e91268e86833 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -810,13 +810,14 @@ int vc_resize(struct vc_data *vc, unsigned int cols, unsigned int lines) * from the top and bottom of cursor position */ old_origin += (vc->vc_y - new_rows/2) * old_row_size; - end = old_origin + new_screen_size; + end = old_origin + (old_row_size * new_rows); } } else /* * Cursor near the top, copy contents from the top of buffer */ - end = (old_rows > new_rows) ? old_origin + new_screen_size : + end = (old_rows > new_rows) ? old_origin + + (old_row_size * new_rows) : vc->vc_scr_end; update_attr(vc); From 541950027f8b7c96a639bc16e48930c590f1b98a Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Thu, 15 Sep 2005 21:52:51 -0400 Subject: [PATCH 063/240] [SCSI] fix use after potential free in scsi_remove_device Signed-off-by: Alan Stern Signed-off-by: James Bottomley --- drivers/scsi/scsi_sysfs.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c index b8052d5206cc..83f87c41b18d 100644 --- a/drivers/scsi/scsi_sysfs.c +++ b/drivers/scsi/scsi_sysfs.c @@ -707,9 +707,11 @@ void __scsi_remove_device(struct scsi_device *sdev) **/ void scsi_remove_device(struct scsi_device *sdev) { - down(&sdev->host->scan_mutex); + struct Scsi_Host *shost = sdev->host; + + down(&shost->scan_mutex); __scsi_remove_device(sdev); - up(&sdev->host->scan_mutex); + up(&shost->scan_mutex); } EXPORT_SYMBOL(scsi_remove_device); From ed4b9f8014db4f343e89b44b7c5ca355f439ce36 Mon Sep 17 00:00:00 2001 From: Jay Vosburgh Date: Wed, 14 Sep 2005 14:52:09 -0700 Subject: [PATCH 064/240] [PATCH] bonding: plug reference count leak Bonding leaks route structures when the ARP monitor is configured to send probes over VLANs. Originally reported by Ian Abel ; his original fix was modified by Jay Vosburgh to correct coding style and to close a leak it missed. Signed-off-by: Jay Vosburgh Signed-off-by: Jeff Garzik --- drivers/net/bonding/bond_main.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 94c9f68dd16b..f8dedb623dc0 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -2879,6 +2879,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave) * This target is not on a VLAN */ if (rt->u.dst.dev == bond->dev) { + ip_rt_put(rt); dprintk("basa: rtdev == bond->dev: arp_send\n"); bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], bond->master_ip, 0); @@ -2898,6 +2899,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave) } if (vlan_id) { + ip_rt_put(rt); bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], vlan->vlan_ip, vlan_id); continue; @@ -2909,6 +2911,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave) bond->dev->name, NIPQUAD(fl.fl4_dst), rt->u.dst.dev ? rt->u.dst.dev->name : "NULL"); } + ip_rt_put(rt); } } From 8b51292764a0ed46d7eb6bead4625ba83ee2ec05 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 14 Sep 2005 09:45:44 -0700 Subject: [PATCH 065/240] [PATCH] 8139cp: allocate statistics space only when needed Don't crash if ethtool statistics are requested and device is down. Fix is to allocate pci space for statistics only when needed. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik --- drivers/net/8139cp.c | 46 ++++++++++++++++++++++---------------------- 1 file changed, 23 insertions(+), 23 deletions(-) diff --git a/drivers/net/8139cp.c b/drivers/net/8139cp.c index 34b80de34fae..bc537440ca02 100644 --- a/drivers/net/8139cp.c +++ b/drivers/net/8139cp.c @@ -353,8 +353,6 @@ struct cp_private { struct net_device_stats net_stats; struct cp_extra_stats cp_stats; - struct cp_dma_stats *nic_stats; - dma_addr_t nic_stats_dma; unsigned rx_tail ____cacheline_aligned; struct cp_desc *rx_ring; @@ -1143,10 +1141,6 @@ static int cp_alloc_rings (struct cp_private *cp) cp->rx_ring = mem; cp->tx_ring = &cp->rx_ring[CP_RX_RING_SIZE]; - mem += (CP_RING_BYTES - CP_STATS_SIZE); - cp->nic_stats = mem; - cp->nic_stats_dma = cp->ring_dma + (CP_RING_BYTES - CP_STATS_SIZE); - return cp_init_rings(cp); } @@ -1187,7 +1181,6 @@ static void cp_free_rings (struct cp_private *cp) pci_free_consistent(cp->pdev, CP_RING_BYTES, cp->rx_ring, cp->ring_dma); cp->rx_ring = NULL; cp->tx_ring = NULL; - cp->nic_stats = NULL; } static int cp_open (struct net_device *dev) @@ -1516,13 +1509,17 @@ static void cp_get_ethtool_stats (struct net_device *dev, struct ethtool_stats *estats, u64 *tmp_stats) { struct cp_private *cp = netdev_priv(dev); + struct cp_dma_stats *nic_stats; + dma_addr_t dma; int i; - memset(cp->nic_stats, 0, sizeof(struct cp_dma_stats)); + nic_stats = pci_alloc_consistent(cp->pdev, sizeof(*nic_stats), &dma); + if (!nic_stats) + return; /* begin NIC statistics dump */ - cpw32(StatsAddr + 4, (cp->nic_stats_dma >> 16) >> 16); - cpw32(StatsAddr, (cp->nic_stats_dma & 0xffffffff) | DumpStats); + cpw32(StatsAddr + 4, (u64)dma >> 32); + cpw32(StatsAddr, ((u64)dma & DMA_32BIT_MASK) | DumpStats); cpr32(StatsAddr); for (i = 0; i < 1000; i++) { @@ -1532,24 +1529,27 @@ static void cp_get_ethtool_stats (struct net_device *dev, } cpw32(StatsAddr, 0); cpw32(StatsAddr + 4, 0); + cpr32(StatsAddr); i = 0; - tmp_stats[i++] = le64_to_cpu(cp->nic_stats->tx_ok); - tmp_stats[i++] = le64_to_cpu(cp->nic_stats->rx_ok); - tmp_stats[i++] = le64_to_cpu(cp->nic_stats->tx_err); - tmp_stats[i++] = le32_to_cpu(cp->nic_stats->rx_err); - tmp_stats[i++] = le16_to_cpu(cp->nic_stats->rx_fifo); - tmp_stats[i++] = le16_to_cpu(cp->nic_stats->frame_align); - tmp_stats[i++] = le32_to_cpu(cp->nic_stats->tx_ok_1col); - tmp_stats[i++] = le32_to_cpu(cp->nic_stats->tx_ok_mcol); - tmp_stats[i++] = le64_to_cpu(cp->nic_stats->rx_ok_phys); - tmp_stats[i++] = le64_to_cpu(cp->nic_stats->rx_ok_bcast); - tmp_stats[i++] = le32_to_cpu(cp->nic_stats->rx_ok_mcast); - tmp_stats[i++] = le16_to_cpu(cp->nic_stats->tx_abort); - tmp_stats[i++] = le16_to_cpu(cp->nic_stats->tx_underrun); + tmp_stats[i++] = le64_to_cpu(nic_stats->tx_ok); + tmp_stats[i++] = le64_to_cpu(nic_stats->rx_ok); + tmp_stats[i++] = le64_to_cpu(nic_stats->tx_err); + tmp_stats[i++] = le32_to_cpu(nic_stats->rx_err); + tmp_stats[i++] = le16_to_cpu(nic_stats->rx_fifo); + tmp_stats[i++] = le16_to_cpu(nic_stats->frame_align); + tmp_stats[i++] = le32_to_cpu(nic_stats->tx_ok_1col); + tmp_stats[i++] = le32_to_cpu(nic_stats->tx_ok_mcol); + tmp_stats[i++] = le64_to_cpu(nic_stats->rx_ok_phys); + tmp_stats[i++] = le64_to_cpu(nic_stats->rx_ok_bcast); + tmp_stats[i++] = le32_to_cpu(nic_stats->rx_ok_mcast); + tmp_stats[i++] = le16_to_cpu(nic_stats->tx_abort); + tmp_stats[i++] = le16_to_cpu(nic_stats->tx_underrun); tmp_stats[i++] = cp->cp_stats.rx_frags; if (i != CP_NUM_STATS) BUG(); + + pci_free_consistent(cp->pdev, sizeof(*nic_stats), nic_stats, dma); } static struct ethtool_ops cp_ethtool_ops = { From 53abbf7eeab22620586cbba272e569fd3e729b0b Mon Sep 17 00:00:00 2001 From: Jens Osterkamp Date: Fri, 16 Sep 2005 08:55:33 +0200 Subject: [PATCH 066/240] [PATCH] net: fix spider_net media detection This patch makes the driver work with any BladeCenter network switch, it used to work only with certain models. Please apply. Signed-off-by: Arnd Bergmann Signed-off-by: Jeff Garzik --- drivers/net/spider_net.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 4e19220473d0..c796f41b4a52 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -1817,6 +1817,10 @@ spider_net_setup_phy(struct spider_net_card *card) /* LEDs active in both modes, autosense prio = fiber */ spider_net_write_phy(card->netdev, 1, MII_NCONFIG, 0x945f); + /* switch off fibre autoneg */ + spider_net_write_phy(card->netdev, 1, MII_NCONFIG, 0xfc01); + spider_net_write_phy(card->netdev, 1, 0x0b, 0x0004); + phy->def->ops->read_link(phy); pr_info("Found %s with %i Mbps, %s-duplex.\n", phy->def->name, phy->speed, phy->duplex==1 ? "Full" : "Half"); From 7fb6ec287a05d7a71ec086d8bc9a452d5e16ff1a Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Fri, 16 Sep 2005 06:01:48 -0400 Subject: [PATCH 067/240] [libata] fix PIO completion race Make sure we that completion is the final action we take; prior to this change, another CPU may have changed ap->pio_task_state before we tested it a final time. Spotted by, and original patch by Albert Lee @ IBM. Also includes a minor optimization: eliminate a ton of unnecessary queue_work() calls, simply by jumping to the beginning of the FSM function ata_pio_task(). --- drivers/scsi/libata-core.c | 37 +++++++++++++++++++++++-------------- 1 file changed, 23 insertions(+), 14 deletions(-) diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c index 5cc53cd9323e..d92273cbe0de 100644 --- a/drivers/scsi/libata-core.c +++ b/drivers/scsi/libata-core.c @@ -2465,9 +2465,12 @@ static unsigned long ata_pio_poll(struct ata_port *ap) * * LOCKING: * None. (executing in kernel thread context) + * + * RETURNS: + * Non-zero if qc completed, zero otherwise. */ -static void ata_pio_complete (struct ata_port *ap) +static int ata_pio_complete (struct ata_port *ap) { struct ata_queued_cmd *qc; u8 drv_stat; @@ -2486,14 +2489,14 @@ static void ata_pio_complete (struct ata_port *ap) if (drv_stat & (ATA_BUSY | ATA_DRQ)) { ap->pio_task_state = PIO_ST_LAST_POLL; ap->pio_task_timeout = jiffies + ATA_TMOUT_PIO; - return; + return 0; } } drv_stat = ata_wait_idle(ap); if (!ata_ok(drv_stat)) { ap->pio_task_state = PIO_ST_ERR; - return; + return 0; } qc = ata_qc_from_tag(ap, ap->active_tag); @@ -2502,6 +2505,10 @@ static void ata_pio_complete (struct ata_port *ap) ap->pio_task_state = PIO_ST_IDLE; ata_poll_qc_complete(qc, drv_stat); + + /* another command may start at this point */ + + return 1; } @@ -2709,7 +2716,7 @@ static void __atapi_pio_bytes(struct ata_queued_cmd *qc, unsigned int bytes) next_sg: if (unlikely(qc->cursg >= qc->n_elem)) { - /* + /* * The end of qc->sg is reached and the device expects * more data to transfer. In order not to overrun qc->sg * and fulfill length specified in the byte count register, @@ -2721,7 +2728,7 @@ next_sg: unsigned int i; if (words) /* warning if bytes > 1 */ - printk(KERN_WARNING "ata%u: %u bytes trailing data\n", + printk(KERN_WARNING "ata%u: %u bytes trailing data\n", ap->id, bytes); for (i = 0; i < words; i++) @@ -2849,9 +2856,7 @@ static void ata_pio_block(struct ata_port *ap) if (is_atapi_taskfile(&qc->tf)) { /* no more data to transfer or unsupported ATAPI command */ if ((status & ATA_DRQ) == 0) { - ap->pio_task_state = PIO_ST_IDLE; - - ata_poll_qc_complete(qc, status); + ap->pio_task_state = PIO_ST_LAST; return; } @@ -2887,7 +2892,12 @@ static void ata_pio_error(struct ata_port *ap) static void ata_pio_task(void *_data) { struct ata_port *ap = _data; - unsigned long timeout = 0; + unsigned long timeout; + int qc_completed; + +fsm_start: + timeout = 0; + qc_completed = 0; switch (ap->pio_task_state) { case PIO_ST_IDLE: @@ -2898,7 +2908,7 @@ static void ata_pio_task(void *_data) break; case PIO_ST_LAST: - ata_pio_complete(ap); + qc_completed = ata_pio_complete(ap); break; case PIO_ST_POLL: @@ -2913,10 +2923,9 @@ static void ata_pio_task(void *_data) } if (timeout) - queue_delayed_work(ata_wq, &ap->pio_task, - timeout); - else - queue_work(ata_wq, &ap->pio_task); + queue_delayed_work(ata_wq, &ap->pio_task, timeout); + else if (!qc_completed) + goto fsm_start; } static void atapi_request_sense(struct ata_port *ap, struct ata_device *dev, From 1832a5862f2e1b4e5835611ee14bc30a8ed3cad5 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Fri, 16 Sep 2005 11:01:14 +0200 Subject: [PATCH 068/240] [SCSI] change port speed definitions for scsi_transport_fc obviously FC Port Speeds in scsi_transport_fc.h are defined according to FC-HBA: #define FC_PORTSPEED_1GBIT 1 #define FC_PORTSPEED_2GBIT 2 #define FC_PORTSPEED_10GBIT 4 #define FC_PORTSPEED_4GBIT 8 Problem is, whoever invented FC-HBA did not care about FC-FS or FC-GS-x. Following FC-FS/FC-GS-x defintions of port speeds would look like: 1 GBit: 0x0001 2 GBit: 0x0002 4 GBit: 0x0004 10GBit: 0x0008 (and new in FC-LS: 8 Gbit: 0x0010 16GBit: 0x0020) I really appreciate if scsi_transport_fc.h would define port speeds according to FC-GS-x/FC-FS. Thus mapping of port speed capabilities to values defined in scsi_transport_fc.h can be avoided in the LLDD. Attached is a patch to change the definitions. Signed-off-by: James Bottomley --- include/scsi/scsi_transport_fc.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h index 115db056dc6b..b0d445437372 100644 --- a/include/scsi/scsi_transport_fc.h +++ b/include/scsi/scsi_transport_fc.h @@ -103,8 +103,8 @@ enum fc_port_state { incapable of reporting */ #define FC_PORTSPEED_1GBIT 1 #define FC_PORTSPEED_2GBIT 2 -#define FC_PORTSPEED_10GBIT 4 -#define FC_PORTSPEED_4GBIT 8 +#define FC_PORTSPEED_4GBIT 4 +#define FC_PORTSPEED_10GBIT 8 #define FC_PORTSPEED_NOT_NEGOTIATED (1 << 15) /* Speed not established */ /* From 24b8e0cc09483adc0fdd9c68914b19597bb9fddc Mon Sep 17 00:00:00 2001 From: Peter Chubb Date: Thu, 15 Sep 2005 15:36:35 +1000 Subject: [PATCH 069/240] [IA64] Remove warnings for gcc 4.0 IA64 compilation. This patch removes some compilation warnings, mostly trivially. acpi.c fix also noted by Kenji Kaneshige. Signed-off-by; Peter Chubb Signed-off-by: Tony Luck --- arch/ia64/kernel/acpi.c | 2 +- arch/ia64/kernel/entry.S | 2 +- arch/ia64/kernel/perfmon.c | 2 +- drivers/char/agp/hp-agp.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index 28a4529fdd60..7e926471e4ec 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -899,7 +899,7 @@ int acpi_register_ioapic(acpi_handle handle, u64 phys_addr, u32 gsi_base) if ((err = iosapic_init(phys_addr, gsi_base))) return err; -#if CONFIG_ACPI_NUMA +#ifdef CONFIG_ACPI_NUMA acpi_map_iosapic(handle, 0, NULL, NULL); #endif /* CONFIG_ACPI_NUMA */ diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index ba0b6a1f429f..0741b066b98f 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -491,7 +491,7 @@ GLOBAL_ENTRY(prefetch_stack) ;; lfetch.fault [r16], 128 br.ret.sptk.many rp -END(prefetch_switch_stack) +END(prefetch_stack) GLOBAL_ENTRY(execve) mov r15=__NR_execve // put syscall number in place diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 1650353e3f77..af42cda6be80 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -574,7 +574,7 @@ pfm_protect_ctx_ctxsw(pfm_context_t *x) return 0UL; } -static inline unsigned long +static inline void pfm_unprotect_ctx_ctxsw(pfm_context_t *x, unsigned long f) { spin_unlock(&(x)->ctx_lock); diff --git a/drivers/char/agp/hp-agp.c b/drivers/char/agp/hp-agp.c index 99762b6c19ae..de5d6d212674 100644 --- a/drivers/char/agp/hp-agp.c +++ b/drivers/char/agp/hp-agp.c @@ -252,7 +252,7 @@ hp_zx1_configure (void) readl(hp->ioc_regs+HP_ZX1_PDIR_BASE); writel(hp->io_tlb_ps, hp->ioc_regs+HP_ZX1_TCNFG); readl(hp->ioc_regs+HP_ZX1_TCNFG); - writel(~(HP_ZX1_IOVA_SIZE-1), hp->ioc_regs+HP_ZX1_IMASK); + writel((unsigned int)(~(HP_ZX1_IOVA_SIZE-1)), hp->ioc_regs+HP_ZX1_IMASK); readl(hp->ioc_regs+HP_ZX1_IMASK); writel(hp->iova_base|1, hp->ioc_regs+HP_ZX1_IBASE); readl(hp->ioc_regs+HP_ZX1_IBASE); From 8ee9e23d41d2c51fafd158861fa4639fb199baf0 Mon Sep 17 00:00:00 2001 From: Keith Owens Date: Fri, 16 Sep 2005 14:49:14 +1000 Subject: [PATCH 070/240] [IA64] Add Documentation/ia64/mca.txt Add Documentation/ia64/mca.txt, an ad-hoc collection of notes on IA64 MCA and INIT processing. Signed-off-by: Keith Owens Signed-off-by: Tony Luck --- Documentation/ia64/mca.txt | 194 +++++++++++++++++++++++++++++++++++++ 1 file changed, 194 insertions(+) create mode 100644 Documentation/ia64/mca.txt diff --git a/Documentation/ia64/mca.txt b/Documentation/ia64/mca.txt new file mode 100644 index 000000000000..a71cc6a67ef7 --- /dev/null +++ b/Documentation/ia64/mca.txt @@ -0,0 +1,194 @@ +An ad-hoc collection of notes on IA64 MCA and INIT processing. Feel +free to update it with notes about any area that is not clear. + +--- + +MCA/INIT are completely asynchronous. They can occur at any time, when +the OS is in any state. Including when one of the cpus is already +holding a spinlock. Trying to get any lock from MCA/INIT state is +asking for deadlock. Also the state of structures that are protected +by locks is indeterminate, including linked lists. + +--- + +The complicated ia64 MCA process. All of this is mandated by Intel's +specification for ia64 SAL, error recovery and and unwind, it is not as +if we have a choice here. + +* MCA occurs on one cpu, usually due to a double bit memory error. + This is the monarch cpu. + +* SAL sends an MCA rendezvous interrupt (which is a normal interrupt) + to all the other cpus, the slaves. + +* Slave cpus that receive the MCA interrupt call down into SAL, they + end up spinning disabled while the MCA is being serviced. + +* If any slave cpu was already spinning disabled when the MCA occurred + then it cannot service the MCA interrupt. SAL waits ~20 seconds then + sends an unmaskable INIT event to the slave cpus that have not + already rendezvoused. + +* Because MCA/INIT can be delivered at any time, including when the cpu + is down in PAL in physical mode, the registers at the time of the + event are _completely_ undefined. In particular the MCA/INIT + handlers cannot rely on the thread pointer, PAL physical mode can + (and does) modify TP. It is allowed to do that as long as it resets + TP on return. However MCA/INIT events expose us to these PAL + internal TP changes. Hence curr_task(). + +* If an MCA/INIT event occurs while the kernel was running (not user + space) and the kernel has called PAL then the MCA/INIT handler cannot + assume that the kernel stack is in a fit state to be used. Mainly + because PAL may or may not maintain the stack pointer internally. + Because the MCA/INIT handlers cannot trust the kernel stack, they + have to use their own, per-cpu stacks. The MCA/INIT stacks are + preformatted with just enough task state to let the relevant handlers + do their job. + +* Unlike most other architectures, the ia64 struct task is embedded in + the kernel stack[1]. So switching to a new kernel stack means that + we switch to a new task as well. Because various bits of the kernel + assume that current points into the struct task, switching to a new + stack also means a new value for current. + +* Once all slaves have rendezvoused and are spinning disabled, the + monarch is entered. The monarch now tries to diagnose the problem + and decide if it can recover or not. + +* Part of the monarch's job is to look at the state of all the other + tasks. The only way to do that on ia64 is to call the unwinder, + as mandated by Intel. + +* The starting point for the unwind depends on whether a task is + running or not. That is, whether it is on a cpu or is blocked. The + monarch has to determine whether or not a task is on a cpu before it + knows how to start unwinding it. The tasks that received an MCA or + INIT event are no longer running, they have been converted to blocked + tasks. But (and its a big but), the cpus that received the MCA + rendezvous interrupt are still running on their normal kernel stacks! + +* To distinguish between these two cases, the monarch must know which + tasks are on a cpu and which are not. Hence each slave cpu that + switches to an MCA/INIT stack, registers its new stack using + set_curr_task(), so the monarch can tell that the _original_ task is + no longer running on that cpu. That gives us a decent chance of + getting a valid backtrace of the _original_ task. + +* MCA/INIT can be nested, to a depth of 2 on any cpu. In the case of a + nested error, we want diagnostics on the MCA/INIT handler that + failed, not on the task that was originally running. Again this + requires set_curr_task() so the MCA/INIT handlers can register their + own stack as running on that cpu. Then a recursive error gets a + trace of the failing handler's "task". + +[1] My (Keith Owens) original design called for ia64 to separate its + struct task and the kernel stacks. Then the MCA/INIT data would be + chained stacks like i386 interrupt stacks. But that required + radical surgery on the rest of ia64, plus extra hard wired TLB + entries with its associated performance degradation. David + Mosberger vetoed that approach. Which meant that separate kernel + stacks meant separate "tasks" for the MCA/INIT handlers. + +--- + +INIT is less complicated than MCA. Pressing the nmi button or using +the equivalent command on the management console sends INIT to all +cpus. SAL picks one one of the cpus as the monarch and the rest are +slaves. All the OS INIT handlers are entered at approximately the same +time. The OS monarch prints the state of all tasks and returns, after +which the slaves return and the system resumes. + +At least that is what is supposed to happen. Alas there are broken +versions of SAL out there. Some drive all the cpus as monarchs. Some +drive them all as slaves. Some drive one cpu as monarch, wait for that +cpu to return from the OS then drive the rest as slaves. Some versions +of SAL cannot even cope with returning from the OS, they spin inside +SAL on resume. The OS INIT code has workarounds for some of these +broken SAL symptoms, but some simply cannot be fixed from the OS side. + +--- + +The scheduler hooks used by ia64 (curr_task, set_curr_task) are layer +violations. Unfortunately MCA/INIT start off as massive layer +violations (can occur at _any_ time) and they build from there. + +At least ia64 makes an attempt at recovering from hardware errors, but +it is a difficult problem because of the asynchronous nature of these +errors. When processing an unmaskable interrupt we sometimes need +special code to cope with our inability to take any locks. + +--- + +How is ia64 MCA/INIT different from x86 NMI? + +* x86 NMI typically gets delivered to one cpu. MCA/INIT gets sent to + all cpus. + +* x86 NMI cannot be nested. MCA/INIT can be nested, to a depth of 2 + per cpu. + +* x86 has a separate struct task which points to one of multiple kernel + stacks. ia64 has the struct task embedded in the single kernel + stack, so switching stack means switching task. + +* x86 does not call the BIOS so the NMI handler does not have to worry + about any registers having changed. MCA/INIT can occur while the cpu + is in PAL in physical mode, with undefined registers and an undefined + kernel stack. + +* i386 backtrace is not very sensitive to whether a process is running + or not. ia64 unwind is very, very sensitive to whether a process is + running or not. + +--- + +What happens when MCA/INIT is delivered what a cpu is running user +space code? + +The user mode registers are stored in the RSE area of the MCA/INIT on +entry to the OS and are restored from there on return to SAL, so user +mode registers are preserved across a recoverable MCA/INIT. Since the +OS has no idea what unwind data is available for the user space stack, +MCA/INIT never tries to backtrace user space. Which means that the OS +does not bother making the user space process look like a blocked task, +i.e. the OS does not copy pt_regs and switch_stack to the user space +stack. Also the OS has no idea how big the user space RSE and memory +stacks are, which makes it too risky to copy the saved state to a user +mode stack. + +--- + +How do we get a backtrace on the tasks that were running when MCA/INIT +was delivered? + +mca.c:::ia64_mca_modify_original_stack(). That identifies and +verifies the original kernel stack, copies the dirty registers from +the MCA/INIT stack's RSE to the original stack's RSE, copies the +skeleton struct pt_regs and switch_stack to the original stack, fills +in the skeleton structures from the PAL minstate area and updates the +original stack's thread.ksp. That makes the original stack look +exactly like any other blocked task, i.e. it now appears to be +sleeping. To get a backtrace, just start with thread.ksp for the +original task and unwind like any other sleeping task. + +--- + +How do we identify the tasks that were running when MCA/INIT was +delivered? + +If the previous task has been verified and converted to a blocked +state, then sos->prev_task on the MCA/INIT stack is updated to point to +the previous task. You can look at that field in dumps or debuggers. +To help distinguish between the handler and the original tasks, +handlers have _TIF_MCA_INIT set in thread_info.flags. + +The sos data is always in the MCA/INIT handler stack, at offset +MCA_SOS_OFFSET. You can get that value from mca_asm.h or calculate it +as KERNEL_STACK_SIZE - sizeof(struct pt_regs) - sizeof(struct +ia64_sal_os_state), with 16 byte alignment for all structures. + +Also the comm field of the MCA/INIT task is modified to include the pid +of the original task, for humans to use. For example, a comm field of +'MCA 12159' means that pid 12159 was running when the MCA was +delivered. From bc9a5154a24d16d41cc79dbf1442de34454bc7db Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 15 Sep 2005 22:53:28 +0100 Subject: [PATCH 071/240] [PATCH] epca iomem annotations + several missing readw() [originally sent to Alan, he had no problems with it] - iomem pointers marked as such - several direct dereferencings of such pointers replaced with read[bw](). Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- drivers/char/epca.c | 84 ++++++++++++++++++++++++--------------------- drivers/char/epca.h | 12 +++---- 2 files changed, 50 insertions(+), 46 deletions(-) diff --git a/drivers/char/epca.c b/drivers/char/epca.c index 58d3738a2b7f..407708a001e4 100644 --- a/drivers/char/epca.c +++ b/drivers/char/epca.c @@ -534,7 +534,7 @@ static void shutdown(struct channel *ch) unsigned long flags; struct tty_struct *tty; - struct board_chan *bc; + struct board_chan __iomem *bc; if (!(ch->asyncflags & ASYNC_INITIALIZED)) return; @@ -618,7 +618,7 @@ static int pc_write(struct tty_struct * tty, struct channel *ch; unsigned long flags; int remain; - struct board_chan *bc; + struct board_chan __iomem *bc; /* ---------------------------------------------------------------- pc_write is primarily called directly by the kernel routine @@ -685,7 +685,7 @@ static int pc_write(struct tty_struct * tty, ------------------------------------------------------------------- */ dataLen = min(bytesAvailable, dataLen); - memcpy(ch->txptr + head, buf, dataLen); + memcpy_toio(ch->txptr + head, buf, dataLen); buf += dataLen; head += dataLen; amountCopied += dataLen; @@ -726,7 +726,7 @@ static int pc_write_room(struct tty_struct *tty) struct channel *ch; unsigned long flags; unsigned int head, tail; - struct board_chan *bc; + struct board_chan __iomem *bc; remain = 0; @@ -773,7 +773,7 @@ static int pc_chars_in_buffer(struct tty_struct *tty) int remain; unsigned long flags; struct channel *ch; - struct board_chan *bc; + struct board_chan __iomem *bc; /* --------------------------------------------------------- verifyChannel returns the channel from the tty struct @@ -830,7 +830,7 @@ static void pc_flush_buffer(struct tty_struct *tty) unsigned int tail; unsigned long flags; struct channel *ch; - struct board_chan *bc; + struct board_chan __iomem *bc; /* --------------------------------------------------------- verifyChannel returns the channel from the tty struct if it is valid. This serves as a sanity check. @@ -976,7 +976,7 @@ static int pc_open(struct tty_struct *tty, struct file * filp) struct channel *ch; unsigned long flags; int line, retval, boardnum; - struct board_chan *bc; + struct board_chan __iomem *bc; unsigned int head; line = tty->index; @@ -1041,7 +1041,7 @@ static int pc_open(struct tty_struct *tty, struct file * filp) ch->statusflags = 0; /* Save boards current modem status */ - ch->imodem = bc->mstat; + ch->imodem = readb(&bc->mstat); /* ---------------------------------------------------------------- Set receive head and tail ptrs to each other. This indicates @@ -1399,10 +1399,10 @@ static void post_fep_init(unsigned int crd) { /* Begin post_fep_init */ int i; - unsigned char *memaddr; - struct global_data *gd; + void __iomem *memaddr; + struct global_data __iomem *gd; struct board_info *bd; - struct board_chan *bc; + struct board_chan __iomem *bc; struct channel *ch; int shrinkmem = 0, lowwater ; @@ -1461,7 +1461,7 @@ static void post_fep_init(unsigned int crd) 8 and 64 of these structures. -------------------------------------------------------------------- */ - bc = (struct board_chan *)(memaddr + CHANSTRUCT); + bc = memaddr + CHANSTRUCT; /* ------------------------------------------------------------------- The below assignment will set gd to point at the BEGINING of @@ -1470,7 +1470,7 @@ static void post_fep_init(unsigned int crd) pointer begins at 0xd10. ---------------------------------------------------------------------- */ - gd = (struct global_data *)(memaddr + GLOBAL); + gd = memaddr + GLOBAL; /* -------------------------------------------------------------------- XEPORTS (address 0xc22) points at the number of channels the @@ -1493,6 +1493,7 @@ static void post_fep_init(unsigned int crd) for (i = 0; i < bd->numports; i++, ch++, bc++) { /* Begin for each port */ unsigned long flags; + u16 tseg, rseg; ch->brdchan = bc; ch->mailbox = gd; @@ -1553,50 +1554,53 @@ static void post_fep_init(unsigned int crd) shrinkmem = 0; } + tseg = readw(&bc->tseg); + rseg = readw(&bc->rseg); + switch (bd->type) { case PCIXEM: case PCIXRJ: case PCIXR: /* Cover all the 2MEG cards */ - ch->txptr = memaddr + (((bc->tseg) << 4) & 0x1fffff); - ch->rxptr = memaddr + (((bc->rseg) << 4) & 0x1fffff); - ch->txwin = FEPWIN | ((bc->tseg) >> 11); - ch->rxwin = FEPWIN | ((bc->rseg) >> 11); + ch->txptr = memaddr + ((tseg << 4) & 0x1fffff); + ch->rxptr = memaddr + ((rseg << 4) & 0x1fffff); + ch->txwin = FEPWIN | (tseg >> 11); + ch->rxwin = FEPWIN | (rseg >> 11); break; case PCXEM: case EISAXEM: /* Cover all the 32K windowed cards */ /* Mask equal to window size - 1 */ - ch->txptr = memaddr + (((bc->tseg) << 4) & 0x7fff); - ch->rxptr = memaddr + (((bc->rseg) << 4) & 0x7fff); - ch->txwin = FEPWIN | ((bc->tseg) >> 11); - ch->rxwin = FEPWIN | ((bc->rseg) >> 11); + ch->txptr = memaddr + ((tseg << 4) & 0x7fff); + ch->rxptr = memaddr + ((rseg << 4) & 0x7fff); + ch->txwin = FEPWIN | (tseg >> 11); + ch->rxwin = FEPWIN | (rseg >> 11); break; case PCXEVE: case PCXE: - ch->txptr = memaddr + (((bc->tseg - bd->memory_seg) << 4) & 0x1fff); - ch->txwin = FEPWIN | ((bc->tseg - bd->memory_seg) >> 9); - ch->rxptr = memaddr + (((bc->rseg - bd->memory_seg) << 4) & 0x1fff); - ch->rxwin = FEPWIN | ((bc->rseg - bd->memory_seg) >>9 ); + ch->txptr = memaddr + (((tseg - bd->memory_seg) << 4) & 0x1fff); + ch->txwin = FEPWIN | ((tseg - bd->memory_seg) >> 9); + ch->rxptr = memaddr + (((rseg - bd->memory_seg) << 4) & 0x1fff); + ch->rxwin = FEPWIN | ((rseg - bd->memory_seg) >>9 ); break; case PCXI: case PC64XE: - ch->txptr = memaddr + ((bc->tseg - bd->memory_seg) << 4); - ch->rxptr = memaddr + ((bc->rseg - bd->memory_seg) << 4); + ch->txptr = memaddr + ((tseg - bd->memory_seg) << 4); + ch->rxptr = memaddr + ((rseg - bd->memory_seg) << 4); ch->txwin = ch->rxwin = 0; break; } /* End switch bd->type */ ch->txbufhead = 0; - ch->txbufsize = bc->tmax + 1; + ch->txbufsize = readw(&bc->tmax) + 1; ch->rxbufhead = 0; - ch->rxbufsize = bc->rmax + 1; + ch->rxbufsize = readw(&bc->rmax) + 1; lowwater = ch->txbufsize >= 2000 ? 1024 : (ch->txbufsize / 2); @@ -1718,11 +1722,11 @@ static void epcapoll(unsigned long ignored) static void doevent(int crd) { /* Begin doevent */ - void *eventbuf; + void __iomem *eventbuf; struct channel *ch, *chan0; static struct tty_struct *tty; struct board_info *bd; - struct board_chan *bc; + struct board_chan __iomem *bc; unsigned int tail, head; int event, channel; int mstat, lstat; @@ -1817,7 +1821,7 @@ static void doevent(int crd) static void fepcmd(struct channel *ch, int cmd, int word_or_byte, int byte2, int ncmds, int bytecmd) { /* Begin fepcmd */ - unchar *memaddr; + unchar __iomem *memaddr; unsigned int head, cmdTail, cmdStart, cmdMax; long count; int n; @@ -2000,7 +2004,7 @@ static void epcaparam(struct tty_struct *tty, struct channel *ch) unsigned int cmdHead; struct termios *ts; - struct board_chan *bc; + struct board_chan __iomem *bc; unsigned mval, hflow, cflag, iflag; bc = ch->brdchan; @@ -2010,7 +2014,7 @@ static void epcaparam(struct tty_struct *tty, struct channel *ch) ts = tty->termios; if ((ts->c_cflag & CBAUD) == 0) { /* Begin CBAUD detected */ cmdHead = readw(&bc->rin); - bc->rout = cmdHead; + writew(cmdHead, &bc->rout); cmdHead = readw(&bc->tin); /* Changing baud in mid-stream transmission can be wonderful */ /* --------------------------------------------------------------- @@ -2116,7 +2120,7 @@ static void receive_data(struct channel *ch) unchar *rptr; struct termios *ts = NULL; struct tty_struct *tty; - struct board_chan *bc; + struct board_chan __iomem *bc; int dataToRead, wrapgap, bytesAvailable; unsigned int tail, head; unsigned int wrapmask; @@ -2154,7 +2158,7 @@ static void receive_data(struct channel *ch) --------------------------------------------------------------------- */ if (!tty || !ts || !(ts->c_cflag & CREAD)) { - bc->rout = head; + writew(head, &bc->rout); return; } @@ -2270,7 +2274,7 @@ static int info_ioctl(struct tty_struct *tty, struct file * file, static int pc_tiocmget(struct tty_struct *tty, struct file *file) { struct channel *ch = (struct channel *) tty->driver_data; - struct board_chan *bc; + struct board_chan __iomem *bc; unsigned int mstat, mflag = 0; unsigned long flags; @@ -2351,7 +2355,7 @@ static int pc_ioctl(struct tty_struct *tty, struct file * file, unsigned long flags; unsigned int mflag, mstat; unsigned char startc, stopc; - struct board_chan *bc; + struct board_chan __iomem *bc; struct channel *ch = (struct channel *) tty->driver_data; void __user *argp = (void __user *)arg; @@ -2633,7 +2637,7 @@ static void pc_start(struct tty_struct *tty) spin_lock_irqsave(&epca_lock, flags); /* Just in case output was resumed because of a change in Digi-flow */ if (ch->statusflags & TXSTOPPED) { /* Begin transmit resume requested */ - struct board_chan *bc; + struct board_chan __iomem *bc; globalwinon(ch); bc = ch->brdchan; if (ch->statusflags & LOWWAIT) @@ -2727,7 +2731,7 @@ void digi_send_break(struct channel *ch, int msec) static void setup_empty_event(struct tty_struct *tty, struct channel *ch) { /* Begin setup_empty_event */ - struct board_chan *bc = ch->brdchan; + struct board_chan __iomem *bc = ch->brdchan; globalwinon(ch); ch->statusflags |= EMPTYWAIT; diff --git a/drivers/char/epca.h b/drivers/char/epca.h index 20eeb5a70e1a..456d6c8f94a8 100644 --- a/drivers/char/epca.h +++ b/drivers/char/epca.h @@ -128,17 +128,17 @@ struct channel unsigned long c_cflag; unsigned long c_lflag; unsigned long c_oflag; - unsigned char *txptr; - unsigned char *rxptr; + unsigned char __iomem *txptr; + unsigned char __iomem *rxptr; unsigned char *tmp_buf; struct board_info *board; - struct board_chan *brdchan; + struct board_chan __iomem *brdchan; struct digi_struct digiext; struct tty_struct *tty; wait_queue_head_t open_wait; wait_queue_head_t close_wait; struct work_struct tqueue; - struct global_data *mailbox; + struct global_data __iomem *mailbox; }; struct board_info @@ -149,8 +149,8 @@ struct board_info unsigned short numports; unsigned long port; unsigned long membase; - unsigned char __iomem *re_map_port; - unsigned char *re_map_membase; + void __iomem *re_map_port; + void __iomem *re_map_membase; unsigned long memory_seg; void ( * memwinon ) (struct board_info *, unsigned int) ; void ( * memwinoff ) (struct board_info *, unsigned int) ; From 20305e59724381365096b6ab0a48054723367c80 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Fri, 16 Sep 2005 13:44:56 +0900 Subject: [PATCH 072/240] [IA64] mca_drv cleanup There were some trailing white spaces, long lines, brackets in weird style etc. This patch cleans them up. Signed-off-by: Hidetoshi Seto Signed-off-by: Tony Luck --- arch/ia64/kernel/mca_drv.c | 114 ++++++++++++++++++++------------- arch/ia64/kernel/mca_drv.h | 2 +- arch/ia64/kernel/mca_drv_asm.S | 48 +++++++------- 3 files changed, 94 insertions(+), 70 deletions(-) diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c index 6e683745af49..80f83d6cdbfc 100644 --- a/arch/ia64/kernel/mca_drv.c +++ b/arch/ia64/kernel/mca_drv.c @@ -84,23 +84,23 @@ mca_page_isolate(unsigned long paddr) struct page *p; /* whether physical address is valid or not */ - if ( !ia64_phys_addr_valid(paddr) ) + if (!ia64_phys_addr_valid(paddr)) return ISOLATE_NG; /* convert physical address to physical page number */ p = pfn_to_page(paddr>>PAGE_SHIFT); /* check whether a page number have been already registered or not */ - for( i = 0; i < num_page_isolate; i++ ) - if( page_isolate[i] == p ) + for (i = 0; i < num_page_isolate; i++) + if (page_isolate[i] == p) return ISOLATE_OK; /* already listed */ /* limitation check */ - if( num_page_isolate == MAX_PAGE_ISOLATE ) + if (num_page_isolate == MAX_PAGE_ISOLATE) return ISOLATE_NG; /* kick pages having attribute 'SLAB' or 'Reserved' */ - if( PageSlab(p) || PageReserved(p) ) + if (PageSlab(p) || PageReserved(p)) return ISOLATE_NG; /* add attribute 'Reserved' and register the page */ @@ -139,10 +139,10 @@ mca_handler_bh(unsigned long paddr) * @peidx: pointer to index of processor error section */ -static void +static void mca_make_peidx(sal_log_processor_info_t *slpi, peidx_table_t *peidx) { - /* + /* * calculate the start address of * "struct cpuid_info" and "sal_processor_static_info_t". */ @@ -164,7 +164,7 @@ mca_make_peidx(sal_log_processor_info_t *slpi, peidx_table_t *peidx) } /** - * mca_make_slidx - Make index of SAL error record + * mca_make_slidx - Make index of SAL error record * @buffer: pointer to SAL error record * @slidx: pointer to index of SAL error record * @@ -172,12 +172,12 @@ mca_make_peidx(sal_log_processor_info_t *slpi, peidx_table_t *peidx) * 1 if record has platform error / 0 if not */ #define LOG_INDEX_ADD_SECT_PTR(sect, ptr) \ - { slidx_list_t *hl = &slidx_pool.buffer[slidx_pool.cur_idx]; \ - hl->hdr = ptr; \ - list_add(&hl->list, &(sect)); \ - slidx_pool.cur_idx = (slidx_pool.cur_idx + 1)%slidx_pool.max_idx; } + {slidx_list_t *hl = &slidx_pool.buffer[slidx_pool.cur_idx]; \ + hl->hdr = ptr; \ + list_add(&hl->list, &(sect)); \ + slidx_pool.cur_idx = (slidx_pool.cur_idx + 1)%slidx_pool.max_idx; } -static int +static int mca_make_slidx(void *buffer, slidx_table_t *slidx) { int platform_err = 0; @@ -214,28 +214,36 @@ mca_make_slidx(void *buffer, slidx_table_t *slidx) sp = (sal_log_section_hdr_t *)((char*)buffer + ercd_pos); if (!efi_guidcmp(sp->guid, SAL_PROC_DEV_ERR_SECT_GUID)) { LOG_INDEX_ADD_SECT_PTR(slidx->proc_err, sp); - } else if (!efi_guidcmp(sp->guid, SAL_PLAT_MEM_DEV_ERR_SECT_GUID)) { + } else if (!efi_guidcmp(sp->guid, + SAL_PLAT_MEM_DEV_ERR_SECT_GUID)) { platform_err = 1; LOG_INDEX_ADD_SECT_PTR(slidx->mem_dev_err, sp); - } else if (!efi_guidcmp(sp->guid, SAL_PLAT_SEL_DEV_ERR_SECT_GUID)) { + } else if (!efi_guidcmp(sp->guid, + SAL_PLAT_SEL_DEV_ERR_SECT_GUID)) { platform_err = 1; LOG_INDEX_ADD_SECT_PTR(slidx->sel_dev_err, sp); - } else if (!efi_guidcmp(sp->guid, SAL_PLAT_PCI_BUS_ERR_SECT_GUID)) { + } else if (!efi_guidcmp(sp->guid, + SAL_PLAT_PCI_BUS_ERR_SECT_GUID)) { platform_err = 1; LOG_INDEX_ADD_SECT_PTR(slidx->pci_bus_err, sp); - } else if (!efi_guidcmp(sp->guid, SAL_PLAT_SMBIOS_DEV_ERR_SECT_GUID)) { + } else if (!efi_guidcmp(sp->guid, + SAL_PLAT_SMBIOS_DEV_ERR_SECT_GUID)) { platform_err = 1; LOG_INDEX_ADD_SECT_PTR(slidx->smbios_dev_err, sp); - } else if (!efi_guidcmp(sp->guid, SAL_PLAT_PCI_COMP_ERR_SECT_GUID)) { + } else if (!efi_guidcmp(sp->guid, + SAL_PLAT_PCI_COMP_ERR_SECT_GUID)) { platform_err = 1; LOG_INDEX_ADD_SECT_PTR(slidx->pci_comp_err, sp); - } else if (!efi_guidcmp(sp->guid, SAL_PLAT_SPECIFIC_ERR_SECT_GUID)) { + } else if (!efi_guidcmp(sp->guid, + SAL_PLAT_SPECIFIC_ERR_SECT_GUID)) { platform_err = 1; LOG_INDEX_ADD_SECT_PTR(slidx->plat_specific_err, sp); - } else if (!efi_guidcmp(sp->guid, SAL_PLAT_HOST_CTLR_ERR_SECT_GUID)) { + } else if (!efi_guidcmp(sp->guid, + SAL_PLAT_HOST_CTLR_ERR_SECT_GUID)) { platform_err = 1; LOG_INDEX_ADD_SECT_PTR(slidx->host_ctlr_err, sp); - } else if (!efi_guidcmp(sp->guid, SAL_PLAT_BUS_ERR_SECT_GUID)) { + } else if (!efi_guidcmp(sp->guid, + SAL_PLAT_BUS_ERR_SECT_GUID)) { platform_err = 1; LOG_INDEX_ADD_SECT_PTR(slidx->plat_bus_err, sp); } else { @@ -253,15 +261,16 @@ mca_make_slidx(void *buffer, slidx_table_t *slidx) * Return value: * 0 on Success / -ENOMEM on Failure */ -static int +static int init_record_index_pools(void) { int i; int rec_max_size; /* Maximum size of SAL error records */ int sect_min_size; /* Minimum size of SAL error sections */ /* minimum size table of each section */ - static int sal_log_sect_min_sizes[] = { - sizeof(sal_log_processor_info_t) + sizeof(sal_processor_static_info_t), + static int sal_log_sect_min_sizes[] = { + sizeof(sal_log_processor_info_t) + + sizeof(sal_processor_static_info_t), sizeof(sal_log_mem_dev_err_info_t), sizeof(sal_log_sel_dev_err_info_t), sizeof(sal_log_pci_bus_err_info_t), @@ -294,7 +303,8 @@ init_record_index_pools(void) /* - 3 - */ slidx_pool.max_idx = (rec_max_size/sect_min_size) * 2 + 1; - slidx_pool.buffer = (slidx_list_t *) kmalloc(slidx_pool.max_idx * sizeof(slidx_list_t), GFP_KERNEL); + slidx_pool.buffer = (slidx_list_t *) + kmalloc(slidx_pool.max_idx * sizeof(slidx_list_t), GFP_KERNEL); return slidx_pool.buffer ? 0 : -ENOMEM; } @@ -308,6 +318,7 @@ init_record_index_pools(void) * is_mca_global - Check whether this MCA is global or not * @peidx: pointer of index of processor error section * @pbci: pointer to pal_bus_check_info_t + * @sos: pointer to hand off struct between SAL and OS * * Return value: * MCA_IS_LOCAL / MCA_IS_GLOBAL @@ -317,11 +328,12 @@ static mca_type_t is_mca_global(peidx_table_t *peidx, pal_bus_check_info_t *pbci, struct ia64_sal_os_state *sos) { - pal_processor_state_info_t *psp = (pal_processor_state_info_t*)peidx_psp(peidx); + pal_processor_state_info_t *psp = + (pal_processor_state_info_t*)peidx_psp(peidx); - /* + /* * PAL can request a rendezvous, if the MCA has a global scope. - * If "rz_always" flag is set, SAL requests MCA rendezvous + * If "rz_always" flag is set, SAL requests MCA rendezvous * in spite of global MCA. * Therefore it is local MCA when rendezvous has not been requested. * Failed to rendezvous, the system must be down. @@ -381,13 +393,15 @@ is_mca_global(peidx_table_t *peidx, pal_bus_check_info_t *pbci, * @slidx: pointer of index of SAL error record * @peidx: pointer of index of processor error section * @pbci: pointer of pal_bus_check_info + * @sos: pointer to hand off struct between SAL and OS * * Return value: * 1 on Success / 0 on Failure */ static int -recover_from_read_error(slidx_table_t *slidx, peidx_table_t *peidx, pal_bus_check_info_t *pbci, +recover_from_read_error(slidx_table_t *slidx, + peidx_table_t *peidx, pal_bus_check_info_t *pbci, struct ia64_sal_os_state *sos) { sal_log_mod_error_info_t *smei; @@ -453,24 +467,28 @@ recover_from_read_error(slidx_table_t *slidx, peidx_table_t *peidx, pal_bus_chec * @slidx: pointer of index of SAL error record * @peidx: pointer of index of processor error section * @pbci: pointer of pal_bus_check_info + * @sos: pointer to hand off struct between SAL and OS * * Return value: * 1 on Success / 0 on Failure */ static int -recover_from_platform_error(slidx_table_t *slidx, peidx_table_t *peidx, pal_bus_check_info_t *pbci, +recover_from_platform_error(slidx_table_t *slidx, peidx_table_t *peidx, + pal_bus_check_info_t *pbci, struct ia64_sal_os_state *sos) { int status = 0; - pal_processor_state_info_t *psp = (pal_processor_state_info_t*)peidx_psp(peidx); + pal_processor_state_info_t *psp = + (pal_processor_state_info_t*)peidx_psp(peidx); if (psp->bc && pbci->eb && pbci->bsi == 0) { switch(pbci->type) { case 1: /* partial read */ case 3: /* full line(cpu) read */ case 9: /* I/O space read */ - status = recover_from_read_error(slidx, peidx, pbci, sos); + status = recover_from_read_error(slidx, peidx, pbci, + sos); break; case 0: /* unknown */ case 2: /* partial write */ @@ -481,7 +499,8 @@ recover_from_platform_error(slidx_table_t *slidx, peidx_table_t *peidx, pal_bus_ case 8: /* write coalescing transactions */ case 10: /* I/O space write */ case 11: /* inter-processor interrupt message(IPI) */ - case 12: /* interrupt acknowledge or external task priority cycle */ + case 12: /* interrupt acknowledge or + external task priority cycle */ default: break; } @@ -496,6 +515,7 @@ recover_from_platform_error(slidx_table_t *slidx, peidx_table_t *peidx, pal_bus_ * @slidx: pointer of index of SAL error record * @peidx: pointer of index of processor error section * @pbci: pointer of pal_bus_check_info + * @sos: pointer to hand off struct between SAL and OS * * Return value: * 1 on Success / 0 on Failure @@ -509,15 +529,17 @@ recover_from_platform_error(slidx_table_t *slidx, peidx_table_t *peidx, pal_bus_ */ static int -recover_from_processor_error(int platform, slidx_table_t *slidx, peidx_table_t *peidx, pal_bus_check_info_t *pbci, +recover_from_processor_error(int platform, slidx_table_t *slidx, + peidx_table_t *peidx, pal_bus_check_info_t *pbci, struct ia64_sal_os_state *sos) { - pal_processor_state_info_t *psp = (pal_processor_state_info_t*)peidx_psp(peidx); + pal_processor_state_info_t *psp = + (pal_processor_state_info_t*)peidx_psp(peidx); - /* + /* * We cannot recover errors with other than bus_check. */ - if (psp->cc || psp->rc || psp->uc) + if (psp->cc || psp->rc || psp->uc) return 0; /* @@ -546,10 +568,10 @@ recover_from_processor_error(int platform, slidx_table_t *slidx, peidx_table_t * * (e.g. a load from poisoned memory) * This means "there are some platform errors". */ - if (platform) + if (platform) return recover_from_platform_error(slidx, peidx, pbci, sos); - /* - * On account of strange SAL error record, we cannot recover. + /* + * On account of strange SAL error record, we cannot recover. */ return 0; } @@ -557,14 +579,14 @@ recover_from_processor_error(int platform, slidx_table_t *slidx, peidx_table_t * /** * mca_try_to_recover - Try to recover from MCA * @rec: pointer to a SAL error record + * @sos: pointer to hand off struct between SAL and OS * * Return value: * 1 on Success / 0 on Failure */ static int -mca_try_to_recover(void *rec, - struct ia64_sal_os_state *sos) +mca_try_to_recover(void *rec, struct ia64_sal_os_state *sos) { int platform_err; int n_proc_err; @@ -588,7 +610,8 @@ mca_try_to_recover(void *rec, } /* Make index of processor error section */ - mca_make_peidx((sal_log_processor_info_t*)slidx_first_entry(&slidx.proc_err)->hdr, &peidx); + mca_make_peidx((sal_log_processor_info_t*) + slidx_first_entry(&slidx.proc_err)->hdr, &peidx); /* Extract Processor BUS_CHECK[0] */ *((u64*)&pbci) = peidx_check_info(&peidx, bus_check, 0); @@ -598,7 +621,8 @@ mca_try_to_recover(void *rec, return 0; /* Try to recover a processor error */ - return recover_from_processor_error(platform_err, &slidx, &peidx, &pbci, sos); + return recover_from_processor_error(platform_err, &slidx, &peidx, + &pbci, sos); } /* @@ -611,7 +635,7 @@ int __init mca_external_handler_init(void) return -ENOMEM; /* register external mca handlers */ - if (ia64_reg_MCA_extension(mca_try_to_recover)){ + if (ia64_reg_MCA_extension(mca_try_to_recover)) { printk(KERN_ERR "ia64_reg_MCA_extension failed.\n"); kfree(slidx_pool.buffer); return -EFAULT; diff --git a/arch/ia64/kernel/mca_drv.h b/arch/ia64/kernel/mca_drv.h index 0227b761f2c4..e2f6fa1e0ef6 100644 --- a/arch/ia64/kernel/mca_drv.h +++ b/arch/ia64/kernel/mca_drv.h @@ -6,7 +6,7 @@ * Copyright (C) Hidetoshi Seto (seto.hidetoshi@jp.fujitsu.com) */ /* - * Processor error section: + * Processor error section: * * +-sal_log_processor_info_t *info-------------+ * | sal_log_section_hdr_t header; | diff --git a/arch/ia64/kernel/mca_drv_asm.S b/arch/ia64/kernel/mca_drv_asm.S index 2d7e0217638d..3f298ee4d00c 100644 --- a/arch/ia64/kernel/mca_drv_asm.S +++ b/arch/ia64/kernel/mca_drv_asm.S @@ -13,45 +13,45 @@ #include GLOBAL_ENTRY(mca_handler_bhhook) - invala // clear RSE ? - ;; // - cover // - ;; // - clrrrb // + invala // clear RSE ? + ;; + cover + ;; + clrrrb ;; - alloc r16=ar.pfs,0,2,1,0 // make a new frame + alloc r16=ar.pfs,0,2,1,0 // make a new frame ;; - mov ar.rsc=0 + mov ar.rsc=0 ;; - mov r13=IA64_KR(CURRENT) // current task pointer + mov r13=IA64_KR(CURRENT) // current task pointer ;; - mov r2=r13 + mov r2=r13 ;; - addl r22=IA64_RBS_OFFSET,r2 + addl r22=IA64_RBS_OFFSET,r2 ;; - mov ar.bspstore=r22 + mov ar.bspstore=r22 ;; - addl sp=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2 + addl sp=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2 ;; - adds r2=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13 + adds r2=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13 ;; - st1 [r2]=r0 // clear current->thread.on_ustack flag - mov loc0=r16 - movl loc1=mca_handler_bh // recovery C function + st1 [r2]=r0 // clear current->thread.on_ustack flag + mov loc0=r16 + movl loc1=mca_handler_bh // recovery C function ;; - mov out0=r8 // poisoned address - mov b6=loc1 + mov out0=r8 // poisoned address + mov b6=loc1 ;; - mov loc1=rp + mov loc1=rp ;; - ssm psr.i + ssm psr.i ;; - br.call.sptk.many rp=b6 // does not return ... + br.call.sptk.many rp=b6 // does not return ... ;; - mov ar.pfs=loc0 - mov rp=loc1 + mov ar.pfs=loc0 + mov rp=loc1 ;; - mov r8=r0 + mov r8=r0 br.ret.sptk.many rp ;; END(mca_handler_bhhook) From a063cf5b7dde94d98f3f7c9f1c951e02c6564022 Mon Sep 17 00:00:00 2001 From: Karsten Keil Date: Fri, 16 Sep 2005 19:32:53 +0200 Subject: [PATCH 073/240] [PATCH] Add PCI IDs for Sitecom DC-105 Sitecom DC-105 PCI work with hfc_pci HiSax driver Signed-off-by: Karsten Keil Signed-off-by: Linus Torvalds --- drivers/isdn/hisax/hfc_pci.c | 1 + include/linux/pci_ids.h | 3 +++ 2 files changed, 4 insertions(+) diff --git a/drivers/isdn/hisax/hfc_pci.c b/drivers/isdn/hisax/hfc_pci.c index 8337b0f26cc4..4866fc32d8d9 100644 --- a/drivers/isdn/hisax/hfc_pci.c +++ b/drivers/isdn/hisax/hfc_pci.c @@ -61,6 +61,7 @@ static const PCI_ENTRY id_list[] = {PCI_VENDOR_ID_DIGI, PCI_DEVICE_ID_DIGI_DF_M_E,"Digi International", "Digi DataFire Micro V (Europe)"}, {PCI_VENDOR_ID_DIGI, PCI_DEVICE_ID_DIGI_DF_M_IOM2_A,"Digi International", "Digi DataFire Micro V IOM2 (North America)"}, {PCI_VENDOR_ID_DIGI, PCI_DEVICE_ID_DIGI_DF_M_A,"Digi International", "Digi DataFire Micro V (North America)"}, + {PCI_VENDOR_ID_SITECOM, PCI_DEVICE_ID_SITECOM_DC105V2, "Sitecom Europe", "DC-105 ISDN PCI"}, {0, 0, NULL, NULL}, }; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 72fe3385743c..b6ef7eb4bcbf 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2252,6 +2252,9 @@ #define PCI_VENDOR_ID_INFINICON 0x1820 +#define PCI_VENDOR_ID_SITECOM 0x182d +#define PCI_DEVICE_ID_SITECOM_DC105V2 0x3069 + #define PCI_VENDOR_ID_TOPSPIN 0x1867 #define PCI_VENDOR_ID_TDI 0x192E From 06168d8a10ceccced51380d683245b33474d428a Mon Sep 17 00:00:00 2001 From: Karsten Keil Date: Fri, 16 Sep 2005 19:34:17 +0200 Subject: [PATCH 074/240] [PATCH] cleanup whitespace in pci_ids.h Signed-off-by: Karsten Keil Signed-off-by: Linus Torvalds --- include/linux/pci_ids.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index b6ef7eb4bcbf..486d1c1676bd 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2254,7 +2254,7 @@ #define PCI_VENDOR_ID_SITECOM 0x182d #define PCI_DEVICE_ID_SITECOM_DC105V2 0x3069 - + #define PCI_VENDOR_ID_TOPSPIN 0x1867 #define PCI_VENDOR_ID_TDI 0x192E From 37f7f421cce13435fdc0d870caf51141e5ebf079 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 16 Sep 2005 16:51:01 -0700 Subject: [PATCH 075/240] [NET]: Do not leak MSG_CMSG_COMPAT into userspace. Noticed by Sridhar Samudrala. Signed-off-by: David S. Miller --- net/socket.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/socket.c b/net/socket.c index c699e93c33d7..f9264472377f 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1862,7 +1862,8 @@ asmlinkage long sys_recvmsg(int fd, struct msghdr __user *msg, unsigned int flag if (err < 0) goto out_freeiov; } - err = __put_user(msg_sys.msg_flags, COMPAT_FLAGS(msg)); + err = __put_user((msg_sys.msg_flags & ~MSG_CMSG_COMPAT), + COMPAT_FLAGS(msg)); if (err) goto out_freeiov; if (MSG_CMSG_COMPAT & flags) From 0c10c5d96865ce611d6a780888eff0ef4fab358b Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Fri, 16 Sep 2005 16:58:33 -0700 Subject: [PATCH 076/240] [DCCP]: More precisely set reset_code when sending RESET packets Moving the setting of DCCP_SKB_CB(skb)->dccpd_reset_code to the places where events happen that trigger sending a RESET packet. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- net/dccp/input.c | 22 +++++++++++----------- net/dccp/ipv4.c | 10 +++++++--- 2 files changed, 18 insertions(+), 14 deletions(-) diff --git a/net/dccp/input.c b/net/dccp/input.c index c74034cf7ede..062e9f8359d0 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -384,9 +384,9 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, } out_invalid_packet: - return 1; /* dccp_v4_do_rcv will send a reset, but... - FIXME: the reset code should be - DCCP_RESET_CODE_PACKET_ERROR */ + /* dccp_v4_do_rcv will send a reset */ + DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_PACKET_ERROR; + return 1; } static int dccp_rcv_respond_partopen_state_process(struct sock *sk, @@ -433,6 +433,7 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, struct dccp_hdr *dh, unsigned len) { struct dccp_sock *dp = dccp_sk(sk); + struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); const int old_state = sk->sk_state; int queued = 0; @@ -473,7 +474,8 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, if (dh->dccph_type == DCCP_PKT_RESET) goto discard; - /* Caller (dccp_v4_do_rcv) will send Reset(No Connection)*/ + /* Caller (dccp_v4_do_rcv) will send Reset */ + dcb->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION; return 1; } @@ -487,8 +489,7 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, if (dccp_parse_options(sk, skb)) goto discard; - if (DCCP_SKB_CB(skb)->dccpd_ack_seq != - DCCP_PKT_WITHOUT_ACK_SEQ) + if (dcb->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ) dccp_event_ack_recv(sk, skb); ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb); @@ -500,7 +501,7 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, */ if (dp->dccps_options.dccpo_send_ack_vector) { if (dccp_ackpkts_add(dp->dccps_hc_rx_ackpkts, sk, - DCCP_SKB_CB(skb)->dccpd_seq, + dcb->dccpd_seq, DCCP_ACKPKTS_STATE_RECEIVED)) goto discard; /* @@ -551,8 +552,7 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, dh->dccph_type == DCCP_PKT_REQUEST) || (sk->sk_state == DCCP_RESPOND && dh->dccph_type == DCCP_PKT_DATA)) { - dccp_send_sync(sk, DCCP_SKB_CB(skb)->dccpd_seq, - DCCP_PKT_SYNC); + dccp_send_sync(sk, dcb->dccpd_seq, DCCP_PKT_SYNC); goto discard; } else if (dh->dccph_type == DCCP_PKT_CLOSEREQ) { dccp_rcv_closereq(sk, skb); @@ -563,13 +563,13 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, } if (unlikely(dh->dccph_type == DCCP_PKT_SYNC)) { - dccp_send_sync(sk, DCCP_SKB_CB(skb)->dccpd_seq, - DCCP_PKT_SYNCACK); + dccp_send_sync(sk, dcb->dccpd_seq, DCCP_PKT_SYNCACK); goto discard; } switch (sk->sk_state) { case DCCP_CLOSED: + dcb->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION; return 1; case DCCP_REQUESTING: diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 2afaa464e7f0..e09907d8b7da 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -669,12 +669,16 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) struct dccp_request_sock *dreq; const __u32 saddr = skb->nh.iph->saddr; const __u32 daddr = skb->nh.iph->daddr; + struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); + __u8 reset_code = DCCP_RESET_CODE_TOO_BUSY; struct dst_entry *dst = NULL; /* Never answer to DCCP_PKT_REQUESTs send to broadcast or multicast */ if (((struct rtable *)skb->dst)->rt_flags & - (RTCF_BROADCAST | RTCF_MULTICAST)) + (RTCF_BROADCAST | RTCF_MULTICAST)) { + reset_code = DCCP_RESET_CODE_NO_CONNECTION; goto drop; + } /* * TW buckets are converted to open requests without @@ -718,7 +722,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) * dccp_create_openreq_child. */ dreq = dccp_rsk(req); - dreq->dreq_isr = DCCP_SKB_CB(skb)->dccpd_seq; + dreq->dreq_isr = dcb->dccpd_seq; dreq->dreq_iss = dccp_v4_init_sequence(sk, skb); dreq->dreq_service = dccp_hdr_request(skb)->dccph_req_service; @@ -735,6 +739,7 @@ drop_and_free: __reqsk_free(req); drop: DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS); + dcb->dccpd_reset_code = reset_code; return -1; } @@ -1005,7 +1010,6 @@ int dccp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION; dccp_v4_ctl_send_reset(skb); discard: kfree_skb(skb); From 67e6b629212fa9ffb7420e8a88a41806af637e28 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Fri, 16 Sep 2005 16:58:40 -0700 Subject: [PATCH 077/240] [DCCP]: Introduce DCCP_SOCKOPT_SERVICE As discussed in the dccp@vger mailing list: Now applications have to use setsockopt(DCCP_SOCKOPT_SERVICE, service[s]), prior to calling listen() and connect(). An array of unsigned ints can be passed meaning that the listening sock accepts connection requests for several services. With this we can ditch struct sockaddr_dccp and use only sockaddr_in (and sockaddr_in6 in the future). Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/dccp.h | 40 +++++++++++++++------ net/dccp/dccp.h | 9 +++-- net/dccp/ipv4.c | 30 ++++++++++++++-- net/dccp/minisocks.c | 6 ++-- net/dccp/output.c | 14 ++++---- net/dccp/proto.c | 85 ++++++++++++++++++++++++++++++++++++++++++-- 6 files changed, 154 insertions(+), 30 deletions(-) diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 8bf4bacb5051..0e72708677e4 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -4,16 +4,6 @@ #include #include -/* Structure describing an Internet (DCCP) socket address. */ -struct sockaddr_dccp { - __u16 sdccp_family; /* Address family */ - __u16 sdccp_port; /* Port number */ - __u32 sdccp_addr; /* Internet address */ - __u32 sdccp_service; /* Service */ - /* Pad to size of `struct sockaddr': 16 bytes . */ - __u32 sdccp_pad; -}; - /** * struct dccp_hdr - generic part of DCCP packet header * @@ -188,6 +178,9 @@ enum { /* DCCP socket options */ #define DCCP_SOCKOPT_PACKET_SIZE 1 +#define DCCP_SOCKOPT_SERVICE 2 + +#define DCCP_SERVICE_LIST_MAX_LEN 32 #ifdef __KERNEL__ @@ -382,6 +375,25 @@ enum dccp_role { DCCP_ROLE_SERVER, }; +struct dccp_service_list { + __u32 dccpsl_nr; + __u32 dccpsl_list[0]; +}; + +#define DCCP_SERVICE_INVALID_VALUE htonl((__u32)-1) + +static inline int dccp_list_has_service(const struct dccp_service_list *sl, + const u32 service) +{ + if (likely(sl != NULL)) { + u32 i = sl->dccpsl_nr; + while (i--) + if (sl->dccpsl_list[i] == service) + return 1; + } + return 0; +} + /** * struct dccp_sock - DCCP socket state * @@ -417,7 +429,8 @@ struct dccp_sock { __u64 dccps_gss; __u64 dccps_gsr; __u64 dccps_gar; - unsigned long dccps_service; + __u32 dccps_service; + struct dccp_service_list *dccps_service_list; struct timeval dccps_timestamp_time; __u32 dccps_timestamp_echo; __u32 dccps_packet_size; @@ -443,6 +456,11 @@ static inline struct dccp_sock *dccp_sk(const struct sock *sk) return (struct dccp_sock *)sk; } +static inline int dccp_service_not_initialized(const struct sock *sk) +{ + return dccp_sk(sk)->dccps_service == DCCP_SERVICE_INVALID_VALUE; +} + static inline const char *dccp_role(const struct sock *sk) { switch (dccp_sk(sk)->dccps_role) { diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index 95c4630b3b18..be7a660b6b24 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -258,13 +258,12 @@ extern int dccp_v4_send_reset(struct sock *sk, extern void dccp_send_close(struct sock *sk, const int active); struct dccp_skb_cb { - __u8 dccpd_type; - __u8 dccpd_reset_code; - __u8 dccpd_service; - __u8 dccpd_ccval; + __u8 dccpd_type:4; + __u8 dccpd_ccval:4; + __u8 dccpd_reset_code; + __u16 dccpd_opt_len; __u64 dccpd_seq; __u64 dccpd_ack_seq; - int dccpd_opt_len; }; #define DCCP_SKB_CB(__skb) ((struct dccp_skb_cb *)&((__skb)->cb[0])) diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index e09907d8b7da..94a440b2685b 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -246,6 +246,9 @@ static int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, dp->dccps_role = DCCP_ROLE_CLIENT; + if (dccp_service_not_initialized(sk)) + return -EPROTO; + if (addr_len < sizeof(struct sockaddr_in)) return -EINVAL; @@ -661,6 +664,16 @@ static inline u64 dccp_v4_init_sequence(const struct sock *sk, dccp_hdr(skb)->dccph_sport); } +static inline int dccp_bad_service_code(const struct sock *sk, + const __u32 service) +{ + const struct dccp_sock *dp = dccp_sk(sk); + + if (dp->dccps_service == service) + return 0; + return !dccp_list_has_service(dp->dccps_service_list, service); +} + int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) { struct inet_request_sock *ireq; @@ -669,6 +682,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) struct dccp_request_sock *dreq; const __u32 saddr = skb->nh.iph->saddr; const __u32 daddr = skb->nh.iph->daddr; + const __u32 service = dccp_hdr_request(skb)->dccph_req_service; struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); __u8 reset_code = DCCP_RESET_CODE_TOO_BUSY; struct dst_entry *dst = NULL; @@ -680,6 +694,10 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) goto drop; } + if (dccp_bad_service_code(sk, service)) { + reset_code = DCCP_RESET_CODE_BAD_SERVICE_CODE; + goto drop; + } /* * TW buckets are converted to open requests without * limitations, they conserve resources and peer is @@ -722,9 +740,9 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) * dccp_create_openreq_child. */ dreq = dccp_rsk(req); - dreq->dreq_isr = dcb->dccpd_seq; - dreq->dreq_iss = dccp_v4_init_sequence(sk, skb); - dreq->dreq_service = dccp_hdr_request(skb)->dccph_req_service; + dreq->dreq_isr = dcb->dccpd_seq; + dreq->dreq_iss = dccp_v4_init_sequence(sk, skb); + dreq->dreq_service = service; if (dccp_v4_send_response(sk, req, dst)) goto drop_and_free; @@ -1284,6 +1302,7 @@ static int dccp_v4_init_sock(struct sock *sk) sk->sk_write_space = dccp_write_space; dp->dccps_mss_cache = 536; dp->dccps_role = DCCP_ROLE_UNDEFINED; + dp->dccps_service = DCCP_SERVICE_INVALID_VALUE; return 0; } @@ -1305,6 +1324,11 @@ static int dccp_v4_destroy_sock(struct sock *sk) if (inet_csk(sk)->icsk_bind_hash != NULL) inet_put_port(&dccp_hashinfo, sk); + if (dp->dccps_service_list != NULL) { + kfree(dp->dccps_service_list); + dp->dccps_service_list = NULL; + } + ccid_hc_rx_exit(dp->dccps_hc_rx_ccid, sk); ccid_hc_tx_exit(dp->dccps_hc_tx_ccid, sk); dccp_ackpkts_free(dp->dccps_hc_rx_ackpkts); diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 18461bc04cbe..933e10db1789 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -93,9 +93,11 @@ struct sock *dccp_create_openreq_child(struct sock *sk, struct inet_connection_sock *newicsk = inet_csk(sk); struct dccp_sock *newdp = dccp_sk(newsk); + newdp->dccps_role = DCCP_ROLE_SERVER; newdp->dccps_hc_rx_ackpkts = NULL; - newdp->dccps_role = DCCP_ROLE_SERVER; - newicsk->icsk_rto = DCCP_TIMEOUT_INIT; + newdp->dccps_service_list = NULL; + newdp->dccps_service = dreq->dreq_service; + newicsk->icsk_rto = DCCP_TIMEOUT_INIT; do_gettimeofday(&newdp->dccps_epoch); if (newdp->dccps_options.dccpo_send_ack_vector) { diff --git a/net/dccp/output.c b/net/dccp/output.c index ea6d0e91e511..156b1d29a156 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -85,7 +85,7 @@ int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) switch (dcb->dccpd_type) { case DCCP_PKT_REQUEST: dccp_hdr_request(skb)->dccph_req_service = - dcb->dccpd_service; + dp->dccps_service; break; case DCCP_PKT_RESET: dccp_hdr_reset(skb)->dccph_reset_code = @@ -270,6 +270,7 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, struct request_sock *req) { struct dccp_hdr *dh; + struct dccp_request_sock *dreq; const int dccp_header_size = sizeof(struct dccp_hdr) + sizeof(struct dccp_hdr_ext) + sizeof(struct dccp_hdr_response); @@ -285,8 +286,9 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, skb->dst = dst_clone(dst); skb->csum = 0; + dreq = dccp_rsk(req); DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_RESPONSE; - DCCP_SKB_CB(skb)->dccpd_seq = dccp_rsk(req)->dreq_iss; + DCCP_SKB_CB(skb)->dccpd_seq = dreq->dreq_iss; dccp_insert_options(sk, skb); skb->h.raw = skb_push(skb, dccp_header_size); @@ -300,8 +302,9 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, DCCP_SKB_CB(skb)->dccpd_opt_len) / 4; dh->dccph_type = DCCP_PKT_RESPONSE; dh->dccph_x = 1; - dccp_hdr_set_seq(dh, dccp_rsk(req)->dreq_iss); - dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), dccp_rsk(req)->dreq_isr); + dccp_hdr_set_seq(dh, dreq->dreq_iss); + dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), dreq->dreq_isr); + dccp_hdr_response(skb)->dccph_resp_service = dreq->dreq_service; dh->dccph_checksum = dccp_v4_checksum(skb, inet_rsk(req)->loc_addr, inet_rsk(req)->rmt_addr); @@ -397,9 +400,6 @@ int dccp_connect(struct sock *sk) skb_reserve(skb, MAX_DCCP_HEADER); DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_REQUEST; - /* FIXME: set service to something meaningful, coming - * from userspace*/ - DCCP_SKB_CB(skb)->dccpd_service = 0; skb->csum = 0; skb_set_owner_w(skb, sk); diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 18a0e69c9dc7..9bda2868eba6 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -94,7 +94,15 @@ EXPORT_SYMBOL_GPL(dccp_state_name); static inline int dccp_listen_start(struct sock *sk) { - dccp_sk(sk)->dccps_role = DCCP_ROLE_LISTEN; + struct dccp_sock *dp = dccp_sk(sk); + + dp->dccps_role = DCCP_ROLE_LISTEN; + /* + * Apps need to use setsockopt(DCCP_SOCKOPT_SERVICE) + * before calling listen() + */ + if (dccp_service_not_initialized(sk)) + return -EPROTO; return inet_csk_listen_start(sk, TCP_SYNQ_HSIZE); } @@ -202,6 +210,42 @@ int dccp_ioctl(struct sock *sk, int cmd, unsigned long arg) return -ENOIOCTLCMD; } +static int dccp_setsockopt_service(struct sock *sk, const u32 service, + char __user *optval, int optlen) +{ + struct dccp_sock *dp = dccp_sk(sk); + struct dccp_service_list *sl = NULL; + + if (service == DCCP_SERVICE_INVALID_VALUE || + optlen > DCCP_SERVICE_LIST_MAX_LEN * sizeof(u32)) + return -EINVAL; + + if (optlen > sizeof(service)) { + sl = kmalloc(optlen, GFP_KERNEL); + if (sl == NULL) + return -ENOMEM; + + sl->dccpsl_nr = optlen / sizeof(u32) - 1; + if (copy_from_user(sl->dccpsl_list, + optval + sizeof(service), + optlen - sizeof(service)) || + dccp_list_has_service(sl, DCCP_SERVICE_INVALID_VALUE)) { + kfree(sl); + return -EFAULT; + } + } + + lock_sock(sk); + dp->dccps_service = service; + + if (dp->dccps_service_list != NULL) + kfree(dp->dccps_service_list); + + dp->dccps_service_list = sl; + release_sock(sk); + return 0; +} + int dccp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { @@ -218,8 +262,10 @@ int dccp_setsockopt(struct sock *sk, int level, int optname, if (get_user(val, (int __user *)optval)) return -EFAULT; - lock_sock(sk); + if (optname == DCCP_SOCKOPT_SERVICE) + return dccp_setsockopt_service(sk, val, optval, optlen); + lock_sock(sk); dp = dccp_sk(sk); err = 0; @@ -236,6 +282,37 @@ int dccp_setsockopt(struct sock *sk, int level, int optname, return err; } +static int dccp_getsockopt_service(struct sock *sk, int len, + u32 __user *optval, + int __user *optlen) +{ + const struct dccp_sock *dp = dccp_sk(sk); + const struct dccp_service_list *sl; + int err = -ENOENT, slen = 0, total_len = sizeof(u32); + + lock_sock(sk); + if (dccp_service_not_initialized(sk)) + goto out; + + if ((sl = dp->dccps_service_list) != NULL) { + slen = sl->dccpsl_nr * sizeof(u32); + total_len += slen; + } + + err = -EINVAL; + if (total_len > len) + goto out; + + err = 0; + if (put_user(total_len, optlen) || + put_user(dp->dccps_service, optval) || + (sl != NULL && copy_to_user(optval + 1, sl->dccpsl_list, slen))) + err = -EFAULT; +out: + release_sock(sk); + return err; +} + int dccp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -248,6 +325,10 @@ int dccp_getsockopt(struct sock *sk, int level, int optname, if (get_user(len, optlen)) return -EFAULT; + if (optname == DCCP_SOCKOPT_SERVICE) + return dccp_getsockopt_service(sk, len, + (u32 __user *)optval, optlen); + len = min_t(unsigned int, len, sizeof(int)); if (len < 0) return -EINVAL; From 1cbf07478bbf3e350a2025bc5ea23fedaa95855a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 16 Sep 2005 16:59:20 -0700 Subject: [PATCH 078/240] [TG3]: Add AMD K8 to list of write-reorder chipsets. Thanks to Andy Stewart for the report and testing debug patches from Michael Chan. Signed-off-by: David S. Miller --- drivers/net/tg3.c | 8 +++++--- include/linux/pci_ids.h | 1 + 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 7599f52e15b3..52eae9f096e2 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -67,8 +67,8 @@ #define DRV_MODULE_NAME "tg3" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "3.39" -#define DRV_MODULE_RELDATE "September 5, 2005" +#define DRV_MODULE_VERSION "3.40" +#define DRV_MODULE_RELDATE "September 15, 2005" #define TG3_DEF_MAC_MODE 0 #define TG3_DEF_RX_MODE 0 @@ -9271,6 +9271,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) static struct pci_device_id write_reorder_chipsets[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_FE_GATE_700C) }, + { PCI_DEVICE(PCI_VENDOR_ID_AMD, + PCI_DEVICE_ID_AMD_K8_NB) }, { }, }; u32 misc_ctrl_reg; @@ -9285,7 +9287,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) tp->tg3_flags2 |= TG3_FLG2_SUN_570X; #endif - /* If we have an AMD 762 chipset, write + /* If we have an AMD 762 or K8 chipset, write * reordering to the mailbox registers done by the host * controller can cause major troubles. We read back from * every mailbox register write to force the writes to be diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 72fe3385743c..3dc161894c7b 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -491,6 +491,7 @@ #define PCI_DEVICE_ID_AMI_MEGARAID2 0x9060 #define PCI_VENDOR_ID_AMD 0x1022 +#define PCI_DEVICE_ID_AMD_K8_NB 0x1100 #define PCI_DEVICE_ID_AMD_LANCE 0x2000 #define PCI_DEVICE_ID_AMD_LANCE_HOME 0x2001 #define PCI_DEVICE_ID_AMD_SCSI 0x2020 From 4451362445b2d83886003f1d739b94e4f000eeeb Mon Sep 17 00:00:00 2001 From: KOVACS Krisztian Date: Fri, 16 Sep 2005 16:59:46 -0700 Subject: [PATCH 079/240] [NETFILTER] CLUSTERIP: introduce reference counting for entries The CLUSTERIP target creates a procfs entry for all different cluster IPs. Although more than one rules can refer to a single cluster IP (and thus a single config structure), removal of the procfs entry is done unconditionally in destroy(). In more complicated situations involving deferred dereferencing of the config structure by procfs and creating a new rule with the same cluster IP it's also possible that no entry will be created for the new rule. This patch fixes the problem by counting the number of entries referencing a given config structure and moving the config list manipulation and procfs entry deletion parts to the clusterip_config_entry_put() function. Signed-off-by: KOVACS Krisztian Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- net/ipv4/netfilter/ipt_CLUSTERIP.c | 80 +++++++++++++++++++++++------- 1 file changed, 62 insertions(+), 18 deletions(-) diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 7d38913754b1..adbf4d752d0f 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -49,6 +49,8 @@ MODULE_DESCRIPTION("iptables target for CLUSTERIP"); struct clusterip_config { struct list_head list; /* list of all configs */ atomic_t refcount; /* reference count */ + atomic_t entries; /* number of entries/rules + * referencing us */ u_int32_t clusterip; /* the IP address */ u_int8_t clustermac[ETH_ALEN]; /* the MAC address */ @@ -76,23 +78,48 @@ static struct proc_dir_entry *clusterip_procdir; #endif static inline void -clusterip_config_get(struct clusterip_config *c) { +clusterip_config_get(struct clusterip_config *c) +{ atomic_inc(&c->refcount); } static inline void -clusterip_config_put(struct clusterip_config *c) { - if (atomic_dec_and_test(&c->refcount)) { +clusterip_config_put(struct clusterip_config *c) +{ + if (atomic_dec_and_test(&c->refcount)) + kfree(c); +} + +/* increase the count of entries(rules) using/referencing this config */ +static inline void +clusterip_config_entry_get(struct clusterip_config *c) +{ + atomic_inc(&c->entries); +} + +/* decrease the count of entries using/referencing this config. If last + * entry(rule) is removed, remove the config from lists, but don't free it + * yet, since proc-files could still be holding references */ +static inline void +clusterip_config_entry_put(struct clusterip_config *c) +{ + if (atomic_dec_and_test(&c->entries)) { write_lock_bh(&clusterip_lock); list_del(&c->list); write_unlock_bh(&clusterip_lock); + dev_mc_delete(c->dev, c->clustermac, ETH_ALEN, 0); dev_put(c->dev); - kfree(c); + + /* In case anyone still accesses the file, the open/close + * functions are also incrementing the refcount on their own, + * so it's safe to remove the entry even if it's in use. */ +#ifdef CONFIG_PROC_FS + remove_proc_entry(c->pde->name, c->pde->parent); +#endif } } - static struct clusterip_config * __clusterip_config_find(u_int32_t clusterip) { @@ -111,7 +138,7 @@ __clusterip_config_find(u_int32_t clusterip) } static inline struct clusterip_config * -clusterip_config_find_get(u_int32_t clusterip) +clusterip_config_find_get(u_int32_t clusterip, int entry) { struct clusterip_config *c; @@ -122,6 +149,8 @@ clusterip_config_find_get(u_int32_t clusterip) return NULL; } atomic_inc(&c->refcount); + if (entry) + atomic_inc(&c->entries); read_unlock_bh(&clusterip_lock); return c; @@ -148,6 +177,7 @@ clusterip_config_init(struct ipt_clusterip_tgt_info *i, u_int32_t ip, c->hash_mode = i->hash_mode; c->hash_initval = i->hash_initval; atomic_set(&c->refcount, 1); + atomic_set(&c->entries, 1); #ifdef CONFIG_PROC_FS /* create proc dir entry */ @@ -415,8 +445,26 @@ checkentry(const char *tablename, /* FIXME: further sanity checks */ - config = clusterip_config_find_get(e->ip.dst.s_addr); - if (!config) { + config = clusterip_config_find_get(e->ip.dst.s_addr, 1); + if (config) { + if (cipinfo->config != NULL) { + /* Case A: This is an entry that gets reloaded, since + * it still has a cipinfo->config pointer. Simply + * increase the entry refcount and return */ + if (cipinfo->config != config) { + printk(KERN_ERR "CLUSTERIP: Reloaded entry " + "has invalid config pointer!\n"); + return 0; + } + clusterip_config_entry_get(cipinfo->config); + } else { + /* Case B: This is a new rule referring to an existing + * clusterip config. */ + cipinfo->config = config; + clusterip_config_entry_get(cipinfo->config); + } + } else { + /* Case C: This is a completely new clusterip config */ if (!(cipinfo->flags & CLUSTERIP_FLAG_NEW)) { printk(KERN_WARNING "CLUSTERIP: no config found for %u.%u.%u.%u, need 'new'\n", NIPQUAD(e->ip.dst.s_addr)); return 0; @@ -443,10 +491,9 @@ checkentry(const char *tablename, } dev_mc_add(config->dev,config->clustermac, ETH_ALEN, 0); } + cipinfo->config = config; } - cipinfo->config = config; - return 1; } @@ -455,13 +502,10 @@ static void destroy(void *matchinfo, unsigned int matchinfosize) { struct ipt_clusterip_tgt_info *cipinfo = matchinfo; - /* we first remove the proc entry and then drop the reference - * count. In case anyone still accesses the file, the open/close - * functions are also incrementing the refcount on their own */ -#ifdef CONFIG_PROC_FS - remove_proc_entry(cipinfo->config->pde->name, - cipinfo->config->pde->parent); -#endif + /* if no more entries are referencing the config, remove it + * from the list and destroy the proc entry */ + clusterip_config_entry_put(cipinfo->config); + clusterip_config_put(cipinfo->config); } @@ -533,7 +577,7 @@ arp_mangle(unsigned int hook, /* if there is no clusterip configuration for the arp reply's * source ip, we don't want to mangle it */ - c = clusterip_config_find_get(payload->src_ip); + c = clusterip_config_find_get(payload->src_ip, 0); if (!c) return NF_ACCEPT; From 136e92bbec0a6d4c2dd1e5b5ac869ab5470547a4 Mon Sep 17 00:00:00 2001 From: KOVACS Krisztian Date: Fri, 16 Sep 2005 17:00:04 -0700 Subject: [PATCH 080/240] [NETFILTER] CLUSTERIP: use a bitmap to store node responsibility data Instead of maintaining an array containing a list of nodes this instance is responsible for let's use a simple bitmap. This provides the following features: * clusterip_responsible() and the add_node()/delete_node() operations become very simple and don't need locking * the config structure is much smaller In spite of the completely different internal data representation the user-space interface remains almost unchanged; the only difference is that the proc file does not list nodes in the order they were added. (The target info structure remains the same.) Signed-off-by: KOVACS Krisztian Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- net/ipv4/netfilter/ipt_CLUSTERIP.c | 141 ++++++++++++----------------- 1 file changed, 60 insertions(+), 81 deletions(-) diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index adbf4d752d0f..9bcb398fbc1f 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -30,7 +31,7 @@ #include #include -#define CLUSTERIP_VERSION "0.7" +#define CLUSTERIP_VERSION "0.8" #define DEBUG_CLUSTERIP @@ -56,8 +57,7 @@ struct clusterip_config { u_int8_t clustermac[ETH_ALEN]; /* the MAC address */ struct net_device *dev; /* device */ u_int16_t num_total_nodes; /* total number of nodes */ - u_int16_t num_local_nodes; /* number of local nodes */ - u_int16_t local_nodes[CLUSTERIP_MAX_NODES]; /* node number array */ + unsigned long local_nodes; /* node number array */ #ifdef CONFIG_PROC_FS struct proc_dir_entry *pde; /* proc dir entry */ @@ -68,8 +68,7 @@ struct clusterip_config { static LIST_HEAD(clusterip_configs); -/* clusterip_lock protects the clusterip_configs list _AND_ the configurable - * data within all structurses (num_local_nodes, local_nodes[]) */ +/* clusterip_lock protects the clusterip_configs list */ static DEFINE_RWLOCK(clusterip_lock); #ifdef CONFIG_PROC_FS @@ -156,6 +155,17 @@ clusterip_config_find_get(u_int32_t clusterip, int entry) return c; } +static void +clusterip_config_init_nodelist(struct clusterip_config *c, + const struct ipt_clusterip_tgt_info *i) +{ + int n; + + for (n = 0; n < i->num_local_nodes; n++) { + set_bit(i->local_nodes[n] - 1, &c->local_nodes); + } +} + static struct clusterip_config * clusterip_config_init(struct ipt_clusterip_tgt_info *i, u_int32_t ip, struct net_device *dev) @@ -172,8 +182,7 @@ clusterip_config_init(struct ipt_clusterip_tgt_info *i, u_int32_t ip, c->clusterip = ip; memcpy(&c->clustermac, &i->clustermac, ETH_ALEN); c->num_total_nodes = i->num_total_nodes; - c->num_local_nodes = i->num_local_nodes; - memcpy(&c->local_nodes, &i->local_nodes, sizeof(c->local_nodes)); + clusterip_config_init_nodelist(c, i); c->hash_mode = i->hash_mode; c->hash_initval = i->hash_initval; atomic_set(&c->refcount, 1); @@ -201,53 +210,28 @@ clusterip_config_init(struct ipt_clusterip_tgt_info *i, u_int32_t ip, static int clusterip_add_node(struct clusterip_config *c, u_int16_t nodenum) { - int i; - write_lock_bh(&clusterip_lock); - - if (c->num_local_nodes >= CLUSTERIP_MAX_NODES - || nodenum > CLUSTERIP_MAX_NODES) { - write_unlock_bh(&clusterip_lock); + if (nodenum == 0 || + nodenum > c->num_total_nodes) return 1; - } - /* check if we alrady have this number in our array */ - for (i = 0; i < c->num_local_nodes; i++) { - if (c->local_nodes[i] == nodenum) { - write_unlock_bh(&clusterip_lock); - return 1; - } - } + /* check if we already have this number in our bitfield */ + if (test_and_set_bit(nodenum - 1, &c->local_nodes)) + return 1; - c->local_nodes[c->num_local_nodes++] = nodenum; - - write_unlock_bh(&clusterip_lock); return 0; } static int clusterip_del_node(struct clusterip_config *c, u_int16_t nodenum) { - int i; - - write_lock_bh(&clusterip_lock); - - if (c->num_local_nodes <= 1 || nodenum > CLUSTERIP_MAX_NODES) { - write_unlock_bh(&clusterip_lock); + if (nodenum == 0 || + nodenum > c->num_total_nodes) return 1; - } - for (i = 0; i < c->num_local_nodes; i++) { - if (c->local_nodes[i] == nodenum) { - int size = sizeof(u_int16_t)*(c->num_local_nodes-(i+1)); - memmove(&c->local_nodes[i], &c->local_nodes[i+1], size); - c->num_local_nodes--; - write_unlock_bh(&clusterip_lock); - return 0; - } - } + if (test_and_clear_bit(nodenum - 1, &c->local_nodes)) + return 0; - write_unlock_bh(&clusterip_lock); return 1; } @@ -315,25 +299,7 @@ clusterip_hashfn(struct sk_buff *skb, struct clusterip_config *config) static inline int clusterip_responsible(struct clusterip_config *config, u_int32_t hash) { - int i; - - read_lock_bh(&clusterip_lock); - - if (config->num_local_nodes == 0) { - read_unlock_bh(&clusterip_lock); - return 0; - } - - for (i = 0; i < config->num_local_nodes; i++) { - if (config->local_nodes[i] == hash) { - read_unlock_bh(&clusterip_lock); - return 1; - } - } - - read_unlock_bh(&clusterip_lock); - - return 0; + return test_bit(hash - 1, &config->local_nodes); } /*********************************************************************** @@ -618,56 +584,69 @@ static struct nf_hook_ops cip_arp_ops = { #ifdef CONFIG_PROC_FS +struct clusterip_seq_position { + unsigned int pos; /* position */ + unsigned int weight; /* number of bits set == size */ + unsigned int bit; /* current bit */ + unsigned long val; /* current value */ +}; + static void *clusterip_seq_start(struct seq_file *s, loff_t *pos) { struct proc_dir_entry *pde = s->private; struct clusterip_config *c = pde->data; - unsigned int *nodeidx; + unsigned int weight; + u_int32_t local_nodes; + struct clusterip_seq_position *idx; - read_lock_bh(&clusterip_lock); - if (*pos >= c->num_local_nodes) + /* FIXME: possible race */ + local_nodes = c->local_nodes; + weight = hweight32(local_nodes); + if (*pos >= weight) return NULL; - nodeidx = kmalloc(sizeof(unsigned int), GFP_KERNEL); - if (!nodeidx) + idx = kmalloc(sizeof(struct clusterip_seq_position), GFP_KERNEL); + if (!idx) return ERR_PTR(-ENOMEM); - *nodeidx = *pos; - return nodeidx; + idx->pos = *pos; + idx->weight = weight; + idx->bit = ffs(local_nodes); + idx->val = local_nodes; + clear_bit(idx->bit - 1, &idx->val); + + return idx; } static void *clusterip_seq_next(struct seq_file *s, void *v, loff_t *pos) { - struct proc_dir_entry *pde = s->private; - struct clusterip_config *c = pde->data; - unsigned int *nodeidx = (unsigned int *)v; + struct clusterip_seq_position *idx = (struct clusterip_seq_position *)v; - *pos = ++(*nodeidx); - if (*pos >= c->num_local_nodes) { + *pos = ++idx->pos; + if (*pos >= idx->weight) { kfree(v); return NULL; } - return nodeidx; + idx->bit = ffs(idx->val); + clear_bit(idx->bit - 1, &idx->val); + return idx; } static void clusterip_seq_stop(struct seq_file *s, void *v) { kfree(v); - - read_unlock_bh(&clusterip_lock); } static int clusterip_seq_show(struct seq_file *s, void *v) { - struct proc_dir_entry *pde = s->private; - struct clusterip_config *c = pde->data; - unsigned int *nodeidx = (unsigned int *)v; + struct clusterip_seq_position *idx = (struct clusterip_seq_position *)v; - if (*nodeidx != 0) + if (idx->pos != 0) seq_putc(s, ','); - seq_printf(s, "%u", c->local_nodes[*nodeidx]); - if (*nodeidx == c->num_local_nodes-1) + seq_printf(s, "%u", idx->bit); + + if (idx->pos == idx->weight - 1) seq_putc(s, '\n'); return 0; From a8f39143ac67ffa2e26ce48aaac6bf5dc7dae95f Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Fri, 16 Sep 2005 17:00:38 -0700 Subject: [PATCH 081/240] [NETFILTER]: Fix oops in conntrack event cache ip_ct_refresh_acct() can be called without a valid "skb" pointer. This used to work, since ct_add_counters() deals with that fact. However, the recently-added event cache doesn't handle this at all. This patch is a quick fix that is supposed to be replaced soon by a cleaner solution during the pending redesign of the event cache. Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_conntrack_core.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index 19cba16e6e1e..f8cd8e42961e 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -1143,7 +1143,10 @@ void ip_ct_refresh_acct(struct ip_conntrack *ct, if (del_timer(&ct->timeout)) { ct->timeout.expires = jiffies + extra_jiffies; add_timer(&ct->timeout); - ip_conntrack_event_cache(IPCT_REFRESH, skb); + /* FIXME: We loose some REFRESH events if this function + * is called without an skb. I'll fix this later -HW */ + if (skb) + ip_conntrack_event_cache(IPCT_REFRESH, skb); } ct_add_counters(ct, ctinfo, skb); write_unlock_bh(&ip_conntrack_lock); From 22abe310bc4b0c684fd3716af6b6116ff1011707 Mon Sep 17 00:00:00 2001 From: Peter Hagervall Date: Fri, 16 Sep 2005 17:01:03 -0700 Subject: [PATCH 082/240] [TG3]: Sparse fixes for tg3 Change 0 to NULL where appropriate. Signed-off-by: Peter Hagervall Acked-by: Jeff Garzik Signed-off-by: David S. Miller --- drivers/net/tg3.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 52eae9f096e2..06b02c59d90b 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -9534,7 +9534,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) tp->write32_rx_mbox = tg3_write_indirect_mbox; iounmap(tp->regs); - tp->regs = 0; + tp->regs = NULL; pci_read_config_word(tp->pdev, PCI_COMMAND, &pci_cmd); pci_cmd &= ~PCI_COMMAND_MEMORY; @@ -10682,7 +10682,7 @@ static int __devinit tg3_init_one(struct pci_dev *pdev, err_out_iounmap: if (tp->regs) { iounmap(tp->regs); - tp->regs = 0; + tp->regs = NULL; } err_out_free_dev: @@ -10707,7 +10707,7 @@ static void __devexit tg3_remove_one(struct pci_dev *pdev) unregister_netdev(dev); if (tp->regs) { iounmap(tp->regs); - tp->regs = 0; + tp->regs = NULL; } free_netdev(dev); pci_release_regions(pdev); From 777ed97f3e549832845d70dcae96cb36c41a543a Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Sat, 17 Sep 2005 00:41:02 -0700 Subject: [PATCH 083/240] [NETFILTER] Fix Kconfig dependencies for nfnetlink/ctnetlink Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- net/ipv4/netfilter/Kconfig | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 30aa8e2ee214..85190de5710a 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -8,6 +8,7 @@ menu "IP: Netfilter Configuration" # connection tracking, helpers and protocols config IP_NF_CONNTRACK tristate "Connection tracking (required for masq/NAT)" + select NETFILTER_NETLINK if IP_NF_CONNTRACK_NETLINK!=n ---help--- Connection tracking keeps a record of what packets have passed through your machine, in order to figure out how they are related @@ -51,6 +52,15 @@ config IP_NF_CONNTRACK_EVENTS IF unsure, say `N'. +config IP_NF_CONNTRACK_NETLINK + tristate 'Connection tracking netlink interface' + depends on IP_NF_CONNTRACK && NETFILTER_NETLINK + default IP_NF_CONNTRACK if NETFILTER_NETLINK=y + default m if NETFILTER_NETLINK=m + help + This option enables support for a netlink-based userspace interface + + config IP_NF_CT_PROTO_SCTP tristate 'SCTP protocol connection tracking support (EXPERIMENTAL)' depends on IP_NF_CONNTRACK && EXPERIMENTAL @@ -774,11 +784,5 @@ config IP_NF_ARP_MANGLE Allows altering the ARP packet payload: source and destination hardware and network addresses. -config IP_NF_CONNTRACK_NETLINK - tristate 'Connection tracking netlink interface' - depends on IP_NF_CONNTRACK && NETFILTER_NETLINK - help - This option enables support for a netlink-based userspace interface - endmenu From 9eb0eec74d9425f4376df653304783339a027d79 Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Sat, 17 Sep 2005 00:41:21 -0700 Subject: [PATCH 084/240] [NETFILTER] move nfnetlink options to right location in kconfig menu Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- net/Kconfig | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/Kconfig b/net/Kconfig index 2bdd5623fdd5..60f6f321bd76 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -140,6 +140,7 @@ config BRIDGE_NETFILTER If unsure, say N. +source "net/netfilter/Kconfig" source "net/ipv4/netfilter/Kconfig" source "net/ipv6/netfilter/Kconfig" source "net/decnet/netfilter/Kconfig" @@ -206,8 +207,6 @@ config NET_PKTGEN To compile this code as a module, choose M here: the module will be called pktgen. -source "net/netfilter/Kconfig" - endmenu endmenu From eb8edb085716621605cc2e7131a6369d2223d992 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sat, 17 Sep 2005 00:42:26 -0700 Subject: [PATCH 085/240] [DCCP]: Add MAINTAINERS and CREDITS entries Also remove the SPX entry in MAINTAINERS, forgot to do that when I removed it. Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- CREDITS | 26 ++++++++++++++------------ MAINTAINERS | 13 +++++++------ 2 files changed, 21 insertions(+), 18 deletions(-) diff --git a/CREDITS b/CREDITS index f553f8cfaa62..a347520bef2d 100644 --- a/CREDITS +++ b/CREDITS @@ -2211,6 +2211,15 @@ D: OV511 driver S: (address available on request) S: USA +N: Ian McDonald +E: iam4@cs.waikato.ac.nz +E: imcdnzl@gmail.com +W: http://wand.net.nz/~iam4 +W: http://imcdnzl.blogspot.com +D: DCCP, CCID3 +S: Hamilton +S: New Zealand + N: Patrick McHardy E: kaber@trash.net P: 1024D/12155E80 B128 7DE6 FF0A C2B2 48BE AB4C C9D4 964E 1215 5E80 @@ -2246,19 +2255,12 @@ S: D-90453 Nuernberg S: Germany N: Arnaldo Carvalho de Melo -E: acme@conectiva.com.br -E: acme@kernel.org -E: acme@gnu.org -W: http://bazar2.conectiva.com.br/~acme -W: http://advogato.org/person/acme +E: acme@mandriva.com +E: acme@ghostprotocols.net +W: http://oops.ghostprotocols.net:81/blog/ P: 1024D/9224DF01 D5DF E3BB E3C8 BCBB F8AD 841A B6AB 4681 9224 DF01 -D: wanrouter hacking -D: misc Makefile, Config.in, drivers and network stacks fixes -D: IPX & LLC network stacks maintainer -D: Cyclom 2X synchronous card driver -D: wl3501 PCMCIA wireless card driver -D: i18n for minicom, net-tools, util-linux, fetchmail, etc -S: Conectiva S.A. +D: IPX, LLC, DCCP, cyc2x, wl3501_cs, net/ hacks +S: Mandriva S: R. Tocantins, 89 - Cristo Rei S: 80050-430 - Curitiba - Paraná S: Brazil diff --git a/MAINTAINERS b/MAINTAINERS index d1e0eb46d201..dc8f3babcabd 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -686,6 +686,13 @@ P: Guennadi Liakhovetski M: g.liakhovetski@gmx.de S: Maintained +DCCP PROTOCOL +P: Arnaldo Carvalho de Melo +M: acme@mandriva.com +L: dccp@vger.kernel.org +W: http://www.wlug.org.nz/DCCP +S: Maintained + DECnet NETWORK LAYER P: Patrick Caulfield M: patrick@tykepenguin.com @@ -2271,12 +2278,6 @@ M: R.E.Wolff@BitWizard.nl L: linux-kernel@vger.kernel.org ? S: Supported -SPX NETWORK LAYER -P: Jay Schulist -M: jschlst@samba.org -L: netdev@vger.kernel.org -S: Supported - SRM (Alpha) environment access P: Jan-Benedict Glaw M: jbglaw@lug-owl.de From c58ec93245a1fb7354f9e331960380827b9f41db Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Sat, 17 Sep 2005 00:46:27 -0700 Subject: [PATCH 086/240] [TG3]: Fix 4GB boundary tx handling Fix and simplify the workaround code for the 4GB boundary tx buffer hardware bug. 1. Need to unmap the original SKB's dma addresses if a new SKB cannot be allocated. 2. Need to pass the base flag to tigon3_4gb_hwbug_workaround() or TSO won't work properly. 3. The guilty entry and length parameters for tigon3_4gb_hwbug_workaround() are removed as they are not necessary. 4. Remove assumption that only one fragment can hit the 4GB boundary. Another fragment can hit 8GB for example. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/tg3.c | 94 ++++++++++++++++++++--------------------------- 1 file changed, 39 insertions(+), 55 deletions(-) diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 06b02c59d90b..81f4aedf534c 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -3442,31 +3442,47 @@ static void tg3_tx_timeout(struct net_device *dev) schedule_work(&tp->reset_task); } +/* Test for DMA buffers crossing any 4GB boundaries: 4G, 8G, etc */ +static inline int tg3_4g_overflow_test(dma_addr_t mapping, int len) +{ + u32 base = (u32) mapping & 0xffffffff; + + return ((base > 0xffffdcc0) && + (base + len + 8 < base)); +} + static void tg3_set_txd(struct tg3 *, int, dma_addr_t, int, u32, u32); static int tigon3_4gb_hwbug_workaround(struct tg3 *tp, struct sk_buff *skb, - u32 guilty_entry, int guilty_len, - u32 last_plus_one, u32 *start, u32 mss) + u32 last_plus_one, u32 *start, + u32 base_flags, u32 mss) { struct sk_buff *new_skb = skb_copy(skb, GFP_ATOMIC); - dma_addr_t new_addr; + dma_addr_t new_addr = 0; u32 entry = *start; - int i; + int i, ret = 0; if (!new_skb) { - dev_kfree_skb(skb); - return -1; + ret = -1; + } else { + /* New SKB is guaranteed to be linear. */ + entry = *start; + new_addr = pci_map_single(tp->pdev, new_skb->data, new_skb->len, + PCI_DMA_TODEVICE); + /* Make sure new skb does not cross any 4G boundaries. + * Drop the packet if it does. + */ + if (tg3_4g_overflow_test(new_addr, new_skb->len)) { + ret = -1; + dev_kfree_skb(new_skb); + new_skb = NULL; + } else { + tg3_set_txd(tp, entry, new_addr, new_skb->len, + base_flags, 1 | (mss << 1)); + *start = NEXT_TX(entry); + } } - /* New SKB is guaranteed to be linear. */ - entry = *start; - new_addr = pci_map_single(tp->pdev, new_skb->data, new_skb->len, - PCI_DMA_TODEVICE); - tg3_set_txd(tp, entry, new_addr, new_skb->len, - (skb->ip_summed == CHECKSUM_HW) ? - TXD_FLAG_TCPUDP_CSUM : 0, 1 | (mss << 1)); - *start = NEXT_TX(entry); - /* Now clean up the sw ring entries. */ i = 0; while (entry != last_plus_one) { @@ -3491,7 +3507,7 @@ static int tigon3_4gb_hwbug_workaround(struct tg3 *tp, struct sk_buff *skb, dev_kfree_skb(skb); - return 0; + return ret; } static void tg3_set_txd(struct tg3 *tp, int entry, @@ -3517,19 +3533,10 @@ static void tg3_set_txd(struct tg3 *tp, int entry, txd->vlan_tag = vlan_tag << TXD_VLAN_TAG_SHIFT; } -static inline int tg3_4g_overflow_test(dma_addr_t mapping, int len) -{ - u32 base = (u32) mapping & 0xffffffff; - - return ((base > 0xffffdcc0) && - (base + len + 8 < base)); -} - static int tg3_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct tg3 *tp = netdev_priv(dev); dma_addr_t mapping; - unsigned int i; u32 len, entry, base_flags, mss; int would_hit_hwbug; @@ -3624,7 +3631,7 @@ static int tg3_start_xmit(struct sk_buff *skb, struct net_device *dev) would_hit_hwbug = 0; if (tg3_4g_overflow_test(mapping, len)) - would_hit_hwbug = entry + 1; + would_hit_hwbug = 1; tg3_set_txd(tp, entry, mapping, len, base_flags, (skb_shinfo(skb)->nr_frags == 0) | (mss << 1)); @@ -3648,12 +3655,8 @@ static int tg3_start_xmit(struct sk_buff *skb, struct net_device *dev) tp->tx_buffers[entry].skb = NULL; pci_unmap_addr_set(&tp->tx_buffers[entry], mapping, mapping); - if (tg3_4g_overflow_test(mapping, len)) { - /* Only one should match. */ - if (would_hit_hwbug) - BUG(); - would_hit_hwbug = entry + 1; - } + if (tg3_4g_overflow_test(mapping, len)) + would_hit_hwbug = 1; if (tp->tg3_flags2 & TG3_FLG2_HW_TSO) tg3_set_txd(tp, entry, mapping, len, @@ -3669,34 +3672,15 @@ static int tg3_start_xmit(struct sk_buff *skb, struct net_device *dev) if (would_hit_hwbug) { u32 last_plus_one = entry; u32 start; - unsigned int len = 0; - would_hit_hwbug -= 1; - entry = entry - 1 - skb_shinfo(skb)->nr_frags; - entry &= (TG3_TX_RING_SIZE - 1); - start = entry; - i = 0; - while (entry != last_plus_one) { - if (i == 0) - len = skb_headlen(skb); - else - len = skb_shinfo(skb)->frags[i-1].size; - - if (entry == would_hit_hwbug) - break; - - i++; - entry = NEXT_TX(entry); - - } + start = entry - 1 - skb_shinfo(skb)->nr_frags; + start &= (TG3_TX_RING_SIZE - 1); /* If the workaround fails due to memory/mapping * failure, silently drop this packet. */ - if (tigon3_4gb_hwbug_workaround(tp, skb, - entry, len, - last_plus_one, - &start, mss)) + if (tigon3_4gb_hwbug_workaround(tp, skb, last_plus_one, + &start, base_flags, mss)) goto out_unlock; entry = start; From d7e70ba45fd9850296ebb78ff5827f6a375f650c Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Fri, 16 Sep 2005 19:27:29 -0700 Subject: [PATCH 087/240] [PATCH] RAID6 Altivec fix This patch fixes a signedness bug with RAID6 for Altivec, and makes the Altivec code testable in userspace. Signed-off-by: H. Peter Anvin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/raid6.h | 4 ++++ drivers/md/raid6algos.c | 1 + drivers/md/raid6altivec.uc | 18 +++++++++++++----- drivers/md/raid6test/Makefile | 27 ++++++++++++++++++++++----- 4 files changed, 40 insertions(+), 10 deletions(-) diff --git a/drivers/md/raid6.h b/drivers/md/raid6.h index f80ee6350edf..31cbee71365f 100644 --- a/drivers/md/raid6.h +++ b/drivers/md/raid6.h @@ -69,9 +69,13 @@ extern const char raid6_empty_zero_page[PAGE_SIZE]; #define __init #define __exit #define __attribute_const__ __attribute__((const)) +#define noinline __attribute__((noinline)) #define preempt_enable() #define preempt_disable() +#define cpu_has_feature(x) 1 +#define enable_kernel_altivec() +#define disable_kernel_altivec() #endif /* __KERNEL__ */ diff --git a/drivers/md/raid6algos.c b/drivers/md/raid6algos.c index acf386fc4b4f..51c63c0cf1c9 100644 --- a/drivers/md/raid6algos.c +++ b/drivers/md/raid6algos.c @@ -19,6 +19,7 @@ #include "raid6.h" #ifndef __KERNEL__ #include +#include #endif struct raid6_calls raid6_call; diff --git a/drivers/md/raid6altivec.uc b/drivers/md/raid6altivec.uc index 1de8f030eee0..b9afd35b8812 100644 --- a/drivers/md/raid6altivec.uc +++ b/drivers/md/raid6altivec.uc @@ -27,16 +27,20 @@ #ifdef CONFIG_ALTIVEC #include -#include -#include +#ifdef __KERNEL__ +# include +# include +#endif /* - * This is the C data type to use + * This is the C data type to use. We use a vector of + * signed char so vec_cmpgt() will generate the right + * instruction. */ -typedef vector unsigned char unative_t; +typedef vector signed char unative_t; -#define NBYTES(x) ((vector unsigned char) {x,x,x,x, x,x,x,x, x,x,x,x, x,x,x,x}) +#define NBYTES(x) ((vector signed char) {x,x,x,x, x,x,x,x, x,x,x,x, x,x,x,x}) #define NSIZE sizeof(unative_t) /* @@ -108,7 +112,11 @@ int raid6_have_altivec(void); int raid6_have_altivec(void) { /* This assumes either all CPUs have Altivec or none does */ +# ifdef __KERNEL__ return cpu_has_feature(CPU_FTR_ALTIVEC); +# else + return 1; +# endif } #endif diff --git a/drivers/md/raid6test/Makefile b/drivers/md/raid6test/Makefile index 557806728609..78e0396adf2a 100644 --- a/drivers/md/raid6test/Makefile +++ b/drivers/md/raid6test/Makefile @@ -8,6 +8,8 @@ OPTFLAGS = -O2 # Adjust as desired CFLAGS = -I.. -g $(OPTFLAGS) LD = ld PERL = perl +AR = ar +RANLIB = ranlib .c.o: $(CC) $(CFLAGS) -c -o $@ $< @@ -18,18 +20,33 @@ PERL = perl %.uc: ../%.uc cp -f $< $@ -all: raid6.o raid6test +all: raid6.a raid6test -raid6.o: raid6int1.o raid6int2.o raid6int4.o raid6int8.o raid6int16.o \ +raid6.a: raid6int1.o raid6int2.o raid6int4.o raid6int8.o raid6int16.o \ raid6int32.o \ raid6mmx.o raid6sse1.o raid6sse2.o \ + raid6altivec1.o raid6altivec2.o raid6altivec4.o raid6altivec8.o \ raid6recov.o raid6algos.o \ raid6tables.o - $(LD) -r -o $@ $^ + rm -f $@ + $(AR) cq $@ $^ + $(RANLIB) $@ -raid6test: raid6.o test.c +raid6test: test.c raid6.a $(CC) $(CFLAGS) -o raid6test $^ +raid6altivec1.c: raid6altivec.uc ../unroll.pl + $(PERL) ../unroll.pl 1 < raid6altivec.uc > $@ + +raid6altivec2.c: raid6altivec.uc ../unroll.pl + $(PERL) ../unroll.pl 2 < raid6altivec.uc > $@ + +raid6altivec4.c: raid6altivec.uc ../unroll.pl + $(PERL) ../unroll.pl 4 < raid6altivec.uc > $@ + +raid6altivec8.c: raid6altivec.uc ../unroll.pl + $(PERL) ../unroll.pl 8 < raid6altivec.uc > $@ + raid6int1.c: raid6int.uc ../unroll.pl $(PERL) ../unroll.pl 1 < raid6int.uc > $@ @@ -52,7 +69,7 @@ raid6tables.c: mktables ./mktables > raid6tables.c clean: - rm -f *.o mktables mktables.c raid6int.uc raid6*.c raid6test + rm -f *.o *.a mktables mktables.c raid6int.uc raid6*.c raid6test spotless: clean rm -f *~ From 12e8780813c8c515b685b86c1306d169a4cae5fe Mon Sep 17 00:00:00 2001 From: Richard Purdie Date: Fri, 16 Sep 2005 19:27:30 -0700 Subject: [PATCH 088/240] [PATCH] SharpSL: Add missing hunk from backlight update This hunk from the sharpsl/corgi backlight update appears to have got lost somewhere along the way. Its needed to match the other changes. Signed-Off-By: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/backlight/corgi_bl.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/video/backlight/corgi_bl.c b/drivers/video/backlight/corgi_bl.c index a32817678552..630f2dfa9699 100644 --- a/drivers/video/backlight/corgi_bl.c +++ b/drivers/video/backlight/corgi_bl.c @@ -19,17 +19,18 @@ #include #include -#include -#include +#include +#include -#define CORGI_MAX_INTENSITY 0x3e #define CORGI_DEFAULT_INTENSITY 0x1f -#define CORGI_LIMIT_MASK 0x0b +#define CORGI_LIMIT_MASK 0x0b static int corgibl_powermode = FB_BLANK_UNBLANK; static int current_intensity = 0; static int corgibl_limit = 0; +static void (*corgibl_mach_set_intensity)(int intensity); static spinlock_t bl_lock = SPIN_LOCK_UNLOCKED; +static struct backlight_properties corgibl_data; static void corgibl_send_intensity(int intensity) { From 62052d42a5327281fc43fd7bcfb73ac7d36ffc2e Mon Sep 17 00:00:00 2001 From: Richard Purdie Date: Fri, 16 Sep 2005 19:27:31 -0700 Subject: [PATCH 089/240] [PATCH] MTD: Update SharpSL partition definitions Add partition definitions for the new Sharp Zaurus models Spitz (SL-C3000), Akita (SL-C1000) and Borzoi (SL-C3100) Signed-off-by: Richard Purdie Signed-off-by: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/mtd/maps/sharpsl-flash.c | 2 +- drivers/mtd/nand/sharpsl.c | 10 ++++++++-- 2 files changed, 9 insertions(+), 3 deletions(-) diff --git a/drivers/mtd/maps/sharpsl-flash.c b/drivers/mtd/maps/sharpsl-flash.c index d15da6fd84c1..b7f093fbf9b0 100644 --- a/drivers/mtd/maps/sharpsl-flash.c +++ b/drivers/mtd/maps/sharpsl-flash.c @@ -82,7 +82,7 @@ int __init init_sharpsl(void) } else if (machine_is_tosa()) { sharpsl_partitions[0].size=0x006a0000; sharpsl_partitions[0].offset=0x00160000; - } else if (machine_is_spitz()) { + } else if (machine_is_spitz() || machine_is_akita() || machine_is_borzoi()) { sharpsl_partitions[0].size=0x006b0000; sharpsl_partitions[0].offset=0x00140000; } else { diff --git a/drivers/mtd/nand/sharpsl.c b/drivers/mtd/nand/sharpsl.c index 9853b87bb756..88b5b5b40b43 100644 --- a/drivers/mtd/nand/sharpsl.c +++ b/drivers/mtd/nand/sharpsl.c @@ -221,10 +221,16 @@ sharpsl_nand_init(void) sharpsl_partition_info[1].size=25 * 1024 * 1024; } else if (machine_is_husky()) { sharpsl_partition_info[1].size=53 * 1024 * 1024; - } + } else if (machine_is_spitz()) { + sharpsl_partition_info[1].size=5 * 1024 * 1024; + } else if (machine_is_akita()) { + sharpsl_partition_info[1].size=58 * 1024 * 1024; + } else if (machine_is_borzoi()) { + sharpsl_partition_info[1].size=32 * 1024 * 1024; + } } - if (machine_is_husky()) { + if (machine_is_husky() || machine_is_borzoi()) { /* Need to use small eraseblock size for backward compatibility */ sharpsl_mtd->flags |= MTD_NO_VIRTBLOCKS; } From 1306a2b2be35179449db737ac0562b067cb96ddb Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 16 Sep 2005 19:27:32 -0700 Subject: [PATCH 090/240] [PATCH] s390: default configuration Update default configuration of s390. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/defconfig | 36 +++++++++++++++++++++++++----------- 1 file changed, 25 insertions(+), 11 deletions(-) diff --git a/arch/s390/defconfig b/arch/s390/defconfig index 0865251a3f44..45d44c6bb39d 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.13-rc4 -# Fri Jul 29 14:49:30 2005 +# Linux kernel version: 2.6.14-rc1 +# Wed Sep 14 16:46:19 2005 # CONFIG_MMU=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y @@ -21,6 +21,7 @@ CONFIG_INIT_ENV_ARG_LIMIT=32 # General setup # CONFIG_LOCALVERSION="" +CONFIG_LOCALVERSION_AUTO=y CONFIG_SWAP=y CONFIG_SYSVIPC=y CONFIG_POSIX_MQUEUE=y @@ -33,6 +34,7 @@ CONFIG_KOBJECT_UEVENT=y CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y # CONFIG_CPUSETS is not set +CONFIG_INITRAMFS_SOURCE="" # CONFIG_EMBEDDED is not set CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_ALL is not set @@ -94,6 +96,7 @@ CONFIG_FLATMEM_MANUAL=y # CONFIG_SPARSEMEM_MANUAL is not set CONFIG_FLATMEM=y CONFIG_FLAT_NODE_MEM_MAP=y +# CONFIG_SPARSEMEM_STATIC is not set # # I/O subsystem configuration @@ -151,8 +154,8 @@ CONFIG_IP_FIB_HASH=y # CONFIG_INET_ESP is not set # CONFIG_INET_IPCOMP is not set # CONFIG_INET_TUNNEL is not set -CONFIG_IP_TCPDIAG=y -CONFIG_IP_TCPDIAG_IPV6=y +CONFIG_INET_DIAG=y +CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set CONFIG_TCP_CONG_BIC=y CONFIG_IPV6=y @@ -164,6 +167,11 @@ CONFIG_IPV6=y # CONFIG_IPV6_TUNNEL is not set # CONFIG_NETFILTER is not set +# +# DCCP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_DCCP is not set + # # SCTP Configuration (EXPERIMENTAL) # @@ -217,9 +225,11 @@ CONFIG_NET_CLS_POLICE=y # Network testing # # CONFIG_NET_PKTGEN is not set +# CONFIG_NETFILTER_NETLINK is not set # CONFIG_HAMRADIO is not set # CONFIG_IRDA is not set # CONFIG_BT is not set +# CONFIG_IEEE80211 is not set # CONFIG_PCMCIA is not set # @@ -233,6 +243,7 @@ CONFIG_PREVENT_FIRMWARE_BUILD=y # # SCSI device support # +# CONFIG_RAID_ATTRS is not set CONFIG_SCSI=y CONFIG_SCSI_PROC_FS=y @@ -260,6 +271,7 @@ CONFIG_SCSI_LOGGING=y # CONFIG_SCSI_SPI_ATTRS is not set CONFIG_SCSI_FC_ATTRS=y # CONFIG_SCSI_ISCSI_ATTRS is not set +# CONFIG_SCSI_SAS_ATTRS is not set # # SCSI low-level drivers @@ -280,7 +292,6 @@ CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=16 CONFIG_BLK_DEV_RAM_SIZE=4096 CONFIG_BLK_DEV_INITRD=y -CONFIG_INITRAMFS_SOURCE="" # CONFIG_LBD is not set # CONFIG_CDROM_PKTCDVD is not set @@ -383,6 +394,10 @@ CONFIG_BONDING=m CONFIG_EQUALIZER=m CONFIG_TUN=m +# +# PHY device support +# + # # Ethernet (10 or 100Mbit) # @@ -453,10 +468,6 @@ CONFIG_FS_MBCACHE=y # CONFIG_REISERFS_FS is not set # CONFIG_JFS_FS is not set # CONFIG_FS_POSIX_ACL is not set - -# -# XFS support -# # CONFIG_XFS_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set @@ -465,6 +476,7 @@ CONFIG_INOTIFY=y CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set # CONFIG_AUTOFS4_FS is not set +# CONFIG_FUSE_FS is not set # # CD-ROM/DVD Filesystems @@ -485,11 +497,10 @@ CONFIG_DNOTIFY=y CONFIG_PROC_FS=y CONFIG_PROC_KCORE=y CONFIG_SYSFS=y -# CONFIG_DEVPTS_FS_XATTR is not set CONFIG_TMPFS=y -# CONFIG_TMPFS_XATTR is not set # CONFIG_HUGETLB_PAGE is not set CONFIG_RAMFS=y +# CONFIG_RELAYFS_FS is not set # # Miscellaneous filesystems @@ -533,6 +544,7 @@ CONFIG_SUNRPC=y # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set +# CONFIG_9P_FS is not set # # Partition Types @@ -572,6 +584,7 @@ CONFIG_MSDOS_PARTITION=y CONFIG_DEBUG_KERNEL=y CONFIG_MAGIC_SYSRQ=y CONFIG_LOG_BUF_SHIFT=17 +CONFIG_DETECT_SOFTLOCKUP=y # CONFIG_SCHEDSTATS is not set # CONFIG_DEBUG_SLAB is not set CONFIG_DEBUG_PREEMPT=y @@ -626,5 +639,6 @@ CONFIG_CRYPTO=y # Library routines # # CONFIG_CRC_CCITT is not set +# CONFIG_CRC16 is not set CONFIG_CRC32=m # CONFIG_LIBCRC32C is not set From d176795b3eedc18f95160ecc0d7cdf0d2bc5a578 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 16 Sep 2005 19:27:33 -0700 Subject: [PATCH 091/240] [PATCH] s390: bl_dev array size Calculate correct size for bl_dev array. It should be 8KB instead of 512KB for 2^16 bits. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/s390/cio/blacklist.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/s390/cio/blacklist.c b/drivers/s390/cio/blacklist.c index aac83ce6469c..a1c52a682191 100644 --- a/drivers/s390/cio/blacklist.c +++ b/drivers/s390/cio/blacklist.c @@ -1,7 +1,7 @@ /* * drivers/s390/cio/blacklist.c * S/390 common I/O routines -- blacklisting of specific devices - * $Revision: 1.34 $ + * $Revision: 1.35 $ * * Copyright (C) 1999-2002 IBM Deutschland Entwicklung GmbH, * IBM Corporation @@ -35,7 +35,7 @@ */ /* 65536 bits to indicate if a devno is blacklisted or not */ -#define __BL_DEV_WORDS (__MAX_SUBCHANNELS + (8*sizeof(long) - 1) / \ +#define __BL_DEV_WORDS ((__MAX_SUBCHANNELS + (8*sizeof(long) - 1)) / \ (8*sizeof(long))) static unsigned long bl_dev[__BL_DEV_WORDS]; typedef enum {add, free} range_action; From 90f4e12112284f1a41594e2950e6eced5b064efc Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 16 Sep 2005 19:27:34 -0700 Subject: [PATCH 092/240] [PATCH] s390: crypto driver patch take 2 Got confused with the crypto update. The last patch added a call to destroy_workqueue() for a non-existent workqueue with the comment "Remove device workqueue on module unload". This is nonsense. Remove the offending hunk again. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/s390/crypto/z90main.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/s390/crypto/z90main.c b/drivers/s390/crypto/z90main.c index 6aeef3bacc33..0cb47eca91f3 100644 --- a/drivers/s390/crypto/z90main.c +++ b/drivers/s390/crypto/z90main.c @@ -682,9 +682,6 @@ z90crypt_cleanup_module(void) del_timer(&config_timer); del_timer(&cleanup_timer); - if (z90_device_work) - destroy_workqueue(z90_device_work); - destroy_z90crypt(); PRINTKN("Unloaded.\n"); From b7ae9dd80a3ecf070ea0c5fe1d405cd7687d6c51 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 16 Sep 2005 19:27:34 -0700 Subject: [PATCH 093/240] [PATCH] s390: show_cpuinfo fix Disable preemption in show_cpuinfo to avoid problems and the warning about smp_processor_id. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/setup.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 5ba5a5485da9..907d16ac9dc9 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -634,6 +634,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) struct cpuinfo_S390 *cpuinfo; unsigned long n = (unsigned long) v - 1; + preempt_disable(); if (!n) { seq_printf(m, "vendor_id : IBM/S390\n" "# processors : %i\n" @@ -658,6 +659,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) cpuinfo->cpu_id.ident, cpuinfo->cpu_id.machine); } + preempt_enable(); return 0; } From c782268be8892962a1ae2c09a2e87e7849454ef1 Mon Sep 17 00:00:00 2001 From: Volker Sameske Date: Fri, 16 Sep 2005 19:27:35 -0700 Subject: [PATCH 094/240] [PATCH] s390: diag 0x308 reipl Add code to support the re-IPL method using diagnose 0x308. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/Makefile | 2 +- arch/s390/kernel/reipl_diag.c | 39 +++++++++++++++++++++++++++++++++++ arch/s390/kernel/setup.c | 3 +++ arch/s390/kernel/smp.c | 3 +++ 4 files changed, 46 insertions(+), 1 deletion(-) create mode 100644 arch/s390/kernel/reipl_diag.c diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index ab1e49d2e518..8584dd823218 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -6,7 +6,7 @@ EXTRA_AFLAGS := -traditional obj-y := bitmap.o traps.o time.o process.o \ setup.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o \ - semaphore.o s390_ext.o debug.o profile.o irq.o + semaphore.o s390_ext.o debug.o profile.o irq.o reipl_diag.o extra-$(CONFIG_ARCH_S390_31) += head.o extra-$(CONFIG_ARCH_S390X) += head64.o diff --git a/arch/s390/kernel/reipl_diag.c b/arch/s390/kernel/reipl_diag.c new file mode 100644 index 000000000000..83cb42bc0b76 --- /dev/null +++ b/arch/s390/kernel/reipl_diag.c @@ -0,0 +1,39 @@ +/* + * This file contains the implementation of the + * Linux re-IPL support + * + * (C) Copyright IBM Corp. 2005 + * + * Author(s): Volker Sameske (sameske@de.ibm.com) + * + */ + +#include + +static unsigned int reipl_diag_rc1; +static unsigned int reipl_diag_rc2; + +/* + * re-IPL the system using the last used IPL parameters + */ +void reipl_diag(void) +{ + asm volatile ( + " la %%r4,0\n" + " la %%r5,0\n" + " diag %%r4,%2,0x308\n" + "0:\n" + " st %%r4,%0\n" + " st %%r5,%1\n" + ".section __ex_table,\"a\"\n" +#ifdef __s390x__ + " .align 8\n" + " .quad 0b, 0b\n" +#else + " .align 4\n" + " .long 0b, 0b\n" +#endif + ".previous\n" + : "=m" (reipl_diag_rc1), "=m" (reipl_diag_rc2) + : "d" (3) : "cc", "4", "5" ); +} diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 907d16ac9dc9..5204778b8e5e 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -261,8 +261,11 @@ void (*_machine_power_off)(void) = machine_power_off_smp; * Reboot, halt and power_off routines for non SMP. */ extern void reipl(unsigned long devno); +extern void reipl_diag(void); static void do_machine_restart_nonsmp(char * __unused) { + reipl_diag(); + if (MACHINE_IS_VM) cpcmd ("IPL", NULL, 0); else diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 85222fee4361..e13c87b446b2 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -65,6 +65,7 @@ extern char vmhalt_cmd[]; extern char vmpoff_cmd[]; extern void reipl(unsigned long devno); +extern void reipl_diag(void); static void smp_ext_bitcall(int, ec_bit_sig); static void smp_ext_bitcall_others(ec_bit_sig); @@ -283,6 +284,8 @@ static void do_machine_restart(void * __unused) * interrupted by an external interrupt and s390irq * locks are always held disabled). */ + reipl_diag(); + if (MACHINE_IS_VM) cpcmd ("IPL", NULL, 0, NULL); else From fa72e26a640e551637f91122d782d5a85d514f5b Mon Sep 17 00:00:00 2001 From: Domen Puncer Date: Fri, 16 Sep 2005 19:27:36 -0700 Subject: [PATCH 095/240] [PATCH] Remove arch/arm26/boot/compressed/hw-bse.c Remove nowhere referenced file (egrep "hw-bse\." didn't find anything). Signed-off-by: Domen Puncer Signed-off-by: Alexey Dobriyan Acked-by: Ian Molton Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm26/boot/compressed/hw-bse.c | 74 ----------------------------- 1 file changed, 74 deletions(-) delete mode 100644 arch/arm26/boot/compressed/hw-bse.c diff --git a/arch/arm26/boot/compressed/hw-bse.c b/arch/arm26/boot/compressed/hw-bse.c deleted file mode 100644 index 3e8f07f8e08a..000000000000 --- a/arch/arm26/boot/compressed/hw-bse.c +++ /dev/null @@ -1,74 +0,0 @@ -/* - * Bright Star Engineering Inc. - * - * code for readng parameters from the - * parameter blocks of the boot block - * flash memory - * - */ - -static int strcmp(const char *s1, const char *s2) -{ - while (*s1 != '\0' && *s1 == *s2) - { - s1++; - s2++; - } - - return (*(unsigned char *) s1) - (*(unsigned char *) s2); -} - -struct pblk_t { - char type; - unsigned short size; -}; - -static char *bse_getflashparam(char *name) { - unsigned int esize; - char *q,*r; - unsigned char *p,*e; - struct pblk_t *thepb = (struct pblk_t *) 0x00004000; - struct pblk_t *altpb = (struct pblk_t *) 0x00006000; - if (thepb->type&1) { - if (altpb->type&1) { - /* no valid param block */ - return (char*)0; - } else { - /* altpb is valid */ - struct pblk_t *tmp; - tmp = thepb; - thepb = altpb; - altpb = tmp; - } - } - p = (char*)thepb + sizeof(struct pblk_t); - e = p + thepb->size; - while (p < e) { - q = p; - esize = *p; - if (esize == 0xFF) break; - if (esize == 0) break; - if (esize > 127) { - esize = (esize&0x7F)<<8 | p[1]; - q++; - } - q++; - r=q; - if (*r && ((name == 0) || (!strcmp(name,r)))) { - while (*q++) ; - return q; - } - p+=esize; - } - return (char*)0; -} - -void bse_setup(void) { - /* extract the linux cmdline from flash */ - char *name=bse_getflashparam("linuxboot"); - char *x = (char *)0xc0000100; - if (name) { - while (*name) *x++=*name++; - } - *x=0; -} From 26ff6ad9786abf6f40a6d3cbb89753b4fa50cb00 Mon Sep 17 00:00:00 2001 From: Srivatsa Vaddagiri Date: Fri, 16 Sep 2005 19:27:40 -0700 Subject: [PATCH 096/240] [PATCH] CPU hotplug breaks wake_up_new_task Fix a problem wherein a new-born task is added to a dead CPU. Signed-off-by: Srivatsa Vaddagiri Acked-by: Nick Piggin Acked-by: Shaohua Li Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/fork.c b/kernel/fork.c index 8149f3602881..533ce27f4b2c 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1062,7 +1062,8 @@ static task_t *copy_process(unsigned long clone_flags, * parent's CPU). This avoids alot of nasty races. */ p->cpus_allowed = current->cpus_allowed; - if (unlikely(!cpu_isset(task_cpu(p), p->cpus_allowed))) + if (unlikely(!cpu_isset(task_cpu(p), p->cpus_allowed) || + !cpu_online(task_cpu(p)))) set_task_cpu(p, smp_processor_id()); /* From 6add9f7f529b124938d14ae9db5a952d1272075a Mon Sep 17 00:00:00 2001 From: Peter Oberparleiter Date: Fri, 16 Sep 2005 19:27:42 -0700 Subject: [PATCH 097/240] [PATCH] s390: kernel stack corruption When an asynchronous interruption occurs during the execution of the 'critical section' within the generic interruption handling code (entry.S), a faulty check for a userspace PSW may result in a corrupted kernel stack pointer which subsequently triggers a stack overflow check. Signed-off-by: Peter Oberparleiter Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/entry.S | 2 +- arch/s390/kernel/entry64.S | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 58fc7fbcb40e..9b30f4cf32c4 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -108,7 +108,7 @@ STACK_SIZE = 1 << STACK_SHIFT bl BASED(0f) l %r14,BASED(.Lcleanup_critical) basr %r14,%r14 - tm 0(%r12),0x01 # retest problem state after cleanup + tm 1(%r12),0x01 # retest problem state after cleanup bnz BASED(1f) 0: l %r14,__LC_ASYNC_STACK # are we already on the async stack ? slr %r14,%r15 diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index d0c9ffaa25db..7b9b4a2ba1d7 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -101,7 +101,7 @@ _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_MCCK_PENDING) clc \psworg+8(8),BASED(.Lcritical_start) jl 0f brasl %r14,cleanup_critical - tm 0(%r12),0x01 # retest problem state after cleanup + tm 1(%r12),0x01 # retest problem state after cleanup jnz 1f 0: lg %r14,__LC_ASYNC_STACK # are we already on the async. stack ? slgr %r14,%r15 From f6e34c6af6f18bd6c66bfb1c6a7c57068412aa73 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:27:43 -0700 Subject: [PATCH 098/240] [PATCH] uml: _switch_to code consolidation This patch moves code that is in both switch_to_tt and switch_to_skas to the top level _switch_to function, keeping us from duplicating code. It is required for the stack trace patch to work properly. Signed-off-by: Allan Graves Signed-off-by: Jeff Dike Cc: Paolo Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/process_kern.c | 12 ++++++++++-- arch/um/kernel/skas/include/mode_kern-skas.h | 2 +- arch/um/kernel/skas/process_kern.c | 7 +------ arch/um/kernel/tt/include/mode_kern-tt.h | 2 +- arch/um/kernel/tt/process_kern.c | 7 +------ 5 files changed, 14 insertions(+), 16 deletions(-) diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c index c23d8a08d0ff..1143f5e522b2 100644 --- a/arch/um/kernel/process_kern.c +++ b/arch/um/kernel/process_kern.c @@ -113,8 +113,16 @@ void set_current(void *t) void *_switch_to(void *prev, void *next, void *last) { - return(CHOOSE_MODE(switch_to_tt(prev, next), - switch_to_skas(prev, next))); + struct task_struct *from = prev; + struct task_struct *to= next; + + to->thread.prev_sched = from; + set_current(to); + + CHOOSE_MODE_PROC(switch_to_tt, switch_to_skas, prev, next); + + return(current->thread.prev_sched); + } void interrupt_end(void) diff --git a/arch/um/kernel/skas/include/mode_kern-skas.h b/arch/um/kernel/skas/include/mode_kern-skas.h index e48490028111..c97a80dfe370 100644 --- a/arch/um/kernel/skas/include/mode_kern-skas.h +++ b/arch/um/kernel/skas/include/mode_kern-skas.h @@ -11,7 +11,7 @@ #include "asm/ptrace.h" extern void flush_thread_skas(void); -extern void *switch_to_skas(void *prev, void *next); +extern void switch_to_skas(void *prev, void *next); extern void start_thread_skas(struct pt_regs *regs, unsigned long eip, unsigned long esp); extern int copy_thread_skas(int nr, unsigned long clone_flags, diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c index 3d1b227226e6..efe92e8aa2a9 100644 --- a/arch/um/kernel/skas/process_kern.c +++ b/arch/um/kernel/skas/process_kern.c @@ -24,7 +24,7 @@ #include "proc_mm.h" #include "registers.h" -void *switch_to_skas(void *prev, void *next) +void switch_to_skas(void *prev, void *next) { struct task_struct *from, *to; @@ -35,16 +35,11 @@ void *switch_to_skas(void *prev, void *next) if(current->pid == 0) switch_timers(0); - to->thread.prev_sched = from; - set_current(to); - switch_threads(&from->thread.mode.skas.switch_buf, to->thread.mode.skas.switch_buf); if(current->pid == 0) switch_timers(1); - - return(current->thread.prev_sched); } extern void schedule_tail(struct task_struct *prev); diff --git a/arch/um/kernel/tt/include/mode_kern-tt.h b/arch/um/kernel/tt/include/mode_kern-tt.h index e0ca0e0b2516..2a35b15c5fef 100644 --- a/arch/um/kernel/tt/include/mode_kern-tt.h +++ b/arch/um/kernel/tt/include/mode_kern-tt.h @@ -11,7 +11,7 @@ #include "asm/ptrace.h" #include "asm/uaccess.h" -extern void *switch_to_tt(void *prev, void *next); +extern void switch_to_tt(void *prev, void *next); extern void flush_thread_tt(void); extern void start_thread_tt(struct pt_regs *regs, unsigned long eip, unsigned long esp); diff --git a/arch/um/kernel/tt/process_kern.c b/arch/um/kernel/tt/process_kern.c index a189a2b92935..0de05a268b22 100644 --- a/arch/um/kernel/tt/process_kern.c +++ b/arch/um/kernel/tt/process_kern.c @@ -26,7 +26,7 @@ #include "init.h" #include "tt.h" -void *switch_to_tt(void *prev, void *next, void *last) +int switch_to_tt(void *prev, void *next, void *last) { struct task_struct *from, *to, *prev_sched; unsigned long flags; @@ -36,8 +36,6 @@ void *switch_to_tt(void *prev, void *next, void *last) from = prev; to = next; - to->thread.prev_sched = from; - cpu = from->thread_info->cpu; if(cpu == 0) forward_interrupts(to->thread.mode.tt.extern_pid); @@ -53,7 +51,6 @@ void *switch_to_tt(void *prev, void *next, void *last) forward_pending_sigio(to->thread.mode.tt.extern_pid); c = 0; - set_current(to); err = os_write_file(to->thread.mode.tt.switch_pipe[1], &c, sizeof(c)); if(err != sizeof(c)) @@ -85,8 +82,6 @@ void *switch_to_tt(void *prev, void *next, void *last) flush_tlb_all(); local_irq_restore(flags); - - return(current->thread.prev_sched); } void release_thread_tt(struct task_struct *task) From 3eddddcf239c89bbd3c50d1440001a3d384ed40a Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:27:46 -0700 Subject: [PATCH 099/240] [PATCH] uml: breakpoint an arbitrary thread This patch implements a stack trace for a thread, not unlike sysrq-t does. The advantage to this is that a break point can be placed on showreqs, so that upon showing the stack, you jump immediately into the debugger. While sysrq-t does the same thing, sysrq-t shows *all* threads stacks. It also doesn't work right now. In the future, I thought it might be acceptable to make this show all pids stacks, but perhaps leaving well enough alone and just using sysrq-t would be okay. For now, upon receiving the stack command, UML switches context to that thread, dumps its registers, and then switches context back to the original thread. Since UML compacts all threads into one of 4 host threads, this sort of mechanism could be expanded in the future to include other debugging helpers that sysrq does not cover. Note by jdike - The main benefit to this is that it brings an arbitrary thread back into context, where it can be examined by gdb. The fact that it dumps it stack is secondary. This provides the capability to examine a sleeping thread, which has existed in tt mode, but not in skas mode until now. Also, the other threads, that sysrq doesn't cover, can be gdb-ed directly anyway. Signed-off-by: Allan Graves Signed-off-by: Jeff Dike Cc: Paolo Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/mconsole_kern.c | 52 ++++++++++++++++++++++++++++++ arch/um/drivers/mconsole_user.c | 1 + arch/um/include/mconsole.h | 1 + arch/um/kernel/process_kern.c | 9 +++++- include/asm-um/processor-generic.h | 1 + 5 files changed, 63 insertions(+), 1 deletion(-) diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index c190c2414197..12c95368124a 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -32,6 +32,7 @@ #include "os.h" #include "umid.h" #include "irq_kern.h" +#include "choose-mode.h" static int do_unlink_socket(struct notifier_block *notifier, unsigned long what, void *data) @@ -276,6 +277,7 @@ void mconsole_proc(struct mc_request *req) go - continue the UML after a 'stop' \n\ log - make UML enter into the kernel log\n\ proc - returns the contents of the UML's /proc/\n\ + stack - returns the stack of the specified pid\n\ " void mconsole_help(struct mc_request *req) @@ -479,6 +481,56 @@ void mconsole_sysrq(struct mc_request *req) } #endif +/* Mconsole stack trace + * Added by Allan Graves, Jeff Dike + * Dumps a stacks registers to the linux console. + * Usage stack . + */ +void do_stack(struct mc_request *req) +{ + char *ptr = req->request.data; + int pid_requested= -1; + struct task_struct *from = NULL; + struct task_struct *to = NULL; + + /* Would be nice: + * 1) Send showregs output to mconsole. + * 2) Add a way to stack dump all pids. + */ + + ptr += strlen("stack"); + while(isspace(*ptr)) ptr++; + + /* Should really check for multiple pids or reject bad args here */ + /* What do the arguments in mconsole_reply mean? */ + if(sscanf(ptr, "%d", &pid_requested) == 0){ + mconsole_reply(req, "Please specify a pid", 1, 0); + return; + } + + from = current; + to = find_task_by_pid(pid_requested); + + if((to == NULL) || (pid_requested == 0)) { + mconsole_reply(req, "Couldn't find that pid", 1, 0); + return; + } + to->thread.saved_task = current; + + switch_to(from, to, from); + mconsole_reply(req, "Stack Dumped to console and message log", 0, 0); +} + +void mconsole_stack(struct mc_request *req) +{ + /* This command doesn't work in TT mode, so let's check and then + * get out of here + */ + CHOOSE_MODE(mconsole_reply(req, "Sorry, this doesn't work in TT mode", + 1, 0), + do_stack(req)); +} + /* Changed by mconsole_setup, which is __setup, and called before SMP is * active. */ diff --git a/arch/um/drivers/mconsole_user.c b/arch/um/drivers/mconsole_user.c index fe5afb13252c..a5b8aeade1c5 100644 --- a/arch/um/drivers/mconsole_user.c +++ b/arch/um/drivers/mconsole_user.c @@ -30,6 +30,7 @@ static struct mconsole_command commands[] = { { "go", mconsole_go, MCONSOLE_INTR }, { "log", mconsole_log, MCONSOLE_INTR }, { "proc", mconsole_proc, MCONSOLE_PROC }, + { "stack", mconsole_stack, MCONSOLE_INTR }, }; /* Initialized in mconsole_init, which is an initcall */ diff --git a/arch/um/include/mconsole.h b/arch/um/include/mconsole.h index cfa368e045a5..b1b512f47035 100644 --- a/arch/um/include/mconsole.h +++ b/arch/um/include/mconsole.h @@ -81,6 +81,7 @@ extern void mconsole_stop(struct mc_request *req); extern void mconsole_go(struct mc_request *req); extern void mconsole_log(struct mc_request *req); extern void mconsole_proc(struct mc_request *req); +extern void mconsole_stack(struct mc_request *req); extern int mconsole_get_request(int fd, struct mc_request *req); extern int mconsole_notify(char *sock_name, int type, const void *data, diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c index 1143f5e522b2..39cf568ccfaf 100644 --- a/arch/um/kernel/process_kern.c +++ b/arch/um/kernel/process_kern.c @@ -119,7 +119,14 @@ void *_switch_to(void *prev, void *next, void *last) to->thread.prev_sched = from; set_current(to); - CHOOSE_MODE_PROC(switch_to_tt, switch_to_skas, prev, next); + do { + current->thread.saved_task = NULL ; + CHOOSE_MODE_PROC(switch_to_tt, switch_to_skas, prev, next); + if(current->thread.saved_task) + show_regs(&(current->thread.regs)); + next= current->thread.saved_task; + prev= current; + } while(current->thread.saved_task); return(current->thread.prev_sched); diff --git a/include/asm-um/processor-generic.h b/include/asm-um/processor-generic.h index b2fc94fbc2d9..2d242360c3d6 100644 --- a/include/asm-um/processor-generic.h +++ b/include/asm-um/processor-generic.h @@ -21,6 +21,7 @@ struct thread_struct { * copy_thread) to mark that we are begin called from userspace (fork / * vfork / clone), and reset to 0 after. It is left to 0 when called * from kernelspace (i.e. kernel_thread() or fork_idle(), as of 2.6.11). */ + struct task_struct *saved_task; int forking; int nsyscalls; struct pt_regs regs; From edcc2205d33d6baefc773c915b6e7ce517bc0b19 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:27:47 -0700 Subject: [PATCH 100/240] [PATCH] uml: Remove an unused file This removes a file which is no longer used. Signed-off-by: Jeff Dike Cc: Paolo Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/ubd_user.c | 75 -------------------------------------- 1 file changed, 75 deletions(-) delete mode 100644 arch/um/drivers/ubd_user.c diff --git a/arch/um/drivers/ubd_user.c b/arch/um/drivers/ubd_user.c deleted file mode 100644 index b94d2bc4fe06..000000000000 --- a/arch/um/drivers/ubd_user.c +++ /dev/null @@ -1,75 +0,0 @@ -/* - * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) - * Copyright (C) 2001 Ridgerun,Inc (glonnon@ridgerun.com) - * Licensed under the GPL - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "asm/types.h" -#include "user_util.h" -#include "kern_util.h" -#include "user.h" -#include "ubd_user.h" -#include "os.h" -#include "cow.h" - -#include -#include - -void ignore_sigwinch_sig(void) -{ - signal(SIGWINCH, SIG_IGN); -} - -int start_io_thread(unsigned long sp, int *fd_out) -{ - int pid, fds[2], err; - - err = os_pipe(fds, 1, 1); - if(err < 0){ - printk("start_io_thread - os_pipe failed, err = %d\n", -err); - goto out; - } - - kernel_fd = fds[0]; - *fd_out = fds[1]; - - pid = clone(io_thread, (void *) sp, CLONE_FILES | CLONE_VM | SIGCHLD, - NULL); - if(pid < 0){ - printk("start_io_thread - clone failed : errno = %d\n", errno); - err = -errno; - goto out_close; - } - - return(pid); - - out_close: - os_close_file(fds[0]); - os_close_file(fds[1]); - kernel_fd = -1; - *fd_out = -1; - out: - return(err); -} - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ From d9f0d9151083a9a77331a020a7fee498cf828e3b Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:27:47 -0700 Subject: [PATCH 101/240] [PATCH] uml: Remove a useless include linux/inet.h isn't needed, and on my system, is empty. Signed-off-by: Jeff Dike Cc: Paolo Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/mcast_user.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/um/drivers/mcast_user.c b/arch/um/drivers/mcast_user.c index 7a0d115b29d0..3fd69067a5f5 100644 --- a/arch/um/drivers/mcast_user.c +++ b/arch/um/drivers/mcast_user.c @@ -13,7 +13,6 @@ #include #include -#include #include #include #include From 64b7673f91c0c3614028c5942b0d6a91d0b64a98 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:27:48 -0700 Subject: [PATCH 102/240] [PATCH] uml: Remove some build warnings These ugly double-casts are the result of gdb complaining about size Signed-off-by: Jeff Dike Cc: Paolo Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/os-Linux/aio.c | 2 +- arch/um/os-Linux/elf_aux.c | 4 +++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/um/os-Linux/aio.c b/arch/um/os-Linux/aio.c index b04897cd995d..182905be869e 100644 --- a/arch/um/os-Linux/aio.c +++ b/arch/um/os-Linux/aio.c @@ -142,7 +142,7 @@ static int aio_thread(void *arg) "errno = %d\n", errno); } else { - aio = (struct aio_context *) event.data; + aio = (struct aio_context *) (long) event.data; if(update_aio(aio, event.res)){ do_aio(ctx, aio); continue; diff --git a/arch/um/os-Linux/elf_aux.c b/arch/um/os-Linux/elf_aux.c index 1399520a8588..5c02e3395225 100644 --- a/arch/um/os-Linux/elf_aux.c +++ b/arch/um/os-Linux/elf_aux.c @@ -9,6 +9,7 @@ */ #include #include +#include #include "init.h" #include "elf_user.h" #include "mem_user.h" @@ -54,7 +55,8 @@ __init void scan_elf_aux( char **envp) * a_un, so we have to use a_val, which is * all that's left. */ - elf_aux_platform = (char *) auxv->a_un.a_val; + elf_aux_platform = + (char *) (long) auxv->a_un.a_val; break; case AT_PAGESZ: page_size = auxv->a_un.a_val; From b4fd310e163477236a241580b3b8c29aee65f4cc Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:27:49 -0700 Subject: [PATCH 103/240] [PATCH] uml: preserve errno in error paths The poster child for this patch is the third tuntap_user hunk. When an ioctl fails, it properly closes the opened file descriptor and returns. However, the close resets errno to 0, and the 'return errno' that follows returns 0 rather than the value that ioctl set. This caused the caller to believe that the device open succeeded and had opened file descriptor 0, which caused no end of interesting behavior. The rest of this patch is a pass through the UML sources looking for places where errno could be reset before being passed back out. A common culprit is printk, which could call write, being called before errno is returned. In some cases, where the code ends up being much smaller, I just deleted the printk. There was another case where a caller of run_helper looked at errno after a failure, rather than the return value of run_helper, which was the errno value that it wanted. Signed-off-by: Jeff Dike Cc: Paolo Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/mcast_user.c | 11 +++- arch/um/drivers/mconsole_user.c | 6 +- arch/um/drivers/pty.c | 3 +- arch/um/drivers/xterm.c | 6 +- arch/um/kernel/helper.c | 12 ++-- arch/um/kernel/user_util.c | 12 ++-- arch/um/os-Linux/aio.c | 5 +- arch/um/os-Linux/drivers/tuntap_user.c | 8 ++- arch/um/os-Linux/file.c | 82 ++++++++++---------------- 9 files changed, 67 insertions(+), 78 deletions(-) diff --git a/arch/um/drivers/mcast_user.c b/arch/um/drivers/mcast_user.c index 3fd69067a5f5..5db136e2651c 100644 --- a/arch/um/drivers/mcast_user.c +++ b/arch/um/drivers/mcast_user.c @@ -54,7 +54,7 @@ static int mcast_open(void *data) struct mcast_data *pri = data; struct sockaddr_in *sin = pri->mcast_addr; struct ip_mreq mreq; - int fd = -EINVAL, yes = 1, err = -EINVAL;; + int fd, yes = 1, err = 0; if ((sin->sin_addr.s_addr == 0) || (sin->sin_port == 0)) @@ -65,13 +65,14 @@ static int mcast_open(void *data) if (fd < 0){ printk("mcast_open : data socket failed, errno = %d\n", errno); - fd = -errno; + err = -errno; goto out; } if (setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, &yes, sizeof(yes)) < 0) { printk("mcast_open: SO_REUSEADDR failed, errno = %d\n", errno); + err = -errno; goto out_close; } @@ -80,6 +81,7 @@ static int mcast_open(void *data) sizeof(pri->ttl)) < 0) { printk("mcast_open: IP_MULTICAST_TTL failed, error = %d\n", errno); + err = -errno; goto out_close; } @@ -87,12 +89,14 @@ static int mcast_open(void *data) if (setsockopt(fd, SOL_IP, IP_MULTICAST_LOOP, &yes, sizeof(yes)) < 0) { printk("mcast_open: IP_MULTICAST_LOOP failed, error = %d\n", errno); + err = -errno; goto out_close; } /* bind socket to mcast address */ if (bind(fd, (struct sockaddr *) sin, sizeof(*sin)) < 0) { printk("mcast_open : data bind failed, errno = %d\n", errno); + err = -errno; goto out_close; } @@ -107,14 +111,15 @@ static int mcast_open(void *data) "interface on the host.\n"); printk("eth0 should be configured in order to use the " "multicast transport.\n"); + err = -errno; goto out_close; } - out: return fd; out_close: os_close_file(fd); + out: return err; } diff --git a/arch/um/drivers/mconsole_user.c b/arch/um/drivers/mconsole_user.c index a5b8aeade1c5..310c1f823f26 100644 --- a/arch/um/drivers/mconsole_user.c +++ b/arch/um/drivers/mconsole_user.c @@ -173,9 +173,9 @@ int mconsole_notify(char *sock_name, int type, const void *data, int len) if(notify_sock < 0){ notify_sock = socket(PF_UNIX, SOCK_DGRAM, 0); if(notify_sock < 0){ - printk("mconsole_notify - socket failed, errno = %d\n", - errno); err = -errno; + printk("mconsole_notify - socket failed, errno = %d\n", + err); } } unlock_notify(); @@ -198,8 +198,8 @@ int mconsole_notify(char *sock_name, int type, const void *data, int len) n = sendto(notify_sock, &packet, len, 0, (struct sockaddr *) &target, sizeof(target)); if(n < 0){ - printk("mconsole_notify - sendto failed, errno = %d\n", errno); err = -errno; + printk("mconsole_notify - sendto failed, errno = %d\n", errno); } return(err); } diff --git a/arch/um/drivers/pty.c b/arch/um/drivers/pty.c index ed84d01df6cc..0306a1b215b7 100644 --- a/arch/um/drivers/pty.c +++ b/arch/um/drivers/pty.c @@ -43,8 +43,9 @@ static int pts_open(int input, int output, int primary, void *d, fd = get_pty(); if(fd < 0){ + err = -errno; printk("open_pts : Failed to open pts\n"); - return(-errno); + return err; } if(data->raw){ CATCH_EINTR(err = tcgetattr(fd, &data->tt)); diff --git a/arch/um/drivers/xterm.c b/arch/um/drivers/xterm.c index 93dc1911363f..90e0e5ff451e 100644 --- a/arch/um/drivers/xterm.c +++ b/arch/um/drivers/xterm.c @@ -110,13 +110,15 @@ int xterm_open(int input, int output, int primary, void *d, fd = mkstemp(file); if(fd < 0){ + err = -errno; printk("xterm_open : mkstemp failed, errno = %d\n", errno); - return(-errno); + return err; } if(unlink(file)){ + err = -errno; printk("xterm_open : unlink failed, errno = %d\n", errno); - return(-errno); + return err; } os_close_file(fd); diff --git a/arch/um/kernel/helper.c b/arch/um/kernel/helper.c index f83e1e8e2392..33fb0bd3b11a 100644 --- a/arch/um/kernel/helper.c +++ b/arch/um/kernel/helper.c @@ -85,8 +85,8 @@ int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv, data.fd = fds[1]; pid = clone(helper_child, (void *) sp, CLONE_VM | SIGCHLD, &data); if(pid < 0){ - printk("run_helper : clone failed, errno = %d\n", errno); ret = -errno; + printk("run_helper : clone failed, errno = %d\n", errno); goto out_close; } @@ -122,7 +122,7 @@ int run_helper_thread(int (*proc)(void *), void *arg, unsigned int flags, unsigned long *stack_out, int stack_order) { unsigned long stack, sp; - int pid, status; + int pid, status, err; stack = alloc_stack(stack_order, um_in_interrupt()); if(stack == 0) return(-ENOMEM); @@ -130,16 +130,18 @@ int run_helper_thread(int (*proc)(void *), void *arg, unsigned int flags, sp = stack + (page_size() << stack_order) - sizeof(void *); pid = clone(proc, (void *) sp, flags | SIGCHLD, arg); if(pid < 0){ + err = -errno; printk("run_helper_thread : clone failed, errno = %d\n", errno); - return(-errno); + return err; } if(stack_out == NULL){ CATCH_EINTR(pid = waitpid(pid, &status, 0)); if(pid < 0){ + err = -errno; printk("run_helper_thread - wait failed, errno = %d\n", errno); - pid = -errno; + pid = err; } if(!WIFEXITED(status) || (WEXITSTATUS(status) != 0)) printk("run_helper_thread - thread returned status " @@ -156,8 +158,8 @@ int helper_wait(int pid) CATCH_EINTR(ret = waitpid(pid, NULL, WNOHANG)); if(ret < 0){ + ret = -errno; printk("helper_wait : waitpid failed, errno = %d\n", errno); - return(-errno); } return(ret); } diff --git a/arch/um/kernel/user_util.c b/arch/um/kernel/user_util.c index 954ff67cc8b3..a25f3ea11fd7 100644 --- a/arch/um/kernel/user_util.c +++ b/arch/um/kernel/user_util.c @@ -109,18 +109,14 @@ int raw(int fd) int err; CATCH_EINTR(err = tcgetattr(fd, &tt)); - if (err < 0) { - printk("tcgetattr failed, errno = %d\n", errno); - return(-errno); - } + if(err < 0) + return -errno; cfmakeraw(&tt); CATCH_EINTR(err = tcsetattr(fd, TCSADRAIN, &tt)); - if (err < 0) { - printk("tcsetattr failed, errno = %d\n", errno); - return(-errno); - } + if(err < 0) + return -errno; /* XXX tcsetattr could have applied only some changes * (and cfmakeraw() is a set of changes) */ diff --git a/arch/um/os-Linux/aio.c b/arch/um/os-Linux/aio.c index 182905be869e..e942beb4e108 100644 --- a/arch/um/os-Linux/aio.c +++ b/arch/um/os-Linux/aio.c @@ -313,15 +313,16 @@ static int init_aio_26(void) int err; if(io_setup(256, &ctx)){ + err = -errno; printk("aio_thread failed to initialize context, err = %d\n", errno); - return -errno; + return err; } err = run_helper_thread(aio_thread, NULL, CLONE_FILES | CLONE_VM | SIGCHLD, &stack, 0); if(err < 0) - return -errno; + return err; aio_pid = err; diff --git a/arch/um/os-Linux/drivers/tuntap_user.c b/arch/um/os-Linux/drivers/tuntap_user.c index 4b83c6c3f48d..4ba9b17adf13 100644 --- a/arch/um/os-Linux/drivers/tuntap_user.c +++ b/arch/um/os-Linux/drivers/tuntap_user.c @@ -75,7 +75,7 @@ static int tuntap_open_tramp(char *gate, int *fd_out, int me, int remote, struct msghdr msg; struct cmsghdr *cmsg; struct iovec iov; - int pid, n; + int pid, n, err; sprintf(version_buf, "%d", UML_NET_VERSION); @@ -105,9 +105,10 @@ static int tuntap_open_tramp(char *gate, int *fd_out, int me, int remote, n = recvmsg(me, &msg, 0); *used_out = n; if(n < 0){ + err = -errno; printk("tuntap_open_tramp : recvmsg failed - errno = %d\n", errno); - return(-errno); + return err; } CATCH_EINTR(waitpid(pid, NULL, 0)); @@ -147,9 +148,10 @@ static int tuntap_open(void *data) ifr.ifr_flags = IFF_TAP | IFF_NO_PI; strlcpy(ifr.ifr_name, pri->dev_name, sizeof(ifr.ifr_name)); if(ioctl(pri->fd, TUNSETIFF, (void *) &ifr) < 0){ + err = -errno; printk("TUNSETIFF failed, errno = %d\n", errno); os_close_file(pri->fd); - return(-errno); + return err; } } else { diff --git a/arch/um/os-Linux/file.c b/arch/um/os-Linux/file.c index fd45bb260907..f55773c819e6 100644 --- a/arch/um/os-Linux/file.c +++ b/arch/um/os-Linux/file.c @@ -119,15 +119,11 @@ int os_window_size(int fd, int *rows, int *cols) int os_new_tty_pgrp(int fd, int pid) { - if(ioctl(fd, TIOCSCTTY, 0) < 0){ - printk("TIOCSCTTY failed, errno = %d\n", errno); - return(-errno); - } + if(ioctl(fd, TIOCSCTTY, 0) < 0) + return -errno; - if(tcsetpgrp(fd, pid) < 0){ - printk("tcsetpgrp failed, errno = %d\n", errno); - return(-errno); - } + if(tcsetpgrp(fd, pid) < 0) + return -errno; return(0); } @@ -146,18 +142,12 @@ int os_set_slip(int fd) int disc, sencap; disc = N_SLIP; - if(ioctl(fd, TIOCSETD, &disc) < 0){ - printk("Failed to set slip line discipline - " - "errno = %d\n", errno); - return(-errno); - } + if(ioctl(fd, TIOCSETD, &disc) < 0) + return -errno; sencap = 0; - if(ioctl(fd, SIOCSIFENCAP, &sencap) < 0){ - printk("Failed to set slip encapsulation - " - "errno = %d\n", errno); - return(-errno); - } + if(ioctl(fd, SIOCSIFENCAP, &sencap) < 0) + return -errno; return(0); } @@ -180,22 +170,15 @@ int os_sigio_async(int master, int slave) int flags; flags = fcntl(master, F_GETFL); - if(flags < 0) { - printk("fcntl F_GETFL failed, errno = %d\n", errno); - return(-errno); - } + if(flags < 0) + return errno; if((fcntl(master, F_SETFL, flags | O_NONBLOCK | O_ASYNC) < 0) || - (fcntl(master, F_SETOWN, os_getpid()) < 0)){ - printk("fcntl F_SETFL or F_SETOWN failed, errno = %d\n", - errno); - return(-errno); - } + (fcntl(master, F_SETOWN, os_getpid()) < 0)) + return -errno; - if((fcntl(slave, F_SETFL, flags | O_NONBLOCK) < 0)){ - printk("fcntl F_SETFL failed, errno = %d\n", errno); - return(-errno); - } + if((fcntl(slave, F_SETFL, flags | O_NONBLOCK) < 0)) + return -errno; return(0); } @@ -255,7 +238,7 @@ int os_file_mode(char *file, struct openflags *mode_out) int os_open_file(char *file, struct openflags flags, int mode) { - int fd, f = 0; + int fd, err, f = 0; if(flags.r && flags.w) f = O_RDWR; else if(flags.r) f = O_RDONLY; @@ -272,8 +255,9 @@ int os_open_file(char *file, struct openflags flags, int mode) return(-errno); if(flags.cl && fcntl(fd, F_SETFD, 1)){ + err = -errno; os_close_file(fd); - return(-errno); + return err; } return(fd); @@ -383,9 +367,9 @@ int os_file_size(char *file, unsigned long long *size_out) return(fd); } if(ioctl(fd, BLKGETSIZE, &blocks) < 0){ + err = -errno; printk("Couldn't get the block size of \"%s\", " "errno = %d\n", file, errno); - err = -errno; os_close_file(fd); return(err); } @@ -473,11 +457,14 @@ int os_pipe(int *fds, int stream, int close_on_exec) int os_set_fd_async(int fd, int owner) { + int err; + /* XXX This should do F_GETFL first */ if(fcntl(fd, F_SETFL, O_ASYNC | O_NONBLOCK) < 0){ + err = -errno; printk("os_set_fd_async : failed to set O_ASYNC and " "O_NONBLOCK on fd # %d, errno = %d\n", fd, errno); - return(-errno); + return err; } #ifdef notdef if(fcntl(fd, F_SETFD, 1) < 0){ @@ -488,10 +475,11 @@ int os_set_fd_async(int fd, int owner) if((fcntl(fd, F_SETSIG, SIGIO) < 0) || (fcntl(fd, F_SETOWN, owner) < 0)){ + err = -errno; printk("os_set_fd_async : Failed to fcntl F_SETOWN " "(or F_SETSIG) fd %d to pid %d, errno = %d\n", fd, owner, errno); - return(-errno); + return err; } return(0); @@ -516,11 +504,9 @@ int os_set_fd_block(int fd, int blocking) if(blocking) flags &= ~O_NONBLOCK; else flags |= O_NONBLOCK; - if(fcntl(fd, F_SETFL, flags) < 0){ - printk("Failed to change blocking on fd # %d, errno = %d\n", - fd, errno); - return(-errno); - } + if(fcntl(fd, F_SETFL, flags) < 0) + return -errno; + return(0); } @@ -609,11 +595,8 @@ int os_create_unix_socket(char *file, int len, int close_on_exec) int sock, err; sock = socket(PF_UNIX, SOCK_DGRAM, 0); - if (sock < 0){ - printk("create_unix_socket - socket failed, errno = %d\n", - errno); - return(-errno); - } + if(sock < 0) + return -errno; if(close_on_exec) { err = os_set_exec_close(sock, 1); @@ -628,11 +611,8 @@ int os_create_unix_socket(char *file, int len, int close_on_exec) snprintf(addr.sun_path, len, "%s", file); err = bind(sock, (struct sockaddr *) &addr, sizeof(addr)); - if (err < 0){ - printk("create_listening_socket at '%s' - bind failed, " - "errno = %d\n", file, errno); - return(-errno); - } + if(err < 0) + return -errno; return(sock); } From 0f80bc85c587e8fdeecece4f294a47eca4922ea2 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:27:50 -0700 Subject: [PATCH 104/240] [PATCH] uml: move libc code out of mem_user.c and tempfile.c The serial UML OS-abstraction layer patch (um/kernel dir). This moves all system calls from mem_user.c and tempfile.c files under os-Linux dir. Signed-off-by: Gennady Sharapov Signed-off-by: Jeff Dike Cc: Paolo Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/mem_user.h | 15 --- arch/um/include/os.h | 5 + arch/um/kernel/Makefile | 6 +- arch/um/kernel/init_task.c | 5 +- arch/um/kernel/mem_user.c | 205 +---------------------------------- arch/um/kernel/tempfile.c | 82 -------------- arch/um/kernel/tlb.c | 2 +- arch/um/kernel/tt/mem_user.c | 1 + arch/um/kernel/um_arch.c | 5 - arch/um/os-Linux/Makefile | 8 +- arch/um/os-Linux/mem.c | 161 +++++++++++++++++++++++++++ arch/um/os-Linux/start_up.c | 109 ++++++++++++++++++- arch/um/os-Linux/tt.c | 14 +++ 13 files changed, 300 insertions(+), 318 deletions(-) delete mode 100644 arch/um/kernel/tempfile.c create mode 100644 arch/um/os-Linux/mem.c diff --git a/arch/um/include/mem_user.h b/arch/um/include/mem_user.h index d6404bb64662..9fef4123a65a 100644 --- a/arch/um/include/mem_user.h +++ b/arch/um/include/mem_user.h @@ -51,7 +51,6 @@ extern unsigned long task_size; extern void check_devanon(void); extern int init_mem_user(void); -extern int create_mem_file(unsigned long len); extern void setup_memory(void *entry); extern unsigned long find_iomem(char *driver, unsigned long *len_out); extern int init_maps(unsigned long physmem, unsigned long iomem, @@ -64,20 +63,6 @@ extern unsigned long phys_offset(unsigned long phys); extern void unmap_physmem(void); extern void map_memory(unsigned long virt, unsigned long phys, unsigned long len, int r, int w, int x); -extern int protect_memory(unsigned long addr, unsigned long len, - int r, int w, int x, int must_succeed); extern unsigned long get_kmem_end(void); -extern void check_tmpexec(void); #endif - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/include/os.h b/arch/um/include/os.h index 4c362458052c..583329d0a539 100644 --- a/arch/um/include/os.h +++ b/arch/um/include/os.h @@ -157,6 +157,9 @@ extern int os_lock_file(int fd, int excl); extern void os_early_checks(void); extern int can_do_skas(void); +/* mem.c */ +extern int create_mem_file(unsigned long len); + /* process.c */ extern unsigned long os_process_pc(int pid); extern int os_process_parent(int pid); @@ -181,6 +184,8 @@ extern unsigned long long os_usecs(void); /* tt.c * for tt mode only (will be deleted in future...) */ +extern int protect_memory(unsigned long addr, unsigned long len, + int r, int w, int x, int must_succeed); extern void forward_pending_sigio(int target); extern int start_fork_tramp(void *arg, unsigned long temp_stack, int clone_flags, int (*tramp)(void *)); diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile index 614b8ebeb0ed..7eb52c5a7cab 100644 --- a/arch/um/kernel/Makefile +++ b/arch/um/kernel/Makefile @@ -10,7 +10,7 @@ obj-y = config.o exec_kern.o exitcode.o \ helper.o init_task.o irq.o irq_user.o ksyms.o main.o mem.o mem_user.o \ physmem.o process_kern.o ptrace.o reboot.o resource.o sigio_user.o \ sigio_kern.o signal_kern.o signal_user.o smp.o syscall_kern.o sysrq.o \ - tempfile.o time.o time_kern.o tlb.o trap_kern.o trap_user.o \ + time.o time_kern.o tlb.o trap_kern.o trap_user.o \ uaccess_user.o um_arch.o umid.o user_util.o obj-$(CONFIG_BLK_DEV_INITRD) += initrd.o @@ -24,8 +24,8 @@ obj-$(CONFIG_MODE_SKAS) += skas/ user-objs-$(CONFIG_TTY_LOG) += tty_log.o -USER_OBJS := $(user-objs-y) config.o helper.o main.o tempfile.o time.o \ - tty_log.o umid.o user_util.o +USER_OBJS := $(user-objs-y) config.o helper.o main.o time.o tty_log.o umid.o \ + user_util.o include arch/um/scripts/Makefile.rules diff --git a/arch/um/kernel/init_task.c b/arch/um/kernel/init_task.c index cd7c85be0a1b..49ed5ddf0704 100644 --- a/arch/um/kernel/init_task.c +++ b/arch/um/kernel/init_task.c @@ -13,6 +13,7 @@ #include "asm/pgtable.h" #include "user_util.h" #include "mem_user.h" +#include "os.h" static struct fs_struct init_fs = INIT_FS; struct mm_struct init_mm = INIT_MM(init_mm); @@ -45,8 +46,8 @@ __attribute__((__section__(".data.init_task"))) = void unprotect_stack(unsigned long stack) { - protect_memory(stack, (1 << CONFIG_KERNEL_STACK_ORDER) * PAGE_SIZE, - 1, 1, 0, 1); + os_protect_memory((void *) stack, (1 << CONFIG_KERNEL_STACK_ORDER) * PAGE_SIZE, + 1, 1, 0); } /* diff --git a/arch/um/kernel/mem_user.c b/arch/um/kernel/mem_user.c index 4a663fd434bb..d49640d9923f 100644 --- a/arch/um/kernel/mem_user.c +++ b/arch/um/kernel/mem_user.c @@ -48,219 +48,16 @@ #include "tempfile.h" #include "kern_constants.h" -#define TEMPNAME_TEMPLATE "vm_file-XXXXXX" - -static int create_tmp_file(unsigned long len) -{ - int fd, err; - char zero; - - fd = make_tempfile(TEMPNAME_TEMPLATE, NULL, 1); - if(fd < 0) { - os_print_error(fd, "make_tempfile"); - exit(1); - } - - err = os_mode_fd(fd, 0777); - if(err < 0){ - os_print_error(err, "os_mode_fd"); - exit(1); - } - err = os_seek_file(fd, len); - if(err < 0){ - os_print_error(err, "os_seek_file"); - exit(1); - } - zero = 0; - err = os_write_file(fd, &zero, 1); - if(err != 1){ - os_print_error(err, "os_write_file"); - exit(1); - } - - return(fd); -} - -void check_tmpexec(void) -{ - void *addr; - int err, fd = create_tmp_file(UM_KERN_PAGE_SIZE); - - addr = mmap(NULL, UM_KERN_PAGE_SIZE, - PROT_READ | PROT_WRITE | PROT_EXEC, MAP_PRIVATE, fd, 0); - printf("Checking PROT_EXEC mmap in /tmp..."); - fflush(stdout); - if(addr == MAP_FAILED){ - err = errno; - perror("failed"); - if(err == EPERM) - printf("/tmp must be not mounted noexec\n"); - exit(1); - } - printf("OK\n"); - munmap(addr, UM_KERN_PAGE_SIZE); - - os_close_file(fd); -} - -static int have_devanon = 0; - -void check_devanon(void) -{ - int fd; - - printk("Checking for /dev/anon on the host..."); - fd = open("/dev/anon", O_RDWR); - if(fd < 0){ - printk("Not available (open failed with errno %d)\n", errno); - return; - } - - printk("OK\n"); - have_devanon = 1; -} - -static int create_anon_file(unsigned long len) -{ - void *addr; - int fd; - - fd = open("/dev/anon", O_RDWR); - if(fd < 0) { - os_print_error(fd, "opening /dev/anon"); - exit(1); - } - - addr = mmap(NULL, len, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0); - if(addr == MAP_FAILED){ - perror("mapping physmem file"); - exit(1); - } - munmap(addr, len); - - return(fd); -} - -int create_mem_file(unsigned long len) -{ - int err, fd; - - if(have_devanon) - fd = create_anon_file(len); - else fd = create_tmp_file(len); - - err = os_set_exec_close(fd, 1); - if(err < 0) - os_print_error(err, "exec_close"); - return(fd); -} - struct iomem_region *iomem_regions = NULL; int iomem_size = 0; -static int __init parse_iomem(char *str, int *add) -{ - struct iomem_region *new; - struct uml_stat buf; - char *file, *driver; - int fd, err, size; - - driver = str; - file = strchr(str,','); - if(file == NULL){ - printf("parse_iomem : failed to parse iomem\n"); - goto out; - } - *file = '\0'; - file++; - fd = os_open_file(file, of_rdwr(OPENFLAGS()), 0); - if(fd < 0){ - os_print_error(fd, "parse_iomem - Couldn't open io file"); - goto out; - } - - err = os_stat_fd(fd, &buf); - if(err < 0){ - os_print_error(err, "parse_iomem - cannot stat_fd file"); - goto out_close; - } - - new = malloc(sizeof(*new)); - if(new == NULL){ - perror("Couldn't allocate iomem_region struct"); - goto out_close; - } - - size = (buf.ust_size + UM_KERN_PAGE_SIZE) & ~(UM_KERN_PAGE_SIZE - 1); - - *new = ((struct iomem_region) { .next = iomem_regions, - .driver = driver, - .fd = fd, - .size = size, - .phys = 0, - .virt = 0 }); - iomem_regions = new; - iomem_size += new->size + UM_KERN_PAGE_SIZE; - - return(0); - out_close: - os_close_file(fd); - out: - return(1); -} +extern int parse_iomem(char *str, int *add) __init; __uml_setup("iomem=", parse_iomem, "iomem=,\n" " Configure as an IO memory region named .\n\n" ); -int protect_memory(unsigned long addr, unsigned long len, int r, int w, int x, - int must_succeed) -{ - int err; - - err = os_protect_memory((void *) addr, len, r, w, x); - if(err < 0){ - if(must_succeed) - panic("protect failed, err = %d", -err); - else return(err); - } - return(0); -} - -#if 0 -/* Debugging facility for dumping stuff out to the host, avoiding the timing - * problems that come with printf and breakpoints. - * Enable in case of emergency. - */ - -int logging = 1; -int logging_fd = -1; - -int logging_line = 0; -char logging_buf[512]; - -void log(char *fmt, ...) -{ - va_list ap; - struct timeval tv; - struct openflags flags; - - if(logging == 0) return; - if(logging_fd < 0){ - flags = of_create(of_trunc(of_rdwr(OPENFLAGS()))); - logging_fd = os_open_file("log", flags, 0644); - } - gettimeofday(&tv, NULL); - sprintf(logging_buf, "%d\t %u.%u ", logging_line++, tv.tv_sec, - tv.tv_usec); - va_start(ap, fmt); - vsprintf(&logging_buf[strlen(logging_buf)], fmt, ap); - va_end(ap); - write(logging_fd, logging_buf, strlen(logging_buf)); -} -#endif - /* * Overrides for Emacs so that we follow Linus's tabbing style. * Emacs will notice this stuff at the end of the file and automatically diff --git a/arch/um/kernel/tempfile.c b/arch/um/kernel/tempfile.c deleted file mode 100644 index b1674bc1395d..000000000000 --- a/arch/um/kernel/tempfile.c +++ /dev/null @@ -1,82 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#include -#include -#include -#include -#include -#include -#include "init.h" - -/* Modified from create_mem_file and start_debugger */ -static char *tempdir = NULL; - -static void __init find_tempdir(void) -{ - char *dirs[] = { "TMP", "TEMP", "TMPDIR", NULL }; - int i; - char *dir = NULL; - - if(tempdir != NULL) return; /* We've already been called */ - for(i = 0; dirs[i]; i++){ - dir = getenv(dirs[i]); - if((dir != NULL) && (*dir != '\0')) - break; - } - if((dir == NULL) || (*dir == '\0')) - dir = "/tmp"; - - tempdir = malloc(strlen(dir) + 2); - if(tempdir == NULL){ - fprintf(stderr, "Failed to malloc tempdir, " - "errno = %d\n", errno); - return; - } - strcpy(tempdir, dir); - strcat(tempdir, "/"); -} - -int make_tempfile(const char *template, char **out_tempname, int do_unlink) -{ - char tempname[MAXPATHLEN]; - int fd; - - find_tempdir(); - if (*template != '/') - strcpy(tempname, tempdir); - else - *tempname = 0; - strcat(tempname, template); - fd = mkstemp(tempname); - if(fd < 0){ - fprintf(stderr, "open - cannot create %s: %s\n", tempname, - strerror(errno)); - return -1; - } - if(do_unlink && (unlink(tempname) < 0)){ - perror("unlink"); - return -1; - } - if(out_tempname){ - *out_tempname = strdup(tempname); - if(*out_tempname == NULL){ - perror("strdup"); - return -1; - } - } - return(fd); -} - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c index 80ed6188e8a2..0a562c3c0fd8 100644 --- a/arch/um/kernel/tlb.c +++ b/arch/um/kernel/tlb.c @@ -307,7 +307,7 @@ int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) } else if(pte_newprot(*pte)){ updated = 1; - protect_memory(addr, PAGE_SIZE, 1, 1, 1, 1); + os_protect_memory((void *) addr, PAGE_SIZE, 1, 1, 1); } addr += PAGE_SIZE; } diff --git a/arch/um/kernel/tt/mem_user.c b/arch/um/kernel/tt/mem_user.c index 3085267459b1..03e589895388 100644 --- a/arch/um/kernel/tt/mem_user.c +++ b/arch/um/kernel/tt/mem_user.c @@ -12,6 +12,7 @@ #include "tt.h" #include "mem_user.h" #include "user_util.h" +#include "os.h" void remap_data(void *segment_start, void *segment_end, int w) { diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c index 09f6f7ce4695..f0a275947d34 100644 --- a/arch/um/kernel/um_arch.c +++ b/arch/um/kernel/um_arch.c @@ -361,11 +361,6 @@ int linux_main(int argc, char **argv) uml_start = CHOOSE_MODE_PROC(set_task_sizes_tt, set_task_sizes_skas, 0, &host_task_size, &task_size); - /* Need to check this early because mmapping happens before the - * kernel is running. - */ - check_tmpexec(); - brk_start = (unsigned long) sbrk(0); CHOOSE_MODE_PROC(before_mem_tt, before_mem_skas, brk_start); /* Increase physical memory size for exec-shield users diff --git a/arch/um/os-Linux/Makefile b/arch/um/os-Linux/Makefile index 7a1662419c0c..d15ec2af6a22 100644 --- a/arch/um/os-Linux/Makefile +++ b/arch/um/os-Linux/Makefile @@ -3,11 +3,11 @@ # Licensed under the GPL # -obj-y = aio.o elf_aux.o file.o process.o signal.o start_up.o time.o tt.o \ - tty.o user_syms.o drivers/ sys-$(SUBARCH)/ +obj-y = aio.o elf_aux.o file.o mem.o process.o signal.o start_up.o time.o \ + tt.o tty.o user_syms.o drivers/ sys-$(SUBARCH)/ -USER_OBJS := aio.o elf_aux.o file.o process.o signal.o start_up.o time.o tt.o \ - tty.o +USER_OBJS := aio.o elf_aux.o file.o mem.o process.o signal.o start_up.o \ + time.o tt.o tty.o elf_aux.o: $(ARCH_DIR)/kernel-offsets.h CFLAGS_elf_aux.o += -I$(objtree)/arch/um diff --git a/arch/um/os-Linux/mem.c b/arch/um/os-Linux/mem.c new file mode 100644 index 000000000000..8e71edaaf80b --- /dev/null +++ b/arch/um/os-Linux/mem.c @@ -0,0 +1,161 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "kern_util.h" +#include "user.h" +#include "user_util.h" +#include "mem_user.h" +#include "init.h" +#include "os.h" +#include "tempfile.h" +#include "kern_constants.h" + +#include + +static char *tempdir = NULL; + +static void __init find_tempdir(void) +{ + char *dirs[] = { "TMP", "TEMP", "TMPDIR", NULL }; + int i; + char *dir = NULL; + + if(tempdir != NULL) return; /* We've already been called */ + for(i = 0; dirs[i]; i++){ + dir = getenv(dirs[i]); + if((dir != NULL) && (*dir != '\0')) + break; + } + if((dir == NULL) || (*dir == '\0')) + dir = "/tmp"; + + tempdir = malloc(strlen(dir) + 2); + if(tempdir == NULL){ + fprintf(stderr, "Failed to malloc tempdir, " + "errno = %d\n", errno); + return; + } + strcpy(tempdir, dir); + strcat(tempdir, "/"); +} + +/* + * This proc still used in tt-mode + * (file: kernel/tt/ptproxy/proxy.c, proc: start_debugger). + * So it isn't 'static' yet. + */ +int make_tempfile(const char *template, char **out_tempname, int do_unlink) +{ + char tempname[MAXPATHLEN]; + int fd; + + find_tempdir(); + if (*template != '/') + strcpy(tempname, tempdir); + else + *tempname = 0; + strcat(tempname, template); + fd = mkstemp(tempname); + if(fd < 0){ + fprintf(stderr, "open - cannot create %s: %s\n", tempname, + strerror(errno)); + return -1; + } + if(do_unlink && (unlink(tempname) < 0)){ + perror("unlink"); + return -1; + } + if(out_tempname){ + *out_tempname = strdup(tempname); + if(*out_tempname == NULL){ + perror("strdup"); + return -1; + } + } + return(fd); +} + +#define TEMPNAME_TEMPLATE "vm_file-XXXXXX" + +/* + * This proc is used in start_up.c + * So it isn't 'static'. + */ +int create_tmp_file(unsigned long len) +{ + int fd, err; + char zero; + + fd = make_tempfile(TEMPNAME_TEMPLATE, NULL, 1); + if(fd < 0) { + exit(1); + } + + err = fchmod(fd, 0777); + if(err < 0){ + perror("os_mode_fd"); + exit(1); + } + + if (lseek64(fd, len, SEEK_SET) < 0) { + perror("os_seek_file"); + exit(1); + } + + zero = 0; + + err = os_write_file(fd, &zero, 1); + if(err != 1){ + errno = -err; + perror("os_write_file"); + exit(1); + } + + return(fd); +} + +static int create_anon_file(unsigned long len) +{ + void *addr; + int fd; + + fd = open("/dev/anon", O_RDWR); + if(fd < 0) { + perror("opening /dev/anon"); + exit(1); + } + + addr = mmap(NULL, len, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0); + if(addr == MAP_FAILED){ + perror("mapping physmem file"); + exit(1); + } + munmap(addr, len); + + return(fd); +} + +extern int have_devanon; + +int create_mem_file(unsigned long len) +{ + int err, fd; + + if(have_devanon) + fd = create_anon_file(len); + else fd = create_tmp_file(len); + + err = os_set_exec_close(fd, 1); + if(err < 0){ + errno = -err; + perror("exec_close"); + } + return(fd); +} diff --git a/arch/um/os-Linux/start_up.c b/arch/um/os-Linux/start_up.c index 040cc1472bc7..6af83171ca4e 100644 --- a/arch/um/os-Linux/start_up.c +++ b/arch/um/os-Linux/start_up.c @@ -4,18 +4,22 @@ */ #include +#include +#include +#include +#include #include #include #include +#include #include -#include -#include #include #include #include #include #include #include +#include #include "user_util.h" #include "kern_util.h" #include "user.h" @@ -25,6 +29,7 @@ #include "sysdep/sigcontext.h" #include "irq_user.h" #include "ptrace_user.h" +#include "mem_user.h" #include "time_user.h" #include "init.h" #include "os.h" @@ -32,6 +37,8 @@ #include "choose-mode.h" #include "mode.h" #include "tempfile.h" +#include "kern_constants.h" + #ifdef UML_CONFIG_MODE_SKAS #include "skas.h" #include "skas_ptrace.h" @@ -276,9 +283,38 @@ static void __init check_ptrace(void) check_sysemu(); } +extern int create_tmp_file(unsigned long len); + +static void check_tmpexec(void) +{ + void *addr; + int err, fd = create_tmp_file(UM_KERN_PAGE_SIZE); + + addr = mmap(NULL, UM_KERN_PAGE_SIZE, + PROT_READ | PROT_WRITE | PROT_EXEC, MAP_PRIVATE, fd, 0); + printf("Checking PROT_EXEC mmap in /tmp..."); + fflush(stdout); + if(addr == MAP_FAILED){ + err = errno; + perror("failed"); + if(err == EPERM) + printf("/tmp must be not mounted noexec\n"); + exit(1); + } + printf("OK\n"); + munmap(addr, UM_KERN_PAGE_SIZE); + + close(fd); +} + void os_early_checks(void) { check_ptrace(); + + /* Need to check this early because mmapping happens before the + * kernel is running. + */ + check_tmpexec(); } static int __init noprocmm_cmd_param(char *str, int* add) @@ -357,3 +393,72 @@ int can_do_skas(void) return(0); } #endif + +int have_devanon = 0; + +void check_devanon(void) +{ + int fd; + + printk("Checking for /dev/anon on the host..."); + fd = open("/dev/anon", O_RDWR); + if(fd < 0){ + printk("Not available (open failed with errno %d)\n", errno); + return; + } + + printk("OK\n"); + have_devanon = 1; +} + +int __init parse_iomem(char *str, int *add) +{ + struct iomem_region *new; + struct uml_stat buf; + char *file, *driver; + int fd, err, size; + + driver = str; + file = strchr(str,','); + if(file == NULL){ + printf("parse_iomem : failed to parse iomem\n"); + goto out; + } + *file = '\0'; + file++; + fd = os_open_file(file, of_rdwr(OPENFLAGS()), 0); + if(fd < 0){ + os_print_error(fd, "parse_iomem - Couldn't open io file"); + goto out; + } + + err = os_stat_fd(fd, &buf); + if(err < 0){ + os_print_error(err, "parse_iomem - cannot stat_fd file"); + goto out_close; + } + + new = malloc(sizeof(*new)); + if(new == NULL){ + perror("Couldn't allocate iomem_region struct"); + goto out_close; + } + + size = (buf.ust_size + UM_KERN_PAGE_SIZE) & ~(UM_KERN_PAGE_SIZE - 1); + + *new = ((struct iomem_region) { .next = iomem_regions, + .driver = driver, + .fd = fd, + .size = size, + .phys = 0, + .virt = 0 }); + iomem_regions = new; + iomem_size += new->size + UM_KERN_PAGE_SIZE; + + return(0); + out_close: + os_close_file(fd); + out: + return(1); +} + diff --git a/arch/um/os-Linux/tt.c b/arch/um/os-Linux/tt.c index 5b047ab8416a..a6db8877931a 100644 --- a/arch/um/os-Linux/tt.c +++ b/arch/um/os-Linux/tt.c @@ -36,6 +36,20 @@ #include "mode.h" #include "tempfile.h" +int protect_memory(unsigned long addr, unsigned long len, int r, int w, int x, + int must_succeed) +{ + int err; + + err = os_protect_memory((void *) addr, len, r, w, x); + if(err < 0){ + if(must_succeed) + panic("protect failed, err = %d", -err); + else return(err); + } + return(0); +} + /* *------------------------- * only for tt mode (will be deleted in future...) From 114069f7384c15a7a9f8d3e333caf4d470dfe520 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:27:51 -0700 Subject: [PATCH 105/240] [PATCH] uml: merge mem_user.c and mem.c The serial UML OS-abstraction layer patch (um/kernel dir). This joins mem_user.c and mem.c files. Signed-off-by: Gennady Sharapov Signed-off-by: Jeff Dike Cc: Paolo Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/Makefile | 12 +++---- arch/um/kernel/mem.c | 16 ++++++++- arch/um/kernel/mem_user.c | 70 --------------------------------------- 3 files changed, 21 insertions(+), 77 deletions(-) delete mode 100644 arch/um/kernel/mem_user.c diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile index 7eb52c5a7cab..1a0001b3850c 100644 --- a/arch/um/kernel/Makefile +++ b/arch/um/kernel/Makefile @@ -1,4 +1,4 @@ -# +# # Copyright (C) 2002 Jeff Dike (jdike@karaya.com) # Licensed under the GPL # @@ -7,11 +7,11 @@ extra-y := vmlinux.lds clean-files := obj-y = config.o exec_kern.o exitcode.o \ - helper.o init_task.o irq.o irq_user.o ksyms.o main.o mem.o mem_user.o \ - physmem.o process_kern.o ptrace.o reboot.o resource.o sigio_user.o \ - sigio_kern.o signal_kern.o signal_user.o smp.o syscall_kern.o sysrq.o \ - time.o time_kern.o tlb.o trap_kern.o trap_user.o \ - uaccess_user.o um_arch.o umid.o user_util.o + helper.o init_task.o irq.o irq_user.o ksyms.o main.o mem.o physmem.o \ + process_kern.o ptrace.o reboot.o resource.o sigio_user.o sigio_kern.o \ + signal_kern.o signal_user.o smp.o syscall_kern.o sysrq.o time.o \ + time_kern.o tlb.o trap_kern.o trap_user.o uaccess_user.o um_arch.o \ + umid.o user_util.o obj-$(CONFIG_BLK_DEV_INITRD) += initrd.o obj-$(CONFIG_GPROF) += gprof_syms.o diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c index 64fa062cc119..ea008b031a8f 100644 --- a/arch/um/kernel/mem.c +++ b/arch/um/kernel/mem.c @@ -1,4 +1,4 @@ -/* +/* * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) * Licensed under the GPL */ @@ -19,6 +19,10 @@ #include "mem_user.h" #include "uml_uaccess.h" #include "os.h" +#include "linux/types.h" +#include "linux/string.h" +#include "init.h" +#include "kern_constants.h" extern char __binary_start; @@ -368,6 +372,16 @@ struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address) return pte; } +struct iomem_region *iomem_regions = NULL; +int iomem_size = 0; + +extern int parse_iomem(char *str, int *add) __init; + +__uml_setup("iomem=", parse_iomem, +"iomem=,\n" +" Configure as an IO memory region named .\n\n" +); + /* * Overrides for Emacs so that we follow Linus's tabbing style. * Emacs will notice this stuff at the end of the file and automatically diff --git a/arch/um/kernel/mem_user.c b/arch/um/kernel/mem_user.c deleted file mode 100644 index d49640d9923f..000000000000 --- a/arch/um/kernel/mem_user.c +++ /dev/null @@ -1,70 +0,0 @@ -/* - * arch/um/kernel/mem_user.c - * - * BRIEF MODULE DESCRIPTION - * user side memory routines for supporting IO memory inside user mode linux - * - * Copyright (C) 2001 RidgeRun, Inc. - * Author: RidgeRun, Inc. - * Greg Lonnon glonnon@ridgerun.com or info@ridgerun.com - * - * 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 SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED - * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF - * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN - * NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT - * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF - * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON - * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF - * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - * - * 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., - * 675 Mass Ave, Cambridge, MA 02139, USA. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "kern_util.h" -#include "user.h" -#include "user_util.h" -#include "mem_user.h" -#include "init.h" -#include "os.h" -#include "tempfile.h" -#include "kern_constants.h" - -struct iomem_region *iomem_regions = NULL; -int iomem_size = 0; - -extern int parse_iomem(char *str, int *add) __init; - -__uml_setup("iomem=", parse_iomem, -"iomem=,\n" -" Configure as an IO memory region named .\n\n" -); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ From 2867ace6751adc20bb37cfb09478510c49110b27 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:27:51 -0700 Subject: [PATCH 106/240] [PATCH] uml: return a real error code do_aio used to return -1 on error instead of errno. Signed-off-by: Jeff Dike Cc: Paolo Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/os-Linux/aio.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/um/os-Linux/aio.c b/arch/um/os-Linux/aio.c index e942beb4e108..298d5632128b 100644 --- a/arch/um/os-Linux/aio.c +++ b/arch/um/os-Linux/aio.c @@ -117,6 +117,8 @@ static int do_aio(aio_context_t ctx, struct aio_context *aio) err = io_submit(ctx, 1, &iocbp); if(err > 0) err = 0; + else + err = -errno; out: return err; From 1f87b1e6e6050c0afe8b4afc279121f99043dc54 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:27:52 -0700 Subject: [PATCH 107/240] [PATCH] uml: remove include of asm/elf.h asm/elf.h is bad on x86_64, and i386 doesn't need it any more after Al's cleanup. Signed-off-by: Jeff Dike Cc: Paolo Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/os-Linux/elf_aux.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/um/os-Linux/elf_aux.c b/arch/um/os-Linux/elf_aux.c index 5c02e3395225..e770cb029578 100644 --- a/arch/um/os-Linux/elf_aux.c +++ b/arch/um/os-Linux/elf_aux.c @@ -9,7 +9,6 @@ */ #include #include -#include #include "init.h" #include "elf_user.h" #include "mem_user.h" From e36d394deb1b59d004ab057e0b5c505ffc5d8c0a Mon Sep 17 00:00:00 2001 From: Richard Purdie Date: Fri, 16 Sep 2005 19:27:53 -0700 Subject: [PATCH 108/240] [PATCH] Fix up some pm_message_t types Fix up some pm_message_t types Signed-Off-By: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/common/locomo.c | 2 +- arch/ppc/syslib/open_pic2.c | 2 +- drivers/serial/mpc52xx_uart.c | 2 +- drivers/video/imxfb.c | 2 +- sound/arm/aaci.c | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/arm/common/locomo.c b/arch/arm/common/locomo.c index 2786f7c34b3f..a7bd85700152 100644 --- a/arch/arm/common/locomo.c +++ b/arch/arm/common/locomo.c @@ -551,7 +551,7 @@ struct locomo_save_data { u16 LCM_SPIMD; }; -static int locomo_suspend(struct device *dev, u32 pm_message_t, u32 level) +static int locomo_suspend(struct device *dev, pm_message_t state, u32 level) { struct locomo *lchip = dev_get_drvdata(dev); struct locomo_save_data *save; diff --git a/arch/ppc/syslib/open_pic2.c b/arch/ppc/syslib/open_pic2.c index 2e0ea92144f6..9a7e8748e2b2 100644 --- a/arch/ppc/syslib/open_pic2.c +++ b/arch/ppc/syslib/open_pic2.c @@ -575,7 +575,7 @@ static void openpic2_cached_disable_irq(u_int irq) * we need something better to deal with that... Maybe switch to S1 for * cpufreq changes */ -int openpic2_suspend(struct sys_device *sysdev, u32 state) +int openpic2_suspend(struct sys_device *sysdev, pm_message_t state) { int i; unsigned long flags; diff --git a/drivers/serial/mpc52xx_uart.c b/drivers/serial/mpc52xx_uart.c index a3cd0ee8486d..0585ab27ffde 100644 --- a/drivers/serial/mpc52xx_uart.c +++ b/drivers/serial/mpc52xx_uart.c @@ -781,7 +781,7 @@ mpc52xx_uart_remove(struct device *dev) #ifdef CONFIG_PM static int -mpc52xx_uart_suspend(struct device *dev, u32 state, u32 level) +mpc52xx_uart_suspend(struct device *dev, pm_message_t state, u32 level) { struct uart_port *port = (struct uart_port *) dev_get_drvdata(dev); diff --git a/drivers/video/imxfb.c b/drivers/video/imxfb.c index cabd53cec991..6c2244cf0e74 100644 --- a/drivers/video/imxfb.c +++ b/drivers/video/imxfb.c @@ -425,7 +425,7 @@ static void imxfb_setup_gpio(struct imxfb_info *fbi) * Power management hooks. Note that we won't be called from IRQ context, * unlike the blank functions above, so we may sleep. */ -static int imxfb_suspend(struct device *dev, u32 state, u32 level) +static int imxfb_suspend(struct device *dev, pm_message_t state, u32 level) { struct imxfb_info *fbi = dev_get_drvdata(dev); pr_debug("%s\n",__FUNCTION__); diff --git a/sound/arm/aaci.c b/sound/arm/aaci.c index 34195b748608..b2d5db20ec8c 100644 --- a/sound/arm/aaci.c +++ b/sound/arm/aaci.c @@ -650,7 +650,7 @@ static int aaci_do_resume(snd_card_t *card, unsigned int state) return 0; } -static int aaci_suspend(struct amba_device *dev, u32 state) +static int aaci_suspend(struct amba_device *dev, pm_message_t state) { snd_card_t *card = amba_get_drvdata(dev); return card ? aaci_do_suspend(card) : 0; From b9491ac835829e6a34e2bbaa8adad261c71bf990 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 16 Sep 2005 19:27:54 -0700 Subject: [PATCH 109/240] [PATCH] x86_64: e820.c needs module.h For EXPORT_SYMBOL. Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/e820.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index 4e34b0f9d613..ab3f87aaff70 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -17,6 +17,8 @@ #include #include #include +#include + #include #include #include From f3519f91942f2b43942400348c16d63fe9327f04 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 16 Sep 2005 19:27:54 -0700 Subject: [PATCH 110/240] [PATCH] fix mm/Kconfig spelling Signed-off-by: Dave Hansen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/Kconfig b/mm/Kconfig index 4e9937ac3529..391ffc54d136 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -29,7 +29,7 @@ config FLATMEM_MANUAL If unsure, choose this option (Flat Memory) over any other. config DISCONTIGMEM_MANUAL - bool "Discontigious Memory" + bool "Discontiguous Memory" depends on ARCH_DISCONTIGMEM_ENABLE help This option provides enhanced support for discontiguous @@ -52,7 +52,7 @@ config SPARSEMEM_MANUAL memory hotplug systems. This is normal. For many other systems, this will be an alternative to - "Discontigious Memory". This option provides some potential + "Discontiguous Memory". This option provides some potential performance benefits, along with decreased code complexity, but it is newer, and more experimental. From 73a0b538ee573a76cba59cdc9f177a71776d4678 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 16 Sep 2005 19:27:55 -0700 Subject: [PATCH 111/240] [PATCH] x86_64: desc.h-needs smp.h include/asm/desc.h: In function `load_LDT': include/asm/desc.h:209: warning: implicit declaration of function `get_cpu' include/asm/desc.h:211: warning: implicit declaration of function `put_cpu' Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-x86_64/desc.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/asm-x86_64/desc.h b/include/asm-x86_64/desc.h index 594e610f4a1e..68ac3c62fe3d 100644 --- a/include/asm-x86_64/desc.h +++ b/include/asm-x86_64/desc.h @@ -8,6 +8,8 @@ #ifndef __ASSEMBLY__ #include +#include + #include #include From d15c5749eb81dee94d40fe12584ca8461858b4cb Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 16 Sep 2005 19:27:56 -0700 Subject: [PATCH 112/240] [PATCH] seclvl-use-securityfs tidy We don't put braces around single statements, thanks. Cc: Serge Hallyn Cc: James Morris Cc: Chris Wright Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- security/seclvl.c | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/security/seclvl.c b/security/seclvl.c index dc4e17b6eaf6..241093d6f0b8 100644 --- a/security/seclvl.c +++ b/security/seclvl.c @@ -253,9 +253,8 @@ passwd_write_file(struct file * file, const char __user * buf, if (count < 0 || count >= PAGE_SIZE) return -ENOMEM; - if (*ppos != 0) { + if (*ppos != 0) return -EINVAL; - } page = (char *)get_zeroed_page(GFP_KERNEL); if (!page) return -ENOMEM; @@ -265,9 +264,8 @@ passwd_write_file(struct file * file, const char __user * buf, len = strlen(page); /* ``echo "secret" > seclvl/passwd'' includes a newline */ - if (page[len - 1] == '\n') { + if (page[len - 1] == '\n') len--; - } /* Hash the password, then compare the hashed values */ if ((rc = plaintext_to_sha1(tmp, page, len))) { seclvl_printk(0, KERN_ERR, "Error hashing password: rc = " @@ -275,9 +273,8 @@ passwd_write_file(struct file * file, const char __user * buf, return rc; } for (i = 0; i < SHA1_DIGEST_SIZE; i++) { - if (hashedPassword[i] != tmp[i]) { + if (hashedPassword[i] != tmp[i]) return -EPERM; - } } seclvl_printk(0, KERN_INFO, "Password accepted; seclvl reduced to 0.\n"); @@ -482,9 +479,8 @@ static void seclvl_file_free_security(struct file *filp) */ static int seclvl_umount(struct vfsmount *mnt, int flags) { - if (current->pid == 1) { + if (current->pid == 1) return 0; - } if (seclvl == 2) { seclvl_printk(1, KERN_WARNING, "Attempt to unmount in secure " "level %d\n", seclvl); @@ -638,9 +634,8 @@ static int __init seclvl_init(void) static void __exit seclvl_exit(void) { securityfs_remove(seclvl_ino); - if (*passwd || *sha1_passwd) { + if (*passwd || *sha1_passwd) securityfs_remove(passwd_ino); - } securityfs_remove(dir_ino); if (secondary == 1) { mod_unreg_security(MY_NAME, &seclvl_ops); From 9afa57b04ca08ff061e54787e3becf5c40283149 Mon Sep 17 00:00:00 2001 From: Serge Hallyn Date: Fri, 16 Sep 2005 19:27:57 -0700 Subject: [PATCH 113/240] [PATCH] seclvl: use securityfs (fix) That should be -EINVAL for both. Signed-off-by: Serge Hallyn Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- security/seclvl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/security/seclvl.c b/security/seclvl.c index 241093d6f0b8..1caac0164643 100644 --- a/security/seclvl.c +++ b/security/seclvl.c @@ -252,7 +252,7 @@ passwd_write_file(struct file * file, const char __user * buf, } if (count < 0 || count >= PAGE_SIZE) - return -ENOMEM; + return -EINVAL; if (*ppos != 0) return -EINVAL; page = (char *)get_zeroed_page(GFP_KERNEL); From 8a70ce8a4b1c9d39e0a180173728a8d102274631 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Fri, 16 Sep 2005 19:27:58 -0700 Subject: [PATCH 114/240] [PATCH] ppc64: build fix I forgot to include siginfo.h when I added data breakpoint support. We must include it in a round-a-bout way in mainline. Signed-off-by: Anton Blanchard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc64/mm/fault.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/ppc64/mm/fault.c b/arch/ppc64/mm/fault.c index 7fbc68bbb739..be3f25cf3e9f 100644 --- a/arch/ppc64/mm/fault.c +++ b/arch/ppc64/mm/fault.c @@ -38,6 +38,7 @@ #include #include #include +#include /* * Check whether the instruction at regs->nip is a store using From 0faf3d3d08bb20d803ba090a1bc1ffedaea6ced6 Mon Sep 17 00:00:00 2001 From: Thomas Maguin Date: Fri, 16 Sep 2005 19:27:58 -0700 Subject: [PATCH 115/240] [PATCH] scsi_ioctl: Add WRITE_LONG_2 as write safe command Add WRITE_LONG_2 as write safe commands, which which allows normal users to make a c1-, c2- and cu-scan (so called cxscan) with readcd on cxscan-capable cd/dvd-writers Signed-off-by: Jens Axboe Cc: James Bottomley Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/scsi_ioctl.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/block/scsi_ioctl.c b/drivers/block/scsi_ioctl.c index 856c2278e9d0..079ec344eb47 100644 --- a/drivers/block/scsi_ioctl.c +++ b/drivers/block/scsi_ioctl.c @@ -168,6 +168,7 @@ static int verify_command(struct file *file, unsigned char *cmd) safe_for_write(WRITE_VERIFY_12), safe_for_write(WRITE_16), safe_for_write(WRITE_LONG), + safe_for_write(WRITE_LONG_2), safe_for_write(ERASE), safe_for_write(GPCMD_MODE_SELECT_10), safe_for_write(MODE_SELECT), From 514ccd4e6c414d8064d53235f7fc09fc02ec2078 Mon Sep 17 00:00:00 2001 From: Vitaly Bordug Date: Fri, 16 Sep 2005 19:28:00 -0700 Subject: [PATCH 116/240] [PATCH] ppc32: Add ppc_sys descriptions for PowerQUICC I devices Added ppc_sys device and system definitions for PowerQUICC I devices. This will allow drivers for PQI to be proper platform device drivers. Currently sys section contains only MPC885 and MPC866. Identification should be done with identify_ppc_sys_by_name call, with board-specific "name" string passed, since PQI do not have any register that could identify the SOC. Signed-off-by: Vitaly Bordug Signed-off-by: Kumar Gala Signed-off-by: Marcelo Tosatti Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc/syslib/Makefile | 3 +- arch/ppc/syslib/mpc8xx_devices.c | 224 +++++++++++++++++++++++++++++++ arch/ppc/syslib/mpc8xx_sys.c | 61 +++++++++ include/asm-ppc/irq.h | 10 ++ include/asm-ppc/mpc8xx.h | 16 +++ include/asm-ppc/ppc_sys.h | 2 + 6 files changed, 315 insertions(+), 1 deletion(-) create mode 100644 arch/ppc/syslib/mpc8xx_devices.c create mode 100644 arch/ppc/syslib/mpc8xx_sys.c diff --git a/arch/ppc/syslib/Makefile b/arch/ppc/syslib/Makefile index 8b9b226005d1..b8d08f33f7ee 100644 --- a/arch/ppc/syslib/Makefile +++ b/arch/ppc/syslib/Makefile @@ -34,7 +34,8 @@ ifeq ($(CONFIG_40x),y) obj-$(CONFIG_PCI) += indirect_pci.o pci_auto.o ppc405_pci.o endif endif -obj-$(CONFIG_8xx) += m8xx_setup.o ppc8xx_pic.o $(wdt-mpc8xx-y) +obj-$(CONFIG_8xx) += m8xx_setup.o ppc8xx_pic.o $(wdt-mpc8xx-y) \ + ppc_sys.o mpc8xx_devices.o mpc8xx_sys.o ifeq ($(CONFIG_8xx),y) obj-$(CONFIG_PCI) += qspan_pci.o i8259.o endif diff --git a/arch/ppc/syslib/mpc8xx_devices.c b/arch/ppc/syslib/mpc8xx_devices.c new file mode 100644 index 000000000000..2b5f0e701687 --- /dev/null +++ b/arch/ppc/syslib/mpc8xx_devices.c @@ -0,0 +1,224 @@ +/* + * arch/ppc/syslib/mpc8xx_devices.c + * + * MPC8xx Device descriptions + * + * Maintainer: Kumar Gala + * + * Copyright 2005 MontaVista Software, Inc. by Vitaly Bordug + * + * 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. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* We use offsets for IORESOURCE_MEM to do not set dependences at compile time. + * They will get fixed up by mach_mpc8xx_fixup + */ + +struct platform_device ppc_sys_platform_devices[] = { + [MPC8xx_CPM_FEC1] = { + .name = "fsl-cpm-fec", + .id = 1, + .num_resources = 2, + .resource = (struct resource[]) { + { + .name = "regs", + .start = 0xe00, + .end = 0xe88, + .flags = IORESOURCE_MEM, + }, + { + .name = "interrupt", + .start = MPC8xx_INT_FEC1, + .end = MPC8xx_INT_FEC1, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC8xx_CPM_FEC2] = { + .name = "fsl-cpm-fec", + .id = 2, + .num_resources = 2, + .resource = (struct resource[]) { + { + .name = "regs", + .start = 0x1e00, + .end = 0x1e88, + .flags = IORESOURCE_MEM, + }, + { + .name = "interrupt", + .start = MPC8xx_INT_FEC2, + .end = MPC8xx_INT_FEC2, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC8xx_CPM_SCC1] = { + .name = "fsl-cpm-scc", + .id = 1, + .num_resources = 3, + .resource = (struct resource[]) { + { + .name = "regs", + .start = 0xa00, + .end = 0xa18, + .flags = IORESOURCE_MEM, + }, + { + .name = "pram", + .start = 0x3c00, + .end = 0x3c80, + .flags = IORESOURCE_MEM, + }, + { + .name = "interrupt", + .start = MPC8xx_INT_SCC1, + .end = MPC8xx_INT_SCC1, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC8xx_CPM_SCC2] = { + .name = "fsl-cpm-scc", + .id = 2, + .num_resources = 3, + .resource = (struct resource[]) { + { + .name = "regs", + .start = 0xa20, + .end = 0xa38, + .flags = IORESOURCE_MEM, + }, + { + .name = "pram", + .start = 0x3d00, + .end = 0x3d80, + .flags = IORESOURCE_MEM, + }, + + { + .name = "interrupt", + .start = MPC8xx_INT_SCC2, + .end = MPC8xx_INT_SCC2, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC8xx_CPM_SCC3] = { + .name = "fsl-cpm-scc", + .id = 3, + .num_resources = 3, + .resource = (struct resource[]) { + { + .name = "regs", + .start = 0xa40, + .end = 0xa58, + .flags = IORESOURCE_MEM, + }, + { + .name = "pram", + .start = 0x3e00, + .end = 0x3e80, + .flags = IORESOURCE_MEM, + }, + + { + .name = "interrupt", + .start = MPC8xx_INT_SCC3, + .end = MPC8xx_INT_SCC3, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC8xx_CPM_SCC4] = { + .name = "fsl-cpm-scc", + .id = 4, + .num_resources = 3, + .resource = (struct resource[]) { + { + .name = "regs", + .start = 0xa60, + .end = 0xa78, + .flags = IORESOURCE_MEM, + }, + { + .name = "pram", + .start = 0x3f00, + .end = 0x3f80, + .flags = IORESOURCE_MEM, + }, + + { + .name = "interrupt", + .start = MPC8xx_INT_SCC4, + .end = MPC8xx_INT_SCC4, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC8xx_CPM_SMC1] = { + .name = "fsl-cpm-smc", + .id = 1, + .num_resources = 2, + .resource = (struct resource[]) { + { + .name = "regs", + .start = 0xa82, + .end = 0xa91, + .flags = IORESOURCE_MEM, + }, + { + .name = "interrupt", + .start = MPC8xx_INT_SMC1, + .end = MPC8xx_INT_SMC1, + .flags = IORESOURCE_IRQ, + }, + }, + }, + [MPC8xx_CPM_SMC2] = { + .name = "fsl-cpm-smc", + .id = 2, + .num_resources = 2, + .resource = (struct resource[]) { + { + .name = "regs", + .start = 0xa92, + .end = 0xaa1, + .flags = IORESOURCE_MEM, + }, + { + .name = "interrupt", + .start = MPC8xx_INT_SMC2, + .end = MPC8xx_INT_SMC2, + .flags = IORESOURCE_IRQ, + }, + }, + }, +}; + +static int __init mach_mpc8xx_fixup(struct platform_device *pdev) +{ + ppc_sys_fixup_mem_resource (pdev, IMAP_ADDR); + return 0; +} + +static int __init mach_mpc8xx_init(void) +{ + ppc_sys_device_fixup = mach_mpc8xx_fixup; + return 0; +} + +postcore_initcall(mach_mpc8xx_init); diff --git a/arch/ppc/syslib/mpc8xx_sys.c b/arch/ppc/syslib/mpc8xx_sys.c new file mode 100644 index 000000000000..a532ccc861c0 --- /dev/null +++ b/arch/ppc/syslib/mpc8xx_sys.c @@ -0,0 +1,61 @@ +/* + * arch/ppc/platforms/mpc8xx_sys.c + * + * MPC8xx System descriptions + * + * Maintainer: Kumar Gala + * + * Copyright 2005 MontaVista Software, Inc. by Vitaly Bordug + * + * 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. + */ + +#include +#include +#include +#include + +struct ppc_sys_spec *cur_ppc_sys_spec; +struct ppc_sys_spec ppc_sys_specs[] = { + { + .ppc_sys_name = "MPC86X", + .mask = 0xFFFFFFFF, + .value = 0x00000000, + .num_devices = 2, + .device_list = (enum ppc_sys_devices[]) + { + MPC8xx_CPM_FEC1, + MPC8xx_CPM_SCC1, + MPC8xx_CPM_SCC2, + MPC8xx_CPM_SCC3, + MPC8xx_CPM_SCC4, + MPC8xx_CPM_SMC1, + MPC8xx_CPM_SMC2, + }, + }, + { + .ppc_sys_name = "MPC885", + .mask = 0xFFFFFFFF, + .value = 0x00000000, + .num_devices = 3, + .device_list = (enum ppc_sys_devices[]) + { + MPC8xx_CPM_FEC1, + MPC8xx_CPM_FEC2, + MPC8xx_CPM_SCC1, + MPC8xx_CPM_SCC2, + MPC8xx_CPM_SCC3, + MPC8xx_CPM_SCC4, + MPC8xx_CPM_SMC1, + MPC8xx_CPM_SMC2, + }, + }, + { /* default match */ + .ppc_sys_name = "", + .mask = 0x00000000, + .value = 0x00000000, + }, +}; diff --git a/include/asm-ppc/irq.h b/include/asm-ppc/irq.h index 55752474d0d9..bd9674807f05 100644 --- a/include/asm-ppc/irq.h +++ b/include/asm-ppc/irq.h @@ -138,6 +138,16 @@ irq_canonicalize(int irq) #define SIU_IRQ7 (14) #define SIU_LEVEL7 (15) +#define MPC8xx_INT_FEC1 SIU_LEVEL1 +#define MPC8xx_INT_FEC2 SIU_LEVEL3 + +#define MPC8xx_INT_SCC1 (CPM_IRQ_OFFSET + CPMVEC_SCC1) +#define MPC8xx_INT_SCC2 (CPM_IRQ_OFFSET + CPMVEC_SCC2) +#define MPC8xx_INT_SCC3 (CPM_IRQ_OFFSET + CPMVEC_SCC3) +#define MPC8xx_INT_SCC4 (CPM_IRQ_OFFSET + CPMVEC_SCC4) +#define MPC8xx_INT_SMC1 (CPM_IRQ_OFFSET + CPMVEC_SMC1) +#define MPC8xx_INT_SMC2 (CPM_IRQ_OFFSET + CPMVEC_SMC2) + /* The internal interrupts we can configure as we see fit. * My personal preference is CPM at level 2, which puts it above the * MBX PCI/ISA/IDE interrupts. diff --git a/include/asm-ppc/mpc8xx.h b/include/asm-ppc/mpc8xx.h index dc8e59896050..208a2e11daee 100644 --- a/include/asm-ppc/mpc8xx.h +++ b/include/asm-ppc/mpc8xx.h @@ -97,6 +97,22 @@ extern unsigned char __res[]; struct pt_regs; +enum ppc_sys_devices { + MPC8xx_CPM_FEC1, + MPC8xx_CPM_FEC2, + MPC8xx_CPM_I2C, + MPC8xx_CPM_SCC1, + MPC8xx_CPM_SCC2, + MPC8xx_CPM_SCC3, + MPC8xx_CPM_SCC4, + MPC8xx_CPM_SPI, + MPC8xx_CPM_MCC1, + MPC8xx_CPM_MCC2, + MPC8xx_CPM_SMC1, + MPC8xx_CPM_SMC2, + MPC8xx_CPM_USB, +}; + #endif /* !__ASSEMBLY__ */ #endif /* CONFIG_8xx */ #endif /* __CONFIG_8xx_DEFS */ diff --git a/include/asm-ppc/ppc_sys.h b/include/asm-ppc/ppc_sys.h index 048f7c8596ee..549f44843c5e 100644 --- a/include/asm-ppc/ppc_sys.h +++ b/include/asm-ppc/ppc_sys.h @@ -27,6 +27,8 @@ #include #elif defined(CONFIG_85xx) #include +#elif defined(CONFIG_8xx) +#include #elif defined(CONFIG_PPC_MPC52xx) #include #elif defined(CONFIG_MPC10X_BRIDGE) From afeda2c24e74cbddde376e06fdd82c215f9cb637 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 16 Sep 2005 19:28:01 -0700 Subject: [PATCH 117/240] [PATCH] relayfs documentation typo Small typo in relayfs documentation. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/filesystems/relayfs.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/filesystems/relayfs.txt b/Documentation/filesystems/relayfs.txt index d24e1b0d4f39..d803abed29f0 100644 --- a/Documentation/filesystems/relayfs.txt +++ b/Documentation/filesystems/relayfs.txt @@ -15,7 +15,7 @@ retrieve the data as it becomes available. The format of the data logged into the channel buffers is completely up to the relayfs client; relayfs does however provide hooks which -allow clients to impose some stucture on the buffer data. Nor does +allow clients to impose some structure on the buffer data. Nor does relayfs implement any form of data filtering - this also is left to the client. The purpose is to keep relayfs as simple as possible. From 2030c0fd3db3332378f95f479c286181e3920644 Mon Sep 17 00:00:00 2001 From: Michael Kerrisk Date: Fri, 16 Sep 2005 19:28:02 -0700 Subject: [PATCH 118/240] [PATCH] PR_GET_DUMPABLE returns incorrect info 2.6.13 incorporated Alan Cox's patch for /proc/sys/fs/suid_dumpable (one version of this patch can be found here http://marc.theaimsgroup.com/?l=linux-kernel&m=109647550421014&w=2 ). This patch also made corresponding changes in kernel/sys.c to change the prctl() PR_SET_DUMPABLE operation so that the permitted range of 'arg2' was modified from 0..1 to 0..2. However, a corresponding change was not made for PR_GET_DUMPABLE: if the dumpable flag is non-zero, then PR_GET_DUMPABLE always returns 1, so that the caller can't determine the true setting of this flag. Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sys.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/kernel/sys.c b/kernel/sys.c index c80412be2302..f723522e6986 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1728,8 +1728,7 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, error = put_user(current->pdeath_signal, (int __user *)arg2); break; case PR_GET_DUMPABLE: - if (current->mm->dumpable) - error = 1; + error = current->mm->dumpable; break; case PR_SET_DUMPABLE: if (arg2 < 0 || arg2 > 2) { From 3013449243adc3421b507696e5d247a3d292ee0c Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 16 Sep 2005 19:28:03 -0700 Subject: [PATCH 119/240] [PATCH] uml: UML/i386 cmpxchg fix Using native cmpxchg offers a slight performance improvement in uml/i386. Signed-off-by: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/Kconfig.i386 | 4 ++++ include/asm-um/system-i386.h | 2 -- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/um/Kconfig.i386 b/arch/um/Kconfig.i386 index 8ad156a00499..5d92cacd56c6 100644 --- a/arch/um/Kconfig.i386 +++ b/arch/um/Kconfig.i386 @@ -42,3 +42,7 @@ config ARCH_HAS_SC_SIGNALS config ARCH_REUSE_HOST_VSYSCALL_AREA bool default y + +config X86_CMPXCHG + bool + default y diff --git a/include/asm-um/system-i386.h b/include/asm-um/system-i386.h index ea8381de3cc9..c436263e67ba 100644 --- a/include/asm-um/system-i386.h +++ b/include/asm-um/system-i386.h @@ -3,6 +3,4 @@ #include "asm/system-generic.h" -#define __HAVE_ARCH_CMPXCHG 1 - #endif From e61c0e336f3931842f09e6709d76146bfd81184e Mon Sep 17 00:00:00 2001 From: Abhay Salunke Date: Fri, 16 Sep 2005 19:28:04 -0700 Subject: [PATCH 120/240] [PATCH] dell_rbu: enhancements and fixes BUG fixes: The driver used to allocate memory with spinlock held which has been fixed in this patch. The driver was printing the entire buffer when it received a invalid entry in image_type. The fix is to only print a warning message and not the buffer. Usability enhancements: It is possible that due to user error the /sys/class/firmware/dell_rbu entries might be missing, this can happen if the user does the following echo 1 > /sys/class/firmware/dell_rbu/loading echo 0 > /sys/class/firmware/dell_rbu/loading This will make the entries in /sys/class/firmware/ to disappear and the only way get them back was bby unloading and loading the driver. This patch makes the user recreate these entries by echoing init in to image_type. This patch has been tested with Libsmbios and Dell OpenManage. Signed-off-by: Abhay Salunke Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/dell_rbu.txt | 18 ++- drivers/firmware/dell_rbu.c | 265 ++++++++++++++++++++++-------------- 2 files changed, 174 insertions(+), 109 deletions(-) diff --git a/Documentation/dell_rbu.txt b/Documentation/dell_rbu.txt index bcfa5c35036b..95d7f62e4dbc 100644 --- a/Documentation/dell_rbu.txt +++ b/Documentation/dell_rbu.txt @@ -13,6 +13,8 @@ the BIOS on Dell servers (starting from servers sold since 1999), desktops and notebooks (starting from those sold in 2005). Please go to http://support.dell.com register and you can find info on OpenManage and Dell Update packages (DUP). +Libsmbios can also be used to update BIOS on Dell systems go to +http://linux.dell.com/libsmbios/ for details. Dell_RBU driver supports BIOS update using the monilothic image and packetized image methods. In case of moniolithic the driver allocates a contiguous chunk @@ -22,8 +24,8 @@ would place each packet in contiguous physical memory. The driver also maintains a link list of packets for reading them back. If the dell_rbu driver is unloaded all the allocated memory is freed. -The rbu driver needs to have an application which will inform the BIOS to -enable the update in the next system reboot. +The rbu driver needs to have an application (as mentioned above)which will +inform the BIOS to enable the update in the next system reboot. The user should not unload the rbu driver after downloading the BIOS image or updating. @@ -42,9 +44,11 @@ In case of packet mechanism the single memory can be broken in smaller chuks of contiguous memory and the BIOS image is scattered in these packets. By default the driver uses monolithic memory for the update type. This can be -changed to contiguous during the driver load time by specifying the load +changed to packets during the driver load time by specifying the load parameter image_type=packet. This can also be changed later as below echo packet > /sys/devices/platform/dell_rbu/image_type +Also echoing either mono ,packet or init in to image_type will free up the +memory allocated by the driver. Do the steps below to download the BIOS image. 1) echo 1 > /sys/class/firmware/dell_rbu/loading @@ -53,9 +57,13 @@ Do the steps below to download the BIOS image. The /sys/class/firmware/dell_rbu/ entries will remain till the following is done. -echo -1 > /sys/class/firmware/dell_rbu/loading - +echo -1 > /sys/class/firmware/dell_rbu/loading. Until this step is completed the drivr cannot be unloaded. +If an user by accident executes steps 1 and 3 above without executing step 2; +it will make the /sys/class/firmware/dell_rbu/ entries to disappear. +The entries can be recreated by doing the following +echo init > /sys/devices/platform/dell_rbu/image_type +NOTE: echoing init in image_type does not change it original value. Also the driver provides /sys/devices/platform/dell_rbu/data readonly file to read back the image downloaded. This is useful in case of packet update diff --git a/drivers/firmware/dell_rbu.c b/drivers/firmware/dell_rbu.c index 3b865f34a095..9c2c2b199454 100644 --- a/drivers/firmware/dell_rbu.c +++ b/drivers/firmware/dell_rbu.c @@ -50,7 +50,7 @@ MODULE_AUTHOR("Abhay Salunke "); MODULE_DESCRIPTION("Driver for updating BIOS image on DELL systems"); MODULE_LICENSE("GPL"); -MODULE_VERSION("1.0"); +MODULE_VERSION("2.0"); #define BIOS_SCAN_LIMIT 0xffffffff #define MAX_IMAGE_LENGTH 16 @@ -65,10 +65,11 @@ static struct _rbu_data { unsigned long packet_write_count; unsigned long num_packets; unsigned long packetsize; + int entry_created; } rbu_data; -static char image_type[MAX_IMAGE_LENGTH] = "mono"; -module_param_string(image_type, image_type, sizeof(image_type), 0); +static char image_type[MAX_IMAGE_LENGTH + 1] = "mono"; +module_param_string(image_type, image_type, sizeof (image_type), 0); MODULE_PARM_DESC(image_type, "BIOS image type. choose- mono or packet"); struct packet_data { @@ -84,7 +85,8 @@ static struct platform_device *rbu_device; static int context; static dma_addr_t dell_rbu_dmaaddr; -static void init_packet_head(void) +static void +init_packet_head(void) { INIT_LIST_HEAD(&packet_data_head.list); rbu_data.packet_write_count = 0; @@ -93,7 +95,8 @@ static void init_packet_head(void) rbu_data.packetsize = 0; } -static int fill_last_packet(void *data, size_t length) +static int +fill_last_packet(void *data, size_t length) { struct list_head *ptemp_list; struct packet_data *packet = NULL; @@ -114,7 +117,7 @@ static int fill_last_packet(void *data, size_t length) if ((rbu_data.packet_write_count + length) > rbu_data.packetsize) { pr_debug("dell_rbu:%s: packet size data " - "overrun\n", __FUNCTION__); + "overrun\n", __FUNCTION__); return -EINVAL; } @@ -135,7 +138,8 @@ static int fill_last_packet(void *data, size_t length) return 0; } -static int create_packet(size_t length) +static int +create_packet(size_t length) { struct packet_data *newpacket; int ordernum = 0; @@ -146,12 +150,14 @@ static int create_packet(size_t length) pr_debug("create_packet: packetsize not specified\n"); return -EINVAL; } + spin_unlock(&rbu_data.lock); + newpacket = kmalloc(sizeof (struct packet_data), GFP_KERNEL); + spin_lock(&rbu_data.lock); - newpacket = kmalloc(sizeof(struct packet_data), GFP_KERNEL); if (!newpacket) { printk(KERN_WARNING - "dell_rbu:%s: failed to allocate new " - "packet\n", __FUNCTION__); + "dell_rbu:%s: failed to allocate new " + "packet\n", __FUNCTION__); return -ENOMEM; } @@ -160,15 +166,17 @@ static int create_packet(size_t length) * there is no upper limit on memory * address for packetized mechanism */ - newpacket->data = (unsigned char *)__get_free_pages(GFP_KERNEL, - ordernum); + spin_unlock(&rbu_data.lock); + newpacket->data = (unsigned char *) __get_free_pages(GFP_KERNEL, + ordernum); + spin_lock(&rbu_data.lock); pr_debug("create_packet: newpacket %p\n", newpacket->data); if (!newpacket->data) { printk(KERN_WARNING - "dell_rbu:%s: failed to allocate new " - "packet\n", __FUNCTION__); + "dell_rbu:%s: failed to allocate new " + "packet\n", __FUNCTION__); kfree(newpacket); return -ENOMEM; } @@ -190,7 +198,8 @@ static int create_packet(size_t length) return 0; } -static int packetize_data(void *data, size_t length) +static int +packetize_data(void *data, size_t length) { int rc = 0; @@ -206,7 +215,7 @@ static int packetize_data(void *data, size_t length) static int do_packet_read(char *data, struct list_head *ptemp_list, - int length, int bytes_read, int *list_read_count) + int length, int bytes_read, int *list_read_count) { void *ptemp_buf; struct packet_data *newpacket = NULL; @@ -239,7 +248,8 @@ do_packet_read(char *data, struct list_head *ptemp_list, return bytes_copied; } -static int packet_read_list(char *data, size_t * pread_length) +static int +packet_read_list(char *data, size_t * pread_length) { struct list_head *ptemp_list; int temp_count = 0; @@ -258,8 +268,7 @@ static int packet_read_list(char *data, size_t * pread_length) ptemp_list = (&packet_data_head.list)->next; while (!list_empty(ptemp_list)) { bytes_copied = do_packet_read(pdest, ptemp_list, - remaining_bytes, bytes_read, - &temp_count); + remaining_bytes, bytes_read, &temp_count); remaining_bytes -= bytes_copied; bytes_read += bytes_copied; pdest += bytes_copied; @@ -278,7 +287,8 @@ static int packet_read_list(char *data, size_t * pread_length) return 0; } -static void packet_empty_list(void) +static void +packet_empty_list(void) { struct list_head *ptemp_list; struct list_head *pnext_list; @@ -287,7 +297,7 @@ static void packet_empty_list(void) ptemp_list = (&packet_data_head.list)->next; while (!list_empty(ptemp_list)) { newpacket = - list_entry(ptemp_list, struct packet_data, list); + list_entry(ptemp_list, struct packet_data, list); pnext_list = ptemp_list->next; list_del(ptemp_list); ptemp_list = pnext_list; @@ -296,8 +306,8 @@ static void packet_empty_list(void) * to make sure there are no stale RBU packets left in memory */ memset(newpacket->data, 0, rbu_data.packetsize); - free_pages((unsigned long)newpacket->data, - newpacket->ordernum); + free_pages((unsigned long) newpacket->data, + newpacket->ordernum); kfree(newpacket); } rbu_data.packet_write_count = 0; @@ -310,7 +320,8 @@ static void packet_empty_list(void) * img_update_free: Frees the buffer allocated for storing BIOS image * Always called with lock held and returned with lock held */ -static void img_update_free(void) +static void +img_update_free(void) { if (!rbu_data.image_update_buffer) return; @@ -319,14 +330,13 @@ static void img_update_free(void) * BIOS image copied in memory. */ memset(rbu_data.image_update_buffer, 0, - rbu_data.image_update_buffer_size); + rbu_data.image_update_buffer_size); if (rbu_data.dma_alloc == 1) dma_free_coherent(NULL, rbu_data.bios_image_size, - rbu_data.image_update_buffer, - dell_rbu_dmaaddr); + rbu_data.image_update_buffer, dell_rbu_dmaaddr); else - free_pages((unsigned long)rbu_data.image_update_buffer, - rbu_data.image_update_ordernum); + free_pages((unsigned long) rbu_data.image_update_buffer, + rbu_data.image_update_ordernum); /* * Re-initialize the rbu_data variables after a free @@ -348,7 +358,8 @@ static void img_update_free(void) * already allocated size, then that memory is reused. This function is * called with lock held and returns with lock held. */ -static int img_update_realloc(unsigned long size) +static int +img_update_realloc(unsigned long size) { unsigned char *image_update_buffer = NULL; unsigned long rc; @@ -366,7 +377,7 @@ static int img_update_realloc(unsigned long size) */ if ((size != 0) && (rbu_data.image_update_buffer == NULL)) { printk(KERN_ERR "dell_rbu:%s: corruption " - "check failed\n", __FUNCTION__); + "check failed\n", __FUNCTION__); return -EINVAL; } /* @@ -385,17 +396,16 @@ static int img_update_realloc(unsigned long size) ordernum = get_order(size); image_update_buffer = - (unsigned char *)__get_free_pages(GFP_KERNEL, ordernum); + (unsigned char *) __get_free_pages(GFP_KERNEL, ordernum); img_buf_phys_addr = - (unsigned long)virt_to_phys(image_update_buffer); + (unsigned long) virt_to_phys(image_update_buffer); if (img_buf_phys_addr > BIOS_SCAN_LIMIT) { - free_pages((unsigned long)image_update_buffer, ordernum); + free_pages((unsigned long) image_update_buffer, ordernum); ordernum = -1; image_update_buffer = dma_alloc_coherent(NULL, size, - &dell_rbu_dmaaddr, - GFP_KERNEL); + &dell_rbu_dmaaddr, GFP_KERNEL); dma_alloc = 1; } @@ -405,20 +415,21 @@ static int img_update_realloc(unsigned long size) rbu_data.image_update_buffer = image_update_buffer; rbu_data.image_update_buffer_size = size; rbu_data.bios_image_size = - rbu_data.image_update_buffer_size; + rbu_data.image_update_buffer_size; rbu_data.image_update_ordernum = ordernum; rbu_data.dma_alloc = dma_alloc; rc = 0; } else { pr_debug("Not enough memory for image update:" - "size = %ld\n", size); + "size = %ld\n", size); rc = -ENOMEM; } return rc; } -static ssize_t read_packet_data(char *buffer, loff_t pos, size_t count) +static ssize_t +read_packet_data(char *buffer, loff_t pos, size_t count) { int retval; size_t bytes_left; @@ -438,7 +449,7 @@ static ssize_t read_packet_data(char *buffer, loff_t pos, size_t count) if (pos > imagesize) { retval = 0; printk(KERN_WARNING "dell_rbu:read_packet_data: " - "data underrun\n"); + "data underrun\n"); goto read_rbu_data_exit; } @@ -459,7 +470,8 @@ static ssize_t read_packet_data(char *buffer, loff_t pos, size_t count) return retval; } -static ssize_t read_rbu_mono_data(char *buffer, loff_t pos, size_t count) +static ssize_t +read_rbu_mono_data(char *buffer, loff_t pos, size_t count) { unsigned char *ptemp = NULL; size_t bytes_left = 0; @@ -468,11 +480,11 @@ static ssize_t read_rbu_mono_data(char *buffer, loff_t pos, size_t count) /* check to see if we have something to return */ if ((rbu_data.image_update_buffer == NULL) || - (rbu_data.bios_image_size == 0)) { + (rbu_data.bios_image_size == 0)) { pr_debug("read_rbu_data_mono: image_update_buffer %p ," - "bios_image_size %lu\n", - rbu_data.image_update_buffer, - rbu_data.bios_image_size); + "bios_image_size %lu\n", + rbu_data.image_update_buffer, + rbu_data.bios_image_size); ret_count = -ENOMEM; goto read_rbu_data_exit; } @@ -515,62 +527,21 @@ read_rbu_data(struct kobject *kobj, char *buffer, loff_t pos, size_t count) return ret_count; } -static ssize_t -read_rbu_image_type(struct kobject *kobj, char *buffer, loff_t pos, - size_t count) -{ - int size = 0; - if (!pos) - size = sprintf(buffer, "%s\n", image_type); - return size; -} - -static ssize_t -write_rbu_image_type(struct kobject *kobj, char *buffer, loff_t pos, - size_t count) -{ - int rc = count; - spin_lock(&rbu_data.lock); - - if (strlen(buffer) < MAX_IMAGE_LENGTH) - sscanf(buffer, "%s", image_type); - else - printk(KERN_WARNING "dell_rbu: image_type is invalid" - "max chars = %d, \n incoming str--%s-- \n", - MAX_IMAGE_LENGTH, buffer); - - /* we must free all previous allocations */ - packet_empty_list(); - img_update_free(); - - spin_unlock(&rbu_data.lock); - return rc; - -} - -static struct bin_attribute rbu_data_attr = { - .attr = {.name = "data",.owner = THIS_MODULE,.mode = 0444}, - .read = read_rbu_data, -}; - -static struct bin_attribute rbu_image_type_attr = { - .attr = {.name = "image_type",.owner = THIS_MODULE,.mode = 0644}, - .read = read_rbu_image_type, - .write = write_rbu_image_type, -}; - -static void callbackfn_rbu(const struct firmware *fw, void *context) +static void +callbackfn_rbu(const struct firmware *fw, void *context) { int rc = 0; - if (!fw || !fw->size) + if (!fw || !fw->size) { + rbu_data.entry_created = 0; return; + } spin_lock(&rbu_data.lock); if (!strcmp(image_type, "mono")) { if (!img_update_realloc(fw->size)) memcpy(rbu_data.image_update_buffer, - fw->data, fw->size); + fw->data, fw->size); } else if (!strcmp(image_type, "packet")) { if (!rbu_data.packetsize) rbu_data.packetsize = fw->size; @@ -584,26 +555,110 @@ static void callbackfn_rbu(const struct firmware *fw, void *context) spin_unlock(&rbu_data.lock); rc = request_firmware_nowait(THIS_MODULE, FW_ACTION_NOHOTPLUG, - "dell_rbu", &rbu_device->dev, - &context, callbackfn_rbu); + "dell_rbu", &rbu_device->dev, &context, callbackfn_rbu); if (rc) printk(KERN_ERR - "dell_rbu:%s request_firmware_nowait failed" - " %d\n", __FUNCTION__, rc); + "dell_rbu:%s request_firmware_nowait failed" + " %d\n", __FUNCTION__, rc); + else + rbu_data.entry_created = 1; } -static int __init dcdrbu_init(void) +static ssize_t +read_rbu_image_type(struct kobject *kobj, char *buffer, loff_t pos, + size_t count) +{ + int size = 0; + if (!pos) + size = sprintf(buffer, "%s\n", image_type); + return size; +} + +static ssize_t +write_rbu_image_type(struct kobject *kobj, char *buffer, loff_t pos, + size_t count) +{ + int rc = count; + int req_firm_rc = 0; + int i; + spin_lock(&rbu_data.lock); + /* + * Find the first newline or space + */ + for (i = 0; i < count; ++i) + if (buffer[i] == '\n' || buffer[i] == ' ') { + buffer[i] = '\0'; + break; + } + if (i == count) + buffer[count] = '\0'; + + if (strstr(buffer, "mono")) + strcpy(image_type, "mono"); + else if (strstr(buffer, "packet")) + strcpy(image_type, "packet"); + else if (strstr(buffer, "init")) { + /* + * If due to the user error the driver gets in a bad + * state where even though it is loaded , the + * /sys/class/firmware/dell_rbu entries are missing. + * to cover this situation the user can recreate entries + * by writing init to image_type. + */ + if (!rbu_data.entry_created) { + spin_unlock(&rbu_data.lock); + req_firm_rc = request_firmware_nowait(THIS_MODULE, + FW_ACTION_NOHOTPLUG, "dell_rbu", + &rbu_device->dev, &context, + callbackfn_rbu); + if (req_firm_rc) { + printk(KERN_ERR + "dell_rbu:%s request_firmware_nowait" + " failed %d\n", __FUNCTION__, rc); + rc = -EIO; + } else + rbu_data.entry_created = 1; + + spin_lock(&rbu_data.lock); + } + } else { + printk(KERN_WARNING "dell_rbu: image_type is invalid\n"); + spin_unlock(&rbu_data.lock); + return -EINVAL; + } + + /* we must free all previous allocations */ + packet_empty_list(); + img_update_free(); + spin_unlock(&rbu_data.lock); + + return rc; +} + +static struct bin_attribute rbu_data_attr = { + .attr = {.name = "data",.owner = THIS_MODULE,.mode = 0444}, + .read = read_rbu_data, +}; + +static struct bin_attribute rbu_image_type_attr = { + .attr = {.name = "image_type",.owner = THIS_MODULE,.mode = 0644}, + .read = read_rbu_image_type, + .write = write_rbu_image_type, +}; + +static int __init +dcdrbu_init(void) { int rc = 0; spin_lock_init(&rbu_data.lock); init_packet_head(); rbu_device = - platform_device_register_simple("dell_rbu", -1, NULL, 0); + platform_device_register_simple("dell_rbu", -1, NULL, 0); if (!rbu_device) { printk(KERN_ERR - "dell_rbu:%s:platform_device_register_simple " - "failed\n", __FUNCTION__); + "dell_rbu:%s:platform_device_register_simple " + "failed\n", __FUNCTION__); return -EIO; } @@ -611,17 +666,19 @@ static int __init dcdrbu_init(void) sysfs_create_bin_file(&rbu_device->dev.kobj, &rbu_image_type_attr); rc = request_firmware_nowait(THIS_MODULE, FW_ACTION_NOHOTPLUG, - "dell_rbu", &rbu_device->dev, - &context, callbackfn_rbu); + "dell_rbu", &rbu_device->dev, &context, callbackfn_rbu); if (rc) printk(KERN_ERR "dell_rbu:%s:request_firmware_nowait" - " failed %d\n", __FUNCTION__, rc); + " failed %d\n", __FUNCTION__, rc); + else + rbu_data.entry_created = 1; return rc; } -static __exit void dcdrbu_exit(void) +static __exit void +dcdrbu_exit(void) { spin_lock(&rbu_data.lock); packet_empty_list(); From dda8577fb5a00507e5aea737833190a10516b257 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 16 Sep 2005 19:28:05 -0700 Subject: [PATCH 121/240] [PATCH] dell_rbu tidy Whitespace standardisation. Cc: Abhay Salunke Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/firmware/dell_rbu.c | 68 ++++++++++++++++--------------------- 1 file changed, 30 insertions(+), 38 deletions(-) diff --git a/drivers/firmware/dell_rbu.c b/drivers/firmware/dell_rbu.c index 9c2c2b199454..b66782398258 100644 --- a/drivers/firmware/dell_rbu.c +++ b/drivers/firmware/dell_rbu.c @@ -85,8 +85,7 @@ static struct platform_device *rbu_device; static int context; static dma_addr_t dell_rbu_dmaaddr; -static void -init_packet_head(void) +static void init_packet_head(void) { INIT_LIST_HEAD(&packet_data_head.list); rbu_data.packet_write_count = 0; @@ -95,8 +94,7 @@ init_packet_head(void) rbu_data.packetsize = 0; } -static int -fill_last_packet(void *data, size_t length) +static int fill_last_packet(void *data, size_t length) { struct list_head *ptemp_list; struct packet_data *packet = NULL; @@ -138,8 +136,7 @@ fill_last_packet(void *data, size_t length) return 0; } -static int -create_packet(size_t length) +static int create_packet(size_t length) { struct packet_data *newpacket; int ordernum = 0; @@ -198,8 +195,7 @@ create_packet(size_t length) return 0; } -static int -packetize_data(void *data, size_t length) +static int packetize_data(void *data, size_t length) { int rc = 0; @@ -213,8 +209,7 @@ packetize_data(void *data, size_t length) return rc; } -static int -do_packet_read(char *data, struct list_head *ptemp_list, +static int do_packet_read(char *data, struct list_head *ptemp_list, int length, int bytes_read, int *list_read_count) { void *ptemp_buf; @@ -248,8 +243,7 @@ do_packet_read(char *data, struct list_head *ptemp_list, return bytes_copied; } -static int -packet_read_list(char *data, size_t * pread_length) +static int packet_read_list(char *data, size_t *pread_length) { struct list_head *ptemp_list; int temp_count = 0; @@ -287,8 +281,7 @@ packet_read_list(char *data, size_t * pread_length) return 0; } -static void -packet_empty_list(void) +static void packet_empty_list(void) { struct list_head *ptemp_list; struct list_head *pnext_list; @@ -320,8 +313,7 @@ packet_empty_list(void) * img_update_free: Frees the buffer allocated for storing BIOS image * Always called with lock held and returned with lock held */ -static void -img_update_free(void) +static void img_update_free(void) { if (!rbu_data.image_update_buffer) return; @@ -358,8 +350,7 @@ img_update_free(void) * already allocated size, then that memory is reused. This function is * called with lock held and returns with lock held. */ -static int -img_update_realloc(unsigned long size) +static int img_update_realloc(unsigned long size) { unsigned char *image_update_buffer = NULL; unsigned long rc; @@ -428,8 +419,7 @@ img_update_realloc(unsigned long size) return rc; } -static ssize_t -read_packet_data(char *buffer, loff_t pos, size_t count) +static ssize_t read_packet_data(char *buffer, loff_t pos, size_t count) { int retval; size_t bytes_left; @@ -470,8 +460,7 @@ read_packet_data(char *buffer, loff_t pos, size_t count) return retval; } -static ssize_t -read_rbu_mono_data(char *buffer, loff_t pos, size_t count) +static ssize_t read_rbu_mono_data(char *buffer, loff_t pos, size_t count) { unsigned char *ptemp = NULL; size_t bytes_left = 0; @@ -509,8 +498,8 @@ read_rbu_mono_data(char *buffer, loff_t pos, size_t count) return ret_count; } -static ssize_t -read_rbu_data(struct kobject *kobj, char *buffer, loff_t pos, size_t count) +static ssize_t read_rbu_data(struct kobject *kobj, char *buffer, + loff_t pos, size_t count) { ssize_t ret_count = 0; @@ -527,8 +516,7 @@ read_rbu_data(struct kobject *kobj, char *buffer, loff_t pos, size_t count) return ret_count; } -static void -callbackfn_rbu(const struct firmware *fw, void *context) +static void callbackfn_rbu(const struct firmware *fw, void *context) { int rc = 0; @@ -564,9 +552,8 @@ callbackfn_rbu(const struct firmware *fw, void *context) rbu_data.entry_created = 1; } -static ssize_t -read_rbu_image_type(struct kobject *kobj, char *buffer, loff_t pos, - size_t count) +static ssize_t read_rbu_image_type(struct kobject *kobj, char *buffer, + loff_t pos, size_t count) { int size = 0; if (!pos) @@ -574,9 +561,8 @@ read_rbu_image_type(struct kobject *kobj, char *buffer, loff_t pos, return size; } -static ssize_t -write_rbu_image_type(struct kobject *kobj, char *buffer, loff_t pos, - size_t count) +static ssize_t write_rbu_image_type(struct kobject *kobj, char *buffer, + loff_t pos, size_t count) { int rc = count; int req_firm_rc = 0; @@ -636,18 +622,25 @@ write_rbu_image_type(struct kobject *kobj, char *buffer, loff_t pos, } static struct bin_attribute rbu_data_attr = { - .attr = {.name = "data",.owner = THIS_MODULE,.mode = 0444}, + .attr = { + .name = "data", + .owner = THIS_MODULE, + .mode = 0444, + }, .read = read_rbu_data, }; static struct bin_attribute rbu_image_type_attr = { - .attr = {.name = "image_type",.owner = THIS_MODULE,.mode = 0644}, + .attr = { + .name = "image_type", + .owner = THIS_MODULE, + .mode = 0644, + }, .read = read_rbu_image_type, .write = write_rbu_image_type, }; -static int __init -dcdrbu_init(void) +static int __init dcdrbu_init(void) { int rc = 0; spin_lock_init(&rbu_data.lock); @@ -677,8 +670,7 @@ dcdrbu_init(void) } -static __exit void -dcdrbu_exit(void) +static __exit void dcdrbu_exit(void) { spin_lock(&rbu_data.lock); packet_empty_list(); From 53d2be79d5981b7efc8c5ec1169613bba95bde20 Mon Sep 17 00:00:00 2001 From: Davide Libenzi Date: Fri, 16 Sep 2005 19:28:06 -0700 Subject: [PATCH 122/240] [PATCH] epoll: fix delayed initialization bug Al found a potential problem in epoll_create(), where the file->private_data member was set after fd_install(). This is obviously wrong since another thread might do a close() on that fd# before we set the file->private_data member. This goes over 2.6.13 and passes a few basic tests I've done here. (akpm: snuck in a kzalloc() cleanup too) Signed-off-by: Davide Libenzi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/eventpoll.c | 40 ++++++++++++++++++++-------------------- 1 file changed, 20 insertions(+), 20 deletions(-) diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 6ab1dd0ca904..403b90a1213d 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -231,8 +231,9 @@ struct ep_pqueue { static void ep_poll_safewake_init(struct poll_safewake *psw); static void ep_poll_safewake(struct poll_safewake *psw, wait_queue_head_t *wq); -static int ep_getfd(int *efd, struct inode **einode, struct file **efile); -static int ep_file_init(struct file *file); +static int ep_getfd(int *efd, struct inode **einode, struct file **efile, + struct eventpoll *ep); +static int ep_alloc(struct eventpoll **pep); static void ep_free(struct eventpoll *ep); static struct epitem *ep_find(struct eventpoll *ep, struct file *file, int fd); static void ep_use_epitem(struct epitem *epi); @@ -501,38 +502,37 @@ void eventpoll_release_file(struct file *file) asmlinkage long sys_epoll_create(int size) { int error, fd; + struct eventpoll *ep; struct inode *inode; struct file *file; DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d)\n", current, size)); - /* Sanity check on the size parameter */ + /* + * Sanity check on the size parameter, and create the internal data + * structure ( "struct eventpoll" ). + */ error = -EINVAL; - if (size <= 0) + if (size <= 0 || (error = ep_alloc(&ep)) != 0) goto eexit_1; /* * Creates all the items needed to setup an eventpoll file. That is, * a file structure, and inode and a free file descriptor. */ - error = ep_getfd(&fd, &inode, &file); - if (error) - goto eexit_1; - - /* Setup the file internal data structure ( "struct eventpoll" ) */ - error = ep_file_init(file); + error = ep_getfd(&fd, &inode, &file, ep); if (error) goto eexit_2; - DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n", current, size, fd)); return fd; eexit_2: - sys_close(fd); + ep_free(ep); + kfree(ep); eexit_1: DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n", current, size, error)); @@ -706,7 +706,8 @@ eexit_1: /* * Creates the file descriptor to be used by the epoll interface. */ -static int ep_getfd(int *efd, struct inode **einode, struct file **efile) +static int ep_getfd(int *efd, struct inode **einode, struct file **efile, + struct eventpoll *ep) { struct qstr this; char name[32]; @@ -756,7 +757,7 @@ static int ep_getfd(int *efd, struct inode **einode, struct file **efile) file->f_op = &eventpoll_fops; file->f_mode = FMODE_READ; file->f_version = 0; - file->private_data = NULL; + file->private_data = ep; /* Install the new setup file into the allocated fd. */ fd_install(fd, file); @@ -777,14 +778,13 @@ eexit_1: } -static int ep_file_init(struct file *file) +static int ep_alloc(struct eventpoll **pep) { - struct eventpoll *ep; + struct eventpoll *ep = kzalloc(sizeof(*ep), GFP_KERNEL); - if (!(ep = kmalloc(sizeof(struct eventpoll), GFP_KERNEL))) + if (!ep) return -ENOMEM; - memset(ep, 0, sizeof(*ep)); rwlock_init(&ep->lock); init_rwsem(&ep->sem); init_waitqueue_head(&ep->wq); @@ -792,9 +792,9 @@ static int ep_file_init(struct file *file) INIT_LIST_HEAD(&ep->rdllist); ep->rbr = RB_ROOT; - file->private_data = ep; + *pep = ep; - DNPRINTK(3, (KERN_INFO "[%p] eventpoll: ep_file_init() ep=%p\n", + DNPRINTK(3, (KERN_INFO "[%p] eventpoll: ep_alloc() ep=%p\n", current, ep)); return 0; } From 393ad299658d8464149820363ee09bdf3fd45566 Mon Sep 17 00:00:00 2001 From: Robert Love Date: Fri, 16 Sep 2005 19:28:07 -0700 Subject: [PATCH 123/240] [PATCH] hdaps driver update - Remove the relative input device - Add an absolute input device - Misc. cleanup and bug fixing The patch is sizable due to the cleanup from removing the relative input device (net -112 lines). Signed-off-by: Robert Love Cc: Dmitry Torokhov Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/hwmon/hdaps.c | 356 ++++++++++++++---------------------------- 1 file changed, 118 insertions(+), 238 deletions(-) diff --git a/drivers/hwmon/hdaps.c b/drivers/hwmon/hdaps.c index eaebfc14c933..4c56411f3993 100644 --- a/drivers/hwmon/hdaps.c +++ b/drivers/hwmon/hdaps.c @@ -5,10 +5,10 @@ * Copyright (C) 2005 Jesper Juhl * * The HardDisk Active Protection System (hdaps) is present in the IBM ThinkPad - * T41, T42, T43, R51, and X40, at least. It provides a basic two-axis - * accelerometer and other data, such as the device's temperature. + * T41, T42, T43, R50, R50p, R51, and X40, at least. It provides a basic + * two-axis accelerometer and other data, such as the device's temperature. * - * Based on the document by Mark A. Smith available at + * This driver is based on the document by Mark A. Smith available at * http://www.almaden.ibm.com/cs/people/marksmith/tpaps.html and a lot of trial * and error. * @@ -36,12 +36,7 @@ #include #define HDAPS_LOW_PORT 0x1600 /* first port used by hdaps */ -#define HDAPS_NR_PORTS 0x30 /* 0x1600 - 0x162f */ - -#define STATE_FRESH 0x50 /* accelerometer data is fresh */ - -#define REFRESH_ASYNC 0x00 /* do asynchronous refresh */ -#define REFRESH_SYNC 0x01 /* do synchronous refresh */ +#define HDAPS_NR_PORTS 0x30 /* number of ports: 0x1600 - 0x162f */ #define HDAPS_PORT_STATE 0x1611 /* device state */ #define HDAPS_PORT_YPOS 0x1612 /* y-axis position */ @@ -53,7 +48,7 @@ #define HDAPS_PORT_UNKNOWN 0x161c /* what is this? */ #define HDAPS_PORT_KMACT 0x161d /* keyboard or mouse activity */ -#define HDAPS_READ_MASK 0xff /* some reads have the low 8 bits set */ +#define STATE_FRESH 0x50 /* accelerometer data is fresh */ #define KEYBD_MASK 0x20 /* set if keyboard activity */ #define MOUSE_MASK 0x40 /* set if mouse activity */ @@ -63,12 +58,11 @@ #define INIT_TIMEOUT_MSECS 4000 /* wait up to 4s for device init ... */ #define INIT_WAIT_MSECS 200 /* ... in 200ms increments */ -static struct platform_device *pdev; -static struct input_dev hdaps_idev; +#define HDAPS_POLL_PERIOD (HZ/20) /* poll for input every 1/20s */ +#define HDAPS_INPUT_FUZZ 4 /* input event threshold */ + static struct timer_list hdaps_timer; -static unsigned int hdaps_mousedev_threshold = 4; -static unsigned long hdaps_poll_ms = 50; -static unsigned int hdaps_mousedev; +static struct platform_device *pdev; static unsigned int hdaps_invert; static u8 km_activity; static int rest_x; @@ -81,14 +75,14 @@ static DECLARE_MUTEX(hdaps_sem); */ static inline u8 __get_latch(u16 port) { - return inb(port) & HDAPS_READ_MASK; + return inb(port) & 0xff; } /* - * __check_latch - Check a port latch for a given value. Callers must hold - * hdaps_sem. Returns zero if the port contains the given value. + * __check_latch - Check a port latch for a given value. Returns zero if the + * port contains the given value. Callers must hold hdaps_sem. */ -static inline unsigned int __check_latch(u16 port, u8 val) +static inline int __check_latch(u16 port, u8 val) { if (__get_latch(port) == val) return 0; @@ -99,7 +93,7 @@ static inline unsigned int __check_latch(u16 port, u8 val) * __wait_latch - Wait up to 100us for a port latch to get a certain value, * returning zero if the value is obtained. Callers must hold hdaps_sem. */ -static unsigned int __wait_latch(u16 port, u8 val) +static int __wait_latch(u16 port, u8 val) { unsigned int i; @@ -109,59 +103,42 @@ static unsigned int __wait_latch(u16 port, u8 val) udelay(5); } - return -EINVAL; + return -EIO; } /* - * __device_refresh - Request a refresh from the accelerometer. - * - * If sync is REFRESH_SYNC, we perform a synchronous refresh and will wait. - * Returns zero if successful and nonzero on error. - * - * If sync is REFRESH_ASYNC, we merely kick off a new refresh if the device is - * not up-to-date. Always returns zero. - * - * Callers must hold hdaps_sem. + * __device_refresh - request a refresh from the accelerometer. Does not wait + * for refresh to complete. Callers must hold hdaps_sem. */ -static int __device_refresh(unsigned int sync) +static void __device_refresh(void) { - u8 state; - - udelay(100); - - state = inb(0x1604); - if (state == STATE_FRESH) - return 0; - - outb(0x11, 0x1610); - outb(0x01, 0x161f); - if (sync == REFRESH_ASYNC) - return 0; + udelay(200); + if (inb(0x1604) != STATE_FRESH) { + outb(0x11, 0x1610); + outb(0x01, 0x161f); + } +} +/* + * __device_refresh_sync - request a synchronous refresh from the + * accelerometer. We wait for the refresh to complete. Returns zero if + * successful and nonzero on error. Callers must hold hdaps_sem. + */ +static int __device_refresh_sync(void) +{ + __device_refresh(); return __wait_latch(0x1604, STATE_FRESH); } /* - * __device_complete - Indicate to the accelerometer that we are done reading + * __device_complete - indicate to the accelerometer that we are done reading * data, and then initiate an async refresh. Callers must hold hdaps_sem. */ static inline void __device_complete(void) { inb(0x161f); inb(0x1604); - __device_refresh(REFRESH_ASYNC); -} - -static int __hdaps_readb_one(unsigned int port, u8 *val) -{ - /* do a sync refresh -- we need to be sure that we read fresh data */ - if (__device_refresh(REFRESH_SYNC)) - return -EIO; - - *val = inb(port); - __device_complete(); - - return 0; + __device_refresh(); } /* @@ -174,17 +151,26 @@ static int hdaps_readb_one(unsigned int port, u8 *val) int ret; down(&hdaps_sem); - ret = __hdaps_readb_one(port, val); - up(&hdaps_sem); + /* do a sync refresh -- we need to be sure that we read fresh data */ + ret = __device_refresh_sync(); + if (ret) + goto out; + + *val = inb(port); + __device_complete(); + +out: + up(&hdaps_sem); return ret; } +/* __hdaps_read_pair - internal lockless helper for hdaps_read_pair(). */ static int __hdaps_read_pair(unsigned int port1, unsigned int port2, int *x, int *y) { /* do a sync refresh -- we need to be sure that we read fresh data */ - if (__device_refresh(REFRESH_SYNC)) + if (__device_refresh_sync()) return -EIO; *y = inw(port2); @@ -217,11 +203,13 @@ static int hdaps_read_pair(unsigned int port1, unsigned int port2, return ret; } -/* initialize the accelerometer */ +/* + * hdaps_device_init - initialize the accelerometer. Returns zero on success + * and negative error code on failure. Can sleep. + */ static int hdaps_device_init(void) { - unsigned int total_msecs = INIT_TIMEOUT_MSECS; - int ret = -ENXIO; + int total, ret = -ENXIO; down(&hdaps_sem); @@ -231,8 +219,10 @@ static int hdaps_device_init(void) goto out; /* - * The 0x03 value appears to only work on some thinkpads, such as the - * T42p. Others return 0x01. + * Most ThinkPads return 0x01. + * + * Others--namely the R50p, T41p, and T42p--return 0x03. These laptops + * have "inverted" axises. * * The 0x02 value occurs when the chip has been previously initialized. */ @@ -267,24 +257,23 @@ static int hdaps_device_init(void) outb(0x01, 0x161f); if (__wait_latch(0x161f, 0x00)) goto out; - if (__device_refresh(REFRESH_SYNC)) + if (__device_refresh_sync()) goto out; if (__wait_latch(0x1611, 0x00)) goto out; /* we have done our dance, now let's wait for the applause */ - while (total_msecs > 0) { - u8 ignored; + for (total = INIT_TIMEOUT_MSECS; total > 0; total -= INIT_WAIT_MSECS) { + int x, y; /* a read of the device helps push it into action */ - __hdaps_readb_one(HDAPS_PORT_UNKNOWN, &ignored); + __hdaps_read_pair(HDAPS_PORT_XPOS, HDAPS_PORT_YPOS, &x, &y); if (!__wait_latch(0x1611, 0x02)) { ret = 0; break; } msleep(INIT_WAIT_MSECS); - total_msecs -= INIT_WAIT_MSECS; } out: @@ -293,96 +282,6 @@ out: } -/* Input class stuff */ - -/* - * hdaps_calibrate - Zero out our "resting" values. Callers must hold hdaps_sem. - */ -static void hdaps_calibrate(void) -{ - int x, y; - - if (__hdaps_read_pair(HDAPS_PORT_XPOS, HDAPS_PORT_YPOS, &x, &y)) - return; - - rest_x = x; - rest_y = y; -} - -static void hdaps_mousedev_poll(unsigned long unused) -{ - int x, y; - - /* Cannot sleep. Try nonblockingly. If we fail, try again later. */ - if (down_trylock(&hdaps_sem)) { - mod_timer(&hdaps_timer,jiffies+msecs_to_jiffies(hdaps_poll_ms)); - return; - } - - if (__hdaps_read_pair(HDAPS_PORT_XPOS, HDAPS_PORT_YPOS, &x, &y)) - goto out; - - x -= rest_x; - y -= rest_y; - if (abs(x) > hdaps_mousedev_threshold) - input_report_rel(&hdaps_idev, REL_X, x); - if (abs(y) > hdaps_mousedev_threshold) - input_report_rel(&hdaps_idev, REL_Y, y); - input_sync(&hdaps_idev); - - mod_timer(&hdaps_timer, jiffies + msecs_to_jiffies(hdaps_poll_ms)); - -out: - up(&hdaps_sem); -} - -/* - * hdaps_mousedev_enable - enable the input class device. Can sleep. - */ -static void hdaps_mousedev_enable(void) -{ - down(&hdaps_sem); - - /* calibrate the device before enabling */ - hdaps_calibrate(); - - /* initialize the input class */ - init_input_dev(&hdaps_idev); - hdaps_idev.dev = &pdev->dev; - hdaps_idev.evbit[0] = BIT(EV_KEY) | BIT(EV_REL); - hdaps_idev.relbit[0] = BIT(REL_X) | BIT(REL_Y); - hdaps_idev.keybit[LONG(BTN_LEFT)] = BIT(BTN_LEFT); - input_register_device(&hdaps_idev); - - /* start up our timer */ - init_timer(&hdaps_timer); - hdaps_timer.function = hdaps_mousedev_poll; - hdaps_timer.expires = jiffies + msecs_to_jiffies(hdaps_poll_ms); - add_timer(&hdaps_timer); - - hdaps_mousedev = 1; - - up(&hdaps_sem); - - printk(KERN_INFO "hdaps: input device enabled.\n"); -} - -/* - * hdaps_mousedev_disable - disable the input class device. Caller must hold - * hdaps_sem. - */ -static void hdaps_mousedev_disable(void) -{ - down(&hdaps_sem); - if (hdaps_mousedev) { - hdaps_mousedev = 0; - del_timer_sync(&hdaps_timer); - input_unregister_device(&hdaps_idev); - } - up(&hdaps_sem); -} - - /* Device model stuff */ static int hdaps_probe(struct device *dev) @@ -412,6 +311,49 @@ static struct device_driver hdaps_driver = { .resume = hdaps_resume }; +/* Input class stuff */ + +static struct input_dev hdaps_idev = { + .name = "hdaps", + .evbit = { BIT(EV_ABS) }, + .absbit = { BIT(ABS_X) | BIT(ABS_Y) }, + .absmin = { [ABS_X] = -256, [ABS_Y] = -256 }, + .absmax = { [ABS_X] = 256, [ABS_Y] = 256 }, + .absfuzz = { [ABS_X] = HDAPS_INPUT_FUZZ, [ABS_Y] = HDAPS_INPUT_FUZZ }, + .absflat = { [ABS_X] = HDAPS_INPUT_FUZZ, [ABS_Y] = HDAPS_INPUT_FUZZ }, +}; + +/* + * hdaps_calibrate - Set our "resting" values. Callers must hold hdaps_sem. + */ +static void hdaps_calibrate(void) +{ + __hdaps_read_pair(HDAPS_PORT_XPOS, HDAPS_PORT_YPOS, &rest_x, &rest_y); +} + +static void hdaps_mousedev_poll(unsigned long unused) +{ + int x, y; + + /* Cannot sleep. Try nonblockingly. If we fail, try again later. */ + if (down_trylock(&hdaps_sem)) { + mod_timer(&hdaps_timer,jiffies + HDAPS_POLL_PERIOD); + return; + } + + if (__hdaps_read_pair(HDAPS_PORT_XPOS, HDAPS_PORT_YPOS, &x, &y)) + goto out; + + input_report_abs(&hdaps_idev, ABS_X, x - rest_x); + input_report_abs(&hdaps_idev, ABS_Y, y - rest_y); + input_sync(&hdaps_idev); + + mod_timer(&hdaps_timer, jiffies + HDAPS_POLL_PERIOD); + +out: + up(&hdaps_sem); +} + /* Sysfs Files */ @@ -517,69 +459,6 @@ static ssize_t hdaps_invert_store(struct device *dev, return count; } -static ssize_t hdaps_mousedev_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - return sprintf(buf, "%d\n", hdaps_mousedev); -} - -static ssize_t hdaps_mousedev_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int enable; - - if (sscanf(buf, "%d", &enable) != 1) - return -EINVAL; - - if (enable == 1) - hdaps_mousedev_enable(); - else if (enable == 0) - hdaps_mousedev_disable(); - else - return -EINVAL; - - return count; -} - -static ssize_t hdaps_poll_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - return sprintf(buf, "%lu\n", hdaps_poll_ms); -} - -static ssize_t hdaps_poll_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) -{ - unsigned int poll; - - if (sscanf(buf, "%u", &poll) != 1 || poll == 0) - return -EINVAL; - hdaps_poll_ms = poll; - - return count; -} - -static ssize_t hdaps_threshold_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - return sprintf(buf, "%u\n", hdaps_mousedev_threshold); -} - -static ssize_t hdaps_threshold_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) -{ - unsigned int threshold; - - if (sscanf(buf, "%u", &threshold) != 1 || threshold == 0) - return -EINVAL; - hdaps_mousedev_threshold = threshold; - - return count; -} - static DEVICE_ATTR(position, 0444, hdaps_position_show, NULL); static DEVICE_ATTR(variance, 0444, hdaps_variance_show, NULL); static DEVICE_ATTR(temp1, 0444, hdaps_temp1_show, NULL); @@ -588,10 +467,6 @@ static DEVICE_ATTR(keyboard_activity, 0444, hdaps_keyboard_activity_show, NULL); static DEVICE_ATTR(mouse_activity, 0444, hdaps_mouse_activity_show, NULL); static DEVICE_ATTR(calibrate, 0644, hdaps_calibrate_show,hdaps_calibrate_store); static DEVICE_ATTR(invert, 0644, hdaps_invert_show, hdaps_invert_store); -static DEVICE_ATTR(mousedev, 0644, hdaps_mousedev_show, hdaps_mousedev_store); -static DEVICE_ATTR(mousedev_poll_ms, 0644, hdaps_poll_show, hdaps_poll_store); -static DEVICE_ATTR(mousedev_threshold, 0644, hdaps_threshold_show, - hdaps_threshold_store); static struct attribute *hdaps_attributes[] = { &dev_attr_position.attr, @@ -601,9 +476,6 @@ static struct attribute *hdaps_attributes[] = { &dev_attr_keyboard_activity.attr, &dev_attr_mouse_activity.attr, &dev_attr_calibrate.attr, - &dev_attr_mousedev.attr, - &dev_attr_mousedev_threshold.attr, - &dev_attr_mousedev_poll_ms.attr, &dev_attr_invert.attr, NULL, }; @@ -619,7 +491,7 @@ static struct attribute_group hdaps_attribute_group = { * XXX: We should be able to return nonzero and halt the detection process. * But there is a bug in dmi_check_system() where a nonzero return from the * first match will result in a return of failure from dmi_check_system(). - * I fixed this; the patch is in 2.6-mm. Once in Linus's tree we can make + * I fixed this; the patch is 2.6-git. Once in a released tree, we can make * hdaps_dmi_match_invert() return hdaps_dmi_match(), which in turn returns 1. */ static int hdaps_dmi_match(struct dmi_system_id *id) @@ -668,6 +540,7 @@ static int __init hdaps_init(void) HDAPS_DMI_MATCH_NORMAL("ThinkPad T42"), HDAPS_DMI_MATCH_NORMAL("ThinkPad T43"), HDAPS_DMI_MATCH_NORMAL("ThinkPad X40"), + HDAPS_DMI_MATCH_NORMAL("ThinkPad X41 Tablet"), { .ident = NULL } }; @@ -696,8 +569,18 @@ static int __init hdaps_init(void) if (ret) goto out_device; - if (hdaps_mousedev) - hdaps_mousedev_enable(); + /* initial calibrate for the input device */ + hdaps_calibrate(); + + /* initialize the input class */ + hdaps_idev.dev = &pdev->dev; + input_register_device(&hdaps_idev); + + /* start up our timer for the input device */ + init_timer(&hdaps_timer); + hdaps_timer.function = hdaps_mousedev_poll; + hdaps_timer.expires = jiffies + HDAPS_POLL_PERIOD; + add_timer(&hdaps_timer); printk(KERN_INFO "hdaps: driver successfully loaded.\n"); return 0; @@ -715,8 +598,8 @@ out: static void __exit hdaps_exit(void) { - hdaps_mousedev_disable(); - + del_timer_sync(&hdaps_timer); + input_unregister_device(&hdaps_idev); sysfs_remove_group(&pdev->dev.kobj, &hdaps_attribute_group); platform_device_unregister(pdev); driver_unregister(&hdaps_driver); @@ -728,9 +611,6 @@ static void __exit hdaps_exit(void) module_init(hdaps_init); module_exit(hdaps_exit); -module_param_named(mousedev, hdaps_mousedev, bool, 0); -MODULE_PARM_DESC(mousedev, "enable the input class device"); - module_param_named(invert, hdaps_invert, bool, 0); MODULE_PARM_DESC(invert, "invert data along each axis"); From 8ac2120d90273c590cf7662f03d103519101685b Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Fri, 16 Sep 2005 19:28:08 -0700 Subject: [PATCH 124/240] [PATCH] i2c: kill an unused i2c_adapter struct member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Kill an unused member of the i2c_adapter structure. This additionally fixes a potential bug, because doesn't include , so different files including could see a different definition of the i2c_adapter structure, depending on them including (or other header files themselves including ) before , or not. Credits go to Jörn Engel for pointing me to the problem. Signed-off-by: Jean Delvare Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/i2c.h | 5 ----- 1 file changed, 5 deletions(-) diff --git a/include/linux/i2c.h b/include/linux/i2c.h index be35332b67e6..3d49a305bf88 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -230,11 +230,6 @@ struct i2c_adapter { struct device dev; /* the adapter device */ struct class_device class_dev; /* the class device */ -#ifdef CONFIG_PROC_FS - /* No need to set this when you initialize the adapter */ - int inode; -#endif /* def CONFIG_PROC_FS */ - int nr; struct list_head clients; struct list_head list; From a464adeb7e8f1cd65ca911e20a7c02e452dc2c17 Mon Sep 17 00:00:00 2001 From: Zach Brown Date: Fri, 16 Sep 2005 19:28:09 -0700 Subject: [PATCH 125/240] [PATCH] Add smp_mb__after_clear_bit() to unlock_kiocb() Add smp_mb__after_clear_bit() to unlock_kiocb() AIO's use of wait_on_bit_lock()/wake_up_bit() forgot to add a barrier between clearing its lock bit and calling wake_up_bit() so wake_up_bit()'s unlocked waitqueue_active() can race. This puts AIO's use in line with the others and the comment above wake_up_bit(). Signed-off-by: Zach Brown Acked-by: Benjamin LaHaise Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/aio.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/aio.c b/fs/aio.c index 38f62680fd63..0e11e31dbb77 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -562,6 +562,7 @@ static inline void lock_kiocb(struct kiocb *iocb) static inline void unlock_kiocb(struct kiocb *iocb) { kiocbClearLocked(iocb); + smp_mb__after_clear_bit(); wake_up_bit(&iocb->ki_flags, KIF_LOCKED); } From f647e08a55d2c88c4e7ab17a0a8e3fcf568fbc65 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 16 Sep 2005 19:28:10 -0700 Subject: [PATCH 126/240] [PATCH] joystick-vs-x.org fix Fix http://bugzilla.kernel.org/show_bug.cgi?id=5241 2.6.13 broke compilation of the xorg tree, which apprarently insists on including that file. Cc: Vojtech Pavlik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/joystick.h | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) diff --git a/include/linux/joystick.h b/include/linux/joystick.h index 06b9af77eb7f..5fd20ddd7ae3 100644 --- a/include/linux/joystick.h +++ b/include/linux/joystick.h @@ -111,29 +111,30 @@ struct js_corr { #define JS_SET_ALL 8 struct JS_DATA_TYPE { - __s32 buttons; - __s32 x; - __s32 y; + int32_t buttons; + int32_t x; + int32_t y; }; struct JS_DATA_SAVE_TYPE_32 { - __s32 JS_TIMEOUT; - __s32 BUSY; - __s32 JS_EXPIRETIME; - __s32 JS_TIMELIMIT; + int32_t JS_TIMEOUT; + int32_t BUSY; + int32_t JS_EXPIRETIME; + int32_t JS_TIMELIMIT; struct JS_DATA_TYPE JS_SAVE; struct JS_DATA_TYPE JS_CORR; }; struct JS_DATA_SAVE_TYPE_64 { - __s32 JS_TIMEOUT; - __s32 BUSY; - __s64 JS_EXPIRETIME; - __s64 JS_TIMELIMIT; + int32_t JS_TIMEOUT; + int32_t BUSY; + int64_t JS_EXPIRETIME; + int64_t JS_TIMELIMIT; struct JS_DATA_TYPE JS_SAVE; struct JS_DATA_TYPE JS_CORR; }; +#ifdef __KERNEL__ #if BITS_PER_LONG == 64 #define JS_DATA_SAVE_TYPE JS_DATA_SAVE_TYPE_64 #elif BITS_PER_LONG == 32 @@ -141,5 +142,6 @@ struct JS_DATA_SAVE_TYPE_64 { #else #error Unexpected BITS_PER_LONG #endif +#endif #endif /* _LINUX_JOYSTICK_H */ From af4e5a218e18ad588d60a4f9d6f8fb5db1a32587 Mon Sep 17 00:00:00 2001 From: Pekka J Enberg Date: Fri, 16 Sep 2005 19:28:11 -0700 Subject: [PATCH 127/240] [PATCH] CodingStyle: memory allocation This patch adds a new chapter on memory allocation to Documentation/CodingStyle. Signed-off-by: Pekka Enberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/CodingStyle | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) diff --git a/Documentation/CodingStyle b/Documentation/CodingStyle index 22e5f9036f3c..eb7db3c19227 100644 --- a/Documentation/CodingStyle +++ b/Documentation/CodingStyle @@ -410,7 +410,26 @@ Kernel messages do not have to be terminated with a period. Printing numbers in parentheses (%d) adds no value and should be avoided. - Chapter 13: References + Chapter 13: Allocating memory + +The kernel provides the following general purpose memory allocators: +kmalloc(), kzalloc(), kcalloc(), and vmalloc(). Please refer to the API +documentation for further information about them. + +The preferred form for passing a size of a struct is the following: + + p = kmalloc(sizeof(*p), ...); + +The alternative form where struct name is spelled out hurts readability and +introduces an opportunity for a bug when the pointer variable type is changed +but the corresponding sizeof that is passed to a memory allocator is not. + +Casting the return value which is a void pointer is redundant. The conversion +from void pointer to any other pointer type is guaranteed by the C programming +language. + + + Chapter 14: References The C Programming Language, Second Edition by Brian W. Kernighan and Dennis M. Ritchie. From 4fb3a53860cee2aaaf81186c451b7da0b95b45c1 Mon Sep 17 00:00:00 2001 From: Dipankar Sarma Date: Fri, 16 Sep 2005 19:28:13 -0700 Subject: [PATCH 128/240] [PATCH] files: fix preemption issues With the new fdtable locking rules, you have to protect fdtable with either ->file_lock or rcu_read_lock/unlock(). There are some places where we aren't doing either. This patch fixes those places. Signed-off-by: Dipankar Sarma Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/kernel/osf_sys.c | 7 ++++++- arch/ia64/kernel/perfmon.c | 3 ++- fs/locks.c | 3 +++ fs/proc/array.c | 3 +++ kernel/exit.c | 6 ++++++ 5 files changed, 20 insertions(+), 2 deletions(-) diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index 0636116210d2..01fe990d3e54 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -37,6 +37,7 @@ #include #include #include +#include #include #include @@ -975,6 +976,7 @@ osf_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, long timeout; int ret = -EINVAL; struct fdtable *fdt; + int max_fdset; timeout = MAX_SCHEDULE_TIMEOUT; if (tvp) { @@ -996,8 +998,11 @@ osf_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, } } + rcu_read_lock(); fdt = files_fdtable(current->files); - if (n < 0 || n > fdt->max_fdset) + max_fdset = fdt->max_fdset; + rcu_read_unlock(); + if (n < 0 || n > max_fdset) goto out_nofds; /* diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index af42cda6be80..d71731ee5b61 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -2218,12 +2218,13 @@ static void pfm_free_fd(int fd, struct file *file) { struct files_struct *files = current->files; - struct fdtable *fdt = files_fdtable(files); + struct fdtable *fdt; /* * there ie no fd_uninstall(), so we do it here */ spin_lock(&files->file_lock); + fdt = files_fdtable(files); rcu_assign_pointer(fdt->fd[fd], NULL); spin_unlock(&files->file_lock); diff --git a/fs/locks.c b/fs/locks.c index c2c09b4798d6..f7daa5f48949 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -124,6 +124,7 @@ #include #include #include +#include #include #include @@ -2205,6 +2206,7 @@ void steal_locks(fl_owner_t from) lock_kernel(); j = 0; + rcu_read_lock(); fdt = files_fdtable(files); for (;;) { unsigned long set; @@ -2222,6 +2224,7 @@ void steal_locks(fl_owner_t from) set >>= 1; } } + rcu_read_unlock(); unlock_kernel(); } EXPORT_SYMBOL(steal_locks); diff --git a/fs/proc/array.c b/fs/proc/array.c index d88d518d30f6..d84eecacbeaf 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -74,6 +74,7 @@ #include #include #include +#include #include #include @@ -180,12 +181,14 @@ static inline char * task_state(struct task_struct *p, char *buffer) p->gid, p->egid, p->sgid, p->fsgid); read_unlock(&tasklist_lock); task_lock(p); + rcu_read_lock(); if (p->files) fdt = files_fdtable(p->files); buffer += sprintf(buffer, "FDSize:\t%d\n" "Groups:\t", fdt ? fdt->max_fds : 0); + rcu_read_unlock(); group_info = p->group_info; get_group_info(group_info); diff --git a/kernel/exit.c b/kernel/exit.c index 6d2089a1bce7..ee6d8b8abef5 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -371,6 +371,12 @@ static inline void close_files(struct files_struct * files) struct fdtable *fdt; j = 0; + + /* + * It is safe to dereference the fd table without RCU or + * ->file_lock because this is the last reference to the + * files structure. + */ fdt = files_fdtable(files); for (;;) { unsigned long set; From ef402268f7c9ab1872cafa1e638eb78a75b7c18f Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Fri, 16 Sep 2005 19:28:13 -0700 Subject: [PATCH 129/240] [PATCH] FAT: miss-sync issues on sync mount (miss-sync on write) This patch fixes miss-sync issue on write() system call. This updates inode attrs flags, mtime and ctime on every comit_write call, due to locking. Signed-off-by: Hiroyuki Machida Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/file.c | 37 ++----------------------------------- fs/fat/inode.c | 15 ++++++++++++++- 2 files changed, 16 insertions(+), 36 deletions(-) diff --git a/fs/fat/file.c b/fs/fat/file.c index 62ffa9139400..7134403d5be2 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -12,39 +12,6 @@ #include #include -static ssize_t fat_file_aio_write(struct kiocb *iocb, const char __user *buf, - size_t count, loff_t pos) -{ - struct inode *inode = iocb->ki_filp->f_dentry->d_inode; - int retval; - - retval = generic_file_aio_write(iocb, buf, count, pos); - if (retval > 0) { - inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; - MSDOS_I(inode)->i_attrs |= ATTR_ARCH; - mark_inode_dirty(inode); -// check the locking rules -// if (IS_SYNC(inode)) -// fat_sync_inode(inode); - } - return retval; -} - -static ssize_t fat_file_writev(struct file *filp, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos) -{ - struct inode *inode = filp->f_dentry->d_inode; - int retval; - - retval = generic_file_writev(filp, iov, nr_segs, ppos); - if (retval > 0) { - inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; - MSDOS_I(inode)->i_attrs |= ATTR_ARCH; - mark_inode_dirty(inode); - } - return retval; -} - int fat_generic_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg) { @@ -148,9 +115,9 @@ struct file_operations fat_file_operations = { .read = do_sync_read, .write = do_sync_write, .readv = generic_file_readv, - .writev = fat_file_writev, + .writev = generic_file_writev, .aio_read = generic_file_aio_read, - .aio_write = fat_file_aio_write, + .aio_write = generic_file_aio_write, .mmap = generic_file_mmap, .ioctl = fat_generic_ioctl, .fsync = file_fsync, diff --git a/fs/fat/inode.c b/fs/fat/inode.c index a7cbe68e2259..51b1d15d9d5c 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -102,6 +102,19 @@ static int fat_prepare_write(struct file *file, struct page *page, &MSDOS_I(page->mapping->host)->mmu_private); } +static int fat_commit_write(struct file *file, struct page *page, + unsigned from, unsigned to) +{ + struct inode *inode = page->mapping->host; + int err = generic_commit_write(file, page, from, to); + if (!err && !(MSDOS_I(inode)->i_attrs & ATTR_ARCH)) { + inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; + MSDOS_I(inode)->i_attrs |= ATTR_ARCH; + mark_inode_dirty(inode); + } + return err; +} + static sector_t _fat_bmap(struct address_space *mapping, sector_t block) { return generic_block_bmap(mapping, block, fat_get_block); @@ -112,7 +125,7 @@ static struct address_space_operations fat_aops = { .writepage = fat_writepage, .sync_page = block_sync_page, .prepare_write = fat_prepare_write, - .commit_write = generic_commit_write, + .commit_write = fat_commit_write, .bmap = _fat_bmap }; From 9564df1ff3c61f41bdc54283715417ccdfb6782b Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 16 Sep 2005 19:28:15 -0700 Subject: [PATCH 130/240] [PATCH] fix pf request handling Here's the patch from http://bugzilla.kernel.org/show_bug.cgi?id=4853 It is a feeble attempt at fixing the request handling in pf, it is totally foobar right now. Signed-off-by: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/paride/pf.c | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) diff --git a/drivers/block/paride/pf.c b/drivers/block/paride/pf.c index 711d2f314ac3..94af920465b5 100644 --- a/drivers/block/paride/pf.c +++ b/drivers/block/paride/pf.c @@ -750,6 +750,14 @@ static int pf_ready(void) static struct request_queue *pf_queue; +static void pf_end_request(int uptodate) +{ + if (pf_req) { + end_request(pf_req, uptodate); + pf_req = NULL; + } +} + static void do_pf_request(request_queue_t * q) { if (pf_busy) @@ -765,7 +773,7 @@ repeat: pf_count = pf_req->current_nr_sectors; if (pf_block + pf_count > get_capacity(pf_req->rq_disk)) { - end_request(pf_req, 0); + pf_end_request(0); goto repeat; } @@ -780,7 +788,7 @@ repeat: pi_do_claimed(pf_current->pi, do_pf_write); else { pf_busy = 0; - end_request(pf_req, 0); + pf_end_request(0); goto repeat; } } @@ -798,9 +806,11 @@ static int pf_next_buf(void) if (!pf_count) return 1; spin_lock_irqsave(&pf_spin_lock, saved_flags); - end_request(pf_req, 1); - pf_count = pf_req->current_nr_sectors; - pf_buf = pf_req->buffer; + pf_end_request(1); + if (pf_req) { + pf_count = pf_req->current_nr_sectors; + pf_buf = pf_req->buffer; + } spin_unlock_irqrestore(&pf_spin_lock, saved_flags); return 1; } @@ -810,7 +820,7 @@ static inline void next_request(int success) unsigned long saved_flags; spin_lock_irqsave(&pf_spin_lock, saved_flags); - end_request(pf_req, success); + pf_end_request(success); pf_busy = 0; do_pf_request(pf_queue); spin_unlock_irqrestore(&pf_spin_lock, saved_flags); From ef0f6a437f92fbd96c725086555b37b29371e107 Mon Sep 17 00:00:00 2001 From: Timothy Thelin Date: Fri, 16 Sep 2005 19:28:16 -0700 Subject: [PATCH 131/240] [PATCH] ide: fix null request pointer for taskfile ioctl When doing ioctl HDIO_DRIVE_TASKFILE, the ide_task_t's request pointer is never set, but flagged_taskfile and do_rw_taskfile pass it as a parameter to the prehandler. The kernel will oops taskfile pio-out commands because of this (taskfile pio-in doesn't use a prehandler). This fix sets the request pointer at the time the request is created to stop this oops. Signed-off-by: Timothy Thelin Cc: "Bartlomiej Zolnierkiewicz" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/ide/ide-taskfile.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c index d04f62ab5de1..ace8edad6e96 100644 --- a/drivers/ide/ide-taskfile.c +++ b/drivers/ide/ide-taskfile.c @@ -500,6 +500,7 @@ static int ide_diag_taskfile(ide_drive_t *drive, ide_task_t *args, unsigned long } rq.special = args; + args->rq = &rq; return ide_do_drive_cmd(drive, &rq, ide_wait); } From d648daca1170945600756b31866d0362a67f2aae Mon Sep 17 00:00:00 2001 From: Adam Kropelin Date: Fri, 16 Sep 2005 19:28:17 -0700 Subject: [PATCH 132/240] [PATCH] ibmphp: Use dword accessors for PCI_ROM_ADDRESS PCI_ROM_ADDRESS is a 32 bit register and as such should be accessed using pci_bus_{read,write}_config_dword(). A recent audit of drivers/ turned up several cases of byte- and word-sized accesses. The harmful ones were fixed by Linus directly. This patches up one of the remaining harmless-but-still-wrong cases caught in the dragnet. Signed-off-by: Adam Kropelin Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/pci/hotplug/ibmphp_pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/hotplug/ibmphp_pci.c b/drivers/pci/hotplug/ibmphp_pci.c index 8122fe734aa7..b1ba429e0a2d 100644 --- a/drivers/pci/hotplug/ibmphp_pci.c +++ b/drivers/pci/hotplug/ibmphp_pci.c @@ -558,7 +558,7 @@ static int configure_device (struct pci_func *func) pci_bus_write_config_byte (ibmphp_pci_bus, devfn, PCI_CACHE_LINE_SIZE, CACHE); pci_bus_write_config_byte (ibmphp_pci_bus, devfn, PCI_LATENCY_TIMER, LATENCY); - pci_bus_write_config_word (ibmphp_pci_bus, devfn, PCI_ROM_ADDRESS, 0x00L); + pci_bus_write_config_dword (ibmphp_pci_bus, devfn, PCI_ROM_ADDRESS, 0x00L); pci_bus_write_config_word (ibmphp_pci_bus, devfn, PCI_COMMAND, DEVICEENABLE); return 0; From c2fa4f4ad8ddf4b9fac344f1da7a25d4868f08f5 Mon Sep 17 00:00:00 2001 From: Adam Kropelin Date: Fri, 16 Sep 2005 19:28:18 -0700 Subject: [PATCH 133/240] [PATCH] pciehp: Use dword accessors for PCI_ROM_ADDRESS PCI_ROM_ADDRESS is a 32 bit register and as such should be accessed using pci_bus_{read,write}_config_dword(). A recent audit of drivers/ turned up several cases of byte- and word-sized accesses. The harmful ones were fixed by Linus directly. This patches up one of the remaining harmless-but-still-wrong cases caught in the dragnet. Signed-off-by: Adam Kropelin Cc: Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/pci/hotplug/pciehp_ctrl.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index 0e0947601526..898f6da6f0de 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -2526,7 +2526,6 @@ configure_new_function(struct controller *ctrl, struct pci_func *func, int cloop; u8 temp_byte; u8 class_code; - u16 temp_word; u32 rc; u32 temp_register; u32 base; @@ -2682,8 +2681,7 @@ configure_new_function(struct controller *ctrl, struct pci_func *func, } /* End of base register loop */ /* disable ROM base Address */ - temp_word = 0x00L; - rc = pci_bus_write_config_word (pci_bus, devfn, PCI_ROM_ADDRESS, temp_word); + rc = pci_bus_write_config_dword (pci_bus, devfn, PCI_ROM_ADDRESS, 0x00); /* Set HP parameters (Cache Line Size, Latency Timer) */ rc = pciehprm_set_hpp(ctrl, func, PCI_HEADER_TYPE_NORMAL); From 06c6d271f41ffa20f2dadc9bfe100a89f7f1dd1d Mon Sep 17 00:00:00 2001 From: Adam Kropelin Date: Fri, 16 Sep 2005 19:28:19 -0700 Subject: [PATCH 134/240] [PATCH] shpchp: Use dword accessors for PCI_ROM_ADDRESS PCI_ROM_ADDRESS is a 32 bit register and as such should be accessed using pci_bus_{read,write}_config_dword(). A recent audit of drivers/ turned up several cases of byte- and word-sized accesses. The harmful ones were fixed by Linus directly. This patches up one of the remaining harmless-but-still-wrong cases caught in the dragnet. Signed-off-by: Adam Kropelin Cc: Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/pci/hotplug/shpchp_ctrl.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/pci/hotplug/shpchp_ctrl.c b/drivers/pci/hotplug/shpchp_ctrl.c index 783b5abb0717..91c9903e621f 100644 --- a/drivers/pci/hotplug/shpchp_ctrl.c +++ b/drivers/pci/hotplug/shpchp_ctrl.c @@ -2824,8 +2824,7 @@ static int configure_new_function (struct controller * ctrl, struct pci_func * f } #endif /* Disable ROM base Address */ - temp_word = 0x00L; - rc = pci_bus_write_config_word (pci_bus, devfn, PCI_ROM_ADDRESS, temp_word); + rc = pci_bus_write_config_dword (pci_bus, devfn, PCI_ROM_ADDRESS, 0x00); /* Set HP parameters (Cache Line Size, Latency Timer) */ rc = shpchprm_set_hpp(ctrl, func, PCI_HEADER_TYPE_NORMAL); From 27b2f6792f3b482e4636818c7efaf4c43fef32a3 Mon Sep 17 00:00:00 2001 From: Adam Kropelin Date: Fri, 16 Sep 2005 19:28:20 -0700 Subject: [PATCH 135/240] [PATCH] qla2xxx: Use dword accessors for PCI_ROM_ADDRESS PCI_ROM_ADDRESS is a 32 bit register and as such should be accessed using pci_bus_{read,write}_config_dword(). A recent audit of drivers/ turned up several cases of byte- and word-sized accesses. The harmful ones were fixed by Linus directly. This patches up one of the remaining harmless-but-still-wrong cases caught in the dragnet. Signed-off-by: Adam Kropelin Cc: Andrew Vasquez Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/scsi/qla2xxx/qla_init.c | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index 3e9b64137873..23d095d3817b 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -201,6 +201,7 @@ int qla2100_pci_config(scsi_qla_host_t *ha) { uint16_t w, mwi; + uint32_t d; unsigned long flags; struct device_reg_2xxx __iomem *reg = &ha->iobase->isp; @@ -215,9 +216,9 @@ qla2100_pci_config(scsi_qla_host_t *ha) pci_write_config_word(ha->pdev, PCI_COMMAND, w); /* Reset expansion ROM address decode enable */ - pci_read_config_word(ha->pdev, PCI_ROM_ADDRESS, &w); - w &= ~PCI_ROM_ADDRESS_ENABLE; - pci_write_config_word(ha->pdev, PCI_ROM_ADDRESS, w); + pci_read_config_dword(ha->pdev, PCI_ROM_ADDRESS, &d); + d &= ~PCI_ROM_ADDRESS_ENABLE; + pci_write_config_dword(ha->pdev, PCI_ROM_ADDRESS, d); /* Get PCI bus information. */ spin_lock_irqsave(&ha->hardware_lock, flags); @@ -237,6 +238,7 @@ int qla2300_pci_config(scsi_qla_host_t *ha) { uint16_t w, mwi; + uint32_t d; unsigned long flags = 0; uint32_t cnt; struct device_reg_2xxx __iomem *reg = &ha->iobase->isp; @@ -302,9 +304,9 @@ qla2300_pci_config(scsi_qla_host_t *ha) pci_write_config_byte(ha->pdev, PCI_LATENCY_TIMER, 0x80); /* Reset expansion ROM address decode enable */ - pci_read_config_word(ha->pdev, PCI_ROM_ADDRESS, &w); - w &= ~PCI_ROM_ADDRESS_ENABLE; - pci_write_config_word(ha->pdev, PCI_ROM_ADDRESS, w); + pci_read_config_dword(ha->pdev, PCI_ROM_ADDRESS, &d); + d &= ~PCI_ROM_ADDRESS_ENABLE; + pci_write_config_dword(ha->pdev, PCI_ROM_ADDRESS, d); /* Get PCI bus information. */ spin_lock_irqsave(&ha->hardware_lock, flags); @@ -324,6 +326,7 @@ int qla24xx_pci_config(scsi_qla_host_t *ha) { uint16_t w, mwi; + uint32_t d; unsigned long flags = 0; struct device_reg_24xx __iomem *reg = &ha->iobase->isp24; int pcix_cmd_reg, pcie_dctl_reg; @@ -366,9 +369,9 @@ qla24xx_pci_config(scsi_qla_host_t *ha) } /* Reset expansion ROM address decode enable */ - pci_read_config_word(ha->pdev, PCI_ROM_ADDRESS, &w); - w &= ~PCI_ROM_ADDRESS_ENABLE; - pci_write_config_word(ha->pdev, PCI_ROM_ADDRESS, w); + pci_read_config_dword(ha->pdev, PCI_ROM_ADDRESS, &d); + d &= ~PCI_ROM_ADDRESS_ENABLE; + pci_write_config_dword(ha->pdev, PCI_ROM_ADDRESS, d); /* Get PCI bus information. */ spin_lock_irqsave(&ha->hardware_lock, flags); From b95be99d52ce4f9db9ff0bd5f10e9e2066da6d2e Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Thu, 15 Sep 2005 13:52:51 -0400 Subject: [PATCH 136/240] [SCSI] fix oops in scsi_release_buffers() I found one other thing that needs to be fixed. The call to scsi_release_buffers in scsi_unprep_request causes an oops, because the sgtable has already been freed in scsi_io_completion. The following patch is needed. Signed-off-by: Alan Stern Signed-off-by: James Bottomley --- drivers/scsi/scsi_lib.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 0c6924eecf59..f065cbc1a6e5 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -118,7 +118,6 @@ static void scsi_unprep_request(struct request *req) req->flags &= ~REQ_DONTPREP; req->special = (req->flags & REQ_SPECIAL) ? cmd->sc_request : NULL; - scsi_release_buffers(cmd); scsi_put_command(cmd); } @@ -1512,7 +1511,6 @@ static void scsi_request_fn(struct request_queue *q) * cases (host limits or settings) should run the queue at some * later time. */ - scsi_unprep_request(req); spin_lock_irq(q->queue_lock); blk_requeue_request(q, req); sdev->device_busy--; From 61ffcafafb3d985e1ab8463be0187b421614775c Mon Sep 17 00:00:00 2001 From: Karsten Keil Date: Sat, 17 Sep 2005 23:52:42 +0200 Subject: [PATCH 137/240] [PATCH] Fix ST 5481 USB driver The old driver was not fully adapted to new USB ABI and does not work. Signed-off-by: Karsten Keil Signed-off-by: Linus Torvalds --- drivers/isdn/hisax/st5481.h | 4 +- drivers/isdn/hisax/st5481_b.c | 22 ++++++---- drivers/isdn/hisax/st5481_d.c | 26 +++++++----- drivers/isdn/hisax/st5481_init.c | 4 +- drivers/isdn/hisax/st5481_usb.c | 70 ++++++++++++++++++++------------ 5 files changed, 76 insertions(+), 50 deletions(-) diff --git a/drivers/isdn/hisax/st5481.h b/drivers/isdn/hisax/st5481.h index 0fda5c89429b..9ffaae7c657a 100644 --- a/drivers/isdn/hisax/st5481.h +++ b/drivers/isdn/hisax/st5481.h @@ -466,10 +466,10 @@ void st5481_stop(struct st5481_adapter *adapter); #define __debug_variable st5481_debug #include "hisax_debug.h" -#ifdef CONFIG_HISAX_DEBUG - extern int st5481_debug; +#ifdef CONFIG_HISAX_DEBUG + #define DBG_ISO_PACKET(level,urb) \ if (level & __debug_variable) dump_iso_packet(__FUNCTION__,urb) diff --git a/drivers/isdn/hisax/st5481_b.c b/drivers/isdn/hisax/st5481_b.c index 2fcd093921d8..0a2536d62402 100644 --- a/drivers/isdn/hisax/st5481_b.c +++ b/drivers/isdn/hisax/st5481_b.c @@ -172,14 +172,18 @@ static void usb_b_out_complete(struct urb *urb, struct pt_regs *regs) test_and_clear_bit(buf_nr, &b_out->busy); if (unlikely(urb->status < 0)) { - if (urb->status != -ENOENT && urb->status != -ESHUTDOWN) { - WARN("urb status %d",urb->status); - if (b_out->busy == 0) { - st5481_usb_pipe_reset(adapter, (bcs->channel+1)*2 | USB_DIR_OUT, NULL, NULL); - } - } else { - DBG(1,"urb killed"); - return; // Give up + switch (urb->status) { + case -ENOENT: + case -ESHUTDOWN: + case -ECONNRESET: + DBG(4,"urb killed status %d", urb->status); + return; // Give up + default: + WARN("urb status %d",urb->status); + if (b_out->busy == 0) { + st5481_usb_pipe_reset(adapter, (bcs->channel+1)*2 | USB_DIR_OUT, NULL, NULL); + } + break; } } @@ -205,7 +209,9 @@ static void st5481B_mode(struct st5481_bcs *bcs, int mode) bcs->mode = mode; // Cancel all USB transfers on this B channel + b_out->urb[0]->transfer_flags |= URB_ASYNC_UNLINK; usb_unlink_urb(b_out->urb[0]); + b_out->urb[1]->transfer_flags |= URB_ASYNC_UNLINK; usb_unlink_urb(b_out->urb[1]); b_out->busy = 0; diff --git a/drivers/isdn/hisax/st5481_d.c b/drivers/isdn/hisax/st5481_d.c index 071b1d31999f..941f7022ada1 100644 --- a/drivers/isdn/hisax/st5481_d.c +++ b/drivers/isdn/hisax/st5481_d.c @@ -382,16 +382,20 @@ static void usb_d_out_complete(struct urb *urb, struct pt_regs *regs) test_and_clear_bit(buf_nr, &d_out->busy); if (unlikely(urb->status < 0)) { - if (urb->status != -ENOENT && urb->status != -ESHUTDOWN) { - WARN("urb status %d",urb->status); - if (d_out->busy == 0) { - st5481_usb_pipe_reset(adapter, EP_D_OUT | USB_DIR_OUT, fifo_reseted, adapter); - } - return; - } else { - DBG(1,"urb killed"); - return; // Give up + switch (urb->status) { + case -ENOENT: + case -ESHUTDOWN: + case -ECONNRESET: + DBG(1,"urb killed status %d", urb->status); + break; + default: + WARN("urb status %d",urb->status); + if (d_out->busy == 0) { + st5481_usb_pipe_reset(adapter, EP_D_OUT | USB_DIR_OUT, fifo_reseted, adapter); + } + break; } + return; // Give up } FsmEvent(&adapter->d_out.fsm, EV_DOUT_COMPLETE, (void *) buf_nr); @@ -709,14 +713,14 @@ int st5481_setup_d(struct st5481_adapter *adapter) adapter->l1m.fsm = &l1fsm; adapter->l1m.state = ST_L1_F3; - adapter->l1m.debug = 1; + adapter->l1m.debug = st5481_debug & 0x100; adapter->l1m.userdata = adapter; adapter->l1m.printdebug = l1m_debug; FsmInitTimer(&adapter->l1m, &adapter->timer); adapter->d_out.fsm.fsm = &dout_fsm; adapter->d_out.fsm.state = ST_DOUT_NONE; - adapter->d_out.fsm.debug = 1; + adapter->d_out.fsm.debug = st5481_debug & 0x100; adapter->d_out.fsm.userdata = adapter; adapter->d_out.fsm.printdebug = dout_debug; diff --git a/drivers/isdn/hisax/st5481_init.c b/drivers/isdn/hisax/st5481_init.c index 7aa810d5d333..2cf5d1a6df6c 100644 --- a/drivers/isdn/hisax/st5481_init.c +++ b/drivers/isdn/hisax/st5481_init.c @@ -43,10 +43,10 @@ static int number_of_leds = 2; /* 2 LEDs on the adpater default */ module_param(number_of_leds, int, 0); #ifdef CONFIG_HISAX_DEBUG -static int debug = 0x1; +static int debug = 0; module_param(debug, int, 0); -int st5481_debug; #endif +int st5481_debug; static LIST_HEAD(adapter_list); diff --git a/drivers/isdn/hisax/st5481_usb.c b/drivers/isdn/hisax/st5481_usb.c index ab62223297a5..ffd5b2d45552 100644 --- a/drivers/isdn/hisax/st5481_usb.c +++ b/drivers/isdn/hisax/st5481_usb.c @@ -132,11 +132,15 @@ static void usb_ctrl_complete(struct urb *urb, struct pt_regs *regs) struct ctrl_msg *ctrl_msg; if (unlikely(urb->status < 0)) { - if (urb->status != -ENOENT && urb->status != -ESHUTDOWN) { - WARN("urb status %d",urb->status); - } else { - DBG(1,"urb killed"); - return; // Give up + switch (urb->status) { + case -ENOENT: + case -ESHUTDOWN: + case -ECONNRESET: + DBG(1,"urb killed status %d", urb->status); + return; // Give up + default: + WARN("urb status %d",urb->status); + break; } } @@ -184,22 +188,22 @@ static void usb_int_complete(struct urb *urb, struct pt_regs *regs) int status; switch (urb->status) { - case 0: - /* success */ - break; - case -ECONNRESET: - case -ENOENT: - case -ESHUTDOWN: - /* this urb is terminated, clean up */ - DBG(1, "urb shutting down with status: %d", urb->status); - return; - default: - WARN("nonzero urb status received: %d", urb->status); - goto exit; + case 0: + /* success */ + break; + case -ECONNRESET: + case -ENOENT: + case -ESHUTDOWN: + /* this urb is terminated, clean up */ + DBG(2, "urb shutting down with status: %d", urb->status); + return; + default: + WARN("nonzero urb status received: %d", urb->status); + goto exit; } - DBG_PACKET(1, data, INT_PKT_SIZE); + DBG_PACKET(2, data, INT_PKT_SIZE); if (urb->actual_length == 0) { goto exit; @@ -250,7 +254,7 @@ int st5481_setup_usb(struct st5481_adapter *adapter) struct urb *urb; u8 *buf; - DBG(1,""); + DBG(2,""); if ((status = usb_reset_configuration (dev)) < 0) { WARN("reset_configuration failed,status=%d",status); @@ -330,15 +334,17 @@ void st5481_release_usb(struct st5481_adapter *adapter) DBG(1,""); // Stop and free Control and Interrupt URBs - usb_unlink_urb(ctrl->urb); + usb_kill_urb(ctrl->urb); if (ctrl->urb->transfer_buffer) kfree(ctrl->urb->transfer_buffer); usb_free_urb(ctrl->urb); + ctrl->urb = NULL; - usb_unlink_urb(intr->urb); + usb_kill_urb(intr->urb); if (intr->urb->transfer_buffer) kfree(intr->urb->transfer_buffer); usb_free_urb(intr->urb); + ctrl->urb = NULL; } /* @@ -406,6 +412,7 @@ fill_isoc_urb(struct urb *urb, struct usb_device *dev, spin_lock_init(&urb->lock); urb->dev=dev; urb->pipe=pipe; + urb->interval = 1; urb->transfer_buffer=buf; urb->number_of_packets = num_packets; urb->transfer_buffer_length=num_packets*packet_size; @@ -452,7 +459,9 @@ st5481_setup_isocpipes(struct urb* urb[2], struct usb_device *dev, if (urb[j]) { if (urb[j]->transfer_buffer) kfree(urb[j]->transfer_buffer); + urb[j]->transfer_buffer = NULL; usb_free_urb(urb[j]); + urb[j] = NULL; } } return retval; @@ -463,10 +472,11 @@ void st5481_release_isocpipes(struct urb* urb[2]) int j; for (j = 0; j < 2; j++) { - usb_unlink_urb(urb[j]); + usb_kill_urb(urb[j]); if (urb[j]->transfer_buffer) kfree(urb[j]->transfer_buffer); usb_free_urb(urb[j]); + urb[j] = NULL; } } @@ -485,11 +495,15 @@ static void usb_in_complete(struct urb *urb, struct pt_regs *regs) int len, count, status; if (unlikely(urb->status < 0)) { - if (urb->status != -ENOENT && urb->status != -ESHUTDOWN) { - WARN("urb status %d",urb->status); - } else { - DBG(1,"urb killed"); - return; // Give up + switch (urb->status) { + case -ENOENT: + case -ESHUTDOWN: + case -ECONNRESET: + DBG(1,"urb killed status %d", urb->status); + return; // Give up + default: + WARN("urb status %d",urb->status); + break; } } @@ -631,7 +645,9 @@ void st5481_in_mode(struct st5481_in *in, int mode) in->mode = mode; + in->urb[0]->transfer_flags |= URB_ASYNC_UNLINK; usb_unlink_urb(in->urb[0]); + in->urb[1]->transfer_flags |= URB_ASYNC_UNLINK; usb_unlink_urb(in->urb[1]); if (in->mode != L1_MODE_NULL) { From bc5e8fdfc622b03acf5ac974a1b8b26da6511c99 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 17 Sep 2005 15:41:04 -0700 Subject: [PATCH 138/240] x86-64/smp: fix random SIGSEGV issues They seem to have been due to AMD errata 63/122; the fix is to disable TLB flush filtering in SMP configurations. Confirmed to fix the problem by Andrew Walrond [ Let's see if we'll have a better fix eventually, this is the Q&D "let's get this fixed and out there" version ] Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/setup.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index 351d8d64c2fb..238f73e1a834 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -831,11 +831,26 @@ static void __init amd_detect_cmp(struct cpuinfo_x86 *c) #endif } +#define HWCR 0xc0010015 + static int __init init_amd(struct cpuinfo_x86 *c) { int r; int level; +#ifdef CONFIG_SMP + unsigned long value; + + // Disable TLB flush filter by setting HWCR.FFDIS: + // bit 6 of msr C001_0015 + // + // Errata 63 for SH-B3 steppings + // Errata 122 for all(?) steppings + rdmsrl(HWCR, value); + value |= 1 << 6; + wrmsrl(HWCR, value); +#endif + /* Bit 31 in normal CPUID used for nonstandard 3DNow ID; 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */ clear_bit(0*32+31, &c->x86_capability); From ae31c3399d17b1f7bc1742724f70476b5417744f Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sun, 18 Sep 2005 00:17:51 -0700 Subject: [PATCH 139/240] [DCCP]: Move the ack vector code to net/dccp/ackvec.[ch] Isolating it, that will be used when we introduce a CCID2 (TCP-Like) implementation. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/dccp.h | 14 +- net/dccp/Makefile | 2 + net/dccp/ackvec.c | 419 ++++++++++++++++++++++++++++++++++++++++ net/dccp/ackvec.h | 133 +++++++++++++ net/dccp/dccp.h | 82 ++------ net/dccp/input.c | 69 ++----- net/dccp/ipv4.c | 59 ++---- net/dccp/minisocks.c | 13 +- net/dccp/options.c | 440 +------------------------------------------ net/dccp/output.c | 12 +- 10 files changed, 612 insertions(+), 631 deletions(-) create mode 100644 net/dccp/ackvec.c create mode 100644 net/dccp/ackvec.h diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 0e72708677e4..8c8e029095a5 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -353,14 +353,8 @@ static inline struct dccp_request_sock *dccp_rsk(const struct request_sock *req) extern struct inet_timewait_death_row dccp_death_row; -/* Read about the ECN nonce to see why it is 253 */ -#define DCCP_MAX_ACK_VECTOR_LEN 253 - struct dccp_options_received { - u32 dccpor_ndp:24, - dccpor_ack_vector_len:8; - u32 dccpor_ack_vector_idx:10; - /* 22 bits hole, try to pack */ + u32 dccpor_ndp; /* only 24 bits */ u32 dccpor_timestamp; u32 dccpor_timestamp_echo; u32 dccpor_elapsed_time; @@ -394,6 +388,8 @@ static inline int dccp_list_has_service(const struct dccp_service_list *sl, return 0; } +struct dccp_ackvec; + /** * struct dccp_sock - DCCP socket state * @@ -414,7 +410,7 @@ static inline int dccp_list_has_service(const struct dccp_service_list *sl, * @dccps_packet_size - Set thru setsockopt * @dccps_role - Role of this sock, one of %dccp_role * @dccps_ndp_count - number of Non Data Packets since last data packet - * @dccps_hc_rx_ackpkts - receiver half connection acked packets + * @dccps_hc_rx_ackvec - rx half connection ack vector */ struct dccp_sock { /* inet_connection_sock has to be the first member of dccp_sock */ @@ -439,7 +435,7 @@ struct dccp_sock { __u32 dccps_pmtu_cookie; __u32 dccps_mss_cache; struct dccp_options dccps_options; - struct dccp_ackpkts *dccps_hc_rx_ackpkts; + struct dccp_ackvec *dccps_hc_rx_ackvec; void *dccps_hc_rx_ccid_private; void *dccps_hc_tx_ccid_private; struct ccid *dccps_hc_rx_ccid; diff --git a/net/dccp/Makefile b/net/dccp/Makefile index fb97bb042455..344a8da153fc 100644 --- a/net/dccp/Makefile +++ b/net/dccp/Makefile @@ -3,6 +3,8 @@ obj-$(CONFIG_IP_DCCP) += dccp.o dccp-y := ccid.o input.o ipv4.o minisocks.o options.o output.o proto.o \ timer.o +dccp-$(CONFIG_IP_DCCP_ACKVEC) += ackvec.o + obj-$(CONFIG_INET_DCCP_DIAG) += dccp_diag.o dccp_diag-y := diag.o diff --git a/net/dccp/ackvec.c b/net/dccp/ackvec.c new file mode 100644 index 000000000000..6530283eafca --- /dev/null +++ b/net/dccp/ackvec.c @@ -0,0 +1,419 @@ +/* + * net/dccp/ackvec.c + * + * An implementation of the DCCP protocol + * Copyright (c) 2005 Arnaldo Carvalho de Melo + * + * 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; version 2 of the License; + */ + +#include "ackvec.h" +#include "dccp.h" + +#include +#include + +#include + +int dccp_insert_option_ackvec(struct sock *sk, struct sk_buff *skb) +{ + struct dccp_sock *dp = dccp_sk(sk); + struct dccp_ackvec *av = dp->dccps_hc_rx_ackvec; + int len = av->dccpav_vec_len + 2; + struct timeval now; + u32 elapsed_time; + unsigned char *to, *from; + + dccp_timestamp(sk, &now); + elapsed_time = timeval_delta(&now, &av->dccpav_time) / 10; + + if (elapsed_time != 0) + dccp_insert_option_elapsed_time(sk, skb, elapsed_time); + + if (DCCP_SKB_CB(skb)->dccpd_opt_len + len > DCCP_MAX_OPT_LEN) + return -1; + + /* + * XXX: now we have just one ack vector sent record, so + * we have to wait for it to be cleared. + * + * Of course this is not acceptable, but this is just for + * basic testing now. + */ + if (av->dccpav_ack_seqno != DCCP_MAX_SEQNO + 1) + return -1; + + DCCP_SKB_CB(skb)->dccpd_opt_len += len; + + to = skb_push(skb, len); + *to++ = DCCPO_ACK_VECTOR_0; + *to++ = len; + + len = av->dccpav_vec_len; + from = av->dccpav_buf + av->dccpav_buf_head; + + /* Check if buf_head wraps */ + if (av->dccpav_buf_head + len > av->dccpav_vec_len) { + const u32 tailsize = (av->dccpav_vec_len - av->dccpav_buf_head); + + memcpy(to, from, tailsize); + to += tailsize; + len -= tailsize; + from = av->dccpav_buf; + } + + memcpy(to, from, len); + /* + * From draft-ietf-dccp-spec-11.txt: + * + * For each acknowledgement it sends, the HC-Receiver will add an + * acknowledgement record. ack_seqno will equal the HC-Receiver + * sequence number it used for the ack packet; ack_ptr will equal + * buf_head; ack_ackno will equal buf_ackno; and ack_nonce will + * equal buf_nonce. + * + * This implemention uses just one ack record for now. + */ + av->dccpav_ack_seqno = DCCP_SKB_CB(skb)->dccpd_seq; + av->dccpav_ack_ptr = av->dccpav_buf_head; + av->dccpav_ack_ackno = av->dccpav_buf_ackno; + av->dccpav_ack_nonce = av->dccpav_buf_nonce; + av->dccpav_sent_len = av->dccpav_vec_len; + + dccp_pr_debug("%sACK Vector 0, len=%d, ack_seqno=%llu, " + "ack_ackno=%llu\n", + debug_prefix, av->dccpav_sent_len, + (unsigned long long)av->dccpav_ack_seqno, + (unsigned long long)av->dccpav_ack_ackno); + return -1; +} + +struct dccp_ackvec *dccp_ackvec_alloc(const unsigned int len, + const unsigned int __nocast priority) +{ + struct dccp_ackvec *av = kmalloc(sizeof(*av) + len, priority); + + if (av != NULL) { + av->dccpav_buf_len = len; + av->dccpav_buf_head = + av->dccpav_buf_tail = av->dccpav_buf_len - 1; + av->dccpav_buf_ackno = + av->dccpav_ack_ackno = av->dccpav_ack_seqno = ~0LLU; + av->dccpav_buf_nonce = av->dccpav_buf_nonce = 0; + av->dccpav_ack_ptr = 0; + av->dccpav_time.tv_sec = 0; + av->dccpav_time.tv_usec = 0; + av->dccpav_sent_len = av->dccpav_vec_len = 0; + } + + return av; +} + +void dccp_ackvec_free(struct dccp_ackvec *av) +{ + kfree(av); +} + +static inline u8 dccp_ackvec_state(const struct dccp_ackvec *av, + const unsigned int index) +{ + return av->dccpav_buf[index] & DCCP_ACKVEC_STATE_MASK; +} + +static inline u8 dccp_ackvec_len(const struct dccp_ackvec *av, + const unsigned int index) +{ + return av->dccpav_buf[index] & DCCP_ACKVEC_LEN_MASK; +} + +/* + * If several packets are missing, the HC-Receiver may prefer to enter multiple + * bytes with run length 0, rather than a single byte with a larger run length; + * this simplifies table updates if one of the missing packets arrives. + */ +static inline int dccp_ackvec_set_buf_head_state(struct dccp_ackvec *av, + const unsigned int packets, + const unsigned char state) +{ + unsigned int gap; + signed long new_head; + + if (av->dccpav_vec_len + packets > av->dccpav_buf_len) + return -ENOBUFS; + + gap = packets - 1; + new_head = av->dccpav_buf_head - packets; + + if (new_head < 0) { + if (gap > 0) { + memset(av->dccpav_buf, DCCP_ACKVEC_STATE_NOT_RECEIVED, + gap + new_head + 1); + gap = -new_head; + } + new_head += av->dccpav_buf_len; + } + + av->dccpav_buf_head = new_head; + + if (gap > 0) + memset(av->dccpav_buf + av->dccpav_buf_head + 1, + DCCP_ACKVEC_STATE_NOT_RECEIVED, gap); + + av->dccpav_buf[av->dccpav_buf_head] = state; + av->dccpav_vec_len += packets; + return 0; +} + +/* + * Implements the draft-ietf-dccp-spec-11.txt Appendix A + */ +int dccp_ackvec_add(struct dccp_ackvec *av, const struct sock *sk, + const u64 ackno, const u8 state) +{ + /* + * Check at the right places if the buffer is full, if it is, tell the + * caller to start dropping packets till the HC-Sender acks our ACK + * vectors, when we will free up space in dccpav_buf. + * + * We may well decide to do buffer compression, etc, but for now lets + * just drop. + * + * From Appendix A: + * + * Of course, the circular buffer may overflow, either when the + * HC-Sender is sending data at a very high rate, when the + * HC-Receiver's acknowledgements are not reaching the HC-Sender, + * or when the HC-Sender is forgetting to acknowledge those acks + * (so the HC-Receiver is unable to clean up old state). In this + * case, the HC-Receiver should either compress the buffer (by + * increasing run lengths when possible), transfer its state to + * a larger buffer, or, as a last resort, drop all received + * packets, without processing them whatsoever, until its buffer + * shrinks again. + */ + + /* See if this is the first ackno being inserted */ + if (av->dccpav_vec_len == 0) { + av->dccpav_buf[av->dccpav_buf_head] = state; + av->dccpav_vec_len = 1; + } else if (after48(ackno, av->dccpav_buf_ackno)) { + const u64 delta = dccp_delta_seqno(av->dccpav_buf_ackno, + ackno); + + /* + * Look if the state of this packet is the same as the + * previous ackno and if so if we can bump the head len. + */ + if (delta == 1 && + dccp_ackvec_state(av, av->dccpav_buf_head) == state && + (dccp_ackvec_len(av, av->dccpav_buf_head) < + DCCP_ACKVEC_LEN_MASK)) + av->dccpav_buf[av->dccpav_buf_head]++; + else if (dccp_ackvec_set_buf_head_state(av, delta, state)) + return -ENOBUFS; + } else { + /* + * A.1.2. Old Packets + * + * When a packet with Sequence Number S arrives, and + * S <= buf_ackno, the HC-Receiver will scan the table + * for the byte corresponding to S. (Indexing structures + * could reduce the complexity of this scan.) + */ + u64 delta = dccp_delta_seqno(ackno, av->dccpav_buf_ackno); + unsigned int index = av->dccpav_buf_head; + + while (1) { + const u8 len = dccp_ackvec_len(av, index); + const u8 state = dccp_ackvec_state(av, index); + /* + * valid packets not yet in dccpav_buf have a reserved + * entry, with a len equal to 0. + */ + if (state == DCCP_ACKVEC_STATE_NOT_RECEIVED && + len == 0 && delta == 0) { /* Found our + reserved seat! */ + dccp_pr_debug("Found %llu reserved seat!\n", + (unsigned long long)ackno); + av->dccpav_buf[index] = state; + goto out; + } + /* len == 0 means one packet */ + if (delta < len + 1) + goto out_duplicate; + + delta -= len + 1; + if (++index == av->dccpav_buf_len) + index = 0; + } + } + + av->dccpav_buf_ackno = ackno; + dccp_timestamp(sk, &av->dccpav_time); +out: + dccp_pr_debug(""); + return 0; + +out_duplicate: + /* Duplicate packet */ + dccp_pr_debug("Received a dup or already considered lost " + "packet: %llu\n", (unsigned long long)ackno); + return -EILSEQ; +} + +#ifdef CONFIG_IP_DCCP_DEBUG +void dccp_ackvector_print(const u64 ackno, const unsigned char *vector, int len) +{ + if (!dccp_debug) + return; + + printk("ACK vector len=%d, ackno=%llu |", len, + (unsigned long long)ackno); + + while (len--) { + const u8 state = (*vector & DCCP_ACKVEC_STATE_MASK) >> 6; + const u8 rl = *vector & DCCP_ACKVEC_LEN_MASK; + + printk("%d,%d|", state, rl); + ++vector; + } + + printk("\n"); +} + +void dccp_ackvec_print(const struct dccp_ackvec *av) +{ + dccp_ackvector_print(av->dccpav_buf_ackno, + av->dccpav_buf + av->dccpav_buf_head, + av->dccpav_vec_len); +} +#endif + +static void dccp_ackvec_trow_away_ack_record(struct dccp_ackvec *av) +{ + /* + * As we're keeping track of the ack vector size (dccpav_vec_len) and + * the sent ack vector size (dccpav_sent_len) we don't need + * dccpav_buf_tail at all, but keep this code here as in the future + * we'll implement a vector of ack records, as suggested in + * draft-ietf-dccp-spec-11.txt Appendix A. -acme + */ +#if 0 + av->dccpav_buf_tail = av->dccpav_ack_ptr + 1; + if (av->dccpav_buf_tail >= av->dccpav_vec_len) + av->dccpav_buf_tail -= av->dccpav_vec_len; +#endif + av->dccpav_vec_len -= av->dccpav_sent_len; +} + +void dccp_ackvec_check_rcv_ackno(struct dccp_ackvec *av, struct sock *sk, + const u64 ackno) +{ + /* Check if we actually sent an ACK vector */ + if (av->dccpav_ack_seqno == DCCP_MAX_SEQNO + 1) + return; + + if (ackno == av->dccpav_ack_seqno) { +#ifdef CONFIG_IP_DCCP_DEBUG + struct dccp_sock *dp = dccp_sk(sk); + const char *debug_prefix = dp->dccps_role == DCCP_ROLE_CLIENT ? + "CLIENT rx ack: " : "server rx ack: "; +#endif + dccp_pr_debug("%sACK packet 0, len=%d, ack_seqno=%llu, " + "ack_ackno=%llu, ACKED!\n", + debug_prefix, 1, + (unsigned long long)av->dccpav_ack_seqno, + (unsigned long long)av->dccpav_ack_ackno); + dccp_ackvec_trow_away_ack_record(av); + av->dccpav_ack_seqno = DCCP_MAX_SEQNO + 1; + } +} + +static void dccp_ackvec_check_rcv_ackvector(struct dccp_ackvec *av, + struct sock *sk, u64 ackno, + const unsigned char len, + const unsigned char *vector) +{ + unsigned char i; + + /* Check if we actually sent an ACK vector */ + if (av->dccpav_ack_seqno == DCCP_MAX_SEQNO + 1) + return; + /* + * We're in the receiver half connection, so if the received an ACK + * vector ackno (e.g. 50) before dccpav_ack_seqno (e.g. 52), we're + * not interested. + * + * Extra explanation with example: + * + * if we received an ACK vector with ackno 50, it can only be acking + * 50, 49, 48, etc, not 52 (the seqno for the ACK vector we sent). + */ + /* dccp_pr_debug("is %llu < %llu? ", ackno, av->dccpav_ack_seqno); */ + if (before48(ackno, av->dccpav_ack_seqno)) { + /* dccp_pr_debug_cat("yes\n"); */ + return; + } + /* dccp_pr_debug_cat("no\n"); */ + + i = len; + while (i--) { + const u8 rl = *vector & DCCP_ACKVEC_LEN_MASK; + u64 ackno_end_rl; + + dccp_set_seqno(&ackno_end_rl, ackno - rl); + + /* + * dccp_pr_debug("is %llu <= %llu <= %llu? ", ackno_end_rl, + * av->dccpav_ack_seqno, ackno); + */ + if (between48(av->dccpav_ack_seqno, ackno_end_rl, ackno)) { + const u8 state = (*vector & + DCCP_ACKVEC_STATE_MASK) >> 6; + /* dccp_pr_debug_cat("yes\n"); */ + + if (state != DCCP_ACKVEC_STATE_NOT_RECEIVED) { +#ifdef CONFIG_IP_DCCP_DEBUG + struct dccp_sock *dp = dccp_sk(sk); + const char *debug_prefix = + dp->dccps_role == DCCP_ROLE_CLIENT ? + "CLIENT rx ack: " : "server rx ack: "; +#endif + dccp_pr_debug("%sACK vector 0, len=%d, " + "ack_seqno=%llu, ack_ackno=%llu, " + "ACKED!\n", + debug_prefix, len, + (unsigned long long) + av->dccpav_ack_seqno, + (unsigned long long) + av->dccpav_ack_ackno); + dccp_ackvec_trow_away_ack_record(av); + } + /* + * If dccpav_ack_seqno was not received, no problem + * we'll send another ACK vector. + */ + av->dccpav_ack_seqno = DCCP_MAX_SEQNO + 1; + break; + } + /* dccp_pr_debug_cat("no\n"); */ + + dccp_set_seqno(&ackno, ackno_end_rl - 1); + ++vector; + } +} + +int dccp_ackvec_parse(struct sock *sk, const struct sk_buff *skb, + const u8 opt, const u8 *value, const u8 len) +{ + if (len > DCCP_MAX_ACKVEC_LEN) + return -1; + + /* dccp_ackvector_print(DCCP_SKB_CB(skb)->dccpd_ack_seq, value, len); */ + dccp_ackvec_check_rcv_ackvector(dccp_sk(sk)->dccps_hc_rx_ackvec, sk, + DCCP_SKB_CB(skb)->dccpd_ack_seq, + len, value); + return 0; +} diff --git a/net/dccp/ackvec.h b/net/dccp/ackvec.h new file mode 100644 index 000000000000..8ca51c9191f7 --- /dev/null +++ b/net/dccp/ackvec.h @@ -0,0 +1,133 @@ +#ifndef _ACKVEC_H +#define _ACKVEC_H +/* + * net/dccp/ackvec.h + * + * An implementation of the DCCP protocol + * Copyright (c) 2005 Arnaldo Carvalho de Melo + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +/* Read about the ECN nonce to see why it is 253 */ +#define DCCP_MAX_ACKVEC_LEN 253 + +#define DCCP_ACKVEC_STATE_RECEIVED 0 +#define DCCP_ACKVEC_STATE_ECN_MARKED (1 << 6) +#define DCCP_ACKVEC_STATE_NOT_RECEIVED (3 << 6) + +#define DCCP_ACKVEC_STATE_MASK 0xC0 /* 11000000 */ +#define DCCP_ACKVEC_LEN_MASK 0x3F /* 00111111 */ + +/** struct dccp_ackvec - ack vector + * + * This data structure is the one defined in the DCCP draft + * Appendix A. + * + * @dccpav_buf_head - circular buffer head + * @dccpav_buf_tail - circular buffer tail + * @dccpav_buf_ackno - ack # of the most recent packet acknowledgeable in the + * buffer (i.e. %dccpav_buf_head) + * @dccpav_buf_nonce - the one-bit sum of the ECN Nonces on all packets acked + * by the buffer with State 0 + * + * Additionally, the HC-Receiver must keep some information about the + * Ack Vectors it has recently sent. For each packet sent carrying an + * Ack Vector, it remembers four variables: + * + * @dccpav_ack_seqno - the Sequence Number used for the packet + * (HC-Receiver seqno) + * @dccpav_ack_ptr - the value of buf_head at the time of acknowledgement. + * @dccpav_ack_ackno - the Acknowledgement Number used for the packet + * (HC-Sender seqno) + * @dccpav_ack_nonce - the one-bit sum of the ECN Nonces for all State 0. + * + * @dccpav_buf_len - circular buffer length + * @dccpav_time - the time in usecs + * @dccpav_buf - circular buffer of acknowledgeable packets + */ +struct dccp_ackvec { + unsigned int dccpav_buf_head; + unsigned int dccpav_buf_tail; + u64 dccpav_buf_ackno; + u64 dccpav_ack_seqno; + u64 dccpav_ack_ackno; + unsigned int dccpav_ack_ptr; + unsigned int dccpav_sent_len; + unsigned int dccpav_vec_len; + unsigned int dccpav_buf_len; + struct timeval dccpav_time; + u8 dccpav_buf_nonce; + u8 dccpav_ack_nonce; + u8 dccpav_buf[0]; +}; + +struct sock; +struct sk_buff; + +#ifdef CONFIG_IP_DCCP_ACKVEC +extern struct dccp_ackvec *dccp_ackvec_alloc(unsigned int len, + const unsigned int __nocast priority); +extern void dccp_ackvec_free(struct dccp_ackvec *av); + +extern int dccp_ackvec_add(struct dccp_ackvec *av, const struct sock *sk, + const u64 ackno, const u8 state); + +extern void dccp_ackvec_check_rcv_ackno(struct dccp_ackvec *av, + struct sock *sk, const u64 ackno); +extern int dccp_ackvec_parse(struct sock *sk, const struct sk_buff *skb, + const u8 opt, const u8 *value, const u8 len); + +extern int dccp_insert_option_ackvec(struct sock *sk, struct sk_buff *skb); + +static inline int dccp_ackvec_pending(const struct dccp_ackvec *av) +{ + return av->dccpav_sent_len != av->dccpav_vec_len; +} +#else /* CONFIG_IP_DCCP_ACKVEC */ +static inline struct dccp_ackvec *dccp_ackvec_alloc(unsigned int len, + const unsigned int __nocast priority) +{ + return NULL; +} + +static inline void dccp_ackvec_free(struct dccp_ackvec *av) +{ +} + +static inline int dccp_ackvec_add(struct dccp_ackvec *av, const struct sock *sk, + const u64 ackno, const u8 state) +{ + return -1; +} + +static inline void dccp_ackvec_check_rcv_ackno(struct dccp_ackvec *av, + struct sock *sk, const u64 ackno) +{ +} + +static inline int dccp_ackvec_parse(struct sock *sk, const struct sk_buff *skb, + const u8 opt, const u8 *value, const u8 len) +{ + return -1; +} + +static inline int dccp_insert_option_ackvec(const struct sock *sk, + const struct sk_buff *skb) +{ + return -1; +} + +static inline int dccp_ackvec_pending(const struct dccp_ackvec *av) +{ + return 0; +} +#endif /* CONFIG_IP_DCCP_ACKVEC */ +#endif /* _ACKVEC_H */ diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index be7a660b6b24..5871c027f9dc 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -17,6 +17,7 @@ #include #include #include +#include "ackvec.h" #ifdef CONFIG_IP_DCCP_DEBUG extern int dccp_debug; @@ -358,6 +359,17 @@ static inline void dccp_update_gss(struct sock *sk, u64 seq) (dp->dccps_gss - dp->dccps_options.dccpo_sequence_window + 1)); } + +static inline int dccp_ack_pending(const struct sock *sk) +{ + const struct dccp_sock *dp = dccp_sk(sk); + return dp->dccps_timestamp_echo != 0 || +#ifdef CONFIG_IP_DCCP_ACKVEC + (dp->dccps_options.dccpo_send_ack_vector && + dccp_ackvec_pending(dp->dccps_hc_rx_ackvec)) || +#endif + inet_csk_ack_scheduled(sk); +} extern void dccp_insert_options(struct sock *sk, struct sk_buff *skb); extern void dccp_insert_option_elapsed_time(struct sock *sk, @@ -371,65 +383,6 @@ extern void dccp_insert_option(struct sock *sk, struct sk_buff *skb, extern struct socket *dccp_ctl_socket; -#define DCCP_ACKPKTS_STATE_RECEIVED 0 -#define DCCP_ACKPKTS_STATE_ECN_MARKED (1 << 6) -#define DCCP_ACKPKTS_STATE_NOT_RECEIVED (3 << 6) - -#define DCCP_ACKPKTS_STATE_MASK 0xC0 /* 11000000 */ -#define DCCP_ACKPKTS_LEN_MASK 0x3F /* 00111111 */ - -/** struct dccp_ackpkts - acknowledgeable packets - * - * This data structure is the one defined in the DCCP draft - * Appendix A. - * - * @dccpap_buf_head - circular buffer head - * @dccpap_buf_tail - circular buffer tail - * @dccpap_buf_ackno - ack # of the most recent packet acknowledgeable in the - * buffer (i.e. %dccpap_buf_head) - * @dccpap_buf_nonce - the one-bit sum of the ECN Nonces on all packets acked - * by the buffer with State 0 - * - * Additionally, the HC-Receiver must keep some information about the - * Ack Vectors it has recently sent. For each packet sent carrying an - * Ack Vector, it remembers four variables: - * - * @dccpap_ack_seqno - the Sequence Number used for the packet - * (HC-Receiver seqno) - * @dccpap_ack_ptr - the value of buf_head at the time of acknowledgement. - * @dccpap_ack_ackno - the Acknowledgement Number used for the packet - * (HC-Sender seqno) - * @dccpap_ack_nonce - the one-bit sum of the ECN Nonces for all State 0. - * - * @dccpap_buf_len - circular buffer length - * @dccpap_time - the time in usecs - * @dccpap_buf - circular buffer of acknowledgeable packets - */ -struct dccp_ackpkts { - unsigned int dccpap_buf_head; - unsigned int dccpap_buf_tail; - u64 dccpap_buf_ackno; - u64 dccpap_ack_seqno; - u64 dccpap_ack_ackno; - unsigned int dccpap_ack_ptr; - unsigned int dccpap_buf_vector_len; - unsigned int dccpap_ack_vector_len; - unsigned int dccpap_buf_len; - struct timeval dccpap_time; - u8 dccpap_buf_nonce; - u8 dccpap_ack_nonce; - u8 dccpap_buf[0]; -}; - -extern struct dccp_ackpkts * - dccp_ackpkts_alloc(unsigned int len, - const unsigned int __nocast priority); -extern void dccp_ackpkts_free(struct dccp_ackpkts *ap); -extern int dccp_ackpkts_add(struct dccp_ackpkts *ap, const struct sock *sk, - u64 ackno, u8 state); -extern void dccp_ackpkts_check_rcv_ackno(struct dccp_ackpkts *ap, - struct sock *sk, u64 ackno); - extern void dccp_timestamp(const struct sock *sk, struct timeval *tv); static inline suseconds_t timeval_usecs(const struct timeval *tv) @@ -470,15 +423,4 @@ static inline void timeval_sub_usecs(struct timeval *tv, } } -#ifdef CONFIG_IP_DCCP_DEBUG -extern void dccp_ackvector_print(const u64 ackno, - const unsigned char *vector, int len); -extern void dccp_ackpkts_print(const struct dccp_ackpkts *ap); -#else -static inline void dccp_ackvector_print(const u64 ackno, - const unsigned char *vector, - int len) { } -static inline void dccp_ackpkts_print(const struct dccp_ackpkts *ap) { } -#endif - #endif /* _DCCP_H */ diff --git a/net/dccp/input.c b/net/dccp/input.c index 062e9f8359d0..1b6b2cb12376 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -16,6 +16,7 @@ #include +#include "ackvec.h" #include "ccid.h" #include "dccp.h" @@ -60,8 +61,8 @@ static inline void dccp_event_ack_recv(struct sock *sk, struct sk_buff *skb) struct dccp_sock *dp = dccp_sk(sk); if (dp->dccps_options.dccpo_send_ack_vector) - dccp_ackpkts_check_rcv_ackno(dp->dccps_hc_rx_ackpkts, sk, - DCCP_SKB_CB(skb)->dccpd_ack_seq); + dccp_ackvec_check_rcv_ackno(dp->dccps_hc_rx_ackvec, sk, + DCCP_SKB_CB(skb)->dccpd_ack_seq); } static int dccp_check_seqno(struct sock *sk, struct sk_buff *skb) @@ -164,37 +165,11 @@ int dccp_rcv_established(struct sock *sk, struct sk_buff *skb, if (DCCP_SKB_CB(skb)->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ) dccp_event_ack_recv(sk, skb); - /* - * FIXME: check ECN to see if we should use - * DCCP_ACKPKTS_STATE_ECN_MARKED - */ - if (dp->dccps_options.dccpo_send_ack_vector) { - struct dccp_ackpkts *ap = dp->dccps_hc_rx_ackpkts; - - if (dccp_ackpkts_add(dp->dccps_hc_rx_ackpkts, sk, - DCCP_SKB_CB(skb)->dccpd_seq, - DCCP_ACKPKTS_STATE_RECEIVED)) { - LIMIT_NETDEBUG(KERN_WARNING "DCCP: acknowledgeable " - "packets buffer full!\n"); - ap->dccpap_ack_seqno = DCCP_MAX_SEQNO + 1; - inet_csk_schedule_ack(sk); - inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, - TCP_DELACK_MIN, - DCCP_RTO_MAX); - goto discard; - } - - /* - * FIXME: this activation is probably wrong, have to study more - * TCP delack machinery and how it fits into DCCP draft, but - * for now it kinda "works" 8) - */ - if (!inet_csk_ack_scheduled(sk)) { - inet_csk_schedule_ack(sk); - inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, 5 * HZ, - DCCP_RTO_MAX); - } - } + if (dp->dccps_options.dccpo_send_ack_vector && + dccp_ackvec_add(dp->dccps_hc_rx_ackvec, sk, + DCCP_SKB_CB(skb)->dccpd_seq, + DCCP_ACKVEC_STATE_RECEIVED)) + goto discard; ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb); ccid_hc_tx_packet_recv(dp->dccps_hc_tx_ccid, sk, skb); @@ -495,29 +470,11 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb); ccid_hc_tx_packet_recv(dp->dccps_hc_tx_ccid, sk, skb); - /* - * FIXME: check ECN to see if we should use - * DCCP_ACKPKTS_STATE_ECN_MARKED - */ - if (dp->dccps_options.dccpo_send_ack_vector) { - if (dccp_ackpkts_add(dp->dccps_hc_rx_ackpkts, sk, - dcb->dccpd_seq, - DCCP_ACKPKTS_STATE_RECEIVED)) - goto discard; - /* - * FIXME: this activation is probably wrong, have to - * study more TCP delack machinery and how it fits into - * DCCP draft, but for now it kinda "works" 8) - */ - if ((dp->dccps_hc_rx_ackpkts->dccpap_ack_seqno == - DCCP_MAX_SEQNO + 1) && - !inet_csk_ack_scheduled(sk)) { - inet_csk_schedule_ack(sk); - inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, - TCP_DELACK_MIN, - DCCP_RTO_MAX); - } - } + if (dp->dccps_options.dccpo_send_ack_vector && + dccp_ackvec_add(dp->dccps_hc_rx_ackvec, sk, + DCCP_SKB_CB(skb)->dccpd_seq, + DCCP_ACKVEC_STATE_RECEIVED)) + goto discard; } /* diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 94a440b2685b..82434e4a42df 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -23,6 +23,7 @@ #include #include +#include "ackvec.h" #include "ccid.h" #include "dccp.h" @@ -1112,45 +1113,7 @@ int dccp_v4_rcv(struct sk_buff *skb) goto discard_it; dh = dccp_hdr(skb); -#if 0 - /* - * Use something like this to simulate some DATA/DATAACK loss to test - * dccp_ackpkts_add, you'll get something like this on a session that - * sends 10 DATA/DATAACK packets: - * - * ackpkts_print: 281473596467422 |0,0|3,0|0,0|3,0|0,0|3,0|0,0|3,0|0,1| - * - * 0, 0 means: DCCP_ACKPKTS_STATE_RECEIVED, RLE == just this packet - * 0, 1 means: DCCP_ACKPKTS_STATE_RECEIVED, RLE == two adjacent packets - * with the same state - * 3, 0 means: DCCP_ACKPKTS_STATE_NOT_RECEIVED, RLE == just this packet - * - * So... - * - * 281473596467422 was received - * 281473596467421 was not received - * 281473596467420 was received - * 281473596467419 was not received - * 281473596467418 was received - * 281473596467417 was not received - * 281473596467416 was received - * 281473596467415 was not received - * 281473596467414 was received - * 281473596467413 was received (this one was the 3way handshake - * RESPONSE) - * - */ - if (dh->dccph_type == DCCP_PKT_DATA || - dh->dccph_type == DCCP_PKT_DATAACK) { - static int discard = 0; - if (discard) { - discard = 0; - goto discard_it; - } - discard = 1; - } -#endif DCCP_SKB_CB(skb)->dccpd_seq = dccp_hdr_seq(skb); DCCP_SKB_CB(skb)->dccpd_type = dh->dccph_type; @@ -1264,11 +1227,9 @@ static int dccp_v4_init_sock(struct sock *sk) do_gettimeofday(&dp->dccps_epoch); if (dp->dccps_options.dccpo_send_ack_vector) { - dp->dccps_hc_rx_ackpkts = - dccp_ackpkts_alloc(DCCP_MAX_ACK_VECTOR_LEN, - GFP_KERNEL); - - if (dp->dccps_hc_rx_ackpkts == NULL) + dp->dccps_hc_rx_ackvec = dccp_ackvec_alloc(DCCP_MAX_ACKVEC_LEN, + GFP_KERNEL); + if (dp->dccps_hc_rx_ackvec == NULL) return -ENOMEM; } @@ -1288,8 +1249,10 @@ static int dccp_v4_init_sock(struct sock *sk) dp->dccps_hc_tx_ccid == NULL) { ccid_exit(dp->dccps_hc_rx_ccid, sk); ccid_exit(dp->dccps_hc_tx_ccid, sk); - dccp_ackpkts_free(dp->dccps_hc_rx_ackpkts); - dp->dccps_hc_rx_ackpkts = NULL; + if (dp->dccps_options.dccpo_send_ack_vector) { + dccp_ackvec_free(dp->dccps_hc_rx_ackvec); + dp->dccps_hc_rx_ackvec = NULL; + } dp->dccps_hc_rx_ccid = dp->dccps_hc_tx_ccid = NULL; return -ENOMEM; } @@ -1331,8 +1294,10 @@ static int dccp_v4_destroy_sock(struct sock *sk) ccid_hc_rx_exit(dp->dccps_hc_rx_ccid, sk); ccid_hc_tx_exit(dp->dccps_hc_tx_ccid, sk); - dccp_ackpkts_free(dp->dccps_hc_rx_ackpkts); - dp->dccps_hc_rx_ackpkts = NULL; + if (dp->dccps_options.dccpo_send_ack_vector) { + dccp_ackvec_free(dp->dccps_hc_rx_ackvec); + dp->dccps_hc_rx_ackvec = NULL; + } ccid_exit(dp->dccps_hc_rx_ccid, sk); ccid_exit(dp->dccps_hc_tx_ccid, sk); dp->dccps_hc_rx_ccid = dp->dccps_hc_tx_ccid = NULL; diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 933e10db1789..1393461898bb 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -19,6 +19,7 @@ #include #include +#include "ackvec.h" #include "ccid.h" #include "dccp.h" @@ -94,23 +95,23 @@ struct sock *dccp_create_openreq_child(struct sock *sk, struct dccp_sock *newdp = dccp_sk(newsk); newdp->dccps_role = DCCP_ROLE_SERVER; - newdp->dccps_hc_rx_ackpkts = NULL; + newdp->dccps_hc_rx_ackvec = NULL; newdp->dccps_service_list = NULL; newdp->dccps_service = dreq->dreq_service; newicsk->icsk_rto = DCCP_TIMEOUT_INIT; do_gettimeofday(&newdp->dccps_epoch); if (newdp->dccps_options.dccpo_send_ack_vector) { - newdp->dccps_hc_rx_ackpkts = - dccp_ackpkts_alloc(DCCP_MAX_ACK_VECTOR_LEN, - GFP_ATOMIC); + newdp->dccps_hc_rx_ackvec = + dccp_ackvec_alloc(DCCP_MAX_ACKVEC_LEN, + GFP_ATOMIC); /* * XXX: We're using the same CCIDs set on the parent, * i.e. sk_clone copied the master sock and left the * CCID pointers for this child, that is why we do the * __ccid_get calls. */ - if (unlikely(newdp->dccps_hc_rx_ackpkts == NULL)) + if (unlikely(newdp->dccps_hc_rx_ackvec == NULL)) goto out_free; } @@ -118,7 +119,7 @@ struct sock *dccp_create_openreq_child(struct sock *sk, newsk) != 0 || ccid_hc_tx_init(newdp->dccps_hc_tx_ccid, newsk) != 0)) { - dccp_ackpkts_free(newdp->dccps_hc_rx_ackpkts); + dccp_ackvec_free(newdp->dccps_hc_rx_ackvec); ccid_hc_rx_exit(newdp->dccps_hc_rx_ccid, newsk); ccid_hc_tx_exit(newdp->dccps_hc_tx_ccid, newsk); out_free: diff --git a/net/dccp/options.c b/net/dccp/options.c index d4c4242d8dd7..c480c506a4a4 100644 --- a/net/dccp/options.c +++ b/net/dccp/options.c @@ -18,15 +18,10 @@ #include #include +#include "ackvec.h" #include "ccid.h" #include "dccp.h" -static void dccp_ackpkts_check_rcv_ackvector(struct dccp_ackpkts *ap, - struct sock *sk, - const u64 ackno, - const unsigned char len, - const unsigned char *vector); - /* stores the default values for new connection. may be changed with sysctl */ static const struct dccp_options dccpo_default_values = { .dccpo_sequence_window = DCCPF_INITIAL_SEQUENCE_WINDOW, @@ -113,25 +108,13 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb) opt_recv->dccpor_ndp); break; case DCCPO_ACK_VECTOR_0: - if (len > DCCP_MAX_ACK_VECTOR_LEN) - goto out_invalid_option; - + case DCCPO_ACK_VECTOR_1: if (pkt_type == DCCP_PKT_DATA) continue; - opt_recv->dccpor_ack_vector_len = len; - opt_recv->dccpor_ack_vector_idx = value - options; - - dccp_pr_debug("%sACK vector 0, len=%d, ack_ackno=%llu\n", - debug_prefix, len, - (unsigned long long) - DCCP_SKB_CB(skb)->dccpd_ack_seq); - dccp_ackvector_print(DCCP_SKB_CB(skb)->dccpd_ack_seq, - value, len); - dccp_ackpkts_check_rcv_ackvector(dp->dccps_hc_rx_ackpkts, - sk, - DCCP_SKB_CB(skb)->dccpd_ack_seq, - len, value); + if (dp->dccps_options.dccpo_send_ack_vector && + dccp_ackvec_parse(sk, skb, opt, value, len)) + goto out_invalid_option; break; case DCCPO_TIMESTAMP: if (len != 4) @@ -352,86 +335,6 @@ void dccp_insert_option_elapsed_time(struct sock *sk, EXPORT_SYMBOL_GPL(dccp_insert_option_elapsed_time); -static void dccp_insert_option_ack_vector(struct sock *sk, struct sk_buff *skb) -{ - struct dccp_sock *dp = dccp_sk(sk); -#ifdef CONFIG_IP_DCCP_DEBUG - const char *debug_prefix = dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT TX opt: " : "server TX opt: "; -#endif - struct dccp_ackpkts *ap = dp->dccps_hc_rx_ackpkts; - int len = ap->dccpap_buf_vector_len + 2; - struct timeval now; - u32 elapsed_time; - unsigned char *to, *from; - - dccp_timestamp(sk, &now); - elapsed_time = timeval_delta(&now, &ap->dccpap_time) / 10; - - if (elapsed_time != 0) - dccp_insert_option_elapsed_time(sk, skb, elapsed_time); - - if (DCCP_SKB_CB(skb)->dccpd_opt_len + len > DCCP_MAX_OPT_LEN) { - LIMIT_NETDEBUG(KERN_INFO "DCCP: packet too small to " - "insert ACK Vector!\n"); - return; - } - - /* - * XXX: now we have just one ack vector sent record, so - * we have to wait for it to be cleared. - * - * Of course this is not acceptable, but this is just for - * basic testing now. - */ - if (ap->dccpap_ack_seqno != DCCP_MAX_SEQNO + 1) - return; - - DCCP_SKB_CB(skb)->dccpd_opt_len += len; - - to = skb_push(skb, len); - *to++ = DCCPO_ACK_VECTOR_0; - *to++ = len; - - len = ap->dccpap_buf_vector_len; - from = ap->dccpap_buf + ap->dccpap_buf_head; - - /* Check if buf_head wraps */ - if (ap->dccpap_buf_head + len > ap->dccpap_buf_len) { - const unsigned int tailsize = (ap->dccpap_buf_len - - ap->dccpap_buf_head); - - memcpy(to, from, tailsize); - to += tailsize; - len -= tailsize; - from = ap->dccpap_buf; - } - - memcpy(to, from, len); - /* - * From draft-ietf-dccp-spec-11.txt: - * - * For each acknowledgement it sends, the HC-Receiver will add an - * acknowledgement record. ack_seqno will equal the HC-Receiver - * sequence number it used for the ack packet; ack_ptr will equal - * buf_head; ack_ackno will equal buf_ackno; and ack_nonce will - * equal buf_nonce. - * - * This implemention uses just one ack record for now. - */ - ap->dccpap_ack_seqno = DCCP_SKB_CB(skb)->dccpd_seq; - ap->dccpap_ack_ptr = ap->dccpap_buf_head; - ap->dccpap_ack_ackno = ap->dccpap_buf_ackno; - ap->dccpap_ack_nonce = ap->dccpap_buf_nonce; - ap->dccpap_ack_vector_len = ap->dccpap_buf_vector_len; - - dccp_pr_debug("%sACK Vector 0, len=%d, ack_seqno=%llu, " - "ack_ackno=%llu\n", - debug_prefix, ap->dccpap_ack_vector_len, - (unsigned long long) ap->dccpap_ack_seqno, - (unsigned long long) ap->dccpap_ack_ackno); -} - void dccp_timestamp(const struct sock *sk, struct timeval *tv) { const struct dccp_sock *dp = dccp_sk(sk); @@ -528,9 +431,8 @@ void dccp_insert_options(struct sock *sk, struct sk_buff *skb) if (!dccp_packet_without_ack(skb)) { if (dp->dccps_options.dccpo_send_ack_vector && - (dp->dccps_hc_rx_ackpkts->dccpap_buf_ackno != - DCCP_MAX_SEQNO + 1)) - dccp_insert_option_ack_vector(sk, skb); + dccp_ackvec_pending(dp->dccps_hc_rx_ackvec)) + dccp_insert_option_ackvec(sk, skb); if (dp->dccps_timestamp_echo != 0) dccp_insert_option_timestamp_echo(sk, skb); } @@ -557,331 +459,3 @@ void dccp_insert_options(struct sock *sk, struct sk_buff *skb) } } } - -struct dccp_ackpkts *dccp_ackpkts_alloc(const unsigned int len, - const unsigned int __nocast priority) -{ - struct dccp_ackpkts *ap = kmalloc(sizeof(*ap) + len, priority); - - if (ap != NULL) { -#ifdef CONFIG_IP_DCCP_DEBUG - memset(ap->dccpap_buf, 0xFF, len); -#endif - ap->dccpap_buf_len = len; - ap->dccpap_buf_head = - ap->dccpap_buf_tail = - ap->dccpap_buf_len - 1; - ap->dccpap_buf_ackno = - ap->dccpap_ack_ackno = - ap->dccpap_ack_seqno = DCCP_MAX_SEQNO + 1; - ap->dccpap_buf_nonce = ap->dccpap_buf_nonce = 0; - ap->dccpap_ack_ptr = 0; - ap->dccpap_time.tv_sec = 0; - ap->dccpap_time.tv_usec = 0; - ap->dccpap_buf_vector_len = ap->dccpap_ack_vector_len = 0; - } - - return ap; -} - -void dccp_ackpkts_free(struct dccp_ackpkts *ap) -{ - if (ap != NULL) { -#ifdef CONFIG_IP_DCCP_DEBUG - memset(ap, 0xFF, sizeof(*ap) + ap->dccpap_buf_len); -#endif - kfree(ap); - } -} - -static inline u8 dccp_ackpkts_state(const struct dccp_ackpkts *ap, - const unsigned int index) -{ - return ap->dccpap_buf[index] & DCCP_ACKPKTS_STATE_MASK; -} - -static inline u8 dccp_ackpkts_len(const struct dccp_ackpkts *ap, - const unsigned int index) -{ - return ap->dccpap_buf[index] & DCCP_ACKPKTS_LEN_MASK; -} - -/* - * If several packets are missing, the HC-Receiver may prefer to enter multiple - * bytes with run length 0, rather than a single byte with a larger run length; - * this simplifies table updates if one of the missing packets arrives. - */ -static inline int dccp_ackpkts_set_buf_head_state(struct dccp_ackpkts *ap, - const unsigned int packets, - const unsigned char state) -{ - unsigned int gap; - signed long new_head; - - if (ap->dccpap_buf_vector_len + packets > ap->dccpap_buf_len) - return -ENOBUFS; - - gap = packets - 1; - new_head = ap->dccpap_buf_head - packets; - - if (new_head < 0) { - if (gap > 0) { - memset(ap->dccpap_buf, DCCP_ACKPKTS_STATE_NOT_RECEIVED, - gap + new_head + 1); - gap = -new_head; - } - new_head += ap->dccpap_buf_len; - } - - ap->dccpap_buf_head = new_head; - - if (gap > 0) - memset(ap->dccpap_buf + ap->dccpap_buf_head + 1, - DCCP_ACKPKTS_STATE_NOT_RECEIVED, gap); - - ap->dccpap_buf[ap->dccpap_buf_head] = state; - ap->dccpap_buf_vector_len += packets; - return 0; -} - -/* - * Implements the draft-ietf-dccp-spec-11.txt Appendix A - */ -int dccp_ackpkts_add(struct dccp_ackpkts *ap, const struct sock *sk, - u64 ackno, u8 state) -{ - /* - * Check at the right places if the buffer is full, if it is, tell the - * caller to start dropping packets till the HC-Sender acks our ACK - * vectors, when we will free up space in dccpap_buf. - * - * We may well decide to do buffer compression, etc, but for now lets - * just drop. - * - * From Appendix A: - * - * Of course, the circular buffer may overflow, either when the - * HC-Sender is sending data at a very high rate, when the - * HC-Receiver's acknowledgements are not reaching the HC-Sender, - * or when the HC-Sender is forgetting to acknowledge those acks - * (so the HC-Receiver is unable to clean up old state). In this - * case, the HC-Receiver should either compress the buffer (by - * increasing run lengths when possible), transfer its state to - * a larger buffer, or, as a last resort, drop all received - * packets, without processing them whatsoever, until its buffer - * shrinks again. - */ - - /* See if this is the first ackno being inserted */ - if (ap->dccpap_buf_vector_len == 0) { - ap->dccpap_buf[ap->dccpap_buf_head] = state; - ap->dccpap_buf_vector_len = 1; - } else if (after48(ackno, ap->dccpap_buf_ackno)) { - const u64 delta = dccp_delta_seqno(ap->dccpap_buf_ackno, - ackno); - - /* - * Look if the state of this packet is the same as the - * previous ackno and if so if we can bump the head len. - */ - if (delta == 1 && - dccp_ackpkts_state(ap, ap->dccpap_buf_head) == state && - (dccp_ackpkts_len(ap, ap->dccpap_buf_head) < - DCCP_ACKPKTS_LEN_MASK)) - ap->dccpap_buf[ap->dccpap_buf_head]++; - else if (dccp_ackpkts_set_buf_head_state(ap, delta, state)) - return -ENOBUFS; - } else { - /* - * A.1.2. Old Packets - * - * When a packet with Sequence Number S arrives, and - * S <= buf_ackno, the HC-Receiver will scan the table - * for the byte corresponding to S. (Indexing structures - * could reduce the complexity of this scan.) - */ - u64 delta = dccp_delta_seqno(ackno, ap->dccpap_buf_ackno); - unsigned int index = ap->dccpap_buf_head; - - while (1) { - const u8 len = dccp_ackpkts_len(ap, index); - const u8 state = dccp_ackpkts_state(ap, index); - /* - * valid packets not yet in dccpap_buf have a reserved - * entry, with a len equal to 0. - */ - if (state == DCCP_ACKPKTS_STATE_NOT_RECEIVED && - len == 0 && delta == 0) { /* Found our - reserved seat! */ - dccp_pr_debug("Found %llu reserved seat!\n", - (unsigned long long) ackno); - ap->dccpap_buf[index] = state; - goto out; - } - /* len == 0 means one packet */ - if (delta < len + 1) - goto out_duplicate; - - delta -= len + 1; - if (++index == ap->dccpap_buf_len) - index = 0; - } - } - - ap->dccpap_buf_ackno = ackno; - dccp_timestamp(sk, &ap->dccpap_time); -out: - dccp_pr_debug(""); - dccp_ackpkts_print(ap); - return 0; - -out_duplicate: - /* Duplicate packet */ - dccp_pr_debug("Received a dup or already considered lost " - "packet: %llu\n", (unsigned long long) ackno); - return -EILSEQ; -} - -#ifdef CONFIG_IP_DCCP_DEBUG -void dccp_ackvector_print(const u64 ackno, const unsigned char *vector, - int len) -{ - if (!dccp_debug) - return; - - printk("ACK vector len=%d, ackno=%llu |", len, - (unsigned long long) ackno); - - while (len--) { - const u8 state = (*vector & DCCP_ACKPKTS_STATE_MASK) >> 6; - const u8 rl = (*vector & DCCP_ACKPKTS_LEN_MASK); - - printk("%d,%d|", state, rl); - ++vector; - } - - printk("\n"); -} - -void dccp_ackpkts_print(const struct dccp_ackpkts *ap) -{ - dccp_ackvector_print(ap->dccpap_buf_ackno, - ap->dccpap_buf + ap->dccpap_buf_head, - ap->dccpap_buf_vector_len); -} -#endif - -static void dccp_ackpkts_trow_away_ack_record(struct dccp_ackpkts *ap) -{ - /* - * As we're keeping track of the ack vector size - * (dccpap_buf_vector_len) and the sent ack vector size - * (dccpap_ack_vector_len) we don't need dccpap_buf_tail at all, but - * keep this code here as in the future we'll implement a vector of - * ack records, as suggested in draft-ietf-dccp-spec-11.txt - * Appendix A. -acme - */ -#if 0 - ap->dccpap_buf_tail = ap->dccpap_ack_ptr + 1; - if (ap->dccpap_buf_tail >= ap->dccpap_buf_len) - ap->dccpap_buf_tail -= ap->dccpap_buf_len; -#endif - ap->dccpap_buf_vector_len -= ap->dccpap_ack_vector_len; -} - -void dccp_ackpkts_check_rcv_ackno(struct dccp_ackpkts *ap, struct sock *sk, - u64 ackno) -{ - /* Check if we actually sent an ACK vector */ - if (ap->dccpap_ack_seqno == DCCP_MAX_SEQNO + 1) - return; - - if (ackno == ap->dccpap_ack_seqno) { -#ifdef CONFIG_IP_DCCP_DEBUG - struct dccp_sock *dp = dccp_sk(sk); - const char *debug_prefix = dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT rx ack: " : "server rx ack: "; -#endif - dccp_pr_debug("%sACK packet 0, len=%d, ack_seqno=%llu, " - "ack_ackno=%llu, ACKED!\n", - debug_prefix, 1, - (unsigned long long) ap->dccpap_ack_seqno, - (unsigned long long) ap->dccpap_ack_ackno); - dccp_ackpkts_trow_away_ack_record(ap); - ap->dccpap_ack_seqno = DCCP_MAX_SEQNO + 1; - } -} - -static void dccp_ackpkts_check_rcv_ackvector(struct dccp_ackpkts *ap, - struct sock *sk, u64 ackno, - const unsigned char len, - const unsigned char *vector) -{ - unsigned char i; - - /* Check if we actually sent an ACK vector */ - if (ap->dccpap_ack_seqno == DCCP_MAX_SEQNO + 1) - return; - /* - * We're in the receiver half connection, so if the received an ACK - * vector ackno (e.g. 50) before dccpap_ack_seqno (e.g. 52), we're - * not interested. - * - * Extra explanation with example: - * - * if we received an ACK vector with ackno 50, it can only be acking - * 50, 49, 48, etc, not 52 (the seqno for the ACK vector we sent). - */ - /* dccp_pr_debug("is %llu < %llu? ", ackno, ap->dccpap_ack_seqno); */ - if (before48(ackno, ap->dccpap_ack_seqno)) { - /* dccp_pr_debug_cat("yes\n"); */ - return; - } - /* dccp_pr_debug_cat("no\n"); */ - - i = len; - while (i--) { - const u8 rl = (*vector & DCCP_ACKPKTS_LEN_MASK); - u64 ackno_end_rl; - - dccp_set_seqno(&ackno_end_rl, ackno - rl); - - /* - * dccp_pr_debug("is %llu <= %llu <= %llu? ", ackno_end_rl, - * ap->dccpap_ack_seqno, ackno); - */ - if (between48(ap->dccpap_ack_seqno, ackno_end_rl, ackno)) { - const u8 state = (*vector & - DCCP_ACKPKTS_STATE_MASK) >> 6; - /* dccp_pr_debug_cat("yes\n"); */ - - if (state != DCCP_ACKPKTS_STATE_NOT_RECEIVED) { -#ifdef CONFIG_IP_DCCP_DEBUG - struct dccp_sock *dp = dccp_sk(sk); - const char *debug_prefix = - dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT rx ack: " : "server rx ack: "; -#endif - dccp_pr_debug("%sACK vector 0, len=%d, " - "ack_seqno=%llu, ack_ackno=%llu, " - "ACKED!\n", - debug_prefix, len, - (unsigned long long) - ap->dccpap_ack_seqno, - (unsigned long long) - ap->dccpap_ack_ackno); - dccp_ackpkts_trow_away_ack_record(ap); - } - /* - * If dccpap_ack_seqno was not received, no problem - * we'll send another ACK vector. - */ - ap->dccpap_ack_seqno = DCCP_MAX_SEQNO + 1; - break; - } - /* dccp_pr_debug_cat("no\n"); */ - - dccp_set_seqno(&ackno, ackno_end_rl - 1); - ++vector; - } -} diff --git a/net/dccp/output.c b/net/dccp/output.c index 156b1d29a156..4786bdcddcc9 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -16,6 +16,7 @@ #include +#include "ackvec.h" #include "ccid.h" #include "dccp.h" @@ -225,7 +226,6 @@ int dccp_write_xmit(struct sock *sk, struct sk_buff *skb, long *timeo) err = dccp_wait_for_ccid(sk, skb, timeo); if (err == 0) { - const struct dccp_ackpkts *ap = dp->dccps_hc_rx_ackpkts; struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); const int len = skb->len; @@ -236,15 +236,7 @@ int dccp_write_xmit(struct sock *sk, struct sk_buff *skb, long *timeo) inet_csk(sk)->icsk_rto, DCCP_RTO_MAX); dcb->dccpd_type = DCCP_PKT_DATAACK; - /* - * FIXME: we really should have a - * dccps_ack_pending or use icsk. - */ - } else if (inet_csk_ack_scheduled(sk) || - dp->dccps_timestamp_echo != 0 || - (dp->dccps_options.dccpo_send_ack_vector && - ap->dccpap_buf_ackno != DCCP_MAX_SEQNO + 1 && - ap->dccpap_ack_seqno == DCCP_MAX_SEQNO + 1)) + } else if (dccp_ack_pending(sk)) dcb->dccpd_type = DCCP_PKT_DATAACK; else dcb->dccpd_type = DCCP_PKT_DATA; From 65299d6c3cfb49cc3eee4fc483e7edd23ea7b2ed Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sun, 18 Sep 2005 00:18:32 -0700 Subject: [PATCH 140/240] [CCID3]: Introduce include/linux/tfrc.h Moving the TFRC sender and receiver variables to separate structs, so that we can copy these structs to userspace thru getsockopt, dccp_diag, etc. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/tfrc.h | 35 +++++++++++++++++++++++++++++++++++ net/dccp/ccids/ccid3.h | 23 +++++++++++++---------- 2 files changed, 48 insertions(+), 10 deletions(-) create mode 100644 include/linux/tfrc.h diff --git a/include/linux/tfrc.h b/include/linux/tfrc.h new file mode 100644 index 000000000000..7dab7831c3cb --- /dev/null +++ b/include/linux/tfrc.h @@ -0,0 +1,35 @@ +#ifndef _LINUX_TFRC_H_ +#define _LINUX_TFRC_H_ +/* + * include/linux/tfrc.h + * + * Copyright (c) 2005 The University of Waikato, Hamilton, New Zealand. + * Copyright (c) 2005 Ian McDonald + * Copyright (c) 2005 Arnaldo Carvalho de Melo + * Copyright (c) 2003 Nils-Erik Mattsson, Joacim Haggmark, Magnus Erixzon + * + * 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. + */ + +#include + +struct tfrc_rx_info { + __u32 tfrcrx_x_recv; + __u32 tfrcrx_rtt; + __u32 tfrcrx_p; +}; + +struct tfrc_tx_info { + __u32 tfrctx_x; + __u32 tfrctx_x_recv; + __u32 tfrctx_x_calc; + __u32 tfrctx_rtt; + __u32 tfrctx_p; + __u32 tfrctx_rto; + __u32 tfrctx_ipi; +}; + +#endif /* _LINUX_TFRC_H_ */ diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index eb248778eea3..0bde4583d091 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -40,6 +40,7 @@ #include #include #include +#include #define TFRC_MIN_PACKET_SIZE 16 #define TFRC_STD_PACKET_SIZE 256 @@ -93,12 +94,15 @@ struct ccid3_options_received { * @ccid3hctx_hist - Packet history */ struct ccid3_hc_tx_sock { - u32 ccid3hctx_x; - u32 ccid3hctx_x_recv; - u32 ccid3hctx_x_calc; + struct tfrc_tx_info ccid3hctx_tfrc; +#define ccid3hctx_x ccid3hctx_tfrc.tfrctx_x +#define ccid3hctx_x_recv ccid3hctx_tfrc.tfrctx_x_recv +#define ccid3hctx_x_calc ccid3hctx_tfrc.tfrctx_x_calc +#define ccid3hctx_rtt ccid3hctx_tfrc.tfrctx_rtt +#define ccid3hctx_p ccid3hctx_tfrc.tfrctx_p +#define ccid3hctx_t_rto ccid3hctx_tfrc.tfrctx_rto +#define ccid3hctx_t_ipi ccid3hctx_tfrc.tfrctx_ipi u16 ccid3hctx_s; - u32 ccid3hctx_rtt; - u32 ccid3hctx_p; u8 ccid3hctx_state; u8 ccid3hctx_last_win_count; u8 ccid3hctx_idle; @@ -106,19 +110,19 @@ struct ccid3_hc_tx_sock { struct timer_list ccid3hctx_no_feedback_timer; struct timeval ccid3hctx_t_ld; struct timeval ccid3hctx_t_nom; - u32 ccid3hctx_t_rto; - u32 ccid3hctx_t_ipi; u32 ccid3hctx_delta; struct list_head ccid3hctx_hist; struct ccid3_options_received ccid3hctx_options_received; }; struct ccid3_hc_rx_sock { + struct tfrc_rx_info ccid3hcrx_tfrc; +#define ccid3hcrx_x_recv ccid3hcrx_tfrc.tfrcrx_x_recv +#define ccid3hcrx_rtt ccid3hcrx_tfrc.tfrcrx_rtt +#define ccid3hcrx_p ccid3hcrx_tfrc.tfrcrx_p u64 ccid3hcrx_seqno_last_counter:48, ccid3hcrx_state:8, ccid3hcrx_last_counter:4; - u32 ccid3hcrx_rtt; - u32 ccid3hcrx_p; u32 ccid3hcrx_bytes_recv; struct timeval ccid3hcrx_tstamp_last_feedback; struct timeval ccid3hcrx_tstamp_last_ack; @@ -127,7 +131,6 @@ struct ccid3_hc_rx_sock { u16 ccid3hcrx_s; u32 ccid3hcrx_pinv; u32 ccid3hcrx_elapsed_time; - u32 ccid3hcrx_x_recv; }; static inline struct ccid3_hc_tx_sock *ccid3_hc_tx_sk(const struct sock *sk) From 561713cf475de1f671cc89c437927ec008a20209 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sun, 18 Sep 2005 00:18:52 -0700 Subject: [PATCH 141/240] [DCCP]: Don't use necessarily the same CCID for tx and rx Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/dccp.h | 3 ++- net/dccp/ipv4.c | 4 ++-- net/dccp/options.c | 3 ++- 3 files changed, 6 insertions(+), 4 deletions(-) diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 8c8e029095a5..13f9b78483fc 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -330,7 +330,8 @@ static inline unsigned int dccp_hdr_len(const struct sk_buff *skb) */ struct dccp_options { __u64 dccpo_sequence_window; - __u8 dccpo_ccid; + __u8 dccpo_rx_ccid; + __u8 dccpo_tx_ccid; __u8 dccpo_send_ack_vector; __u8 dccpo_send_ndp_count; }; diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 82434e4a42df..40fe6afacde6 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -1241,9 +1241,9 @@ static int dccp_v4_init_sock(struct sock *sk) * setsockopt(CCIDs-I-want/accept). -acme */ if (likely(!dccp_ctl_socket_init)) { - dp->dccps_hc_rx_ccid = ccid_init(dp->dccps_options.dccpo_ccid, + dp->dccps_hc_rx_ccid = ccid_init(dp->dccps_options.dccpo_rx_ccid, sk); - dp->dccps_hc_tx_ccid = ccid_init(dp->dccps_options.dccpo_ccid, + dp->dccps_hc_tx_ccid = ccid_init(dp->dccps_options.dccpo_tx_ccid, sk); if (dp->dccps_hc_rx_ccid == NULL || dp->dccps_hc_tx_ccid == NULL) { diff --git a/net/dccp/options.c b/net/dccp/options.c index c480c506a4a4..0a76426c9aea 100644 --- a/net/dccp/options.c +++ b/net/dccp/options.c @@ -25,7 +25,8 @@ /* stores the default values for new connection. may be changed with sysctl */ static const struct dccp_options dccpo_default_values = { .dccpo_sequence_window = DCCPF_INITIAL_SEQUENCE_WINDOW, - .dccpo_ccid = DCCPF_INITIAL_CCID, + .dccpo_rx_ccid = DCCPF_INITIAL_CCID, + .dccpo_tx_ccid = DCCPF_INITIAL_CCID, .dccpo_send_ack_vector = DCCPF_INITIAL_SEND_ACK_VECTOR, .dccpo_send_ndp_count = DCCPF_INITIAL_SEND_NDP_COUNT, }; From 88f964db6ef728982734356bf4c406270ea29c1d Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sun, 18 Sep 2005 00:19:32 -0700 Subject: [PATCH 142/240] [DCCP]: Introduce CCID getsockopt for the CCIDs Allocation for the optnames is similar to the DCCP options, with a range for rx and tx half connection CCIDs. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/dccp.h | 2 ++ net/dccp/ccid.h | 31 +++++++++++++++++++++++ net/dccp/ccids/ccid3.c | 56 ++++++++++++++++++++++++++++++++++++++++++ net/dccp/proto.c | 17 ++++++++----- 4 files changed, 100 insertions(+), 6 deletions(-) diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 13f9b78483fc..71fab4311e92 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -179,6 +179,8 @@ enum { /* DCCP socket options */ #define DCCP_SOCKOPT_PACKET_SIZE 1 #define DCCP_SOCKOPT_SERVICE 2 +#define DCCP_SOCKOPT_CCID_RX_INFO 128 +#define DCCP_SOCKOPT_CCID_TX_INFO 192 #define DCCP_SERVICE_LIST_MAX_LEN 32 diff --git a/net/dccp/ccid.h b/net/dccp/ccid.h index 962f1e9e2f7e..21e55142dcd3 100644 --- a/net/dccp/ccid.h +++ b/net/dccp/ccid.h @@ -14,6 +14,7 @@ */ #include +#include #include #include #include @@ -54,6 +55,14 @@ struct ccid { struct tcp_info *info); void (*ccid_hc_tx_get_info)(struct sock *sk, struct tcp_info *info); + int (*ccid_hc_rx_getsockopt)(struct sock *sk, + const int optname, int len, + u32 __user *optval, + int __user *optlen); + int (*ccid_hc_tx_getsockopt)(struct sock *sk, + const int optname, int len, + u32 __user *optval, + int __user *optlen); }; extern int ccid_register(struct ccid *ccid); @@ -177,4 +186,26 @@ static inline void ccid_hc_tx_get_info(struct ccid *ccid, struct sock *sk, if (ccid->ccid_hc_tx_get_info != NULL) ccid->ccid_hc_tx_get_info(sk, info); } + +static inline int ccid_hc_rx_getsockopt(struct ccid *ccid, struct sock *sk, + const int optname, int len, + u32 __user *optval, int __user *optlen) +{ + int rc = -ENOPROTOOPT; + if (ccid->ccid_hc_rx_getsockopt != NULL) + rc = ccid->ccid_hc_rx_getsockopt(sk, optname, len, + optval, optlen); + return rc; +} + +static inline int ccid_hc_tx_getsockopt(struct ccid *ccid, struct sock *sk, + const int optname, int len, + u32 __user *optval, int __user *optlen) +{ + int rc = -ENOPROTOOPT; + if (ccid->ccid_hc_tx_getsockopt != NULL) + rc = ccid->ccid_hc_tx_getsockopt(sk, optname, len, + optval, optlen); + return rc; +} #endif /* _CCID_H */ diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 38aa84986118..aa68e0ab274d 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -1120,6 +1120,60 @@ static void ccid3_hc_tx_get_info(struct sock *sk, struct tcp_info *info) info->tcpi_rtt = hctx->ccid3hctx_rtt; } +static int ccid3_hc_rx_getsockopt(struct sock *sk, const int optname, int len, + u32 __user *optval, int __user *optlen) +{ + const struct ccid3_hc_rx_sock *hcrx = ccid3_hc_rx_sk(sk); + const void *val; + + /* Listen socks doesn't have a private CCID block */ + if (sk->sk_state == DCCP_LISTEN) + return -EINVAL; + + switch (optname) { + case DCCP_SOCKOPT_CCID_RX_INFO: + if (len < sizeof(hcrx->ccid3hcrx_tfrc)) + return -EINVAL; + len = sizeof(hcrx->ccid3hcrx_tfrc); + val = &hcrx->ccid3hcrx_tfrc; + break; + default: + return -ENOPROTOOPT; + } + + if (put_user(len, optlen) || copy_to_user(optval, val, len)) + return -EFAULT; + + return 0; +} + +static int ccid3_hc_tx_getsockopt(struct sock *sk, const int optname, int len, + u32 __user *optval, int __user *optlen) +{ + const struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); + const void *val; + + /* Listen socks doesn't have a private CCID block */ + if (sk->sk_state == DCCP_LISTEN) + return -EINVAL; + + switch (optname) { + case DCCP_SOCKOPT_CCID_TX_INFO: + if (len < sizeof(hctx->ccid3hctx_tfrc)) + return -EINVAL; + len = sizeof(hctx->ccid3hctx_tfrc); + val = &hctx->ccid3hctx_tfrc; + break; + default: + return -ENOPROTOOPT; + } + + if (put_user(len, optlen) || copy_to_user(optval, val, len)) + return -EFAULT; + + return 0; +} + static struct ccid ccid3 = { .ccid_id = 3, .ccid_name = "ccid3", @@ -1139,6 +1193,8 @@ static struct ccid ccid3 = { .ccid_hc_rx_packet_recv = ccid3_hc_rx_packet_recv, .ccid_hc_rx_get_info = ccid3_hc_rx_get_info, .ccid_hc_tx_get_info = ccid3_hc_tx_get_info, + .ccid_hc_rx_getsockopt = ccid3_hc_rx_getsockopt, + .ccid_hc_tx_getsockopt = ccid3_hc_tx_getsockopt, }; module_param(ccid3_debug, int, 0444); diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 9bda2868eba6..a1cfd0e9e3bc 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -325,12 +325,7 @@ int dccp_getsockopt(struct sock *sk, int level, int optname, if (get_user(len, optlen)) return -EFAULT; - if (optname == DCCP_SOCKOPT_SERVICE) - return dccp_getsockopt_service(sk, len, - (u32 __user *)optval, optlen); - - len = min_t(unsigned int, len, sizeof(int)); - if (len < 0) + if (len < sizeof(int)) return -EINVAL; dp = dccp_sk(sk); @@ -338,7 +333,17 @@ int dccp_getsockopt(struct sock *sk, int level, int optname, switch (optname) { case DCCP_SOCKOPT_PACKET_SIZE: val = dp->dccps_packet_size; + len = sizeof(dp->dccps_packet_size); break; + case DCCP_SOCKOPT_SERVICE: + return dccp_getsockopt_service(sk, len, + (u32 __user *)optval, optlen); + case 128 ... 191: + return ccid_hc_rx_getsockopt(dp->dccps_hc_rx_ccid, sk, optname, + len, (u32 __user *)optval, optlen); + case 192 ... 255: + return ccid_hc_tx_getsockopt(dp->dccps_hc_tx_ccid, sk, optname, + len, (u32 __user *)optval, optlen); default: return -ENOPROTOOPT; } From 40abc27066c49b2c13c817154d438431b0303b96 Mon Sep 17 00:00:00 2001 From: Florin Malita Date: Sun, 18 Sep 2005 00:24:12 -0700 Subject: [PATCH 143/240] [BOND]: Fix bond_init() error path handling. From: Florin Malita bond_init() is not releasing rtnl_sem after register_netdevice() and before calling unregister_netdevice() (from bond_free_all()) in the exception path. As the device registration is not completed (dev->reg_state == NETREG_REGISTERING), the call to unregister_netdevice() triggers BUG_ON(dev->reg_state != NETREG_REGISTERED). Signed-off-by: Florin Malita Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index f8dedb623dc0..90449a0f2a6c 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -5039,6 +5039,14 @@ static int __init bonding_init(void) return 0; out_err: + /* + * rtnl_unlock() will run netdev_run_todo(), putting the + * thus-far-registered bonding devices into a state which + * unregigister_netdevice() will accept + */ + rtnl_unlock(); + rtnl_lock(); + /* free and unregister all bonds that were successfully added */ bond_free_all(); From 987905ded3d19c196dae25cac46c569cac9594b8 Mon Sep 17 00:00:00 2001 From: Mitsuru KANDA Date: Sun, 18 Sep 2005 00:30:08 -0700 Subject: [PATCH 144/240] [IPV6]: Check connect(2) status for IPv6 UDP socket (Re: xfrm_lookup) I think we should cache the per-socket route(dst_entry) only when the IPv6 UDP socket is connect(2)'ed. (which is same as IPv4 UDP send behavior) Signed-off-by: Mitsuru KANDA Signed-off-by: David S. Miller --- net/ipv6/udp.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 2b9bf9bd177f..6001948600f3 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -639,6 +639,7 @@ static int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, int tclass = -1; int corkreq = up->corkflag || msg->msg_flags&MSG_MORE; int err; + int connected = 0; /* destination address check */ if (sin6) { @@ -748,6 +749,7 @@ do_udp_sendmsg: fl->fl_ip_dport = inet->dport; daddr = &np->daddr; fl->fl6_flowlabel = np->flow_label; + connected = 1; } if (!fl->oif) @@ -770,6 +772,7 @@ do_udp_sendmsg: } if (!(opt->opt_nflen|opt->opt_flen)) opt = NULL; + connected = 0; } if (opt == NULL) opt = np->opt; @@ -787,10 +790,13 @@ do_udp_sendmsg: ipv6_addr_copy(&final, &fl->fl6_dst); ipv6_addr_copy(&fl->fl6_dst, rt0->addr); final_p = &final; + connected = 0; } - if (!fl->oif && ipv6_addr_is_multicast(&fl->fl6_dst)) + if (!fl->oif && ipv6_addr_is_multicast(&fl->fl6_dst)) { fl->oif = np->mcast_oif; + connected = 0; + } err = ip6_dst_lookup(sk, &dst, fl); if (err) @@ -846,7 +852,7 @@ do_append_data: else if (!corkreq) err = udp_v6_push_pending_frames(sk, up); - if (dst) + if (dst && connected) ip6_dst_store(sk, dst, ipv6_addr_equal(&fl->fl6_dst, &np->daddr) ? &np->daddr : NULL); From 628f87f3d585bd0c2b0e39df039585d7a5831cc9 Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Sun, 18 Sep 2005 00:33:02 -0700 Subject: [PATCH 145/240] [NETFILTER]: Solve Kconfig dependency problem As suggested by Roman Zippel. Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- net/ipv4/netfilter/Kconfig | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 85190de5710a..e2162d270073 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -8,7 +8,6 @@ menu "IP: Netfilter Configuration" # connection tracking, helpers and protocols config IP_NF_CONNTRACK tristate "Connection tracking (required for masq/NAT)" - select NETFILTER_NETLINK if IP_NF_CONNTRACK_NETLINK!=n ---help--- Connection tracking keeps a record of what packets have passed through your machine, in order to figure out how they are related @@ -55,8 +54,7 @@ config IP_NF_CONNTRACK_EVENTS config IP_NF_CONNTRACK_NETLINK tristate 'Connection tracking netlink interface' depends on IP_NF_CONNTRACK && NETFILTER_NETLINK - default IP_NF_CONNTRACK if NETFILTER_NETLINK=y - default m if NETFILTER_NETLINK=m + depends on IP_NF_CONNTRACK!=y || NETFILTER_NETLINK!=m help This option enables support for a netlink-based userspace interface From 7f8c0fd78dccaf30e60cb4303bd7a21c7d9e6da3 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 18 Sep 2005 21:11:08 +0100 Subject: [PATCH 146/240] [ARM] Fix warning in asm/futex.h The recently added futex.h contains an unused variable, which gcc naturally warns about. Remove this unused variable. Signed-off-by: Russell King --- include/asm-arm/futex.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-arm/futex.h b/include/asm-arm/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-arm/futex.h +++ b/include/asm-arm/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; From e0a200894efa33ea1ac9957717ae151afa02bb6f Mon Sep 17 00:00:00 2001 From: Deepak Saxena Date: Sun, 18 Sep 2005 21:11:56 +0100 Subject: [PATCH 147/240] [ARM] 2917/1: Make IXP4xx mach_desc's depend on config options Patch from Deepak Saxena Building a kernel for IXDP425 currently includes the machine descriptors for IXDP465 and PRPMC1100 even if those machines are not configured. This means we can build a kernel that boots on those machines even though the machine_is_xxx() macro will always return 0 and other bits such as PCI won't be compiled in. This can lead to many wasted hours wondering what you have done to your kernel to make it randomly crash thus requireing large quantities of beer to be consumed. While I am all for consumption of large quantities of beer, there are better reasons to do so then stupid kernel bugs. Signed-off-by: Deepak Saxena Signed-off-by: Russell King --- arch/arm/mach-ixp4xx/ixdp425-setup.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/arm/mach-ixp4xx/ixdp425-setup.c b/arch/arm/mach-ixp4xx/ixdp425-setup.c index ae1fa099d5fa..39b06ed80646 100644 --- a/arch/arm/mach-ixp4xx/ixdp425-setup.c +++ b/arch/arm/mach-ixp4xx/ixdp425-setup.c @@ -123,6 +123,7 @@ static void __init ixdp425_init(void) platform_add_devices(ixdp425_devices, ARRAY_SIZE(ixdp425_devices)); } +#ifdef CONFIG_ARCH_IXDP465 MACHINE_START(IXDP425, "Intel IXDP425 Development Platform") /* Maintainer: MontaVista Software, Inc. */ .phys_ram = PHYS_OFFSET, @@ -134,7 +135,9 @@ MACHINE_START(IXDP425, "Intel IXDP425 Development Platform") .boot_params = 0x0100, .init_machine = ixdp425_init, MACHINE_END +#endif +#ifdef CONFIG_MACH_IXDP465 MACHINE_START(IXDP465, "Intel IXDP465 Development Platform") /* Maintainer: MontaVista Software, Inc. */ .phys_ram = PHYS_OFFSET, @@ -146,7 +149,9 @@ MACHINE_START(IXDP465, "Intel IXDP465 Development Platform") .boot_params = 0x0100, .init_machine = ixdp425_init, MACHINE_END +#endif +#ifdef CONFIG_ARCH_PRPMC1100 MACHINE_START(IXCDP1100, "Intel IXCDP1100 Development Platform") /* Maintainer: MontaVista Software, Inc. */ .phys_ram = PHYS_OFFSET, @@ -158,6 +163,7 @@ MACHINE_START(IXCDP1100, "Intel IXCDP1100 Development Platform") .boot_params = 0x0100, .init_machine = ixdp425_init, MACHINE_END +#endif /* * Avila is functionally equivalent to IXDP425 except that it adds From a64358db1253b35d508a411e80a3ad23b859ec88 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 26 Jul 2005 10:27:10 -0400 Subject: [PATCH 148/240] [SCSI] SCSI scanning and removal fixes This patch (as545) fixes the list traversals in __scsi_remove_target and scsi_forget_host. In each case the existing code list_for_each_entry_safe in an _unsafe_ manner, because the list was not protected from outside modification while the iteration was running. The new scsi_forget_host routine takes the moderately controversial step of iterating over devices for removal rather than iterating over targets. This makes more sense to me because the current scheme treats targets as second-class citizens, created and removed on demand, rather than as objects corresponding to actual hardware. (Also I couldn't figure out any safe way to iterate over the target list, since it's not so easy to tell when a target has already been removed.) Signed-off-by: Alan Stern Signed-off-by: James Bottomley --- drivers/scsi/scsi_scan.c | 20 +++++++------------- drivers/scsi/scsi_sysfs.c | 9 ++++++--- 2 files changed, 13 insertions(+), 16 deletions(-) diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index b86f170fa8ed..fcf9f6cbb142 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -1466,23 +1466,17 @@ EXPORT_SYMBOL(scsi_scan_single_target); void scsi_forget_host(struct Scsi_Host *shost) { - struct scsi_target *starget, *tmp; + struct scsi_device *sdev; unsigned long flags; - /* - * Ok, this look a bit strange. We always look for the first device - * on the list as scsi_remove_device removes them from it - thus we - * also have to release the lock. - * We don't need to get another reference to the device before - * releasing the lock as we already own the reference from - * scsi_register_device that's release in scsi_remove_device. And - * after that we don't look at sdev anymore. - */ + restart: spin_lock_irqsave(shost->host_lock, flags); - list_for_each_entry_safe(starget, tmp, &shost->__targets, siblings) { + list_for_each_entry(sdev, &shost->__devices, siblings) { + if (sdev->sdev_state == SDEV_DEL) + continue; spin_unlock_irqrestore(shost->host_lock, flags); - scsi_remove_target(&starget->dev); - spin_lock_irqsave(shost->host_lock, flags); + __scsi_remove_device(sdev); + goto restart; } spin_unlock_irqrestore(shost->host_lock, flags); } diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c index 83f87c41b18d..1e47b7eddef4 100644 --- a/drivers/scsi/scsi_sysfs.c +++ b/drivers/scsi/scsi_sysfs.c @@ -719,17 +719,20 @@ void __scsi_remove_target(struct scsi_target *starget) { struct Scsi_Host *shost = dev_to_shost(starget->dev.parent); unsigned long flags; - struct scsi_device *sdev, *tmp; + struct scsi_device *sdev; spin_lock_irqsave(shost->host_lock, flags); starget->reap_ref++; - list_for_each_entry_safe(sdev, tmp, &shost->__devices, siblings) { + restart: + list_for_each_entry(sdev, &shost->__devices, siblings) { if (sdev->channel != starget->channel || - sdev->id != starget->id) + sdev->id != starget->id || + sdev->sdev_state == SDEV_DEL) continue; spin_unlock_irqrestore(shost->host_lock, flags); scsi_remove_device(sdev); spin_lock_irqsave(shost->host_lock, flags); + goto restart; } spin_unlock_irqrestore(shost->host_lock, flags); scsi_target_reap(starget); From 63150fcf73dbfb752a850a5e2fe3aed570154433 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 18 Sep 2005 21:49:29 +0100 Subject: [PATCH 149/240] [ARM] Fix warning in arch/arm/kernel/semaphore.c Newer binutils complains: /tmp/cc07pbI9.s:146: Warning: ignoring changed section type for .sched.text Fix this warning by adding %progbits to the .section. Signed-off-by: Russell King --- arch/arm/kernel/semaphore.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/kernel/semaphore.c b/arch/arm/kernel/semaphore.c index ac423e3e224b..4c31f2923055 100644 --- a/arch/arm/kernel/semaphore.c +++ b/arch/arm/kernel/semaphore.c @@ -178,7 +178,7 @@ int __down_trylock(struct semaphore * sem) * registers (r0 to r3 and lr), but not ip, as we use it as a return * value in some cases.. */ -asm(" .section .sched.text,\"ax\" \n\ +asm(" .section .sched.text,\"ax\",%progbits \n\ .align 5 \n\ .globl __down_failed \n\ __down_failed: \n\ From 118ec0b3e02e7d22cf27b9f5a2836fa6f395b61a Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 18 Sep 2005 22:16:41 +0100 Subject: [PATCH 150/240] [ARM] Fix warning in arch/arm/mach-sa1100/generic.c Fix: arch/arm/mach-sa1100/generic.c:224: warning: 'struct mcp_plat_data' declared inside parameter list caused by mussing structure and function declaration. Signed-off-by: Russell King --- arch/arm/mach-sa1100/generic.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/arm/mach-sa1100/generic.h b/arch/arm/mach-sa1100/generic.h index 279e3afa3c39..f085d68e568e 100644 --- a/arch/arm/mach-sa1100/generic.h +++ b/arch/arm/mach-sa1100/generic.h @@ -39,3 +39,6 @@ extern void sa11x0_set_ssp_data(struct sa11x0_ssp_plat_ops *ops); struct irda_platform_data; void sa11x0_set_irda_data(struct irda_platform_data *irda); + +struct mcp_plat_data; +void sa11x0_set_mcp_data(struct mcp_plat_data *data); From 51574e0398a2d93cbf7f26e36b673cd919062268 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 12 Sep 2005 09:52:28 -0700 Subject: [PATCH 151/240] [PATCH] IPoIB: fix module removal race Since ipoib uses queue_delayed_work to run flush task on port state events, it must flush scheduled work after unregistering the event handler. Signed-off-by: Michael S. Tsirkin Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/ipoib/ipoib_main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 49d120d2b92c..704f48e0b6a7 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -1005,6 +1005,7 @@ debug_failed: register_failed: ib_unregister_event_handler(&priv->event_handler); + flush_scheduled_work(); event_failed: ipoib_dev_cleanup(priv->dev); @@ -1057,6 +1058,7 @@ static void ipoib_remove_one(struct ib_device *device) list_for_each_entry_safe(priv, tmp, dev_list, list) { ib_unregister_event_handler(&priv->event_handler); + flush_scheduled_work(); unregister_netdev(priv->dev); ipoib_dev_cleanup(priv->dev); From bb4a7f0da75ce9f3e933880428d39b2aa1f16961 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Mon, 12 Sep 2005 14:08:51 -0700 Subject: [PATCH 152/240] [PATCH] IB/mthca: assign ACK timeout field correctly The hardware reads the ACK timeout field from the most significant 5 bits of struct mthca_qp_path's ackto field, not the least significant bits. This fix has the driver put the timeout in the right place. Without this, we get a timeout that is 2^8 times too small. Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mthca/mthca_qp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c index bcef06bf15e7..2c883c1fd53a 100644 --- a/drivers/infiniband/hw/mthca/mthca_qp.c +++ b/drivers/infiniband/hw/mthca/mthca_qp.c @@ -687,7 +687,7 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask) } if (attr_mask & IB_QP_TIMEOUT) { - qp_context->pri_path.ackto = attr->timeout; + qp_context->pri_path.ackto = attr->timeout << 3; qp_param->opt_param_mask |= cpu_to_be32(MTHCA_QP_OPTPAR_ACK_TIMEOUT); } From d6cff021e24515255b296d399ec517a68bf2ed08 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Tue, 13 Sep 2005 10:41:03 -0700 Subject: [PATCH 153/240] [PATCH] IB/mthca: fix posting of first work request Fix posting first WQE for mem-free HCAs: we need to link to previous WQE even in that case. While we're at it, simplify code for Tavor-mode HCAs. We don't really need the conditional test there either; we can similarly always link to the previous WQE. Based on Michael S. Tsirkin's analogous fix for userspace libmthca. Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mthca/mthca_qp.c | 48 ++++++++++++------------- drivers/infiniband/hw/mthca/mthca_srq.c | 14 ++++---- 2 files changed, 28 insertions(+), 34 deletions(-) diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c index 2c883c1fd53a..eaa521857630 100644 --- a/drivers/infiniband/hw/mthca/mthca_qp.c +++ b/drivers/infiniband/hw/mthca/mthca_qp.c @@ -227,7 +227,6 @@ static void mthca_wq_init(struct mthca_wq *wq) wq->last_comp = wq->max - 1; wq->head = 0; wq->tail = 0; - wq->last = NULL; } void mthca_qp_event(struct mthca_dev *dev, u32 qpn, @@ -1103,6 +1102,9 @@ static int mthca_alloc_qp_common(struct mthca_dev *dev, } } + qp->sq.last = get_send_wqe(qp, qp->sq.max - 1); + qp->rq.last = get_recv_wqe(qp, qp->rq.max - 1); + return 0; } @@ -1583,15 +1585,13 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, goto out; } - if (prev_wqe) { - ((struct mthca_next_seg *) prev_wqe)->nda_op = - cpu_to_be32(((ind << qp->sq.wqe_shift) + - qp->send_wqe_offset) | - mthca_opcode[wr->opcode]); - wmb(); - ((struct mthca_next_seg *) prev_wqe)->ee_nds = - cpu_to_be32((size0 ? 0 : MTHCA_NEXT_DBD) | size); - } + ((struct mthca_next_seg *) prev_wqe)->nda_op = + cpu_to_be32(((ind << qp->sq.wqe_shift) + + qp->send_wqe_offset) | + mthca_opcode[wr->opcode]); + wmb(); + ((struct mthca_next_seg *) prev_wqe)->ee_nds = + cpu_to_be32((size0 ? 0 : MTHCA_NEXT_DBD) | size); if (!size0) { size0 = size; @@ -1688,13 +1688,11 @@ int mthca_tavor_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, qp->wrid[ind] = wr->wr_id; - if (likely(prev_wqe)) { - ((struct mthca_next_seg *) prev_wqe)->nda_op = - cpu_to_be32((ind << qp->rq.wqe_shift) | 1); - wmb(); - ((struct mthca_next_seg *) prev_wqe)->ee_nds = - cpu_to_be32(MTHCA_NEXT_DBD | size); - } + ((struct mthca_next_seg *) prev_wqe)->nda_op = + cpu_to_be32((ind << qp->rq.wqe_shift) | 1); + wmb(); + ((struct mthca_next_seg *) prev_wqe)->ee_nds = + cpu_to_be32(MTHCA_NEXT_DBD | size); if (!size0) size0 = size; @@ -1905,15 +1903,13 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, goto out; } - if (likely(prev_wqe)) { - ((struct mthca_next_seg *) prev_wqe)->nda_op = - cpu_to_be32(((ind << qp->sq.wqe_shift) + - qp->send_wqe_offset) | - mthca_opcode[wr->opcode]); - wmb(); - ((struct mthca_next_seg *) prev_wqe)->ee_nds = - cpu_to_be32(MTHCA_NEXT_DBD | size); - } + ((struct mthca_next_seg *) prev_wqe)->nda_op = + cpu_to_be32(((ind << qp->sq.wqe_shift) + + qp->send_wqe_offset) | + mthca_opcode[wr->opcode]); + wmb(); + ((struct mthca_next_seg *) prev_wqe)->ee_nds = + cpu_to_be32(MTHCA_NEXT_DBD | size); if (!size0) { size0 = size; diff --git a/drivers/infiniband/hw/mthca/mthca_srq.c b/drivers/infiniband/hw/mthca/mthca_srq.c index 75cd2d84ef12..fe06cc0df936 100644 --- a/drivers/infiniband/hw/mthca/mthca_srq.c +++ b/drivers/infiniband/hw/mthca/mthca_srq.c @@ -189,7 +189,6 @@ int mthca_alloc_srq(struct mthca_dev *dev, struct mthca_pd *pd, srq->max = attr->max_wr; srq->max_gs = attr->max_sge; - srq->last = NULL; srq->counter = 0; if (mthca_is_memfree(dev)) @@ -264,6 +263,7 @@ int mthca_alloc_srq(struct mthca_dev *dev, struct mthca_pd *pd, srq->first_free = 0; srq->last_free = srq->max - 1; + srq->last = get_wqe(srq, srq->max - 1); return 0; @@ -446,13 +446,11 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, ((struct mthca_data_seg *) wqe)->addr = 0; } - if (likely(prev_wqe)) { - ((struct mthca_next_seg *) prev_wqe)->nda_op = - cpu_to_be32((ind << srq->wqe_shift) | 1); - wmb(); - ((struct mthca_next_seg *) prev_wqe)->ee_nds = - cpu_to_be32(MTHCA_NEXT_DBD); - } + ((struct mthca_next_seg *) prev_wqe)->nda_op = + cpu_to_be32((ind << srq->wqe_shift) | 1); + wmb(); + ((struct mthca_next_seg *) prev_wqe)->ee_nds = + cpu_to_be32(MTHCA_NEXT_DBD); srq->wrid[ind] = wr->wr_id; srq->first_free = next_ind; From ce5b65cc9626feac0d4ffb96f798407e50c45575 Mon Sep 17 00:00:00 2001 From: Hal Rosenstock Date: Sun, 18 Sep 2005 13:47:53 -0700 Subject: [PATCH 154/240] [PATCH] IPoIB: Fix SA client retransmission strategy We got a little mixed up with what the backoff member holds in the IPoIB multicast group structure: sometimes it was used as a number of seconds, and sometimes it was used as a number of jiffies. Fix the code so that backoff is always in seconds. Signed-off-by: Hal Rosenstock Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index aca7aea18a69..3aab4d85abfc 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -145,7 +145,7 @@ static struct ipoib_mcast *ipoib_mcast_alloc(struct net_device *dev, mcast->dev = dev; mcast->created = jiffies; - mcast->backoff = HZ; + mcast->backoff = 1; mcast->logcount = 0; INIT_LIST_HEAD(&mcast->list); @@ -396,7 +396,7 @@ static void ipoib_mcast_join_complete(int status, IPOIB_GID_ARG(mcast->mcmember.mgid), status); if (!status && !ipoib_mcast_join_finish(mcast, mcmember)) { - mcast->backoff = HZ; + mcast->backoff = 1; down(&mcast_mutex); if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) queue_work(ipoib_workqueue, &priv->mcast_task); @@ -496,7 +496,7 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast, if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) queue_delayed_work(ipoib_workqueue, &priv->mcast_task, - mcast->backoff); + mcast->backoff * HZ); up(&mcast_mutex); } else mcast->query_id = ret; From c915033fc62d7186d243d89f88782d6be33fd8f6 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Sun, 18 Sep 2005 13:52:06 -0700 Subject: [PATCH 155/240] [PATCH] IB/mthca: Initialize eq->nent before we use it In mthca_create_eq(), we call get_eqe() before setting eq->nent. This is wrong, because get_eqe() uses eq->nent. Fix this, and clean up the code a little while we're at it. (We got lucky with the current code, because eq->nent was cleared to 0, which get_eqe() made happen to do the right thing) Pointed out by Michael S. Tsirkin Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mthca/mthca_eq.c | 16 +++++----------- 1 file changed, 5 insertions(+), 11 deletions(-) diff --git a/drivers/infiniband/hw/mthca/mthca_eq.c b/drivers/infiniband/hw/mthca/mthca_eq.c index 18f0981eb0c1..78152a8ad17d 100644 --- a/drivers/infiniband/hw/mthca/mthca_eq.c +++ b/drivers/infiniband/hw/mthca/mthca_eq.c @@ -476,12 +476,8 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev, int i; u8 status; - /* Make sure EQ size is aligned to a power of 2 size. */ - for (i = 1; i < nent; i <<= 1) - ; /* nothing */ - nent = i; - - eq->dev = dev; + eq->dev = dev; + eq->nent = roundup_pow_of_two(max(nent, 2)); eq->page_list = kmalloc(npages * sizeof *eq->page_list, GFP_KERNEL); @@ -512,7 +508,7 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev, memset(eq->page_list[i].buf, 0, PAGE_SIZE); } - for (i = 0; i < nent; ++i) + for (i = 0; i < eq->nent; ++i) set_eqe_hw(get_eqe(eq, i)); eq->eqn = mthca_alloc(&dev->eq_table.alloc); @@ -528,8 +524,6 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev, if (err) goto err_out_free_eq; - eq->nent = nent; - memset(eq_context, 0, sizeof *eq_context); eq_context->flags = cpu_to_be32(MTHCA_EQ_STATUS_OK | MTHCA_EQ_OWNER_HW | @@ -538,7 +532,7 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev, if (mthca_is_memfree(dev)) eq_context->flags |= cpu_to_be32(MTHCA_EQ_STATE_ARBEL); - eq_context->logsize_usrpage = cpu_to_be32((ffs(nent) - 1) << 24); + eq_context->logsize_usrpage = cpu_to_be32((ffs(eq->nent) - 1) << 24); if (mthca_is_memfree(dev)) { eq_context->arbel_pd = cpu_to_be32(dev->driver_pd.pd_num); } else { @@ -569,7 +563,7 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev, dev->eq_table.arm_mask |= eq->eqn_mask; mthca_dbg(dev, "Allocated EQ %d with %d entries\n", - eq->eqn, nent); + eq->eqn, eq->nent); return err; From 3853194c2e174cee4da093c67bd54cbf9a38559a Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Sun, 18 Sep 2005 14:00:17 -0700 Subject: [PATCH 156/240] [PATCH] IB/mthca: Fix posting work requests to shared receive queues The error handling paths in mthca_tavor_post_srq_recv() and mthca_arbel_post_srq_recv() are quite bogus, the result of a screwed up merge. Fix them so they work as intended. Pointed out by Michael S. Tsirkin Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mthca/mthca_srq.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/infiniband/hw/mthca/mthca_srq.c b/drivers/infiniband/hw/mthca/mthca_srq.c index fe06cc0df936..45dedd2787a4 100644 --- a/drivers/infiniband/hw/mthca/mthca_srq.c +++ b/drivers/infiniband/hw/mthca/mthca_srq.c @@ -409,7 +409,7 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, mthca_err(dev, "SRQ %06x full\n", srq->srqn); err = -ENOMEM; *bad_wr = wr; - return nreq; + break; } wqe = get_wqe(srq, ind); @@ -427,7 +427,7 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, err = -EINVAL; *bad_wr = wr; srq->last = prev_wqe; - return nreq; + break; } for (i = 0; i < wr->num_sge; ++i) { @@ -456,8 +456,6 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, srq->first_free = next_ind; } - return nreq; - if (likely(nreq)) { __be32 doorbell[2]; @@ -501,7 +499,7 @@ int mthca_arbel_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, mthca_err(dev, "SRQ %06x full\n", srq->srqn); err = -ENOMEM; *bad_wr = wr; - return nreq; + break; } wqe = get_wqe(srq, ind); @@ -517,7 +515,7 @@ int mthca_arbel_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, if (unlikely(wr->num_sge > srq->max_gs)) { err = -EINVAL; *bad_wr = wr; - return nreq; + break; } for (i = 0; i < wr->num_sge; ++i) { From 5c9f6de3b80ca46000bd1b63d892820f9ee32138 Mon Sep 17 00:00:00 2001 From: Anton Altaparmakov Date: Mon, 19 Sep 2005 09:33:40 +0100 Subject: [PATCH 157/240] NTFS: Fix various bugs in the runlist merging code. (Based on libntfs changes by Richard Russon.) Signed-off-by: Anton Altaparmakov --- fs/ntfs/ChangeLog | 2 + fs/ntfs/runlist.c | 132 ++++++++++++++++++++++++---------------------- 2 files changed, 70 insertions(+), 64 deletions(-) diff --git a/fs/ntfs/ChangeLog b/fs/ntfs/ChangeLog index 49eafbdb15c1..c7e9237379c2 100644 --- a/fs/ntfs/ChangeLog +++ b/fs/ntfs/ChangeLog @@ -92,6 +92,8 @@ ToDo/Notes: an octal number to conform to how chmod(1) works, too. Thanks to Giuseppe Bilotta and Horst von Brand for pointing out the errors of my ways. + - Fix various bugs in the runlist merging code. (Based on libntfs + changes by Richard Russon.) 2.1.23 - Implement extension of resident files and make writing safe as well as many bug fixes, cleanups, and enhancements... diff --git a/fs/ntfs/runlist.c b/fs/ntfs/runlist.c index f5b2ac929081..e2665d011d72 100644 --- a/fs/ntfs/runlist.c +++ b/fs/ntfs/runlist.c @@ -2,7 +2,7 @@ * runlist.c - NTFS runlist handling code. Part of the Linux-NTFS project. * * Copyright (c) 2001-2005 Anton Altaparmakov - * Copyright (c) 2002 Richard Russon + * Copyright (c) 2002-2005 Richard Russon * * This program/include file is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License as published @@ -214,8 +214,8 @@ static inline void __ntfs_rl_merge(runlist_element *dst, runlist_element *src) static inline runlist_element *ntfs_rl_append(runlist_element *dst, int dsize, runlist_element *src, int ssize, int loc) { - BOOL right; - int magic; + BOOL right; /* Right end of @src needs merging. */ + int marker; /* End of the inserted runs. */ BUG_ON(!dst); BUG_ON(!src); @@ -236,18 +236,19 @@ static inline runlist_element *ntfs_rl_append(runlist_element *dst, if (right) __ntfs_rl_merge(src + ssize - 1, dst + loc + 1); - magic = loc + ssize; + /* First run after the @src runs that have been inserted. */ + marker = loc + ssize + 1; /* Move the tail of @dst out of the way, then copy in @src. */ - ntfs_rl_mm(dst, magic + 1, loc + 1 + right, dsize - loc - 1 - right); + ntfs_rl_mm(dst, marker, loc + 1 + right, dsize - (loc + 1 + right)); ntfs_rl_mc(dst, loc + 1, src, 0, ssize); /* Adjust the size of the preceding hole. */ dst[loc].length = dst[loc + 1].vcn - dst[loc].vcn; /* We may have changed the length of the file, so fix the end marker */ - if (dst[magic + 1].lcn == LCN_ENOENT) - dst[magic + 1].vcn = dst[magic].vcn + dst[magic].length; + if (dst[marker].lcn == LCN_ENOENT) + dst[marker].vcn = dst[marker - 1].vcn + dst[marker - 1].length; return dst; } @@ -279,18 +280,17 @@ static inline runlist_element *ntfs_rl_append(runlist_element *dst, static inline runlist_element *ntfs_rl_insert(runlist_element *dst, int dsize, runlist_element *src, int ssize, int loc) { - BOOL left = FALSE; - BOOL disc = FALSE; /* Discontinuity */ - BOOL hole = FALSE; /* Following a hole */ - int magic; + BOOL left = FALSE; /* Left end of @src needs merging. */ + BOOL disc = FALSE; /* Discontinuity between @dst and @src. */ + int marker; /* End of the inserted runs. */ BUG_ON(!dst); BUG_ON(!src); - /* disc => Discontinuity between the end of @dst and the start of @src. - * This means we might need to insert a hole. - * hole => @dst ends with a hole or an unmapped region which we can - * extend to match the discontinuity. */ + /* + * disc => Discontinuity between the end of @dst and the start of @src. + * This means we might need to insert a "not mapped" run. + */ if (loc == 0) disc = (src[0].vcn > 0); else { @@ -303,58 +303,49 @@ static inline runlist_element *ntfs_rl_insert(runlist_element *dst, merged_length += src->length; disc = (src[0].vcn > dst[loc - 1].vcn + merged_length); - if (disc) - hole = (dst[loc - 1].lcn == LCN_HOLE); } - - /* Space required: @dst size + @src size, less one if we merged, plus - * one if there was a discontinuity, less one for a trailing hole. */ - dst = ntfs_rl_realloc(dst, dsize, dsize + ssize - left + disc - hole); + /* + * Space required: @dst size + @src size, less one if we merged, plus + * one if there was a discontinuity. + */ + dst = ntfs_rl_realloc(dst, dsize, dsize + ssize - left + disc); if (IS_ERR(dst)) return dst; /* * We are guaranteed to succeed from here so can start modifying the * original runlist. */ - if (left) __ntfs_rl_merge(dst + loc - 1, src); - - magic = loc + ssize - left + disc - hole; + /* + * First run after the @src runs that have been inserted. + * Nominally, @marker equals @loc + @ssize, i.e. location + number of + * runs in @src. However, if @left, then the first run in @src has + * been merged with one in @dst. And if @disc, then @dst and @src do + * not meet and we need an extra run to fill the gap. + */ + marker = loc + ssize - left + disc; /* Move the tail of @dst out of the way, then copy in @src. */ - ntfs_rl_mm(dst, magic, loc, dsize - loc); - ntfs_rl_mc(dst, loc + disc - hole, src, left, ssize - left); + ntfs_rl_mm(dst, marker, loc, dsize - loc); + ntfs_rl_mc(dst, loc + disc, src, left, ssize - left); - /* Adjust the VCN of the last run ... */ - if (dst[magic].lcn <= LCN_HOLE) - dst[magic].vcn = dst[magic - 1].vcn + dst[magic - 1].length; + /* Adjust the VCN of the first run after the insertion... */ + dst[marker].vcn = dst[marker - 1].vcn + dst[marker - 1].length; /* ... and the length. */ - if (dst[magic].lcn == LCN_HOLE || dst[magic].lcn == LCN_RL_NOT_MAPPED) - dst[magic].length = dst[magic + 1].vcn - dst[magic].vcn; + if (dst[marker].lcn == LCN_HOLE || dst[marker].lcn == LCN_RL_NOT_MAPPED) + dst[marker].length = dst[marker + 1].vcn - dst[marker].vcn; - /* Writing beyond the end of the file and there's a discontinuity. */ + /* Writing beyond the end of the file and there is a discontinuity. */ if (disc) { - if (hole) - dst[loc - 1].length = dst[loc].vcn - dst[loc - 1].vcn; - else { - if (loc > 0) { - dst[loc].vcn = dst[loc - 1].vcn + - dst[loc - 1].length; - dst[loc].length = dst[loc + 1].vcn - - dst[loc].vcn; - } else { - dst[loc].vcn = 0; - dst[loc].length = dst[loc + 1].vcn; - } - dst[loc].lcn = LCN_RL_NOT_MAPPED; + if (loc > 0) { + dst[loc].vcn = dst[loc - 1].vcn + dst[loc - 1].length; + dst[loc].length = dst[loc + 1].vcn - dst[loc].vcn; + } else { + dst[loc].vcn = 0; + dst[loc].length = dst[loc + 1].vcn; } - - magic += hole; - - if (dst[magic].lcn == LCN_ENOENT) - dst[magic].vcn = dst[magic - 1].vcn + - dst[magic - 1].length; + dst[loc].lcn = LCN_RL_NOT_MAPPED; } return dst; } @@ -385,9 +376,10 @@ static inline runlist_element *ntfs_rl_insert(runlist_element *dst, static inline runlist_element *ntfs_rl_replace(runlist_element *dst, int dsize, runlist_element *src, int ssize, int loc) { - BOOL left = FALSE; - BOOL right; - int magic; + BOOL left = FALSE; /* Left end of @src needs merging. */ + BOOL right; /* Right end of @src needs merging. */ + int tail; /* Start of tail of @dst. */ + int marker; /* End of the inserted runs. */ BUG_ON(!dst); BUG_ON(!src); @@ -396,9 +388,10 @@ static inline runlist_element *ntfs_rl_replace(runlist_element *dst, right = ntfs_are_rl_mergeable(src + ssize - 1, dst + loc + 1); if (loc > 0) left = ntfs_are_rl_mergeable(dst + loc - 1, src); - - /* Allocate some space. We'll need less if the left, right, or both - * ends were merged. */ + /* + * Allocate some space. We will need less if the left, right, or both + * ends were merged. + */ dst = ntfs_rl_realloc(dst, dsize, dsize + ssize - left - right); if (IS_ERR(dst)) return dst; @@ -410,17 +403,28 @@ static inline runlist_element *ntfs_rl_replace(runlist_element *dst, __ntfs_rl_merge(src + ssize - 1, dst + loc + 1); if (left) __ntfs_rl_merge(dst + loc - 1, src); - - /* FIXME: What does this mean? (AIA) */ - magic = loc + ssize - left; + /* + * First run of @dst that needs to be moved out of the way to make + * space for the runs to be copied from @src, i.e. the first run of the + * tail of @dst. + */ + tail = loc + right + 1; + /* + * First run after the @src runs that have been inserted, i.e. where + * the tail of @dst needs to be moved to. + * Nominally, marker equals @loc + @ssize, i.e. location + number of + * runs in @src). However, if @left, then the first run in @src has + * been merged with one in @dst. + */ + marker = loc + ssize - left; /* Move the tail of @dst out of the way, then copy in @src. */ - ntfs_rl_mm(dst, magic, loc + right + 1, dsize - loc - right - 1); + ntfs_rl_mm(dst, marker, tail, dsize - tail); ntfs_rl_mc(dst, loc, src, left, ssize - left); - /* We may have changed the length of the file, so fix the end marker */ - if (dst[magic].lcn == LCN_ENOENT) - dst[magic].vcn = dst[magic - 1].vcn + dst[magic - 1].length; + /* We may have changed the length of the file, so fix the end marker. */ + if (dsize - tail > 0 && dst[marker].lcn == LCN_ENOENT) + dst[marker].vcn = dst[marker - 1].vcn + dst[marker - 1].length; return dst; } From 4e64c88693fde1b1cbaa4cfecad43a0c3fad354e Mon Sep 17 00:00:00 2001 From: Anton Altaparmakov Date: Mon, 19 Sep 2005 09:38:41 +0100 Subject: [PATCH 158/240] NTFS: Fix handling of compressed directories that I broke in earlier changeset. Signed-off-by: Anton Altaparmakov --- fs/ntfs/aops.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c index b6cc8cf24626..5fd516f42eec 100644 --- a/fs/ntfs/aops.c +++ b/fs/ntfs/aops.c @@ -389,9 +389,11 @@ retry_readpage: * Only $DATA attributes can be encrypted and only unnamed $DATA * attributes can be compressed. Index root can have the flags set but * this means to create compressed/encrypted files, not that the - * attribute is compressed/encrypted. + * attribute is compressed/encrypted. Note we need to check for + * AT_INDEX_ALLOCATION since this is the type of both directory and + * index inodes. */ - if (ni->type != AT_INDEX_ROOT) { + if (ni->type != AT_INDEX_ALLOCATION) { /* If attribute is encrypted, deny access, just like NT4. */ if (NInoEncrypted(ni)) { BUG_ON(ni->type != AT_DATA); @@ -1341,9 +1343,11 @@ retry_writepage: * Only $DATA attributes can be encrypted and only unnamed $DATA * attributes can be compressed. Index root can have the flags set but * this means to create compressed/encrypted files, not that the - * attribute is compressed/encrypted. + * attribute is compressed/encrypted. Note we need to check for + * AT_INDEX_ALLOCATION since this is the type of both directory and + * index inodes. */ - if (ni->type != AT_INDEX_ROOT) { + if (ni->type != AT_INDEX_ALLOCATION) { /* If file is encrypted, deny access, just like NT4. */ if (NInoEncrypted(ni)) { unlock_page(page); From f6098cf449b81c14a51e48dd22ae47d03126a1de Mon Sep 17 00:00:00 2001 From: Anton Altaparmakov Date: Mon, 19 Sep 2005 09:41:39 +0100 Subject: [PATCH 159/240] NTFS: Fix ntfs_{read,write}page() to cope with concurrent truncates better. Signed-off-by: Anton Altaparmakov --- fs/ntfs/aops.c | 108 ++++++++++++++++++++++++++++++++--------------- fs/ntfs/inode.c | 9 ++-- fs/ntfs/malloc.h | 2 +- 3 files changed, 79 insertions(+), 40 deletions(-) diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c index 5fd516f42eec..5e80c07c6a4d 100644 --- a/fs/ntfs/aops.c +++ b/fs/ntfs/aops.c @@ -59,39 +59,49 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate) unsigned long flags; struct buffer_head *first, *tmp; struct page *page; + struct inode *vi; ntfs_inode *ni; int page_uptodate = 1; page = bh->b_page; - ni = NTFS_I(page->mapping->host); + vi = page->mapping->host; + ni = NTFS_I(vi); if (likely(uptodate)) { - s64 file_ofs, initialized_size; + loff_t i_size; + s64 file_ofs, init_size; set_buffer_uptodate(bh); file_ofs = ((s64)page->index << PAGE_CACHE_SHIFT) + bh_offset(bh); read_lock_irqsave(&ni->size_lock, flags); - initialized_size = ni->initialized_size; + init_size = ni->initialized_size; + i_size = i_size_read(vi); read_unlock_irqrestore(&ni->size_lock, flags); + if (unlikely(init_size > i_size)) { + /* Race with shrinking truncate. */ + init_size = i_size; + } /* Check for the current buffer head overflowing. */ - if (file_ofs + bh->b_size > initialized_size) { - char *addr; - int ofs = 0; + if (unlikely(file_ofs + bh->b_size > init_size)) { + u8 *kaddr; + int ofs; - if (file_ofs < initialized_size) - ofs = initialized_size - file_ofs; - addr = kmap_atomic(page, KM_BIO_SRC_IRQ); - memset(addr + bh_offset(bh) + ofs, 0, bh->b_size - ofs); + ofs = 0; + if (file_ofs < init_size) + ofs = init_size - file_ofs; + kaddr = kmap_atomic(page, KM_BIO_SRC_IRQ); + memset(kaddr + bh_offset(bh) + ofs, 0, + bh->b_size - ofs); + kunmap_atomic(kaddr, KM_BIO_SRC_IRQ); flush_dcache_page(page); - kunmap_atomic(addr, KM_BIO_SRC_IRQ); } } else { clear_buffer_uptodate(bh); SetPageError(page); - ntfs_error(ni->vol->sb, "Buffer I/O error, logical block %llu.", - (unsigned long long)bh->b_blocknr); + ntfs_error(ni->vol->sb, "Buffer I/O error, logical block " + "0x%llx.", (unsigned long long)bh->b_blocknr); } first = page_buffers(page); local_irq_save(flags); @@ -124,7 +134,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate) if (likely(page_uptodate && !PageError(page))) SetPageUptodate(page); } else { - char *addr; + u8 *kaddr; unsigned int i, recs; u32 rec_size; @@ -132,12 +142,12 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate) recs = PAGE_CACHE_SIZE / rec_size; /* Should have been verified before we got here... */ BUG_ON(!recs); - addr = kmap_atomic(page, KM_BIO_SRC_IRQ); + kaddr = kmap_atomic(page, KM_BIO_SRC_IRQ); for (i = 0; i < recs; i++) - post_read_mst_fixup((NTFS_RECORD*)(addr + + post_read_mst_fixup((NTFS_RECORD*)(kaddr + i * rec_size), rec_size); + kunmap_atomic(kaddr, KM_BIO_SRC_IRQ); flush_dcache_page(page); - kunmap_atomic(addr, KM_BIO_SRC_IRQ); if (likely(page_uptodate && !PageError(page))) SetPageUptodate(page); } @@ -168,8 +178,11 @@ still_busy: */ static int ntfs_read_block(struct page *page) { + loff_t i_size; VCN vcn; LCN lcn; + s64 init_size; + struct inode *vi; ntfs_inode *ni; ntfs_volume *vol; runlist_element *rl; @@ -180,7 +193,8 @@ static int ntfs_read_block(struct page *page) int i, nr; unsigned char blocksize_bits; - ni = NTFS_I(page->mapping->host); + vi = page->mapping->host; + ni = NTFS_I(vi); vol = ni->vol; /* $MFT/$DATA must have its complete runlist in memory at all times. */ @@ -199,11 +213,28 @@ static int ntfs_read_block(struct page *page) bh = head = page_buffers(page); BUG_ON(!bh); + /* + * We may be racing with truncate. To avoid some of the problems we + * now take a snapshot of the various sizes and use those for the whole + * of the function. In case of an extending truncate it just means we + * may leave some buffers unmapped which are now allocated. This is + * not a problem since these buffers will just get mapped when a write + * occurs. In case of a shrinking truncate, we will detect this later + * on due to the runlist being incomplete and if the page is being + * fully truncated, truncate will throw it away as soon as we unlock + * it so no need to worry what we do with it. + */ iblock = (s64)page->index << (PAGE_CACHE_SHIFT - blocksize_bits); read_lock_irqsave(&ni->size_lock, flags); lblock = (ni->allocated_size + blocksize - 1) >> blocksize_bits; - zblock = (ni->initialized_size + blocksize - 1) >> blocksize_bits; + init_size = ni->initialized_size; + i_size = i_size_read(vi); read_unlock_irqrestore(&ni->size_lock, flags); + if (unlikely(init_size > i_size)) { + /* Race with shrinking truncate. */ + init_size = i_size; + } + zblock = (init_size + blocksize - 1) >> blocksize_bits; /* Loop through all the buffers in the page. */ rl = NULL; @@ -366,6 +397,8 @@ handle_zblock: */ static int ntfs_readpage(struct file *file, struct page *page) { + loff_t i_size; + struct inode *vi; ntfs_inode *ni, *base_ni; u8 *kaddr; ntfs_attr_search_ctx *ctx; @@ -384,7 +417,8 @@ retry_readpage: unlock_page(page); return 0; } - ni = NTFS_I(page->mapping->host); + vi = page->mapping->host; + ni = NTFS_I(vi); /* * Only $DATA attributes can be encrypted and only unnamed $DATA * attributes can be compressed. Index root can have the flags set but @@ -458,7 +492,12 @@ retry_readpage: read_lock_irqsave(&ni->size_lock, flags); if (unlikely(attr_len > ni->initialized_size)) attr_len = ni->initialized_size; + i_size = i_size_read(vi); read_unlock_irqrestore(&ni->size_lock, flags); + if (unlikely(attr_len > i_size)) { + /* Race with shrinking truncate. */ + attr_len = i_size; + } kaddr = kmap_atomic(page, KM_USER0); /* Copy the data to the page. */ memcpy(kaddr, (u8*)ctx->attr + @@ -1383,8 +1422,8 @@ retry_writepage: unsigned int ofs = i_size & ~PAGE_CACHE_MASK; kaddr = kmap_atomic(page, KM_USER0); memset(kaddr + ofs, 0, PAGE_CACHE_SIZE - ofs); - flush_dcache_page(page); kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(page); } /* Handle mst protected attributes. */ if (NInoMstProtected(ni)) @@ -1447,34 +1486,33 @@ retry_writepage: BUG_ON(PageWriteback(page)); set_page_writeback(page); unlock_page(page); - /* - * Here, we do not need to zero the out of bounds area everytime - * because the below memcpy() already takes care of the - * mmap-at-end-of-file requirements. If the file is converted to a - * non-resident one, then the code path use is switched to the - * non-resident one where the zeroing happens on each ntfs_writepage() - * invocation. - */ attr_len = le32_to_cpu(ctx->attr->data.resident.value_length); i_size = i_size_read(vi); if (unlikely(attr_len > i_size)) { + /* Race with shrinking truncate or a failed truncate. */ attr_len = i_size; - ctx->attr->data.resident.value_length = cpu_to_le32(attr_len); + /* + * If the truncate failed, fix it up now. If a concurrent + * truncate, we do its job, so it does not have to do anything. + */ + err = ntfs_resident_attr_value_resize(ctx->mrec, ctx->attr, + attr_len); + /* Shrinking cannot fail. */ + BUG_ON(err); } kaddr = kmap_atomic(page, KM_USER0); /* Copy the data from the page to the mft record. */ memcpy((u8*)ctx->attr + le16_to_cpu(ctx->attr->data.resident.value_offset), kaddr, attr_len); - flush_dcache_mft_record_page(ctx->ntfs_ino); /* Zero out of bounds area in the page cache page. */ memset(kaddr + attr_len, 0, PAGE_CACHE_SIZE - attr_len); - flush_dcache_page(page); kunmap_atomic(kaddr, KM_USER0); - + flush_dcache_mft_record_page(ctx->ntfs_ino); + flush_dcache_page(page); + /* We are done with the page. */ end_page_writeback(page); - - /* Mark the mft record dirty, so it gets written back. */ + /* Finally, mark the mft record dirty, so it gets written back. */ mark_mft_record_dirty(ctx->ntfs_ino); ntfs_attr_put_search_ctx(ctx); unmap_mft_record(base_ni); diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index dc4bbe3acf5c..7ec045131808 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -1166,6 +1166,8 @@ err_out: * * Return 0 on success and -errno on error. In the error case, the inode will * have had make_bad_inode() executed on it. + * + * Note this cannot be called for AT_INDEX_ALLOCATION. */ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi) { @@ -1242,8 +1244,8 @@ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi) } } /* - * The encryption flag set in an index root just means to - * compress all files. + * The compressed/sparse flag set in an index root just means + * to compress all files. */ if (NInoMstProtected(ni) && ni->type != AT_INDEX_ROOT) { ntfs_error(vi->i_sb, "Found mst protected attribute " @@ -1319,8 +1321,7 @@ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi) "the mapping pairs array."); goto unm_err_out; } - if ((NInoCompressed(ni) || NInoSparse(ni)) && - ni->type != AT_INDEX_ROOT) { + if (NInoCompressed(ni) || NInoSparse(ni)) { if (a->data.non_resident.compression_unit != 4) { ntfs_error(vi->i_sb, "Found nonstandard " "compression unit (%u instead " diff --git a/fs/ntfs/malloc.h b/fs/ntfs/malloc.h index 3288bcc2c4aa..006946efca8c 100644 --- a/fs/ntfs/malloc.h +++ b/fs/ntfs/malloc.h @@ -1,7 +1,7 @@ /* * malloc.h - NTFS kernel memory handling. Part of the Linux-NTFS project. * - * Copyright (c) 2001-2004 Anton Altaparmakov + * Copyright (c) 2001-2005 Anton Altaparmakov * * This program/include file is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License as published From 939647ee308e0ad924e776657704c7bedd498664 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Sun, 18 Sep 2005 15:05:20 -0500 Subject: [PATCH 160/240] [SCSI] fix oops on usb storage device disconnect We fix the oops by enforcing the host state model. There have also been two extra states added: SHOST_CANCEL_RECOVERY and SHOST_DEL_RECOVERY so we can take the model through host removal while the recovery thread is active. Signed-off-by: James Bottomley --- drivers/scsi/hosts.c | 35 +++++++++++++++++++++++++++++++++-- drivers/scsi/scsi.c | 5 ++--- drivers/scsi/scsi_error.c | 23 ++++++++++++++++------- drivers/scsi/scsi_ioctl.c | 2 +- drivers/scsi/scsi_lib.c | 4 ++-- drivers/scsi/scsi_sysfs.c | 2 ++ drivers/scsi/sg.c | 2 +- include/scsi/scsi_host.h | 9 +++++++++ 8 files changed, 66 insertions(+), 16 deletions(-) diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index 85503fad789a..f2a72d33132c 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -98,6 +98,7 @@ int scsi_host_set_state(struct Scsi_Host *shost, enum scsi_host_state state) switch (oldstate) { case SHOST_CREATED: case SHOST_RUNNING: + case SHOST_CANCEL_RECOVERY: break; default: goto illegal; @@ -107,12 +108,31 @@ int scsi_host_set_state(struct Scsi_Host *shost, enum scsi_host_state state) case SHOST_DEL: switch (oldstate) { case SHOST_CANCEL: + case SHOST_DEL_RECOVERY: break; default: goto illegal; } break; + case SHOST_CANCEL_RECOVERY: + switch (oldstate) { + case SHOST_CANCEL: + case SHOST_RECOVERY: + break; + default: + goto illegal; + } + break; + + case SHOST_DEL_RECOVERY: + switch (oldstate) { + case SHOST_CANCEL_RECOVERY: + break; + default: + goto illegal; + } + break; } shost->shost_state = state; return 0; @@ -134,13 +154,24 @@ EXPORT_SYMBOL(scsi_host_set_state); **/ void scsi_remove_host(struct Scsi_Host *shost) { + unsigned long flags; down(&shost->scan_mutex); - scsi_host_set_state(shost, SHOST_CANCEL); + spin_lock_irqsave(shost->host_lock, flags); + if (scsi_host_set_state(shost, SHOST_CANCEL)) + if (scsi_host_set_state(shost, SHOST_CANCEL_RECOVERY)) { + spin_unlock_irqrestore(shost->host_lock, flags); + up(&shost->scan_mutex); + return; + } + spin_unlock_irqrestore(shost->host_lock, flags); up(&shost->scan_mutex); scsi_forget_host(shost); scsi_proc_host_rm(shost); - scsi_host_set_state(shost, SHOST_DEL); + spin_lock_irqsave(shost->host_lock, flags); + if (scsi_host_set_state(shost, SHOST_DEL)) + BUG_ON(scsi_host_set_state(shost, SHOST_DEL_RECOVERY)); + spin_unlock_irqrestore(shost->host_lock, flags); transport_unregister_device(&shost->shost_gendev); class_device_unregister(&shost->shost_classdev); diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index a780546eda9c..1f0ebabf6d47 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -1265,9 +1265,8 @@ int scsi_device_cancel(struct scsi_device *sdev, int recovery) list_for_each_safe(lh, lh_sf, &active_list) { scmd = list_entry(lh, struct scsi_cmnd, eh_entry); list_del_init(lh); - if (recovery) { - scsi_eh_scmd_add(scmd, SCSI_EH_CANCEL_CMD); - } else { + if (recovery && + !scsi_eh_scmd_add(scmd, SCSI_EH_CANCEL_CMD)) { scmd->result = (DID_ABORT << 16); scsi_finish_command(scmd); } diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 895c9452be4c..af589fac814d 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -68,19 +68,24 @@ int scsi_eh_scmd_add(struct scsi_cmnd *scmd, int eh_flag) { struct Scsi_Host *shost = scmd->device->host; unsigned long flags; + int ret = 0; if (shost->eh_wait == NULL) return 0; spin_lock_irqsave(shost->host_lock, flags); + if (scsi_host_set_state(shost, SHOST_RECOVERY)) + if (scsi_host_set_state(shost, SHOST_CANCEL_RECOVERY)) + goto out_unlock; + ret = 1; scmd->eh_eflags |= eh_flag; list_add_tail(&scmd->eh_entry, &shost->eh_cmd_q); - scsi_host_set_state(shost, SHOST_RECOVERY); shost->host_failed++; scsi_eh_wakeup(shost); + out_unlock: spin_unlock_irqrestore(shost->host_lock, flags); - return 1; + return ret; } /** @@ -176,8 +181,8 @@ void scsi_times_out(struct scsi_cmnd *scmd) } if (unlikely(!scsi_eh_scmd_add(scmd, SCSI_EH_CANCEL_CMD))) { - panic("Error handler thread not present at %p %p %s %d", - scmd, scmd->device->host, __FILE__, __LINE__); + scmd->result |= DID_TIME_OUT << 16; + __scsi_done(scmd); } } @@ -196,8 +201,7 @@ int scsi_block_when_processing_errors(struct scsi_device *sdev) { int online; - wait_event(sdev->host->host_wait, (sdev->host->shost_state != - SHOST_RECOVERY)); + wait_event(sdev->host->host_wait, !scsi_host_in_recovery(sdev->host)); online = scsi_device_online(sdev); @@ -1441,6 +1445,7 @@ static void scsi_eh_lock_door(struct scsi_device *sdev) static void scsi_restart_operations(struct Scsi_Host *shost) { struct scsi_device *sdev; + unsigned long flags; /* * If the door was locked, we need to insert a door lock request @@ -1460,7 +1465,11 @@ static void scsi_restart_operations(struct Scsi_Host *shost) SCSI_LOG_ERROR_RECOVERY(3, printk("%s: waking up host to restart\n", __FUNCTION__)); - scsi_host_set_state(shost, SHOST_RUNNING); + spin_lock_irqsave(shost->host_lock, flags); + if (scsi_host_set_state(shost, SHOST_RUNNING)) + if (scsi_host_set_state(shost, SHOST_CANCEL)) + BUG_ON(scsi_host_set_state(shost, SHOST_DEL)); + spin_unlock_irqrestore(shost->host_lock, flags); wake_up(&shost->host_wait); diff --git a/drivers/scsi/scsi_ioctl.c b/drivers/scsi/scsi_ioctl.c index b7fddac81347..de7f98cc38fe 100644 --- a/drivers/scsi/scsi_ioctl.c +++ b/drivers/scsi/scsi_ioctl.c @@ -458,7 +458,7 @@ int scsi_nonblockable_ioctl(struct scsi_device *sdev, int cmd, * error processing, as long as the device was opened * non-blocking */ if (filp && filp->f_flags & O_NONBLOCK) { - if (sdev->host->shost_state == SHOST_RECOVERY) + if (scsi_host_in_recovery(sdev->host)) return -ENODEV; } else if (!scsi_block_when_processing_errors(sdev)) return -ENODEV; diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index f065cbc1a6e5..dc9c772bc874 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -447,7 +447,7 @@ void scsi_device_unbusy(struct scsi_device *sdev) spin_lock_irqsave(shost->host_lock, flags); shost->host_busy--; - if (unlikely((shost->shost_state == SHOST_RECOVERY) && + if (unlikely(scsi_host_in_recovery(shost) && shost->host_failed)) scsi_eh_wakeup(shost); spin_unlock(shost->host_lock); @@ -1339,7 +1339,7 @@ static inline int scsi_host_queue_ready(struct request_queue *q, struct Scsi_Host *shost, struct scsi_device *sdev) { - if (shost->shost_state == SHOST_RECOVERY) + if (scsi_host_in_recovery(shost)) return 0; if (shost->host_busy == 0 && shost->host_blocked) { /* diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c index 1e47b7eddef4..72a6550a056c 100644 --- a/drivers/scsi/scsi_sysfs.c +++ b/drivers/scsi/scsi_sysfs.c @@ -57,6 +57,8 @@ static struct { { SHOST_CANCEL, "cancel" }, { SHOST_DEL, "deleted" }, { SHOST_RECOVERY, "recovery" }, + { SHOST_CANCEL_RECOVERY, "cancel/recovery" }, + { SHOST_DEL_RECOVERY, "deleted/recovery", }, }; const char *scsi_host_state_name(enum scsi_host_state state) { diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index 9ea4765d1d12..4d09a6e4dd2e 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -1027,7 +1027,7 @@ sg_ioctl(struct inode *inode, struct file *filp, if (sdp->detached) return -ENODEV; if (filp->f_flags & O_NONBLOCK) { - if (sdp->device->host->shost_state == SHOST_RECOVERY) + if (scsi_host_in_recovery(sdp->device->host)) return -EBUSY; } else if (!scsi_block_when_processing_errors(sdp->device)) return -EBUSY; diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h index 916144be208b..540369ff815c 100644 --- a/include/scsi/scsi_host.h +++ b/include/scsi/scsi_host.h @@ -439,6 +439,8 @@ enum scsi_host_state { SHOST_CANCEL, SHOST_DEL, SHOST_RECOVERY, + SHOST_CANCEL_RECOVERY, + SHOST_DEL_RECOVERY, }; struct Scsi_Host { @@ -621,6 +623,13 @@ static inline struct Scsi_Host *dev_to_shost(struct device *dev) return container_of(dev, struct Scsi_Host, shost_gendev); } +static inline int scsi_host_in_recovery(struct Scsi_Host *shost) +{ + return shost->shost_state == SHOST_RECOVERY || + shost->shost_state == SHOST_CANCEL_RECOVERY || + shost->shost_state == SHOST_DEL_RECOVERY; +} + extern int scsi_queue_work(struct Scsi_Host *, struct work_struct *); extern void scsi_flush_work(struct Scsi_Host *); From 3ed7a4704beb66a155acd67b78b7e9a5674d55fb Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Mon, 19 Sep 2005 09:50:04 -0500 Subject: [PATCH 161/240] [SCSI] Fix thread termination for the SCSI error handle From: Alan Stern This patch (as561) fixes the error handler's thread-exit code. The kthread_stop call won't wake the thread from a down_interruptible, so the patch gets rid of the semaphore and simply does set_current_state(TASK_INTERRUPTIBLE); Signed-off-by: Alan Stern Modified to simplify the termination loop and correct the sleep condition. Signed-off-by: James Bottomley --- drivers/scsi/scsi_error.c | 55 ++++++++++++++++----------------------- include/scsi/scsi_host.h | 2 -- 2 files changed, 23 insertions(+), 34 deletions(-) diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index af589fac814d..ad5342165079 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -50,7 +50,7 @@ void scsi_eh_wakeup(struct Scsi_Host *shost) { if (shost->host_busy == shost->host_failed) { - up(shost->eh_wait); + wake_up_process(shost->ehandler); SCSI_LOG_ERROR_RECOVERY(5, printk("Waking error handler thread\n")); } @@ -70,7 +70,7 @@ int scsi_eh_scmd_add(struct scsi_cmnd *scmd, int eh_flag) unsigned long flags; int ret = 0; - if (shost->eh_wait == NULL) + if (!shost->ehandler) return 0; spin_lock_irqsave(shost->host_lock, flags); @@ -1591,40 +1591,31 @@ int scsi_error_handler(void *data) { struct Scsi_Host *shost = (struct Scsi_Host *) data; int rtn; - DECLARE_MUTEX_LOCKED(sem); current->flags |= PF_NOFREEZE; - shost->eh_wait = &sem; + /* - * Wake up the thread that created us. + * Note - we always use TASK_INTERRUPTIBLE even if the module + * was loaded as part of the kernel. The reason is that + * UNINTERRUPTIBLE would cause this thread to be counted in + * the load average as a running process, and an interruptible + * wait doesn't. */ - SCSI_LOG_ERROR_RECOVERY(3, printk("Wake up parent of" - " scsi_eh_%d\n",shost->host_no)); - - while (1) { - /* - * If we get a signal, it means we are supposed to go - * away and die. This typically happens if the user is - * trying to unload a module. - */ - SCSI_LOG_ERROR_RECOVERY(1, printk("Error handler" - " scsi_eh_%d" - " sleeping\n",shost->host_no)); - - /* - * Note - we always use down_interruptible with the semaphore - * even if the module was loaded as part of the kernel. The - * reason is that down() will cause this thread to be counted - * in the load average as a running process, and down - * interruptible doesn't. Given that we need to allow this - * thread to die if the driver was loaded as a module, using - * semaphores isn't unreasonable. - */ - down_interruptible(&sem); - if (kthread_should_stop()) - break; + set_current_state(TASK_INTERRUPTIBLE); + while (!kthread_should_stop()) { + if (shost->host_failed == 0 || + shost->host_failed != shost->host_busy) { + SCSI_LOG_ERROR_RECOVERY(1, printk("Error handler" + " scsi_eh_%d" + " sleeping\n", + shost->host_no)); + schedule(); + set_current_state(TASK_INTERRUPTIBLE); + continue; + } + __set_current_state(TASK_RUNNING); SCSI_LOG_ERROR_RECOVERY(1, printk("Error handler" " scsi_eh_%d waking" " up\n",shost->host_no)); @@ -1651,7 +1642,7 @@ int scsi_error_handler(void *data) * which are still online. */ scsi_restart_operations(shost); - + set_current_state(TASK_INTERRUPTIBLE); } SCSI_LOG_ERROR_RECOVERY(1, printk("Error handler scsi_eh_%d" @@ -1660,7 +1651,7 @@ int scsi_error_handler(void *data) /* * Make sure that nobody tries to wake us up again. */ - shost->eh_wait = NULL; + shost->ehandler = NULL; return 0; } diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h index 540369ff815c..69313ba7505b 100644 --- a/include/scsi/scsi_host.h +++ b/include/scsi/scsi_host.h @@ -467,8 +467,6 @@ struct Scsi_Host { struct list_head eh_cmd_q; struct task_struct * ehandler; /* Error recovery thread. */ - struct semaphore * eh_wait; /* The error recovery thread waits - on this. */ struct semaphore * eh_action; /* Wait for specific actions on the host. */ unsigned int eh_active:1; /* Indicates the eh thread is awake and active if From 6577ae51cf52f5fb0e4a85e673dd7bf2d0074e3e Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Mon, 19 Sep 2005 09:17:56 -0700 Subject: [PATCH 162/240] [PATCH] IB/mthca: Don't try to set srq->last for userspace SRQs Userspace SRQs don't have a buffer allocated for them in the kernel, so it doesn't make sense to set srq->last during initialization. In fact, this can crash trying to follow a nonexistent buffer pointer. Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mthca/mthca_srq.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/mthca/mthca_srq.c b/drivers/infiniband/hw/mthca/mthca_srq.c index 45dedd2787a4..18998d48c53e 100644 --- a/drivers/infiniband/hw/mthca/mthca_srq.c +++ b/drivers/infiniband/hw/mthca/mthca_srq.c @@ -172,6 +172,8 @@ static int mthca_alloc_srq_buf(struct mthca_dev *dev, struct mthca_pd *pd, scatter->lkey = cpu_to_be32(MTHCA_INVAL_LKEY); } + srq->last = get_wqe(srq, srq->max - 1); + return 0; } @@ -263,7 +265,6 @@ int mthca_alloc_srq(struct mthca_dev *dev, struct mthca_pd *pd, srq->first_free = 0; srq->last_free = srq->max - 1; - srq->last = get_wqe(srq, srq->max - 1); return 0; From 997a51ae373df6484cdd4a5fc61a9c9bec82cd68 Mon Sep 17 00:00:00 2001 From: Ivan Kokshaysky Date: Mon, 19 Sep 2005 18:55:51 +0400 Subject: [PATCH 163/240] [PATCH] Alpha: ISA IRQs fixup for dp264 Basically, this extends original dp264 fixup to all dp264 variations. Here is one minor change: mask out bits 4-7 of a value assigned by SRM, because - newer consoles report ISA IRQs with offset 0xe0; - even if console IRQ value is bogus, we'll have a value < 16 so it should be harmless as it won't clash with native IRQs. Particularly this fixes USB interrupt problem on xp1000 and es40. Signed-off-by: Ivan Kokshaysky Acked-by: Richard Henderson Signed-off-by: Linus Torvalds --- arch/alpha/kernel/sys_dp264.c | 41 ++++++++++++++++++++--------------- 1 file changed, 23 insertions(+), 18 deletions(-) diff --git a/arch/alpha/kernel/sys_dp264.c b/arch/alpha/kernel/sys_dp264.c index 9e36b07fa940..d5da6b1b28ee 100644 --- a/arch/alpha/kernel/sys_dp264.c +++ b/arch/alpha/kernel/sys_dp264.c @@ -394,6 +394,22 @@ clipper_init_irq(void) * 10 64 bit PCI option slot 3 (not bus 0) */ +static int __init +isa_irq_fixup(struct pci_dev *dev, int irq) +{ + u8 irq8; + + if (irq > 0) + return irq; + + /* This interrupt is routed via ISA bridge, so we'll + just have to trust whatever value the console might + have assigned. */ + pci_read_config_byte(dev, PCI_INTERRUPT_LINE, &irq8); + + return irq8 & 0xf; +} + static int __init dp264_map_irq(struct pci_dev *dev, u8 slot, u8 pin) { @@ -407,25 +423,13 @@ dp264_map_irq(struct pci_dev *dev, u8 slot, u8 pin) { 16+ 3, 16+ 3, 16+ 2, 16+ 1, 16+ 0} /* IdSel 10 slot 3 */ }; const long min_idsel = 5, max_idsel = 10, irqs_per_slot = 5; - struct pci_controller *hose = dev->sysdata; int irq = COMMON_TABLE_LOOKUP; - if (irq > 0) { + if (irq > 0) irq += 16 * hose->index; - } else { - /* ??? The Contaq IDE controller on the ISA bridge uses - "legacy" interrupts 14 and 15. I don't know if anything - can wind up at the same slot+pin on hose1, so we'll - just have to trust whatever value the console might - have assigned. */ - u8 irq8; - pci_read_config_byte(dev, PCI_INTERRUPT_LINE, &irq8); - irq = irq8; - } - - return irq; + return isa_irq_fixup(dev, irq); } static int __init @@ -453,7 +457,8 @@ monet_map_irq(struct pci_dev *dev, u8 slot, u8 pin) { 24, 24, 25, 26, 27} /* IdSel 15 slot 5 PCI2*/ }; const long min_idsel = 3, max_idsel = 15, irqs_per_slot = 5; - return COMMON_TABLE_LOOKUP; + + return isa_irq_fixup(dev, COMMON_TABLE_LOOKUP); } static u8 __init @@ -507,7 +512,8 @@ webbrick_map_irq(struct pci_dev *dev, u8 slot, u8 pin) { 47, 47, 46, 45, 44}, /* IdSel 17 slot 3 */ }; const long min_idsel = 7, max_idsel = 17, irqs_per_slot = 5; - return COMMON_TABLE_LOOKUP; + + return isa_irq_fixup(dev, COMMON_TABLE_LOOKUP); } static int __init @@ -524,14 +530,13 @@ clipper_map_irq(struct pci_dev *dev, u8 slot, u8 pin) { -1, -1, -1, -1, -1} /* IdSel 7 ISA Bridge */ }; const long min_idsel = 1, max_idsel = 7, irqs_per_slot = 5; - struct pci_controller *hose = dev->sysdata; int irq = COMMON_TABLE_LOOKUP; if (irq > 0) irq += 16 * hose->index; - return irq; + return isa_irq_fixup(dev, irq); } static void __init From 82ffb67164064752a56669511545316075b41e1d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 9 Sep 2005 16:25:54 +0200 Subject: [PATCH 164/240] [SCSI] fusion core changes for SAS support - various bits for SAS support from the LSI driver. - use the device private data for the fusion target private data. this should be using the midlayer target data framework, but we can't move over to that until fusion has been switched to the generic DV code - use target ID and channel from the fusion target private data, because those in scsi_device will be different for mptsas Signed-off-by: Christoph Hellwig Signed-off-by: James Bottomley --- drivers/message/fusion/mptbase.c | 514 +++++++++++++++++++++++++----- drivers/message/fusion/mptbase.h | 28 +- drivers/message/fusion/mptscsih.c | 15 +- 3 files changed, 471 insertions(+), 86 deletions(-) diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c index f517d0692d5f..14d62d96ca41 100644 --- a/drivers/message/fusion/mptbase.c +++ b/drivers/message/fusion/mptbase.c @@ -141,7 +141,7 @@ static int GetPortFacts(MPT_ADAPTER *ioc, int portnum, int sleepFlag); static int SendIocInit(MPT_ADAPTER *ioc, int sleepFlag); static int SendPortEnable(MPT_ADAPTER *ioc, int portnum, int sleepFlag); static int mpt_do_upload(MPT_ADAPTER *ioc, int sleepFlag); -static int mpt_downloadboot(MPT_ADAPTER *ioc, int sleepFlag); +static int mpt_downloadboot(MPT_ADAPTER *ioc, MpiFwHeader_t *pFwHeader, int sleepFlag); static int mpt_diag_reset(MPT_ADAPTER *ioc, int ignore, int sleepFlag); static int KickStart(MPT_ADAPTER *ioc, int ignore, int sleepFlag); static int SendIocReset(MPT_ADAPTER *ioc, u8 reset_type, int sleepFlag); @@ -152,6 +152,7 @@ static int WaitForDoorbellReply(MPT_ADAPTER *ioc, int howlong, int sleepFlag); static int GetLanConfigPages(MPT_ADAPTER *ioc); static int GetFcPortPage0(MPT_ADAPTER *ioc, int portnum); static int GetIoUnitPage2(MPT_ADAPTER *ioc); +int mptbase_sas_persist_operation(MPT_ADAPTER *ioc, u8 persist_opcode); static int mpt_GetScsiPortSettings(MPT_ADAPTER *ioc, int portnum); static int mpt_readScsiDevicePageHeaders(MPT_ADAPTER *ioc, int portnum); static void mpt_read_ioc_pg_1(MPT_ADAPTER *ioc); @@ -159,6 +160,8 @@ static void mpt_read_ioc_pg_4(MPT_ADAPTER *ioc); static void mpt_timer_expired(unsigned long data); static int SendEventNotification(MPT_ADAPTER *ioc, u8 EvSwitch); static int SendEventAck(MPT_ADAPTER *ioc, EventNotificationReply_t *evnp); +static int mpt_host_page_access_control(MPT_ADAPTER *ioc, u8 access_control_value, int sleepFlag); +static int mpt_host_page_alloc(MPT_ADAPTER *ioc, pIOCInit_t ioc_init); #ifdef CONFIG_PROC_FS static int procmpt_summary_read(char *buf, char **start, off_t offset, @@ -509,6 +512,14 @@ mpt_base_reply(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR *reply) pCfg->wait_done = 1; wake_up(&mpt_waitq); } + } else if (func == MPI_FUNCTION_SAS_IO_UNIT_CONTROL) { + /* we should be always getting a reply frame */ + memcpy(ioc->persist_reply_frame, reply, + min(MPT_DEFAULT_FRAME_SIZE, + 4*reply->u.reply.MsgLength)); + del_timer(&ioc->persist_timer); + ioc->persist_wait_done = 1; + wake_up(&mpt_waitq); } else { printk(MYIOC_s_ERR_FMT "Unexpected msg function (=%02Xh) reply received!\n", ioc->name, func); @@ -750,6 +761,7 @@ mpt_get_msg_frame(int handle, MPT_ADAPTER *ioc) mf = list_entry(ioc->FreeQ.next, MPT_FRAME_HDR, u.frame.linkage.list); list_del(&mf->u.frame.linkage.list); + mf->u.frame.linkage.arg1 = 0; mf->u.frame.hwhdr.msgctxu.fld.cb_idx = handle; /* byte */ req_offset = (u8 *)mf - (u8 *)ioc->req_frames; /* u16! */ @@ -845,6 +857,7 @@ mpt_free_msg_frame(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf) /* Put Request back on FreeQ! */ spin_lock_irqsave(&ioc->FreeQlock, flags); + mf->u.frame.linkage.arg1 = 0xdeadbeaf; /* signature to know if this mf is freed */ list_add_tail(&mf->u.frame.linkage.list, &ioc->FreeQ); #ifdef MFCNT ioc->mfcnt--; @@ -971,10 +984,121 @@ mpt_send_handshake_request(int handle, MPT_ADAPTER *ioc, int reqBytes, u32 *req, /* Make sure there are no doorbells */ CHIPREG_WRITE32(&ioc->chip->IntStatus, 0); - + return r; } +/*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ +/** + * mpt_host_page_access_control - provides mechanism for the host + * driver to control the IOC's Host Page Buffer access. + * @ioc: Pointer to MPT adapter structure + * @access_control_value: define bits below + * + * Access Control Value - bits[15:12] + * 0h Reserved + * 1h Enable Access { MPI_DB_HPBAC_ENABLE_ACCESS } + * 2h Disable Access { MPI_DB_HPBAC_DISABLE_ACCESS } + * 3h Free Buffer { MPI_DB_HPBAC_FREE_BUFFER } + * + * Returns 0 for success, non-zero for failure. + */ + +static int +mpt_host_page_access_control(MPT_ADAPTER *ioc, u8 access_control_value, int sleepFlag) +{ + int r = 0; + + /* return if in use */ + if (CHIPREG_READ32(&ioc->chip->Doorbell) + & MPI_DOORBELL_ACTIVE) + return -1; + + CHIPREG_WRITE32(&ioc->chip->IntStatus, 0); + + CHIPREG_WRITE32(&ioc->chip->Doorbell, + ((MPI_FUNCTION_HOST_PAGEBUF_ACCESS_CONTROL + <HostPageBuffer) { + + host_page_buffer_sz = + le32_to_cpu(ioc->facts.HostPageBufferSGE.FlagsLength) & 0xFFFFFF; + + if(!host_page_buffer_sz) + return 0; /* fw doesn't need any host buffers */ + + /* spin till we get enough memory */ + while(host_page_buffer_sz > 0) { + + if((ioc->HostPageBuffer = pci_alloc_consistent( + ioc->pcidev, + host_page_buffer_sz, + &ioc->HostPageBuffer_dma)) != NULL) { + + dinitprintk((MYIOC_s_INFO_FMT + "host_page_buffer @ %p, dma @ %x, sz=%d bytes\n", + ioc->name, + ioc->HostPageBuffer, + ioc->HostPageBuffer_dma, + hst_page_buffer_sz)); + ioc->alloc_total += host_page_buffer_sz; + ioc->HostPageBuffer_sz = host_page_buffer_sz; + break; + } + + host_page_buffer_sz -= (4*1024); + } + } + + if(!ioc->HostPageBuffer) { + printk(MYIOC_s_ERR_FMT + "Failed to alloc memory for host_page_buffer!\n", + ioc->name); + return -999; + } + + psge = (char *)&ioc_init->HostPageBufferSGE; + flags_length = MPI_SGE_FLAGS_SIMPLE_ELEMENT | + MPI_SGE_FLAGS_SYSTEM_ADDRESS | + MPI_SGE_FLAGS_32_BIT_ADDRESSING | + MPI_SGE_FLAGS_HOST_TO_IOC | + MPI_SGE_FLAGS_END_OF_BUFFER; + if (sizeof(dma_addr_t) == sizeof(u64)) { + flags_length |= MPI_SGE_FLAGS_64_BIT_ADDRESSING; + } + flags_length = flags_length << MPI_SGE_FLAGS_SHIFT; + flags_length |= ioc->HostPageBuffer_sz; + mpt_add_sge(psge, flags_length, ioc->HostPageBuffer_dma); + ioc->facts.HostPageBufferSGE = ioc_init->HostPageBufferSGE; + +return 0; +} + /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** * mpt_verify_adapter - Given a unique IOC identifier, set pointer to @@ -1213,6 +1337,33 @@ mpt_attach(struct pci_dev *pdev, const struct pci_device_id *id) ioc->prod_name = "LSI53C1035"; ioc->bus_type = SCSI; } + else if (pdev->device == MPI_MANUFACTPAGE_DEVID_SAS1064) { + ioc->prod_name = "LSISAS1064"; + ioc->bus_type = SAS; + ioc->errata_flag_1064 = 1; + } + else if (pdev->device == MPI_MANUFACTPAGE_DEVID_SAS1066) { + ioc->prod_name = "LSISAS1066"; + ioc->bus_type = SAS; + ioc->errata_flag_1064 = 1; + } + else if (pdev->device == MPI_MANUFACTPAGE_DEVID_SAS1068) { + ioc->prod_name = "LSISAS1068"; + ioc->bus_type = SAS; + ioc->errata_flag_1064 = 1; + } + else if (pdev->device == MPI_MANUFACTPAGE_DEVID_SAS1064E) { + ioc->prod_name = "LSISAS1064E"; + ioc->bus_type = SAS; + } + else if (pdev->device == MPI_MANUFACTPAGE_DEVID_SAS1066E) { + ioc->prod_name = "LSISAS1066E"; + ioc->bus_type = SAS; + } + else if (pdev->device == MPI_MANUFACTPAGE_DEVID_SAS1068E) { + ioc->prod_name = "LSISAS1068E"; + ioc->bus_type = SAS; + } if (ioc->errata_flag_1064) pci_disable_io_access(pdev); @@ -1640,7 +1791,22 @@ mpt_do_ioc_recovery(MPT_ADAPTER *ioc, u32 reason, int sleepFlag) * and we try GetLanConfigPages again... */ if ((ret == 0) && (reason == MPT_HOSTEVENT_IOC_BRINGUP)) { - if (ioc->bus_type == FC) { + if (ioc->bus_type == SAS) { + + /* clear persistency table */ + if(ioc->facts.IOCExceptions & + MPI_IOCFACTS_EXCEPT_PERSISTENT_TABLE_FULL) { + ret = mptbase_sas_persist_operation(ioc, + MPI_SAS_OP_CLEAR_NOT_PRESENT); + if(ret != 0) + return -1; + } + + /* Find IM volumes + */ + mpt_findImVolumes(ioc); + + } else if (ioc->bus_type == FC) { /* * Pre-fetch FC port WWN and stuff... * (FCPortPage0_t stuff) @@ -1783,7 +1949,7 @@ mpt_adapter_disable(MPT_ADAPTER *ioc) if (ioc->cached_fw != NULL) { ddlprintk((KERN_INFO MYNAM ": mpt_adapter_disable: Pushing FW onto adapter\n")); - if ((ret = mpt_downloadboot(ioc, NO_SLEEP)) < 0) { + if ((ret = mpt_downloadboot(ioc, (MpiFwHeader_t *)ioc->cached_fw, NO_SLEEP)) < 0) { printk(KERN_WARNING MYNAM ": firmware downloadboot failure (%d)!\n", ret); } @@ -1852,6 +2018,23 @@ mpt_adapter_disable(MPT_ADAPTER *ioc) kfree(ioc->ChainToChain); ioc->ChainToChain = NULL; + + if (ioc->HostPageBuffer != NULL) { + if((ret = mpt_host_page_access_control(ioc, + MPI_DB_HPBAC_FREE_BUFFER, NO_SLEEP)) != 0) { + printk(KERN_ERR MYNAM + ": %s: host page buffers free failed (%d)!\n", + __FUNCTION__, ret); + } + dexitprintk((KERN_INFO MYNAM ": %s HostPageBuffer free @ %p, sz=%d bytes\n", + ioc->name, ioc->HostPageBuffer, ioc->HostPageBuffer_sz)); + pci_free_consistent(ioc->pcidev, ioc->HostPageBuffer_sz, + ioc->HostPageBuffer, + ioc->HostPageBuffer_dma); + ioc->HostPageBuffer = NULL; + ioc->HostPageBuffer_sz = 0; + ioc->alloc_total -= ioc->HostPageBuffer_sz; + } } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ @@ -2034,7 +2217,7 @@ MakeIocReady(MPT_ADAPTER *ioc, int force, int sleepFlag) * Loop here waiting for IOC to come READY. */ ii = 0; - cntdn = ((sleepFlag == CAN_SLEEP) ? HZ : 1000) * 15; /* 15 seconds */ + cntdn = ((sleepFlag == CAN_SLEEP) ? HZ : 1000) * 5; /* 5 seconds */ while ((ioc_state = mpt_GetIocState(ioc, 1)) != MPI_IOC_STATE_READY) { if (ioc_state == MPI_IOC_STATE_OPERATIONAL) { @@ -2212,6 +2395,7 @@ GetIocFacts(MPT_ADAPTER *ioc, int sleepFlag, int reason) le32_to_cpu(facts->CurrentSenseBufferHighAddr); facts->CurReplyFrameSize = le16_to_cpu(facts->CurReplyFrameSize); + facts->IOCCapabilities = le32_to_cpu(facts->IOCCapabilities); /* * Handle NEW (!) IOCFactsReply fields in MPI-1.01.xx @@ -2383,13 +2567,25 @@ SendIocInit(MPT_ADAPTER *ioc, int sleepFlag) ddlprintk((MYIOC_s_INFO_FMT "upload_fw %d facts.Flags=%x\n", ioc->name, ioc->upload_fw, ioc->facts.Flags)); - if (ioc->bus_type == FC) + if(ioc->bus_type == SAS) + ioc_init.MaxDevices = ioc->facts.MaxDevices; + else if(ioc->bus_type == FC) ioc_init.MaxDevices = MPT_MAX_FC_DEVICES; else ioc_init.MaxDevices = MPT_MAX_SCSI_DEVICES; - ioc_init.MaxBuses = MPT_MAX_BUS; + dinitprintk((MYIOC_s_INFO_FMT "facts.MsgVersion=%x\n", + ioc->name, ioc->facts.MsgVersion)); + if (ioc->facts.MsgVersion >= MPI_VERSION_01_05) { + // set MsgVersion and HeaderVersion host driver was built with + ioc_init.MsgVersion = cpu_to_le16(MPI_VERSION); + ioc_init.HeaderVersion = cpu_to_le16(MPI_HEADER_VERSION); + if (ioc->facts.Flags & MPI_IOCFACTS_FLAGS_HOST_PAGE_BUFFER_PERSISTENT) { + ioc_init.HostPageBufferSGE = ioc->facts.HostPageBufferSGE; + } else if(mpt_host_page_alloc(ioc, &ioc_init)) + return -99; + } ioc_init.ReplyFrameSize = cpu_to_le16(ioc->reply_sz); /* in BYTES */ if (sizeof(dma_addr_t) == sizeof(u64)) { @@ -2403,17 +2599,21 @@ SendIocInit(MPT_ADAPTER *ioc, int sleepFlag) ioc_init.HostMfaHighAddr = cpu_to_le32(0); ioc_init.SenseBufferHighAddr = cpu_to_le32(0); } - + ioc->facts.CurrentHostMfaHighAddr = ioc_init.HostMfaHighAddr; ioc->facts.CurrentSenseBufferHighAddr = ioc_init.SenseBufferHighAddr; + ioc->facts.MaxDevices = ioc_init.MaxDevices; + ioc->facts.MaxBuses = ioc_init.MaxBuses; dhsprintk((MYIOC_s_INFO_FMT "Sending IOCInit (req @ %p)\n", ioc->name, &ioc_init)); r = mpt_handshake_req_reply_wait(ioc, sizeof(IOCInit_t), (u32*)&ioc_init, sizeof(MPIDefaultReply_t), (u16*)&init_reply, 10 /*seconds*/, sleepFlag); - if (r != 0) + if (r != 0) { + printk(MYIOC_s_ERR_FMT "Sending IOCInit failed(%d)!\n",ioc->name, r); return r; + } /* No need to byte swap the multibyte fields in the reply * since we don't even look at it's contents. @@ -2472,7 +2672,7 @@ SendPortEnable(MPT_ADAPTER *ioc, int portnum, int sleepFlag) { PortEnable_t port_enable; MPIDefaultReply_t reply_buf; - int ii; + int rc; int req_sz; int reply_sz; @@ -2494,22 +2694,15 @@ SendPortEnable(MPT_ADAPTER *ioc, int portnum, int sleepFlag) /* RAID FW may take a long time to enable */ - if (ioc->bus_type == FC) { - ii = mpt_handshake_req_reply_wait(ioc, req_sz, (u32*)&port_enable, - reply_sz, (u16*)&reply_buf, 65 /*seconds*/, sleepFlag); - } else { - ii = mpt_handshake_req_reply_wait(ioc, req_sz, (u32*)&port_enable, + if ( (ioc->facts.ProductID & MPI_FW_HEADER_PID_PROD_MASK) + > MPI_FW_HEADER_PID_PROD_TARGET_SCSI ) { + rc = mpt_handshake_req_reply_wait(ioc, req_sz, (u32*)&port_enable, reply_sz, (u16*)&reply_buf, 300 /*seconds*/, sleepFlag); + } else { + rc = mpt_handshake_req_reply_wait(ioc, req_sz, (u32*)&port_enable, + reply_sz, (u16*)&reply_buf, 30 /*seconds*/, sleepFlag); } - - if (ii != 0) - return ii; - - /* We do not even look at the reply, so we need not - * swap the multi-byte fields. - */ - - return 0; + return rc; } /* @@ -2666,9 +2859,8 @@ mpt_do_upload(MPT_ADAPTER *ioc, int sleepFlag) * <0 for fw upload failure. */ static int -mpt_downloadboot(MPT_ADAPTER *ioc, int sleepFlag) +mpt_downloadboot(MPT_ADAPTER *ioc, MpiFwHeader_t *pFwHeader, int sleepFlag) { - MpiFwHeader_t *pFwHeader; MpiExtImageHeader_t *pExtImage; u32 fwSize; u32 diag0val; @@ -2679,18 +2871,8 @@ mpt_downloadboot(MPT_ADAPTER *ioc, int sleepFlag) u32 load_addr; u32 ioc_state=0; - ddlprintk((MYIOC_s_INFO_FMT "downloadboot: fw size 0x%x, ioc FW Ptr %p\n", - ioc->name, ioc->facts.FWImageSize, ioc->cached_fw)); - - if ( ioc->facts.FWImageSize == 0 ) - return -1; - - if (ioc->cached_fw == NULL) - return -2; - - /* prevent a second downloadboot and memory free with alt_ioc */ - if (ioc->alt_ioc && ioc->alt_ioc->cached_fw) - ioc->alt_ioc->cached_fw = NULL; + ddlprintk((MYIOC_s_INFO_FMT "downloadboot: fw size 0x%x (%d), FW Ptr %p\n", + ioc->name, pFwHeader->ImageSize, pFwHeader->ImageSize, pFwHeader)); CHIPREG_WRITE32(&ioc->chip->WriteSequence, 0xFF); CHIPREG_WRITE32(&ioc->chip->WriteSequence, MPI_WRSEQ_1ST_KEY_VALUE); @@ -2718,16 +2900,17 @@ mpt_downloadboot(MPT_ADAPTER *ioc, int sleepFlag) ioc->name, count)); break; } - /* wait 1 sec */ + /* wait .1 sec */ if (sleepFlag == CAN_SLEEP) { - msleep_interruptible (1000); + msleep_interruptible (100); } else { - mdelay (1000); + mdelay (100); } } if ( count == 30 ) { - ddlprintk((MYIOC_s_INFO_FMT "downloadboot failed! Unable to RESET_ADAPTER diag0val=%x\n", + ddlprintk((MYIOC_s_INFO_FMT "downloadboot failed! " + "Unable to get MPI_DIAG_DRWE mode, diag0val=%x\n", ioc->name, diag0val)); return -3; } @@ -2742,7 +2925,6 @@ mpt_downloadboot(MPT_ADAPTER *ioc, int sleepFlag) /* Set the DiagRwEn and Disable ARM bits */ CHIPREG_WRITE32(&ioc->chip->Diagnostic, (MPI_DIAG_RW_ENABLE | MPI_DIAG_DISABLE_ARM)); - pFwHeader = (MpiFwHeader_t *) ioc->cached_fw; fwSize = (pFwHeader->ImageSize + 3)/4; ptrFw = (u32 *) pFwHeader; @@ -2792,19 +2974,38 @@ mpt_downloadboot(MPT_ADAPTER *ioc, int sleepFlag) /* Clear the internal flash bad bit - autoincrementing register, * so must do two writes. */ - CHIPREG_PIO_WRITE32(&ioc->pio_chip->DiagRwAddress, 0x3F000000); - diagRwData = CHIPREG_PIO_READ32(&ioc->pio_chip->DiagRwData); - diagRwData |= 0x4000000; - CHIPREG_PIO_WRITE32(&ioc->pio_chip->DiagRwAddress, 0x3F000000); - CHIPREG_PIO_WRITE32(&ioc->pio_chip->DiagRwData, diagRwData); + if (ioc->bus_type == SCSI) { + /* + * 1030 and 1035 H/W errata, workaround to access + * the ClearFlashBadSignatureBit + */ + CHIPREG_PIO_WRITE32(&ioc->pio_chip->DiagRwAddress, 0x3F000000); + diagRwData = CHIPREG_PIO_READ32(&ioc->pio_chip->DiagRwData); + diagRwData |= 0x40000000; + CHIPREG_PIO_WRITE32(&ioc->pio_chip->DiagRwAddress, 0x3F000000); + CHIPREG_PIO_WRITE32(&ioc->pio_chip->DiagRwData, diagRwData); + + } else /* if((ioc->bus_type == SAS) || (ioc->bus_type == FC)) */ { + diag0val = CHIPREG_READ32(&ioc->chip->Diagnostic); + CHIPREG_WRITE32(&ioc->chip->Diagnostic, diag0val | + MPI_DIAG_CLEAR_FLASH_BAD_SIG); + + /* wait 1 msec */ + if (sleepFlag == CAN_SLEEP) { + msleep_interruptible (1); + } else { + mdelay (1); + } + } if (ioc->errata_flag_1064) pci_disable_io_access(ioc->pcidev); diag0val = CHIPREG_READ32(&ioc->chip->Diagnostic); - ddlprintk((MYIOC_s_INFO_FMT "downloadboot diag0val=%x, turning off PREVENT_IOC_BOOT, DISABLE_ARM\n", + ddlprintk((MYIOC_s_INFO_FMT "downloadboot diag0val=%x, " + "turning off PREVENT_IOC_BOOT, DISABLE_ARM, RW_ENABLE\n", ioc->name, diag0val)); - diag0val &= ~(MPI_DIAG_PREVENT_IOC_BOOT | MPI_DIAG_DISABLE_ARM); + diag0val &= ~(MPI_DIAG_PREVENT_IOC_BOOT | MPI_DIAG_DISABLE_ARM | MPI_DIAG_RW_ENABLE); ddlprintk((MYIOC_s_INFO_FMT "downloadboot now diag0val=%x\n", ioc->name, diag0val)); CHIPREG_WRITE32(&ioc->chip->Diagnostic, diag0val); @@ -2812,10 +3013,23 @@ mpt_downloadboot(MPT_ADAPTER *ioc, int sleepFlag) /* Write 0xFF to reset the sequencer */ CHIPREG_WRITE32(&ioc->chip->WriteSequence, 0xFF); + if (ioc->bus_type == SAS) { + ioc_state = mpt_GetIocState(ioc, 0); + if ( (GetIocFacts(ioc, sleepFlag, + MPT_HOSTEVENT_IOC_BRINGUP)) != 0 ) { + ddlprintk((MYIOC_s_INFO_FMT "GetIocFacts failed: IocState=%x\n", + ioc->name, ioc_state)); + return -EFAULT; + } + } + for (count=0; countname, count, ioc_state)); + if (ioc->bus_type == SAS) { + return 0; + } if ((SendIocInit(ioc, sleepFlag)) != 0) { ddlprintk((MYIOC_s_INFO_FMT "downloadboot: SendIocInit failed\n", ioc->name)); @@ -3049,12 +3263,13 @@ mpt_diag_reset(MPT_ADAPTER *ioc, int ignore, int sleepFlag) /* wait 1 sec */ if (sleepFlag == CAN_SLEEP) { - ssleep(1); + msleep_interruptible (1000); } else { mdelay (1000); } } - if ((count = mpt_downloadboot(ioc, sleepFlag)) < 0) { + if ((count = mpt_downloadboot(ioc, + (MpiFwHeader_t *)ioc->cached_fw, sleepFlag)) < 0) { printk(KERN_WARNING MYNAM ": firmware downloadboot failure (%d)!\n", count); } @@ -3999,6 +4214,85 @@ GetFcPortPage0(MPT_ADAPTER *ioc, int portnum) return rc; } +/*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ +/* + * mptbase_sas_persist_operation - Perform operation on SAS Persitent Table + * @ioc: Pointer to MPT_ADAPTER structure + * @sas_address: 64bit SAS Address for operation. + * @target_id: specified target for operation + * @bus: specified bus for operation + * @persist_opcode: see below + * + * MPI_SAS_OP_CLEAR_NOT_PRESENT - Free all persist TargetID mappings for + * devices not currently present. + * MPI_SAS_OP_CLEAR_ALL_PERSISTENT - Clear al persist TargetID mappings + * + * NOTE: Don't use not this function during interrupt time. + * + * Returns: 0 for success, non-zero error + */ + +/*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ +int +mptbase_sas_persist_operation(MPT_ADAPTER *ioc, u8 persist_opcode) +{ + SasIoUnitControlRequest_t *sasIoUnitCntrReq; + SasIoUnitControlReply_t *sasIoUnitCntrReply; + MPT_FRAME_HDR *mf = NULL; + MPIHeader_t *mpi_hdr; + + + /* insure garbage is not sent to fw */ + switch(persist_opcode) { + + case MPI_SAS_OP_CLEAR_NOT_PRESENT: + case MPI_SAS_OP_CLEAR_ALL_PERSISTENT: + break; + + default: + return -1; + break; + } + + printk("%s: persist_opcode=%x\n",__FUNCTION__, persist_opcode); + + /* Get a MF for this command. + */ + if ((mf = mpt_get_msg_frame(mpt_base_index, ioc)) == NULL) { + printk("%s: no msg frames!\n",__FUNCTION__); + return -1; + } + + mpi_hdr = (MPIHeader_t *) mf; + sasIoUnitCntrReq = (SasIoUnitControlRequest_t *)mf; + memset(sasIoUnitCntrReq,0,sizeof(SasIoUnitControlRequest_t)); + sasIoUnitCntrReq->Function = MPI_FUNCTION_SAS_IO_UNIT_CONTROL; + sasIoUnitCntrReq->MsgContext = mpi_hdr->MsgContext; + sasIoUnitCntrReq->Operation = persist_opcode; + + init_timer(&ioc->persist_timer); + ioc->persist_timer.data = (unsigned long) ioc; + ioc->persist_timer.function = mpt_timer_expired; + ioc->persist_timer.expires = jiffies + HZ*10 /* 10 sec */; + ioc->persist_wait_done=0; + add_timer(&ioc->persist_timer); + mpt_put_msg_frame(mpt_base_index, ioc, mf); + wait_event(mpt_waitq, ioc->persist_wait_done); + + sasIoUnitCntrReply = + (SasIoUnitControlReply_t *)ioc->persist_reply_frame; + if (le16_to_cpu(sasIoUnitCntrReply->IOCStatus) != MPI_IOCSTATUS_SUCCESS) { + printk("%s: IOCStatus=0x%X IOCLogInfo=0x%X\n", + __FUNCTION__, + sasIoUnitCntrReply->IOCStatus, + sasIoUnitCntrReply->IOCLogInfo); + return -1; + } + + printk("%s: success\n",__FUNCTION__); + return 0; +} + /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /* * GetIoUnitPage2 - Retrieve BIOS version and boot order information. @@ -5366,8 +5660,8 @@ mpt_HardResetHandler(MPT_ADAPTER *ioc, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -static char * -EventDescriptionStr(u8 event, u32 evData0) +static void +EventDescriptionStr(u8 event, u32 evData0, char *evStr) { char *ds; @@ -5420,8 +5714,95 @@ EventDescriptionStr(u8 event, u32 evData0) ds = "Events(OFF) Change"; break; case MPI_EVENT_INTEGRATED_RAID: - ds = "Integrated Raid"; + { + u8 ReasonCode = (u8)(evData0 >> 16); + switch (ReasonCode) { + case MPI_EVENT_RAID_RC_VOLUME_CREATED : + ds = "Integrated Raid: Volume Created"; + break; + case MPI_EVENT_RAID_RC_VOLUME_DELETED : + ds = "Integrated Raid: Volume Deleted"; + break; + case MPI_EVENT_RAID_RC_VOLUME_SETTINGS_CHANGED : + ds = "Integrated Raid: Volume Settings Changed"; + break; + case MPI_EVENT_RAID_RC_VOLUME_STATUS_CHANGED : + ds = "Integrated Raid: Volume Status Changed"; + break; + case MPI_EVENT_RAID_RC_VOLUME_PHYSDISK_CHANGED : + ds = "Integrated Raid: Volume Physdisk Changed"; + break; + case MPI_EVENT_RAID_RC_PHYSDISK_CREATED : + ds = "Integrated Raid: Physdisk Created"; + break; + case MPI_EVENT_RAID_RC_PHYSDISK_DELETED : + ds = "Integrated Raid: Physdisk Deleted"; + break; + case MPI_EVENT_RAID_RC_PHYSDISK_SETTINGS_CHANGED : + ds = "Integrated Raid: Physdisk Settings Changed"; + break; + case MPI_EVENT_RAID_RC_PHYSDISK_STATUS_CHANGED : + ds = "Integrated Raid: Physdisk Status Changed"; + break; + case MPI_EVENT_RAID_RC_DOMAIN_VAL_NEEDED : + ds = "Integrated Raid: Domain Validation Needed"; + break; + case MPI_EVENT_RAID_RC_SMART_DATA : + ds = "Integrated Raid; Smart Data"; + break; + case MPI_EVENT_RAID_RC_REPLACE_ACTION_STARTED : + ds = "Integrated Raid: Replace Action Started"; + break; + default: + ds = "Integrated Raid"; break; + } + break; + } + case MPI_EVENT_SCSI_DEVICE_STATUS_CHANGE: + ds = "SCSI Device Status Change"; + break; + case MPI_EVENT_SAS_DEVICE_STATUS_CHANGE: + { + u8 ReasonCode = (u8)(evData0 >> 16); + switch (ReasonCode) { + case MPI_EVENT_SAS_DEV_STAT_RC_ADDED: + ds = "SAS Device Status Change: Added"; + break; + case MPI_EVENT_SAS_DEV_STAT_RC_NOT_RESPONDING: + ds = "SAS Device Status Change: Deleted"; + break; + case MPI_EVENT_SAS_DEV_STAT_RC_SMART_DATA: + ds = "SAS Device Status Change: SMART Data"; + break; + case MPI_EVENT_SAS_DEV_STAT_RC_NO_PERSIST_ADDED: + ds = "SAS Device Status Change: No Persistancy Added"; + break; + default: + ds = "SAS Device Status Change: Unknown"; + break; + } + break; + } + case MPI_EVENT_ON_BUS_TIMER_EXPIRED: + ds = "Bus Timer Expired"; + break; + case MPI_EVENT_QUEUE_FULL: + ds = "Queue Full"; + break; + case MPI_EVENT_SAS_SES: + ds = "SAS SES Event"; + break; + case MPI_EVENT_PERSISTENT_TABLE_FULL: + ds = "Persistent Table Full"; + break; + case MPI_EVENT_SAS_PHY_LINK_STATUS: + ds = "SAS PHY Link Status"; + break; + case MPI_EVENT_SAS_DISCOVERY_ERROR: + ds = "SAS Discovery Error"; + break; + /* * MPT base "custom" events may be added here... */ @@ -5429,7 +5810,7 @@ EventDescriptionStr(u8 event, u32 evData0) ds = "Unknown"; break; } - return ds; + strcpy(evStr,ds); } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ @@ -5451,7 +5832,7 @@ ProcessEventNotification(MPT_ADAPTER *ioc, EventNotificationReply_t *pEventReply int ii; int r = 0; int handlers = 0; - char *evStr; + char evStr[100]; u8 event; /* @@ -5464,7 +5845,7 @@ ProcessEventNotification(MPT_ADAPTER *ioc, EventNotificationReply_t *pEventReply evData0 = le32_to_cpu(pEventReply->Data[0]); } - evStr = EventDescriptionStr(event, evData0); + EventDescriptionStr(event, evData0, evStr); devtprintk((MYIOC_s_INFO_FMT "MPT event (%s=%02Xh) detected!\n", ioc->name, evStr, @@ -5481,20 +5862,6 @@ ProcessEventNotification(MPT_ADAPTER *ioc, EventNotificationReply_t *pEventReply * Do general / base driver event processing */ switch(event) { - case MPI_EVENT_NONE: /* 00 */ - case MPI_EVENT_LOG_DATA: /* 01 */ - case MPI_EVENT_STATE_CHANGE: /* 02 */ - case MPI_EVENT_UNIT_ATTENTION: /* 03 */ - case MPI_EVENT_IOC_BUS_RESET: /* 04 */ - case MPI_EVENT_EXT_BUS_RESET: /* 05 */ - case MPI_EVENT_RESCAN: /* 06 */ - case MPI_EVENT_LINK_STATUS_CHANGE: /* 07 */ - case MPI_EVENT_LOOP_STATE_CHANGE: /* 08 */ - case MPI_EVENT_LOGOUT: /* 09 */ - case MPI_EVENT_INTEGRATED_RAID: /* 0B */ - case MPI_EVENT_SCSI_DEVICE_STATUS_CHANGE: /* 0C */ - default: - break; case MPI_EVENT_EVENT_CHANGE: /* 0A */ if (evDataLen) { u8 evState = evData0 & 0xFF; @@ -5507,6 +5874,8 @@ ProcessEventNotification(MPT_ADAPTER *ioc, EventNotificationReply_t *pEventReply } } break; + default: + break; } /* @@ -5814,6 +6183,7 @@ EXPORT_SYMBOL(mpt_findImVolumes); EXPORT_SYMBOL(mpt_read_ioc_pg_3); EXPORT_SYMBOL(mpt_alloc_fw_memory); EXPORT_SYMBOL(mpt_free_fw_memory); +EXPORT_SYMBOL(mptbase_sas_persist_operation); /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ diff --git a/drivers/message/fusion/mptbase.h b/drivers/message/fusion/mptbase.h index f4827d923731..bbd21d74ce5c 100644 --- a/drivers/message/fusion/mptbase.h +++ b/drivers/message/fusion/mptbase.h @@ -65,6 +65,7 @@ #include "lsi/mpi_fc.h" /* Fibre Channel (lowlevel) support */ #include "lsi/mpi_targ.h" /* SCSI/FCP Target protcol support */ #include "lsi/mpi_tool.h" /* Tools support */ +#include "lsi/mpi_sas.h" /* SAS support */ /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ @@ -477,6 +478,14 @@ typedef struct _ScsiCfgData { u8 rsvd[1]; } ScsiCfgData; +typedef struct _SasCfgData { + u8 ptClear; /* 1 to automatically clear the + * persistent table. + * 0 to disable + * automatic clearing. + */ +}SasCfgData; + /* * Adapter Structure - pci_dev specific. Maximum: MPT_MAX_ADAPTERS */ @@ -530,11 +539,15 @@ typedef struct _MPT_ADAPTER u8 *sense_buf_pool; dma_addr_t sense_buf_pool_dma; u32 sense_buf_low_dma; + u8 *HostPageBuffer; /* SAS - host page buffer support */ + u32 HostPageBuffer_sz; + dma_addr_t HostPageBuffer_dma; int mtrr_reg; struct pci_dev *pcidev; /* struct pci_dev pointer */ u8 __iomem *memmap; /* mmap address */ struct Scsi_Host *sh; /* Scsi Host pointer */ ScsiCfgData spi_data; /* Scsi config. data */ + SasCfgData sas_data; /* Sas config. data */ MPT_IOCTL *ioctl; /* ioctl data pointer */ struct proc_dir_entry *ioc_dentry; struct _MPT_ADAPTER *alt_ioc; /* ptr to 929 bound adapter port */ @@ -554,31 +567,35 @@ typedef struct _MPT_ADAPTER #else u32 mfcnt; #endif - u32 NB_for_64_byte_frame; + u32 NB_for_64_byte_frame; u32 hs_req[MPT_MAX_FRAME_SIZE/sizeof(u32)]; u16 hs_reply[MPT_MAX_FRAME_SIZE/sizeof(u16)]; IOCFactsReply_t facts; PortFactsReply_t pfacts[2]; FCPortPage0_t fc_port_page0[2]; + struct timer_list persist_timer; /* persist table timer */ + int persist_wait_done; /* persist completion flag */ + u8 persist_reply_frame[MPT_DEFAULT_FRAME_SIZE]; /* persist reply */ LANPage0_t lan_cnfg_page0; LANPage1_t lan_cnfg_page1; - /* + /* * Description: errata_flag_1064 * If a PCIX read occurs within 1 or 2 cycles after the chip receives * a split completion for a read data, an internal address pointer incorrectly * increments by 32 bytes */ - int errata_flag_1064; + int errata_flag_1064; u8 FirstWhoInit; u8 upload_fw; /* If set, do a fw upload */ u8 reload_fw; /* Force a FW Reload on next reset */ - u8 NBShiftFactor; /* NB Shift Factor based on Block Size (Facts) */ + u8 NBShiftFactor; /* NB Shift Factor based on Block Size (Facts) */ u8 pad1[4]; int DoneCtx; int TaskCtx; int InternalCtx; - struct list_head list; + struct list_head list; struct net_device *netdev; + struct list_head sas_topology; } MPT_ADAPTER; /* @@ -964,6 +981,7 @@ extern void mpt_alloc_fw_memory(MPT_ADAPTER *ioc, int size); extern void mpt_free_fw_memory(MPT_ADAPTER *ioc); extern int mpt_findImVolumes(MPT_ADAPTER *ioc); extern int mpt_read_ioc_pg_3(MPT_ADAPTER *ioc); +extern int mptbase_sas_persist_operation(MPT_ADAPTER *ioc, u8 persist_opcode); /* * Public data decl's... diff --git a/drivers/message/fusion/mptscsih.c b/drivers/message/fusion/mptscsih.c index 4a003dc5fde8..58b5fdee009a 100644 --- a/drivers/message/fusion/mptscsih.c +++ b/drivers/message/fusion/mptscsih.c @@ -1256,8 +1256,7 @@ mptscsih_qcmd(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *)) MPT_SCSI_HOST *hd; MPT_FRAME_HDR *mf; SCSIIORequest_t *pScsiReq; - VirtDevice *pTarget; - int target; + VirtDevice *pTarget = SCpnt->device->hostdata; int lun; u32 datalen; u32 scsictl; @@ -1267,12 +1266,9 @@ mptscsih_qcmd(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *)) int ii; hd = (MPT_SCSI_HOST *) SCpnt->device->host->hostdata; - target = SCpnt->device->id; lun = SCpnt->device->lun; SCpnt->scsi_done = done; - pTarget = hd->Targets[target]; - dmfprintk((MYIOC_s_INFO_FMT "qcmd: SCpnt=%p, done()=%p\n", (hd && hd->ioc) ? hd->ioc->name : "ioc?", SCpnt, done)); @@ -1315,7 +1311,7 @@ mptscsih_qcmd(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *)) /* Default to untagged. Once a target structure has been allocated, * use the Inquiry data to determine if device supports tagged. */ - if ( pTarget + if (pTarget && (pTarget->tflags & MPT_TARGET_FLAGS_Q_YES) && (SCpnt->device->tagged_supported)) { scsictl = scsidir | MPI_SCSIIO_CONTROL_SIMPLEQ; @@ -1325,8 +1321,8 @@ mptscsih_qcmd(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *)) /* Use the above information to set up the message frame */ - pScsiReq->TargetID = (u8) target; - pScsiReq->Bus = (u8) SCpnt->device->channel; + pScsiReq->TargetID = (u8) pTarget->target_id; + pScsiReq->Bus = pTarget->bus_id; pScsiReq->ChainOffset = 0; pScsiReq->Function = MPI_FUNCTION_SCSI_IO_REQUEST; pScsiReq->CDBLength = SCpnt->cmd_len; @@ -1378,7 +1374,7 @@ mptscsih_qcmd(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *)) #ifdef MPTSCSIH_ENABLE_DOMAIN_VALIDATION if (hd->ioc->bus_type == SCSI) { - int dvStatus = hd->ioc->spi_data.dvStatus[target]; + int dvStatus = hd->ioc->spi_data.dvStatus[pTarget->target_id]; int issueCmd = 1; if (dvStatus || hd->ioc->spi_data.forceDv) { @@ -2180,6 +2176,7 @@ mptscsih_slave_alloc(struct scsi_device *device) out: vdev->num_luns++; + device->hostdata = vdev; return 0; } From 0c33b27deb93178f10778b3d2669af1674793cef Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 9 Sep 2005 16:27:19 +0200 Subject: [PATCH 165/240] [SCSI] fusion SAS support (mptsas driver) Adds the actual mptsas driver, based upon the LSI driver with new work for SAS transport class integration from Eric Moore and me. This obviously depends on the SAS transport class. Signed-off-by: Christoph Hellwig Signed-off-by: James Bottomley --- drivers/message/fusion/Kconfig | 17 + drivers/message/fusion/Makefile | 1 + drivers/message/fusion/mptsas.c | 1235 +++++++++++++++++++++++++++++++ 3 files changed, 1253 insertions(+) create mode 100644 drivers/message/fusion/mptsas.c diff --git a/drivers/message/fusion/Kconfig b/drivers/message/fusion/Kconfig index 33f209a39cb6..1883d22cffeb 100644 --- a/drivers/message/fusion/Kconfig +++ b/drivers/message/fusion/Kconfig @@ -35,6 +35,23 @@ config FUSION_FC LSIFC929X LSIFC929XL +config FUSION_SAS + tristate "Fusion MPT ScsiHost drivers for SAS" + depends on PCI && SCSI + select FUSION + select SCSI_SAS_ATTRS + ---help--- + SCSI HOST support for a SAS host adapters. + + List of supported controllers: + + LSISAS1064 + LSISAS1066 + LSISAS1068 + LSISAS1064E + LSISAS1066E + LSISAS1068E + config FUSION_MAX_SGE int "Maximum number of scatter gather entries (16 - 128)" depends on FUSION diff --git a/drivers/message/fusion/Makefile b/drivers/message/fusion/Makefile index 1d2f9db813c1..8a2e2657f4c2 100644 --- a/drivers/message/fusion/Makefile +++ b/drivers/message/fusion/Makefile @@ -34,5 +34,6 @@ obj-$(CONFIG_FUSION_SPI) += mptbase.o mptscsih.o mptspi.o obj-$(CONFIG_FUSION_FC) += mptbase.o mptscsih.o mptfc.o +obj-$(CONFIG_FUSION_SAS) += mptbase.o mptscsih.o mptsas.o obj-$(CONFIG_FUSION_CTL) += mptctl.o obj-$(CONFIG_FUSION_LAN) += mptlan.o diff --git a/drivers/message/fusion/mptsas.c b/drivers/message/fusion/mptsas.c new file mode 100644 index 000000000000..0d9a192e1bd4 --- /dev/null +++ b/drivers/message/fusion/mptsas.c @@ -0,0 +1,1235 @@ +/* + * linux/drivers/message/fusion/mptsas.c + * For use with LSI Logic PCI chip/adapter(s) + * running LSI Logic Fusion MPT (Message Passing Technology) firmware. + * + * Copyright (c) 1999-2005 LSI Logic Corporation + * (mailto:mpt_linux_developer@lsil.com) + * Copyright (c) 2005 Dell + */ +/*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ +/* + 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; version 2 of the License. + + 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. + + NO WARRANTY + THE PROGRAM IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OR + CONDITIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED INCLUDING, WITHOUT + LIMITATION, ANY WARRANTIES OR CONDITIONS OF TITLE, NON-INFRINGEMENT, + MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Each Recipient is + solely responsible for determining the appropriateness of using and + distributing the Program and assumes all risks associated with its + exercise of rights under this Agreement, including but not limited to + the risks and costs of program errors, damage to or loss of data, + programs or equipment, and unavailability or interruption of operations. + + DISCLAIMER OF LIABILITY + NEITHER RECIPIENT NOR ANY CONTRIBUTORS SHALL HAVE ANY LIABILITY FOR ANY + DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + DAMAGES (INCLUDING WITHOUT LIMITATION LOST PROFITS), HOWEVER CAUSED AND + ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR + TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE + USE OR DISTRIBUTION OF THE PROGRAM OR THE EXERCISE OF ANY RIGHTS GRANTED + HEREUNDER, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES + + 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 + +#include +#include +#include +#include + +#include "mptbase.h" +#include "mptscsih.h" + + +#define my_NAME "Fusion MPT SAS Host driver" +#define my_VERSION MPT_LINUX_VERSION_COMMON +#define MYNAM "mptsas" + +MODULE_AUTHOR(MODULEAUTHOR); +MODULE_DESCRIPTION(my_NAME); +MODULE_LICENSE("GPL"); + +static int mpt_pq_filter; +module_param(mpt_pq_filter, int, 0); +MODULE_PARM_DESC(mpt_pq_filter, + "Enable peripheral qualifier filter: enable=1 " + "(default=0)"); + +static int mpt_pt_clear; +module_param(mpt_pt_clear, int, 0); +MODULE_PARM_DESC(mpt_pt_clear, + "Clear persistency table: enable=1 " + "(default=MPTSCSIH_PT_CLEAR=0)"); + +static int mptsasDoneCtx = -1; +static int mptsasTaskCtx = -1; +static int mptsasInternalCtx = -1; /* Used only for internal commands */ + + +/* + * SAS topology structures + * + * The MPT Fusion firmware interface spreads information about the + * SAS topology over many manufacture pages, thus we need some data + * structure to collect it and process it for the SAS transport class. + */ + +struct mptsas_devinfo { + u16 handle; /* unique id to address this device */ + u8 phy_id; /* phy number of parent device */ + u8 port_id; /* sas physical port this device + is assoc'd with */ + u8 target; /* logical target id of this device */ + u8 bus; /* logical bus number of this device */ + u64 sas_address; /* WWN of this device, + SATA is assigned by HBA,expander */ + u32 device_info; /* bitfield detailed info about this device */ +}; + +struct mptsas_phyinfo { + u8 phy_id; /* phy index */ + u8 port_id; /* port number this phy is part of */ + u8 negotiated_link_rate; /* nego'd link rate for this phy */ + u8 hw_link_rate; /* hardware max/min phys link rate */ + u8 programmed_link_rate; /* programmed max/min phy link rate */ + struct mptsas_devinfo identify; /* point to phy device info */ + struct mptsas_devinfo attached; /* point to attached device info */ + struct sas_rphy *rphy; +}; + +struct mptsas_portinfo { + struct list_head list; + u16 handle; /* unique id to address this */ + u8 num_phys; /* number of phys */ + struct mptsas_phyinfo *phy_info; +}; + +/* + * This is pretty ugly. We will be able to seriously clean it up + * once the DV code in mptscsih goes away and we can properly + * implement ->target_alloc. + */ +static int +mptsas_slave_alloc(struct scsi_device *device) +{ + struct Scsi_Host *host = device->host; + MPT_SCSI_HOST *hd = (MPT_SCSI_HOST *)host->hostdata; + struct sas_rphy *rphy; + struct mptsas_portinfo *p; + VirtDevice *vdev; + uint target = device->id; + int i; + + if ((vdev = hd->Targets[target]) != NULL) + goto out; + + vdev = kmalloc(sizeof(VirtDevice), GFP_KERNEL); + if (!vdev) { + printk(MYIOC_s_ERR_FMT "slave_alloc kmalloc(%zd) FAILED!\n", + hd->ioc->name, sizeof(VirtDevice)); + return -ENOMEM; + } + + memset(vdev, 0, sizeof(VirtDevice)); + vdev->tflags = MPT_TARGET_FLAGS_Q_YES|MPT_TARGET_FLAGS_VALID_INQUIRY; + vdev->ioc_id = hd->ioc->id; + + rphy = dev_to_rphy(device->sdev_target->dev.parent); + list_for_each_entry(p, &hd->ioc->sas_topology, list) { + for (i = 0; i < p->num_phys; i++) { + if (p->phy_info[i].attached.sas_address == + rphy->identify.sas_address) { + vdev->target_id = + p->phy_info[i].attached.target; + vdev->bus_id = p->phy_info[i].attached.bus; + hd->Targets[device->id] = vdev; + goto out; + } + } + } + + printk("No matching SAS device found!!\n"); + kfree(vdev); + return -ENODEV; + + out: + vdev->num_luns++; + device->hostdata = vdev; + return 0; +} + +static struct scsi_host_template mptsas_driver_template = { + .proc_name = "mptsas", + .proc_info = mptscsih_proc_info, + .name = "MPT SPI Host", + .info = mptscsih_info, + .queuecommand = mptscsih_qcmd, + .slave_alloc = mptsas_slave_alloc, + .slave_configure = mptscsih_slave_configure, + .slave_destroy = mptscsih_slave_destroy, + .change_queue_depth = mptscsih_change_queue_depth, + .eh_abort_handler = mptscsih_abort, + .eh_device_reset_handler = mptscsih_dev_reset, + .eh_bus_reset_handler = mptscsih_bus_reset, + .eh_host_reset_handler = mptscsih_host_reset, + .bios_param = mptscsih_bios_param, + .can_queue = MPT_FC_CAN_QUEUE, + .this_id = -1, + .sg_tablesize = MPT_SCSI_SG_DEPTH, + .max_sectors = 8192, + .cmd_per_lun = 7, + .use_clustering = ENABLE_CLUSTERING, +}; + +static struct sas_function_template mptsas_transport_functions = { +}; + +static struct scsi_transport_template *mptsas_transport_template; + +#ifdef SASDEBUG +static void mptsas_print_phy_data(MPI_SAS_IO_UNIT0_PHY_DATA *phy_data) +{ + printk("---- IO UNIT PAGE 0 ------------\n"); + printk("Handle=0x%X\n", + le16_to_cpu(phy_data->AttachedDeviceHandle)); + printk("Controller Handle=0x%X\n", + le16_to_cpu(phy_data->ControllerDevHandle)); + printk("Port=0x%X\n", phy_data->Port); + printk("Port Flags=0x%X\n", phy_data->PortFlags); + printk("PHY Flags=0x%X\n", phy_data->PhyFlags); + printk("Negotiated Link Rate=0x%X\n", phy_data->NegotiatedLinkRate); + printk("Controller PHY Device Info=0x%X\n", + le32_to_cpu(phy_data->ControllerPhyDeviceInfo)); + printk("DiscoveryStatus=0x%X\n", + le32_to_cpu(phy_data->DiscoveryStatus)); + printk("\n"); +} + +static void mptsas_print_phy_pg0(SasPhyPage0_t *pg0) +{ + __le64 sas_address; + + memcpy(&sas_address, &pg0->SASAddress, sizeof(__le64)); + + printk("---- SAS PHY PAGE 0 ------------\n"); + printk("Attached Device Handle=0x%X\n", + le16_to_cpu(pg0->AttachedDevHandle)); + printk("SAS Address=0x%llX\n", + (unsigned long long)le64_to_cpu(sas_address)); + printk("Attached PHY Identifier=0x%X\n", pg0->AttachedPhyIdentifier); + printk("Attached Device Info=0x%X\n", + le32_to_cpu(pg0->AttachedDeviceInfo)); + printk("Programmed Link Rate=0x%X\n", pg0->ProgrammedLinkRate); + printk("Change Count=0x%X\n", pg0->ChangeCount); + printk("PHY Info=0x%X\n", le32_to_cpu(pg0->PhyInfo)); + printk("\n"); +} + +static void mptsas_print_device_pg0(SasDevicePage0_t *pg0) +{ + __le64 sas_address; + + memcpy(&sas_address, &pg0->SASAddress, sizeof(__le64)); + + printk("---- SAS DEVICE PAGE 0 ---------\n"); + printk("Handle=0x%X\n" ,le16_to_cpu(pg0->DevHandle)); + printk("Enclosure Handle=0x%X\n", le16_to_cpu(pg0->EnclosureHandle)); + printk("Slot=0x%X\n", le16_to_cpu(pg0->Slot)); + printk("SAS Address=0x%llX\n", le64_to_cpu(sas_address)); + printk("Target ID=0x%X\n", pg0->TargetID); + printk("Bus=0x%X\n", pg0->Bus); + printk("PhyNum=0x%X\n", pg0->PhyNum); + printk("AccessStatus=0x%X\n", le16_to_cpu(pg0->AccessStatus)); + printk("Device Info=0x%X\n", le32_to_cpu(pg0->DeviceInfo)); + printk("Flags=0x%X\n", le16_to_cpu(pg0->Flags)); + printk("Physical Port=0x%X\n", pg0->PhysicalPort); + printk("\n"); +} + +static void mptsas_print_expander_pg1(SasExpanderPage1_t *pg1) +{ + printk("---- SAS EXPANDER PAGE 1 ------------\n"); + + printk("Physical Port=0x%X\n", pg1->PhysicalPort); + printk("PHY Identifier=0x%X\n", pg1->Phy); + printk("Negotiated Link Rate=0x%X\n", pg1->NegotiatedLinkRate); + printk("Programmed Link Rate=0x%X\n", pg1->ProgrammedLinkRate); + printk("Hardware Link Rate=0x%X\n", pg1->HwLinkRate); + printk("Owner Device Handle=0x%X\n", + le16_to_cpu(pg1->OwnerDevHandle)); + printk("Attached Device Handle=0x%X\n", + le16_to_cpu(pg1->AttachedDevHandle)); +} +#else +#define mptsas_print_phy_data(phy_data) do { } while (0) +#define mptsas_print_phy_pg0(pg0) do { } while (0) +#define mptsas_print_device_pg0(pg0) do { } while (0) +#define mptsas_print_expander_pg1(pg1) do { } while (0) +#endif + +static int +mptsas_sas_io_unit_pg0(MPT_ADAPTER *ioc, struct mptsas_portinfo *port_info) +{ + ConfigExtendedPageHeader_t hdr; + CONFIGPARMS cfg; + SasIOUnitPage0_t *buffer; + dma_addr_t dma_handle; + int error, i; + + hdr.PageVersion = MPI_SASIOUNITPAGE0_PAGEVERSION; + hdr.ExtPageLength = 0; + hdr.PageNumber = 0; + hdr.Reserved1 = 0; + hdr.Reserved2 = 0; + hdr.PageType = MPI_CONFIG_PAGETYPE_EXTENDED; + hdr.ExtPageType = MPI_CONFIG_EXTPAGETYPE_SAS_IO_UNIT; + + cfg.cfghdr.ehdr = &hdr; + cfg.physAddr = -1; + cfg.pageAddr = 0; + cfg.action = MPI_CONFIG_ACTION_PAGE_HEADER; + cfg.dir = 0; /* read */ + cfg.timeout = 10; + + error = mpt_config(ioc, &cfg); + if (error) + goto out; + if (!hdr.ExtPageLength) { + error = -ENXIO; + goto out; + } + + buffer = pci_alloc_consistent(ioc->pcidev, hdr.ExtPageLength * 4, + &dma_handle); + if (!buffer) { + error = -ENOMEM; + goto out; + } + + cfg.physAddr = dma_handle; + cfg.action = MPI_CONFIG_ACTION_PAGE_READ_CURRENT; + + error = mpt_config(ioc, &cfg); + if (error) + goto out_free_consistent; + + port_info->num_phys = buffer->NumPhys; + port_info->phy_info = kcalloc(port_info->num_phys, + sizeof(struct mptsas_phyinfo),GFP_KERNEL); + if (!port_info->phy_info) { + error = -ENOMEM; + goto out_free_consistent; + } + + for (i = 0; i < port_info->num_phys; i++) { + mptsas_print_phy_data(&buffer->PhyData[i]); + port_info->phy_info[i].phy_id = i; + port_info->phy_info[i].port_id = + buffer->PhyData[i].Port; + port_info->phy_info[i].negotiated_link_rate = + buffer->PhyData[i].NegotiatedLinkRate; + } + + out_free_consistent: + pci_free_consistent(ioc->pcidev, hdr.ExtPageLength * 4, + buffer, dma_handle); + out: + return error; +} + +static int +mptsas_sas_phy_pg0(MPT_ADAPTER *ioc, struct mptsas_phyinfo *phy_info, + u32 form, u32 form_specific) +{ + ConfigExtendedPageHeader_t hdr; + CONFIGPARMS cfg; + SasPhyPage0_t *buffer; + dma_addr_t dma_handle; + int error; + + hdr.PageVersion = MPI_SASPHY0_PAGEVERSION; + hdr.ExtPageLength = 0; + hdr.PageNumber = 0; + hdr.Reserved1 = 0; + hdr.Reserved2 = 0; + hdr.PageType = MPI_CONFIG_PAGETYPE_EXTENDED; + hdr.ExtPageType = MPI_CONFIG_EXTPAGETYPE_SAS_PHY; + + cfg.cfghdr.ehdr = &hdr; + cfg.dir = 0; /* read */ + cfg.timeout = 10; + + /* Get Phy Pg 0 for each Phy. */ + cfg.physAddr = -1; + cfg.pageAddr = form + form_specific; + cfg.action = MPI_CONFIG_ACTION_PAGE_HEADER; + + error = mpt_config(ioc, &cfg); + if (error) + goto out; + + if (!hdr.ExtPageLength) { + error = -ENXIO; + goto out; + } + + buffer = pci_alloc_consistent(ioc->pcidev, hdr.ExtPageLength * 4, + &dma_handle); + if (!buffer) { + error = -ENOMEM; + goto out; + } + + cfg.physAddr = dma_handle; + cfg.action = MPI_CONFIG_ACTION_PAGE_READ_CURRENT; + + error = mpt_config(ioc, &cfg); + if (error) + goto out_free_consistent; + + mptsas_print_phy_pg0(buffer); + + phy_info->hw_link_rate = buffer->HwLinkRate; + phy_info->programmed_link_rate = buffer->ProgrammedLinkRate; + phy_info->identify.handle = le16_to_cpu(buffer->OwnerDevHandle); + phy_info->attached.handle = le16_to_cpu(buffer->AttachedDevHandle); + + out_free_consistent: + pci_free_consistent(ioc->pcidev, hdr.ExtPageLength * 4, + buffer, dma_handle); + out: + return error; +} + +static int +mptsas_sas_device_pg0(MPT_ADAPTER *ioc, struct mptsas_devinfo *device_info, + u32 form, u32 form_specific) +{ + ConfigExtendedPageHeader_t hdr; + CONFIGPARMS cfg; + SasDevicePage0_t *buffer; + dma_addr_t dma_handle; + __le64 sas_address; + int error; + + hdr.PageVersion = MPI_SASDEVICE0_PAGEVERSION; + hdr.ExtPageLength = 0; + hdr.PageNumber = 0; + hdr.Reserved1 = 0; + hdr.Reserved2 = 0; + hdr.PageType = MPI_CONFIG_PAGETYPE_EXTENDED; + hdr.ExtPageType = MPI_CONFIG_EXTPAGETYPE_SAS_DEVICE; + + cfg.cfghdr.ehdr = &hdr; + cfg.pageAddr = form + form_specific; + cfg.physAddr = -1; + cfg.action = MPI_CONFIG_ACTION_PAGE_HEADER; + cfg.dir = 0; /* read */ + cfg.timeout = 10; + + error = mpt_config(ioc, &cfg); + if (error) + goto out; + if (!hdr.ExtPageLength) { + error = -ENXIO; + goto out; + } + + buffer = pci_alloc_consistent(ioc->pcidev, hdr.ExtPageLength * 4, + &dma_handle); + if (!buffer) { + error = -ENOMEM; + goto out; + } + + cfg.physAddr = dma_handle; + cfg.action = MPI_CONFIG_ACTION_PAGE_READ_CURRENT; + + error = mpt_config(ioc, &cfg); + if (error) + goto out_free_consistent; + + mptsas_print_device_pg0(buffer); + + device_info->handle = le16_to_cpu(buffer->DevHandle); + device_info->phy_id = buffer->PhyNum; + device_info->port_id = buffer->PhysicalPort; + device_info->target = buffer->TargetID; + device_info->bus = buffer->Bus; + memcpy(&sas_address, &buffer->SASAddress, sizeof(__le64)); + device_info->sas_address = le64_to_cpu(sas_address); + device_info->device_info = + le32_to_cpu(buffer->DeviceInfo); + + out_free_consistent: + pci_free_consistent(ioc->pcidev, hdr.ExtPageLength * 4, + buffer, dma_handle); + out: + return error; +} + +static int +mptsas_sas_expander_pg0(MPT_ADAPTER *ioc, struct mptsas_portinfo *port_info, + u32 form, u32 form_specific) +{ + ConfigExtendedPageHeader_t hdr; + CONFIGPARMS cfg; + SasExpanderPage0_t *buffer; + dma_addr_t dma_handle; + int error; + + hdr.PageVersion = MPI_SASEXPANDER0_PAGEVERSION; + hdr.ExtPageLength = 0; + hdr.PageNumber = 0; + hdr.Reserved1 = 0; + hdr.Reserved2 = 0; + hdr.PageType = MPI_CONFIG_PAGETYPE_EXTENDED; + hdr.ExtPageType = MPI_CONFIG_EXTPAGETYPE_SAS_EXPANDER; + + cfg.cfghdr.ehdr = &hdr; + cfg.physAddr = -1; + cfg.pageAddr = form + form_specific; + cfg.action = MPI_CONFIG_ACTION_PAGE_HEADER; + cfg.dir = 0; /* read */ + cfg.timeout = 10; + + error = mpt_config(ioc, &cfg); + if (error) + goto out; + + if (!hdr.ExtPageLength) { + error = -ENXIO; + goto out; + } + + buffer = pci_alloc_consistent(ioc->pcidev, hdr.ExtPageLength * 4, + &dma_handle); + if (!buffer) { + error = -ENOMEM; + goto out; + } + + cfg.physAddr = dma_handle; + cfg.action = MPI_CONFIG_ACTION_PAGE_READ_CURRENT; + + error = mpt_config(ioc, &cfg); + if (error) + goto out_free_consistent; + + /* save config data */ + port_info->num_phys = buffer->NumPhys; + port_info->handle = le16_to_cpu(buffer->DevHandle); + port_info->phy_info = kcalloc(port_info->num_phys, + sizeof(struct mptsas_phyinfo),GFP_KERNEL); + if (!port_info->phy_info) { + error = -ENOMEM; + goto out_free_consistent; + } + + out_free_consistent: + pci_free_consistent(ioc->pcidev, hdr.ExtPageLength * 4, + buffer, dma_handle); + out: + return error; +} + +static int +mptsas_sas_expander_pg1(MPT_ADAPTER *ioc, struct mptsas_phyinfo *phy_info, + u32 form, u32 form_specific) +{ + ConfigExtendedPageHeader_t hdr; + CONFIGPARMS cfg; + SasExpanderPage1_t *buffer; + dma_addr_t dma_handle; + int error; + + hdr.PageVersion = MPI_SASEXPANDER0_PAGEVERSION; + hdr.ExtPageLength = 0; + hdr.PageNumber = 1; + hdr.Reserved1 = 0; + hdr.Reserved2 = 0; + hdr.PageType = MPI_CONFIG_PAGETYPE_EXTENDED; + hdr.ExtPageType = MPI_CONFIG_EXTPAGETYPE_SAS_EXPANDER; + + cfg.cfghdr.ehdr = &hdr; + cfg.physAddr = -1; + cfg.pageAddr = form + form_specific; + cfg.action = MPI_CONFIG_ACTION_PAGE_HEADER; + cfg.dir = 0; /* read */ + cfg.timeout = 10; + + error = mpt_config(ioc, &cfg); + if (error) + goto out; + + if (!hdr.ExtPageLength) { + error = -ENXIO; + goto out; + } + + buffer = pci_alloc_consistent(ioc->pcidev, hdr.ExtPageLength * 4, + &dma_handle); + if (!buffer) { + error = -ENOMEM; + goto out; + } + + cfg.physAddr = dma_handle; + cfg.action = MPI_CONFIG_ACTION_PAGE_READ_CURRENT; + + error = mpt_config(ioc, &cfg); + if (error) + goto out_free_consistent; + + + mptsas_print_expander_pg1(buffer); + + /* save config data */ + phy_info->phy_id = buffer->Phy; + phy_info->port_id = buffer->PhysicalPort; + phy_info->negotiated_link_rate = buffer->NegotiatedLinkRate; + phy_info->programmed_link_rate = buffer->ProgrammedLinkRate; + phy_info->hw_link_rate = buffer->HwLinkRate; + phy_info->identify.handle = le16_to_cpu(buffer->OwnerDevHandle); + phy_info->attached.handle = le16_to_cpu(buffer->AttachedDevHandle); + + + out_free_consistent: + pci_free_consistent(ioc->pcidev, hdr.ExtPageLength * 4, + buffer, dma_handle); + out: + return error; +} + +static void +mptsas_parse_device_info(struct sas_identify *identify, + struct mptsas_devinfo *device_info) +{ + u16 protocols; + + identify->sas_address = device_info->sas_address; + identify->phy_identifier = device_info->phy_id; + + /* + * Fill in Phy Initiator Port Protocol. + * Bits 6:3, more than one bit can be set, fall through cases. + */ + protocols = device_info->device_info & 0x78; + identify->initiator_port_protocols = 0; + if (protocols & MPI_SAS_DEVICE_INFO_SSP_INITIATOR) + identify->initiator_port_protocols |= SAS_PROTOCOL_SSP; + if (protocols & MPI_SAS_DEVICE_INFO_STP_INITIATOR) + identify->initiator_port_protocols |= SAS_PROTOCOL_STP; + if (protocols & MPI_SAS_DEVICE_INFO_SMP_INITIATOR) + identify->initiator_port_protocols |= SAS_PROTOCOL_SMP; + if (protocols & MPI_SAS_DEVICE_INFO_SATA_HOST) + identify->initiator_port_protocols |= SAS_PROTOCOL_SATA; + + /* + * Fill in Phy Target Port Protocol. + * Bits 10:7, more than one bit can be set, fall through cases. + */ + protocols = device_info->device_info & 0x780; + identify->target_port_protocols = 0; + if (protocols & MPI_SAS_DEVICE_INFO_SSP_TARGET) + identify->target_port_protocols |= SAS_PROTOCOL_SSP; + if (protocols & MPI_SAS_DEVICE_INFO_STP_TARGET) + identify->target_port_protocols |= SAS_PROTOCOL_STP; + if (protocols & MPI_SAS_DEVICE_INFO_SMP_TARGET) + identify->target_port_protocols |= SAS_PROTOCOL_SMP; + if (protocols & MPI_SAS_DEVICE_INFO_SATA_DEVICE) + identify->target_port_protocols |= SAS_PROTOCOL_SATA; + + /* + * Fill in Attached device type. + */ + switch (device_info->device_info & + MPI_SAS_DEVICE_INFO_MASK_DEVICE_TYPE) { + case MPI_SAS_DEVICE_INFO_NO_DEVICE: + identify->device_type = SAS_PHY_UNUSED; + break; + case MPI_SAS_DEVICE_INFO_END_DEVICE: + identify->device_type = SAS_END_DEVICE; + break; + case MPI_SAS_DEVICE_INFO_EDGE_EXPANDER: + identify->device_type = SAS_EDGE_EXPANDER_DEVICE; + break; + case MPI_SAS_DEVICE_INFO_FANOUT_EXPANDER: + identify->device_type = SAS_FANOUT_EXPANDER_DEVICE; + break; + } +} + +static int mptsas_probe_one_phy(struct device *dev, + struct mptsas_phyinfo *phy_info, int index) +{ + struct sas_phy *port; + int error; + + port = sas_phy_alloc(dev, index); + if (!port) + return -ENOMEM; + + port->port_identifier = phy_info->port_id; + mptsas_parse_device_info(&port->identify, &phy_info->identify); + + /* + * Set Negotiated link rate. + */ + switch (phy_info->negotiated_link_rate) { + case MPI_SAS_IOUNIT0_RATE_PHY_DISABLED: + port->negotiated_linkrate = SAS_PHY_DISABLED; + break; + case MPI_SAS_IOUNIT0_RATE_FAILED_SPEED_NEGOTIATION: + port->negotiated_linkrate = SAS_LINK_RATE_FAILED; + break; + case MPI_SAS_IOUNIT0_RATE_1_5: + port->negotiated_linkrate = SAS_LINK_RATE_1_5_GBPS; + break; + case MPI_SAS_IOUNIT0_RATE_3_0: + port->negotiated_linkrate = SAS_LINK_RATE_3_0_GBPS; + break; + case MPI_SAS_IOUNIT0_RATE_SATA_OOB_COMPLETE: + case MPI_SAS_IOUNIT0_RATE_UNKNOWN: + default: + port->negotiated_linkrate = SAS_LINK_RATE_UNKNOWN; + break; + } + + /* + * Set Max hardware link rate. + */ + switch (phy_info->hw_link_rate & MPI_SAS_PHY0_PRATE_MAX_RATE_MASK) { + case MPI_SAS_PHY0_HWRATE_MAX_RATE_1_5: + port->maximum_linkrate_hw = SAS_LINK_RATE_1_5_GBPS; + break; + case MPI_SAS_PHY0_PRATE_MAX_RATE_3_0: + port->maximum_linkrate_hw = SAS_LINK_RATE_3_0_GBPS; + break; + default: + break; + } + + /* + * Set Max programmed link rate. + */ + switch (phy_info->programmed_link_rate & + MPI_SAS_PHY0_PRATE_MAX_RATE_MASK) { + case MPI_SAS_PHY0_PRATE_MAX_RATE_1_5: + port->maximum_linkrate = SAS_LINK_RATE_1_5_GBPS; + break; + case MPI_SAS_PHY0_PRATE_MAX_RATE_3_0: + port->maximum_linkrate = SAS_LINK_RATE_3_0_GBPS; + break; + default: + break; + } + + /* + * Set Min hardware link rate. + */ + switch (phy_info->hw_link_rate & MPI_SAS_PHY0_HWRATE_MIN_RATE_MASK) { + case MPI_SAS_PHY0_HWRATE_MIN_RATE_1_5: + port->minimum_linkrate_hw = SAS_LINK_RATE_1_5_GBPS; + break; + case MPI_SAS_PHY0_PRATE_MIN_RATE_3_0: + port->minimum_linkrate_hw = SAS_LINK_RATE_3_0_GBPS; + break; + default: + break; + } + + /* + * Set Min programmed link rate. + */ + switch (phy_info->programmed_link_rate & + MPI_SAS_PHY0_PRATE_MIN_RATE_MASK) { + case MPI_SAS_PHY0_PRATE_MIN_RATE_1_5: + port->minimum_linkrate = SAS_LINK_RATE_1_5_GBPS; + break; + case MPI_SAS_PHY0_PRATE_MIN_RATE_3_0: + port->minimum_linkrate = SAS_LINK_RATE_3_0_GBPS; + break; + default: + break; + } + + error = sas_phy_add(port); + if (error) { + sas_phy_free(port); + return error; + } + + if (phy_info->attached.handle) { + struct sas_rphy *rphy; + + rphy = sas_rphy_alloc(port); + if (!rphy) + return 0; /* non-fatal: an rphy can be added later */ + + mptsas_parse_device_info(&rphy->identify, &phy_info->attached); + error = sas_rphy_add(rphy); + if (error) { + sas_rphy_free(rphy); + return error; + } + + phy_info->rphy = rphy; + } + + return 0; +} + +static int +mptsas_probe_hba_phys(MPT_ADAPTER *ioc, int *index) +{ + struct mptsas_portinfo *port_info; + u32 handle = 0xFFFF; + int error = -ENOMEM, i; + + port_info = kmalloc(sizeof(*port_info), GFP_KERNEL); + if (!port_info) + goto out; + memset(port_info, 0, sizeof(*port_info)); + + error = mptsas_sas_io_unit_pg0(ioc, port_info); + if (error) + goto out_free_port_info; + + list_add_tail(&port_info->list, &ioc->sas_topology); + + for (i = 0; i < port_info->num_phys; i++) { + mptsas_sas_phy_pg0(ioc, &port_info->phy_info[i], + (MPI_SAS_PHY_PGAD_FORM_PHY_NUMBER << + MPI_SAS_PHY_PGAD_FORM_SHIFT), i); + + mptsas_sas_device_pg0(ioc, &port_info->phy_info[i].identify, + (MPI_SAS_DEVICE_PGAD_FORM_GET_NEXT_HANDLE << + MPI_SAS_DEVICE_PGAD_FORM_SHIFT), handle); + handle = port_info->phy_info[i].identify.handle; + + if (port_info->phy_info[i].attached.handle) { + mptsas_sas_device_pg0(ioc, + &port_info->phy_info[i].attached, + (MPI_SAS_DEVICE_PGAD_FORM_HANDLE << + MPI_SAS_DEVICE_PGAD_FORM_SHIFT), + port_info->phy_info[i].attached.handle); + } + + mptsas_probe_one_phy(&ioc->sh->shost_gendev, + &port_info->phy_info[i], *index); + (*index)++; + } + + return 0; + + out_free_port_info: + kfree(port_info); + out: + return error; +} + +static int +mptsas_probe_expander_phys(MPT_ADAPTER *ioc, u32 *handle, int *index) +{ + struct mptsas_portinfo *port_info, *p; + int error = -ENOMEM, i, j; + + port_info = kmalloc(sizeof(*port_info), GFP_KERNEL); + if (!port_info) + goto out; + memset(port_info, 0, sizeof(*port_info)); + + error = mptsas_sas_expander_pg0(ioc, port_info, + (MPI_SAS_EXPAND_PGAD_FORM_GET_NEXT_HANDLE << + MPI_SAS_EXPAND_PGAD_FORM_SHIFT), *handle); + if (error) + goto out_free_port_info; + + *handle = port_info->handle; + + list_add_tail(&port_info->list, &ioc->sas_topology); + for (i = 0; i < port_info->num_phys; i++) { + struct device *parent; + + mptsas_sas_expander_pg1(ioc, &port_info->phy_info[i], + (MPI_SAS_EXPAND_PGAD_FORM_HANDLE_PHY_NUM << + MPI_SAS_EXPAND_PGAD_FORM_SHIFT), (i << 16) + *handle); + + if (port_info->phy_info[i].identify.handle) { + mptsas_sas_device_pg0(ioc, + &port_info->phy_info[i].identify, + (MPI_SAS_DEVICE_PGAD_FORM_HANDLE << + MPI_SAS_DEVICE_PGAD_FORM_SHIFT), + port_info->phy_info[i].identify.handle); + } + + if (port_info->phy_info[i].attached.handle) { + mptsas_sas_device_pg0(ioc, + &port_info->phy_info[i].attached, + (MPI_SAS_DEVICE_PGAD_FORM_HANDLE << + MPI_SAS_DEVICE_PGAD_FORM_SHIFT), + port_info->phy_info[i].attached.handle); + } + + /* + * If we find a parent port handle this expander is + * attached to another expander, else it hangs of the + * HBA phys. + */ + parent = &ioc->sh->shost_gendev; + list_for_each_entry(p, &ioc->sas_topology, list) { + for (j = 0; j < p->num_phys; j++) { + if (port_info->phy_info[i].identify.handle == + p->phy_info[j].attached.handle) + parent = &p->phy_info[j].rphy->dev; + } + } + + mptsas_probe_one_phy(parent, &port_info->phy_info[i], *index); + (*index)++; + } + + return 0; + + out_free_port_info: + kfree(port_info); + out: + return error; +} + +static void +mptsas_scan_sas_topology(MPT_ADAPTER *ioc) +{ + u32 handle = 0xFFFF; + int index = 0; + + mptsas_probe_hba_phys(ioc, &index); + while (!mptsas_probe_expander_phys(ioc, &handle, &index)) + ; +} + +static int +mptsas_probe(struct pci_dev *pdev, const struct pci_device_id *id) +{ + struct Scsi_Host *sh; + MPT_SCSI_HOST *hd; + MPT_ADAPTER *ioc; + unsigned long flags; + int sz, ii; + int numSGE = 0; + int scale; + int ioc_cap; + u8 *mem; + int error=0; + int r; + + r = mpt_attach(pdev,id); + if (r) + return r; + + ioc = pci_get_drvdata(pdev); + ioc->DoneCtx = mptsasDoneCtx; + ioc->TaskCtx = mptsasTaskCtx; + ioc->InternalCtx = mptsasInternalCtx; + + /* Added sanity check on readiness of the MPT adapter. + */ + if (ioc->last_state != MPI_IOC_STATE_OPERATIONAL) { + printk(MYIOC_s_WARN_FMT + "Skipping because it's not operational!\n", + ioc->name); + return -ENODEV; + } + + if (!ioc->active) { + printk(MYIOC_s_WARN_FMT "Skipping because it's disabled!\n", + ioc->name); + return -ENODEV; + } + + /* Sanity check - ensure at least 1 port is INITIATOR capable + */ + ioc_cap = 0; + for (ii = 0; ii < ioc->facts.NumberOfPorts; ii++) { + if (ioc->pfacts[ii].ProtocolFlags & + MPI_PORTFACTS_PROTOCOL_INITIATOR) + ioc_cap++; + } + + if (!ioc_cap) { + printk(MYIOC_s_WARN_FMT + "Skipping ioc=%p because SCSI Initiator mode " + "is NOT enabled!\n", ioc->name, ioc); + return -ENODEV; + } + + sh = scsi_host_alloc(&mptsas_driver_template, sizeof(MPT_SCSI_HOST)); + if (!sh) { + printk(MYIOC_s_WARN_FMT + "Unable to register controller with SCSI subsystem\n", + ioc->name); + return -1; + } + + spin_lock_irqsave(&ioc->FreeQlock, flags); + + /* Attach the SCSI Host to the IOC structure + */ + ioc->sh = sh; + + sh->io_port = 0; + sh->n_io_port = 0; + sh->irq = 0; + + /* set 16 byte cdb's */ + sh->max_cmd_len = 16; + + sh->max_id = ioc->pfacts->MaxDevices + 1; + + sh->transportt = mptsas_transport_template; + + sh->max_lun = MPT_LAST_LUN + 1; + sh->max_channel = 0; + sh->this_id = ioc->pfacts[0].PortSCSIID; + + /* Required entry. + */ + sh->unique_id = ioc->id; + + INIT_LIST_HEAD(&ioc->sas_topology); + + /* Verify that we won't exceed the maximum + * number of chain buffers + * We can optimize: ZZ = req_sz/sizeof(SGE) + * For 32bit SGE's: + * numSGE = 1 + (ZZ-1)*(maxChain -1) + ZZ + * + (req_sz - 64)/sizeof(SGE) + * A slightly different algorithm is required for + * 64bit SGEs. + */ + scale = ioc->req_sz/(sizeof(dma_addr_t) + sizeof(u32)); + if (sizeof(dma_addr_t) == sizeof(u64)) { + numSGE = (scale - 1) * + (ioc->facts.MaxChainDepth-1) + scale + + (ioc->req_sz - 60) / (sizeof(dma_addr_t) + + sizeof(u32)); + } else { + numSGE = 1 + (scale - 1) * + (ioc->facts.MaxChainDepth-1) + scale + + (ioc->req_sz - 64) / (sizeof(dma_addr_t) + + sizeof(u32)); + } + + if (numSGE < sh->sg_tablesize) { + /* Reset this value */ + dprintk((MYIOC_s_INFO_FMT + "Resetting sg_tablesize to %d from %d\n", + ioc->name, numSGE, sh->sg_tablesize)); + sh->sg_tablesize = numSGE; + } + + spin_unlock_irqrestore(&ioc->FreeQlock, flags); + + hd = (MPT_SCSI_HOST *) sh->hostdata; + hd->ioc = ioc; + + /* SCSI needs scsi_cmnd lookup table! + * (with size equal to req_depth*PtrSz!) + */ + sz = ioc->req_depth * sizeof(void *); + mem = kmalloc(sz, GFP_ATOMIC); + if (mem == NULL) { + error = -ENOMEM; + goto mptsas_probe_failed; + } + + memset(mem, 0, sz); + hd->ScsiLookup = (struct scsi_cmnd **) mem; + + dprintk((MYIOC_s_INFO_FMT "ScsiLookup @ %p, sz=%d\n", + ioc->name, hd->ScsiLookup, sz)); + + /* Allocate memory for the device structures. + * A non-Null pointer at an offset + * indicates a device exists. + * max_id = 1 + maximum id (hosts.h) + */ + sz = sh->max_id * sizeof(void *); + mem = kmalloc(sz, GFP_ATOMIC); + if (mem == NULL) { + error = -ENOMEM; + goto mptsas_probe_failed; + } + + memset(mem, 0, sz); + hd->Targets = (VirtDevice **) mem; + + dprintk((KERN_INFO + " Targets @ %p, sz=%d\n", hd->Targets, sz)); + + /* Clear the TM flags + */ + hd->tmPending = 0; + hd->tmState = TM_STATE_NONE; + hd->resetPending = 0; + hd->abortSCpnt = NULL; + + /* Clear the pointer used to store + * single-threaded commands, i.e., those + * issued during a bus scan, dv and + * configuration pages. + */ + hd->cmdPtr = NULL; + + /* Initialize this SCSI Hosts' timers + * To use, set the timer expires field + * and add_timer + */ + init_timer(&hd->timer); + hd->timer.data = (unsigned long) hd; + hd->timer.function = mptscsih_timer_expired; + + hd->mpt_pq_filter = mpt_pq_filter; + ioc->sas_data.ptClear = mpt_pt_clear; + + if (ioc->sas_data.ptClear==1) { + mptbase_sas_persist_operation( + ioc, MPI_SAS_OP_CLEAR_ALL_PERSISTENT); + } + + ddvprintk((MYIOC_s_INFO_FMT + "mpt_pq_filter %x mpt_pq_filter %x\n", + ioc->name, + mpt_pq_filter, + mpt_pq_filter)); + + init_waitqueue_head(&hd->scandv_waitq); + hd->scandv_wait_done = 0; + hd->last_queue_full = 0; + + error = scsi_add_host(sh, &ioc->pcidev->dev); + if (error) { + dprintk((KERN_ERR MYNAM + "scsi_add_host failed\n")); + goto mptsas_probe_failed; + } + + mptsas_scan_sas_topology(ioc); + + return 0; + +mptsas_probe_failed: + + mptscsih_remove(pdev); + return error; +} + +static void __devexit mptsas_remove(struct pci_dev *pdev) +{ + MPT_ADAPTER *ioc = pci_get_drvdata(pdev); + struct mptsas_portinfo *p, *n; + + sas_remove_host(ioc->sh); + + list_for_each_entry_safe(p, n, &ioc->sas_topology, list) { + list_del(&p->list); + kfree(p); + } + + mptscsih_remove(pdev); +} + +static struct pci_device_id mptsas_pci_table[] = { + { PCI_VENDOR_ID_LSI_LOGIC, PCI_DEVICE_ID_LSI_SAS1064, + PCI_ANY_ID, PCI_ANY_ID }, + { PCI_VENDOR_ID_LSI_LOGIC, PCI_DEVICE_ID_LSI_SAS1066, + PCI_ANY_ID, PCI_ANY_ID }, + { PCI_VENDOR_ID_LSI_LOGIC, PCI_DEVICE_ID_LSI_SAS1068, + PCI_ANY_ID, PCI_ANY_ID }, + { PCI_VENDOR_ID_LSI_LOGIC, PCI_DEVICE_ID_LSI_SAS1064E, + PCI_ANY_ID, PCI_ANY_ID }, + { PCI_VENDOR_ID_LSI_LOGIC, PCI_DEVICE_ID_LSI_SAS1066E, + PCI_ANY_ID, PCI_ANY_ID }, + { PCI_VENDOR_ID_LSI_LOGIC, PCI_DEVICE_ID_LSI_SAS1068E, + PCI_ANY_ID, PCI_ANY_ID }, + {0} /* Terminating entry */ +}; +MODULE_DEVICE_TABLE(pci, mptsas_pci_table); + + +static struct pci_driver mptsas_driver = { + .name = "mptsas", + .id_table = mptsas_pci_table, + .probe = mptsas_probe, + .remove = __devexit_p(mptsas_remove), + .shutdown = mptscsih_shutdown, +#ifdef CONFIG_PM + .suspend = mptscsih_suspend, + .resume = mptscsih_resume, +#endif +}; + +static int __init +mptsas_init(void) +{ + show_mptmod_ver(my_NAME, my_VERSION); + + mptsas_transport_template = + sas_attach_transport(&mptsas_transport_functions); + if (!mptsas_transport_template) + return -ENODEV; + + mptsasDoneCtx = mpt_register(mptscsih_io_done, MPTSAS_DRIVER); + mptsasTaskCtx = mpt_register(mptscsih_taskmgmt_complete, MPTSAS_DRIVER); + mptsasInternalCtx = + mpt_register(mptscsih_scandv_complete, MPTSAS_DRIVER); + + if (mpt_event_register(mptsasDoneCtx, mptscsih_event_process) == 0) { + devtprintk((KERN_INFO MYNAM + ": Registered for IOC event notifications\n")); + } + + if (mpt_reset_register(mptsasDoneCtx, mptscsih_ioc_reset) == 0) { + dprintk((KERN_INFO MYNAM + ": Registered for IOC reset notifications\n")); + } + + return pci_register_driver(&mptsas_driver); +} + +static void __exit +mptsas_exit(void) +{ + pci_unregister_driver(&mptsas_driver); + sas_release_transport(mptsas_transport_template); + + mpt_reset_deregister(mptsasDoneCtx); + mpt_event_deregister(mptsasDoneCtx); + + mpt_deregister(mptsasInternalCtx); + mpt_deregister(mptsasTaskCtx); + mpt_deregister(mptsasDoneCtx); +} + +module_init(mptsas_init); +module_exit(mptsas_exit); From 466544d8898fc87ed6e2e62ac14af7c50ab7a1a4 Mon Sep 17 00:00:00 2001 From: "Moore, Eric Dean" Date: Wed, 14 Sep 2005 18:09:10 -0600 Subject: [PATCH 166/240] [SCSI] fusion SAS support (mptsas driver) updates Summary of Changes: * splitting mpt_interrupt per Christophs suggestion about a month ago * rename ScsiCfgData to SpiCfgData structure, then move all the raid related info into new structure called RaidCfgData. This is done because SAS supports RAID, as well as SPI, so the raid stuff should be seperate. * incorrect timeout calculation for cntdn inside WaitForDoorbellAck and WaitForDoortbellInt * add support for interpreting SAS Log Info * Increase Event Log Size from 0xA to 0x32 * Fix bug in mptsas/mptfc/mptspi - when controller has Initiator Mode Disabled, and only running in TargetMode, the mptctl would panic when loading. The fix is to return 0, instead of -ENODEV, in SCSI LLD respective probe routines * Fix bug in mptlan.c - driver will panic if there is host reset, due to dev being set to zero in mpt_lan_ioc_reset * Fix's for SPI - Echo Buffer * Several fix's in mptscsih_io_done - FCP Response info, RESIDUAL_MISMATCH, Data Underrun, etc. * Cleanup Error Handling - EH handlers, mptscsih_flush_cmds, and zeroing out ScsiLookup from mptscsih_qcmd * Cleanup asyn event handling from mptscsih -> mptscsih_event_process. Also added support for SAS Persistent Table Full, an asyn event Signed-off-by: Eric Moore Signed-off-by: James Bottomley --- drivers/message/fusion/mptbase.c | 449 +++++++++++++++++++----------- drivers/message/fusion/mptbase.h | 28 +- drivers/message/fusion/mptctl.c | 4 +- drivers/message/fusion/mptfc.c | 2 +- drivers/message/fusion/mptlan.c | 7 +- drivers/message/fusion/mptsas.c | 2 +- drivers/message/fusion/mptscsih.c | 447 +++++++++++++++-------------- drivers/message/fusion/mptscsih.h | 8 +- drivers/message/fusion/mptspi.c | 2 +- 9 files changed, 549 insertions(+), 400 deletions(-) diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c index 14d62d96ca41..790a2932ded9 100644 --- a/drivers/message/fusion/mptbase.c +++ b/drivers/message/fusion/mptbase.c @@ -135,7 +135,6 @@ static void mpt_adapter_dispose(MPT_ADAPTER *ioc); static void MptDisplayIocCapabilities(MPT_ADAPTER *ioc); static int MakeIocReady(MPT_ADAPTER *ioc, int force, int sleepFlag); -//static u32 mpt_GetIocState(MPT_ADAPTER *ioc, int cooked); static int GetIocFacts(MPT_ADAPTER *ioc, int sleepFlag, int reason); static int GetPortFacts(MPT_ADAPTER *ioc, int portnum, int sleepFlag); static int SendIocInit(MPT_ADAPTER *ioc, int sleepFlag); @@ -178,6 +177,7 @@ static int ProcessEventNotification(MPT_ADAPTER *ioc, EventNotificationReply_t * static void mpt_sp_ioc_info(MPT_ADAPTER *ioc, u32 ioc_status, MPT_FRAME_HDR *mf); static void mpt_fc_log_info(MPT_ADAPTER *ioc, u32 log_info); static void mpt_sp_log_info(MPT_ADAPTER *ioc, u32 log_info); +static void mpt_sas_log_info(MPT_ADAPTER *ioc, u32 log_info); /* module entry point */ static int __init fusion_init (void); @@ -209,6 +209,144 @@ pci_enable_io_access(struct pci_dev *pdev) pci_write_config_word(pdev, PCI_COMMAND, command_reg); } +/* + * Process turbo (context) reply... + */ +static void +mpt_turbo_reply(MPT_ADAPTER *ioc, u32 pa) +{ + MPT_FRAME_HDR *mf = NULL; + MPT_FRAME_HDR *mr = NULL; + int req_idx = 0; + int cb_idx; + + dmfprintk((MYIOC_s_INFO_FMT "Got TURBO reply req_idx=%08x\n", + ioc->name, pa)); + + switch (pa >> MPI_CONTEXT_REPLY_TYPE_SHIFT) { + case MPI_CONTEXT_REPLY_TYPE_SCSI_INIT: + req_idx = pa & 0x0000FFFF; + cb_idx = (pa & 0x00FF0000) >> 16; + mf = MPT_INDEX_2_MFPTR(ioc, req_idx); + break; + case MPI_CONTEXT_REPLY_TYPE_LAN: + cb_idx = mpt_lan_index; + /* + * Blind set of mf to NULL here was fatal + * after lan_reply says "freeme" + * Fix sort of combined with an optimization here; + * added explicit check for case where lan_reply + * was just returning 1 and doing nothing else. + * For this case skip the callback, but set up + * proper mf value first here:-) + */ + if ((pa & 0x58000000) == 0x58000000) { + req_idx = pa & 0x0000FFFF; + mf = MPT_INDEX_2_MFPTR(ioc, req_idx); + mpt_free_msg_frame(ioc, mf); + mb(); + return; + break; + } + mr = (MPT_FRAME_HDR *) CAST_U32_TO_PTR(pa); + break; + case MPI_CONTEXT_REPLY_TYPE_SCSI_TARGET: + cb_idx = mpt_stm_index; + mr = (MPT_FRAME_HDR *) CAST_U32_TO_PTR(pa); + break; + default: + cb_idx = 0; + BUG(); + } + + /* Check for (valid) IO callback! */ + if (cb_idx < 1 || cb_idx >= MPT_MAX_PROTOCOL_DRIVERS || + MptCallbacks[cb_idx] == NULL) { + printk(MYIOC_s_WARN_FMT "%s: Invalid cb_idx (%d)!\n", + __FUNCTION__, ioc->name, cb_idx); + goto out; + } + + if (MptCallbacks[cb_idx](ioc, mf, mr)) + mpt_free_msg_frame(ioc, mf); + out: + mb(); +} + +static void +mpt_reply(MPT_ADAPTER *ioc, u32 pa) +{ + MPT_FRAME_HDR *mf; + MPT_FRAME_HDR *mr; + int req_idx; + int cb_idx; + int freeme; + + u32 reply_dma_low; + u16 ioc_stat; + + /* non-TURBO reply! Hmmm, something may be up... + * Newest turbo reply mechanism; get address + * via left shift 1 (get rid of MPI_ADDRESS_REPLY_A_BIT)! + */ + + /* Map DMA address of reply header to cpu address. + * pa is 32 bits - but the dma address may be 32 or 64 bits + * get offset based only only the low addresses + */ + + reply_dma_low = (pa <<= 1); + mr = (MPT_FRAME_HDR *)((u8 *)ioc->reply_frames + + (reply_dma_low - ioc->reply_frames_low_dma)); + + req_idx = le16_to_cpu(mr->u.frame.hwhdr.msgctxu.fld.req_idx); + cb_idx = mr->u.frame.hwhdr.msgctxu.fld.cb_idx; + mf = MPT_INDEX_2_MFPTR(ioc, req_idx); + + dmfprintk((MYIOC_s_INFO_FMT "Got non-TURBO reply=%p req_idx=%x cb_idx=%x Function=%x\n", + ioc->name, mr, req_idx, cb_idx, mr->u.hdr.Function)); + DBG_DUMP_REPLY_FRAME(mr) + + /* Check/log IOC log info + */ + ioc_stat = le16_to_cpu(mr->u.reply.IOCStatus); + if (ioc_stat & MPI_IOCSTATUS_FLAG_LOG_INFO_AVAILABLE) { + u32 log_info = le32_to_cpu(mr->u.reply.IOCLogInfo); + if (ioc->bus_type == FC) + mpt_fc_log_info(ioc, log_info); + else if (ioc->bus_type == SCSI) + mpt_sp_log_info(ioc, log_info); + else if (ioc->bus_type == SAS) + mpt_sas_log_info(ioc, log_info); + } + if (ioc_stat & MPI_IOCSTATUS_MASK) { + if (ioc->bus_type == SCSI && + cb_idx != mpt_stm_index && + cb_idx != mpt_lan_index) + mpt_sp_ioc_info(ioc, (u32)ioc_stat, mf); + } + + + /* Check for (valid) IO callback! */ + if (cb_idx < 1 || cb_idx >= MPT_MAX_PROTOCOL_DRIVERS || + MptCallbacks[cb_idx] == NULL) { + printk(MYIOC_s_WARN_FMT "%s: Invalid cb_idx (%d)!\n", + __FUNCTION__, ioc->name, cb_idx); + freeme = 0; + goto out; + } + + freeme = MptCallbacks[cb_idx](ioc, mf, mr); + + out: + /* Flush (non-TURBO) reply with a WRITE! */ + CHIPREG_WRITE32(&ioc->chip->ReplyFifo, pa); + + if (freeme) + mpt_free_msg_frame(ioc, mf); + mb(); +} + /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /* * mpt_interrupt - MPT adapter (IOC) specific interrupt handler. @@ -230,164 +368,21 @@ pci_enable_io_access(struct pci_dev *pdev) static irqreturn_t mpt_interrupt(int irq, void *bus_id, struct pt_regs *r) { - MPT_ADAPTER *ioc; - MPT_FRAME_HDR *mf; - MPT_FRAME_HDR *mr; - u32 pa; - int req_idx; - int cb_idx; - int type; - int freeme; - - ioc = (MPT_ADAPTER *)bus_id; + MPT_ADAPTER *ioc = bus_id; + u32 pa; /* * Drain the reply FIFO! - * - * NOTES: I've seen up to 10 replies processed in this loop, so far... - * Update: I've seen up to 9182 replies processed in this loop! ?? - * Update: Limit ourselves to processing max of N replies - * (bottom of loop). */ while (1) { - - if ((pa = CHIPREG_READ32_dmasync(&ioc->chip->ReplyFifo)) == 0xFFFFFFFF) + pa = CHIPREG_READ32_dmasync(&ioc->chip->ReplyFifo); + if (pa == 0xFFFFFFFF) return IRQ_HANDLED; - - cb_idx = 0; - freeme = 0; - - /* - * Check for non-TURBO reply! - */ - if (pa & MPI_ADDRESS_REPLY_A_BIT) { - u32 reply_dma_low; - u16 ioc_stat; - - /* non-TURBO reply! Hmmm, something may be up... - * Newest turbo reply mechanism; get address - * via left shift 1 (get rid of MPI_ADDRESS_REPLY_A_BIT)! - */ - - /* Map DMA address of reply header to cpu address. - * pa is 32 bits - but the dma address may be 32 or 64 bits - * get offset based only only the low addresses - */ - reply_dma_low = (pa = (pa << 1)); - mr = (MPT_FRAME_HDR *)((u8 *)ioc->reply_frames + - (reply_dma_low - ioc->reply_frames_low_dma)); - - req_idx = le16_to_cpu(mr->u.frame.hwhdr.msgctxu.fld.req_idx); - cb_idx = mr->u.frame.hwhdr.msgctxu.fld.cb_idx; - mf = MPT_INDEX_2_MFPTR(ioc, req_idx); - - dmfprintk((MYIOC_s_INFO_FMT "Got non-TURBO reply=%p req_idx=%x cb_idx=%x Function=%x\n", - ioc->name, mr, req_idx, cb_idx, mr->u.hdr.Function)); - DBG_DUMP_REPLY_FRAME(mr) - - /* Check/log IOC log info - */ - ioc_stat = le16_to_cpu(mr->u.reply.IOCStatus); - if (ioc_stat & MPI_IOCSTATUS_FLAG_LOG_INFO_AVAILABLE) { - u32 log_info = le32_to_cpu(mr->u.reply.IOCLogInfo); - if (ioc->bus_type == FC) - mpt_fc_log_info(ioc, log_info); - else if (ioc->bus_type == SCSI) - mpt_sp_log_info(ioc, log_info); - } - if (ioc_stat & MPI_IOCSTATUS_MASK) { - if (ioc->bus_type == SCSI) - mpt_sp_ioc_info(ioc, (u32)ioc_stat, mf); - } - } else { - /* - * Process turbo (context) reply... - */ - dmfprintk((MYIOC_s_INFO_FMT "Got TURBO reply req_idx=%08x\n", ioc->name, pa)); - type = (pa >> MPI_CONTEXT_REPLY_TYPE_SHIFT); - if (type == MPI_CONTEXT_REPLY_TYPE_SCSI_TARGET) { - cb_idx = mpt_stm_index; - mf = NULL; - mr = (MPT_FRAME_HDR *) CAST_U32_TO_PTR(pa); - } else if (type == MPI_CONTEXT_REPLY_TYPE_LAN) { - cb_idx = mpt_lan_index; - /* Blind set of mf to NULL here was fatal - * after lan_reply says "freeme" - * Fix sort of combined with an optimization here; - * added explicit check for case where lan_reply - * was just returning 1 and doing nothing else. - * For this case skip the callback, but set up - * proper mf value first here:-) - */ - if ((pa & 0x58000000) == 0x58000000) { - req_idx = pa & 0x0000FFFF; - mf = MPT_INDEX_2_MFPTR(ioc, req_idx); - freeme = 1; - /* - * IMPORTANT! Invalidate the callback! - */ - cb_idx = 0; - } else { - mf = NULL; - } - mr = (MPT_FRAME_HDR *) CAST_U32_TO_PTR(pa); - } else { - req_idx = pa & 0x0000FFFF; - cb_idx = (pa & 0x00FF0000) >> 16; - mf = MPT_INDEX_2_MFPTR(ioc, req_idx); - mr = NULL; - } - pa = 0; /* No reply flush! */ - } - -#ifdef MPT_DEBUG_IRQ - if (ioc->bus_type == SCSI) { - /* Verify mf, mr are reasonable. - */ - if ((mf) && ((mf >= MPT_INDEX_2_MFPTR(ioc, ioc->req_depth)) - || (mf < ioc->req_frames)) ) { - printk(MYIOC_s_WARN_FMT - "mpt_interrupt: Invalid mf (%p)!\n", ioc->name, (void *)mf); - cb_idx = 0; - pa = 0; - freeme = 0; - } - if ((pa) && (mr) && ((mr >= MPT_INDEX_2_RFPTR(ioc, ioc->req_depth)) - || (mr < ioc->reply_frames)) ) { - printk(MYIOC_s_WARN_FMT - "mpt_interrupt: Invalid rf (%p)!\n", ioc->name, (void *)mr); - cb_idx = 0; - pa = 0; - freeme = 0; - } - if (cb_idx > (MPT_MAX_PROTOCOL_DRIVERS-1)) { - printk(MYIOC_s_WARN_FMT - "mpt_interrupt: Invalid cb_idx (%d)!\n", ioc->name, cb_idx); - cb_idx = 0; - pa = 0; - freeme = 0; - } - } -#endif - - /* Check for (valid) IO callback! */ - if (cb_idx) { - /* Do the callback! */ - freeme = (*(MptCallbacks[cb_idx]))(ioc, mf, mr); - } - - if (pa) { - /* Flush (non-TURBO) reply with a WRITE! */ - CHIPREG_WRITE32(&ioc->chip->ReplyFifo, pa); - } - - if (freeme) { - /* Put Request back on FreeQ! */ - mpt_free_msg_frame(ioc, mf); - } - - mb(); - } /* drain reply FIFO */ + else if (pa & MPI_ADDRESS_REPLY_A_BIT) + mpt_reply(ioc, pa); + else + mpt_turbo_reply(ioc, pa); + } return IRQ_HANDLED; } @@ -1065,7 +1060,7 @@ mpt_host_page_alloc(MPT_ADAPTER *ioc, pIOCInit_t ioc_init) ioc->name, ioc->HostPageBuffer, ioc->HostPageBuffer_dma, - hst_page_buffer_sz)); + host_page_buffer_sz)); ioc->alloc_total += host_page_buffer_sz; ioc->HostPageBuffer_sz = host_page_buffer_sz; break; @@ -1208,7 +1203,7 @@ mpt_attach(struct pci_dev *pdev, const struct pci_device_id *id) /* Initilize SCSI Config Data structure */ - memset(&ioc->spi_data, 0, sizeof(ScsiCfgData)); + memset(&ioc->spi_data, 0, sizeof(SpiCfgData)); /* Initialize the running configQ head. */ @@ -1755,8 +1750,23 @@ mpt_do_ioc_recovery(MPT_ADAPTER *ioc, u32 reason, int sleepFlag) */ if (ret == 0) { rc = mpt_do_upload(ioc, sleepFlag); - if (rc != 0) + if (rc == 0) { + if (ioc->alt_ioc && ioc->alt_ioc->cached_fw) { + /* + * Maintain only one pointer to FW memory + * so there will not be two attempt to + * downloadboot onboard dual function + * chips (mpt_adapter_disable, + * mpt_diag_reset) + */ + ioc->cached_fw = NULL; + ddlprintk((MYIOC_s_INFO_FMT ": mpt_upload: alt_%s has cached_fw=%p \n", + ioc->name, ioc->alt_ioc->name, ioc->alt_ioc->cached_fw)); + } + } else { printk(KERN_WARNING MYNAM ": firmware upload failure!\n"); + ret = -5; + } } } } @@ -1997,9 +2007,9 @@ mpt_adapter_disable(MPT_ADAPTER *ioc) } kfree(ioc->spi_data.nvram); - kfree(ioc->spi_data.pIocPg3); + kfree(ioc->raid_data.pIocPg3); ioc->spi_data.nvram = NULL; - ioc->spi_data.pIocPg3 = NULL; + ioc->raid_data.pIocPg3 = NULL; if (ioc->spi_data.pIocPg4 != NULL) { sz = ioc->spi_data.IocPg4Sz; @@ -3852,7 +3862,7 @@ WaitForDoorbellAck(MPT_ADAPTER *ioc, int howlong, int sleepFlag) int count = 0; u32 intstat=0; - cntdn = ((sleepFlag == CAN_SLEEP) ? HZ : 1000) * howlong; + cntdn = 1000 * howlong; if (sleepFlag == CAN_SLEEP) { while (--cntdn) { @@ -3902,7 +3912,7 @@ WaitForDoorbellInt(MPT_ADAPTER *ioc, int howlong, int sleepFlag) int count = 0; u32 intstat=0; - cntdn = ((sleepFlag == CAN_SLEEP) ? HZ : 1000) * howlong; + cntdn = 1000 * howlong; if (sleepFlag == CAN_SLEEP) { while (--cntdn) { intstat = CHIPREG_READ32(&ioc->chip->IntStatus); @@ -4634,10 +4644,10 @@ mpt_findImVolumes(MPT_ADAPTER *ioc) if (mpt_config(ioc, &cfg) != 0) goto done_and_free; - if ( (mem = (u8 *)ioc->spi_data.pIocPg2) == NULL ) { + if ( (mem = (u8 *)ioc->raid_data.pIocPg2) == NULL ) { mem = kmalloc(iocpage2sz, GFP_ATOMIC); if (mem) { - ioc->spi_data.pIocPg2 = (IOCPage2_t *) mem; + ioc->raid_data.pIocPg2 = (IOCPage2_t *) mem; } else { goto done_and_free; } @@ -4654,7 +4664,7 @@ mpt_findImVolumes(MPT_ADAPTER *ioc) /* At least 1 RAID Volume */ pIocRv = pIoc2->RaidVolume; - ioc->spi_data.isRaid = 0; + ioc->raid_data.isRaid = 0; for (jj = 0; jj < nVols; jj++, pIocRv++) { vid = pIocRv->VolumeID; vbus = pIocRv->VolumeBus; @@ -4663,7 +4673,7 @@ mpt_findImVolumes(MPT_ADAPTER *ioc) /* find the match */ if (vbus == 0) { - ioc->spi_data.isRaid |= (1 << vid); + ioc->raid_data.isRaid |= (1 << vid); } else { /* Error! Always bus 0 */ @@ -4698,8 +4708,8 @@ mpt_read_ioc_pg_3(MPT_ADAPTER *ioc) /* Free the old page */ - kfree(ioc->spi_data.pIocPg3); - ioc->spi_data.pIocPg3 = NULL; + kfree(ioc->raid_data.pIocPg3); + ioc->raid_data.pIocPg3 = NULL; /* There is at least one physical disk. * Read and save IOC Page 3 @@ -4736,7 +4746,7 @@ mpt_read_ioc_pg_3(MPT_ADAPTER *ioc) mem = kmalloc(iocpage3sz, GFP_ATOMIC); if (mem) { memcpy(mem, (u8 *)pIoc3, iocpage3sz); - ioc->spi_data.pIocPg3 = (IOCPage3_t *) mem; + ioc->raid_data.pIocPg3 = (IOCPage3_t *) mem; } } @@ -6022,6 +6032,111 @@ mpt_sp_log_info(MPT_ADAPTER *ioc, u32 log_info) printk(MYIOC_s_INFO_FMT "LogInfo(0x%08x): F/W: %s\n", ioc->name, log_info, desc); } +/* strings for sas loginfo */ + static char *originator_str[] = { + "IOP", /* 00h */ + "PL", /* 01h */ + "IR" /* 02h */ + }; + static char *iop_code_str[] = { + NULL, /* 00h */ + "Invalid SAS Address", /* 01h */ + NULL, /* 02h */ + "Invalid Page", /* 03h */ + NULL, /* 04h */ + "Task Terminated" /* 05h */ + }; + static char *pl_code_str[] = { + NULL, /* 00h */ + "Open Failure", /* 01h */ + "Invalid Scatter Gather List", /* 02h */ + "Wrong Relative Offset or Frame Length", /* 03h */ + "Frame Transfer Error", /* 04h */ + "Transmit Frame Connected Low", /* 05h */ + "SATA Non-NCQ RW Error Bit Set", /* 06h */ + "SATA Read Log Receive Data Error", /* 07h */ + "SATA NCQ Fail All Commands After Error", /* 08h */ + "SATA Error in Receive Set Device Bit FIS", /* 09h */ + "Receive Frame Invalid Message", /* 0Ah */ + "Receive Context Message Valid Error", /* 0Bh */ + "Receive Frame Current Frame Error", /* 0Ch */ + "SATA Link Down", /* 0Dh */ + "Discovery SATA Init W IOS", /* 0Eh */ + "Config Invalid Page", /* 0Fh */ + "Discovery SATA Init Timeout", /* 10h */ + "Reset", /* 11h */ + "Abort", /* 12h */ + "IO Not Yet Executed", /* 13h */ + "IO Executed", /* 14h */ + NULL, /* 15h */ + NULL, /* 16h */ + NULL, /* 17h */ + NULL, /* 18h */ + NULL, /* 19h */ + NULL, /* 1Ah */ + NULL, /* 1Bh */ + NULL, /* 1Ch */ + NULL, /* 1Dh */ + NULL, /* 1Eh */ + NULL, /* 1Fh */ + "Enclosure Management" /* 20h */ + }; + +/*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ +/* + * mpt_sas_log_info - Log information returned from SAS IOC. + * @ioc: Pointer to MPT_ADAPTER structure + * @log_info: U32 LogInfo reply word from the IOC + * + * Refer to lsi/mpi_log_sas.h. + */ +static void +mpt_sas_log_info(MPT_ADAPTER *ioc, u32 log_info) +{ +union loginfo_type { + u32 loginfo; + struct { + u32 subcode:16; + u32 code:8; + u32 originator:4; + u32 bus_type:4; + }dw; +}; + union loginfo_type sas_loginfo; + char *code_desc = NULL; + + sas_loginfo.loginfo = log_info; + if ((sas_loginfo.dw.bus_type != 3 /*SAS*/) && + (sas_loginfo.dw.originator < sizeof(originator_str)/sizeof(char*))) + return; + if ((sas_loginfo.dw.originator == 0 /*IOP*/) && + (sas_loginfo.dw.code < sizeof(iop_code_str)/sizeof(char*))) { + code_desc = iop_code_str[sas_loginfo.dw.code]; + }else if ((sas_loginfo.dw.originator == 1 /*PL*/) && + (sas_loginfo.dw.code < sizeof(pl_code_str)/sizeof(char*) )) { + code_desc = pl_code_str[sas_loginfo.dw.code]; + } + + if (code_desc != NULL) + printk(MYIOC_s_INFO_FMT + "LogInfo(0x%08x): Originator={%s}, Code={%s}," + " SubCode(0x%04x)\n", + ioc->name, + log_info, + originator_str[sas_loginfo.dw.originator], + code_desc, + sas_loginfo.dw.subcode); + else + printk(MYIOC_s_INFO_FMT + "LogInfo(0x%08x): Originator={%s}, Code=(0x%02x)," + " SubCode(0x%04x)\n", + ioc->name, + log_info, + originator_str[sas_loginfo.dw.originator], + sas_loginfo.dw.code, + sas_loginfo.dw.subcode); +} + /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /* * mpt_sp_ioc_info - IOC information returned from SCSI Parallel IOC. diff --git a/drivers/message/fusion/mptbase.h b/drivers/message/fusion/mptbase.h index bbd21d74ce5c..75105277e22f 100644 --- a/drivers/message/fusion/mptbase.h +++ b/drivers/message/fusion/mptbase.h @@ -77,8 +77,8 @@ #define COPYRIGHT "Copyright (c) 1999-2005 " MODULEAUTHOR #endif -#define MPT_LINUX_VERSION_COMMON "3.03.02" -#define MPT_LINUX_PACKAGE_NAME "@(#)mptlinux-3.03.02" +#define MPT_LINUX_VERSION_COMMON "3.03.03" +#define MPT_LINUX_PACKAGE_NAME "@(#)mptlinux-3.03.03" #define WHAT_MAGIC_STRING "@" "(" "#" ")" #define show_mptmod_ver(s,ver) \ @@ -424,7 +424,7 @@ typedef struct _MPT_IOCTL { /* * Event Structure and define */ -#define MPTCTL_EVENT_LOG_SIZE (0x0000000A) +#define MPTCTL_EVENT_LOG_SIZE (0x000000032) typedef struct _mpt_ioctl_events { u32 event; /* Specified by define above */ u32 eventContext; /* Index or counter */ @@ -452,16 +452,13 @@ typedef struct _mpt_ioctl_events { #define MPT_SCSICFG_ALL_IDS 0x02 /* WriteSDP1 to all IDS */ /* #define MPT_SCSICFG_BLK_NEGO 0x10 WriteSDP1 with WDTR and SDTR disabled */ -typedef struct _ScsiCfgData { +typedef struct _SpiCfgData { u32 PortFlags; int *nvram; /* table of device NVRAM values */ - IOCPage2_t *pIocPg2; /* table of Raid Volumes */ - IOCPage3_t *pIocPg3; /* table of physical disks */ IOCPage4_t *pIocPg4; /* SEP devices addressing */ dma_addr_t IocPg4_dma; /* Phys Addr of IOCPage4 data */ int IocPg4Sz; /* IOCPage4 size */ u8 dvStatus[MPT_MAX_SCSI_DEVICES]; - int isRaid; /* bit field, 1 if RAID */ u8 minSyncFactor; /* 0xFF if async */ u8 maxSyncOffset; /* 0 if async */ u8 maxBusWidth; /* 0 if narrow, 1 if wide */ @@ -473,10 +470,14 @@ typedef struct _ScsiCfgData { u8 dvScheduled; /* 1 if scheduled */ u8 forceDv; /* 1 to force DV scheduling */ u8 noQas; /* Disable QAS for this adapter */ - u8 Saf_Te; /* 1 to force all Processors as SAF-TE if Inquiry data length is too short to check for SAF-TE */ + u8 Saf_Te; /* 1 to force all Processors as + * SAF-TE if Inquiry data length + * is too short to check for SAF-TE + */ u8 mpt_dv; /* command line option: enhanced=1, basic=0 */ + u8 bus_reset; /* 1 to allow bus reset */ u8 rsvd[1]; -} ScsiCfgData; +}SpiCfgData; typedef struct _SasCfgData { u8 ptClear; /* 1 to automatically clear the @@ -486,6 +487,12 @@ typedef struct _SasCfgData { */ }SasCfgData; +typedef struct _RaidCfgData { + IOCPage2_t *pIocPg2; /* table of Raid Volumes */ + IOCPage3_t *pIocPg3; /* table of physical disks */ + int isRaid; /* bit field, 1 if RAID */ +}RaidCfgData; + /* * Adapter Structure - pci_dev specific. Maximum: MPT_MAX_ADAPTERS */ @@ -546,7 +553,8 @@ typedef struct _MPT_ADAPTER struct pci_dev *pcidev; /* struct pci_dev pointer */ u8 __iomem *memmap; /* mmap address */ struct Scsi_Host *sh; /* Scsi Host pointer */ - ScsiCfgData spi_data; /* Scsi config. data */ + SpiCfgData spi_data; /* Scsi config. data */ + RaidCfgData raid_data; /* Raid config. data */ SasCfgData sas_data; /* Sas config. data */ MPT_IOCTL *ioctl; /* ioctl data pointer */ struct proc_dir_entry *ioc_dentry; diff --git a/drivers/message/fusion/mptctl.c b/drivers/message/fusion/mptctl.c index 7577c2417e2e..cb2d59d5f5af 100644 --- a/drivers/message/fusion/mptctl.c +++ b/drivers/message/fusion/mptctl.c @@ -1326,7 +1326,7 @@ mptctl_gettargetinfo (unsigned long arg) */ if (hd && hd->Targets) { mpt_findImVolumes(ioc); - pIoc2 = ioc->spi_data.pIocPg2; + pIoc2 = ioc->raid_data.pIocPg2; for ( id = 0; id <= max_id; ) { if ( pIoc2 && pIoc2->NumActiveVolumes ) { if ( id == pIoc2->RaidVolume[0].VolumeID ) { @@ -1348,7 +1348,7 @@ mptctl_gettargetinfo (unsigned long arg) --maxWordsLeft; goto next_id; } else { - pIoc3 = ioc->spi_data.pIocPg3; + pIoc3 = ioc->raid_data.pIocPg3; for ( jj = 0; jj < pIoc3->NumPhysDisks; jj++ ) { if ( pIoc3->PhysDisk[jj].PhysDiskID == id ) goto next_id; diff --git a/drivers/message/fusion/mptfc.c b/drivers/message/fusion/mptfc.c index 13771abea13f..a628be9bbbad 100644 --- a/drivers/message/fusion/mptfc.c +++ b/drivers/message/fusion/mptfc.c @@ -189,7 +189,7 @@ mptfc_probe(struct pci_dev *pdev, const struct pci_device_id *id) printk(MYIOC_s_WARN_FMT "Skipping ioc=%p because SCSI Initiator mode is NOT enabled!\n", ioc->name, ioc); - return -ENODEV; + return 0; } sh = scsi_host_alloc(&mptfc_driver_template, sizeof(MPT_SCSI_HOST)); diff --git a/drivers/message/fusion/mptlan.c b/drivers/message/fusion/mptlan.c index 52794be5a95c..ed3c891e388f 100644 --- a/drivers/message/fusion/mptlan.c +++ b/drivers/message/fusion/mptlan.c @@ -312,7 +312,12 @@ static int mpt_lan_ioc_reset(MPT_ADAPTER *ioc, int reset_phase) { struct net_device *dev = ioc->netdev; - struct mpt_lan_priv *priv = netdev_priv(dev); + struct mpt_lan_priv *priv; + + if (dev == NULL) + return(1); + else + priv = netdev_priv(dev); dlprintk((KERN_INFO MYNAM ": IOC %s_reset routed to LAN driver!\n", reset_phase==MPT_IOC_SETUP_RESET ? "setup" : ( diff --git a/drivers/message/fusion/mptsas.c b/drivers/message/fusion/mptsas.c index 0d9a192e1bd4..429820e48c69 100644 --- a/drivers/message/fusion/mptsas.c +++ b/drivers/message/fusion/mptsas.c @@ -980,7 +980,7 @@ mptsas_probe(struct pci_dev *pdev, const struct pci_device_id *id) printk(MYIOC_s_WARN_FMT "Skipping ioc=%p because SCSI Initiator mode " "is NOT enabled!\n", ioc->name, ioc); - return -ENODEV; + return 0; } sh = scsi_host_alloc(&mptsas_driver_template, sizeof(MPT_SCSI_HOST)); diff --git a/drivers/message/fusion/mptscsih.c b/drivers/message/fusion/mptscsih.c index 58b5fdee009a..8dd25aac5355 100644 --- a/drivers/message/fusion/mptscsih.c +++ b/drivers/message/fusion/mptscsih.c @@ -93,8 +93,9 @@ typedef struct _BIG_SENSE_BUF { #define MPT_ICFLAG_BUF_CAP 0x01 /* ReadBuffer Read Capacity format */ #define MPT_ICFLAG_ECHO 0x02 /* ReadBuffer Echo buffer format */ -#define MPT_ICFLAG_PHYS_DISK 0x04 /* Any SCSI IO but do Phys Disk Format */ -#define MPT_ICFLAG_TAGGED_CMD 0x08 /* Do tagged IO */ +#define MPT_ICFLAG_EBOS 0x04 /* ReadBuffer Echo buffer has EBOS */ +#define MPT_ICFLAG_PHYS_DISK 0x08 /* Any SCSI IO but do Phys Disk Format */ +#define MPT_ICFLAG_TAGGED_CMD 0x10 /* Do tagged IO */ #define MPT_ICFLAG_DID_RESET 0x20 /* Bus Reset occurred with this command */ #define MPT_ICFLAG_RESERVED 0x40 /* Reserved has been issued */ @@ -159,6 +160,8 @@ int mptscsih_scandv_complete(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR static int mptscsih_do_cmd(MPT_SCSI_HOST *hd, INTERNAL_CMD *iocmd); static int mptscsih_synchronize_cache(MPT_SCSI_HOST *hd, int portnum); +static struct work_struct mptscsih_persistTask; + #ifdef MPTSCSIH_ENABLE_DOMAIN_VALIDATION static int mptscsih_do_raid(MPT_SCSI_HOST *hd, u8 action, INTERNAL_CMD *io); static void mptscsih_domainValidation(void *hd); @@ -167,6 +170,7 @@ static void mptscsih_qas_check(MPT_SCSI_HOST *hd, int id); static int mptscsih_doDv(MPT_SCSI_HOST *hd, int channel, int target); static void mptscsih_dv_parms(MPT_SCSI_HOST *hd, DVPARAMETERS *dv,void *pPage); static void mptscsih_fillbuf(char *buffer, int size, int index, int width); +static void mptscsih_set_dvflags_raid(MPT_SCSI_HOST *hd, int id); #endif void mptscsih_remove(struct pci_dev *); @@ -606,11 +610,24 @@ mptscsih_io_done(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR *mr) xfer_cnt = le32_to_cpu(pScsiReply->TransferCount); sc->resid = sc->request_bufflen - xfer_cnt; + /* + * if we get a data underrun indication, yet no data was + * transferred and the SCSI status indicates that the + * command was never started, change the data underrun + * to success + */ + if (status == MPI_IOCSTATUS_SCSI_DATA_UNDERRUN && xfer_cnt == 0 && + (scsi_status == MPI_SCSI_STATUS_BUSY || + scsi_status == MPI_SCSI_STATUS_RESERVATION_CONFLICT || + scsi_status == MPI_SCSI_STATUS_TASK_SET_FULL)) { + status = MPI_IOCSTATUS_SUCCESS; + } + dreplyprintk((KERN_NOTICE "Reply ha=%d id=%d lun=%d:\n" "IOCStatus=%04xh SCSIState=%02xh SCSIStatus=%02xh\n" "resid=%d bufflen=%d xfer_cnt=%d\n", ioc->id, pScsiReq->TargetID, pScsiReq->LUN[1], - status, scsi_state, scsi_status, sc->resid, + status, scsi_state, scsi_status, sc->resid, sc->request_bufflen, xfer_cnt)); if (scsi_state & MPI_SCSI_STATE_AUTOSENSE_VALID) @@ -619,8 +636,11 @@ mptscsih_io_done(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR *mr) /* * Look for + dump FCP ResponseInfo[]! */ - if (scsi_state & MPI_SCSI_STATE_RESPONSE_INFO_VALID) { - printk(KERN_NOTICE " FCP_ResponseInfo=%08xh\n", + if (scsi_state & MPI_SCSI_STATE_RESPONSE_INFO_VALID && + pScsiReply->ResponseInfo) { + printk(KERN_NOTICE "ha=%d id=%d lun=%d: " + "FCP_ResponseInfo=%08xh\n", + ioc->id, pScsiReq->TargetID, pScsiReq->LUN[1], le32_to_cpu(pScsiReply->ResponseInfo)); } @@ -661,23 +681,13 @@ mptscsih_io_done(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR *mr) break; case MPI_IOCSTATUS_SCSI_RESIDUAL_MISMATCH: /* 0x0049 */ - if ( xfer_cnt >= sc->underflow ) { - /* Sufficient data transfer occurred */ + sc->resid = sc->request_bufflen - xfer_cnt; + if((xfer_cnt==0)||(sc->underflow > xfer_cnt)) + sc->result=DID_SOFT_ERROR << 16; + else /* Sufficient data transfer occurred */ sc->result = (DID_OK << 16) | scsi_status; - } else if ( xfer_cnt == 0 ) { - /* A CRC Error causes this condition; retry */ - sc->result = (DRIVER_SENSE << 24) | (DID_OK << 16) | - (CHECK_CONDITION << 1); - sc->sense_buffer[0] = 0x70; - sc->sense_buffer[2] = NO_SENSE; - sc->sense_buffer[12] = 0; - sc->sense_buffer[13] = 0; - } else { - sc->result = DID_SOFT_ERROR << 16; - } - dreplyprintk((KERN_NOTICE - "RESIDUAL_MISMATCH: result=%x on id=%d\n", - sc->result, sc->device->id)); + dreplyprintk((KERN_NOTICE + "RESIDUAL_MISMATCH: result=%x on id=%d\n", sc->result, sc->device->id)); break; case MPI_IOCSTATUS_SCSI_DATA_UNDERRUN: /* 0x0045 */ @@ -692,7 +702,10 @@ mptscsih_io_done(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR *mr) ; } else { if (xfer_cnt < sc->underflow) { - sc->result = DID_SOFT_ERROR << 16; + if (scsi_status == SAM_STAT_BUSY) + sc->result = SAM_STAT_BUSY; + else + sc->result = DID_SOFT_ERROR << 16; } if (scsi_state & (MPI_SCSI_STATE_AUTOSENSE_FAILED | MPI_SCSI_STATE_NO_SCSI_STATUS)) { /* What to do? @@ -717,8 +730,10 @@ mptscsih_io_done(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR *mr) case MPI_IOCSTATUS_SCSI_RECOVERED_ERROR: /* 0x0040 */ case MPI_IOCSTATUS_SUCCESS: /* 0x0000 */ - scsi_status = pScsiReply->SCSIStatus; - sc->result = (DID_OK << 16) | scsi_status; + if (scsi_status == MPI_SCSI_STATUS_BUSY) + sc->result = (DID_BUS_BUSY << 16) | scsi_status; + else + sc->result = (DID_OK << 16) | scsi_status; if (scsi_state == 0) { ; } else if (scsi_state & MPI_SCSI_STATE_AUTOSENSE_VALID) { @@ -890,12 +905,13 @@ mptscsih_search_running_cmds(MPT_SCSI_HOST *hd, uint target, uint lun) SCSIIORequest_t *mf = NULL; int ii; int max = hd->ioc->req_depth; + struct scsi_cmnd *sc; dsprintk((KERN_INFO MYNAM ": search_running target %d lun %d max %d\n", target, lun, max)); for (ii=0; ii < max; ii++) { - if (hd->ScsiLookup[ii] != NULL) { + if ((sc = hd->ScsiLookup[ii]) != NULL) { mf = (SCSIIORequest_t *)MPT_INDEX_2_MFPTR(hd->ioc, ii); @@ -910,9 +926,22 @@ mptscsih_search_running_cmds(MPT_SCSI_HOST *hd, uint target, uint lun) hd->ScsiLookup[ii] = NULL; mptscsih_freeChainBuffers(hd->ioc, ii); mpt_free_msg_frame(hd->ioc, (MPT_FRAME_HDR *)mf); + if (sc->use_sg) { + pci_unmap_sg(hd->ioc->pcidev, + (struct scatterlist *) sc->request_buffer, + sc->use_sg, + sc->sc_data_direction); + } else if (sc->request_bufflen) { + pci_unmap_single(hd->ioc->pcidev, + sc->SCp.dma_handle, + sc->request_bufflen, + sc->sc_data_direction); + } + sc->host_scribble = NULL; + sc->result = DID_NO_CONNECT << 16; + sc->scsi_done(sc); } } - return; } @@ -967,8 +996,10 @@ mptscsih_remove(struct pci_dev *pdev) unsigned long flags; int sz1; - if(!host) + if(!host) { + mpt_detach(pdev); return; + } scsi_remove_host(host); @@ -1422,6 +1453,7 @@ mptscsih_qcmd(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *)) return 0; fail: + hd->ScsiLookup[my_idx] = NULL; mptscsih_freeChainBuffers(hd->ioc, my_idx); mpt_free_msg_frame(hd->ioc, mf); return SCSI_MLQUEUE_HOST_BUSY; @@ -1709,24 +1741,23 @@ mptscsih_abort(struct scsi_cmnd * SCpnt) MPT_FRAME_HDR *mf; u32 ctx2abort; int scpnt_idx; + int retval; /* If we can't locate our host adapter structure, return FAILED status. */ if ((hd = (MPT_SCSI_HOST *) SCpnt->device->host->hostdata) == NULL) { SCpnt->result = DID_RESET << 16; SCpnt->scsi_done(SCpnt); - dfailprintk((KERN_WARNING MYNAM ": mptscsih_abort: " + dfailprintk((KERN_INFO MYNAM ": mptscsih_abort: " "Can't locate host! (sc=%p)\n", SCpnt)); return FAILED; } ioc = hd->ioc; - if (hd->resetPending) + if (hd->resetPending) { return FAILED; - - printk(KERN_WARNING MYNAM ": %s: >> Attempting task abort! (sc=%p)\n", - hd->ioc->name, SCpnt); + } if (hd->timeouts < -1) hd->timeouts++; @@ -1734,16 +1765,20 @@ mptscsih_abort(struct scsi_cmnd * SCpnt) /* Find this command */ if ((scpnt_idx = SCPNT_TO_LOOKUP_IDX(SCpnt)) < 0) { - /* Cmd not found in ScsiLookup. + /* Cmd not found in ScsiLookup. * Do OS callback. */ SCpnt->result = DID_RESET << 16; - dtmprintk((KERN_WARNING MYNAM ": %s: mptscsih_abort: " + dtmprintk((KERN_INFO MYNAM ": %s: mptscsih_abort: " "Command not in the active list! (sc=%p)\n", hd->ioc->name, SCpnt)); return SUCCESS; } + printk(KERN_WARNING MYNAM ": %s: attempting task abort! (sc=%p)\n", + hd->ioc->name, SCpnt); + scsi_print_command(SCpnt); + /* Most important! Set TaskMsgContext to SCpnt's MsgContext! * (the IO to be ABORT'd) * @@ -1756,38 +1791,22 @@ mptscsih_abort(struct scsi_cmnd * SCpnt) hd->abortSCpnt = SCpnt; - if (mptscsih_TMHandler(hd, MPI_SCSITASKMGMT_TASKTYPE_ABORT_TASK, + retval = mptscsih_TMHandler(hd, MPI_SCSITASKMGMT_TASKTYPE_ABORT_TASK, SCpnt->device->channel, SCpnt->device->id, SCpnt->device->lun, - ctx2abort, 2 /* 2 second timeout */) - < 0) { + ctx2abort, 2 /* 2 second timeout */); - /* The TM request failed and the subsequent FW-reload failed! - * Fatal error case. - */ - printk(MYIOC_s_WARN_FMT "Error issuing abort task! (sc=%p)\n", - hd->ioc->name, SCpnt); + printk (KERN_WARNING MYNAM ": %s: task abort: %s (sc=%p)\n", + hd->ioc->name, + ((retval == 0) ? "SUCCESS" : "FAILED" ), SCpnt); - /* We must clear our pending flag before clearing our state. - */ + if (retval == 0) + return SUCCESS; + + if(retval != FAILED ) { hd->tmPending = 0; hd->tmState = TM_STATE_NONE; - - /* Unmap the DMA buffers, if any. */ - if (SCpnt->use_sg) { - pci_unmap_sg(ioc->pcidev, (struct scatterlist *) SCpnt->request_buffer, - SCpnt->use_sg, SCpnt->sc_data_direction); - } else if (SCpnt->request_bufflen) { - pci_unmap_single(ioc->pcidev, SCpnt->SCp.dma_handle, - SCpnt->request_bufflen, SCpnt->sc_data_direction); - } - hd->ScsiLookup[scpnt_idx] = NULL; - SCpnt->result = DID_RESET << 16; - SCpnt->scsi_done(SCpnt); /* Issue the command callback */ - mptscsih_freeChainBuffers(ioc, scpnt_idx); - mpt_free_msg_frame(ioc, mf); - return FAILED; } - return SUCCESS; + return FAILED; } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ @@ -1803,11 +1822,12 @@ int mptscsih_dev_reset(struct scsi_cmnd * SCpnt) { MPT_SCSI_HOST *hd; + int retval; /* If we can't locate our host adapter structure, return FAILED status. */ if ((hd = (MPT_SCSI_HOST *) SCpnt->device->host->hostdata) == NULL){ - dtmprintk((KERN_WARNING MYNAM ": mptscsih_dev_reset: " + dtmprintk((KERN_INFO MYNAM ": mptscsih_dev_reset: " "Can't locate host! (sc=%p)\n", SCpnt)); return FAILED; @@ -1816,24 +1836,26 @@ mptscsih_dev_reset(struct scsi_cmnd * SCpnt) if (hd->resetPending) return FAILED; - printk(KERN_WARNING MYNAM ": %s: >> Attempting target reset! (sc=%p)\n", + printk(KERN_WARNING MYNAM ": %s: attempting target reset! (sc=%p)\n", hd->ioc->name, SCpnt); + scsi_print_command(SCpnt); - if (mptscsih_TMHandler(hd, MPI_SCSITASKMGMT_TASKTYPE_TARGET_RESET, + retval = mptscsih_TMHandler(hd, MPI_SCSITASKMGMT_TASKTYPE_TARGET_RESET, SCpnt->device->channel, SCpnt->device->id, - 0, 0, 5 /* 5 second timeout */) - < 0){ - /* The TM request failed and the subsequent FW-reload failed! - * Fatal error case. - */ - printk(MYIOC_s_WARN_FMT "Error processing TaskMgmt request (sc=%p)\n", - hd->ioc->name, SCpnt); + 0, 0, 5 /* 5 second timeout */); + + printk (KERN_WARNING MYNAM ": %s: target reset: %s (sc=%p)\n", + hd->ioc->name, + ((retval == 0) ? "SUCCESS" : "FAILED" ), SCpnt); + + if (retval == 0) + return SUCCESS; + + if(retval != FAILED ) { hd->tmPending = 0; hd->tmState = TM_STATE_NONE; - return FAILED; } - - return SUCCESS; + return FAILED; } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ @@ -1849,41 +1871,39 @@ int mptscsih_bus_reset(struct scsi_cmnd * SCpnt) { MPT_SCSI_HOST *hd; - spinlock_t *host_lock = SCpnt->device->host->host_lock; + int retval; /* If we can't locate our host adapter structure, return FAILED status. */ if ((hd = (MPT_SCSI_HOST *) SCpnt->device->host->hostdata) == NULL){ - dtmprintk((KERN_WARNING MYNAM ": mptscsih_bus_reset: " + dtmprintk((KERN_INFO MYNAM ": mptscsih_bus_reset: " "Can't locate host! (sc=%p)\n", SCpnt ) ); return FAILED; } - printk(KERN_WARNING MYNAM ": %s: >> Attempting bus reset! (sc=%p)\n", + printk(KERN_WARNING MYNAM ": %s: attempting bus reset! (sc=%p)\n", hd->ioc->name, SCpnt); + scsi_print_command(SCpnt); if (hd->timeouts < -1) hd->timeouts++; - /* We are now ready to execute the task management request. */ - if (mptscsih_TMHandler(hd, MPI_SCSITASKMGMT_TASKTYPE_RESET_BUS, - SCpnt->device->channel, 0, 0, 0, 5 /* 5 second timeout */) - < 0){ + retval = mptscsih_TMHandler(hd, MPI_SCSITASKMGMT_TASKTYPE_RESET_BUS, + SCpnt->device->channel, 0, 0, 0, 5 /* 5 second timeout */); - /* The TM request failed and the subsequent FW-reload failed! - * Fatal error case. - */ - printk(MYIOC_s_WARN_FMT - "Error processing TaskMgmt request (sc=%p)\n", - hd->ioc->name, SCpnt); + printk (KERN_WARNING MYNAM ": %s: bus reset: %s (sc=%p)\n", + hd->ioc->name, + ((retval == 0) ? "SUCCESS" : "FAILED" ), SCpnt); + + if (retval == 0) + return SUCCESS; + + if(retval != FAILED ) { hd->tmPending = 0; hd->tmState = TM_STATE_NONE; - spin_lock_irq(host_lock); - return FAILED; } - - return SUCCESS; + return FAILED; } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ @@ -2165,7 +2185,7 @@ mptscsih_slave_alloc(struct scsi_device *device) vdev->raidVolume = 0; hd->Targets[device->id] = vdev; if (hd->ioc->bus_type == SCSI) { - if (hd->ioc->spi_data.isRaid & (1 << device->id)) { + if (hd->ioc->raid_data.isRaid & (1 << device->id)) { vdev->raidVolume = 1; ddvtprintk((KERN_INFO "RAID Volume @ id %d\n", device->id)); @@ -2180,22 +2200,6 @@ mptscsih_slave_alloc(struct scsi_device *device) return 0; } -static int -mptscsih_is_raid_volume(MPT_SCSI_HOST *hd, uint id) -{ - int i; - - if (!hd->ioc->spi_data.isRaid || !hd->ioc->spi_data.pIocPg3) - return 0; - - for (i = 0; i < hd->ioc->spi_data.pIocPg3->NumPhysDisks; i++) { - if (id == hd->ioc->spi_data.pIocPg3->PhysDisk[i].PhysDiskID) - return 1; - } - - return 0; -} - /* * OS entry point to allow for host driver to free allocated memory * Called if no device present or device being unloaded @@ -2223,7 +2227,7 @@ mptscsih_slave_destroy(struct scsi_device *device) hd->Targets[target] = NULL; if (hd->ioc->bus_type == SCSI) { - if (mptscsih_is_raid_volume(hd, target)) { + if (mptscsih_is_phys_disk(hd->ioc, target)) { hd->ioc->spi_data.forceDv |= MPT_SCSICFG_RELOAD_IOC_PG3; } else { hd->ioc->spi_data.dvStatus[target] = @@ -2436,6 +2440,7 @@ mptscsih_ioc_reset(MPT_ADAPTER *ioc, int reset_phase) { MPT_SCSI_HOST *hd; unsigned long flags; + int ii; dtmprintk((KERN_WARNING MYNAM ": IOC %s_reset routed to SCSI host driver!\n", @@ -2493,11 +2498,8 @@ mptscsih_ioc_reset(MPT_ADAPTER *ioc, int reset_phase) /* ScsiLookup initialization */ - { - int ii; - for (ii=0; ii < hd->ioc->req_depth; ii++) - hd->ScsiLookup[ii] = NULL; - } + for (ii=0; ii < hd->ioc->req_depth; ii++) + hd->ScsiLookup[ii] = NULL; /* 2. Chain Buffer initialization */ @@ -2545,6 +2547,16 @@ mptscsih_ioc_reset(MPT_ADAPTER *ioc, int reset_phase) return 1; /* currently means nothing really */ } +/*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ +/* work queue thread to clear the persitency table */ +static void +mptscsih_sas_persist_clear_table(void * arg) +{ + MPT_ADAPTER *ioc = (MPT_ADAPTER *)arg; + + mptbase_sas_persist_operation(ioc, MPI_SAS_OP_CLEAR_NOT_PRESENT); +} + /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ int mptscsih_event_process(MPT_ADAPTER *ioc, EventNotificationReply_t *pEvReply) @@ -2555,18 +2567,18 @@ mptscsih_event_process(MPT_ADAPTER *ioc, EventNotificationReply_t *pEvReply) devtprintk((MYIOC_s_INFO_FMT "MPT event (=%02Xh) routed to SCSI host driver!\n", ioc->name, event)); + if (ioc->sh == NULL || + ((hd = (MPT_SCSI_HOST *)ioc->sh->hostdata) == NULL)) + return 1; + switch (event) { case MPI_EVENT_UNIT_ATTENTION: /* 03 */ /* FIXME! */ break; case MPI_EVENT_IOC_BUS_RESET: /* 04 */ case MPI_EVENT_EXT_BUS_RESET: /* 05 */ - hd = NULL; - if (ioc->sh) { - hd = (MPT_SCSI_HOST *) ioc->sh->hostdata; - if (hd && (ioc->bus_type == SCSI) && (hd->soft_resets < -1)) - hd->soft_resets++; - } + if (hd && (ioc->bus_type == SCSI) && (hd->soft_resets < -1)) + hd->soft_resets++; break; case MPI_EVENT_LOGOUT: /* 09 */ /* FIXME! */ @@ -2585,69 +2597,24 @@ mptscsih_event_process(MPT_ADAPTER *ioc, EventNotificationReply_t *pEvReply) break; case MPI_EVENT_INTEGRATED_RAID: /* 0B */ + { + pMpiEventDataRaid_t pRaidEventData = + (pMpiEventDataRaid_t) pEvReply->Data; #ifdef MPTSCSIH_ENABLE_DOMAIN_VALIDATION - /* negoNvram set to 0 if DV enabled and to USE_NVRAM if - * if DV disabled. Need to check for target mode. - */ - hd = NULL; - if (ioc->sh) - hd = (MPT_SCSI_HOST *) ioc->sh->hostdata; - - if (hd && (ioc->bus_type == SCSI) && (hd->negoNvram == 0)) { - ScsiCfgData *pSpi; - Ioc3PhysDisk_t *pPDisk; - int numPDisk; - u8 reason; - u8 physDiskNum; - - reason = (le32_to_cpu(pEvReply->Data[0]) & 0x00FF0000) >> 16; - if (reason == MPI_EVENT_RAID_RC_DOMAIN_VAL_NEEDED) { - /* New or replaced disk. - * Set DV flag and schedule DV. - */ - pSpi = &ioc->spi_data; - physDiskNum = (le32_to_cpu(pEvReply->Data[0]) & 0xFF000000) >> 24; - ddvtprintk(("DV requested for phys disk id %d\n", physDiskNum)); - if (pSpi->pIocPg3) { - pPDisk = pSpi->pIocPg3->PhysDisk; - numPDisk =pSpi->pIocPg3->NumPhysDisks; - - while (numPDisk) { - if (physDiskNum == pPDisk->PhysDiskNum) { - pSpi->dvStatus[pPDisk->PhysDiskID] = (MPT_SCSICFG_NEED_DV | MPT_SCSICFG_DV_NOT_DONE); - pSpi->forceDv = MPT_SCSICFG_NEED_DV; - ddvtprintk(("NEED_DV set for phys disk id %d\n", pPDisk->PhysDiskID)); - break; - } - pPDisk++; - numPDisk--; - } - - if (numPDisk == 0) { - /* The physical disk that needs DV was not found - * in the stored IOC Page 3. The driver must reload - * this page. DV routine will set the NEED_DV flag for - * all phys disks that have DV_NOT_DONE set. - */ - pSpi->forceDv = MPT_SCSICFG_NEED_DV | MPT_SCSICFG_RELOAD_IOC_PG3; - ddvtprintk(("phys disk %d not found. Setting reload IOC Pg3 Flag\n", physDiskNum)); - } - } - } - } + /* Domain Validation Needed */ + if (ioc->bus_type == SCSI && + pRaidEventData->ReasonCode == + MPI_EVENT_RAID_RC_DOMAIN_VAL_NEEDED) + mptscsih_set_dvflags_raid(hd, pRaidEventData->PhysDiskNum); #endif + break; + } -#if defined(MPT_DEBUG_DV) || defined(MPT_DEBUG_DV_TINY) - printk("Raid Event RF: "); - { - u32 *m = (u32 *)pEvReply; - int ii; - int n = (int)pEvReply->MsgLength; - for (ii=6; ii < n; ii++) - printk(" %08x", le32_to_cpu(m[ii])); - printk("\n"); - } -#endif + /* Persistent table is full. */ + case MPI_EVENT_PERSISTENT_TABLE_FULL: + INIT_WORK(&mptscsih_persistTask, + mptscsih_sas_persist_clear_table,(void *)ioc); + schedule_work(&mptscsih_persistTask); break; case MPI_EVENT_NONE: /* 00 */ @@ -2684,7 +2651,7 @@ mptscsih_initTarget(MPT_SCSI_HOST *hd, int bus_id, int target_id, u8 lun, char * { int indexed_lun, lun_index; VirtDevice *vdev; - ScsiCfgData *pSpi; + SpiCfgData *pSpi; char data_56; dinitprintk((MYIOC_s_INFO_FMT "initTarget bus=%d id=%d lun=%d hd=%p\n", @@ -2791,7 +2758,7 @@ mptscsih_initTarget(MPT_SCSI_HOST *hd, int bus_id, int target_id, u8 lun, char * static void mptscsih_setTargetNegoParms(MPT_SCSI_HOST *hd, VirtDevice *target, char byte56) { - ScsiCfgData *pspi_data = &hd->ioc->spi_data; + SpiCfgData *pspi_data = &hd->ioc->spi_data; int id = (int) target->target_id; int nvram; VirtDevice *vdev; @@ -2970,11 +2937,13 @@ mptscsih_setTargetNegoParms(MPT_SCSI_HOST *hd, VirtDevice *target, char byte56) static void mptscsih_set_dvflags(MPT_SCSI_HOST *hd, SCSIIORequest_t *pReq) { + MPT_ADAPTER *ioc = hd->ioc; u8 cmd; - ScsiCfgData *pSpi; + SpiCfgData *pSpi; - ddvtprintk((" set_dvflags: id=%d lun=%d negoNvram=%x cmd=%x\n", - pReq->TargetID, pReq->LUN[1], hd->negoNvram, pReq->CDB[0])); + ddvtprintk((MYIOC_s_NOTE_FMT + " set_dvflags: id=%d lun=%d negoNvram=%x cmd=%x\n", + hd->ioc->name, pReq->TargetID, pReq->LUN[1], hd->negoNvram, pReq->CDB[0])); if ((pReq->LUN[1] != 0) || (hd->negoNvram != 0)) return; @@ -2982,12 +2951,12 @@ mptscsih_set_dvflags(MPT_SCSI_HOST *hd, SCSIIORequest_t *pReq) cmd = pReq->CDB[0]; if ((cmd == READ_CAPACITY) || (cmd == MODE_SENSE)) { - pSpi = &hd->ioc->spi_data; - if ((pSpi->isRaid & (1 << pReq->TargetID)) && pSpi->pIocPg3) { + pSpi = &ioc->spi_data; + if ((ioc->raid_data.isRaid & (1 << pReq->TargetID)) && ioc->raid_data.pIocPg3) { /* Set NEED_DV for all hidden disks */ - Ioc3PhysDisk_t *pPDisk = pSpi->pIocPg3->PhysDisk; - int numPDisk = pSpi->pIocPg3->NumPhysDisks; + Ioc3PhysDisk_t *pPDisk = ioc->raid_data.pIocPg3->PhysDisk; + int numPDisk = ioc->raid_data.pIocPg3->NumPhysDisks; while (numPDisk) { pSpi->dvStatus[pPDisk->PhysDiskID] |= MPT_SCSICFG_NEED_DV; @@ -3001,6 +2970,50 @@ mptscsih_set_dvflags(MPT_SCSI_HOST *hd, SCSIIORequest_t *pReq) } } +/* mptscsih_raid_set_dv_flags() + * + * New or replaced disk. Set DV flag and schedule DV. + */ +static void +mptscsih_set_dvflags_raid(MPT_SCSI_HOST *hd, int id) +{ + MPT_ADAPTER *ioc = hd->ioc; + SpiCfgData *pSpi = &ioc->spi_data; + Ioc3PhysDisk_t *pPDisk; + int numPDisk; + + if (hd->negoNvram != 0) + return; + + ddvtprintk(("DV requested for phys disk id %d\n", id)); + if (ioc->raid_data.pIocPg3) { + pPDisk = ioc->raid_data.pIocPg3->PhysDisk; + numPDisk = ioc->raid_data.pIocPg3->NumPhysDisks; + while (numPDisk) { + if (id == pPDisk->PhysDiskNum) { + pSpi->dvStatus[pPDisk->PhysDiskID] = + (MPT_SCSICFG_NEED_DV | MPT_SCSICFG_DV_NOT_DONE); + pSpi->forceDv = MPT_SCSICFG_NEED_DV; + ddvtprintk(("NEED_DV set for phys disk id %d\n", + pPDisk->PhysDiskID)); + break; + } + pPDisk++; + numPDisk--; + } + + if (numPDisk == 0) { + /* The physical disk that needs DV was not found + * in the stored IOC Page 3. The driver must reload + * this page. DV routine will set the NEED_DV flag for + * all phys disks that have DV_NOT_DONE set. + */ + pSpi->forceDv = MPT_SCSICFG_NEED_DV | MPT_SCSICFG_RELOAD_IOC_PG3; + ddvtprintk(("phys disk %d not found. Setting reload IOC Pg3 Flag\n",id)); + } + } +} + /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /* * If no Target, bus reset on 1st I/O. Set the flag to @@ -3088,7 +3101,7 @@ mptscsih_writeSDP1(MPT_SCSI_HOST *hd, int portnum, int target_id, int flags) MPT_ADAPTER *ioc = hd->ioc; Config_t *pReq; SCSIDevicePage1_t *pData; - VirtDevice *pTarget; + VirtDevice *pTarget=NULL; MPT_FRAME_HDR *mf; dma_addr_t dataDma; u16 req_idx; @@ -3187,7 +3200,7 @@ mptscsih_writeSDP1(MPT_SCSI_HOST *hd, int portnum, int target_id, int flags) #endif if (flags & MPT_SCSICFG_BLK_NEGO) - negoFlags = MPT_TARGET_NO_NEGO_WIDE | MPT_TARGET_NO_NEGO_SYNC; + negoFlags |= MPT_TARGET_NO_NEGO_WIDE | MPT_TARGET_NO_NEGO_SYNC; mptscsih_setDevicePage1Flags(width, factor, offset, &requested, &configuration, negoFlags); @@ -4008,7 +4021,7 @@ mptscsih_synchronize_cache(MPT_SCSI_HOST *hd, int portnum) /* If target Ptr NULL or if this target is NOT a disk, skip. */ - if ((pTarget) && (pTarget->tflags & MPT_TARGET_FLAGS_Q_YES)){ + if ((pTarget) && (pTarget->inq_data[0] == TYPE_DISK)){ for (lun=0; lun <= MPT_LAST_LUN; lun++) { /* If LUN present, issue the command */ @@ -4103,9 +4116,9 @@ mptscsih_domainValidation(void *arg) if ((ioc->spi_data.forceDv & MPT_SCSICFG_RELOAD_IOC_PG3) != 0) { mpt_read_ioc_pg_3(ioc); - if (ioc->spi_data.pIocPg3) { - Ioc3PhysDisk_t *pPDisk = ioc->spi_data.pIocPg3->PhysDisk; - int numPDisk = ioc->spi_data.pIocPg3->NumPhysDisks; + if (ioc->raid_data.pIocPg3) { + Ioc3PhysDisk_t *pPDisk = ioc->raid_data.pIocPg3->PhysDisk; + int numPDisk = ioc->raid_data.pIocPg3->NumPhysDisks; while (numPDisk) { if (ioc->spi_data.dvStatus[pPDisk->PhysDiskID] & MPT_SCSICFG_DV_NOT_DONE) @@ -4144,7 +4157,7 @@ mptscsih_domainValidation(void *arg) isPhysDisk = mptscsih_is_phys_disk(ioc, id); if (isPhysDisk) { for (ii=0; ii < MPT_MAX_SCSI_DEVICES; ii++) { - if (hd->ioc->spi_data.isRaid & (1 << ii)) { + if (hd->ioc->raid_data.isRaid & (1 << ii)) { hd->ioc->spi_data.dvStatus[ii] |= MPT_SCSICFG_DV_PENDING; } } @@ -4163,7 +4176,7 @@ mptscsih_domainValidation(void *arg) if (isPhysDisk) { for (ii=0; ii < MPT_MAX_SCSI_DEVICES; ii++) { - if (hd->ioc->spi_data.isRaid & (1 << ii)) { + if (hd->ioc->raid_data.isRaid & (1 << ii)) { hd->ioc->spi_data.dvStatus[ii] &= ~MPT_SCSICFG_DV_PENDING; } } @@ -4185,21 +4198,21 @@ mptscsih_domainValidation(void *arg) /* Search IOC page 3 to determine if this is hidden physical disk */ -static int +/* Search IOC page 3 to determine if this is hidden physical disk + */ +static int mptscsih_is_phys_disk(MPT_ADAPTER *ioc, int id) { - if (ioc->spi_data.pIocPg3) { - Ioc3PhysDisk_t *pPDisk = ioc->spi_data.pIocPg3->PhysDisk; - int numPDisk = ioc->spi_data.pIocPg3->NumPhysDisks; + int i; - while (numPDisk) { - if (pPDisk->PhysDiskID == id) { - return 1; - } - pPDisk++; - numPDisk--; - } + if (!ioc->raid_data.isRaid || !ioc->raid_data.pIocPg3) + return 0; + + for (i = 0; i < ioc->raid_data.pIocPg3->NumPhysDisks; i++) { + if (id == ioc->raid_data.pIocPg3->PhysDisk[i].PhysDiskID) + return 1; } + return 0; } @@ -4405,7 +4418,7 @@ mptscsih_doDv(MPT_SCSI_HOST *hd, int bus_number, int id) /* Skip this ID? Set cfg.cfghdr.hdr to force config page write */ { - ScsiCfgData *pspi_data = &hd->ioc->spi_data; + SpiCfgData *pspi_data = &hd->ioc->spi_data; if (pspi_data->nvram && (pspi_data->nvram[id] != MPT_HOST_NVRAM_INVALID)) { /* Set the factor from nvram */ nfactor = (pspi_data->nvram[id] & MPT_NVRAM_SYNC_MASK) >> 8; @@ -4435,11 +4448,11 @@ mptscsih_doDv(MPT_SCSI_HOST *hd, int bus_number, int id) } /* Finish iocmd inititialization - hidden or visible disk? */ - if (ioc->spi_data.pIocPg3) { + if (ioc->raid_data.pIocPg3) { /* Search IOC page 3 for matching id */ - Ioc3PhysDisk_t *pPDisk = ioc->spi_data.pIocPg3->PhysDisk; - int numPDisk = ioc->spi_data.pIocPg3->NumPhysDisks; + Ioc3PhysDisk_t *pPDisk = ioc->raid_data.pIocPg3->PhysDisk; + int numPDisk = ioc->raid_data.pIocPg3->NumPhysDisks; while (numPDisk) { if (pPDisk->PhysDiskID == id) { @@ -4463,7 +4476,7 @@ mptscsih_doDv(MPT_SCSI_HOST *hd, int bus_number, int id) /* RAID Volume ID's may double for a physical device. If RAID but * not a physical ID as well, skip DV. */ - if ((hd->ioc->spi_data.isRaid & (1 << id)) && !(iocmd.flags & MPT_ICFLAG_PHYS_DISK)) + if ((hd->ioc->raid_data.isRaid & (1 << id)) && !(iocmd.flags & MPT_ICFLAG_PHYS_DISK)) goto target_done; @@ -4812,6 +4825,8 @@ mptscsih_doDv(MPT_SCSI_HOST *hd, int bus_number, int id) notDone = 0; if (iocmd.flags & MPT_ICFLAG_ECHO) { bufsize = ((pbuf1[2] & 0x1F) <<8) | pbuf1[3]; + if (pbuf1[0] & 0x01) + iocmd.flags |= MPT_ICFLAG_EBOS; } else { bufsize = pbuf1[1]<<16 | pbuf1[2]<<8 | pbuf1[3]; } @@ -4908,6 +4923,9 @@ mptscsih_doDv(MPT_SCSI_HOST *hd, int bus_number, int id) } iocmd.flags &= ~MPT_ICFLAG_DID_RESET; + if (iocmd.flags & MPT_ICFLAG_EBOS) + goto skip_Reserve; + repeat = 5; while (repeat && (!(iocmd.flags & MPT_ICFLAG_RESERVED))) { iocmd.cmd = RESERVE; @@ -4951,6 +4969,7 @@ mptscsih_doDv(MPT_SCSI_HOST *hd, int bus_number, int id) } } +skip_Reserve: mptscsih_fillbuf(pbuf1, sz, patt, 1); iocmd.cmd = WRITE_BUFFER; iocmd.data_dma = buf1_dma; @@ -5195,11 +5214,12 @@ mptscsih_dv_parms(MPT_SCSI_HOST *hd, DVPARAMETERS *dv,void *pPage) * If not an LVD bus, the adapter minSyncFactor has been * already throttled back. */ + negoFlags = hd->ioc->spi_data.noQas; if ((hd->Targets)&&((pTarget = hd->Targets[(int)id]) != NULL) && !pTarget->raidVolume) { width = pTarget->maxWidth; offset = pTarget->maxOffset; factor = pTarget->minSyncFactor; - negoFlags = pTarget->negoFlags; + negoFlags |= pTarget->negoFlags; } else { if (hd->ioc->spi_data.nvram && (hd->ioc->spi_data.nvram[id] != MPT_HOST_NVRAM_INVALID)) { data = hd->ioc->spi_data.nvram[id]; @@ -5220,7 +5240,6 @@ mptscsih_dv_parms(MPT_SCSI_HOST *hd, DVPARAMETERS *dv,void *pPage) } /* Set the negotiation flags */ - negoFlags = hd->ioc->spi_data.noQas; if (!width) negoFlags |= MPT_TARGET_NO_NEGO_WIDE; diff --git a/drivers/message/fusion/mptscsih.h b/drivers/message/fusion/mptscsih.h index 51c0255ac16e..09389af9845e 100644 --- a/drivers/message/fusion/mptscsih.h +++ b/drivers/message/fusion/mptscsih.h @@ -1,5 +1,5 @@ /* - * linux/drivers/message/fusion/mptscsi.h + * linux/drivers/message/fusion/mptscsih.h * High performance SCSI / Fibre Channel SCSI Host device driver. * For use with PCI chip/adapter(s): * LSIFC9xx/LSI409xx Fibre Channel @@ -53,8 +53,8 @@ * SCSI Public stuff... */ -#define MPT_SCSI_CMD_PER_DEV_HIGH 31 -#define MPT_SCSI_CMD_PER_DEV_LOW 7 +#define MPT_SCSI_CMD_PER_DEV_HIGH 64 +#define MPT_SCSI_CMD_PER_DEV_LOW 32 #define MPT_SCSI_CMD_PER_LUN 7 @@ -77,6 +77,7 @@ #define MPTSCSIH_MAX_WIDTH 1 #define MPTSCSIH_MIN_SYNC 0x08 #define MPTSCSIH_SAF_TE 0 +#define MPTSCSIH_PT_CLEAR 0 #endif @@ -105,3 +106,4 @@ extern int mptscsih_event_process(MPT_ADAPTER *ioc, EventNotificationReply_t *pE extern int mptscsih_ioc_reset(MPT_ADAPTER *ioc, int post_reset); extern int mptscsih_change_queue_depth(struct scsi_device *sdev, int qdepth); extern void mptscsih_timer_expired(unsigned long data); +extern void scsi_print_command(struct scsi_cmnd *cmd); diff --git a/drivers/message/fusion/mptspi.c b/drivers/message/fusion/mptspi.c index 587d1274fd74..5c0e307d1d5d 100644 --- a/drivers/message/fusion/mptspi.c +++ b/drivers/message/fusion/mptspi.c @@ -199,7 +199,7 @@ mptspi_probe(struct pci_dev *pdev, const struct pci_device_id *id) printk(MYIOC_s_WARN_FMT "Skipping ioc=%p because SCSI Initiator mode is NOT enabled!\n", ioc->name, ioc); - return -ENODEV; + return 0; } sh = scsi_host_alloc(&mptspi_driver_template, sizeof(MPT_SCSI_HOST)); From e0fc15bef0e8c6b5abad6e10cfe3d42e278ae8e8 Mon Sep 17 00:00:00 2001 From: "Moore, Eric Dean" Date: Thu, 15 Sep 2005 13:17:14 -0600 Subject: [PATCH 167/240] [SCSI] fusion SAS support (mptsas driver) minor fix On Thursday, September 15, 2005 6:22 AM, Christoph Hellwig wrote: > Looks good to me, except for the spurious scsi_print_command prototype > in mptscsih.h. The attached patch addresses that concern. Signed-off-by: Eric Moore Signed-off-by: James Bottomley --- drivers/message/fusion/mptscsih.c | 1 + drivers/message/fusion/mptscsih.h | 1 - 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/message/fusion/mptscsih.c b/drivers/message/fusion/mptscsih.c index 8dd25aac5355..5cb07eb224d7 100644 --- a/drivers/message/fusion/mptscsih.c +++ b/drivers/message/fusion/mptscsih.c @@ -62,6 +62,7 @@ #include #include #include +#include #include "mptbase.h" #include "mptscsih.h" diff --git a/drivers/message/fusion/mptscsih.h b/drivers/message/fusion/mptscsih.h index 09389af9845e..971fda4b8b57 100644 --- a/drivers/message/fusion/mptscsih.h +++ b/drivers/message/fusion/mptscsih.h @@ -106,4 +106,3 @@ extern int mptscsih_event_process(MPT_ADAPTER *ioc, EventNotificationReply_t *pE extern int mptscsih_ioc_reset(MPT_ADAPTER *ioc, int post_reset); extern int mptscsih_change_queue_depth(struct scsi_device *sdev, int qdepth); extern void mptscsih_timer_expired(unsigned long data); -extern void scsi_print_command(struct scsi_cmnd *cmd); From 3734d24b2e8d85796de70c13705cfb7cbb1d77df Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Tue, 13 Sep 2005 21:47:11 +0200 Subject: [PATCH 168/240] [SCSI] zfcp: fix race conditions when accessing erp_action lists o always use locking when changing erp_action lists, o avoid escalation to ERP_ACTION_REOPEN_PORT_FORCED if erp_action is still in use for ERP_ACTION_REOPEN_PORT Signed-off-by: Andreas Herrmann Signed-off-by: James Bottomley --- drivers/s390/scsi/zfcp_erp.c | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/drivers/s390/scsi/zfcp_erp.c b/drivers/s390/scsi/zfcp_erp.c index cb4f612550ba..376cb0f6cb74 100644 --- a/drivers/s390/scsi/zfcp_erp.c +++ b/drivers/s390/scsi/zfcp_erp.c @@ -886,7 +886,7 @@ static int zfcp_erp_strategy_check_fsfreq(struct zfcp_erp_action *erp_action) { int retval = 0; - struct zfcp_fsf_req *fsf_req; + struct zfcp_fsf_req *fsf_req = NULL; struct zfcp_adapter *adapter = erp_action->adapter; if (erp_action->fsf_req) { @@ -896,7 +896,7 @@ zfcp_erp_strategy_check_fsfreq(struct zfcp_erp_action *erp_action) list_for_each_entry(fsf_req, &adapter->fsf_req_list_head, list) if (fsf_req == erp_action->fsf_req) break; - if (fsf_req == erp_action->fsf_req) { + if (fsf_req && (fsf_req->erp_action == erp_action)) { /* fsf_req still exists */ debug_text_event(adapter->erp_dbf, 3, "a_ca_req"); debug_event(adapter->erp_dbf, 3, &fsf_req, @@ -2291,7 +2291,9 @@ zfcp_erp_adapter_strategy_open_fsf_xconfig(struct zfcp_erp_action *erp_action) atomic_clear_mask(ZFCP_STATUS_ADAPTER_HOST_CON_INIT, &adapter->status); ZFCP_LOG_DEBUG("Doing exchange config data\n"); + write_lock(&adapter->erp_lock); zfcp_erp_action_to_running(erp_action); + write_unlock(&adapter->erp_lock); zfcp_erp_timeout_init(erp_action); if (zfcp_fsf_exchange_config_data(erp_action)) { retval = ZFCP_ERP_FAILED; @@ -3194,11 +3196,19 @@ zfcp_erp_action_enqueue(int action, /* fall through !!! */ case ZFCP_ERP_ACTION_REOPEN_PORT_FORCED: - if (atomic_test_mask - (ZFCP_STATUS_COMMON_ERP_INUSE, &port->status) - && port->erp_action.action == - ZFCP_ERP_ACTION_REOPEN_PORT_FORCED) { - debug_text_event(adapter->erp_dbf, 4, "pf_actenq_drp"); + if (atomic_test_mask(ZFCP_STATUS_COMMON_ERP_INUSE, + &port->status)) { + if (port->erp_action.action != + ZFCP_ERP_ACTION_REOPEN_PORT_FORCED) { + ZFCP_LOG_INFO("dropped erp action %i (port " + "0x%016Lx, action in use: %i)\n", + action, port->wwpn, + port->erp_action.action); + debug_text_event(adapter->erp_dbf, 4, + "pf_actenq_drp"); + } else + debug_text_event(adapter->erp_dbf, 4, + "pf_actenq_drpcp"); debug_event(adapter->erp_dbf, 4, &port->wwpn, sizeof (wwn_t)); goto out; From 059c97d0434834d291eff94669ca2dd3eaac9d28 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Tue, 13 Sep 2005 21:47:52 +0200 Subject: [PATCH 169/240] [SCSI] zfcp: remove union zfcp_req_data, use unit refcount for FCP commands o union zfcp_req_data removed o increment unit refcount when processing FCP commands (This fixes a theoretical race: When all scsi commands of a unit are aborted and the scsi_device is removed then the unit could be removed before all fsf_requests of that unit are completely processed.) Signed-off-by: Andreas Herrmann Signed-off-by: James Bottomley --- drivers/s390/scsi/zfcp_aux.c | 8 +- drivers/s390/scsi/zfcp_def.h | 69 +------------ drivers/s390/scsi/zfcp_fsf.c | 101 ++++++++++--------- drivers/s390/scsi/zfcp_scsi.c | 176 +++++----------------------------- 4 files changed, 80 insertions(+), 274 deletions(-) diff --git a/drivers/s390/scsi/zfcp_aux.c b/drivers/s390/scsi/zfcp_aux.c index bfe3ba73bc0f..17f9989238f8 100644 --- a/drivers/s390/scsi/zfcp_aux.c +++ b/drivers/s390/scsi/zfcp_aux.c @@ -141,7 +141,7 @@ zfcp_cmd_dbf_event_fsf(const char *text, struct zfcp_fsf_req *fsf_req, spin_lock_irqsave(&adapter->dbf_lock, flags); if (zfcp_fsf_req_is_scsi_cmnd(fsf_req)) { - scsi_cmnd = fsf_req->data.send_fcp_command_task.scsi_cmnd; + scsi_cmnd = (struct scsi_cmnd*) fsf_req->data; debug_text_event(adapter->cmd_dbf, level, "fsferror"); debug_text_event(adapter->cmd_dbf, level, text); debug_event(adapter->cmd_dbf, level, &fsf_req, @@ -167,14 +167,12 @@ void zfcp_cmd_dbf_event_scsi(const char *text, struct scsi_cmnd *scsi_cmnd) { struct zfcp_adapter *adapter; - union zfcp_req_data *req_data; struct zfcp_fsf_req *fsf_req; int level = ((host_byte(scsi_cmnd->result) != 0) ? 1 : 5); unsigned long flags; adapter = (struct zfcp_adapter *) scsi_cmnd->device->host->hostdata[0]; - req_data = (union zfcp_req_data *) scsi_cmnd->host_scribble; - fsf_req = (req_data ? req_data->send_fcp_command_task.fsf_req : NULL); + fsf_req = (struct zfcp_fsf_req *) scsi_cmnd->host_scribble; spin_lock_irqsave(&adapter->dbf_lock, flags); debug_text_event(adapter->cmd_dbf, level, "hostbyte"); debug_text_event(adapter->cmd_dbf, level, text); @@ -1609,7 +1607,7 @@ zfcp_fsf_incoming_els(struct zfcp_fsf_req *fsf_req) u32 els_type; struct zfcp_adapter *adapter; - status_buffer = fsf_req->data.status_read.buffer; + status_buffer = (struct fsf_status_read_buffer *) fsf_req->data; els_type = *(u32 *) (status_buffer->payload); adapter = fsf_req->adapter; diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h index 455e902533a9..9160e68c4a21 100644 --- a/drivers/s390/scsi/zfcp_def.h +++ b/drivers/s390/scsi/zfcp_def.h @@ -635,45 +635,6 @@ struct zfcp_adapter_mempool { mempool_t *data_gid_pn; }; -struct zfcp_exchange_config_data{ -}; - -struct zfcp_open_port { - struct zfcp_port *port; -}; - -struct zfcp_close_port { - struct zfcp_port *port; -}; - -struct zfcp_open_unit { - struct zfcp_unit *unit; -}; - -struct zfcp_close_unit { - struct zfcp_unit *unit; -}; - -struct zfcp_close_physical_port { - struct zfcp_port *port; -}; - -struct zfcp_send_fcp_command_task { - struct zfcp_fsf_req *fsf_req; - struct zfcp_unit *unit; - struct scsi_cmnd *scsi_cmnd; - unsigned long start_jiffies; -}; - -struct zfcp_send_fcp_command_task_management { - struct zfcp_unit *unit; -}; - -struct zfcp_abort_fcp_command { - struct zfcp_fsf_req *fsf_req; - struct zfcp_unit *unit; -}; - /* * header for CT_IU */ @@ -781,33 +742,6 @@ struct zfcp_send_els { int status; }; -struct zfcp_status_read { - struct fsf_status_read_buffer *buffer; -}; - -struct zfcp_fsf_done { - struct completion *complete; - int status; -}; - -/* request specific data */ -union zfcp_req_data { - struct zfcp_exchange_config_data exchange_config_data; - struct zfcp_open_port open_port; - struct zfcp_close_port close_port; - struct zfcp_open_unit open_unit; - struct zfcp_close_unit close_unit; - struct zfcp_close_physical_port close_physical_port; - struct zfcp_send_fcp_command_task send_fcp_command_task; - struct zfcp_send_fcp_command_task_management - send_fcp_command_task_management; - struct zfcp_abort_fcp_command abort_fcp_command; - struct zfcp_send_ct *send_ct; - struct zfcp_send_els *send_els; - struct zfcp_status_read status_read; - struct fsf_qtcb_bottom_port *port_data; -}; - struct zfcp_qdio_queue { struct qdio_buffer *buffer[QDIO_MAX_BUFFERS_PER_Q]; /* SBALs */ u8 free_index; /* index of next free bfr @@ -963,11 +897,12 @@ struct zfcp_fsf_req { u32 fsf_command; /* FSF Command copy */ struct fsf_qtcb *qtcb; /* address of associated QTCB */ u32 seq_no; /* Sequence number of request */ - union zfcp_req_data data; /* Info fields of request */ + unsigned long data; /* private data of request */ struct zfcp_erp_action *erp_action; /* used if this request is issued on behalf of erp */ mempool_t *pool; /* used if request was alloacted from emergency pool */ + struct zfcp_unit *unit; }; typedef void zfcp_fsf_req_handler_t(struct zfcp_fsf_req*); diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index c007b6424e74..64e2c3480189 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -821,7 +821,7 @@ zfcp_fsf_status_read(struct zfcp_adapter *adapter, int req_flags) goto failed_buf; } memset(status_buffer, 0, sizeof (struct fsf_status_read_buffer)); - fsf_req->data.status_read.buffer = status_buffer; + fsf_req->data = (unsigned long) status_buffer; /* insert pointer to respective buffer */ sbale = zfcp_qdio_sbale_curr(fsf_req); @@ -859,7 +859,7 @@ zfcp_fsf_status_read_port_closed(struct zfcp_fsf_req *fsf_req) struct zfcp_port *port; unsigned long flags; - status_buffer = fsf_req->data.status_read.buffer; + status_buffer = (struct fsf_status_read_buffer *) fsf_req->data; adapter = fsf_req->adapter; read_lock_irqsave(&zfcp_data.config_lock, flags); @@ -918,7 +918,7 @@ zfcp_fsf_status_read_handler(struct zfcp_fsf_req *fsf_req) int retval = 0; struct zfcp_adapter *adapter = fsf_req->adapter; struct fsf_status_read_buffer *status_buffer = - fsf_req->data.status_read.buffer; + (struct fsf_status_read_buffer *) fsf_req->data; if (fsf_req->status & ZFCP_STATUS_FSFREQ_DISMISSED) { mempool_free(status_buffer, adapter->pool.data_status_read); @@ -1093,7 +1093,7 @@ zfcp_fsf_abort_fcp_command(unsigned long old_req_id, sbale[0].flags |= SBAL_FLAGS0_TYPE_READ; sbale[1].flags |= SBAL_FLAGS_LAST_ENTRY; - fsf_req->data.abort_fcp_command.unit = unit; + fsf_req->data = (unsigned long) unit; /* set handles of unit and its parent port in QTCB */ fsf_req->qtcb->header.lun_handle = unit->handle; @@ -1139,7 +1139,7 @@ static int zfcp_fsf_abort_fcp_command_handler(struct zfcp_fsf_req *new_fsf_req) { int retval = -EINVAL; - struct zfcp_unit *unit = new_fsf_req->data.abort_fcp_command.unit; + struct zfcp_unit *unit; unsigned char status_qual = new_fsf_req->qtcb->header.fsf_status_qual.word[0]; @@ -1150,6 +1150,8 @@ zfcp_fsf_abort_fcp_command_handler(struct zfcp_fsf_req *new_fsf_req) goto skip_fsfstatus; } + unit = (struct zfcp_unit *) new_fsf_req->data; + /* evaluate FSF status in QTCB */ switch (new_fsf_req->qtcb->header.fsf_status) { @@ -1414,7 +1416,7 @@ zfcp_fsf_send_ct(struct zfcp_send_ct *ct, mempool_t *pool, fsf_req->qtcb->header.port_handle = port->handle; fsf_req->qtcb->bottom.support.service_class = adapter->fc_service_class; fsf_req->qtcb->bottom.support.timeout = ct->timeout; - fsf_req->data.send_ct = ct; + fsf_req->data = (unsigned long) ct; /* start QDIO request for this FSF request */ ret = zfcp_fsf_req_send(fsf_req, ct->timer); @@ -1445,10 +1447,10 @@ zfcp_fsf_send_ct(struct zfcp_send_ct *ct, mempool_t *pool, * zfcp_fsf_send_ct_handler - handler for Generic Service requests * @fsf_req: pointer to struct zfcp_fsf_req * - * Data specific for the Generic Service request is passed by - * fsf_req->data.send_ct - * Usually a specific handler for the request is called via - * fsf_req->data.send_ct->handler at end of this function. + * Data specific for the Generic Service request is passed using + * fsf_req->data. There we find the pointer to struct zfcp_send_ct. + * Usually a specific handler for the CT request is called which is + * found in this structure. */ static int zfcp_fsf_send_ct_handler(struct zfcp_fsf_req *fsf_req) @@ -1462,7 +1464,7 @@ zfcp_fsf_send_ct_handler(struct zfcp_fsf_req *fsf_req) u16 subtable, rule, counter; adapter = fsf_req->adapter; - send_ct = fsf_req->data.send_ct; + send_ct = (struct zfcp_send_ct *) fsf_req->data; port = send_ct->port; header = &fsf_req->qtcb->header; bottom = &fsf_req->qtcb->bottom.support; @@ -1714,7 +1716,7 @@ zfcp_fsf_send_els(struct zfcp_send_els *els) fsf_req->qtcb->bottom.support.d_id = d_id; fsf_req->qtcb->bottom.support.service_class = adapter->fc_service_class; fsf_req->qtcb->bottom.support.timeout = ZFCP_ELS_TIMEOUT; - fsf_req->data.send_els = els; + fsf_req->data = (unsigned long) els; sbale = zfcp_qdio_sbale_req(fsf_req, fsf_req->sbal_curr, 0); @@ -1746,10 +1748,10 @@ zfcp_fsf_send_els(struct zfcp_send_els *els) * zfcp_fsf_send_els_handler - handler for ELS commands * @fsf_req: pointer to struct zfcp_fsf_req * - * Data specific for the ELS command is passed by - * fsf_req->data.send_els - * Usually a specific handler for the command is called via - * fsf_req->data.send_els->handler at end of this function. + * Data specific for the ELS command is passed using + * fsf_req->data. There we find the pointer to struct zfcp_send_els. + * Usually a specific handler for the ELS command is called which is + * found in this structure. */ static int zfcp_fsf_send_els_handler(struct zfcp_fsf_req *fsf_req) { @@ -1762,7 +1764,7 @@ static int zfcp_fsf_send_els_handler(struct zfcp_fsf_req *fsf_req) int retval = -EINVAL; u16 subtable, rule, counter; - send_els = fsf_req->data.send_els; + send_els = (struct zfcp_send_els *) fsf_req->data; adapter = send_els->adapter; port = send_els->port; d_id = send_els->d_id; @@ -2211,12 +2213,12 @@ zfcp_fsf_exchange_port_data(struct zfcp_adapter *adapter, goto out; } + fsf_req->data = (unsigned long) data; + sbale = zfcp_qdio_sbale_req(fsf_req, fsf_req->sbal_curr, 0); sbale[0].flags |= SBAL_FLAGS0_TYPE_READ; sbale[1].flags |= SBAL_FLAGS_LAST_ENTRY; - fsf_req->data.port_data = data; - init_timer(timer); timer->function = zfcp_fsf_request_timeout_handler; timer->data = (unsigned long) adapter; @@ -2257,7 +2259,9 @@ static void zfcp_fsf_exchange_port_data_handler(struct zfcp_fsf_req *fsf_req) { struct fsf_qtcb_bottom_port *bottom; - struct fsf_qtcb_bottom_port *data = fsf_req->data.port_data; + struct fsf_qtcb_bottom_port *data; + + data = (struct fsf_qtcb_bottom_port*) fsf_req->data; if (fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR) return; @@ -2312,7 +2316,7 @@ zfcp_fsf_open_port(struct zfcp_erp_action *erp_action) erp_action->fsf_req->qtcb->bottom.support.d_id = erp_action->port->d_id; atomic_set_mask(ZFCP_STATUS_COMMON_OPENING, &erp_action->port->status); - erp_action->fsf_req->data.open_port.port = erp_action->port; + erp_action->fsf_req->data = (unsigned long) erp_action->port; erp_action->fsf_req->erp_action = erp_action; /* start QDIO request for this FSF request */ @@ -2353,7 +2357,7 @@ zfcp_fsf_open_port_handler(struct zfcp_fsf_req *fsf_req) struct fsf_qtcb_header *header; u16 subtable, rule, counter; - port = fsf_req->data.open_port.port; + port = (struct zfcp_port *) fsf_req->data; header = &fsf_req->qtcb->header; if (fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR) { @@ -2566,7 +2570,7 @@ zfcp_fsf_close_port(struct zfcp_erp_action *erp_action) sbale[1].flags |= SBAL_FLAGS_LAST_ENTRY; atomic_set_mask(ZFCP_STATUS_COMMON_CLOSING, &erp_action->port->status); - erp_action->fsf_req->data.close_port.port = erp_action->port; + erp_action->fsf_req->data = (unsigned long) erp_action->port; erp_action->fsf_req->erp_action = erp_action; erp_action->fsf_req->qtcb->header.port_handle = erp_action->port->handle; @@ -2606,7 +2610,7 @@ zfcp_fsf_close_port_handler(struct zfcp_fsf_req *fsf_req) int retval = -EINVAL; struct zfcp_port *port; - port = fsf_req->data.close_port.port; + port = (struct zfcp_port *) fsf_req->data; if (fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR) { /* don't change port status in our bookkeeping */ @@ -2703,7 +2707,7 @@ zfcp_fsf_close_physical_port(struct zfcp_erp_action *erp_action) atomic_set_mask(ZFCP_STATUS_PORT_PHYS_CLOSING, &erp_action->port->status); /* save a pointer to this port */ - erp_action->fsf_req->data.close_physical_port.port = erp_action->port; + erp_action->fsf_req->data = (unsigned long) erp_action->port; /* port to be closeed */ erp_action->fsf_req->qtcb->header.port_handle = erp_action->port->handle; @@ -2747,7 +2751,7 @@ zfcp_fsf_close_physical_port_handler(struct zfcp_fsf_req *fsf_req) struct fsf_qtcb_header *header; u16 subtable, rule, counter; - port = fsf_req->data.close_physical_port.port; + port = (struct zfcp_port *) fsf_req->data; header = &fsf_req->qtcb->header; if (fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR) { @@ -2911,7 +2915,7 @@ zfcp_fsf_open_unit(struct zfcp_erp_action *erp_action) erp_action->fsf_req->qtcb->bottom.support.option = FSF_OPEN_LUN_SUPPRESS_BOXING; atomic_set_mask(ZFCP_STATUS_COMMON_OPENING, &erp_action->unit->status); - erp_action->fsf_req->data.open_unit.unit = erp_action->unit; + erp_action->fsf_req->data = (unsigned long) erp_action->unit; erp_action->fsf_req->erp_action = erp_action; /* start QDIO request for this FSF request */ @@ -2957,7 +2961,7 @@ zfcp_fsf_open_unit_handler(struct zfcp_fsf_req *fsf_req) u16 subtable, rule, counter; u32 allowed, exclusive, readwrite; - unit = fsf_req->data.open_unit.unit; + unit = (struct zfcp_unit *) fsf_req->data; if (fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR) { /* don't change unit status in our bookkeeping */ @@ -3242,7 +3246,7 @@ zfcp_fsf_close_unit(struct zfcp_erp_action *erp_action) erp_action->port->handle; erp_action->fsf_req->qtcb->header.lun_handle = erp_action->unit->handle; atomic_set_mask(ZFCP_STATUS_COMMON_CLOSING, &erp_action->unit->status); - erp_action->fsf_req->data.close_unit.unit = erp_action->unit; + erp_action->fsf_req->data = (unsigned long) erp_action->unit; erp_action->fsf_req->erp_action = erp_action; /* start QDIO request for this FSF request */ @@ -3281,7 +3285,7 @@ zfcp_fsf_close_unit_handler(struct zfcp_fsf_req *fsf_req) int retval = -EINVAL; struct zfcp_unit *unit; - unit = fsf_req->data.close_unit.unit; /* restore unit */ + unit = (struct zfcp_unit *) fsf_req->data; if (fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR) { /* don't change unit status in our bookkeeping */ @@ -3436,21 +3440,14 @@ zfcp_fsf_send_fcp_command_task(struct zfcp_adapter *adapter, goto failed_req_create; } - /* - * associate FSF request with SCSI request - * (need this for look up on abort) - */ - fsf_req->data.send_fcp_command_task.fsf_req = fsf_req; - scsi_cmnd->host_scribble = (char *) &(fsf_req->data); + zfcp_unit_get(unit); + fsf_req->unit = unit; - /* - * associate SCSI command with FSF request - * (need this for look up on normal command completion) - */ - fsf_req->data.send_fcp_command_task.scsi_cmnd = scsi_cmnd; - fsf_req->data.send_fcp_command_task.start_jiffies = jiffies; - fsf_req->data.send_fcp_command_task.unit = unit; - ZFCP_LOG_DEBUG("unit=%p, fcp_lun=0x%016Lx\n", unit, unit->fcp_lun); + /* associate FSF request with SCSI request (for look up on abort) */ + scsi_cmnd->host_scribble = (char *) fsf_req; + + /* associate SCSI command with FSF request */ + fsf_req->data = (unsigned long) scsi_cmnd; /* set handles of unit and its parent port in QTCB */ fsf_req->qtcb->header.lun_handle = unit->handle; @@ -3584,6 +3581,7 @@ zfcp_fsf_send_fcp_command_task(struct zfcp_adapter *adapter, send_failed: no_fit: failed_scsi_cmnd: + zfcp_unit_put(unit); zfcp_fsf_req_free(fsf_req); fsf_req = NULL; scsi_cmnd->host_scribble = NULL; @@ -3640,7 +3638,7 @@ zfcp_fsf_send_fcp_command_task_management(struct zfcp_adapter *adapter, * hold a pointer to the unit being target of this * task management request */ - fsf_req->data.send_fcp_command_task_management.unit = unit; + fsf_req->data = (unsigned long) unit; /* set FSF related fields in QTCB */ fsf_req->qtcb->header.lun_handle = unit->handle; @@ -3706,9 +3704,9 @@ zfcp_fsf_send_fcp_command_handler(struct zfcp_fsf_req *fsf_req) header = &fsf_req->qtcb->header; if (unlikely(fsf_req->status & ZFCP_STATUS_FSFREQ_TASK_MANAGEMENT)) - unit = fsf_req->data.send_fcp_command_task_management.unit; + unit = (struct zfcp_unit *) fsf_req->data; else - unit = fsf_req->data.send_fcp_command_task.unit; + unit = fsf_req->unit; if (unlikely(fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR)) { /* go directly to calls of special handlers */ @@ -3947,6 +3945,8 @@ zfcp_fsf_send_fcp_command_handler(struct zfcp_fsf_req *fsf_req) zfcp_fsf_send_fcp_command_task_management_handler(fsf_req); } else { retval = zfcp_fsf_send_fcp_command_task_handler(fsf_req); + fsf_req->unit = NULL; + zfcp_unit_put(unit); } return retval; } @@ -3970,10 +3970,10 @@ zfcp_fsf_send_fcp_command_task_handler(struct zfcp_fsf_req *fsf_req) u32 sns_len; char *fcp_rsp_info = zfcp_get_fcp_rsp_info_ptr(fcp_rsp_iu); unsigned long flags; - struct zfcp_unit *unit = fsf_req->data.send_fcp_command_task.unit; + struct zfcp_unit *unit = fsf_req->unit; read_lock_irqsave(&fsf_req->adapter->abort_lock, flags); - scpnt = fsf_req->data.send_fcp_command_task.scsi_cmnd; + scpnt = (struct scsi_cmnd *) fsf_req->data; if (unlikely(!scpnt)) { ZFCP_LOG_DEBUG ("Command with fsf_req %p is not associated to " @@ -4198,8 +4198,7 @@ zfcp_fsf_send_fcp_command_task_management_handler(struct zfcp_fsf_req *fsf_req) struct fcp_rsp_iu *fcp_rsp_iu = (struct fcp_rsp_iu *) &(fsf_req->qtcb->bottom.io.fcp_rsp); char *fcp_rsp_info = zfcp_get_fcp_rsp_info_ptr(fcp_rsp_iu); - struct zfcp_unit *unit = - fsf_req->data.send_fcp_command_task_management.unit; + struct zfcp_unit *unit = (struct zfcp_unit *) fsf_req->data; del_timer(&fsf_req->adapter->scsi_er_timer); if (fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR) { diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index 31a76065cf28..fffd12399010 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -414,67 +414,37 @@ zfcp_port_lookup(struct zfcp_adapter *adapter, int channel, scsi_id_t id) return (struct zfcp_port *) NULL; } -/* - * function: zfcp_scsi_eh_abort_handler +/** + * zfcp_scsi_eh_abort_handler - abort the specified SCSI command + * @scpnt: pointer to scsi_cmnd to be aborted + * Return: SUCCESS - command has been aborted and cleaned up in internal + * bookkeeping, SCSI stack won't be called for aborted command + * FAILED - otherwise * - * purpose: tries to abort the specified (timed out) SCSI command - * - * note: We do not need to care for a SCSI command which completes - * normally but late during this abort routine runs. - * We are allowed to return late commands to the SCSI stack. - * It tracks the state of commands and will handle late commands. - * (Usually, the normal completion of late commands is ignored with - * respect to the running abort operation. Grep for 'done_late' - * in the SCSI stacks sources.) - * - * returns: SUCCESS - command has been aborted and cleaned up in internal - * bookkeeping, - * SCSI stack won't be called for aborted command - * FAILED - otherwise + * We do not need to care for a SCSI command which completes normally + * but late during this abort routine runs. We are allowed to return + * late commands to the SCSI stack. It tracks the state of commands and + * will handle late commands. (Usually, the normal completion of late + * commands is ignored with respect to the running abort operation.) */ int -__zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt) +zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt) { + struct Scsi_Host *scsi_host; + struct zfcp_adapter *adapter; + struct zfcp_unit *unit; int retval = SUCCESS; struct zfcp_fsf_req *new_fsf_req, *old_fsf_req; - struct zfcp_adapter *adapter = (struct zfcp_adapter *) scpnt->device->host->hostdata[0]; - struct zfcp_unit *unit = (struct zfcp_unit *) scpnt->device->hostdata; - struct zfcp_port *port = unit->port; - struct Scsi_Host *scsi_host = scpnt->device->host; - union zfcp_req_data *req_data = NULL; unsigned long flags; - u32 status = 0; - /* the components of a abort_dbf record (fixed size record) */ - u64 dbf_scsi_cmnd = (unsigned long) scpnt; - char dbf_opcode[ZFCP_ABORT_DBF_LENGTH]; - wwn_t dbf_wwn = port->wwpn; - fcp_lun_t dbf_fcp_lun = unit->fcp_lun; - u64 dbf_retries = scpnt->retries; - u64 dbf_allowed = scpnt->allowed; - u64 dbf_timeout = 0; - u64 dbf_fsf_req = 0; - u64 dbf_fsf_status = 0; - u64 dbf_fsf_qual[2] = { 0, 0 }; - char dbf_result[ZFCP_ABORT_DBF_LENGTH] = "##undef"; - - memset(dbf_opcode, 0, ZFCP_ABORT_DBF_LENGTH); - memcpy(dbf_opcode, - scpnt->cmnd, - min(scpnt->cmd_len, (unsigned char) ZFCP_ABORT_DBF_LENGTH)); + scsi_host = scpnt->device->host; + adapter = (struct zfcp_adapter *) scsi_host->hostdata[0]; + unit = (struct zfcp_unit *) scpnt->device->hostdata; ZFCP_LOG_INFO("aborting scsi_cmnd=%p on adapter %s\n", scpnt, zfcp_get_busid_by_adapter(adapter)); - spin_unlock_irq(scsi_host->host_lock); - - /* - * Race condition between normal (late) completion and abort has - * to be avoided. - * The entirity of all accesses to scsi_req have to be atomic. - * scsi_req is usually part of the fsf_req and thus we block the - * release of fsf_req as long as we need to access scsi_req. - */ + /* avoid race condition between late normal completion and abort */ write_lock_irqsave(&adapter->abort_lock, flags); /* @@ -484,144 +454,48 @@ __zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt) * this routine returns. (scpnt is parameter passed to this routine * and must not disappear during abort even on late completion.) */ - req_data = (union zfcp_req_data *) scpnt->host_scribble; - /* DEBUG */ - ZFCP_LOG_DEBUG("req_data=%p\n", req_data); - if (!req_data) { - ZFCP_LOG_DEBUG("late command completion overtook abort\n"); - /* - * That's it. - * Do not initiate abort but return SUCCESS. - */ - write_unlock_irqrestore(&adapter->abort_lock, flags); - retval = SUCCESS; - strncpy(dbf_result, "##late1", ZFCP_ABORT_DBF_LENGTH); - goto out; - } - - /* Figure out which fsf_req needs to be aborted. */ - old_fsf_req = req_data->send_fcp_command_task.fsf_req; - - dbf_fsf_req = (unsigned long) old_fsf_req; - dbf_timeout = - (jiffies - req_data->send_fcp_command_task.start_jiffies) / HZ; - - ZFCP_LOG_DEBUG("old_fsf_req=%p\n", old_fsf_req); + old_fsf_req = (struct zfcp_fsf_req *) scpnt->host_scribble; if (!old_fsf_req) { write_unlock_irqrestore(&adapter->abort_lock, flags); ZFCP_LOG_NORMAL("bug: no old fsf request found\n"); - ZFCP_LOG_NORMAL("req_data:\n"); - ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_NORMAL, - (char *) req_data, sizeof (union zfcp_req_data)); ZFCP_LOG_NORMAL("scsi_cmnd:\n"); ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_NORMAL, (char *) scpnt, sizeof (struct scsi_cmnd)); retval = FAILED; - strncpy(dbf_result, "##bug:r", ZFCP_ABORT_DBF_LENGTH); goto out; } - old_fsf_req->data.send_fcp_command_task.scsi_cmnd = NULL; - /* mark old request as being aborted */ + old_fsf_req->data = 0; old_fsf_req->status |= ZFCP_STATUS_FSFREQ_ABORTING; - /* - * We have to collect all information (e.g. unit) needed by - * zfcp_fsf_abort_fcp_command before calling that routine - * since that routine is not allowed to access - * fsf_req which it is going to abort. - * This is because of we need to release fsf_req_list_lock - * before calling zfcp_fsf_abort_fcp_command. - * Since this lock will not be held, fsf_req may complete - * late and may be released meanwhile. - */ - ZFCP_LOG_DEBUG("unit 0x%016Lx (%p)\n", unit->fcp_lun, unit); - /* - * We block (call schedule) - * That's why we must release the lock and enable the - * interrupts before. - * On the other hand we do not need the lock anymore since - * all critical accesses to scsi_req are done. - */ + /* don't access old_fsf_req after releasing the abort_lock */ write_unlock_irqrestore(&adapter->abort_lock, flags); /* call FSF routine which does the abort */ new_fsf_req = zfcp_fsf_abort_fcp_command((unsigned long) old_fsf_req, adapter, unit, 0); - ZFCP_LOG_DEBUG("new_fsf_req=%p\n", new_fsf_req); if (!new_fsf_req) { retval = FAILED; ZFCP_LOG_NORMAL("error: initiation of Abort FCP Cmnd " "failed\n"); - strncpy(dbf_result, "##nores", ZFCP_ABORT_DBF_LENGTH); goto out; } /* wait for completion of abort */ - ZFCP_LOG_DEBUG("waiting for cleanup...\n"); -#if 1 - /* - * FIXME: - * copying zfcp_fsf_req_wait_and_cleanup code is not really nice - */ __wait_event(new_fsf_req->completion_wq, new_fsf_req->status & ZFCP_STATUS_FSFREQ_COMPLETED); - status = new_fsf_req->status; - dbf_fsf_status = new_fsf_req->qtcb->header.fsf_status; - /* - * Ralphs special debug load provides timestamps in the FSF - * status qualifier. This might be specified later if being - * useful for debugging aborts. - */ - dbf_fsf_qual[0] = - *(u64 *) & new_fsf_req->qtcb->header.fsf_status_qual.word[0]; - dbf_fsf_qual[1] = - *(u64 *) & new_fsf_req->qtcb->header.fsf_status_qual.word[2]; zfcp_fsf_req_free(new_fsf_req); -#else - retval = zfcp_fsf_req_wait_and_cleanup(new_fsf_req, - ZFCP_UNINTERRUPTIBLE, &status); -#endif - ZFCP_LOG_DEBUG("Waiting for cleanup complete, status=0x%x\n", status); + /* status should be valid since signals were not permitted */ - if (status & ZFCP_STATUS_FSFREQ_ABORTSUCCEEDED) { + if (new_fsf_req->status & ZFCP_STATUS_FSFREQ_ABORTSUCCEEDED) { retval = SUCCESS; - strncpy(dbf_result, "##succ", ZFCP_ABORT_DBF_LENGTH); - } else if (status & ZFCP_STATUS_FSFREQ_ABORTNOTNEEDED) { + } else if (new_fsf_req->status & ZFCP_STATUS_FSFREQ_ABORTNOTNEEDED) { retval = SUCCESS; - strncpy(dbf_result, "##late2", ZFCP_ABORT_DBF_LENGTH); } else { retval = FAILED; - strncpy(dbf_result, "##fail", ZFCP_ABORT_DBF_LENGTH); } - out: - debug_event(adapter->abort_dbf, 1, &dbf_scsi_cmnd, sizeof (u64)); - debug_event(adapter->abort_dbf, 1, &dbf_opcode, ZFCP_ABORT_DBF_LENGTH); - debug_event(adapter->abort_dbf, 1, &dbf_wwn, sizeof (wwn_t)); - debug_event(adapter->abort_dbf, 1, &dbf_fcp_lun, sizeof (fcp_lun_t)); - debug_event(adapter->abort_dbf, 1, &dbf_retries, sizeof (u64)); - debug_event(adapter->abort_dbf, 1, &dbf_allowed, sizeof (u64)); - debug_event(adapter->abort_dbf, 1, &dbf_timeout, sizeof (u64)); - debug_event(adapter->abort_dbf, 1, &dbf_fsf_req, sizeof (u64)); - debug_event(adapter->abort_dbf, 1, &dbf_fsf_status, sizeof (u64)); - debug_event(adapter->abort_dbf, 1, &dbf_fsf_qual[0], sizeof (u64)); - debug_event(adapter->abort_dbf, 1, &dbf_fsf_qual[1], sizeof (u64)); - debug_text_event(adapter->abort_dbf, 1, dbf_result); - - spin_lock_irq(scsi_host->host_lock); return retval; } -int -zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt) -{ - int rc; - struct Scsi_Host *scsi_host = scpnt->device->host; - spin_lock_irq(scsi_host->host_lock); - rc = __zfcp_scsi_eh_abort_handler(scpnt); - spin_unlock_irq(scsi_host->host_lock); - return rc; -} - /* * function: zfcp_scsi_eh_device_reset_handler * From 77eb1699c76177af2f3d65c8ae7934cf304e0254 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Tue, 13 Sep 2005 21:48:33 +0200 Subject: [PATCH 170/240] [SCSI] zfcp: remove function zfcp_fsf_req_wait_and_cleanup Signed-off-by: Andreas Herrmann Signed-off-by: James Bottomley --- drivers/s390/scsi/zfcp_ext.h | 1 - drivers/s390/scsi/zfcp_fsf.c | 46 ----------------------------------- drivers/s390/scsi/zfcp_scsi.c | 21 +++++++--------- 3 files changed, 9 insertions(+), 59 deletions(-) diff --git a/drivers/s390/scsi/zfcp_ext.h b/drivers/s390/scsi/zfcp_ext.h index cd98a2de9f8f..c400e3b9de97 100644 --- a/drivers/s390/scsi/zfcp_ext.h +++ b/drivers/s390/scsi/zfcp_ext.h @@ -109,7 +109,6 @@ extern int zfcp_fsf_req_create(struct zfcp_adapter *, u32, int, mempool_t *, extern int zfcp_fsf_send_ct(struct zfcp_send_ct *, mempool_t *, struct zfcp_erp_action *); extern int zfcp_fsf_send_els(struct zfcp_send_els *); -extern int zfcp_fsf_req_wait_and_cleanup(struct zfcp_fsf_req *, int, u32 *); extern int zfcp_fsf_send_fcp_command_task(struct zfcp_adapter *, struct zfcp_unit *, struct scsi_cmnd *, diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index 64e2c3480189..929f6c9cad39 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -4548,52 +4548,6 @@ skip_fsfstatus: return retval; } - -/* - * function: zfcp_fsf_req_wait_and_cleanup - * - * purpose: - * - * FIXME(design): signal seems to be <0 !!! - * returns: 0 - request completed (*status is valid), cleanup succ. - * <0 - request completed (*status is valid), cleanup failed - * >0 - signal which interrupted waiting (*status invalid), - * request not completed, no cleanup - * - * *status is a copy of status of completed fsf_req - */ -int -zfcp_fsf_req_wait_and_cleanup(struct zfcp_fsf_req *fsf_req, - int interruptible, u32 * status) -{ - int retval = 0; - int signal = 0; - - if (interruptible) { - __wait_event_interruptible(fsf_req->completion_wq, - fsf_req->status & - ZFCP_STATUS_FSFREQ_COMPLETED, - signal); - if (signal) { - ZFCP_LOG_DEBUG("Caught signal %i while waiting for the " - "completion of the request at %p\n", - signal, fsf_req); - retval = signal; - goto out; - } - } else { - __wait_event(fsf_req->completion_wq, - fsf_req->status & ZFCP_STATUS_FSFREQ_COMPLETED); - } - - *status = fsf_req->status; - - /* cleanup request */ - zfcp_fsf_req_free(fsf_req); - out: - return retval; -} - static inline int zfcp_fsf_req_sbal_check(unsigned long *flags, struct zfcp_qdio_queue *queue, int needed) diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index fffd12399010..3b5d48c61534 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -558,9 +558,8 @@ static int zfcp_task_management_function(struct zfcp_unit *unit, u8 tm_flags) { struct zfcp_adapter *adapter = unit->port->adapter; - int retval; - int status; struct zfcp_fsf_req *fsf_req; + int retval = 0; /* issue task management function */ fsf_req = zfcp_fsf_send_fcp_command_task_management @@ -574,18 +573,16 @@ zfcp_task_management_function(struct zfcp_unit *unit, u8 tm_flags) goto out; } - retval = zfcp_fsf_req_wait_and_cleanup(fsf_req, - ZFCP_UNINTERRUPTIBLE, &status); - /* - * check completion status of task management function - * (status should always be valid since no signals permitted) - */ - if (status & ZFCP_STATUS_FSFREQ_TMFUNCFAILED) + __wait_event(fsf_req->completion_wq, + fsf_req->status & ZFCP_STATUS_FSFREQ_COMPLETED); + + /* check completion status of task management function */ + if (fsf_req->status & ZFCP_STATUS_FSFREQ_TMFUNCFAILED) retval = -EIO; - else if (status & ZFCP_STATUS_FSFREQ_TMFUNCNOTSUPP) + else if (fsf_req->status & ZFCP_STATUS_FSFREQ_TMFUNCNOTSUPP) retval = -ENOTSUPP; - else - retval = 0; + + zfcp_fsf_req_free(fsf_req); out: return retval; } From 810f1e3ea5cc0a812816af97020a27c73441f8e9 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Tue, 13 Sep 2005 21:49:52 +0200 Subject: [PATCH 171/240] [SCSI] zfcp: shorten eh_bus_reset and eh_host_reset handlers Signed-off-by: Andreas Herrmann Signed-off-by: James Bottomley --- drivers/s390/scsi/zfcp_scsi.c | 40 ++++++++++++----------------------- 1 file changed, 14 insertions(+), 26 deletions(-) diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index 3b5d48c61534..eeb5152854ed 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -587,50 +587,38 @@ zfcp_task_management_function(struct zfcp_unit *unit, u8 tm_flags) return retval; } -/* - * function: zfcp_scsi_eh_bus_reset_handler - * - * purpose: - * - * returns: +/** + * zfcp_scsi_eh_bus_reset_handler - reset bus (reopen adapter) */ int zfcp_scsi_eh_bus_reset_handler(struct scsi_cmnd *scpnt) { - int retval = 0; - struct zfcp_unit *unit; + struct zfcp_unit *unit = (struct zfcp_unit*) scpnt->device->hostdata; + struct zfcp_adapter *adapter = unit->port->adapter; - unit = (struct zfcp_unit *) scpnt->device->hostdata; ZFCP_LOG_NORMAL("bus reset because of problems with " "unit 0x%016Lx\n", unit->fcp_lun); - zfcp_erp_adapter_reopen(unit->port->adapter, 0); - zfcp_erp_wait(unit->port->adapter); - retval = SUCCESS; + zfcp_erp_adapter_reopen(adapter, 0); + zfcp_erp_wait(adapter); - return retval; + return SUCCESS; } -/* - * function: zfcp_scsi_eh_host_reset_handler - * - * purpose: - * - * returns: +/** + * zfcp_scsi_eh_host_reset_handler - reset host (reopen adapter) */ int zfcp_scsi_eh_host_reset_handler(struct scsi_cmnd *scpnt) { - int retval = 0; - struct zfcp_unit *unit; + struct zfcp_unit *unit = (struct zfcp_unit*) scpnt->device->hostdata; + struct zfcp_adapter *adapter = unit->port->adapter; - unit = (struct zfcp_unit *) scpnt->device->hostdata; ZFCP_LOG_NORMAL("host reset because of problems with " "unit 0x%016Lx\n", unit->fcp_lun); - zfcp_erp_adapter_reopen(unit->port->adapter, 0); - zfcp_erp_wait(unit->port->adapter); - retval = SUCCESS; + zfcp_erp_adapter_reopen(adapter, 0); + zfcp_erp_wait(adapter); - return retval; + return SUCCESS; } /* From 8a36e4532ea10471f0a8605207d071361d7be2c3 Mon Sep 17 00:00:00 2001 From: Maxim Shchetynin Date: Tue, 13 Sep 2005 21:50:38 +0200 Subject: [PATCH 172/240] [SCSI] zfcp: enhancement of zfcp debug features Debug features (DBFs) els_dbf, cmd_dbf and abt_dbf were removed and san_dbf, hba_dbf and scsi_dbf were introduced. The erp_dbf did not change. The new traces improve debugging of problems with zfcp, scsi-stack, multipath and hardware in the SAN. san_dbf traces things like ELS and CT commands, hba_dbf saves HBA specific information of requests, and scsi_dbf saves FCP and SCSI specific information of requests. Common to all new DBFs is that they provide a so called structured view. This significantly improves readability of the traces. Signed-off-by: Andreas Herrmann Signed-off-by: James Bottomley --- drivers/s390/scsi/Makefile | 2 +- drivers/s390/scsi/zfcp_aux.c | 180 +------ drivers/s390/scsi/zfcp_dbf.c | 988 ++++++++++++++++++++++++++++++++++ drivers/s390/scsi/zfcp_def.h | 213 ++++++-- drivers/s390/scsi/zfcp_ext.h | 26 +- drivers/s390/scsi/zfcp_fsf.c | 241 +++------ drivers/s390/scsi/zfcp_fsf.h | 3 + drivers/s390/scsi/zfcp_qdio.c | 30 +- drivers/s390/scsi/zfcp_scsi.c | 49 +- 9 files changed, 1297 insertions(+), 435 deletions(-) create mode 100644 drivers/s390/scsi/zfcp_dbf.c diff --git a/drivers/s390/scsi/Makefile b/drivers/s390/scsi/Makefile index fc145307a7d4..d6a78f1a2f16 100644 --- a/drivers/s390/scsi/Makefile +++ b/drivers/s390/scsi/Makefile @@ -3,7 +3,7 @@ # zfcp-objs := zfcp_aux.o zfcp_ccw.o zfcp_scsi.o zfcp_erp.o zfcp_qdio.o \ - zfcp_fsf.o zfcp_sysfs_adapter.o zfcp_sysfs_port.o \ + zfcp_fsf.o zfcp_dbf.o zfcp_sysfs_adapter.o zfcp_sysfs_port.o \ zfcp_sysfs_unit.o zfcp_sysfs_driver.o obj-$(CONFIG_ZFCP) += zfcp.o diff --git a/drivers/s390/scsi/zfcp_aux.c b/drivers/s390/scsi/zfcp_aux.c index 17f9989238f8..0b5087f7cabc 100644 --- a/drivers/s390/scsi/zfcp_aux.c +++ b/drivers/s390/scsi/zfcp_aux.c @@ -122,93 +122,6 @@ _zfcp_hex_dump(char *addr, int count) #define ZFCP_LOG_AREA ZFCP_LOG_AREA_OTHER -static inline int -zfcp_fsf_req_is_scsi_cmnd(struct zfcp_fsf_req *fsf_req) -{ - return ((fsf_req->fsf_command == FSF_QTCB_FCP_CMND) && - !(fsf_req->status & ZFCP_STATUS_FSFREQ_TASK_MANAGEMENT)); -} - -void -zfcp_cmd_dbf_event_fsf(const char *text, struct zfcp_fsf_req *fsf_req, - void *add_data, int add_length) -{ - struct zfcp_adapter *adapter = fsf_req->adapter; - struct scsi_cmnd *scsi_cmnd; - int level = 3; - int i; - unsigned long flags; - - spin_lock_irqsave(&adapter->dbf_lock, flags); - if (zfcp_fsf_req_is_scsi_cmnd(fsf_req)) { - scsi_cmnd = (struct scsi_cmnd*) fsf_req->data; - debug_text_event(adapter->cmd_dbf, level, "fsferror"); - debug_text_event(adapter->cmd_dbf, level, text); - debug_event(adapter->cmd_dbf, level, &fsf_req, - sizeof (unsigned long)); - debug_event(adapter->cmd_dbf, level, &fsf_req->seq_no, - sizeof (u32)); - debug_event(adapter->cmd_dbf, level, &scsi_cmnd, - sizeof (unsigned long)); - debug_event(adapter->cmd_dbf, level, &scsi_cmnd->cmnd, - min(ZFCP_CMD_DBF_LENGTH, (int)scsi_cmnd->cmd_len)); - for (i = 0; i < add_length; i += ZFCP_CMD_DBF_LENGTH) - debug_event(adapter->cmd_dbf, - level, - (char *) add_data + i, - min(ZFCP_CMD_DBF_LENGTH, add_length - i)); - } - spin_unlock_irqrestore(&adapter->dbf_lock, flags); -} - -/* XXX additionally log unit if available */ -/* ---> introduce new parameter for unit, see 2.4 code */ -void -zfcp_cmd_dbf_event_scsi(const char *text, struct scsi_cmnd *scsi_cmnd) -{ - struct zfcp_adapter *adapter; - struct zfcp_fsf_req *fsf_req; - int level = ((host_byte(scsi_cmnd->result) != 0) ? 1 : 5); - unsigned long flags; - - adapter = (struct zfcp_adapter *) scsi_cmnd->device->host->hostdata[0]; - fsf_req = (struct zfcp_fsf_req *) scsi_cmnd->host_scribble; - spin_lock_irqsave(&adapter->dbf_lock, flags); - debug_text_event(adapter->cmd_dbf, level, "hostbyte"); - debug_text_event(adapter->cmd_dbf, level, text); - debug_event(adapter->cmd_dbf, level, &scsi_cmnd->result, sizeof (u32)); - debug_event(adapter->cmd_dbf, level, &scsi_cmnd, - sizeof (unsigned long)); - debug_event(adapter->cmd_dbf, level, &scsi_cmnd->cmnd, - min(ZFCP_CMD_DBF_LENGTH, (int)scsi_cmnd->cmd_len)); - if (likely(fsf_req)) { - debug_event(adapter->cmd_dbf, level, &fsf_req, - sizeof (unsigned long)); - debug_event(adapter->cmd_dbf, level, &fsf_req->seq_no, - sizeof (u32)); - } else { - debug_text_event(adapter->cmd_dbf, level, ""); - debug_text_event(adapter->cmd_dbf, level, ""); - } - spin_unlock_irqrestore(&adapter->dbf_lock, flags); -} - -void -zfcp_in_els_dbf_event(struct zfcp_adapter *adapter, const char *text, - struct fsf_status_read_buffer *status_buffer, int length) -{ - int level = 1; - int i; - - debug_text_event(adapter->in_els_dbf, level, text); - debug_event(adapter->in_els_dbf, level, &status_buffer->d_id, 8); - for (i = 0; i < length; i += ZFCP_IN_ELS_DBF_LENGTH) - debug_event(adapter->in_els_dbf, - level, - (char *) status_buffer->payload + i, - min(ZFCP_IN_ELS_DBF_LENGTH, length - i)); -} - /** * zfcp_device_setup - setup function * @str: pointer to parameter string @@ -1015,81 +928,6 @@ zfcp_free_low_mem_buffers(struct zfcp_adapter *adapter) mempool_destroy(adapter->pool.data_gid_pn); } -/** - * zfcp_adapter_debug_register - registers debug feature for an adapter - * @adapter: pointer to adapter for which debug features should be registered - * return: -ENOMEM on error, 0 otherwise - */ -int -zfcp_adapter_debug_register(struct zfcp_adapter *adapter) -{ - char dbf_name[20]; - - /* debug feature area which records SCSI command failures (hostbyte) */ - spin_lock_init(&adapter->dbf_lock); - - sprintf(dbf_name, ZFCP_CMD_DBF_NAME "%s", - zfcp_get_busid_by_adapter(adapter)); - adapter->cmd_dbf = debug_register(dbf_name, ZFCP_CMD_DBF_INDEX, - ZFCP_CMD_DBF_AREAS, - ZFCP_CMD_DBF_LENGTH); - debug_register_view(adapter->cmd_dbf, &debug_hex_ascii_view); - debug_set_level(adapter->cmd_dbf, ZFCP_CMD_DBF_LEVEL); - - /* debug feature area which records SCSI command aborts */ - sprintf(dbf_name, ZFCP_ABORT_DBF_NAME "%s", - zfcp_get_busid_by_adapter(adapter)); - adapter->abort_dbf = debug_register(dbf_name, ZFCP_ABORT_DBF_INDEX, - ZFCP_ABORT_DBF_AREAS, - ZFCP_ABORT_DBF_LENGTH); - debug_register_view(adapter->abort_dbf, &debug_hex_ascii_view); - debug_set_level(adapter->abort_dbf, ZFCP_ABORT_DBF_LEVEL); - - /* debug feature area which records incoming ELS commands */ - sprintf(dbf_name, ZFCP_IN_ELS_DBF_NAME "%s", - zfcp_get_busid_by_adapter(adapter)); - adapter->in_els_dbf = debug_register(dbf_name, ZFCP_IN_ELS_DBF_INDEX, - ZFCP_IN_ELS_DBF_AREAS, - ZFCP_IN_ELS_DBF_LENGTH); - debug_register_view(adapter->in_els_dbf, &debug_hex_ascii_view); - debug_set_level(adapter->in_els_dbf, ZFCP_IN_ELS_DBF_LEVEL); - - /* debug feature area which records erp events */ - sprintf(dbf_name, ZFCP_ERP_DBF_NAME "%s", - zfcp_get_busid_by_adapter(adapter)); - adapter->erp_dbf = debug_register(dbf_name, ZFCP_ERP_DBF_INDEX, - ZFCP_ERP_DBF_AREAS, - ZFCP_ERP_DBF_LENGTH); - debug_register_view(adapter->erp_dbf, &debug_hex_ascii_view); - debug_set_level(adapter->erp_dbf, ZFCP_ERP_DBF_LEVEL); - - if (!(adapter->cmd_dbf && adapter->abort_dbf && - adapter->in_els_dbf && adapter->erp_dbf)) { - zfcp_adapter_debug_unregister(adapter); - return -ENOMEM; - } - - return 0; - -} - -/** - * zfcp_adapter_debug_unregister - unregisters debug feature for an adapter - * @adapter: pointer to adapter for which debug features should be unregistered - */ -void -zfcp_adapter_debug_unregister(struct zfcp_adapter *adapter) -{ - debug_unregister(adapter->abort_dbf); - debug_unregister(adapter->cmd_dbf); - debug_unregister(adapter->erp_dbf); - debug_unregister(adapter->in_els_dbf); - adapter->abort_dbf = NULL; - adapter->cmd_dbf = NULL; - adapter->erp_dbf = NULL; - adapter->in_els_dbf = NULL; -} - void zfcp_dummy_release(struct device *dev) { @@ -1460,10 +1298,6 @@ zfcp_fsf_incoming_els_rscn(struct zfcp_adapter *adapter, /* see FC-FS */ no_entries = (fcp_rscn_head->payload_len / 4); - zfcp_in_els_dbf_event(adapter, "##rscn", status_buffer, - fcp_rscn_head->payload_len); - - debug_text_event(adapter->erp_dbf, 1, "unsol_els_rscn:"); for (i = 1; i < no_entries; i++) { /* skip head and start with 1st element */ fcp_rscn_element++; @@ -1495,8 +1329,6 @@ zfcp_fsf_incoming_els_rscn(struct zfcp_adapter *adapter, (ZFCP_STATUS_PORT_DID_DID, &port->status)) { ZFCP_LOG_INFO("incoming RSCN, trying to open " "port 0x%016Lx\n", port->wwpn); - debug_text_event(adapter->erp_dbf, 1, - "unsol_els_rscnu:"); zfcp_erp_port_reopen(port, ZFCP_STATUS_COMMON_ERP_FAILED); continue; @@ -1522,8 +1354,6 @@ zfcp_fsf_incoming_els_rscn(struct zfcp_adapter *adapter, */ ZFCP_LOG_INFO("incoming RSCN, trying to open " "port 0x%016Lx\n", port->wwpn); - debug_text_event(adapter->erp_dbf, 1, - "unsol_els_rscnk:"); zfcp_test_link(port); } } @@ -1539,8 +1369,6 @@ zfcp_fsf_incoming_els_plogi(struct zfcp_adapter *adapter, struct zfcp_port *port; unsigned long flags; - zfcp_in_els_dbf_event(adapter, "##plogi", status_buffer, 28); - read_lock_irqsave(&zfcp_data.config_lock, flags); list_for_each_entry(port, &adapter->port_list_head, list) { if (port->wwpn == (*(wwn_t *) & els_logi->nport_wwn)) @@ -1554,8 +1382,6 @@ zfcp_fsf_incoming_els_plogi(struct zfcp_adapter *adapter, status_buffer->d_id, zfcp_get_busid_by_adapter(adapter)); } else { - debug_text_event(adapter->erp_dbf, 1, "unsol_els_plogi:"); - debug_event(adapter->erp_dbf, 1, &els_logi->nport_wwn, 8); zfcp_erp_port_forced_reopen(port, 0); } } @@ -1568,8 +1394,6 @@ zfcp_fsf_incoming_els_logo(struct zfcp_adapter *adapter, struct zfcp_port *port; unsigned long flags; - zfcp_in_els_dbf_event(adapter, "##logo", status_buffer, 16); - read_lock_irqsave(&zfcp_data.config_lock, flags); list_for_each_entry(port, &adapter->port_list_head, list) { if (port->wwpn == els_logo->nport_wwpn) @@ -1583,8 +1407,6 @@ zfcp_fsf_incoming_els_logo(struct zfcp_adapter *adapter, status_buffer->d_id, zfcp_get_busid_by_adapter(adapter)); } else { - debug_text_event(adapter->erp_dbf, 1, "unsol_els_logo:"); - debug_event(adapter->erp_dbf, 1, &els_logo->nport_wwpn, 8); zfcp_erp_port_forced_reopen(port, 0); } } @@ -1593,7 +1415,6 @@ static void zfcp_fsf_incoming_els_unknown(struct zfcp_adapter *adapter, struct fsf_status_read_buffer *status_buffer) { - zfcp_in_els_dbf_event(adapter, "##undef", status_buffer, 24); ZFCP_LOG_NORMAL("warning: unknown incoming ELS 0x%08x " "for adapter %s\n", *(u32 *) (status_buffer->payload), zfcp_get_busid_by_adapter(adapter)); @@ -1611,6 +1432,7 @@ zfcp_fsf_incoming_els(struct zfcp_fsf_req *fsf_req) els_type = *(u32 *) (status_buffer->payload); adapter = fsf_req->adapter; + zfcp_san_dbf_event_incoming_els(fsf_req); if (els_type == LS_PLOGI) zfcp_fsf_incoming_els_plogi(adapter, status_buffer); else if (els_type == LS_LOGO) diff --git a/drivers/s390/scsi/zfcp_dbf.c b/drivers/s390/scsi/zfcp_dbf.c new file mode 100644 index 000000000000..fff1537335c7 --- /dev/null +++ b/drivers/s390/scsi/zfcp_dbf.c @@ -0,0 +1,988 @@ +/* + * + * linux/drivers/s390/scsi/zfcp_dbf.c + * + * FCP adapter driver for IBM eServer zSeries + * + * Debugging facilities + * + * (C) Copyright IBM Corp. 2005 + * + * 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, 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., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#define ZFCP_DBF_REVISION "$Revision$" + +#include +#include +#include "zfcp_ext.h" + +static u32 dbfsize = 4; + +module_param(dbfsize, uint, 0400); +MODULE_PARM_DESC(dbfsize, + "number of pages for each debug feature area (default 4)"); + +#define ZFCP_LOG_AREA ZFCP_LOG_AREA_OTHER + +static inline int +zfcp_dbf_stck(char *out_buf, const char *label, unsigned long long stck) +{ + unsigned long long sec; + struct timespec xtime; + int len = 0; + + stck -= 0x8126d60e46000000LL - (0x3c26700LL * 1000000 * 4096); + sec = stck >> 12; + do_div(sec, 1000000); + xtime.tv_sec = sec; + stck -= (sec * 1000000) << 12; + xtime.tv_nsec = ((stck * 1000) >> 12); + len += sprintf(out_buf + len, "%-24s%011lu:%06lu\n", + label, xtime.tv_sec, xtime.tv_nsec); + + return len; +} + +static int zfcp_dbf_tag(char *out_buf, const char *label, const char *tag) +{ + int len = 0, i; + + len += sprintf(out_buf + len, "%-24s", label); + for (i = 0; i < ZFCP_DBF_TAG_SIZE; i++) + len += sprintf(out_buf + len, "%c", tag[i]); + len += sprintf(out_buf + len, "\n"); + + return len; +} + +static int +zfcp_dbf_view(char *out_buf, const char *label, const char *format, ...) +{ + va_list arg; + int len = 0; + + len += sprintf(out_buf + len, "%-24s", label); + va_start(arg, format); + len += vsprintf(out_buf + len, format, arg); + va_end(arg); + len += sprintf(out_buf + len, "\n"); + + return len; +} + +static int +zfcp_dbf_view_dump(char *out_buf, const char *label, + char *buffer, int buflen, int offset, int total_size) +{ + int len = 0; + + if (offset == 0) + len += sprintf(out_buf + len, "%-24s ", label); + + while (buflen--) { + if (offset > 0) { + if ((offset % 32) == 0) + len += sprintf(out_buf + len, "\n%-24c ", ' '); + else if ((offset % 4) == 0) + len += sprintf(out_buf + len, " "); + } + len += sprintf(out_buf + len, "%02x", *buffer++); + if (++offset == total_size) { + len += sprintf(out_buf + len, "\n"); + break; + } + } + + if (total_size == 0) + len += sprintf(out_buf + len, "\n"); + + return len; +} + +static inline int +zfcp_dbf_view_header(debug_info_t * id, struct debug_view *view, int area, + debug_entry_t * entry, char *out_buf) +{ + struct zfcp_dbf_dump *dump = (struct zfcp_dbf_dump *)DEBUG_DATA(entry); + int len = 0; + + if (strncmp(dump->tag, "dump", ZFCP_DBF_TAG_SIZE) != 0) { + len += zfcp_dbf_stck(out_buf + len, "timestamp", + entry->id.stck); + len += zfcp_dbf_view(out_buf + len, "cpu", "%02i", + entry->id.fields.cpuid); + } else { + len += zfcp_dbf_view_dump(out_buf + len, NULL, + dump->data, + dump->size, + dump->offset, dump->total_size); + if ((dump->offset + dump->size) == dump->total_size) + len += sprintf(out_buf + len, "\n"); + } + + return len; +} + +inline void zfcp_hba_dbf_event_fsf_response(struct zfcp_fsf_req *fsf_req) +{ + struct zfcp_adapter *adapter = fsf_req->adapter; + struct fsf_qtcb *qtcb = fsf_req->qtcb; + union fsf_prot_status_qual *prot_status_qual = + &qtcb->prefix.prot_status_qual; + union fsf_status_qual *fsf_status_qual = &qtcb->header.fsf_status_qual; + struct scsi_cmnd *scsi_cmnd; + struct zfcp_port *port; + struct zfcp_unit *unit; + struct zfcp_send_els *send_els; + struct zfcp_hba_dbf_record *rec = &adapter->hba_dbf_buf; + struct zfcp_hba_dbf_record_response *response = &rec->type.response; + int level; + unsigned long flags; + + spin_lock_irqsave(&adapter->hba_dbf_lock, flags); + memset(rec, 0, sizeof(struct zfcp_hba_dbf_record)); + strncpy(rec->tag, "resp", ZFCP_DBF_TAG_SIZE); + + if ((qtcb->prefix.prot_status != FSF_PROT_GOOD) && + (qtcb->prefix.prot_status != FSF_PROT_FSF_STATUS_PRESENTED)) { + strncpy(rec->tag2, "perr", ZFCP_DBF_TAG_SIZE); + level = 1; + } else if (qtcb->header.fsf_status != FSF_GOOD) { + strncpy(rec->tag2, "ferr", ZFCP_DBF_TAG_SIZE); + level = 1; + } else if ((fsf_req->fsf_command == FSF_QTCB_OPEN_PORT_WITH_DID) || + (fsf_req->fsf_command == FSF_QTCB_OPEN_LUN)) { + strncpy(rec->tag2, "open", ZFCP_DBF_TAG_SIZE); + level = 4; + } else if ((prot_status_qual->doubleword[0] != 0) || + (prot_status_qual->doubleword[1] != 0) || + (fsf_status_qual->doubleword[0] != 0) || + (fsf_status_qual->doubleword[1] != 0)) { + strncpy(rec->tag2, "qual", ZFCP_DBF_TAG_SIZE); + level = 3; + } else { + strncpy(rec->tag2, "norm", ZFCP_DBF_TAG_SIZE); + level = 6; + } + + response->fsf_command = fsf_req->fsf_command; + response->fsf_reqid = (unsigned long)fsf_req; + response->fsf_seqno = fsf_req->seq_no; + response->fsf_issued = fsf_req->issued; + response->fsf_prot_status = qtcb->prefix.prot_status; + response->fsf_status = qtcb->header.fsf_status; + memcpy(response->fsf_prot_status_qual, + prot_status_qual, FSF_PROT_STATUS_QUAL_SIZE); + memcpy(response->fsf_status_qual, + fsf_status_qual, FSF_STATUS_QUALIFIER_SIZE); + response->fsf_req_status = fsf_req->status; + response->sbal_first = fsf_req->sbal_first; + response->sbal_curr = fsf_req->sbal_curr; + response->sbal_last = fsf_req->sbal_last; + response->pool = fsf_req->pool != NULL; + response->erp_action = (unsigned long)fsf_req->erp_action; + + switch (fsf_req->fsf_command) { + case FSF_QTCB_FCP_CMND: + if (fsf_req->status & ZFCP_STATUS_FSFREQ_TASK_MANAGEMENT) + break; + scsi_cmnd = (struct scsi_cmnd *)fsf_req->data; + if (scsi_cmnd != NULL) { + response->data.send_fcp.scsi_cmnd + = (unsigned long)scsi_cmnd; + response->data.send_fcp.scsi_serial + = scsi_cmnd->serial_number; + } + break; + + case FSF_QTCB_OPEN_PORT_WITH_DID: + case FSF_QTCB_CLOSE_PORT: + case FSF_QTCB_CLOSE_PHYSICAL_PORT: + port = (struct zfcp_port *)fsf_req->data; + response->data.port.wwpn = port->wwpn; + response->data.port.d_id = port->d_id; + response->data.port.port_handle = qtcb->header.port_handle; + break; + + case FSF_QTCB_OPEN_LUN: + case FSF_QTCB_CLOSE_LUN: + unit = (struct zfcp_unit *)fsf_req->data; + port = unit->port; + response->data.unit.wwpn = port->wwpn; + response->data.unit.fcp_lun = unit->fcp_lun; + response->data.unit.port_handle = qtcb->header.port_handle; + response->data.unit.lun_handle = qtcb->header.lun_handle; + break; + + case FSF_QTCB_SEND_ELS: + send_els = (struct zfcp_send_els *)fsf_req->data; + response->data.send_els.d_id = qtcb->bottom.support.d_id; + response->data.send_els.ls_code = send_els->ls_code >> 24; + break; + + case FSF_QTCB_ABORT_FCP_CMND: + case FSF_QTCB_SEND_GENERIC: + case FSF_QTCB_EXCHANGE_CONFIG_DATA: + case FSF_QTCB_EXCHANGE_PORT_DATA: + case FSF_QTCB_DOWNLOAD_CONTROL_FILE: + case FSF_QTCB_UPLOAD_CONTROL_FILE: + break; + } + + debug_event(adapter->hba_dbf, level, + rec, sizeof(struct zfcp_hba_dbf_record)); + spin_unlock_irqrestore(&adapter->hba_dbf_lock, flags); +} + +inline void +zfcp_hba_dbf_event_fsf_unsol(const char *tag, struct zfcp_adapter *adapter, + struct fsf_status_read_buffer *status_buffer) +{ + struct zfcp_hba_dbf_record *rec = &adapter->hba_dbf_buf; + unsigned long flags; + + spin_lock_irqsave(&adapter->hba_dbf_lock, flags); + memset(rec, 0, sizeof(struct zfcp_hba_dbf_record)); + strncpy(rec->tag, "stat", ZFCP_DBF_TAG_SIZE); + strncpy(rec->tag2, tag, ZFCP_DBF_TAG_SIZE); + + rec->type.status.failed = adapter->status_read_failed; + if (status_buffer != NULL) { + rec->type.status.status_type = status_buffer->status_type; + rec->type.status.status_subtype = status_buffer->status_subtype; + memcpy(&rec->type.status.queue_designator, + &status_buffer->queue_designator, + sizeof(struct fsf_queue_designator)); + + switch (status_buffer->status_type) { + case FSF_STATUS_READ_SENSE_DATA_AVAIL: + rec->type.status.payload_size = + ZFCP_DBF_UNSOL_PAYLOAD_SENSE_DATA_AVAIL; + break; + + case FSF_STATUS_READ_BIT_ERROR_THRESHOLD: + rec->type.status.payload_size = + ZFCP_DBF_UNSOL_PAYLOAD_BIT_ERROR_THRESHOLD; + break; + + case FSF_STATUS_READ_LINK_DOWN: + rec->type.status.payload_size = sizeof(u64); + break; + + } + memcpy(&rec->type.status.payload, + &status_buffer->payload, rec->type.status.payload_size); + } + + debug_event(adapter->hba_dbf, 2, + rec, sizeof(struct zfcp_hba_dbf_record)); + spin_unlock_irqrestore(&adapter->hba_dbf_lock, flags); +} + +inline void +zfcp_hba_dbf_event_qdio(struct zfcp_adapter *adapter, unsigned int status, + unsigned int qdio_error, unsigned int siga_error, + int sbal_index, int sbal_count) +{ + struct zfcp_hba_dbf_record *rec = &adapter->hba_dbf_buf; + unsigned long flags; + + spin_lock_irqsave(&adapter->hba_dbf_lock, flags); + memset(rec, 0, sizeof(struct zfcp_hba_dbf_record)); + strncpy(rec->tag, "qdio", ZFCP_DBF_TAG_SIZE); + rec->type.qdio.status = status; + rec->type.qdio.qdio_error = qdio_error; + rec->type.qdio.siga_error = siga_error; + rec->type.qdio.sbal_index = sbal_index; + rec->type.qdio.sbal_count = sbal_count; + debug_event(adapter->hba_dbf, 0, + rec, sizeof(struct zfcp_hba_dbf_record)); + spin_unlock_irqrestore(&adapter->hba_dbf_lock, flags); +} + +static inline int +zfcp_hba_dbf_view_response(char *out_buf, + struct zfcp_hba_dbf_record_response *rec) +{ + int len = 0; + + len += zfcp_dbf_view(out_buf + len, "fsf_command", "0x%08x", + rec->fsf_command); + len += zfcp_dbf_view(out_buf + len, "fsf_reqid", "0x%0Lx", + rec->fsf_reqid); + len += zfcp_dbf_view(out_buf + len, "fsf_seqno", "0x%08x", + rec->fsf_seqno); + len += zfcp_dbf_stck(out_buf + len, "fsf_issued", rec->fsf_issued); + len += zfcp_dbf_view(out_buf + len, "fsf_prot_status", "0x%08x", + rec->fsf_prot_status); + len += zfcp_dbf_view(out_buf + len, "fsf_status", "0x%08x", + rec->fsf_status); + len += zfcp_dbf_view_dump(out_buf + len, "fsf_prot_status_qual", + rec->fsf_prot_status_qual, + FSF_PROT_STATUS_QUAL_SIZE, + 0, FSF_PROT_STATUS_QUAL_SIZE); + len += zfcp_dbf_view_dump(out_buf + len, "fsf_status_qual", + rec->fsf_status_qual, + FSF_STATUS_QUALIFIER_SIZE, + 0, FSF_STATUS_QUALIFIER_SIZE); + len += zfcp_dbf_view(out_buf + len, "fsf_req_status", "0x%08x", + rec->fsf_req_status); + len += zfcp_dbf_view(out_buf + len, "sbal_first", "0x%02x", + rec->sbal_first); + len += zfcp_dbf_view(out_buf + len, "sbal_curr", "0x%02x", + rec->sbal_curr); + len += zfcp_dbf_view(out_buf + len, "sbal_last", "0x%02x", + rec->sbal_last); + len += zfcp_dbf_view(out_buf + len, "pool", "0x%02x", rec->pool); + + switch (rec->fsf_command) { + case FSF_QTCB_FCP_CMND: + if (rec->fsf_req_status & ZFCP_STATUS_FSFREQ_TASK_MANAGEMENT) + break; + len += zfcp_dbf_view(out_buf + len, "scsi_cmnd", "0x%0Lx", + rec->data.send_fcp.scsi_cmnd); + len += zfcp_dbf_view(out_buf + len, "scsi_serial", "0x%016Lx", + rec->data.send_fcp.scsi_serial); + break; + + case FSF_QTCB_OPEN_PORT_WITH_DID: + case FSF_QTCB_CLOSE_PORT: + case FSF_QTCB_CLOSE_PHYSICAL_PORT: + len += zfcp_dbf_view(out_buf + len, "wwpn", "0x%016Lx", + rec->data.port.wwpn); + len += zfcp_dbf_view(out_buf + len, "d_id", "0x%06x", + rec->data.port.d_id); + len += zfcp_dbf_view(out_buf + len, "port_handle", "0x%08x", + rec->data.port.port_handle); + break; + + case FSF_QTCB_OPEN_LUN: + case FSF_QTCB_CLOSE_LUN: + len += zfcp_dbf_view(out_buf + len, "wwpn", "0x%016Lx", + rec->data.unit.wwpn); + len += zfcp_dbf_view(out_buf + len, "fcp_lun", "0x%016Lx", + rec->data.unit.fcp_lun); + len += zfcp_dbf_view(out_buf + len, "port_handle", "0x%08x", + rec->data.unit.port_handle); + len += zfcp_dbf_view(out_buf + len, "lun_handle", "0x%08x", + rec->data.unit.lun_handle); + break; + + case FSF_QTCB_SEND_ELS: + len += zfcp_dbf_view(out_buf + len, "d_id", "0x%06x", + rec->data.send_els.d_id); + len += zfcp_dbf_view(out_buf + len, "ls_code", "0x%02x", + rec->data.send_els.ls_code); + break; + + case FSF_QTCB_ABORT_FCP_CMND: + case FSF_QTCB_SEND_GENERIC: + case FSF_QTCB_EXCHANGE_CONFIG_DATA: + case FSF_QTCB_EXCHANGE_PORT_DATA: + case FSF_QTCB_DOWNLOAD_CONTROL_FILE: + case FSF_QTCB_UPLOAD_CONTROL_FILE: + break; + } + + return len; +} + +static inline int +zfcp_hba_dbf_view_status(char *out_buf, struct zfcp_hba_dbf_record_status *rec) +{ + int len = 0; + + len += zfcp_dbf_view(out_buf + len, "failed", "0x%02x", rec->failed); + len += zfcp_dbf_view(out_buf + len, "status_type", "0x%08x", + rec->status_type); + len += zfcp_dbf_view(out_buf + len, "status_subtype", "0x%08x", + rec->status_subtype); + len += zfcp_dbf_view_dump(out_buf + len, "queue_designator", + (char *)&rec->queue_designator, + sizeof(struct fsf_queue_designator), + 0, sizeof(struct fsf_queue_designator)); + len += zfcp_dbf_view_dump(out_buf + len, "payload", + (char *)&rec->payload, + rec->payload_size, 0, rec->payload_size); + + return len; +} + +static inline int +zfcp_hba_dbf_view_qdio(char *out_buf, struct zfcp_hba_dbf_record_qdio *rec) +{ + int len = 0; + + len += zfcp_dbf_view(out_buf + len, "status", "0x%08x", rec->status); + len += zfcp_dbf_view(out_buf + len, "qdio_error", "0x%08x", + rec->qdio_error); + len += zfcp_dbf_view(out_buf + len, "siga_error", "0x%08x", + rec->siga_error); + len += zfcp_dbf_view(out_buf + len, "sbal_index", "0x%02x", + rec->sbal_index); + len += zfcp_dbf_view(out_buf + len, "sbal_count", "0x%02x", + rec->sbal_count); + + return len; +} + +static int +zfcp_hba_dbf_view_format(debug_info_t * id, struct debug_view *view, + char *out_buf, const char *in_buf) +{ + struct zfcp_hba_dbf_record *rec = (struct zfcp_hba_dbf_record *)in_buf; + int len = 0; + + if (strncmp(rec->tag, "dump", ZFCP_DBF_TAG_SIZE) == 0) + return 0; + + len += zfcp_dbf_tag(out_buf + len, "tag", rec->tag); + if (isalpha(rec->tag2[0])) + len += zfcp_dbf_tag(out_buf + len, "tag2", rec->tag2); + if (strncmp(rec->tag, "resp", ZFCP_DBF_TAG_SIZE) == 0) + len += zfcp_hba_dbf_view_response(out_buf + len, + &rec->type.response); + else if (strncmp(rec->tag, "stat", ZFCP_DBF_TAG_SIZE) == 0) + len += zfcp_hba_dbf_view_status(out_buf + len, + &rec->type.status); + else if (strncmp(rec->tag, "qdio", ZFCP_DBF_TAG_SIZE) == 0) + len += zfcp_hba_dbf_view_qdio(out_buf + len, &rec->type.qdio); + + len += sprintf(out_buf + len, "\n"); + + return len; +} + +struct debug_view zfcp_hba_dbf_view = { + "structured", + NULL, + &zfcp_dbf_view_header, + &zfcp_hba_dbf_view_format, + NULL, + NULL +}; + +inline void +_zfcp_san_dbf_event_common_ct(const char *tag, struct zfcp_fsf_req *fsf_req, + fc_id_t s_id, fc_id_t d_id, + void *buffer, int buflen) +{ + struct zfcp_send_ct *send_ct = (struct zfcp_send_ct *)fsf_req->data; + struct zfcp_port *port = send_ct->port; + struct zfcp_adapter *adapter = port->adapter; + struct ct_hdr *header = (struct ct_hdr *)buffer; + struct zfcp_san_dbf_record *rec = &adapter->san_dbf_buf; + struct zfcp_san_dbf_record_ct *ct = &rec->type.ct; + unsigned long flags; + + spin_lock_irqsave(&adapter->san_dbf_lock, flags); + memset(rec, 0, sizeof(struct zfcp_san_dbf_record)); + strncpy(rec->tag, tag, ZFCP_DBF_TAG_SIZE); + rec->fsf_reqid = (unsigned long)fsf_req; + rec->fsf_seqno = fsf_req->seq_no; + rec->s_id = s_id; + rec->d_id = d_id; + if (strncmp(tag, "octc", ZFCP_DBF_TAG_SIZE) == 0) { + ct->type.request.cmd_req_code = header->cmd_rsp_code; + ct->type.request.revision = header->revision; + ct->type.request.gs_type = header->gs_type; + ct->type.request.gs_subtype = header->gs_subtype; + ct->type.request.options = header->options; + ct->type.request.max_res_size = header->max_res_size; + } else if (strncmp(tag, "rctc", ZFCP_DBF_TAG_SIZE) == 0) { + ct->type.response.cmd_rsp_code = header->cmd_rsp_code; + ct->type.response.revision = header->revision; + ct->type.response.reason_code = header->reason_code; + ct->type.response.reason_code_expl = header->reason_code_expl; + ct->type.response.vendor_unique = header->vendor_unique; + } + ct->payload_size = + min(buflen - (int)sizeof(struct ct_hdr), ZFCP_DBF_CT_PAYLOAD); + memcpy(ct->payload, buffer + sizeof(struct ct_hdr), ct->payload_size); + debug_event(adapter->san_dbf, 3, + rec, sizeof(struct zfcp_san_dbf_record)); + spin_unlock_irqrestore(&adapter->san_dbf_lock, flags); +} + +inline void zfcp_san_dbf_event_ct_request(struct zfcp_fsf_req *fsf_req) +{ + struct zfcp_send_ct *ct = (struct zfcp_send_ct *)fsf_req->data; + struct zfcp_port *port = ct->port; + struct zfcp_adapter *adapter = port->adapter; + + _zfcp_san_dbf_event_common_ct("octc", + fsf_req, adapter->s_id, port->d_id, + zfcp_sg_to_address(ct->req), + ct->req->length); +} + +inline void zfcp_san_dbf_event_ct_response(struct zfcp_fsf_req *fsf_req) +{ + struct zfcp_send_ct *ct = (struct zfcp_send_ct *)fsf_req->data; + struct zfcp_port *port = ct->port; + struct zfcp_adapter *adapter = port->adapter; + + _zfcp_san_dbf_event_common_ct("rctc", + fsf_req, port->d_id, adapter->s_id, + zfcp_sg_to_address(ct->resp), + ct->resp->length); +} + +static inline void +_zfcp_san_dbf_event_common_els(const char *tag, int level, + struct zfcp_fsf_req *fsf_req, + fc_id_t s_id, fc_id_t d_id, u8 ls_code, + void *buffer, int buflen) +{ + struct zfcp_adapter *adapter = fsf_req->adapter; + struct zfcp_san_dbf_record *rec = &adapter->san_dbf_buf; + struct zfcp_dbf_dump *dump = (struct zfcp_dbf_dump *)rec; + unsigned long flags; + int offset = 0; + + spin_lock_irqsave(&adapter->san_dbf_lock, flags); + do { + memset(rec, 0, sizeof(struct zfcp_san_dbf_record)); + if (offset == 0) { + strncpy(rec->tag, tag, ZFCP_DBF_TAG_SIZE); + rec->fsf_reqid = (unsigned long)fsf_req; + rec->fsf_seqno = fsf_req->seq_no; + rec->s_id = s_id; + rec->d_id = d_id; + rec->type.els.ls_code = ls_code; + buflen = min(buflen, ZFCP_DBF_ELS_MAX_PAYLOAD); + rec->type.els.payload_size = buflen; + memcpy(rec->type.els.payload, + buffer, min(buflen, ZFCP_DBF_ELS_PAYLOAD)); + offset += min(buflen, ZFCP_DBF_ELS_PAYLOAD); + } else { + strncpy(dump->tag, "dump", ZFCP_DBF_TAG_SIZE); + dump->total_size = buflen; + dump->offset = offset; + dump->size = min(buflen - offset, + (int)sizeof(struct zfcp_san_dbf_record) + - (int)sizeof(struct zfcp_dbf_dump)); + memcpy(dump->data, buffer + offset, dump->size); + offset += dump->size; + } + debug_event(adapter->san_dbf, level, + rec, sizeof(struct zfcp_san_dbf_record)); + } while (offset < buflen); + spin_unlock_irqrestore(&adapter->san_dbf_lock, flags); +} + +inline void zfcp_san_dbf_event_els_request(struct zfcp_fsf_req *fsf_req) +{ + struct zfcp_send_els *els = (struct zfcp_send_els *)fsf_req->data; + + _zfcp_san_dbf_event_common_els("oels", 2, + fsf_req, els->adapter->s_id, els->d_id, + *(u8 *) zfcp_sg_to_address(els->req), + zfcp_sg_to_address(els->req), + els->req->length); +} + +inline void zfcp_san_dbf_event_els_response(struct zfcp_fsf_req *fsf_req) +{ + struct zfcp_send_els *els = (struct zfcp_send_els *)fsf_req->data; + + _zfcp_san_dbf_event_common_els("rels", 2, + fsf_req, els->d_id, els->adapter->s_id, + *(u8 *) zfcp_sg_to_address(els->req), + zfcp_sg_to_address(els->resp), + els->resp->length); +} + +inline void zfcp_san_dbf_event_incoming_els(struct zfcp_fsf_req *fsf_req) +{ + struct zfcp_adapter *adapter = fsf_req->adapter; + struct fsf_status_read_buffer *status_buffer = + (struct fsf_status_read_buffer *)fsf_req->data; + int length = (int)status_buffer->length - + (int)((void *)&status_buffer->payload - (void *)status_buffer); + + _zfcp_san_dbf_event_common_els("iels", 1, + fsf_req, status_buffer->d_id, + adapter->s_id, + *(u8 *) status_buffer->payload, + (void *)status_buffer->payload, length); +} + +static int +zfcp_san_dbf_view_format(debug_info_t * id, struct debug_view *view, + char *out_buf, const char *in_buf) +{ + struct zfcp_san_dbf_record *rec = (struct zfcp_san_dbf_record *)in_buf; + char *buffer = NULL; + int buflen = 0, total = 0; + int len = 0; + + if (strncmp(rec->tag, "dump", ZFCP_DBF_TAG_SIZE) == 0) + return 0; + + len += zfcp_dbf_tag(out_buf + len, "tag", rec->tag); + len += zfcp_dbf_view(out_buf + len, "fsf_reqid", "0x%0Lx", + rec->fsf_reqid); + len += zfcp_dbf_view(out_buf + len, "fsf_seqno", "0x%08x", + rec->fsf_seqno); + len += zfcp_dbf_view(out_buf + len, "s_id", "0x%06x", rec->s_id); + len += zfcp_dbf_view(out_buf + len, "d_id", "0x%06x", rec->d_id); + + if (strncmp(rec->tag, "octc", ZFCP_DBF_TAG_SIZE) == 0) { + len += zfcp_dbf_view(out_buf + len, "cmd_req_code", "0x%04x", + rec->type.ct.type.request.cmd_req_code); + len += zfcp_dbf_view(out_buf + len, "revision", "0x%02x", + rec->type.ct.type.request.revision); + len += zfcp_dbf_view(out_buf + len, "gs_type", "0x%02x", + rec->type.ct.type.request.gs_type); + len += zfcp_dbf_view(out_buf + len, "gs_subtype", "0x%02x", + rec->type.ct.type.request.gs_subtype); + len += zfcp_dbf_view(out_buf + len, "options", "0x%02x", + rec->type.ct.type.request.options); + len += zfcp_dbf_view(out_buf + len, "max_res_size", "0x%04x", + rec->type.ct.type.request.max_res_size); + total = rec->type.ct.payload_size; + buffer = rec->type.ct.payload; + buflen = min(total, ZFCP_DBF_CT_PAYLOAD); + } else if (strncmp(rec->tag, "rctc", ZFCP_DBF_TAG_SIZE) == 0) { + len += zfcp_dbf_view(out_buf + len, "cmd_rsp_code", "0x%04x", + rec->type.ct.type.response.cmd_rsp_code); + len += zfcp_dbf_view(out_buf + len, "revision", "0x%02x", + rec->type.ct.type.response.revision); + len += zfcp_dbf_view(out_buf + len, "reason_code", "0x%02x", + rec->type.ct.type.response.reason_code); + len += + zfcp_dbf_view(out_buf + len, "reason_code_expl", "0x%02x", + rec->type.ct.type.response.reason_code_expl); + len += + zfcp_dbf_view(out_buf + len, "vendor_unique", "0x%02x", + rec->type.ct.type.response.vendor_unique); + total = rec->type.ct.payload_size; + buffer = rec->type.ct.payload; + buflen = min(total, ZFCP_DBF_CT_PAYLOAD); + } else if (strncmp(rec->tag, "oels", ZFCP_DBF_TAG_SIZE) == 0 || + strncmp(rec->tag, "rels", ZFCP_DBF_TAG_SIZE) == 0 || + strncmp(rec->tag, "iels", ZFCP_DBF_TAG_SIZE) == 0) { + len += zfcp_dbf_view(out_buf + len, "ls_code", "0x%02x", + rec->type.els.ls_code); + total = rec->type.els.payload_size; + buffer = rec->type.els.payload; + buflen = min(total, ZFCP_DBF_ELS_PAYLOAD); + } + + len += zfcp_dbf_view_dump(out_buf + len, "payload", + buffer, buflen, 0, total); + + if (buflen == total) + len += sprintf(out_buf + len, "\n"); + + return len; +} + +struct debug_view zfcp_san_dbf_view = { + "structured", + NULL, + &zfcp_dbf_view_header, + &zfcp_san_dbf_view_format, + NULL, + NULL +}; + +static inline void +_zfcp_scsi_dbf_event_common(const char *tag, const char *tag2, int level, + struct zfcp_adapter *adapter, + struct scsi_cmnd *scsi_cmnd, + struct zfcp_fsf_req *new_fsf_req) +{ + struct zfcp_fsf_req *fsf_req = + (struct zfcp_fsf_req *)scsi_cmnd->host_scribble; + struct zfcp_scsi_dbf_record *rec = &adapter->scsi_dbf_buf; + struct zfcp_dbf_dump *dump = (struct zfcp_dbf_dump *)rec; + unsigned long flags; + struct fcp_rsp_iu *fcp_rsp; + char *fcp_rsp_info = NULL, *fcp_sns_info = NULL; + int offset = 0, buflen = 0; + + spin_lock_irqsave(&adapter->scsi_dbf_lock, flags); + do { + memset(rec, 0, sizeof(struct zfcp_scsi_dbf_record)); + if (offset == 0) { + strncpy(rec->tag, tag, ZFCP_DBF_TAG_SIZE); + strncpy(rec->tag2, tag2, ZFCP_DBF_TAG_SIZE); + if (scsi_cmnd->device) { + rec->scsi_id = scsi_cmnd->device->id; + rec->scsi_lun = scsi_cmnd->device->lun; + } + rec->scsi_result = scsi_cmnd->result; + rec->scsi_cmnd = (unsigned long)scsi_cmnd; + rec->scsi_serial = scsi_cmnd->serial_number; + memcpy(rec->scsi_opcode, + &scsi_cmnd->cmnd, + min((int)scsi_cmnd->cmd_len, + ZFCP_DBF_SCSI_OPCODE)); + rec->scsi_retries = scsi_cmnd->retries; + rec->scsi_allowed = scsi_cmnd->allowed; + if (fsf_req != NULL) { + fcp_rsp = (struct fcp_rsp_iu *) + &(fsf_req->qtcb->bottom.io.fcp_rsp); + fcp_rsp_info = + zfcp_get_fcp_rsp_info_ptr(fcp_rsp); + fcp_sns_info = + zfcp_get_fcp_sns_info_ptr(fcp_rsp); + + rec->type.fcp.rsp_validity = + fcp_rsp->validity.value; + rec->type.fcp.rsp_scsi_status = + fcp_rsp->scsi_status; + rec->type.fcp.rsp_resid = fcp_rsp->fcp_resid; + if (fcp_rsp->validity.bits.fcp_rsp_len_valid) + rec->type.fcp.rsp_code = + *(fcp_rsp_info + 3); + if (fcp_rsp->validity.bits.fcp_sns_len_valid) { + buflen = min((int)fcp_rsp->fcp_sns_len, + ZFCP_DBF_SCSI_MAX_FCP_SNS_INFO); + rec->type.fcp.sns_info_len = buflen; + memcpy(rec->type.fcp.sns_info, + fcp_sns_info, + min(buflen, + ZFCP_DBF_SCSI_FCP_SNS_INFO)); + offset += min(buflen, + ZFCP_DBF_SCSI_FCP_SNS_INFO); + } + + rec->fsf_reqid = (unsigned long)fsf_req; + rec->fsf_seqno = fsf_req->seq_no; + rec->fsf_issued = fsf_req->issued; + } + if (new_fsf_req != NULL) { + rec->type.new_fsf_req.fsf_reqid = + (unsigned long) + new_fsf_req; + rec->type.new_fsf_req.fsf_seqno = + new_fsf_req->seq_no; + rec->type.new_fsf_req.fsf_issued = + new_fsf_req->issued; + } + } else { + strncpy(dump->tag, "dump", ZFCP_DBF_TAG_SIZE); + dump->total_size = buflen; + dump->offset = offset; + dump->size = min(buflen - offset, + (int)sizeof(struct + zfcp_scsi_dbf_record) - + (int)sizeof(struct zfcp_dbf_dump)); + memcpy(dump->data, fcp_sns_info + offset, dump->size); + offset += dump->size; + } + debug_event(adapter->scsi_dbf, level, + rec, sizeof(struct zfcp_scsi_dbf_record)); + } while (offset < buflen); + spin_unlock_irqrestore(&adapter->scsi_dbf_lock, flags); +} + +inline void +zfcp_scsi_dbf_event_result(const char *tag, int level, + struct zfcp_adapter *adapter, + struct scsi_cmnd *scsi_cmnd) +{ + _zfcp_scsi_dbf_event_common("rslt", + tag, level, adapter, scsi_cmnd, NULL); +} + +inline void +zfcp_scsi_dbf_event_abort(const char *tag, struct zfcp_adapter *adapter, + struct scsi_cmnd *scsi_cmnd, + struct zfcp_fsf_req *new_fsf_req) +{ + _zfcp_scsi_dbf_event_common("abrt", + tag, 1, adapter, scsi_cmnd, new_fsf_req); +} + +inline void +zfcp_scsi_dbf_event_devreset(const char *tag, u8 flag, struct zfcp_unit *unit, + struct scsi_cmnd *scsi_cmnd) +{ + struct zfcp_adapter *adapter = unit->port->adapter; + + _zfcp_scsi_dbf_event_common(flag == FCP_TARGET_RESET ? "trst" : "lrst", + tag, 1, adapter, scsi_cmnd, NULL); +} + +static int +zfcp_scsi_dbf_view_format(debug_info_t * id, struct debug_view *view, + char *out_buf, const char *in_buf) +{ + struct zfcp_scsi_dbf_record *rec = + (struct zfcp_scsi_dbf_record *)in_buf; + int len = 0; + + if (strncmp(rec->tag, "dump", ZFCP_DBF_TAG_SIZE) == 0) + return 0; + + len += zfcp_dbf_tag(out_buf + len, "tag", rec->tag); + len += zfcp_dbf_tag(out_buf + len, "tag2", rec->tag2); + len += zfcp_dbf_view(out_buf + len, "scsi_id", "0x%08x", rec->scsi_id); + len += zfcp_dbf_view(out_buf + len, "scsi_lun", "0x%08x", + rec->scsi_lun); + len += zfcp_dbf_view(out_buf + len, "scsi_result", "0x%08x", + rec->scsi_result); + len += zfcp_dbf_view(out_buf + len, "scsi_cmnd", "0x%0Lx", + rec->scsi_cmnd); + len += zfcp_dbf_view(out_buf + len, "scsi_serial", "0x%016Lx", + rec->scsi_serial); + len += zfcp_dbf_view_dump(out_buf + len, "scsi_opcode", + rec->scsi_opcode, + ZFCP_DBF_SCSI_OPCODE, + 0, ZFCP_DBF_SCSI_OPCODE); + len += zfcp_dbf_view(out_buf + len, "scsi_retries", "0x%02x", + rec->scsi_retries); + len += zfcp_dbf_view(out_buf + len, "scsi_allowed", "0x%02x", + rec->scsi_allowed); + len += zfcp_dbf_view(out_buf + len, "fsf_reqid", "0x%0Lx", + rec->fsf_reqid); + len += zfcp_dbf_view(out_buf + len, "fsf_seqno", "0x%08x", + rec->fsf_seqno); + len += zfcp_dbf_stck(out_buf + len, "fsf_issued", rec->fsf_issued); + if (strncmp(rec->tag, "rslt", ZFCP_DBF_TAG_SIZE) == 0) { + len += + zfcp_dbf_view(out_buf + len, "fcp_rsp_validity", "0x%02x", + rec->type.fcp.rsp_validity); + len += + zfcp_dbf_view(out_buf + len, "fcp_rsp_scsi_status", + "0x%02x", rec->type.fcp.rsp_scsi_status); + len += + zfcp_dbf_view(out_buf + len, "fcp_rsp_resid", "0x%08x", + rec->type.fcp.rsp_resid); + len += + zfcp_dbf_view(out_buf + len, "fcp_rsp_code", "0x%08x", + rec->type.fcp.rsp_code); + len += + zfcp_dbf_view(out_buf + len, "fcp_sns_info_len", "0x%08x", + rec->type.fcp.sns_info_len); + len += + zfcp_dbf_view_dump(out_buf + len, "fcp_sns_info", + rec->type.fcp.sns_info, + min((int)rec->type.fcp.sns_info_len, + ZFCP_DBF_SCSI_FCP_SNS_INFO), 0, + rec->type.fcp.sns_info_len); + } else if (strncmp(rec->tag, "abrt", ZFCP_DBF_TAG_SIZE) == 0) { + len += zfcp_dbf_view(out_buf + len, "fsf_reqid_abort", "0x%0Lx", + rec->type.new_fsf_req.fsf_reqid); + len += zfcp_dbf_view(out_buf + len, "fsf_seqno_abort", "0x%08x", + rec->type.new_fsf_req.fsf_seqno); + len += zfcp_dbf_stck(out_buf + len, "fsf_issued", + rec->type.new_fsf_req.fsf_issued); + } else if ((strncmp(rec->tag, "trst", ZFCP_DBF_TAG_SIZE) == 0) || + (strncmp(rec->tag, "lrst", ZFCP_DBF_TAG_SIZE) == 0)) { + len += zfcp_dbf_view(out_buf + len, "fsf_reqid_reset", "0x%0Lx", + rec->type.new_fsf_req.fsf_reqid); + len += zfcp_dbf_view(out_buf + len, "fsf_seqno_reset", "0x%08x", + rec->type.new_fsf_req.fsf_seqno); + len += zfcp_dbf_stck(out_buf + len, "fsf_issued", + rec->type.new_fsf_req.fsf_issued); + } + + len += sprintf(out_buf + len, "\n"); + + return len; +} + +struct debug_view zfcp_scsi_dbf_view = { + "structured", + NULL, + &zfcp_dbf_view_header, + &zfcp_scsi_dbf_view_format, + NULL, + NULL +}; + +/** + * zfcp_adapter_debug_register - registers debug feature for an adapter + * @adapter: pointer to adapter for which debug features should be registered + * return: -ENOMEM on error, 0 otherwise + */ +int zfcp_adapter_debug_register(struct zfcp_adapter *adapter) +{ + char dbf_name[DEBUG_MAX_NAME_LEN]; + + /* debug feature area which records recovery activity */ + spin_lock_init(&adapter->erp_dbf_lock); + sprintf(dbf_name, "zfcp_%s_erp", zfcp_get_busid_by_adapter(adapter)); + adapter->erp_dbf = debug_register(dbf_name, dbfsize, 2, + sizeof(struct zfcp_erp_dbf_record)); + if (!adapter->erp_dbf) + goto failed; + debug_register_view(adapter->erp_dbf, &debug_hex_ascii_view); + debug_set_level(adapter->erp_dbf, 3); + + /* debug feature area which records HBA (FSF and QDIO) conditions */ + spin_lock_init(&adapter->hba_dbf_lock); + sprintf(dbf_name, "zfcp_%s_hba", zfcp_get_busid_by_adapter(adapter)); + adapter->hba_dbf = debug_register(dbf_name, dbfsize, 1, + sizeof(struct zfcp_hba_dbf_record)); + if (!adapter->hba_dbf) + goto failed; + debug_register_view(adapter->hba_dbf, &debug_hex_ascii_view); + debug_register_view(adapter->hba_dbf, &zfcp_hba_dbf_view); + debug_set_level(adapter->hba_dbf, 3); + + /* debug feature area which records SAN command failures and recovery */ + spin_lock_init(&adapter->san_dbf_lock); + sprintf(dbf_name, "zfcp_%s_san", zfcp_get_busid_by_adapter(adapter)); + adapter->san_dbf = debug_register(dbf_name, dbfsize, 1, + sizeof(struct zfcp_san_dbf_record)); + if (!adapter->san_dbf) + goto failed; + debug_register_view(adapter->san_dbf, &debug_hex_ascii_view); + debug_register_view(adapter->san_dbf, &zfcp_san_dbf_view); + debug_set_level(adapter->san_dbf, 6); + + /* debug feature area which records SCSI command failures and recovery */ + spin_lock_init(&adapter->scsi_dbf_lock); + sprintf(dbf_name, "zfcp_%s_scsi", zfcp_get_busid_by_adapter(adapter)); + adapter->scsi_dbf = debug_register(dbf_name, dbfsize, 1, + sizeof(struct zfcp_scsi_dbf_record)); + if (!adapter->scsi_dbf) + goto failed; + debug_register_view(adapter->scsi_dbf, &debug_hex_ascii_view); + debug_register_view(adapter->scsi_dbf, &zfcp_scsi_dbf_view); + debug_set_level(adapter->scsi_dbf, 3); + + return 0; + + failed: + zfcp_adapter_debug_unregister(adapter); + + return -ENOMEM; +} + +/** + * zfcp_adapter_debug_unregister - unregisters debug feature for an adapter + * @adapter: pointer to adapter for which debug features should be unregistered + */ +void zfcp_adapter_debug_unregister(struct zfcp_adapter *adapter) +{ + debug_unregister(adapter->scsi_dbf); + debug_unregister(adapter->san_dbf); + debug_unregister(adapter->hba_dbf); + debug_unregister(adapter->erp_dbf); + adapter->scsi_dbf = NULL; + adapter->san_dbf = NULL; + adapter->hba_dbf = NULL; + adapter->erp_dbf = NULL; +} + +#undef ZFCP_LOG_AREA diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h index 9160e68c4a21..fc5bb6f31808 100644 --- a/drivers/s390/scsi/zfcp_def.h +++ b/drivers/s390/scsi/zfcp_def.h @@ -66,7 +66,7 @@ /********************* GENERAL DEFINES *********************************/ /* zfcp version number, it consists of major, minor, and patch-level number */ -#define ZFCP_VERSION "4.3.0" +#define ZFCP_VERSION "4.4.0" /** * zfcp_sg_to_address - determine kernel address from struct scatterlist @@ -280,6 +280,171 @@ struct fcp_logo { wwn_t nport_wwpn; } __attribute__((packed)); +/* + * DBF stuff + */ +#define ZFCP_DBF_TAG_SIZE 4 + +struct zfcp_dbf_dump { + u8 tag[ZFCP_DBF_TAG_SIZE]; + u32 total_size; /* size of total dump data */ + u32 offset; /* how much data has being already dumped */ + u32 size; /* how much data comes with this record */ + u8 data[]; /* dump data */ +} __attribute__ ((packed)); + +/* FIXME: to be inflated when reworking the erp dbf */ +struct zfcp_erp_dbf_record { + u8 dummy[16]; +} __attribute__ ((packed)); + +struct zfcp_hba_dbf_record_response { + u32 fsf_command; + u64 fsf_reqid; + u32 fsf_seqno; + u64 fsf_issued; + u32 fsf_prot_status; + u32 fsf_status; + u8 fsf_prot_status_qual[FSF_PROT_STATUS_QUAL_SIZE]; + u8 fsf_status_qual[FSF_STATUS_QUALIFIER_SIZE]; + u32 fsf_req_status; + u8 sbal_first; + u8 sbal_curr; + u8 sbal_last; + u8 pool; + u64 erp_action; + union { + struct { + u64 scsi_cmnd; + u64 scsi_serial; + } send_fcp; + struct { + u64 wwpn; + u32 d_id; + u32 port_handle; + } port; + struct { + u64 wwpn; + u64 fcp_lun; + u32 port_handle; + u32 lun_handle; + } unit; + struct { + u32 d_id; + u8 ls_code; + } send_els; + } data; +} __attribute__ ((packed)); + +struct zfcp_hba_dbf_record_status { + u8 failed; + u32 status_type; + u32 status_subtype; + struct fsf_queue_designator + queue_designator; + u32 payload_size; +#define ZFCP_DBF_UNSOL_PAYLOAD 80 +#define ZFCP_DBF_UNSOL_PAYLOAD_SENSE_DATA_AVAIL 32 +#define ZFCP_DBF_UNSOL_PAYLOAD_BIT_ERROR_THRESHOLD 56 +#define ZFCP_DBF_UNSOL_PAYLOAD_FEATURE_UPDATE_ALERT 2 * sizeof(u32) + u8 payload[ZFCP_DBF_UNSOL_PAYLOAD]; +} __attribute__ ((packed)); + +struct zfcp_hba_dbf_record_qdio { + u32 status; + u32 qdio_error; + u32 siga_error; + u8 sbal_index; + u8 sbal_count; +} __attribute__ ((packed)); + +struct zfcp_hba_dbf_record { + u8 tag[ZFCP_DBF_TAG_SIZE]; + u8 tag2[ZFCP_DBF_TAG_SIZE]; + union { + struct zfcp_hba_dbf_record_response response; + struct zfcp_hba_dbf_record_status status; + struct zfcp_hba_dbf_record_qdio qdio; + } type; +} __attribute__ ((packed)); + +struct zfcp_san_dbf_record_ct { + union { + struct { + u16 cmd_req_code; + u8 revision; + u8 gs_type; + u8 gs_subtype; + u8 options; + u16 max_res_size; + } request; + struct { + u16 cmd_rsp_code; + u8 revision; + u8 reason_code; + u8 reason_code_expl; + u8 vendor_unique; + } response; + } type; + u32 payload_size; +#define ZFCP_DBF_CT_PAYLOAD 24 + u8 payload[ZFCP_DBF_CT_PAYLOAD]; +} __attribute__ ((packed)); + +struct zfcp_san_dbf_record_els { + u8 ls_code; + u32 payload_size; +#define ZFCP_DBF_ELS_PAYLOAD 32 +#define ZFCP_DBF_ELS_MAX_PAYLOAD 1024 + u8 payload[ZFCP_DBF_ELS_PAYLOAD]; +} __attribute__ ((packed)); + +struct zfcp_san_dbf_record { + u8 tag[ZFCP_DBF_TAG_SIZE]; + u64 fsf_reqid; + u32 fsf_seqno; + u32 s_id; + u32 d_id; + union { + struct zfcp_san_dbf_record_ct ct; + struct zfcp_san_dbf_record_els els; + } type; +} __attribute__ ((packed)); + +struct zfcp_scsi_dbf_record { + u8 tag[ZFCP_DBF_TAG_SIZE]; + u8 tag2[ZFCP_DBF_TAG_SIZE]; + u32 scsi_id; + u32 scsi_lun; + u32 scsi_result; + u64 scsi_cmnd; + u64 scsi_serial; +#define ZFCP_DBF_SCSI_OPCODE 16 + u8 scsi_opcode[ZFCP_DBF_SCSI_OPCODE]; + u8 scsi_retries; + u8 scsi_allowed; + u64 fsf_reqid; + u32 fsf_seqno; + u64 fsf_issued; + union { + struct { + u64 fsf_reqid; + u32 fsf_seqno; + u64 fsf_issued; + } new_fsf_req; + struct { + u8 rsp_validity; + u8 rsp_scsi_status; + u32 rsp_resid; + u8 rsp_code; +#define ZFCP_DBF_SCSI_FCP_SNS_INFO 16 +#define ZFCP_DBF_SCSI_MAX_FCP_SNS_INFO 256 + u32 sns_info_len; + u8 sns_info[ZFCP_DBF_SCSI_FCP_SNS_INFO]; + } fcp; + } type; +} __attribute__ ((packed)); + /* * FC-FS stuff */ @@ -339,34 +504,6 @@ struct zfcp_rc_entry { */ #define ZFCP_CT_TIMEOUT (3 * R_A_TOV) - -/***************** S390 DEBUG FEATURE SPECIFIC DEFINES ***********************/ - -/* debug feature entries per adapter */ -#define ZFCP_ERP_DBF_INDEX 1 -#define ZFCP_ERP_DBF_AREAS 2 -#define ZFCP_ERP_DBF_LENGTH 16 -#define ZFCP_ERP_DBF_LEVEL 3 -#define ZFCP_ERP_DBF_NAME "zfcperp" - -#define ZFCP_CMD_DBF_INDEX 2 -#define ZFCP_CMD_DBF_AREAS 1 -#define ZFCP_CMD_DBF_LENGTH 8 -#define ZFCP_CMD_DBF_LEVEL 3 -#define ZFCP_CMD_DBF_NAME "zfcpcmd" - -#define ZFCP_ABORT_DBF_INDEX 2 -#define ZFCP_ABORT_DBF_AREAS 1 -#define ZFCP_ABORT_DBF_LENGTH 8 -#define ZFCP_ABORT_DBF_LEVEL 6 -#define ZFCP_ABORT_DBF_NAME "zfcpabt" - -#define ZFCP_IN_ELS_DBF_INDEX 2 -#define ZFCP_IN_ELS_DBF_AREAS 1 -#define ZFCP_IN_ELS_DBF_LENGTH 8 -#define ZFCP_IN_ELS_DBF_LEVEL 6 -#define ZFCP_IN_ELS_DBF_NAME "zfcpels" - /******************** LOGGING MACROS AND DEFINES *****************************/ /* @@ -823,11 +960,18 @@ struct zfcp_adapter { u32 erp_low_mem_count; /* nr of erp actions waiting for memory */ struct zfcp_port *nameserver_port; /* adapter's nameserver */ - debug_info_t *erp_dbf; /* S/390 debug features */ - debug_info_t *abort_dbf; - debug_info_t *in_els_dbf; - debug_info_t *cmd_dbf; - spinlock_t dbf_lock; + debug_info_t *erp_dbf; + debug_info_t *hba_dbf; + debug_info_t *san_dbf; /* debug feature areas */ + debug_info_t *scsi_dbf; + spinlock_t erp_dbf_lock; + spinlock_t hba_dbf_lock; + spinlock_t san_dbf_lock; + spinlock_t scsi_dbf_lock; + struct zfcp_erp_dbf_record erp_dbf_buf; + struct zfcp_hba_dbf_record hba_dbf_buf; + struct zfcp_san_dbf_record san_dbf_buf; + struct zfcp_scsi_dbf_record scsi_dbf_buf; struct zfcp_adapter_mempool pool; /* Adapter memory pools */ struct qdio_initialize qdio_init_data; /* for qdio_establish */ struct device generic_services; /* directory for WKA ports */ @@ -902,6 +1046,7 @@ struct zfcp_fsf_req { issued on behalf of erp */ mempool_t *pool; /* used if request was alloacted from emergency pool */ + unsigned long long issued; /* request sent time (STCK) */ struct zfcp_unit *unit; }; diff --git a/drivers/s390/scsi/zfcp_ext.h b/drivers/s390/scsi/zfcp_ext.h index c400e3b9de97..104b7423fd4d 100644 --- a/drivers/s390/scsi/zfcp_ext.h +++ b/drivers/s390/scsi/zfcp_ext.h @@ -181,9 +181,25 @@ extern void zfcp_erp_port_access_changed(struct zfcp_port *); extern void zfcp_erp_unit_access_changed(struct zfcp_unit *); /******************************** AUX ****************************************/ -extern void zfcp_cmd_dbf_event_fsf(const char *, struct zfcp_fsf_req *, - void *, int); -extern void zfcp_cmd_dbf_event_scsi(const char *, struct scsi_cmnd *); -extern void zfcp_in_els_dbf_event(struct zfcp_adapter *, const char *, - struct fsf_status_read_buffer *, int); +extern void zfcp_hba_dbf_event_fsf_response(struct zfcp_fsf_req *); +extern void zfcp_hba_dbf_event_fsf_unsol(const char *, struct zfcp_adapter *, + struct fsf_status_read_buffer *); +extern void zfcp_hba_dbf_event_qdio(struct zfcp_adapter *, + unsigned int, unsigned int, unsigned int, + int, int); + +extern void zfcp_san_dbf_event_ct_request(struct zfcp_fsf_req *); +extern void zfcp_san_dbf_event_ct_response(struct zfcp_fsf_req *); +extern void zfcp_san_dbf_event_els_request(struct zfcp_fsf_req *); +extern void zfcp_san_dbf_event_els_response(struct zfcp_fsf_req *); +extern void zfcp_san_dbf_event_incoming_els(struct zfcp_fsf_req *); + +extern void zfcp_scsi_dbf_event_result(const char *, int, struct zfcp_adapter *, + struct scsi_cmnd *); +extern void zfcp_scsi_dbf_event_abort(const char *, struct zfcp_adapter *, + struct scsi_cmnd *, + struct zfcp_fsf_req *); +extern void zfcp_scsi_dbf_event_devreset(const char *, u8, struct zfcp_unit *, + struct scsi_cmnd *); + #endif /* ZFCP_EXT_H */ diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index 929f6c9cad39..80ab721ddfff 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -285,51 +285,51 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req) { int retval = 0; struct zfcp_adapter *adapter = fsf_req->adapter; + struct fsf_qtcb *qtcb = fsf_req->qtcb; + union fsf_prot_status_qual *prot_status_qual = + &qtcb->prefix.prot_status_qual; - ZFCP_LOG_DEBUG("QTCB is at %p\n", fsf_req->qtcb); + zfcp_hba_dbf_event_fsf_response(fsf_req); if (fsf_req->status & ZFCP_STATUS_FSFREQ_DISMISSED) { ZFCP_LOG_DEBUG("fsf_req 0x%lx has been dismissed\n", (unsigned long) fsf_req); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR | ZFCP_STATUS_FSFREQ_RETRY; /* only for SCSI cmnds. */ - zfcp_cmd_dbf_event_fsf("dismiss", fsf_req, NULL, 0); goto skip_protstatus; } /* log additional information provided by FSF (if any) */ - if (unlikely(fsf_req->qtcb->header.log_length)) { + if (unlikely(qtcb->header.log_length)) { /* do not trust them ;-) */ - if (fsf_req->qtcb->header.log_start > sizeof(struct fsf_qtcb)) { + if (qtcb->header.log_start > sizeof(struct fsf_qtcb)) { ZFCP_LOG_NORMAL ("bug: ULP (FSF logging) log data starts " "beyond end of packet header. Ignored. " "(start=%i, size=%li)\n", - fsf_req->qtcb->header.log_start, + qtcb->header.log_start, sizeof(struct fsf_qtcb)); goto forget_log; } - if ((size_t) (fsf_req->qtcb->header.log_start + - fsf_req->qtcb->header.log_length) + if ((size_t) (qtcb->header.log_start + qtcb->header.log_length) > sizeof(struct fsf_qtcb)) { ZFCP_LOG_NORMAL("bug: ULP (FSF logging) log data ends " "beyond end of packet header. Ignored. " "(start=%i, length=%i, size=%li)\n", - fsf_req->qtcb->header.log_start, - fsf_req->qtcb->header.log_length, + qtcb->header.log_start, + qtcb->header.log_length, sizeof(struct fsf_qtcb)); goto forget_log; } ZFCP_LOG_TRACE("ULP log data: \n"); ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_TRACE, - (char *) fsf_req->qtcb + - fsf_req->qtcb->header.log_start, - fsf_req->qtcb->header.log_length); + (char *) qtcb + qtcb->header.log_start, + qtcb->header.log_length); } forget_log: /* evaluate FSF Protocol Status */ - switch (fsf_req->qtcb->prefix.prot_status) { + switch (qtcb->prefix.prot_status) { case FSF_PROT_GOOD: case FSF_PROT_FSF_STATUS_PRESENTED: @@ -340,14 +340,9 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req) "microcode of version 0x%x, the device driver " "only supports 0x%x. Aborting.\n", zfcp_get_busid_by_adapter(adapter), - fsf_req->qtcb->prefix.prot_status_qual. - version_error.fsf_version, ZFCP_QTCB_VERSION); - /* stop operation for this adapter */ - debug_text_exception(adapter->erp_dbf, 0, "prot_ver_err"); + prot_status_qual->version_error.fsf_version, + ZFCP_QTCB_VERSION); zfcp_erp_adapter_shutdown(adapter, 0); - zfcp_cmd_dbf_event_fsf("qverserr", fsf_req, - &fsf_req->qtcb->prefix.prot_status_qual, - sizeof (union fsf_prot_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -355,16 +350,10 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req) ZFCP_LOG_NORMAL("bug: Sequence number mismatch between " "driver (0x%x) and adapter %s (0x%x). " "Restarting all operations on this adapter.\n", - fsf_req->qtcb->prefix.req_seq_no, + qtcb->prefix.req_seq_no, zfcp_get_busid_by_adapter(adapter), - fsf_req->qtcb->prefix.prot_status_qual. - sequence_error.exp_req_seq_no); - debug_text_exception(adapter->erp_dbf, 0, "prot_seq_err"); - /* restart operation on this adapter */ + prot_status_qual->sequence_error.exp_req_seq_no); zfcp_erp_adapter_reopen(adapter, 0); - zfcp_cmd_dbf_event_fsf("seqnoerr", fsf_req, - &fsf_req->qtcb->prefix.prot_status_qual, - sizeof (union fsf_prot_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_RETRY; fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -375,22 +364,14 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req) "that used on adapter %s. " "Stopping all operations on this adapter.\n", zfcp_get_busid_by_adapter(adapter)); - debug_text_exception(adapter->erp_dbf, 0, "prot_unsup_qtcb"); zfcp_erp_adapter_shutdown(adapter, 0); - zfcp_cmd_dbf_event_fsf("unsqtcbt", fsf_req, - &fsf_req->qtcb->prefix.prot_status_qual, - sizeof (union fsf_prot_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; case FSF_PROT_HOST_CONNECTION_INITIALIZING: - zfcp_cmd_dbf_event_fsf("hconinit", fsf_req, - &fsf_req->qtcb->prefix.prot_status_qual, - sizeof (union fsf_prot_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; atomic_set_mask(ZFCP_STATUS_ADAPTER_HOST_CON_INIT, &(adapter->status)); - debug_text_event(adapter->erp_dbf, 3, "prot_con_init"); break; case FSF_PROT_DUPLICATE_REQUEST_ID: @@ -413,11 +394,7 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req) fsf_req, zfcp_get_busid_by_adapter(adapter)); } - debug_text_exception(adapter->erp_dbf, 0, "prot_dup_id"); zfcp_erp_adapter_shutdown(adapter, 0); - zfcp_cmd_dbf_event_fsf("dupreqid", fsf_req, - &fsf_req->qtcb->prefix.prot_status_qual, - sizeof (union fsf_prot_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -483,8 +460,7 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req) break; case FSF_PROT_REEST_QUEUE: - debug_text_event(adapter->erp_dbf, 1, "prot_reest_queue"); - ZFCP_LOG_INFO("The local link to adapter with " + ZFCP_LOG_NORMAL("The local link to adapter with " "%s was re-plugged. " "Re-starting operations on this adapter.\n", zfcp_get_busid_by_adapter(adapter)); @@ -495,9 +471,6 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req) zfcp_erp_adapter_reopen(adapter, ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED | ZFCP_STATUS_COMMON_ERP_FAILED); - zfcp_cmd_dbf_event_fsf("reestque", fsf_req, - &fsf_req->qtcb->prefix.prot_status_qual, - sizeof (union fsf_prot_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -507,12 +480,7 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req) "Restarting all operations on this " "adapter.\n", zfcp_get_busid_by_adapter(adapter)); - debug_text_event(adapter->erp_dbf, 0, "prot_err_sta"); - /* restart operation on this adapter */ zfcp_erp_adapter_reopen(adapter, 0); - zfcp_cmd_dbf_event_fsf("proterrs", fsf_req, - &fsf_req->qtcb->prefix.prot_status_qual, - sizeof (union fsf_prot_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_RETRY; fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -524,11 +492,7 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req) "Stopping all operations on this adapter. " "(debug info 0x%x).\n", zfcp_get_busid_by_adapter(adapter), - fsf_req->qtcb->prefix.prot_status); - debug_text_event(adapter->erp_dbf, 0, "prot_inval:"); - debug_exception(adapter->erp_dbf, 0, - &fsf_req->qtcb->prefix.prot_status, - sizeof (u32)); + qtcb->prefix.prot_status); zfcp_erp_adapter_shutdown(adapter, 0); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; } @@ -568,28 +532,18 @@ zfcp_fsf_fsfstatus_eval(struct zfcp_fsf_req *fsf_req) "(debug info 0x%x).\n", zfcp_get_busid_by_adapter(fsf_req->adapter), fsf_req->qtcb->header.fsf_command); - debug_text_exception(fsf_req->adapter->erp_dbf, 0, - "fsf_s_unknown"); zfcp_erp_adapter_shutdown(fsf_req->adapter, 0); - zfcp_cmd_dbf_event_fsf("unknownc", fsf_req, - &fsf_req->qtcb->header.fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; case FSF_FCP_RSP_AVAILABLE: ZFCP_LOG_DEBUG("FCP Sense data will be presented to the " "SCSI stack.\n"); - debug_text_event(fsf_req->adapter->erp_dbf, 3, "fsf_s_rsp"); break; case FSF_ADAPTER_STATUS_AVAILABLE: - debug_text_event(fsf_req->adapter->erp_dbf, 2, "fsf_s_astatus"); zfcp_fsf_fsfstatus_qual_eval(fsf_req); break; - - default: - break; } skip_fsfstatus: @@ -617,44 +571,28 @@ zfcp_fsf_fsfstatus_qual_eval(struct zfcp_fsf_req *fsf_req) switch (fsf_req->qtcb->header.fsf_status_qual.word[0]) { case FSF_SQ_FCP_RSP_AVAILABLE: - debug_text_event(fsf_req->adapter->erp_dbf, 4, "fsf_sq_rsp"); break; case FSF_SQ_RETRY_IF_POSSIBLE: /* The SCSI-stack may now issue retries or escalate */ - debug_text_event(fsf_req->adapter->erp_dbf, 2, "fsf_sq_retry"); - zfcp_cmd_dbf_event_fsf("sqretry", fsf_req, - &fsf_req->qtcb->header.fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; case FSF_SQ_COMMAND_ABORTED: /* Carry the aborted state on to upper layer */ - debug_text_event(fsf_req->adapter->erp_dbf, 2, "fsf_sq_abort"); - zfcp_cmd_dbf_event_fsf("sqabort", fsf_req, - &fsf_req->qtcb->header.fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ABORTED; fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; case FSF_SQ_NO_RECOM: - debug_text_exception(fsf_req->adapter->erp_dbf, 0, - "fsf_sq_no_rec"); ZFCP_LOG_NORMAL("bug: No recommendation could be given for a" "problem on the adapter %s " "Stopping all operations on this adapter. ", zfcp_get_busid_by_adapter(fsf_req->adapter)); zfcp_erp_adapter_shutdown(fsf_req->adapter, 0); - zfcp_cmd_dbf_event_fsf("sqnrecom", fsf_req, - &fsf_req->qtcb->header.fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; case FSF_SQ_ULP_PROGRAMMING_ERROR: ZFCP_LOG_NORMAL("error: not enough SBALs for data transfer " "(adapter %s)\n", zfcp_get_busid_by_adapter(fsf_req->adapter)); - debug_text_exception(fsf_req->adapter->erp_dbf, 0, - "fsf_sq_ulp_err"); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; case FSF_SQ_INVOKE_LINK_TEST_PROCEDURE: @@ -668,13 +606,6 @@ zfcp_fsf_fsfstatus_qual_eval(struct zfcp_fsf_req *fsf_req) ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_NORMAL, (char *) &fsf_req->qtcb->header.fsf_status_qual, sizeof (union fsf_status_qual)); - debug_text_event(fsf_req->adapter->erp_dbf, 0, "fsf_sq_inval:"); - debug_exception(fsf_req->adapter->erp_dbf, 0, - &fsf_req->qtcb->header.fsf_status_qual.word[0], - sizeof (u32)); - zfcp_cmd_dbf_event_fsf("squndef", fsf_req, - &fsf_req->qtcb->header.fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; } @@ -696,11 +627,6 @@ zfcp_fsf_req_dispatch(struct zfcp_fsf_req *fsf_req) struct zfcp_adapter *adapter = fsf_req->adapter; int retval = 0; - if (unlikely(fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR)) { - ZFCP_LOG_TRACE("fsf_req=%p, QTCB=%p\n", fsf_req, fsf_req->qtcb); - ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_TRACE, - (char *) fsf_req->qtcb, sizeof(struct fsf_qtcb)); - } switch (fsf_req->fsf_command) { @@ -760,13 +686,13 @@ zfcp_fsf_req_dispatch(struct zfcp_fsf_req *fsf_req) fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; ZFCP_LOG_NORMAL("bug: Command issued by the device driver is " "not supported by the adapter %s\n", - zfcp_get_busid_by_adapter(fsf_req->adapter)); + zfcp_get_busid_by_adapter(adapter)); if (fsf_req->fsf_command != fsf_req->qtcb->header.fsf_command) ZFCP_LOG_NORMAL ("bug: Command issued by the device driver differs " "from the command returned by the adapter %s " "(debug info 0x%x, 0x%x).\n", - zfcp_get_busid_by_adapter(fsf_req->adapter), + zfcp_get_busid_by_adapter(adapter), fsf_req->fsf_command, fsf_req->qtcb->header.fsf_command); } @@ -774,8 +700,6 @@ zfcp_fsf_req_dispatch(struct zfcp_fsf_req *fsf_req) if (!erp_action) return retval; - debug_text_event(adapter->erp_dbf, 3, "a_frh"); - debug_event(adapter->erp_dbf, 3, &erp_action->action, sizeof (int)); zfcp_erp_async_handler(erp_action, 0); return retval; @@ -846,6 +770,7 @@ zfcp_fsf_status_read(struct zfcp_adapter *adapter, int req_flags) failed_buf: zfcp_fsf_req_free(fsf_req); failed_req_create: + zfcp_hba_dbf_event_fsf_unsol("fail", adapter, NULL); out: write_unlock_irqrestore(&adapter->request_queue.queue_lock, lock_flags); return retval; @@ -921,35 +846,30 @@ zfcp_fsf_status_read_handler(struct zfcp_fsf_req *fsf_req) (struct fsf_status_read_buffer *) fsf_req->data; if (fsf_req->status & ZFCP_STATUS_FSFREQ_DISMISSED) { + zfcp_hba_dbf_event_fsf_unsol("dism", adapter, status_buffer); mempool_free(status_buffer, adapter->pool.data_status_read); zfcp_fsf_req_free(fsf_req); goto out; } + zfcp_hba_dbf_event_fsf_unsol("read", adapter, status_buffer); + switch (status_buffer->status_type) { case FSF_STATUS_READ_PORT_CLOSED: - debug_text_event(adapter->erp_dbf, 3, "unsol_pclosed:"); - debug_event(adapter->erp_dbf, 3, - &status_buffer->d_id, sizeof (u32)); zfcp_fsf_status_read_port_closed(fsf_req); break; case FSF_STATUS_READ_INCOMING_ELS: - debug_text_event(adapter->erp_dbf, 3, "unsol_els:"); zfcp_fsf_incoming_els(fsf_req); break; case FSF_STATUS_READ_SENSE_DATA_AVAIL: - debug_text_event(adapter->erp_dbf, 3, "unsol_sense:"); ZFCP_LOG_INFO("unsolicited sense data received (adapter %s)\n", zfcp_get_busid_by_adapter(adapter)); - ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_NORMAL, (char *) status_buffer, - sizeof(struct fsf_status_read_buffer)); break; case FSF_STATUS_READ_BIT_ERROR_THRESHOLD: - debug_text_event(adapter->erp_dbf, 3, "unsol_bit_err:"); ZFCP_LOG_NORMAL("Bit error threshold data received:\n"); ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_NORMAL, (char *) status_buffer, @@ -980,35 +900,30 @@ zfcp_fsf_status_read_handler(struct zfcp_fsf_req *fsf_req) break; case FSF_STATUS_READ_CFDC_UPDATED: - debug_text_event(adapter->erp_dbf, 2, "unsol_cfdc_update:"); - ZFCP_LOG_INFO("CFDC has been updated on the adapter %s\n", + ZFCP_LOG_NORMAL("CFDC has been updated on the adapter %s\n", zfcp_get_busid_by_adapter(adapter)); zfcp_erp_adapter_access_changed(adapter); break; case FSF_STATUS_READ_CFDC_HARDENED: - debug_text_event(adapter->erp_dbf, 2, "unsol_cfdc_harden:"); switch (status_buffer->status_subtype) { case FSF_STATUS_READ_SUB_CFDC_HARDENED_ON_SE: - ZFCP_LOG_INFO("CFDC of adapter %s saved on SE\n", + ZFCP_LOG_NORMAL("CFDC of adapter %s saved on SE\n", zfcp_get_busid_by_adapter(adapter)); break; case FSF_STATUS_READ_SUB_CFDC_HARDENED_ON_SE2: - ZFCP_LOG_INFO("CFDC of adapter %s has been copied " + ZFCP_LOG_NORMAL("CFDC of adapter %s has been copied " "to the secondary SE\n", zfcp_get_busid_by_adapter(adapter)); break; default: - ZFCP_LOG_INFO("CFDC of adapter %s has been hardened\n", + ZFCP_LOG_NORMAL("CFDC of adapter %s has been hardened\n", zfcp_get_busid_by_adapter(adapter)); } break; default: - debug_text_event(adapter->erp_dbf, 0, "unsol_unknown:"); - debug_exception(adapter->erp_dbf, 0, - &status_buffer->status_type, sizeof (u32)); - ZFCP_LOG_NORMAL("bug: An unsolicited status packet of unknown " + ZFCP_LOG_NORMAL("warning: An unsolicited status packet of unknown " "type was received (debug info 0x%x)\n", status_buffer->status_type); ZFCP_LOG_DEBUG("Dump of status_read_buffer %p:\n", @@ -1418,6 +1333,8 @@ zfcp_fsf_send_ct(struct zfcp_send_ct *ct, mempool_t *pool, fsf_req->qtcb->bottom.support.timeout = ct->timeout; fsf_req->data = (unsigned long) ct; + zfcp_san_dbf_event_ct_request(fsf_req); + /* start QDIO request for this FSF request */ ret = zfcp_fsf_req_send(fsf_req, ct->timer); if (ret) { @@ -1476,6 +1393,7 @@ zfcp_fsf_send_ct_handler(struct zfcp_fsf_req *fsf_req) switch (header->fsf_status) { case FSF_GOOD: + zfcp_san_dbf_event_ct_response(fsf_req); retval = 0; break; @@ -1720,6 +1638,8 @@ zfcp_fsf_send_els(struct zfcp_send_els *els) sbale = zfcp_qdio_sbale_req(fsf_req, fsf_req->sbal_curr, 0); + zfcp_san_dbf_event_els_request(fsf_req); + /* start QDIO request for this FSF request */ ret = zfcp_fsf_req_send(fsf_req, els->timer); if (ret) { @@ -1777,6 +1697,7 @@ static int zfcp_fsf_send_els_handler(struct zfcp_fsf_req *fsf_req) switch (header->fsf_status) { case FSF_GOOD: + zfcp_san_dbf_event_els_response(fsf_req); retval = 0; break; @@ -3309,9 +3230,6 @@ zfcp_fsf_close_unit_handler(struct zfcp_fsf_req *fsf_req) debug_text_event(fsf_req->adapter->erp_dbf, 1, "fsf_s_phand_nv"); zfcp_erp_adapter_reopen(unit->port->adapter, 0); - zfcp_cmd_dbf_event_fsf("porthinv", fsf_req, - &fsf_req->qtcb->header.fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -3330,9 +3248,6 @@ zfcp_fsf_close_unit_handler(struct zfcp_fsf_req *fsf_req) debug_text_event(fsf_req->adapter->erp_dbf, 1, "fsf_s_lhand_nv"); zfcp_erp_port_reopen(unit->port, 0); - zfcp_cmd_dbf_event_fsf("lunhinv", fsf_req, - &fsf_req->qtcb->header.fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -3763,10 +3678,6 @@ zfcp_fsf_send_fcp_command_handler(struct zfcp_fsf_req *fsf_req) debug_text_event(fsf_req->adapter->erp_dbf, 1, "fsf_s_hand_mis"); zfcp_erp_adapter_reopen(unit->port->adapter, 0); - zfcp_cmd_dbf_event_fsf("handmism", - fsf_req, - &header->fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -3787,10 +3698,6 @@ zfcp_fsf_send_fcp_command_handler(struct zfcp_fsf_req *fsf_req) debug_text_exception(fsf_req->adapter->erp_dbf, 0, "fsf_s_class_nsup"); zfcp_erp_adapter_shutdown(unit->port->adapter, 0); - zfcp_cmd_dbf_event_fsf("unsclass", - fsf_req, - &header->fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -3809,10 +3716,6 @@ zfcp_fsf_send_fcp_command_handler(struct zfcp_fsf_req *fsf_req) debug_text_event(fsf_req->adapter->erp_dbf, 1, "fsf_s_fcp_lun_nv"); zfcp_erp_port_reopen(unit->port, 0); - zfcp_cmd_dbf_event_fsf("fluninv", - fsf_req, - &header->fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -3851,10 +3754,6 @@ zfcp_fsf_send_fcp_command_handler(struct zfcp_fsf_req *fsf_req) debug_text_event(fsf_req->adapter->erp_dbf, 0, "fsf_s_dir_ind_nv"); zfcp_erp_adapter_shutdown(unit->port->adapter, 0); - zfcp_cmd_dbf_event_fsf("dirinv", - fsf_req, - &header->fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -3870,10 +3769,6 @@ zfcp_fsf_send_fcp_command_handler(struct zfcp_fsf_req *fsf_req) debug_text_event(fsf_req->adapter->erp_dbf, 0, "fsf_s_cmd_len_nv"); zfcp_erp_adapter_shutdown(unit->port->adapter, 0); - zfcp_cmd_dbf_event_fsf("cleninv", - fsf_req, - &header->fsf_status_qual, - sizeof (union fsf_status_qual)); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -4043,7 +3938,6 @@ zfcp_fsf_send_fcp_command_task_handler(struct zfcp_fsf_req *fsf_req) ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_DEBUG, (char *) &fsf_req->qtcb-> bottom.io.fcp_cmnd, FSF_FCP_CMND_SIZE); - zfcp_cmd_dbf_event_fsf("clenmis", fsf_req, NULL, 0); set_host_byte(&scpnt->result, DID_ERROR); goto skip_fsfstatus; case RSP_CODE_FIELD_INVALID: @@ -4062,7 +3956,6 @@ zfcp_fsf_send_fcp_command_task_handler(struct zfcp_fsf_req *fsf_req) (char *) &fsf_req->qtcb-> bottom.io.fcp_cmnd, FSF_FCP_CMND_SIZE); set_host_byte(&scpnt->result, DID_ERROR); - zfcp_cmd_dbf_event_fsf("codeinv", fsf_req, NULL, 0); goto skip_fsfstatus; case RSP_CODE_RO_MISMATCH: /* hardware bug */ @@ -4079,7 +3972,6 @@ zfcp_fsf_send_fcp_command_task_handler(struct zfcp_fsf_req *fsf_req) ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_DEBUG, (char *) &fsf_req->qtcb-> bottom.io.fcp_cmnd, FSF_FCP_CMND_SIZE); - zfcp_cmd_dbf_event_fsf("codemism", fsf_req, NULL, 0); set_host_byte(&scpnt->result, DID_ERROR); goto skip_fsfstatus; default: @@ -4096,7 +3988,6 @@ zfcp_fsf_send_fcp_command_task_handler(struct zfcp_fsf_req *fsf_req) ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_DEBUG, (char *) &fsf_req->qtcb-> bottom.io.fcp_cmnd, FSF_FCP_CMND_SIZE); - zfcp_cmd_dbf_event_fsf("undeffcp", fsf_req, NULL, 0); set_host_byte(&scpnt->result, DID_ERROR); goto skip_fsfstatus; } @@ -4158,19 +4049,17 @@ zfcp_fsf_send_fcp_command_task_handler(struct zfcp_fsf_req *fsf_req) skip_fsfstatus: ZFCP_LOG_DEBUG("scpnt->result =0x%x\n", scpnt->result); - zfcp_cmd_dbf_event_scsi("response", scpnt); + if (scpnt->result != 0) + zfcp_scsi_dbf_event_result("erro", 3, fsf_req->adapter, scpnt); + else if (scpnt->retries > 0) + zfcp_scsi_dbf_event_result("retr", 4, fsf_req->adapter, scpnt); + else + zfcp_scsi_dbf_event_result("norm", 6, fsf_req->adapter, scpnt); /* cleanup pointer (need this especially for abort) */ scpnt->host_scribble = NULL; - /* - * NOTE: - * according to the outcome of a discussion on linux-scsi we - * don't need to grab the io_request_lock here since we use - * the new eh - */ /* always call back */ - (scpnt->scsi_done) (scpnt); /* @@ -4563,15 +4452,16 @@ zfcp_fsf_req_sbal_check(unsigned long *flags, * set qtcb pointer in fsf_req and initialize QTCB */ static inline void -zfcp_fsf_req_qtcb_init(struct zfcp_fsf_req *fsf_req, u32 fsf_cmd) +zfcp_fsf_req_qtcb_init(struct zfcp_fsf_req *fsf_req) { if (likely(fsf_req->qtcb != NULL)) { + fsf_req->qtcb->prefix.req_seq_no = fsf_req->adapter->fsf_req_seq_no; fsf_req->qtcb->prefix.req_id = (unsigned long)fsf_req; fsf_req->qtcb->prefix.ulp_info = ZFCP_ULP_INFO_VERSION; - fsf_req->qtcb->prefix.qtcb_type = fsf_qtcb_type[fsf_cmd]; + fsf_req->qtcb->prefix.qtcb_type = fsf_qtcb_type[fsf_req->fsf_command]; fsf_req->qtcb->prefix.qtcb_version = ZFCP_QTCB_VERSION; fsf_req->qtcb->header.req_handle = (unsigned long)fsf_req; - fsf_req->qtcb->header.fsf_command = fsf_cmd; + fsf_req->qtcb->header.fsf_command = fsf_req->fsf_command; } } @@ -4639,7 +4529,10 @@ zfcp_fsf_req_create(struct zfcp_adapter *adapter, u32 fsf_cmd, int req_flags, goto failed_fsf_req; } - zfcp_fsf_req_qtcb_init(fsf_req, fsf_cmd); + fsf_req->adapter = adapter; + fsf_req->fsf_command = fsf_cmd; + + zfcp_fsf_req_qtcb_init(fsf_req); /* initialize waitqueue which may be used to wait on this request completion */ @@ -4661,8 +4554,10 @@ zfcp_fsf_req_create(struct zfcp_adapter *adapter, u32 fsf_cmd, int req_flags, goto failed_sbals; } - fsf_req->adapter = adapter; /* pointer to "parent" adapter */ - fsf_req->fsf_command = fsf_cmd; + if (fsf_req->qtcb) { + fsf_req->seq_no = adapter->fsf_req_seq_no; + fsf_req->qtcb->prefix.req_seq_no = adapter->fsf_req_seq_no; + } fsf_req->sbal_number = 1; fsf_req->sbal_first = req_queue->free_index; fsf_req->sbal_curr = req_queue->free_index; @@ -4713,9 +4608,9 @@ zfcp_fsf_req_send(struct zfcp_fsf_req *fsf_req, struct timer_list *timer) struct zfcp_adapter *adapter; struct zfcp_qdio_queue *req_queue; volatile struct qdio_buffer_element *sbale; + int inc_seq_no; int new_distance_from_int; unsigned long flags; - int inc_seq_no = 1; int retval = 0; adapter = fsf_req->adapter; @@ -4729,23 +4624,13 @@ zfcp_fsf_req_send(struct zfcp_fsf_req *fsf_req, struct timer_list *timer) ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_TRACE, (char *) sbale[1].addr, sbale[1].length); - /* set sequence counter in QTCB */ - if (likely(fsf_req->qtcb)) { - fsf_req->qtcb->prefix.req_seq_no = adapter->fsf_req_seq_no; - fsf_req->seq_no = adapter->fsf_req_seq_no; - ZFCP_LOG_TRACE("FSF request %p of adapter %s gets " - "FSF sequence counter value of %i\n", - fsf_req, - zfcp_get_busid_by_adapter(adapter), - fsf_req->qtcb->prefix.req_seq_no); - } else - inc_seq_no = 0; - /* put allocated FSF request at list tail */ spin_lock_irqsave(&adapter->fsf_req_list_lock, flags); list_add_tail(&fsf_req->list, &adapter->fsf_req_list_head); spin_unlock_irqrestore(&adapter->fsf_req_list_lock, flags); + inc_seq_no = (fsf_req->qtcb != NULL); + /* figure out expiration time of timeout and start timeout */ if (unlikely(timer)) { timer->expires += jiffies; @@ -4775,6 +4660,8 @@ zfcp_fsf_req_send(struct zfcp_fsf_req *fsf_req, struct timer_list *timer) req_queue->free_index %= QDIO_MAX_BUFFERS_PER_Q; /* wrap if needed */ new_distance_from_int = zfcp_qdio_determine_pci(req_queue, fsf_req); + fsf_req->issued = get_clock(); + retval = do_QDIO(adapter->ccw_device, QDIO_FLAG_SYNC_OUTPUT, 0, fsf_req->sbal_first, fsf_req->sbal_number, NULL); @@ -4813,15 +4700,11 @@ zfcp_fsf_req_send(struct zfcp_fsf_req *fsf_req, struct timer_list *timer) * routines resulting in missing sequence counter values * otherwise, */ + /* Don't increase for unsolicited status */ - if (likely(inc_seq_no)) { + if (inc_seq_no) adapter->fsf_req_seq_no++; - ZFCP_LOG_TRACE - ("FSF sequence counter value of adapter %s " - "increased to %i\n", - zfcp_get_busid_by_adapter(adapter), - adapter->fsf_req_seq_no); - } + /* count FSF requests pending */ atomic_inc(&adapter->fsf_reqs_active); } diff --git a/drivers/s390/scsi/zfcp_fsf.h b/drivers/s390/scsi/zfcp_fsf.h index 07140dfda2a7..57ce0706007e 100644 --- a/drivers/s390/scsi/zfcp_fsf.h +++ b/drivers/s390/scsi/zfcp_fsf.h @@ -116,6 +116,7 @@ #define FSF_INVALID_COMMAND_OPTION 0x000000E5 /* #define FSF_ERROR 0x000000FF */ +#define FSF_PROT_STATUS_QUAL_SIZE 16 #define FSF_STATUS_QUALIFIER_SIZE 16 /* FSF status qualifier, recommendations */ @@ -311,6 +312,7 @@ struct fsf_qual_locallink_error { } __attribute__ ((packed)); union fsf_prot_status_qual { + u64 doubleword[FSF_PROT_STATUS_QUAL_SIZE / sizeof(u64)]; struct fsf_qual_version_error version_error; struct fsf_qual_sequence_error sequence_error; struct fsf_qual_locallink_error locallink_error; @@ -331,6 +333,7 @@ union fsf_status_qual { u8 byte[FSF_STATUS_QUALIFIER_SIZE]; u16 halfword[FSF_STATUS_QUALIFIER_SIZE / sizeof (u16)]; u32 word[FSF_STATUS_QUALIFIER_SIZE / sizeof (u32)]; + u64 doubleword[FSF_STATUS_QUALIFIER_SIZE / sizeof(u64)]; struct fsf_queue_designator fsf_queue_designator; } __attribute__ ((packed)); diff --git a/drivers/s390/scsi/zfcp_qdio.c b/drivers/s390/scsi/zfcp_qdio.c index 24e16ec331d9..d719f66a29a4 100644 --- a/drivers/s390/scsi/zfcp_qdio.c +++ b/drivers/s390/scsi/zfcp_qdio.c @@ -54,8 +54,7 @@ static inline int zfcp_qdio_sbals_from_buffer static qdio_handler_t zfcp_qdio_request_handler; static qdio_handler_t zfcp_qdio_response_handler; static int zfcp_qdio_handler_error_check(struct zfcp_adapter *, - unsigned int, - unsigned int, unsigned int); + unsigned int, unsigned int, unsigned int, int, int); #define ZFCP_LOG_AREA ZFCP_LOG_AREA_QDIO @@ -214,22 +213,12 @@ zfcp_qdio_allocate(struct zfcp_adapter *adapter) * */ static inline int -zfcp_qdio_handler_error_check(struct zfcp_adapter *adapter, - unsigned int status, - unsigned int qdio_error, unsigned int siga_error) +zfcp_qdio_handler_error_check(struct zfcp_adapter *adapter, unsigned int status, + unsigned int qdio_error, unsigned int siga_error, + int first_element, int elements_processed) { int retval = 0; - if (ZFCP_LOG_CHECK(ZFCP_LOG_LEVEL_TRACE)) { - if (status & QDIO_STATUS_INBOUND_INT) { - ZFCP_LOG_TRACE("status is" - " QDIO_STATUS_INBOUND_INT \n"); - } - if (status & QDIO_STATUS_OUTBOUND_INT) { - ZFCP_LOG_TRACE("status is" - " QDIO_STATUS_OUTBOUND_INT \n"); - } - } if (unlikely(status & QDIO_STATUS_LOOK_FOR_ERROR)) { retval = -EIO; @@ -237,9 +226,10 @@ zfcp_qdio_handler_error_check(struct zfcp_adapter *adapter, "qdio_error=0x%x, siga_error=0x%x)\n", status, qdio_error, siga_error); - /* Restarting IO on the failed adapter from scratch */ - debug_text_event(adapter->erp_dbf, 1, "qdio_err"); + zfcp_hba_dbf_event_qdio(adapter, status, qdio_error, siga_error, + first_element, elements_processed); /* + * Restarting IO on the failed adapter from scratch. * Since we have been using this adapter, it is save to assume * that it is not failed but recoverable. The card seems to * report link-up events by self-initiated queue shutdown. @@ -282,7 +272,8 @@ zfcp_qdio_request_handler(struct ccw_device *ccw_device, first_element, elements_processed); if (unlikely(zfcp_qdio_handler_error_check(adapter, status, qdio_error, - siga_error))) + siga_error, first_element, + elements_processed))) goto out; /* * we stored address of struct zfcp_adapter data structure @@ -334,7 +325,8 @@ zfcp_qdio_response_handler(struct ccw_device *ccw_device, queue = &adapter->response_queue; if (unlikely(zfcp_qdio_handler_error_check(adapter, status, qdio_error, - siga_error))) + siga_error, first_element, + elements_processed))) goto out; /* diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index eeb5152854ed..c1403a23174f 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -44,7 +44,8 @@ static int zfcp_scsi_eh_abort_handler(struct scsi_cmnd *); static int zfcp_scsi_eh_device_reset_handler(struct scsi_cmnd *); static int zfcp_scsi_eh_bus_reset_handler(struct scsi_cmnd *); static int zfcp_scsi_eh_host_reset_handler(struct scsi_cmnd *); -static int zfcp_task_management_function(struct zfcp_unit *, u8); +static int zfcp_task_management_function(struct zfcp_unit *, u8, + struct scsi_cmnd *); static struct zfcp_unit *zfcp_unit_lookup(struct zfcp_adapter *, int, scsi_id_t, scsi_lun_t); @@ -242,7 +243,10 @@ static void zfcp_scsi_command_fail(struct scsi_cmnd *scpnt, int result) { set_host_byte(&scpnt->result, result); - zfcp_cmd_dbf_event_scsi("failing", scpnt); + if ((scpnt->device != NULL) && (scpnt->device->host != NULL)) + zfcp_scsi_dbf_event_result("fail", 4, + (struct zfcp_adapter*) scpnt->device->host->hostdata[0], + scpnt); /* return directly */ scpnt->scsi_done(scpnt); } @@ -434,7 +438,8 @@ zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt) struct zfcp_adapter *adapter; struct zfcp_unit *unit; int retval = SUCCESS; - struct zfcp_fsf_req *new_fsf_req, *old_fsf_req; + struct zfcp_fsf_req *new_fsf_req = NULL; + struct zfcp_fsf_req *old_fsf_req; unsigned long flags; scsi_host = scpnt->device->host; @@ -457,11 +462,8 @@ zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt) old_fsf_req = (struct zfcp_fsf_req *) scpnt->host_scribble; if (!old_fsf_req) { write_unlock_irqrestore(&adapter->abort_lock, flags); - ZFCP_LOG_NORMAL("bug: no old fsf request found\n"); - ZFCP_LOG_NORMAL("scsi_cmnd:\n"); - ZFCP_HEX_DUMP(ZFCP_LOG_LEVEL_NORMAL, - (char *) scpnt, sizeof (struct scsi_cmnd)); - retval = FAILED; + zfcp_scsi_dbf_event_abort("lte1", adapter, scpnt, new_fsf_req); + retval = SUCCESS; goto out; } old_fsf_req->data = 0; @@ -473,25 +475,27 @@ zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt) new_fsf_req = zfcp_fsf_abort_fcp_command((unsigned long) old_fsf_req, adapter, unit, 0); if (!new_fsf_req) { + ZFCP_LOG_INFO("error: initiation of Abort FCP Cmnd failed\n"); retval = FAILED; - ZFCP_LOG_NORMAL("error: initiation of Abort FCP Cmnd " - "failed\n"); goto out; } /* wait for completion of abort */ __wait_event(new_fsf_req->completion_wq, new_fsf_req->status & ZFCP_STATUS_FSFREQ_COMPLETED); - zfcp_fsf_req_free(new_fsf_req); /* status should be valid since signals were not permitted */ if (new_fsf_req->status & ZFCP_STATUS_FSFREQ_ABORTSUCCEEDED) { + zfcp_scsi_dbf_event_abort("okay", adapter, scpnt, new_fsf_req); retval = SUCCESS; } else if (new_fsf_req->status & ZFCP_STATUS_FSFREQ_ABORTNOTNEEDED) { + zfcp_scsi_dbf_event_abort("lte2", adapter, scpnt, new_fsf_req); retval = SUCCESS; } else { + zfcp_scsi_dbf_event_abort("fail", adapter, scpnt, new_fsf_req); retval = FAILED; } + zfcp_fsf_req_free(new_fsf_req); out: return retval; } @@ -525,8 +529,9 @@ zfcp_scsi_eh_device_reset_handler(struct scsi_cmnd *scpnt) */ if (!atomic_test_mask(ZFCP_STATUS_UNIT_NOTSUPPUNITRESET, &unit->status)) { - retval = - zfcp_task_management_function(unit, FCP_LOGICAL_UNIT_RESET); + retval = zfcp_task_management_function(unit, + FCP_LOGICAL_UNIT_RESET, + scpnt); if (retval) { ZFCP_LOG_DEBUG("unit reset failed (unit=%p)\n", unit); if (retval == -ENOTSUPP) @@ -542,7 +547,7 @@ zfcp_scsi_eh_device_reset_handler(struct scsi_cmnd *scpnt) goto out; } } - retval = zfcp_task_management_function(unit, FCP_TARGET_RESET); + retval = zfcp_task_management_function(unit, FCP_TARGET_RESET, scpnt); if (retval) { ZFCP_LOG_DEBUG("target reset failed (unit=%p)\n", unit); retval = FAILED; @@ -555,7 +560,8 @@ zfcp_scsi_eh_device_reset_handler(struct scsi_cmnd *scpnt) } static int -zfcp_task_management_function(struct zfcp_unit *unit, u8 tm_flags) +zfcp_task_management_function(struct zfcp_unit *unit, u8 tm_flags, + struct scsi_cmnd *scpnt) { struct zfcp_adapter *adapter = unit->port->adapter; struct zfcp_fsf_req *fsf_req; @@ -569,6 +575,7 @@ zfcp_task_management_function(struct zfcp_unit *unit, u8 tm_flags) "failed for unit 0x%016Lx on port 0x%016Lx on " "adapter %s\n", unit->fcp_lun, unit->port->wwpn, zfcp_get_busid_by_adapter(adapter)); + zfcp_scsi_dbf_event_devreset("nres", tm_flags, unit, scpnt); retval = -ENOMEM; goto out; } @@ -576,11 +583,17 @@ zfcp_task_management_function(struct zfcp_unit *unit, u8 tm_flags) __wait_event(fsf_req->completion_wq, fsf_req->status & ZFCP_STATUS_FSFREQ_COMPLETED); - /* check completion status of task management function */ - if (fsf_req->status & ZFCP_STATUS_FSFREQ_TMFUNCFAILED) + /* + * check completion status of task management function + */ + if (fsf_req->status & ZFCP_STATUS_FSFREQ_TMFUNCFAILED) { + zfcp_scsi_dbf_event_devreset("fail", tm_flags, unit, scpnt); retval = -EIO; - else if (fsf_req->status & ZFCP_STATUS_FSFREQ_TMFUNCNOTSUPP) + } else if (fsf_req->status & ZFCP_STATUS_FSFREQ_TMFUNCNOTSUPP) { + zfcp_scsi_dbf_event_devreset("nsup", tm_flags, unit, scpnt); retval = -ENOTSUPP; + } else + zfcp_scsi_dbf_event_devreset("okay", tm_flags, unit, scpnt); zfcp_fsf_req_free(fsf_req); out: From aef4a983090fa590481a86d9690dc3fa6bb121fa Mon Sep 17 00:00:00 2001 From: Maxim Shchetynin Date: Tue, 13 Sep 2005 21:51:16 +0200 Subject: [PATCH 173/240] [SCSI] zfcp: provide support for NPIV N_Port ID Virtualization (NPIV) allows a single FCP port to appear as multiple, distinct ports providing separate port identification. NPIV is supported by FC HBAs on System z9. zfcp was adapted to support this new feature. Signed-off-by: Andreas Herrmann Signed-off-by: James Bottomley --- drivers/s390/scsi/zfcp_dbf.c | 11 +- drivers/s390/scsi/zfcp_def.h | 13 +- drivers/s390/scsi/zfcp_erp.c | 95 ++++++- drivers/s390/scsi/zfcp_ext.h | 3 +- drivers/s390/scsi/zfcp_fsf.c | 330 ++++++++++++++++--------- drivers/s390/scsi/zfcp_fsf.h | 51 +++- drivers/s390/scsi/zfcp_sysfs_adapter.c | 4 + 7 files changed, 367 insertions(+), 140 deletions(-) diff --git a/drivers/s390/scsi/zfcp_dbf.c b/drivers/s390/scsi/zfcp_dbf.c index fff1537335c7..64d9b90373fe 100644 --- a/drivers/s390/scsi/zfcp_dbf.c +++ b/drivers/s390/scsi/zfcp_dbf.c @@ -279,9 +279,18 @@ zfcp_hba_dbf_event_fsf_unsol(const char *tag, struct zfcp_adapter *adapter, break; case FSF_STATUS_READ_LINK_DOWN: - rec->type.status.payload_size = sizeof(u64); + switch (status_buffer->status_subtype) { + case FSF_STATUS_READ_SUB_NO_PHYSICAL_LINK: + case FSF_STATUS_READ_SUB_FDISC_FAILED: + rec->type.status.payload_size = + sizeof(struct fsf_link_down_info); + } break; + case FSF_STATUS_READ_FEATURE_UPDATE_ALERT: + rec->type.status.payload_size = + ZFCP_DBF_UNSOL_PAYLOAD_FEATURE_UPDATE_ALERT; + break; } memcpy(&rec->type.status.payload, &status_buffer->payload, rec->type.status.payload_size); diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h index fc5bb6f31808..4b6af8e07e8d 100644 --- a/drivers/s390/scsi/zfcp_def.h +++ b/drivers/s390/scsi/zfcp_def.h @@ -66,7 +66,7 @@ /********************* GENERAL DEFINES *********************************/ /* zfcp version number, it consists of major, minor, and patch-level number */ -#define ZFCP_VERSION "4.4.0" +#define ZFCP_VERSION "4.5.0" /** * zfcp_sg_to_address - determine kernel address from struct scatterlist @@ -154,6 +154,11 @@ typedef u32 scsi_lun_t; #define ZFCP_EXCHANGE_CONFIG_DATA_FIRST_SLEEP 100 #define ZFCP_EXCHANGE_CONFIG_DATA_RETRIES 7 +/* Retry 5 times every 2 second, then every minute */ +#define ZFCP_EXCHANGE_PORT_DATA_SHORT_RETRIES 5 +#define ZFCP_EXCHANGE_PORT_DATA_SHORT_SLEEP 200 +#define ZFCP_EXCHANGE_PORT_DATA_LONG_SLEEP 6000 + /* timeout value for "default timer" for fsf requests */ #define ZFCP_FSF_REQUEST_TIMEOUT (60*HZ); @@ -638,6 +643,7 @@ do { \ #define ZFCP_STATUS_ADAPTER_ERP_THREAD_KILL 0x00000080 #define ZFCP_STATUS_ADAPTER_ERP_PENDING 0x00000100 #define ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED 0x00000200 +#define ZFCP_STATUS_ADAPTER_XPORT_OK 0x00000800 #define ZFCP_STATUS_ADAPTER_SCSI_UP \ (ZFCP_STATUS_COMMON_UNBLOCKED | \ @@ -915,13 +921,16 @@ struct zfcp_adapter { wwn_t peer_wwnn; /* P2P peer WWNN */ wwn_t peer_wwpn; /* P2P peer WWPN */ fc_id_t peer_d_id; /* P2P peer D_ID */ + wwn_t physical_wwpn; /* WWPN of physical port */ + fc_id_t physical_s_id; /* local FC port ID */ struct ccw_device *ccw_device; /* S/390 ccw device */ u8 fc_service_class; u32 fc_topology; /* FC topology */ u32 fc_link_speed; /* FC interface speed */ u32 hydra_version; /* Hydra version */ u32 fsf_lic_version; - u32 supported_features;/* of FCP channel */ + u32 adapter_features; /* FCP channel features */ + u32 connection_features; /* host connection features */ u32 hardware_version; /* of FCP channel */ u8 serial_number[32]; /* of hardware */ struct Scsi_Host *scsi_host; /* Pointer to mid-layer */ diff --git a/drivers/s390/scsi/zfcp_erp.c b/drivers/s390/scsi/zfcp_erp.c index 376cb0f6cb74..c4a6799aed44 100644 --- a/drivers/s390/scsi/zfcp_erp.c +++ b/drivers/s390/scsi/zfcp_erp.c @@ -82,6 +82,7 @@ static int zfcp_erp_adapter_strategy_open(struct zfcp_erp_action *); static int zfcp_erp_adapter_strategy_open_qdio(struct zfcp_erp_action *); static int zfcp_erp_adapter_strategy_open_fsf(struct zfcp_erp_action *); static int zfcp_erp_adapter_strategy_open_fsf_xconfig(struct zfcp_erp_action *); +static int zfcp_erp_adapter_strategy_open_fsf_xport(struct zfcp_erp_action *); static int zfcp_erp_adapter_strategy_open_fsf_statusread( struct zfcp_erp_action *); @@ -2258,16 +2259,21 @@ zfcp_erp_adapter_strategy_close_qdio(struct zfcp_erp_action *erp_action) static int zfcp_erp_adapter_strategy_open_fsf(struct zfcp_erp_action *erp_action) { - int retval; + int xconfig, xport; - /* do 'exchange configuration data' */ - retval = zfcp_erp_adapter_strategy_open_fsf_xconfig(erp_action); - if (retval == ZFCP_ERP_FAILED) - return retval; + if (atomic_test_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, + &erp_action->adapter->status)) { + zfcp_erp_adapter_strategy_open_fsf_xport(erp_action); + atomic_set(&erp_action->adapter->erp_counter, 0); + return ZFCP_ERP_FAILED; + } - /* start the desired number of Status Reads */ - retval = zfcp_erp_adapter_strategy_open_fsf_statusread(erp_action); - return retval; + xconfig = zfcp_erp_adapter_strategy_open_fsf_xconfig(erp_action); + xport = zfcp_erp_adapter_strategy_open_fsf_xport(erp_action); + if ((xconfig == ZFCP_ERP_FAILED) || (xport == ZFCP_ERP_FAILED)) + return ZFCP_ERP_FAILED; + + return zfcp_erp_adapter_strategy_open_fsf_statusread(erp_action); } /* @@ -2350,6 +2356,76 @@ zfcp_erp_adapter_strategy_open_fsf_xconfig(struct zfcp_erp_action *erp_action) return retval; } +static int +zfcp_erp_adapter_strategy_open_fsf_xport(struct zfcp_erp_action *erp_action) +{ + int retval = ZFCP_ERP_SUCCEEDED; + int retries; + int sleep; + struct zfcp_adapter *adapter = erp_action->adapter; + + atomic_clear_mask(ZFCP_STATUS_ADAPTER_XPORT_OK, &adapter->status); + + for (retries = 0; ; retries++) { + ZFCP_LOG_DEBUG("Doing exchange port data\n"); + zfcp_erp_action_to_running(erp_action); + zfcp_erp_timeout_init(erp_action); + if (zfcp_fsf_exchange_port_data(erp_action, adapter, NULL)) { + retval = ZFCP_ERP_FAILED; + debug_text_event(adapter->erp_dbf, 5, "a_fstx_xf"); + ZFCP_LOG_INFO("error: initiation of exchange of " + "port data failed for adapter %s\n", + zfcp_get_busid_by_adapter(adapter)); + break; + } + debug_text_event(adapter->erp_dbf, 6, "a_fstx_xok"); + ZFCP_LOG_DEBUG("Xchange underway\n"); + + /* + * Why this works: + * Both the normal completion handler as well as the timeout + * handler will do an 'up' when the 'exchange port data' + * request completes or times out. Thus, the signal to go on + * won't be lost utilizing this semaphore. + * Furthermore, this 'adapter_reopen' action is + * guaranteed to be the only action being there (highest action + * which prevents other actions from being created). + * Resulting from that, the wake signal recognized here + * _must_ be the one belonging to the 'exchange port + * data' request. + */ + down(&adapter->erp_ready_sem); + if (erp_action->status & ZFCP_STATUS_ERP_TIMEDOUT) { + ZFCP_LOG_INFO("error: exchange of port data " + "for adapter %s timed out\n", + zfcp_get_busid_by_adapter(adapter)); + break; + } + + if (!atomic_test_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, + &adapter->status)) + break; + + ZFCP_LOG_DEBUG("host connection still initialising... " + "waiting and retrying...\n"); + /* sleep a little bit before retry */ + sleep = retries < ZFCP_EXCHANGE_PORT_DATA_SHORT_RETRIES ? + ZFCP_EXCHANGE_PORT_DATA_SHORT_SLEEP : + ZFCP_EXCHANGE_PORT_DATA_LONG_SLEEP; + msleep(jiffies_to_msecs(sleep)); + } + + if (atomic_test_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, + &adapter->status)) { + ZFCP_LOG_INFO("error: exchange of port data for " + "adapter %s failed\n", + zfcp_get_busid_by_adapter(adapter)); + retval = ZFCP_ERP_FAILED; + } + + return retval; +} + /* * function: * @@ -3599,6 +3675,9 @@ zfcp_erp_adapter_access_changed(struct zfcp_adapter *adapter) struct zfcp_port *port; unsigned long flags; + if (adapter->connection_features & FSF_FEATURE_NPIV_MODE) + return; + debug_text_event(adapter->erp_dbf, 3, "a_access_recover"); debug_event(adapter->erp_dbf, 3, &adapter->name, 8); diff --git a/drivers/s390/scsi/zfcp_ext.h b/drivers/s390/scsi/zfcp_ext.h index 104b7423fd4d..c3782261cb5c 100644 --- a/drivers/s390/scsi/zfcp_ext.h +++ b/drivers/s390/scsi/zfcp_ext.h @@ -96,7 +96,8 @@ extern int zfcp_fsf_open_unit(struct zfcp_erp_action *); extern int zfcp_fsf_close_unit(struct zfcp_erp_action *); extern int zfcp_fsf_exchange_config_data(struct zfcp_erp_action *); -extern int zfcp_fsf_exchange_port_data(struct zfcp_adapter *, +extern int zfcp_fsf_exchange_port_data(struct zfcp_erp_action *, + struct zfcp_adapter *, struct fsf_qtcb_bottom_port *); extern int zfcp_fsf_control_file(struct zfcp_adapter *, struct zfcp_fsf_req **, u32, u32, struct zfcp_sg_list *); diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index 80ab721ddfff..f994f453e2d2 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -59,6 +59,8 @@ static int zfcp_fsf_req_send(struct zfcp_fsf_req *, struct timer_list *); static int zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *); static int zfcp_fsf_fsfstatus_eval(struct zfcp_fsf_req *); static int zfcp_fsf_fsfstatus_qual_eval(struct zfcp_fsf_req *); +static void zfcp_fsf_link_down_info_eval(struct zfcp_adapter *, + struct fsf_link_down_info *); static int zfcp_fsf_req_dispatch(struct zfcp_fsf_req *); static void zfcp_fsf_req_dismiss(struct zfcp_fsf_req *); @@ -375,87 +377,19 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req) break; case FSF_PROT_DUPLICATE_REQUEST_ID: - if (fsf_req->qtcb) { ZFCP_LOG_NORMAL("bug: The request identifier 0x%Lx " "to the adapter %s is ambiguous. " - "Stopping all operations on this " - "adapter.\n", - *(unsigned long long *) - (&fsf_req->qtcb->bottom.support. - req_handle), + "Stopping all operations on this adapter.\n", + *(unsigned long long*) + (&qtcb->bottom.support.req_handle), zfcp_get_busid_by_adapter(adapter)); - } else { - ZFCP_LOG_NORMAL("bug: The request identifier %p " - "to the adapter %s is ambiguous. " - "Stopping all operations on this " - "adapter. " - "(bug: got this for an unsolicited " - "status read request)\n", - fsf_req, - zfcp_get_busid_by_adapter(adapter)); - } zfcp_erp_adapter_shutdown(adapter, 0); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; case FSF_PROT_LINK_DOWN: - /* - * 'test and set' is not atomic here - - * it's ok as long as calls to our response queue handler - * (and thus execution of this code here) are serialized - * by the qdio module - */ - if (!atomic_test_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, - &adapter->status)) { - switch (fsf_req->qtcb->prefix.prot_status_qual. - locallink_error.code) { - case FSF_PSQ_LINK_NOLIGHT: - ZFCP_LOG_INFO("The local link to adapter %s " - "is down (no light detected).\n", - zfcp_get_busid_by_adapter( - adapter)); - break; - case FSF_PSQ_LINK_WRAPPLUG: - ZFCP_LOG_INFO("The local link to adapter %s " - "is down (wrap plug detected).\n", - zfcp_get_busid_by_adapter( - adapter)); - break; - case FSF_PSQ_LINK_NOFCP: - ZFCP_LOG_INFO("The local link to adapter %s " - "is down (adjacent node on " - "link does not support FCP).\n", - zfcp_get_busid_by_adapter( - adapter)); - break; - default: - ZFCP_LOG_INFO("The local link to adapter %s " - "is down " - "(warning: unknown reason " - "code).\n", - zfcp_get_busid_by_adapter( - adapter)); - break; - - } - /* - * Due to the 'erp failed' flag the adapter won't - * be recovered but will be just set to 'blocked' - * state. All subordinary devices will have state - * 'blocked' and 'erp failed', too. - * Thus the adapter is still able to provide - * 'link up' status without being flooded with - * requests. - * (note: even 'close port' is not permitted) - */ - ZFCP_LOG_INFO("Stopping all operations for adapter " - "%s.\n", - zfcp_get_busid_by_adapter(adapter)); - atomic_set_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED | - ZFCP_STATUS_COMMON_ERP_FAILED, - &adapter->status); - zfcp_erp_adapter_reopen(adapter, 0); - } + zfcp_fsf_link_down_info_eval(adapter, + &prot_status_qual->link_down_info); fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR; break; @@ -613,6 +547,110 @@ zfcp_fsf_fsfstatus_qual_eval(struct zfcp_fsf_req *fsf_req) return retval; } +/** + * zfcp_fsf_link_down_info_eval - evaluate link down information block + */ +static void +zfcp_fsf_link_down_info_eval(struct zfcp_adapter *adapter, + struct fsf_link_down_info *link_down) +{ + switch (link_down->error_code) { + case FSF_PSQ_LINK_NO_LIGHT: + ZFCP_LOG_NORMAL("The local link to adapter %s is down " + "(no light detected)\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_WRAP_PLUG: + ZFCP_LOG_NORMAL("The local link to adapter %s is down " + "(wrap plug detected)\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_NO_FCP: + ZFCP_LOG_NORMAL("The local link to adapter %s is down " + "(adjacent node on link does not support FCP)\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_FIRMWARE_UPDATE: + ZFCP_LOG_NORMAL("The local link to adapter %s is down " + "(firmware update in progress)\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_INVALID_WWPN: + ZFCP_LOG_NORMAL("The local link to adapter %s is down " + "(duplicate or invalid WWPN detected)\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_NO_NPIV_SUPPORT: + ZFCP_LOG_NORMAL("The local link to adapter %s is down " + "(no support for NPIV by Fabric)\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_NO_FCP_RESOURCES: + ZFCP_LOG_NORMAL("The local link to adapter %s is down " + "(out of resource in FCP daughtercard)\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_NO_FABRIC_RESOURCES: + ZFCP_LOG_NORMAL("The local link to adapter %s is down " + "(out of resource in Fabric)\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_FABRIC_LOGIN_UNABLE: + ZFCP_LOG_NORMAL("The local link to adapter %s is down " + "(unable to Fabric login)\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_WWPN_ASSIGNMENT_CORRUPTED: + ZFCP_LOG_NORMAL("WWPN assignment file corrupted on adapter %s\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_MODE_TABLE_CURRUPTED: + ZFCP_LOG_NORMAL("Mode table corrupted on adapter %s\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_PSQ_LINK_NO_WWPN_ASSIGNMENT: + ZFCP_LOG_NORMAL("No WWPN for assignment table on adapter %s\n", + zfcp_get_busid_by_adapter(adapter)); + break; + default: + ZFCP_LOG_NORMAL("The local link to adapter %s is down " + "(warning: unknown reason code %d)\n", + zfcp_get_busid_by_adapter(adapter), + link_down->error_code); + } + + if (adapter->connection_features & FSF_FEATURE_NPIV_MODE) + ZFCP_LOG_DEBUG("Debug information to link down: " + "primary_status=0x%02x " + "ioerr_code=0x%02x " + "action_code=0x%02x " + "reason_code=0x%02x " + "explanation_code=0x%02x " + "vendor_specific_code=0x%02x\n", + link_down->primary_status, + link_down->ioerr_code, + link_down->action_code, + link_down->reason_code, + link_down->explanation_code, + link_down->vendor_specific_code); + + if (!atomic_test_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, + &adapter->status)) { + atomic_set_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, + &adapter->status); + switch (link_down->error_code) { + case FSF_PSQ_LINK_NO_LIGHT: + case FSF_PSQ_LINK_WRAP_PLUG: + case FSF_PSQ_LINK_NO_FCP: + case FSF_PSQ_LINK_FIRMWARE_UPDATE: + zfcp_erp_adapter_reopen(adapter, 0); + break; + default: + zfcp_erp_adapter_failed(adapter); + } + } +} + /* * function: zfcp_fsf_req_dispatch * @@ -877,17 +915,32 @@ zfcp_fsf_status_read_handler(struct zfcp_fsf_req *fsf_req) break; case FSF_STATUS_READ_LINK_DOWN: - debug_text_event(adapter->erp_dbf, 0, "unsol_link_down:"); - ZFCP_LOG_INFO("Local link to adapter %s is down\n", + switch (status_buffer->status_subtype) { + case FSF_STATUS_READ_SUB_NO_PHYSICAL_LINK: + ZFCP_LOG_INFO("Physical link to adapter %s is down\n", + zfcp_get_busid_by_adapter(adapter)); + break; + case FSF_STATUS_READ_SUB_FDISC_FAILED: + ZFCP_LOG_INFO("Local link to adapter %s is down " + "due to failed FDISC login\n", zfcp_get_busid_by_adapter(adapter)); - atomic_set_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, - &adapter->status); - zfcp_erp_adapter_failed(adapter); + break; + case FSF_STATUS_READ_SUB_FIRMWARE_UPDATE: + ZFCP_LOG_INFO("Local link to adapter %s is down " + "due to firmware update on adapter\n", + zfcp_get_busid_by_adapter(adapter)); + break; + default: + ZFCP_LOG_INFO("Local link to adapter %s is down " + "due to unknown reason\n", + zfcp_get_busid_by_adapter(adapter)); + }; + zfcp_fsf_link_down_info_eval(adapter, + (struct fsf_link_down_info *) &status_buffer->payload); break; case FSF_STATUS_READ_LINK_UP: - debug_text_event(adapter->erp_dbf, 2, "unsol_link_up:"); - ZFCP_LOG_INFO("Local link to adapter %s was replugged. " + ZFCP_LOG_NORMAL("Local link to adapter %s was replugged. " "Restarting operations on this adapter\n", zfcp_get_busid_by_adapter(adapter)); /* All ports should be marked as ready to run again */ @@ -922,6 +975,16 @@ zfcp_fsf_status_read_handler(struct zfcp_fsf_req *fsf_req) } break; + case FSF_STATUS_READ_FEATURE_UPDATE_ALERT: + debug_text_event(adapter->erp_dbf, 2, "unsol_features:"); + ZFCP_LOG_INFO("List of supported features on adapter %s has " + "been changed from 0x%08X to 0x%08X\n", + zfcp_get_busid_by_adapter(adapter), + *(u32*) (status_buffer->payload + 4), + *(u32*) (status_buffer->payload)); + adapter->adapter_features = *(u32*) status_buffer->payload; + break; + default: ZFCP_LOG_NORMAL("warning: An unsolicited status packet of unknown " "type was received (debug info 0x%x)\n", @@ -1281,7 +1344,7 @@ zfcp_fsf_send_ct(struct zfcp_send_ct *ct, mempool_t *pool, sbale[3].addr = zfcp_sg_to_address(&ct->resp[0]); sbale[3].length = ct->resp[0].length; sbale[3].flags |= SBAL_FLAGS_LAST_ENTRY; - } else if (adapter->supported_features & + } else if (adapter->adapter_features & FSF_FEATURE_ELS_CT_CHAINED_SBALS) { /* try to use chained SBALs */ bytes = zfcp_qdio_sbals_from_sg(fsf_req, @@ -1584,7 +1647,7 @@ zfcp_fsf_send_els(struct zfcp_send_els *els) sbale[3].addr = zfcp_sg_to_address(&els->resp[0]); sbale[3].length = els->resp[0].length; sbale[3].flags |= SBAL_FLAGS_LAST_ENTRY; - } else if (adapter->supported_features & + } else if (adapter->adapter_features & FSF_FEATURE_ELS_CT_CHAINED_SBALS) { /* try to use chained SBALs */ bytes = zfcp_qdio_sbals_from_sg(fsf_req, @@ -1877,7 +1940,9 @@ zfcp_fsf_exchange_config_data(struct zfcp_erp_action *erp_action) erp_action->fsf_req->erp_action = erp_action; erp_action->fsf_req->qtcb->bottom.config.feature_selection = - (FSF_FEATURE_CFDC | FSF_FEATURE_LUN_SHARING); + FSF_FEATURE_CFDC | + FSF_FEATURE_LUN_SHARING | + FSF_FEATURE_UPDATE_ALERT; /* start QDIO request for this FSF request */ retval = zfcp_fsf_req_send(erp_action->fsf_req, &erp_action->timer); @@ -1918,7 +1983,8 @@ zfcp_fsf_exchange_config_evaluate(struct zfcp_fsf_req *fsf_req, int xchg_ok) ZFCP_LOG_DEBUG("low/high QTCB version 0x%x/0x%x of FSF\n", bottom->low_qtcb_version, bottom->high_qtcb_version); adapter->fsf_lic_version = bottom->lic_version; - adapter->supported_features = bottom->supported_features; + adapter->adapter_features = bottom->adapter_features; + adapter->connection_features = bottom->connection_features; adapter->peer_wwpn = 0; adapter->peer_wwnn = 0; adapter->peer_d_id = 0; @@ -1930,6 +1996,10 @@ zfcp_fsf_exchange_config_evaluate(struct zfcp_fsf_req *fsf_req, int xchg_ok) adapter->fc_topology = bottom->fc_topology; adapter->fc_link_speed = bottom->fc_link_speed; adapter->hydra_version = bottom->adapter_type; + if (adapter->physical_wwpn == 0) + adapter->physical_wwpn = adapter->wwpn; + if (adapter->physical_s_id == 0) + adapter->physical_s_id = adapter->s_id; } else { adapter->wwnn = 0; adapter->wwpn = 0; @@ -1945,7 +2015,7 @@ zfcp_fsf_exchange_config_evaluate(struct zfcp_fsf_req *fsf_req, int xchg_ok) adapter->peer_wwnn = bottom->plogi_payload.wwnn; } - if(adapter->supported_features & FSF_FEATURE_HBAAPI_MANAGEMENT){ + if (adapter->adapter_features & FSF_FEATURE_HBAAPI_MANAGEMENT) { adapter->hardware_version = bottom->hardware_version; memcpy(adapter->serial_number, bottom->serial_number, 17); EBCASC(adapter->serial_number, sizeof(adapter->serial_number)); @@ -2001,11 +2071,12 @@ zfcp_fsf_exchange_config_data_handler(struct zfcp_fsf_req *fsf_req) { struct fsf_qtcb_bottom_config *bottom; struct zfcp_adapter *adapter = fsf_req->adapter; + struct fsf_qtcb *qtcb = fsf_req->qtcb; if (fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR) return -EIO; - switch (fsf_req->qtcb->header.fsf_status) { + switch (qtcb->header.fsf_status) { case FSF_GOOD: if (zfcp_fsf_exchange_config_evaluate(fsf_req, 1)) @@ -2035,7 +2106,7 @@ zfcp_fsf_exchange_config_data_handler(struct zfcp_fsf_req *fsf_req) zfcp_erp_adapter_shutdown(adapter, 0); return -EIO; case FSF_TOPO_FABRIC: - ZFCP_LOG_INFO("Switched fabric fibrechannel " + ZFCP_LOG_NORMAL("Switched fabric fibrechannel " "network detected at adapter %s.\n", zfcp_get_busid_by_adapter(adapter)); break; @@ -2053,7 +2124,7 @@ zfcp_fsf_exchange_config_data_handler(struct zfcp_fsf_req *fsf_req) zfcp_erp_adapter_shutdown(adapter, 0); return -EIO; } - bottom = &fsf_req->qtcb->bottom.config; + bottom = &qtcb->bottom.config; if (bottom->max_qtcb_size < sizeof(struct fsf_qtcb)) { ZFCP_LOG_NORMAL("bug: Maximum QTCB size (%d bytes) " "allowed by the adapter %s " @@ -2078,12 +2149,10 @@ zfcp_fsf_exchange_config_data_handler(struct zfcp_fsf_req *fsf_req) if (zfcp_fsf_exchange_config_evaluate(fsf_req, 0)) return -EIO; - ZFCP_LOG_INFO("Local link to adapter %s is down\n", - zfcp_get_busid_by_adapter(adapter)); - atomic_set_mask(ZFCP_STATUS_ADAPTER_XCONFIG_OK | - ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, - &adapter->status); - zfcp_erp_adapter_failed(adapter); + atomic_set_mask(ZFCP_STATUS_ADAPTER_XCONFIG_OK, &adapter->status); + + zfcp_fsf_link_down_info_eval(adapter, + &qtcb->header.fsf_status_qual.link_down_info); break; default: debug_text_event(fsf_req->adapter->erp_dbf, 0, "fsf-stat-ng"); @@ -2097,11 +2166,13 @@ zfcp_fsf_exchange_config_data_handler(struct zfcp_fsf_req *fsf_req) /** * zfcp_fsf_exchange_port_data - request information about local port + * @erp_action: ERP action for the adapter for which port data is requested * @adapter: for which port data is requested * @data: response to exchange port data request */ int -zfcp_fsf_exchange_port_data(struct zfcp_adapter *adapter, +zfcp_fsf_exchange_port_data(struct zfcp_erp_action *erp_action, + struct zfcp_adapter *adapter, struct fsf_qtcb_bottom_port *data) { volatile struct qdio_buffer_element *sbale; @@ -2110,7 +2181,7 @@ zfcp_fsf_exchange_port_data(struct zfcp_adapter *adapter, struct zfcp_fsf_req *fsf_req; struct timer_list *timer; - if(!(adapter->supported_features & FSF_FEATURE_HBAAPI_MANAGEMENT)){ + if (!(adapter->adapter_features & FSF_FEATURE_HBAAPI_MANAGEMENT)) { ZFCP_LOG_INFO("error: exchange port data " "command not supported by adapter %s\n", zfcp_get_busid_by_adapter(adapter)); @@ -2134,6 +2205,12 @@ zfcp_fsf_exchange_port_data(struct zfcp_adapter *adapter, goto out; } + if (erp_action) { + erp_action->fsf_req = fsf_req; + fsf_req->erp_action = erp_action; + } + + if (data) fsf_req->data = (unsigned long) data; sbale = zfcp_qdio_sbale_req(fsf_req, fsf_req->sbal_curr, 0); @@ -2151,6 +2228,8 @@ zfcp_fsf_exchange_port_data(struct zfcp_adapter *adapter, "command on the adapter %s\n", zfcp_get_busid_by_adapter(adapter)); zfcp_fsf_req_free(fsf_req); + if (erp_action) + erp_action->fsf_req = NULL; write_unlock_irqrestore(&adapter->request_queue.queue_lock, lock_flags); goto out; @@ -2179,23 +2258,40 @@ zfcp_fsf_exchange_port_data(struct zfcp_adapter *adapter, static void zfcp_fsf_exchange_port_data_handler(struct zfcp_fsf_req *fsf_req) { - struct fsf_qtcb_bottom_port *bottom; - struct fsf_qtcb_bottom_port *data; - - data = (struct fsf_qtcb_bottom_port*) fsf_req->data; + struct zfcp_adapter *adapter = fsf_req->adapter; + struct fsf_qtcb *qtcb = fsf_req->qtcb; + struct fsf_qtcb_bottom_port *bottom, *data; if (fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR) return; - switch (fsf_req->qtcb->header.fsf_status) { + switch (qtcb->header.fsf_status) { case FSF_GOOD: - bottom = &fsf_req->qtcb->bottom.port; - memcpy(data, bottom, sizeof(*data)); + atomic_set_mask(ZFCP_STATUS_ADAPTER_XPORT_OK, &adapter->status); + + bottom = &qtcb->bottom.port; + data = (struct fsf_qtcb_bottom_port*) fsf_req->data; + if (data) + memcpy(data, bottom, sizeof(struct fsf_qtcb_bottom_port)); + if (adapter->connection_features & FSF_FEATURE_NPIV_MODE) { + adapter->physical_wwpn = bottom->wwpn; + adapter->physical_s_id = bottom->fc_port_id; + } else { + adapter->physical_wwpn = adapter->wwpn; + adapter->physical_s_id = adapter->s_id; + } + break; + + case FSF_EXCHANGE_CONFIG_DATA_INCOMPLETE: + atomic_set_mask(ZFCP_STATUS_ADAPTER_XPORT_OK, &adapter->status); + + zfcp_fsf_link_down_info_eval(adapter, + &qtcb->header.fsf_status_qual.link_down_info); break; default: - debug_text_event(fsf_req->adapter->erp_dbf, 0, "xchg-port-ng"); - debug_event(fsf_req->adapter->erp_dbf, 0, + debug_text_event(adapter->erp_dbf, 0, "xchg-port-ng"); + debug_event(adapter->erp_dbf, 0, &fsf_req->qtcb->header.fsf_status, sizeof(u32)); } } @@ -2629,7 +2725,7 @@ zfcp_fsf_close_physical_port(struct zfcp_erp_action *erp_action) &erp_action->port->status); /* save a pointer to this port */ erp_action->fsf_req->data = (unsigned long) erp_action->port; - /* port to be closeed */ + /* port to be closed */ erp_action->fsf_req->qtcb->header.port_handle = erp_action->port->handle; erp_action->fsf_req->erp_action = erp_action; @@ -2833,6 +2929,7 @@ zfcp_fsf_open_unit(struct zfcp_erp_action *erp_action) erp_action->port->handle; erp_action->fsf_req->qtcb->bottom.support.fcp_lun = erp_action->unit->fcp_lun; + if (!(erp_action->adapter->connection_features & FSF_FEATURE_NPIV_MODE)) erp_action->fsf_req->qtcb->bottom.support.option = FSF_OPEN_LUN_SUPPRESS_BOXING; atomic_set_mask(ZFCP_STATUS_COMMON_OPENING, &erp_action->unit->status); @@ -2880,7 +2977,7 @@ zfcp_fsf_open_unit_handler(struct zfcp_fsf_req *fsf_req) struct fsf_qtcb_bottom_support *bottom; struct fsf_queue_designator *queue_designator; u16 subtable, rule, counter; - u32 allowed, exclusive, readwrite; + int exclusive, readwrite; unit = (struct zfcp_unit *) fsf_req->data; @@ -2894,10 +2991,6 @@ zfcp_fsf_open_unit_handler(struct zfcp_fsf_req *fsf_req) bottom = &fsf_req->qtcb->bottom.support; queue_designator = &header->fsf_status_qual.fsf_queue_designator; - allowed = bottom->lun_access_info & FSF_UNIT_ACCESS_OPEN_LUN_ALLOWED; - exclusive = bottom->lun_access_info & FSF_UNIT_ACCESS_EXCLUSIVE; - readwrite = bottom->lun_access_info & FSF_UNIT_ACCESS_OUTBOUND_TRANSFER; - atomic_clear_mask(ZFCP_STATUS_COMMON_ACCESS_DENIED | ZFCP_STATUS_UNIT_SHARED | ZFCP_STATUS_UNIT_READONLY, @@ -3071,10 +3164,15 @@ zfcp_fsf_open_unit_handler(struct zfcp_fsf_req *fsf_req) unit->handle); /* mark unit as open */ atomic_set_mask(ZFCP_STATUS_COMMON_OPEN, &unit->status); - atomic_clear_mask(ZFCP_STATUS_COMMON_ACCESS_DENIED | - ZFCP_STATUS_COMMON_ACCESS_BOXED, - &unit->status); - if (adapter->supported_features & FSF_FEATURE_LUN_SHARING){ + + if (!(adapter->connection_features & FSF_FEATURE_NPIV_MODE) && + (adapter->adapter_features & FSF_FEATURE_LUN_SHARING) && + (adapter->ccw_device->id.dev_model != ZFCP_DEVICE_MODEL_PRIV)) { + exclusive = (bottom->lun_access_info & + FSF_UNIT_ACCESS_EXCLUSIVE); + readwrite = (bottom->lun_access_info & + FSF_UNIT_ACCESS_OUTBOUND_TRANSFER); + if (!exclusive) atomic_set_mask(ZFCP_STATUS_UNIT_SHARED, &unit->status); @@ -4164,7 +4262,7 @@ zfcp_fsf_control_file(struct zfcp_adapter *adapter, int direction; int retval = 0; - if (!(adapter->supported_features & FSF_FEATURE_CFDC)) { + if (!(adapter->adapter_features & FSF_FEATURE_CFDC)) { ZFCP_LOG_INFO("cfdc not supported (adapter %s)\n", zfcp_get_busid_by_adapter(adapter)); retval = -EOPNOTSUPP; diff --git a/drivers/s390/scsi/zfcp_fsf.h b/drivers/s390/scsi/zfcp_fsf.h index 57ce0706007e..48719f055952 100644 --- a/drivers/s390/scsi/zfcp_fsf.h +++ b/drivers/s390/scsi/zfcp_fsf.h @@ -140,9 +140,18 @@ #define FSF_SQ_CFDC_SUBTABLE_LUN 0x0004 /* FSF status qualifier (most significant 4 bytes), local link down */ -#define FSF_PSQ_LINK_NOLIGHT 0x00000004 -#define FSF_PSQ_LINK_WRAPPLUG 0x00000008 -#define FSF_PSQ_LINK_NOFCP 0x00000010 +#define FSF_PSQ_LINK_NO_LIGHT 0x00000004 +#define FSF_PSQ_LINK_WRAP_PLUG 0x00000008 +#define FSF_PSQ_LINK_NO_FCP 0x00000010 +#define FSF_PSQ_LINK_FIRMWARE_UPDATE 0x00000020 +#define FSF_PSQ_LINK_INVALID_WWPN 0x00000100 +#define FSF_PSQ_LINK_NO_NPIV_SUPPORT 0x00000200 +#define FSF_PSQ_LINK_NO_FCP_RESOURCES 0x00000400 +#define FSF_PSQ_LINK_NO_FABRIC_RESOURCES 0x00000800 +#define FSF_PSQ_LINK_FABRIC_LOGIN_UNABLE 0x00001000 +#define FSF_PSQ_LINK_WWPN_ASSIGNMENT_CORRUPTED 0x00002000 +#define FSF_PSQ_LINK_MODE_TABLE_CURRUPTED 0x00004000 +#define FSF_PSQ_LINK_NO_WWPN_ASSIGNMENT 0x00008000 /* payload size in status read buffer */ #define FSF_STATUS_READ_PAYLOAD_SIZE 4032 @@ -155,15 +164,21 @@ #define FSF_STATUS_READ_INCOMING_ELS 0x00000002 #define FSF_STATUS_READ_SENSE_DATA_AVAIL 0x00000003 #define FSF_STATUS_READ_BIT_ERROR_THRESHOLD 0x00000004 -#define FSF_STATUS_READ_LINK_DOWN 0x00000005 /* FIXME: really? */ +#define FSF_STATUS_READ_LINK_DOWN 0x00000005 #define FSF_STATUS_READ_LINK_UP 0x00000006 #define FSF_STATUS_READ_CFDC_UPDATED 0x0000000A #define FSF_STATUS_READ_CFDC_HARDENED 0x0000000B +#define FSF_STATUS_READ_FEATURE_UPDATE_ALERT 0x0000000C /* status subtypes in status read buffer */ #define FSF_STATUS_READ_SUB_CLOSE_PHYS_PORT 0x00000001 #define FSF_STATUS_READ_SUB_ERROR_PORT 0x00000002 +/* status subtypes for link down */ +#define FSF_STATUS_READ_SUB_NO_PHYSICAL_LINK 0x00000000 +#define FSF_STATUS_READ_SUB_FDISC_FAILED 0x00000001 +#define FSF_STATUS_READ_SUB_FIRMWARE_UPDATE 0x00000002 + /* status subtypes for CFDC */ #define FSF_STATUS_READ_SUB_CFDC_HARDENED_ON_SE 0x00000002 #define FSF_STATUS_READ_SUB_CFDC_HARDENED_ON_SE2 0x0000000F @@ -194,11 +209,15 @@ #define FSF_QTCB_LOG_SIZE 1024 /* channel features */ -#define FSF_FEATURE_QTCB_SUPPRESSION 0x00000001 #define FSF_FEATURE_CFDC 0x00000002 #define FSF_FEATURE_LUN_SHARING 0x00000004 #define FSF_FEATURE_HBAAPI_MANAGEMENT 0x00000010 #define FSF_FEATURE_ELS_CT_CHAINED_SBALS 0x00000020 +#define FSF_FEATURE_UPDATE_ALERT 0x00000100 + +/* host connection features */ +#define FSF_FEATURE_NPIV_MODE 0x00000001 +#define FSF_FEATURE_VM_ASSIGNED_WWPN 0x00000002 /* option */ #define FSF_OPEN_LUN_SUPPRESS_BOXING 0x00000001 @@ -306,16 +325,23 @@ struct fsf_qual_sequence_error { u32 res1[3]; } __attribute__ ((packed)); -struct fsf_qual_locallink_error { - u32 code; - u32 res1[3]; +struct fsf_link_down_info { + u32 error_code; + u32 res1; + u8 res2[2]; + u8 primary_status; + u8 ioerr_code; + u8 action_code; + u8 reason_code; + u8 explanation_code; + u8 vendor_specific_code; } __attribute__ ((packed)); union fsf_prot_status_qual { u64 doubleword[FSF_PROT_STATUS_QUAL_SIZE / sizeof(u64)]; struct fsf_qual_version_error version_error; struct fsf_qual_sequence_error sequence_error; - struct fsf_qual_locallink_error locallink_error; + struct fsf_link_down_info link_down_info; } __attribute__ ((packed)); struct fsf_qtcb_prefix { @@ -335,6 +361,7 @@ union fsf_status_qual { u32 word[FSF_STATUS_QUALIFIER_SIZE / sizeof (u32)]; u64 doubleword[FSF_STATUS_QUALIFIER_SIZE / sizeof(u64)]; struct fsf_queue_designator fsf_queue_designator; + struct fsf_link_down_info link_down_info; } __attribute__ ((packed)); struct fsf_qtcb_header { @@ -409,8 +436,8 @@ struct fsf_qtcb_bottom_config { u32 low_qtcb_version; u32 max_qtcb_size; u32 max_data_transfer_size; - u32 supported_features; - u8 res1[4]; + u32 adapter_features; + u32 connection_features; u32 fc_topology; u32 fc_link_speed; u32 adapter_type; @@ -428,7 +455,7 @@ struct fsf_qtcb_bottom_config { } __attribute__ ((packed)); struct fsf_qtcb_bottom_port { - u8 res1[8]; + u64 wwpn; u32 fc_port_id; u32 port_type; u32 port_state; diff --git a/drivers/s390/scsi/zfcp_sysfs_adapter.c b/drivers/s390/scsi/zfcp_sysfs_adapter.c index e7345a74800a..c406ea5c5e95 100644 --- a/drivers/s390/scsi/zfcp_sysfs_adapter.c +++ b/drivers/s390/scsi/zfcp_sysfs_adapter.c @@ -68,6 +68,8 @@ ZFCP_DEFINE_ADAPTER_ATTR(s_id, "0x%06x\n", adapter->s_id); ZFCP_DEFINE_ADAPTER_ATTR(peer_wwnn, "0x%016llx\n", adapter->peer_wwnn); ZFCP_DEFINE_ADAPTER_ATTR(peer_wwpn, "0x%016llx\n", adapter->peer_wwpn); ZFCP_DEFINE_ADAPTER_ATTR(peer_d_id, "0x%06x\n", adapter->peer_d_id); +ZFCP_DEFINE_ADAPTER_ATTR(physical_wwpn, "0x%016llx\n", adapter->physical_wwpn); +ZFCP_DEFINE_ADAPTER_ATTR(physical_s_id, "0x%06x\n", adapter->physical_s_id); ZFCP_DEFINE_ADAPTER_ATTR(card_version, "0x%04x\n", adapter->hydra_version); ZFCP_DEFINE_ADAPTER_ATTR(lic_version, "0x%08x\n", adapter->fsf_lic_version); ZFCP_DEFINE_ADAPTER_ATTR(fc_link_speed, "%d Gb/s\n", adapter->fc_link_speed); @@ -261,6 +263,8 @@ static struct attribute *zfcp_adapter_attrs[] = { &dev_attr_peer_wwnn.attr, &dev_attr_peer_wwpn.attr, &dev_attr_peer_d_id.attr, + &dev_attr_physical_wwpn.attr, + &dev_attr_physical_s_id.attr, &dev_attr_card_version.attr, &dev_attr_lic_version.attr, &dev_attr_fc_link_speed.attr, From 13e1e1f08c1c098c7574c1fa72bd8c67792dc89b Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Mon, 19 Sep 2005 16:56:17 +0200 Subject: [PATCH 174/240] [SCSI] zfcp: add additional fc_host attributes this patch adds some fc host attributes and removes its equivalents from the zfcp_adapter structure and zfcp specific sysfs subtree. Furthermore it removes superfluous calls to fc_remort_port_delete when an adapter is set offline because rports will be removed by fc_remove_host anyway. Signed-off-by: Andreas Herrmann Signed-off-by: James Bottomley --- drivers/s390/scsi/zfcp_ccw.c | 10 ---- drivers/s390/scsi/zfcp_dbf.c | 32 ++++++------- drivers/s390/scsi/zfcp_def.h | 16 ++----- drivers/s390/scsi/zfcp_erp.c | 16 +++---- drivers/s390/scsi/zfcp_fsf.c | 65 ++++++++++++++------------ drivers/s390/scsi/zfcp_scsi.c | 21 ++++----- drivers/s390/scsi/zfcp_sysfs_adapter.c | 10 ---- 7 files changed, 72 insertions(+), 98 deletions(-) diff --git a/drivers/s390/scsi/zfcp_ccw.c b/drivers/s390/scsi/zfcp_ccw.c index b30abab77da3..0fc46381fc22 100644 --- a/drivers/s390/scsi/zfcp_ccw.c +++ b/drivers/s390/scsi/zfcp_ccw.c @@ -202,19 +202,9 @@ static int zfcp_ccw_set_offline(struct ccw_device *ccw_device) { struct zfcp_adapter *adapter; - struct zfcp_port *port; - struct fc_rport *rport; down(&zfcp_data.config_sema); adapter = dev_get_drvdata(&ccw_device->dev); - /* might be racy, but we cannot take config_lock due to the fact that - fc_remote_port_delete might sleep */ - list_for_each_entry(port, &adapter->port_list_head, list) - if (port->rport) { - rport = port->rport; - port->rport = NULL; - fc_remote_port_delete(rport); - } zfcp_erp_adapter_shutdown(adapter, 0); zfcp_erp_wait(adapter); zfcp_adapter_scsi_unregister(adapter); diff --git a/drivers/s390/scsi/zfcp_dbf.c b/drivers/s390/scsi/zfcp_dbf.c index 64d9b90373fe..826fb3b00605 100644 --- a/drivers/s390/scsi/zfcp_dbf.c +++ b/drivers/s390/scsi/zfcp_dbf.c @@ -486,8 +486,7 @@ struct debug_view zfcp_hba_dbf_view = { inline void _zfcp_san_dbf_event_common_ct(const char *tag, struct zfcp_fsf_req *fsf_req, - fc_id_t s_id, fc_id_t d_id, - void *buffer, int buflen) + u32 s_id, u32 d_id, void *buffer, int buflen) { struct zfcp_send_ct *send_ct = (struct zfcp_send_ct *)fsf_req->data; struct zfcp_port *port = send_ct->port; @@ -532,9 +531,9 @@ inline void zfcp_san_dbf_event_ct_request(struct zfcp_fsf_req *fsf_req) struct zfcp_port *port = ct->port; struct zfcp_adapter *adapter = port->adapter; - _zfcp_san_dbf_event_common_ct("octc", - fsf_req, adapter->s_id, port->d_id, - zfcp_sg_to_address(ct->req), + _zfcp_san_dbf_event_common_ct("octc", fsf_req, + fc_host_port_id(adapter->scsi_host), + port->d_id, zfcp_sg_to_address(ct->req), ct->req->length); } @@ -544,17 +543,16 @@ inline void zfcp_san_dbf_event_ct_response(struct zfcp_fsf_req *fsf_req) struct zfcp_port *port = ct->port; struct zfcp_adapter *adapter = port->adapter; - _zfcp_san_dbf_event_common_ct("rctc", - fsf_req, port->d_id, adapter->s_id, + _zfcp_san_dbf_event_common_ct("rctc", fsf_req, port->d_id, + fc_host_port_id(adapter->scsi_host), zfcp_sg_to_address(ct->resp), ct->resp->length); } static inline void _zfcp_san_dbf_event_common_els(const char *tag, int level, - struct zfcp_fsf_req *fsf_req, - fc_id_t s_id, fc_id_t d_id, u8 ls_code, - void *buffer, int buflen) + struct zfcp_fsf_req *fsf_req, u32 s_id, + u32 d_id, u8 ls_code, void *buffer, int buflen) { struct zfcp_adapter *adapter = fsf_req->adapter; struct zfcp_san_dbf_record *rec = &adapter->san_dbf_buf; @@ -597,8 +595,9 @@ inline void zfcp_san_dbf_event_els_request(struct zfcp_fsf_req *fsf_req) { struct zfcp_send_els *els = (struct zfcp_send_els *)fsf_req->data; - _zfcp_san_dbf_event_common_els("oels", 2, - fsf_req, els->adapter->s_id, els->d_id, + _zfcp_san_dbf_event_common_els("oels", 2, fsf_req, + fc_host_port_id(els->adapter->scsi_host), + els->d_id, *(u8 *) zfcp_sg_to_address(els->req), zfcp_sg_to_address(els->req), els->req->length); @@ -608,8 +607,8 @@ inline void zfcp_san_dbf_event_els_response(struct zfcp_fsf_req *fsf_req) { struct zfcp_send_els *els = (struct zfcp_send_els *)fsf_req->data; - _zfcp_san_dbf_event_common_els("rels", 2, - fsf_req, els->d_id, els->adapter->s_id, + _zfcp_san_dbf_event_common_els("rels", 2, fsf_req, els->d_id, + fc_host_port_id(els->adapter->scsi_host), *(u8 *) zfcp_sg_to_address(els->req), zfcp_sg_to_address(els->resp), els->resp->length); @@ -623,9 +622,8 @@ inline void zfcp_san_dbf_event_incoming_els(struct zfcp_fsf_req *fsf_req) int length = (int)status_buffer->length - (int)((void *)&status_buffer->payload - (void *)status_buffer); - _zfcp_san_dbf_event_common_els("iels", 1, - fsf_req, status_buffer->d_id, - adapter->s_id, + _zfcp_san_dbf_event_common_els("iels", 1, fsf_req, status_buffer->d_id, + fc_host_port_id(adapter->scsi_host), *(u8 *) status_buffer->payload, (void *)status_buffer->payload, length); } diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h index 4b6af8e07e8d..d81b737d68cc 100644 --- a/drivers/s390/scsi/zfcp_def.h +++ b/drivers/s390/scsi/zfcp_def.h @@ -165,7 +165,6 @@ typedef u32 scsi_lun_t; /*************** FIBRE CHANNEL PROTOCOL SPECIFIC DEFINES ********************/ typedef unsigned long long wwn_t; -typedef unsigned int fc_id_t; typedef unsigned long long fcp_lun_t; /* data length field may be at variable position in FCP-2 FCP_CMND IU */ typedef unsigned int fcp_dl_t; @@ -806,7 +805,7 @@ struct ct_iu_gid_pn_req { /* FS_ACC IU and data unit for GID_PN nameserver request */ struct ct_iu_gid_pn_resp { struct ct_hdr header; - fc_id_t d_id; + u32 d_id; } __attribute__ ((packed)); typedef void (*zfcp_send_ct_handler_t)(unsigned long); @@ -872,7 +871,7 @@ typedef void (*zfcp_send_els_handler_t)(unsigned long); struct zfcp_send_els { struct zfcp_adapter *adapter; struct zfcp_port *port; - fc_id_t d_id; + u32 d_id; struct scatterlist *req; struct scatterlist *resp; unsigned int req_count; @@ -915,24 +914,19 @@ struct zfcp_adapter { atomic_t refcount; /* reference count */ wait_queue_head_t remove_wq; /* can be used to wait for refcount drop to zero */ - wwn_t wwnn; /* WWNN */ - wwn_t wwpn; /* WWPN */ - fc_id_t s_id; /* N_Port ID */ wwn_t peer_wwnn; /* P2P peer WWNN */ wwn_t peer_wwpn; /* P2P peer WWPN */ - fc_id_t peer_d_id; /* P2P peer D_ID */ + u32 peer_d_id; /* P2P peer D_ID */ wwn_t physical_wwpn; /* WWPN of physical port */ - fc_id_t physical_s_id; /* local FC port ID */ + u32 physical_s_id; /* local FC port ID */ struct ccw_device *ccw_device; /* S/390 ccw device */ u8 fc_service_class; u32 fc_topology; /* FC topology */ - u32 fc_link_speed; /* FC interface speed */ u32 hydra_version; /* Hydra version */ u32 fsf_lic_version; u32 adapter_features; /* FCP channel features */ u32 connection_features; /* host connection features */ u32 hardware_version; /* of FCP channel */ - u8 serial_number[32]; /* of hardware */ struct Scsi_Host *scsi_host; /* Pointer to mid-layer */ unsigned short scsi_host_no; /* Assigned host number */ unsigned char name[9]; @@ -1006,7 +1000,7 @@ struct zfcp_port { atomic_t status; /* status of this remote port */ wwn_t wwnn; /* WWNN if known */ wwn_t wwpn; /* WWPN */ - fc_id_t d_id; /* D_ID */ + u32 d_id; /* D_ID */ u32 handle; /* handle assigned by FSF */ struct zfcp_erp_action erp_action; /* pending error recovery */ atomic_t erp_counter; diff --git a/drivers/s390/scsi/zfcp_erp.c b/drivers/s390/scsi/zfcp_erp.c index c4a6799aed44..023f4e558ae4 100644 --- a/drivers/s390/scsi/zfcp_erp.c +++ b/drivers/s390/scsi/zfcp_erp.c @@ -346,13 +346,13 @@ zfcp_erp_adisc(struct zfcp_port *port) /* acc. to FC-FS, hard_nport_id in ADISC should not be set for ports without FC-AL-2 capability, so we don't set it */ - adisc->wwpn = adapter->wwpn; - adisc->wwnn = adapter->wwnn; - adisc->nport_id = adapter->s_id; + adisc->wwpn = fc_host_port_name(adapter->scsi_host); + adisc->wwnn = fc_host_node_name(adapter->scsi_host); + adisc->nport_id = fc_host_port_id(adapter->scsi_host); ZFCP_LOG_INFO("ADISC request from s_id 0x%08x to d_id 0x%08x " "(wwpn=0x%016Lx, wwnn=0x%016Lx, " "hard_nport_id=0x%08x, nport_id=0x%08x)\n", - adapter->s_id, send_els->d_id, (wwn_t) adisc->wwpn, + adisc->nport_id, send_els->d_id, (wwn_t) adisc->wwpn, (wwn_t) adisc->wwnn, adisc->hard_nport_id, adisc->nport_id); @@ -405,7 +405,7 @@ zfcp_erp_adisc_handler(unsigned long data) struct zfcp_send_els *send_els; struct zfcp_port *port; struct zfcp_adapter *adapter; - fc_id_t d_id; + u32 d_id; struct zfcp_ls_adisc_acc *adisc; send_els = (struct zfcp_send_els *) data; @@ -436,9 +436,9 @@ zfcp_erp_adisc_handler(unsigned long data) ZFCP_LOG_INFO("ADISC response from d_id 0x%08x to s_id " "0x%08x (wwpn=0x%016Lx, wwnn=0x%016Lx, " "hard_nport_id=0x%08x, nport_id=0x%08x)\n", - d_id, adapter->s_id, (wwn_t) adisc->wwpn, - (wwn_t) adisc->wwnn, adisc->hard_nport_id, - adisc->nport_id); + d_id, fc_host_port_id(adapter->scsi_host), + (wwn_t) adisc->wwpn, (wwn_t) adisc->wwnn, + adisc->hard_nport_id, adisc->nport_id); /* set wwnn for port */ if (port->wwnn == 0) diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index f994f453e2d2..3b0fc1163f5f 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -1617,7 +1617,7 @@ zfcp_fsf_send_els(struct zfcp_send_els *els) { volatile struct qdio_buffer_element *sbale; struct zfcp_fsf_req *fsf_req; - fc_id_t d_id; + u32 d_id; struct zfcp_adapter *adapter; unsigned long lock_flags; int bytes; @@ -1740,7 +1740,7 @@ static int zfcp_fsf_send_els_handler(struct zfcp_fsf_req *fsf_req) { struct zfcp_adapter *adapter; struct zfcp_port *port; - fc_id_t d_id; + u32 d_id; struct fsf_qtcb_header *header; struct fsf_qtcb_bottom_support *bottom; struct zfcp_send_els *send_els; @@ -1978,6 +1978,7 @@ zfcp_fsf_exchange_config_evaluate(struct zfcp_fsf_req *fsf_req, int xchg_ok) { struct fsf_qtcb_bottom_config *bottom; struct zfcp_adapter *adapter = fsf_req->adapter; + struct Scsi_Host *shost = adapter->scsi_host; bottom = &fsf_req->qtcb->bottom.config; ZFCP_LOG_DEBUG("low/high QTCB version 0x%x/0x%x of FSF\n", @@ -1990,22 +1991,23 @@ zfcp_fsf_exchange_config_evaluate(struct zfcp_fsf_req *fsf_req, int xchg_ok) adapter->peer_d_id = 0; if (xchg_ok) { - adapter->wwnn = bottom->nport_serv_param.wwnn; - adapter->wwpn = bottom->nport_serv_param.wwpn; - adapter->s_id = bottom->s_id & ZFCP_DID_MASK; + fc_host_node_name(shost) = bottom->nport_serv_param.wwnn; + fc_host_port_name(shost) = bottom->nport_serv_param.wwpn; + fc_host_port_id(shost) = bottom->s_id & ZFCP_DID_MASK; + fc_host_speed(shost) = bottom->fc_link_speed; + fc_host_supported_classes(shost) = FC_COS_CLASS2 | FC_COS_CLASS3; adapter->fc_topology = bottom->fc_topology; - adapter->fc_link_speed = bottom->fc_link_speed; adapter->hydra_version = bottom->adapter_type; if (adapter->physical_wwpn == 0) - adapter->physical_wwpn = adapter->wwpn; + adapter->physical_wwpn = fc_host_port_name(shost); if (adapter->physical_s_id == 0) - adapter->physical_s_id = adapter->s_id; + adapter->physical_s_id = fc_host_port_id(shost); } else { - adapter->wwnn = 0; - adapter->wwpn = 0; - adapter->s_id = 0; + fc_host_node_name(shost) = 0; + fc_host_port_name(shost) = 0; + fc_host_port_id(shost) = 0; + fc_host_speed(shost) = FC_PORTSPEED_UNKNOWN; adapter->fc_topology = 0; - adapter->fc_link_speed = 0; adapter->hydra_version = 0; } @@ -2017,24 +2019,26 @@ zfcp_fsf_exchange_config_evaluate(struct zfcp_fsf_req *fsf_req, int xchg_ok) if (adapter->adapter_features & FSF_FEATURE_HBAAPI_MANAGEMENT) { adapter->hardware_version = bottom->hardware_version; - memcpy(adapter->serial_number, bottom->serial_number, 17); - EBCASC(adapter->serial_number, sizeof(adapter->serial_number)); + memcpy(fc_host_serial_number(shost), bottom->serial_number, + min(FC_SERIAL_NUMBER_SIZE, 17)); + EBCASC(fc_host_serial_number(shost), + min(FC_SERIAL_NUMBER_SIZE, 17)); } ZFCP_LOG_NORMAL("The adapter %s reported the following characteristics:\n" - "WWNN 0x%016Lx, " - "WWPN 0x%016Lx, " - "S_ID 0x%08x,\n" - "adapter version 0x%x, " - "LIC version 0x%x, " - "FC link speed %d Gb/s\n", - zfcp_get_busid_by_adapter(adapter), - adapter->wwnn, - adapter->wwpn, - (unsigned int) adapter->s_id, - adapter->hydra_version, - adapter->fsf_lic_version, - adapter->fc_link_speed); + "WWNN 0x%016Lx, " + "WWPN 0x%016Lx, " + "S_ID 0x%08x,\n" + "adapter version 0x%x, " + "LIC version 0x%x, " + "FC link speed %d Gb/s\n", + zfcp_get_busid_by_adapter(adapter), + (wwn_t) fc_host_node_name(shost), + (wwn_t) fc_host_port_name(shost), + fc_host_port_id(shost), + adapter->hydra_version, + adapter->fsf_lic_version, + fc_host_speed(shost)); if (ZFCP_QTCB_VERSION < bottom->low_qtcb_version) { ZFCP_LOG_NORMAL("error: the adapter %s " "only supports newer control block " @@ -2055,7 +2059,6 @@ zfcp_fsf_exchange_config_evaluate(struct zfcp_fsf_req *fsf_req, int xchg_ok) zfcp_erp_adapter_shutdown(adapter, 0); return -EIO; } - zfcp_set_fc_host_attrs(adapter); return 0; } @@ -2259,6 +2262,7 @@ static void zfcp_fsf_exchange_port_data_handler(struct zfcp_fsf_req *fsf_req) { struct zfcp_adapter *adapter = fsf_req->adapter; + struct Scsi_Host *shost = adapter->scsi_host; struct fsf_qtcb *qtcb = fsf_req->qtcb; struct fsf_qtcb_bottom_port *bottom, *data; @@ -2277,9 +2281,10 @@ zfcp_fsf_exchange_port_data_handler(struct zfcp_fsf_req *fsf_req) adapter->physical_wwpn = bottom->wwpn; adapter->physical_s_id = bottom->fc_port_id; } else { - adapter->physical_wwpn = adapter->wwpn; - adapter->physical_s_id = adapter->s_id; + adapter->physical_wwpn = fc_host_port_name(shost); + adapter->physical_s_id = fc_host_port_id(shost); } + fc_host_maxframe_size(shost) = bottom->maximum_frame_size; break; case FSF_EXCHANGE_CONFIG_DATA_INCOMPLETE: diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index c1403a23174f..3dcd1bfba3b4 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -698,10 +698,16 @@ void zfcp_adapter_scsi_unregister(struct zfcp_adapter *adapter) { struct Scsi_Host *shost; + struct zfcp_port *port; shost = adapter->scsi_host; if (!shost) return; + read_lock_irq(&zfcp_data.config_lock); + list_for_each_entry(port, &adapter->port_list_head, list) + if (port->rport) + port->rport = NULL; + read_unlock_irq(&zfcp_data.config_lock); fc_remove_host(shost); scsi_remove_host(shost); scsi_host_put(shost); @@ -776,18 +782,6 @@ zfcp_get_node_name(struct scsi_target *starget) read_unlock_irqrestore(&zfcp_data.config_lock, flags); } -void -zfcp_set_fc_host_attrs(struct zfcp_adapter *adapter) -{ - struct Scsi_Host *shost = adapter->scsi_host; - - fc_host_node_name(shost) = adapter->wwnn; - fc_host_port_name(shost) = adapter->wwpn; - strncpy(fc_host_serial_number(shost), adapter->serial_number, - min(FC_SERIAL_NUMBER_SIZE, 32)); - fc_host_supported_classes(shost) = FC_COS_CLASS2 | FC_COS_CLASS3; -} - struct fc_function_template zfcp_transport_functions = { .get_starget_port_id = zfcp_get_port_id, .get_starget_port_name = zfcp_get_port_name, @@ -799,7 +793,10 @@ struct fc_function_template zfcp_transport_functions = { .show_host_node_name = 1, .show_host_port_name = 1, .show_host_supported_classes = 1, + .show_host_maxframe_size = 1, .show_host_serial_number = 1, + .show_host_speed = 1, + .show_host_port_id = 1, }; /** diff --git a/drivers/s390/scsi/zfcp_sysfs_adapter.c b/drivers/s390/scsi/zfcp_sysfs_adapter.c index c406ea5c5e95..0cd435280e7d 100644 --- a/drivers/s390/scsi/zfcp_sysfs_adapter.c +++ b/drivers/s390/scsi/zfcp_sysfs_adapter.c @@ -62,9 +62,6 @@ static ssize_t zfcp_sysfs_adapter_##_name##_show(struct device *dev, struct devi static DEVICE_ATTR(_name, S_IRUGO, zfcp_sysfs_adapter_##_name##_show, NULL); ZFCP_DEFINE_ADAPTER_ATTR(status, "0x%08x\n", atomic_read(&adapter->status)); -ZFCP_DEFINE_ADAPTER_ATTR(wwnn, "0x%016llx\n", adapter->wwnn); -ZFCP_DEFINE_ADAPTER_ATTR(wwpn, "0x%016llx\n", adapter->wwpn); -ZFCP_DEFINE_ADAPTER_ATTR(s_id, "0x%06x\n", adapter->s_id); ZFCP_DEFINE_ADAPTER_ATTR(peer_wwnn, "0x%016llx\n", adapter->peer_wwnn); ZFCP_DEFINE_ADAPTER_ATTR(peer_wwpn, "0x%016llx\n", adapter->peer_wwpn); ZFCP_DEFINE_ADAPTER_ATTR(peer_d_id, "0x%06x\n", adapter->peer_d_id); @@ -72,13 +69,11 @@ ZFCP_DEFINE_ADAPTER_ATTR(physical_wwpn, "0x%016llx\n", adapter->physical_wwpn); ZFCP_DEFINE_ADAPTER_ATTR(physical_s_id, "0x%06x\n", adapter->physical_s_id); ZFCP_DEFINE_ADAPTER_ATTR(card_version, "0x%04x\n", adapter->hydra_version); ZFCP_DEFINE_ADAPTER_ATTR(lic_version, "0x%08x\n", adapter->fsf_lic_version); -ZFCP_DEFINE_ADAPTER_ATTR(fc_link_speed, "%d Gb/s\n", adapter->fc_link_speed); ZFCP_DEFINE_ADAPTER_ATTR(fc_service_class, "%d\n", adapter->fc_service_class); ZFCP_DEFINE_ADAPTER_ATTR(fc_topology, "%s\n", fc_topologies[adapter->fc_topology]); ZFCP_DEFINE_ADAPTER_ATTR(hardware_version, "0x%08x\n", adapter->hardware_version); -ZFCP_DEFINE_ADAPTER_ATTR(serial_number, "%17s\n", adapter->serial_number); ZFCP_DEFINE_ADAPTER_ATTR(scsi_host_no, "0x%x\n", adapter->scsi_host_no); ZFCP_DEFINE_ADAPTER_ATTR(in_recovery, "%d\n", atomic_test_mask (ZFCP_STATUS_COMMON_ERP_INUSE, &adapter->status)); @@ -257,9 +252,6 @@ static struct attribute *zfcp_adapter_attrs[] = { &dev_attr_in_recovery.attr, &dev_attr_port_remove.attr, &dev_attr_port_add.attr, - &dev_attr_wwnn.attr, - &dev_attr_wwpn.attr, - &dev_attr_s_id.attr, &dev_attr_peer_wwnn.attr, &dev_attr_peer_wwpn.attr, &dev_attr_peer_d_id.attr, @@ -267,13 +259,11 @@ static struct attribute *zfcp_adapter_attrs[] = { &dev_attr_physical_s_id.attr, &dev_attr_card_version.attr, &dev_attr_lic_version.attr, - &dev_attr_fc_link_speed.attr, &dev_attr_fc_service_class.attr, &dev_attr_fc_topology.attr, &dev_attr_scsi_host_no.attr, &dev_attr_status.attr, &dev_attr_hardware_version.attr, - &dev_attr_serial_number.attr, NULL }; From eff4c654b1a4a5e5493fbdc3affa6dd48765c085 Mon Sep 17 00:00:00 2001 From: Hal Rosenstock Date: Mon, 19 Sep 2005 13:51:01 -0700 Subject: [PATCH 175/240] [PATCH] IB: Fix data length for RMPP SA sends We need to subtract off the header length from our payload length when sending multi-packet SA messages. Signed-off-by: Hal Rosenstock Signed-off-by: Roland Dreier --- drivers/infiniband/core/user_mad.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c index 7c2f03057ddb..a64d6b4dcc16 100644 --- a/drivers/infiniband/core/user_mad.c +++ b/drivers/infiniband/core/user_mad.c @@ -334,10 +334,11 @@ static ssize_t ib_umad_write(struct file *filp, const char __user *buf, ret = -EINVAL; goto err_ah; } - /* Validate that management class can support RMPP */ + + /* Validate that the management class can support RMPP */ if (rmpp_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_ADM) { hdr_len = offsetof(struct ib_sa_mad, data); - data_len = length; + data_len = length - hdr_len; } else if ((rmpp_mad->mad_hdr.mgmt_class >= IB_MGMT_CLASS_VENDOR_RANGE2_START) && (rmpp_mad->mad_hdr.mgmt_class <= IB_MGMT_CLASS_VENDOR_RANGE2_END)) { hdr_len = offsetof(struct ib_vendor_mad, data); From 1d25cd6cc2528e4af12ab18e84fe95ed78f3f21a Mon Sep 17 00:00:00 2001 From: Robert Olsson Date: Mon, 19 Sep 2005 15:29:52 -0700 Subject: [PATCH 176/240] [IPV4]: fib_trie tnode stats refinements * Prints the route tnode and set the stats level deepth as before. Signed-off-by: Robert Olsson Signed-off-by: David S. Miller --- net/ipv4/fib_trie.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 1b63b4824164..3dd70787081e 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -2029,7 +2029,7 @@ static struct node *fib_trie_get_first(struct fib_trie_iter *iter, iter->tnode = (struct tnode *) n; iter->trie = t; iter->index = 0; - iter->depth = 0; + iter->depth = 1; return n; } return NULL; @@ -2274,11 +2274,12 @@ static int fib_trie_seq_show(struct seq_file *seq, void *v) seq_puts(seq, ":\n"); else seq_puts(seq, "
:\n"); - } else { - seq_indent(seq, iter->depth-1); - seq_printf(seq, " +-- %d.%d.%d.%d/%d\n", - NIPQUAD(prf), tn->pos); - } + } + seq_indent(seq, iter->depth-1); + seq_printf(seq, " +-- %d.%d.%d.%d/%d %d %d %d\n", + NIPQUAD(prf), tn->pos, tn->bits, tn->full_children, + tn->empty_children); + } else { struct leaf *l = (struct leaf *) n; int i; From 772cb712b1373d335ef2874ea357ec681edc754b Mon Sep 17 00:00:00 2001 From: Robert Olsson Date: Mon, 19 Sep 2005 15:31:18 -0700 Subject: [PATCH 177/240] [IPV4]: fib_trie RCU refinements * This patch is from Paul McKenney's RCU reviewing. Signed-off-by: Robert Olsson Signed-off-by: David S. Miller --- net/ipv4/fib_trie.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 3dd70787081e..90ae70870a10 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -43,7 +43,7 @@ * 2 of the License, or (at your option) any later version. */ -#define VERSION "0.403" +#define VERSION "0.404" #include #include @@ -224,7 +224,7 @@ static inline int tkey_mismatch(t_key a, int offset, t_key b) Consider a node 'n' and its parent 'tp'. If n is a leaf, every bit in its key is significant. Its presence is - necessitaded by path compression, since during a tree traversal (when + necessitated by path compression, since during a tree traversal (when searching for a leaf - unless we are doing an insertion) we will completely ignore all skipped bits we encounter. Thus we need to verify, at the end of a potentially successful search, that we have indeed been walking the @@ -836,11 +836,12 @@ static void trie_init(struct trie *t) #endif } -/* readside most use rcu_read_lock currently dump routines +/* readside must use rcu_read_lock currently dump routines via get_fa_head and dump */ -static struct leaf_info *find_leaf_info(struct hlist_head *head, int plen) +static struct leaf_info *find_leaf_info(struct leaf *l, int plen) { + struct hlist_head *head = &l->list; struct hlist_node *node; struct leaf_info *li; @@ -853,7 +854,7 @@ static struct leaf_info *find_leaf_info(struct hlist_head *head, int plen) static inline struct list_head * get_fa_head(struct leaf *l, int plen) { - struct leaf_info *li = find_leaf_info(&l->list, plen); + struct leaf_info *li = find_leaf_info(l, plen); if (!li) return NULL; @@ -1248,7 +1249,7 @@ err: } -/* should be clalled with rcu_read_lock */ +/* should be called with rcu_read_lock */ static inline int check_leaf(struct trie *t, struct leaf *l, t_key key, int *plen, const struct flowi *flp, struct fib_result *res) @@ -1590,7 +1591,7 @@ fn_trie_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta, rtmsg_fib(RTM_DELROUTE, htonl(key), fa, plen, tb->tb_id, nlhdr, req); l = fib_find_node(t, key); - li = find_leaf_info(&l->list, plen); + li = find_leaf_info(l, plen); list_del_rcu(&fa->fa_list); @@ -1714,7 +1715,6 @@ static int fn_trie_flush(struct fib_table *tb) t->revision++; - rcu_read_lock(); for (h = 0; (l = nextleaf(t, l)) != NULL; h++) { found += trie_flush_leaf(t, l); @@ -1722,7 +1722,6 @@ static int fn_trie_flush(struct fib_table *tb) trie_leaf_remove(t, ll->key); ll = l; } - rcu_read_unlock(); if (ll && hlist_empty(&ll->list)) trie_leaf_remove(t, ll->key); @@ -2288,7 +2287,7 @@ static int fib_trie_seq_show(struct seq_file *seq, void *v) seq_indent(seq, iter->depth); seq_printf(seq, " |-- %d.%d.%d.%d\n", NIPQUAD(val)); for (i = 32; i >= 0; i--) { - struct leaf_info *li = find_leaf_info(&l->list, i); + struct leaf_info *li = find_leaf_info(l, i); if (li) { struct fib_alias *fa; list_for_each_entry_rcu(fa, &li->falh, fa_list) { @@ -2384,7 +2383,7 @@ static int fib_route_seq_show(struct seq_file *seq, void *v) return 0; for (i=32; i>=0; i--) { - struct leaf_info *li = find_leaf_info(&l->list, i); + struct leaf_info *li = find_leaf_info(l, i); struct fib_alias *fa; u32 mask, prefix; From 926b50f92a30090da2c1a8675de954c2d9b09732 Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Mon, 19 Sep 2005 15:33:08 -0700 Subject: [PATCH 178/240] [NETFILTER]: Add new PPTP conntrack and NAT helper This new "version 3" PPTP conntrack/nat helper is finally ready for mainline inclusion. Special thanks to lots of last-minute bugfixing by Patric McHardy. Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ip_conntrack.h | 12 + .../linux/netfilter_ipv4/ip_conntrack_pptp.h | 332 ++++++++ .../netfilter_ipv4/ip_conntrack_proto_gre.h | 114 +++ .../linux/netfilter_ipv4/ip_conntrack_tuple.h | 7 + include/linux/netfilter_ipv4/ip_nat_pptp.h | 11 + net/ipv4/netfilter/Kconfig | 22 + net/ipv4/netfilter/Makefile | 5 + net/ipv4/netfilter/ip_conntrack_helper_pptp.c | 805 ++++++++++++++++++ net/ipv4/netfilter/ip_conntrack_proto_gre.c | 327 +++++++ net/ipv4/netfilter/ip_nat_helper_pptp.c | 401 +++++++++ net/ipv4/netfilter/ip_nat_proto_gre.c | 214 +++++ 11 files changed, 2250 insertions(+) create mode 100644 include/linux/netfilter_ipv4/ip_conntrack_pptp.h create mode 100644 include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h create mode 100644 include/linux/netfilter_ipv4/ip_nat_pptp.h create mode 100644 net/ipv4/netfilter/ip_conntrack_helper_pptp.c create mode 100644 net/ipv4/netfilter/ip_conntrack_proto_gre.c create mode 100644 net/ipv4/netfilter/ip_nat_helper_pptp.c create mode 100644 net/ipv4/netfilter/ip_nat_proto_gre.c diff --git a/include/linux/netfilter_ipv4/ip_conntrack.h b/include/linux/netfilter_ipv4/ip_conntrack.h index 7e033e9271a8..2df446c952ef 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack.h +++ b/include/linux/netfilter_ipv4/ip_conntrack.h @@ -133,11 +133,13 @@ enum ip_conntrack_expect_events { #include #include +#include #include /* per conntrack: protocol private data */ union ip_conntrack_proto { /* insert conntrack proto private data here */ + struct ip_ct_gre gre; struct ip_ct_sctp sctp; struct ip_ct_tcp tcp; struct ip_ct_icmp icmp; @@ -148,6 +150,7 @@ union ip_conntrack_expect_proto { }; /* Add protocol helper include file here */ +#include #include #include #include @@ -155,12 +158,20 @@ union ip_conntrack_expect_proto { /* per conntrack: application helper private data */ union ip_conntrack_help { /* insert conntrack helper private data (master) here */ + struct ip_ct_pptp_master ct_pptp_info; struct ip_ct_ftp_master ct_ftp_info; struct ip_ct_irc_master ct_irc_info; }; #ifdef CONFIG_IP_NF_NAT_NEEDED #include +#include + +/* per conntrack: nat application helper private data */ +union ip_conntrack_nat_help { + /* insert nat helper private data here */ + struct ip_nat_pptp nat_pptp_info; +}; #endif #include @@ -223,6 +234,7 @@ struct ip_conntrack #ifdef CONFIG_IP_NF_NAT_NEEDED struct { struct ip_nat_info info; + union ip_conntrack_nat_help help; #if defined(CONFIG_IP_NF_TARGET_MASQUERADE) || \ defined(CONFIG_IP_NF_TARGET_MASQUERADE_MODULE) int masq_index; diff --git a/include/linux/netfilter_ipv4/ip_conntrack_pptp.h b/include/linux/netfilter_ipv4/ip_conntrack_pptp.h new file mode 100644 index 000000000000..389e3851d52f --- /dev/null +++ b/include/linux/netfilter_ipv4/ip_conntrack_pptp.h @@ -0,0 +1,332 @@ +/* PPTP constants and structs */ +#ifndef _CONNTRACK_PPTP_H +#define _CONNTRACK_PPTP_H + +/* state of the control session */ +enum pptp_ctrlsess_state { + PPTP_SESSION_NONE, /* no session present */ + PPTP_SESSION_ERROR, /* some session error */ + PPTP_SESSION_STOPREQ, /* stop_sess request seen */ + PPTP_SESSION_REQUESTED, /* start_sess request seen */ + PPTP_SESSION_CONFIRMED, /* session established */ +}; + +/* state of the call inside the control session */ +enum pptp_ctrlcall_state { + PPTP_CALL_NONE, + PPTP_CALL_ERROR, + PPTP_CALL_OUT_REQ, + PPTP_CALL_OUT_CONF, + PPTP_CALL_IN_REQ, + PPTP_CALL_IN_REP, + PPTP_CALL_IN_CONF, + PPTP_CALL_CLEAR_REQ, +}; + + +/* conntrack private data */ +struct ip_ct_pptp_master { + enum pptp_ctrlsess_state sstate; /* session state */ + + /* everything below is going to be per-expectation in newnat, + * since there could be more than one call within one session */ + enum pptp_ctrlcall_state cstate; /* call state */ + u_int16_t pac_call_id; /* call id of PAC, host byte order */ + u_int16_t pns_call_id; /* call id of PNS, host byte order */ + + /* in pre-2.6.11 this used to be per-expect. Now it is per-conntrack + * and therefore imposes a fixed limit on the number of maps */ + struct ip_ct_gre_keymap *keymap_orig, *keymap_reply; +}; + +/* conntrack_expect private member */ +struct ip_ct_pptp_expect { + enum pptp_ctrlcall_state cstate; /* call state */ + u_int16_t pac_call_id; /* call id of PAC */ + u_int16_t pns_call_id; /* call id of PNS */ +}; + + +#ifdef __KERNEL__ + +#define IP_CONNTR_PPTP PPTP_CONTROL_PORT + +#define PPTP_CONTROL_PORT 1723 + +#define PPTP_PACKET_CONTROL 1 +#define PPTP_PACKET_MGMT 2 + +#define PPTP_MAGIC_COOKIE 0x1a2b3c4d + +struct pptp_pkt_hdr { + __u16 packetLength; + __u16 packetType; + __u32 magicCookie; +}; + +/* PptpControlMessageType values */ +#define PPTP_START_SESSION_REQUEST 1 +#define PPTP_START_SESSION_REPLY 2 +#define PPTP_STOP_SESSION_REQUEST 3 +#define PPTP_STOP_SESSION_REPLY 4 +#define PPTP_ECHO_REQUEST 5 +#define PPTP_ECHO_REPLY 6 +#define PPTP_OUT_CALL_REQUEST 7 +#define PPTP_OUT_CALL_REPLY 8 +#define PPTP_IN_CALL_REQUEST 9 +#define PPTP_IN_CALL_REPLY 10 +#define PPTP_IN_CALL_CONNECT 11 +#define PPTP_CALL_CLEAR_REQUEST 12 +#define PPTP_CALL_DISCONNECT_NOTIFY 13 +#define PPTP_WAN_ERROR_NOTIFY 14 +#define PPTP_SET_LINK_INFO 15 + +#define PPTP_MSG_MAX 15 + +/* PptpGeneralError values */ +#define PPTP_ERROR_CODE_NONE 0 +#define PPTP_NOT_CONNECTED 1 +#define PPTP_BAD_FORMAT 2 +#define PPTP_BAD_VALUE 3 +#define PPTP_NO_RESOURCE 4 +#define PPTP_BAD_CALLID 5 +#define PPTP_REMOVE_DEVICE_ERROR 6 + +struct PptpControlHeader { + __u16 messageType; + __u16 reserved; +}; + +/* FramingCapability Bitmap Values */ +#define PPTP_FRAME_CAP_ASYNC 0x1 +#define PPTP_FRAME_CAP_SYNC 0x2 + +/* BearerCapability Bitmap Values */ +#define PPTP_BEARER_CAP_ANALOG 0x1 +#define PPTP_BEARER_CAP_DIGITAL 0x2 + +struct PptpStartSessionRequest { + __u16 protocolVersion; + __u8 reserved1; + __u8 reserved2; + __u32 framingCapability; + __u32 bearerCapability; + __u16 maxChannels; + __u16 firmwareRevision; + __u8 hostName[64]; + __u8 vendorString[64]; +}; + +/* PptpStartSessionResultCode Values */ +#define PPTP_START_OK 1 +#define PPTP_START_GENERAL_ERROR 2 +#define PPTP_START_ALREADY_CONNECTED 3 +#define PPTP_START_NOT_AUTHORIZED 4 +#define PPTP_START_UNKNOWN_PROTOCOL 5 + +struct PptpStartSessionReply { + __u16 protocolVersion; + __u8 resultCode; + __u8 generalErrorCode; + __u32 framingCapability; + __u32 bearerCapability; + __u16 maxChannels; + __u16 firmwareRevision; + __u8 hostName[64]; + __u8 vendorString[64]; +}; + +/* PptpStopReasons */ +#define PPTP_STOP_NONE 1 +#define PPTP_STOP_PROTOCOL 2 +#define PPTP_STOP_LOCAL_SHUTDOWN 3 + +struct PptpStopSessionRequest { + __u8 reason; +}; + +/* PptpStopSessionResultCode */ +#define PPTP_STOP_OK 1 +#define PPTP_STOP_GENERAL_ERROR 2 + +struct PptpStopSessionReply { + __u8 resultCode; + __u8 generalErrorCode; +}; + +struct PptpEchoRequest { + __u32 identNumber; +}; + +/* PptpEchoReplyResultCode */ +#define PPTP_ECHO_OK 1 +#define PPTP_ECHO_GENERAL_ERROR 2 + +struct PptpEchoReply { + __u32 identNumber; + __u8 resultCode; + __u8 generalErrorCode; + __u16 reserved; +}; + +/* PptpFramingType */ +#define PPTP_ASYNC_FRAMING 1 +#define PPTP_SYNC_FRAMING 2 +#define PPTP_DONT_CARE_FRAMING 3 + +/* PptpCallBearerType */ +#define PPTP_ANALOG_TYPE 1 +#define PPTP_DIGITAL_TYPE 2 +#define PPTP_DONT_CARE_BEARER_TYPE 3 + +struct PptpOutCallRequest { + __u16 callID; + __u16 callSerialNumber; + __u32 minBPS; + __u32 maxBPS; + __u32 bearerType; + __u32 framingType; + __u16 packetWindow; + __u16 packetProcDelay; + __u16 reserved1; + __u16 phoneNumberLength; + __u16 reserved2; + __u8 phoneNumber[64]; + __u8 subAddress[64]; +}; + +/* PptpCallResultCode */ +#define PPTP_OUTCALL_CONNECT 1 +#define PPTP_OUTCALL_GENERAL_ERROR 2 +#define PPTP_OUTCALL_NO_CARRIER 3 +#define PPTP_OUTCALL_BUSY 4 +#define PPTP_OUTCALL_NO_DIAL_TONE 5 +#define PPTP_OUTCALL_TIMEOUT 6 +#define PPTP_OUTCALL_DONT_ACCEPT 7 + +struct PptpOutCallReply { + __u16 callID; + __u16 peersCallID; + __u8 resultCode; + __u8 generalErrorCode; + __u16 causeCode; + __u32 connectSpeed; + __u16 packetWindow; + __u16 packetProcDelay; + __u32 physChannelID; +}; + +struct PptpInCallRequest { + __u16 callID; + __u16 callSerialNumber; + __u32 callBearerType; + __u32 physChannelID; + __u16 dialedNumberLength; + __u16 dialingNumberLength; + __u8 dialedNumber[64]; + __u8 dialingNumber[64]; + __u8 subAddress[64]; +}; + +/* PptpInCallResultCode */ +#define PPTP_INCALL_ACCEPT 1 +#define PPTP_INCALL_GENERAL_ERROR 2 +#define PPTP_INCALL_DONT_ACCEPT 3 + +struct PptpInCallReply { + __u16 callID; + __u16 peersCallID; + __u8 resultCode; + __u8 generalErrorCode; + __u16 packetWindow; + __u16 packetProcDelay; + __u16 reserved; +}; + +struct PptpInCallConnected { + __u16 peersCallID; + __u16 reserved; + __u32 connectSpeed; + __u16 packetWindow; + __u16 packetProcDelay; + __u32 callFramingType; +}; + +struct PptpClearCallRequest { + __u16 callID; + __u16 reserved; +}; + +struct PptpCallDisconnectNotify { + __u16 callID; + __u8 resultCode; + __u8 generalErrorCode; + __u16 causeCode; + __u16 reserved; + __u8 callStatistics[128]; +}; + +struct PptpWanErrorNotify { + __u16 peersCallID; + __u16 reserved; + __u32 crcErrors; + __u32 framingErrors; + __u32 hardwareOverRuns; + __u32 bufferOverRuns; + __u32 timeoutErrors; + __u32 alignmentErrors; +}; + +struct PptpSetLinkInfo { + __u16 peersCallID; + __u16 reserved; + __u32 sendAccm; + __u32 recvAccm; +}; + + +struct pptp_priv_data { + __u16 call_id; + __u16 mcall_id; + __u16 pcall_id; +}; + +union pptp_ctrl_union { + struct PptpStartSessionRequest sreq; + struct PptpStartSessionReply srep; + struct PptpStopSessionRequest streq; + struct PptpStopSessionReply strep; + struct PptpOutCallRequest ocreq; + struct PptpOutCallReply ocack; + struct PptpInCallRequest icreq; + struct PptpInCallReply icack; + struct PptpInCallConnected iccon; + struct PptpClearCallRequest clrreq; + struct PptpCallDisconnectNotify disc; + struct PptpWanErrorNotify wanerr; + struct PptpSetLinkInfo setlink; +}; + +extern int +(*ip_nat_pptp_hook_outbound)(struct sk_buff **pskb, + struct ip_conntrack *ct, + enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq); + +extern int +(*ip_nat_pptp_hook_inbound)(struct sk_buff **pskb, + struct ip_conntrack *ct, + enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq); + +extern int +(*ip_nat_pptp_hook_exp_gre)(struct ip_conntrack_expect *exp_orig, + struct ip_conntrack_expect *exp_reply); + +extern void +(*ip_nat_pptp_hook_expectfn)(struct ip_conntrack *ct, + struct ip_conntrack_expect *exp); +#endif /* __KERNEL__ */ +#endif /* _CONNTRACK_PPTP_H */ diff --git a/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h b/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h new file mode 100644 index 000000000000..8d090ef82f5f --- /dev/null +++ b/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h @@ -0,0 +1,114 @@ +#ifndef _CONNTRACK_PROTO_GRE_H +#define _CONNTRACK_PROTO_GRE_H +#include + +/* GRE PROTOCOL HEADER */ + +/* GRE Version field */ +#define GRE_VERSION_1701 0x0 +#define GRE_VERSION_PPTP 0x1 + +/* GRE Protocol field */ +#define GRE_PROTOCOL_PPTP 0x880B + +/* GRE Flags */ +#define GRE_FLAG_C 0x80 +#define GRE_FLAG_R 0x40 +#define GRE_FLAG_K 0x20 +#define GRE_FLAG_S 0x10 +#define GRE_FLAG_A 0x80 + +#define GRE_IS_C(f) ((f)&GRE_FLAG_C) +#define GRE_IS_R(f) ((f)&GRE_FLAG_R) +#define GRE_IS_K(f) ((f)&GRE_FLAG_K) +#define GRE_IS_S(f) ((f)&GRE_FLAG_S) +#define GRE_IS_A(f) ((f)&GRE_FLAG_A) + +/* GRE is a mess: Four different standards */ +struct gre_hdr { +#if defined(__LITTLE_ENDIAN_BITFIELD) + __u16 rec:3, + srr:1, + seq:1, + key:1, + routing:1, + csum:1, + version:3, + reserved:4, + ack:1; +#elif defined(__BIG_ENDIAN_BITFIELD) + __u16 csum:1, + routing:1, + key:1, + seq:1, + srr:1, + rec:3, + ack:1, + reserved:4, + version:3; +#else +#error "Adjust your defines" +#endif + __u16 protocol; +}; + +/* modified GRE header for PPTP */ +struct gre_hdr_pptp { + __u8 flags; /* bitfield */ + __u8 version; /* should be GRE_VERSION_PPTP */ + __u16 protocol; /* should be GRE_PROTOCOL_PPTP */ + __u16 payload_len; /* size of ppp payload, not inc. gre header */ + __u16 call_id; /* peer's call_id for this session */ + __u32 seq; /* sequence number. Present if S==1 */ + __u32 ack; /* seq number of highest packet recieved by */ + /* sender in this session */ +}; + + +/* this is part of ip_conntrack */ +struct ip_ct_gre { + unsigned int stream_timeout; + unsigned int timeout; +}; + +#ifdef __KERNEL__ +struct ip_conntrack_expect; +struct ip_conntrack; + +/* structure for original <-> reply keymap */ +struct ip_ct_gre_keymap { + struct list_head list; + + struct ip_conntrack_tuple tuple; +}; + +/* add new tuple->key_reply pair to keymap */ +int ip_ct_gre_keymap_add(struct ip_conntrack *ct, + struct ip_conntrack_tuple *t, + int reply); + +/* delete keymap entries */ +void ip_ct_gre_keymap_destroy(struct ip_conntrack *ct); + + +/* get pointer to gre key, if present */ +static inline u_int32_t *gre_key(struct gre_hdr *greh) +{ + if (!greh->key) + return NULL; + if (greh->csum || greh->routing) + return (u_int32_t *) (greh+sizeof(*greh)+4); + return (u_int32_t *) (greh+sizeof(*greh)); +} + +/* get pointer ot gre csum, if present */ +static inline u_int16_t *gre_csum(struct gre_hdr *greh) +{ + if (!greh->csum) + return NULL; + return (u_int16_t *) (greh+sizeof(*greh)); +} + +#endif /* __KERNEL__ */ + +#endif /* _CONNTRACK_PROTO_GRE_H */ diff --git a/include/linux/netfilter_ipv4/ip_conntrack_tuple.h b/include/linux/netfilter_ipv4/ip_conntrack_tuple.h index c33f0b5e0d0a..14dc0f7b6556 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_tuple.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_tuple.h @@ -28,6 +28,9 @@ union ip_conntrack_manip_proto struct { u_int16_t port; } sctp; + struct { + u_int16_t key; /* key is 32bit, pptp only uses 16 */ + } gre; }; /* The manipulable part of the tuple. */ @@ -61,6 +64,10 @@ struct ip_conntrack_tuple struct { u_int16_t port; } sctp; + struct { + u_int16_t key; /* key is 32bit, + * pptp only uses 16 */ + } gre; } u; /* The protocol. */ diff --git a/include/linux/netfilter_ipv4/ip_nat_pptp.h b/include/linux/netfilter_ipv4/ip_nat_pptp.h new file mode 100644 index 000000000000..eaf66c2e8f93 --- /dev/null +++ b/include/linux/netfilter_ipv4/ip_nat_pptp.h @@ -0,0 +1,11 @@ +/* PPTP constants and structs */ +#ifndef _NAT_PPTP_H +#define _NAT_PPTP_H + +/* conntrack private data */ +struct ip_nat_pptp { + u_int16_t pns_call_id; /* NAT'ed PNS call id */ + u_int16_t pac_call_id; /* NAT'ed PAC call id */ +}; + +#endif /* _NAT_PPTP_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index e2162d270073..3cf9b451675c 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -137,6 +137,22 @@ config IP_NF_AMANDA To compile it as a module, choose M here. If unsure, say Y. +config IP_NF_PPTP + tristate 'PPTP protocol support' + help + This module adds support for PPTP (Point to Point Tunnelling + Protocol, RFC2637) conncection tracking and NAT. + + If you are running PPTP sessions over a stateful firewall or NAT + box, you may want to enable this feature. + + Please note that not all PPTP modes of operation are supported yet. + For more info, read top of the file + net/ipv4/netfilter/ip_conntrack_pptp.c + + If you want to compile it as a module, say M here and read + Documentation/modules.txt. If unsure, say `N'. + config IP_NF_QUEUE tristate "IP Userspace queueing via NETLINK (OBSOLETE)" help @@ -621,6 +637,12 @@ config IP_NF_NAT_AMANDA default IP_NF_NAT if IP_NF_AMANDA=y default m if IP_NF_AMANDA=m +config IP_NF_NAT_PPTP + tristate + depends on IP_NF_NAT!=n && IP_NF_PPTP!=n + default IP_NF_NAT if IP_NF_PPTP=y + default m if IP_NF_PPTP=m + # mangle + specific targets config IP_NF_MANGLE tristate "Packet mangling" diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 1ba0db746817..3d45d3c0283c 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -6,6 +6,9 @@ ip_conntrack-objs := ip_conntrack_standalone.o ip_conntrack_core.o ip_conntrack_proto_generic.o ip_conntrack_proto_tcp.o ip_conntrack_proto_udp.o ip_conntrack_proto_icmp.o iptable_nat-objs := ip_nat_standalone.o ip_nat_rule.o ip_nat_core.o ip_nat_helper.o ip_nat_proto_unknown.o ip_nat_proto_tcp.o ip_nat_proto_udp.o ip_nat_proto_icmp.o +ip_conntrack_pptp-objs := ip_conntrack_helper_pptp.o ip_conntrack_proto_gre.o +ip_nat_pptp-objs := ip_nat_helper_pptp.o ip_nat_proto_gre.o + # connection tracking obj-$(CONFIG_IP_NF_CONNTRACK) += ip_conntrack.o @@ -17,6 +20,7 @@ obj-$(CONFIG_IP_NF_CONNTRACK_NETLINK) += ip_conntrack_netlink.o obj-$(CONFIG_IP_NF_CT_PROTO_SCTP) += ip_conntrack_proto_sctp.o # connection tracking helpers +obj-$(CONFIG_IP_NF_PPTP) += ip_conntrack_pptp.o obj-$(CONFIG_IP_NF_AMANDA) += ip_conntrack_amanda.o obj-$(CONFIG_IP_NF_TFTP) += ip_conntrack_tftp.o obj-$(CONFIG_IP_NF_FTP) += ip_conntrack_ftp.o @@ -24,6 +28,7 @@ obj-$(CONFIG_IP_NF_IRC) += ip_conntrack_irc.o obj-$(CONFIG_IP_NF_NETBIOS_NS) += ip_conntrack_netbios_ns.o # NAT helpers +obj-$(CONFIG_IP_NF_NAT_PPTP) += ip_nat_pptp.o obj-$(CONFIG_IP_NF_NAT_AMANDA) += ip_nat_amanda.o obj-$(CONFIG_IP_NF_NAT_TFTP) += ip_nat_tftp.o obj-$(CONFIG_IP_NF_NAT_FTP) += ip_nat_ftp.o diff --git a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c new file mode 100644 index 000000000000..79db5b70d5f6 --- /dev/null +++ b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c @@ -0,0 +1,805 @@ +/* + * ip_conntrack_pptp.c - Version 3.0 + * + * Connection tracking support for PPTP (Point to Point Tunneling Protocol). + * PPTP is a a protocol for creating virtual private networks. + * It is a specification defined by Microsoft and some vendors + * working with Microsoft. PPTP is built on top of a modified + * version of the Internet Generic Routing Encapsulation Protocol. + * GRE is defined in RFC 1701 and RFC 1702. Documentation of + * PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + * Limitations: + * - We blindly assume that control connections are always + * established in PNS->PAC direction. This is a violation + * of RFFC2673 + * - We can only support one single call within each session + * + * TODO: + * - testing of incoming PPTP calls + * + * Changes: + * 2002-02-05 - Version 1.3 + * - Call ip_conntrack_unexpect_related() from + * pptp_destroy_siblings() to destroy expectations in case + * CALL_DISCONNECT_NOTIFY or tcp fin packet was seen + * (Philip Craig ) + * - Add Version information at module loadtime + * 2002-02-10 - Version 1.6 + * - move to C99 style initializers + * - remove second expectation if first arrives + * 2004-10-22 - Version 2.0 + * - merge Mandrake's 2.6.x port with recent 2.6.x API changes + * - fix lots of linear skb assumptions from Mandrake's port + * 2005-06-10 - Version 2.1 + * - use ip_conntrack_expect_free() instead of kfree() on the + * expect's (which are from the slab for quite some time) + * 2005-06-10 - Version 3.0 + * - port helper to post-2.6.11 API changes, + * funded by Oxcoda NetBox Blue (http://www.netboxblue.com/) + * 2005-07-30 - Version 3.1 + * - port helper to 2.6.13 API changes + * + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#define IP_CT_PPTP_VERSION "3.1" + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("Netfilter connection tracking helper module for PPTP"); + +static DEFINE_SPINLOCK(ip_pptp_lock); + +int +(*ip_nat_pptp_hook_outbound)(struct sk_buff **pskb, + struct ip_conntrack *ct, + enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq); + +int +(*ip_nat_pptp_hook_inbound)(struct sk_buff **pskb, + struct ip_conntrack *ct, + enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq); + +int +(*ip_nat_pptp_hook_exp_gre)(struct ip_conntrack_expect *expect_orig, + struct ip_conntrack_expect *expect_reply); + +void +(*ip_nat_pptp_hook_expectfn)(struct ip_conntrack *ct, + struct ip_conntrack_expect *exp); + +#if 0 +/* PptpControlMessageType names */ +const char *pptp_msg_name[] = { + "UNKNOWN_MESSAGE", + "START_SESSION_REQUEST", + "START_SESSION_REPLY", + "STOP_SESSION_REQUEST", + "STOP_SESSION_REPLY", + "ECHO_REQUEST", + "ECHO_REPLY", + "OUT_CALL_REQUEST", + "OUT_CALL_REPLY", + "IN_CALL_REQUEST", + "IN_CALL_REPLY", + "IN_CALL_CONNECT", + "CALL_CLEAR_REQUEST", + "CALL_DISCONNECT_NOTIFY", + "WAN_ERROR_NOTIFY", + "SET_LINK_INFO" +}; +EXPORT_SYMBOL(pptp_msg_name); +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, __FUNCTION__, ## args) +#else +#define DEBUGP(format, args...) +#endif + +#define SECS *HZ +#define MINS * 60 SECS +#define HOURS * 60 MINS + +#define PPTP_GRE_TIMEOUT (10 MINS) +#define PPTP_GRE_STREAM_TIMEOUT (5 HOURS) + +static void pptp_expectfn(struct ip_conntrack *ct, + struct ip_conntrack_expect *exp) +{ + DEBUGP("increasing timeouts\n"); + + /* increase timeout of GRE data channel conntrack entry */ + ct->proto.gre.timeout = PPTP_GRE_TIMEOUT; + ct->proto.gre.stream_timeout = PPTP_GRE_STREAM_TIMEOUT; + + /* Can you see how rusty this code is, compared with the pre-2.6.11 + * one? That's what happened to my shiny newnat of 2002 ;( -HW */ + + if (!ip_nat_pptp_hook_expectfn) { + struct ip_conntrack_tuple inv_t; + struct ip_conntrack_expect *exp_other; + + /* obviously this tuple inversion only works until you do NAT */ + invert_tuplepr(&inv_t, &exp->tuple); + DEBUGP("trying to unexpect other dir: "); + DUMP_TUPLE(&inv_t); + + exp_other = ip_conntrack_expect_find(&inv_t); + if (exp_other) { + /* delete other expectation. */ + DEBUGP("found\n"); + ip_conntrack_unexpect_related(exp_other); + ip_conntrack_expect_put(exp_other); + } else { + DEBUGP("not found\n"); + } + } else { + /* we need more than simple inversion */ + ip_nat_pptp_hook_expectfn(ct, exp); + } +} + +static int destroy_sibling_or_exp(const struct ip_conntrack_tuple *t) +{ + struct ip_conntrack_tuple_hash *h; + struct ip_conntrack_expect *exp; + + DEBUGP("trying to timeout ct or exp for tuple "); + DUMP_TUPLE(t); + + h = ip_conntrack_find_get(t, NULL); + if (h) { + struct ip_conntrack *sibling = tuplehash_to_ctrack(h); + DEBUGP("setting timeout of conntrack %p to 0\n", sibling); + sibling->proto.gre.timeout = 0; + sibling->proto.gre.stream_timeout = 0; + /* refresh_acct will not modify counters if skb == NULL */ + if (del_timer(&sibling->timeout)) + sibling->timeout.function((unsigned long)sibling); + ip_conntrack_put(sibling); + return 1; + } else { + exp = ip_conntrack_expect_find(t); + if (exp) { + DEBUGP("unexpect_related of expect %p\n", exp); + ip_conntrack_unexpect_related(exp); + ip_conntrack_expect_put(exp); + return 1; + } + } + + return 0; +} + + +/* timeout GRE data connections */ +static void pptp_destroy_siblings(struct ip_conntrack *ct) +{ + struct ip_conntrack_tuple t; + + /* Since ct->sibling_list has literally rusted away in 2.6.11, + * we now need another way to find out about our sibling + * contrack and expects... -HW */ + + /* try original (pns->pac) tuple */ + memcpy(&t, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, sizeof(t)); + t.dst.protonum = IPPROTO_GRE; + t.src.u.gre.key = htons(ct->help.ct_pptp_info.pns_call_id); + t.dst.u.gre.key = htons(ct->help.ct_pptp_info.pac_call_id); + + if (!destroy_sibling_or_exp(&t)) + DEBUGP("failed to timeout original pns->pac ct/exp\n"); + + /* try reply (pac->pns) tuple */ + memcpy(&t, &ct->tuplehash[IP_CT_DIR_REPLY].tuple, sizeof(t)); + t.dst.protonum = IPPROTO_GRE; + t.src.u.gre.key = htons(ct->help.ct_pptp_info.pac_call_id); + t.dst.u.gre.key = htons(ct->help.ct_pptp_info.pns_call_id); + + if (!destroy_sibling_or_exp(&t)) + DEBUGP("failed to timeout reply pac->pns ct/exp\n"); +} + +/* expect GRE connections (PNS->PAC and PAC->PNS direction) */ +static inline int +exp_gre(struct ip_conntrack *master, + u_int32_t seq, + u_int16_t callid, + u_int16_t peer_callid) +{ + struct ip_conntrack_tuple inv_tuple; + struct ip_conntrack_tuple exp_tuples[] = { + /* tuple in original direction, PNS->PAC */ + { .src = { .ip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip, + .u = { .gre = { .key = peer_callid } } + }, + .dst = { .ip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip, + .u = { .gre = { .key = callid } }, + .protonum = IPPROTO_GRE + }, + }, + /* tuple in reply direction, PAC->PNS */ + { .src = { .ip = master->tuplehash[IP_CT_DIR_REPLY].tuple.src.ip, + .u = { .gre = { .key = callid } } + }, + .dst = { .ip = master->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip, + .u = { .gre = { .key = peer_callid } }, + .protonum = IPPROTO_GRE + }, + } + }; + struct ip_conntrack_expect *exp_orig, *exp_reply; + int ret = 1; + + exp_orig = ip_conntrack_expect_alloc(master); + if (exp_orig == NULL) + goto out; + + exp_reply = ip_conntrack_expect_alloc(master); + if (exp_reply == NULL) + goto out_put_orig; + + memcpy(&exp_orig->tuple, &exp_tuples[0], sizeof(exp_orig->tuple)); + + exp_orig->mask.src.ip = 0xffffffff; + exp_orig->mask.src.u.all = 0; + exp_orig->mask.dst.u.all = 0; + exp_orig->mask.dst.u.gre.key = 0xffff; + exp_orig->mask.dst.ip = 0xffffffff; + exp_orig->mask.dst.protonum = 0xff; + + exp_orig->master = master; + exp_orig->expectfn = pptp_expectfn; + exp_orig->flags = 0; + + exp_orig->dir = IP_CT_DIR_ORIGINAL; + + /* both expectations are identical apart from tuple */ + memcpy(exp_reply, exp_orig, sizeof(*exp_reply)); + memcpy(&exp_reply->tuple, &exp_tuples[1], sizeof(exp_reply->tuple)); + + exp_reply->dir = !exp_orig->dir; + + if (ip_nat_pptp_hook_exp_gre) + ret = ip_nat_pptp_hook_exp_gre(exp_orig, exp_reply); + else { + + DEBUGP("calling expect_related PNS->PAC"); + DUMP_TUPLE(&exp_orig->tuple); + + if (ip_conntrack_expect_related(exp_orig) != 0) { + DEBUGP("cannot expect_related()\n"); + goto out_put_both; + } + + DEBUGP("calling expect_related PAC->PNS"); + DUMP_TUPLE(&exp_reply->tuple); + + if (ip_conntrack_expect_related(exp_reply) != 0) { + DEBUGP("cannot expect_related()\n"); + goto out_unexpect_orig; + } + + /* Add GRE keymap entries */ + if (ip_ct_gre_keymap_add(master, &exp_reply->tuple, 0) != 0) { + DEBUGP("cannot keymap_add() exp\n"); + goto out_unexpect_both; + } + + invert_tuplepr(&inv_tuple, &exp_reply->tuple); + if (ip_ct_gre_keymap_add(master, &inv_tuple, 1) != 0) { + ip_ct_gre_keymap_destroy(master); + DEBUGP("cannot keymap_add() exp_inv\n"); + goto out_unexpect_both; + } + ret = 0; + } + +out_put_both: + ip_conntrack_expect_put(exp_reply); +out_put_orig: + ip_conntrack_expect_put(exp_orig); +out: + return ret; + +out_unexpect_both: + ip_conntrack_unexpect_related(exp_reply); +out_unexpect_orig: + ip_conntrack_unexpect_related(exp_orig); + goto out_put_both; +} + +static inline int +pptp_inbound_pkt(struct sk_buff **pskb, + struct tcphdr *tcph, + unsigned int nexthdr_off, + unsigned int datalen, + struct ip_conntrack *ct, + enum ip_conntrack_info ctinfo) +{ + struct PptpControlHeader _ctlh, *ctlh; + unsigned int reqlen; + union pptp_ctrl_union _pptpReq, *pptpReq; + struct ip_ct_pptp_master *info = &ct->help.ct_pptp_info; + u_int16_t msg, *cid, *pcid; + u_int32_t seq; + + ctlh = skb_header_pointer(*pskb, nexthdr_off, sizeof(_ctlh), &_ctlh); + if (!ctlh) { + DEBUGP("error during skb_header_pointer\n"); + return NF_ACCEPT; + } + nexthdr_off += sizeof(_ctlh); + datalen -= sizeof(_ctlh); + + reqlen = datalen; + if (reqlen > sizeof(*pptpReq)) + reqlen = sizeof(*pptpReq); + pptpReq = skb_header_pointer(*pskb, nexthdr_off, reqlen, &_pptpReq); + if (!pptpReq) { + DEBUGP("error during skb_header_pointer\n"); + return NF_ACCEPT; + } + + msg = ntohs(ctlh->messageType); + DEBUGP("inbound control message %s\n", pptp_msg_name[msg]); + + switch (msg) { + case PPTP_START_SESSION_REPLY: + if (reqlen < sizeof(_pptpReq.srep)) { + DEBUGP("%s: short packet\n", pptp_msg_name[msg]); + break; + } + + /* server confirms new control session */ + if (info->sstate < PPTP_SESSION_REQUESTED) { + DEBUGP("%s without START_SESS_REQUEST\n", + pptp_msg_name[msg]); + break; + } + if (pptpReq->srep.resultCode == PPTP_START_OK) + info->sstate = PPTP_SESSION_CONFIRMED; + else + info->sstate = PPTP_SESSION_ERROR; + break; + + case PPTP_STOP_SESSION_REPLY: + if (reqlen < sizeof(_pptpReq.strep)) { + DEBUGP("%s: short packet\n", pptp_msg_name[msg]); + break; + } + + /* server confirms end of control session */ + if (info->sstate > PPTP_SESSION_STOPREQ) { + DEBUGP("%s without STOP_SESS_REQUEST\n", + pptp_msg_name[msg]); + break; + } + if (pptpReq->strep.resultCode == PPTP_STOP_OK) + info->sstate = PPTP_SESSION_NONE; + else + info->sstate = PPTP_SESSION_ERROR; + break; + + case PPTP_OUT_CALL_REPLY: + if (reqlen < sizeof(_pptpReq.ocack)) { + DEBUGP("%s: short packet\n", pptp_msg_name[msg]); + break; + } + + /* server accepted call, we now expect GRE frames */ + if (info->sstate != PPTP_SESSION_CONFIRMED) { + DEBUGP("%s but no session\n", pptp_msg_name[msg]); + break; + } + if (info->cstate != PPTP_CALL_OUT_REQ && + info->cstate != PPTP_CALL_OUT_CONF) { + DEBUGP("%s without OUTCALL_REQ\n", pptp_msg_name[msg]); + break; + } + if (pptpReq->ocack.resultCode != PPTP_OUTCALL_CONNECT) { + info->cstate = PPTP_CALL_NONE; + break; + } + + cid = &pptpReq->ocack.callID; + pcid = &pptpReq->ocack.peersCallID; + + info->pac_call_id = ntohs(*cid); + + if (htons(info->pns_call_id) != *pcid) { + DEBUGP("%s for unknown callid %u\n", + pptp_msg_name[msg], ntohs(*pcid)); + break; + } + + DEBUGP("%s, CID=%X, PCID=%X\n", pptp_msg_name[msg], + ntohs(*cid), ntohs(*pcid)); + + info->cstate = PPTP_CALL_OUT_CONF; + + seq = ntohl(tcph->seq) + sizeof(struct pptp_pkt_hdr) + + sizeof(struct PptpControlHeader) + + ((void *)pcid - (void *)pptpReq); + + if (exp_gre(ct, seq, *cid, *pcid) != 0) + printk("ip_conntrack_pptp: error during exp_gre\n"); + break; + + case PPTP_IN_CALL_REQUEST: + if (reqlen < sizeof(_pptpReq.icack)) { + DEBUGP("%s: short packet\n", pptp_msg_name[msg]); + break; + } + + /* server tells us about incoming call request */ + if (info->sstate != PPTP_SESSION_CONFIRMED) { + DEBUGP("%s but no session\n", pptp_msg_name[msg]); + break; + } + pcid = &pptpReq->icack.peersCallID; + DEBUGP("%s, PCID=%X\n", pptp_msg_name[msg], ntohs(*pcid)); + info->cstate = PPTP_CALL_IN_REQ; + info->pac_call_id = ntohs(*pcid); + break; + + case PPTP_IN_CALL_CONNECT: + if (reqlen < sizeof(_pptpReq.iccon)) { + DEBUGP("%s: short packet\n", pptp_msg_name[msg]); + break; + } + + /* server tells us about incoming call established */ + if (info->sstate != PPTP_SESSION_CONFIRMED) { + DEBUGP("%s but no session\n", pptp_msg_name[msg]); + break; + } + if (info->sstate != PPTP_CALL_IN_REP + && info->sstate != PPTP_CALL_IN_CONF) { + DEBUGP("%s but never sent IN_CALL_REPLY\n", + pptp_msg_name[msg]); + break; + } + + pcid = &pptpReq->iccon.peersCallID; + cid = &info->pac_call_id; + + if (info->pns_call_id != ntohs(*pcid)) { + DEBUGP("%s for unknown CallID %u\n", + pptp_msg_name[msg], ntohs(*cid)); + break; + } + + DEBUGP("%s, PCID=%X\n", pptp_msg_name[msg], ntohs(*pcid)); + info->cstate = PPTP_CALL_IN_CONF; + + /* we expect a GRE connection from PAC to PNS */ + seq = ntohl(tcph->seq) + sizeof(struct pptp_pkt_hdr) + + sizeof(struct PptpControlHeader) + + ((void *)pcid - (void *)pptpReq); + + if (exp_gre(ct, seq, *cid, *pcid) != 0) + printk("ip_conntrack_pptp: error during exp_gre\n"); + + break; + + case PPTP_CALL_DISCONNECT_NOTIFY: + if (reqlen < sizeof(_pptpReq.disc)) { + DEBUGP("%s: short packet\n", pptp_msg_name[msg]); + break; + } + + /* server confirms disconnect */ + cid = &pptpReq->disc.callID; + DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(*cid)); + info->cstate = PPTP_CALL_NONE; + + /* untrack this call id, unexpect GRE packets */ + pptp_destroy_siblings(ct); + break; + + case PPTP_WAN_ERROR_NOTIFY: + break; + + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* I don't have to explain these ;) */ + break; + default: + DEBUGP("invalid %s (TY=%d)\n", (msg <= PPTP_MSG_MAX) + ? pptp_msg_name[msg]:pptp_msg_name[0], msg); + break; + } + + + if (ip_nat_pptp_hook_inbound) + return ip_nat_pptp_hook_inbound(pskb, ct, ctinfo, ctlh, + pptpReq); + + return NF_ACCEPT; + +} + +static inline int +pptp_outbound_pkt(struct sk_buff **pskb, + struct tcphdr *tcph, + unsigned int nexthdr_off, + unsigned int datalen, + struct ip_conntrack *ct, + enum ip_conntrack_info ctinfo) +{ + struct PptpControlHeader _ctlh, *ctlh; + unsigned int reqlen; + union pptp_ctrl_union _pptpReq, *pptpReq; + struct ip_ct_pptp_master *info = &ct->help.ct_pptp_info; + u_int16_t msg, *cid, *pcid; + + ctlh = skb_header_pointer(*pskb, nexthdr_off, sizeof(_ctlh), &_ctlh); + if (!ctlh) + return NF_ACCEPT; + nexthdr_off += sizeof(_ctlh); + datalen -= sizeof(_ctlh); + + reqlen = datalen; + if (reqlen > sizeof(*pptpReq)) + reqlen = sizeof(*pptpReq); + pptpReq = skb_header_pointer(*pskb, nexthdr_off, reqlen, &_pptpReq); + if (!pptpReq) + return NF_ACCEPT; + + msg = ntohs(ctlh->messageType); + DEBUGP("outbound control message %s\n", pptp_msg_name[msg]); + + switch (msg) { + case PPTP_START_SESSION_REQUEST: + /* client requests for new control session */ + if (info->sstate != PPTP_SESSION_NONE) { + DEBUGP("%s but we already have one", + pptp_msg_name[msg]); + } + info->sstate = PPTP_SESSION_REQUESTED; + break; + case PPTP_STOP_SESSION_REQUEST: + /* client requests end of control session */ + info->sstate = PPTP_SESSION_STOPREQ; + break; + + case PPTP_OUT_CALL_REQUEST: + if (reqlen < sizeof(_pptpReq.ocreq)) { + DEBUGP("%s: short packet\n", pptp_msg_name[msg]); + /* FIXME: break; */ + } + + /* client initiating connection to server */ + if (info->sstate != PPTP_SESSION_CONFIRMED) { + DEBUGP("%s but no session\n", + pptp_msg_name[msg]); + break; + } + info->cstate = PPTP_CALL_OUT_REQ; + /* track PNS call id */ + cid = &pptpReq->ocreq.callID; + DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(*cid)); + info->pns_call_id = ntohs(*cid); + break; + case PPTP_IN_CALL_REPLY: + if (reqlen < sizeof(_pptpReq.icack)) { + DEBUGP("%s: short packet\n", pptp_msg_name[msg]); + break; + } + + /* client answers incoming call */ + if (info->cstate != PPTP_CALL_IN_REQ + && info->cstate != PPTP_CALL_IN_REP) { + DEBUGP("%s without incall_req\n", + pptp_msg_name[msg]); + break; + } + if (pptpReq->icack.resultCode != PPTP_INCALL_ACCEPT) { + info->cstate = PPTP_CALL_NONE; + break; + } + pcid = &pptpReq->icack.peersCallID; + if (info->pac_call_id != ntohs(*pcid)) { + DEBUGP("%s for unknown call %u\n", + pptp_msg_name[msg], ntohs(*pcid)); + break; + } + DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(*pcid)); + /* part two of the three-way handshake */ + info->cstate = PPTP_CALL_IN_REP; + info->pns_call_id = ntohs(pptpReq->icack.callID); + break; + + case PPTP_CALL_CLEAR_REQUEST: + /* client requests hangup of call */ + if (info->sstate != PPTP_SESSION_CONFIRMED) { + DEBUGP("CLEAR_CALL but no session\n"); + break; + } + /* FUTURE: iterate over all calls and check if + * call ID is valid. We don't do this without newnat, + * because we only know about last call */ + info->cstate = PPTP_CALL_CLEAR_REQ; + break; + case PPTP_SET_LINK_INFO: + break; + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* I don't have to explain these ;) */ + break; + default: + DEBUGP("invalid %s (TY=%d)\n", (msg <= PPTP_MSG_MAX)? + pptp_msg_name[msg]:pptp_msg_name[0], msg); + /* unknown: no need to create GRE masq table entry */ + break; + } + + if (ip_nat_pptp_hook_outbound) + return ip_nat_pptp_hook_outbound(pskb, ct, ctinfo, ctlh, + pptpReq); + + return NF_ACCEPT; +} + + +/* track caller id inside control connection, call expect_related */ +static int +conntrack_pptp_help(struct sk_buff **pskb, + struct ip_conntrack *ct, enum ip_conntrack_info ctinfo) + +{ + struct pptp_pkt_hdr _pptph, *pptph; + struct tcphdr _tcph, *tcph; + u_int32_t tcplen = (*pskb)->len - (*pskb)->nh.iph->ihl * 4; + u_int32_t datalen; + int dir = CTINFO2DIR(ctinfo); + struct ip_ct_pptp_master *info = &ct->help.ct_pptp_info; + unsigned int nexthdr_off; + + int oldsstate, oldcstate; + int ret; + + /* don't do any tracking before tcp handshake complete */ + if (ctinfo != IP_CT_ESTABLISHED + && ctinfo != IP_CT_ESTABLISHED+IP_CT_IS_REPLY) { + DEBUGP("ctinfo = %u, skipping\n", ctinfo); + return NF_ACCEPT; + } + + nexthdr_off = (*pskb)->nh.iph->ihl*4; + tcph = skb_header_pointer(*pskb, nexthdr_off, sizeof(_tcph), &_tcph); + BUG_ON(!tcph); + nexthdr_off += tcph->doff * 4; + datalen = tcplen - tcph->doff * 4; + + if (tcph->fin || tcph->rst) { + DEBUGP("RST/FIN received, timeouting GRE\n"); + /* can't do this after real newnat */ + info->cstate = PPTP_CALL_NONE; + + /* untrack this call id, unexpect GRE packets */ + pptp_destroy_siblings(ct); + } + + pptph = skb_header_pointer(*pskb, nexthdr_off, sizeof(_pptph), &_pptph); + if (!pptph) { + DEBUGP("no full PPTP header, can't track\n"); + return NF_ACCEPT; + } + nexthdr_off += sizeof(_pptph); + datalen -= sizeof(_pptph); + + /* if it's not a control message we can't do anything with it */ + if (ntohs(pptph->packetType) != PPTP_PACKET_CONTROL || + ntohl(pptph->magicCookie) != PPTP_MAGIC_COOKIE) { + DEBUGP("not a control packet\n"); + return NF_ACCEPT; + } + + oldsstate = info->sstate; + oldcstate = info->cstate; + + spin_lock_bh(&ip_pptp_lock); + + /* FIXME: We just blindly assume that the control connection is always + * established from PNS->PAC. However, RFC makes no guarantee */ + if (dir == IP_CT_DIR_ORIGINAL) + /* client -> server (PNS -> PAC) */ + ret = pptp_outbound_pkt(pskb, tcph, nexthdr_off, datalen, ct, + ctinfo); + else + /* server -> client (PAC -> PNS) */ + ret = pptp_inbound_pkt(pskb, tcph, nexthdr_off, datalen, ct, + ctinfo); + DEBUGP("sstate: %d->%d, cstate: %d->%d\n", + oldsstate, info->sstate, oldcstate, info->cstate); + spin_unlock_bh(&ip_pptp_lock); + + return ret; +} + +/* control protocol helper */ +static struct ip_conntrack_helper pptp = { + .list = { NULL, NULL }, + .name = "pptp", + .me = THIS_MODULE, + .max_expected = 2, + .timeout = 5 * 60, + .tuple = { .src = { .ip = 0, + .u = { .tcp = { .port = + __constant_htons(PPTP_CONTROL_PORT) } } + }, + .dst = { .ip = 0, + .u = { .all = 0 }, + .protonum = IPPROTO_TCP + } + }, + .mask = { .src = { .ip = 0, + .u = { .tcp = { .port = 0xffff } } + }, + .dst = { .ip = 0, + .u = { .all = 0 }, + .protonum = 0xff + } + }, + .help = conntrack_pptp_help +}; + +extern void __exit ip_ct_proto_gre_fini(void); +extern int __init ip_ct_proto_gre_init(void); + +/* ip_conntrack_pptp initialization */ +static int __init init(void) +{ + int retcode; + + retcode = ip_ct_proto_gre_init(); + if (retcode < 0) + return retcode; + + DEBUGP(" registering helper\n"); + if ((retcode = ip_conntrack_helper_register(&pptp))) { + printk(KERN_ERR "Unable to register conntrack application " + "helper for pptp: %d\n", retcode); + ip_ct_proto_gre_fini(); + return retcode; + } + + printk("ip_conntrack_pptp version %s loaded\n", IP_CT_PPTP_VERSION); + return 0; +} + +static void __exit fini(void) +{ + ip_conntrack_helper_unregister(&pptp); + ip_ct_proto_gre_fini(); + printk("ip_conntrack_pptp version %s unloaded\n", IP_CT_PPTP_VERSION); +} + +module_init(init); +module_exit(fini); + +EXPORT_SYMBOL(ip_nat_pptp_hook_outbound); +EXPORT_SYMBOL(ip_nat_pptp_hook_inbound); +EXPORT_SYMBOL(ip_nat_pptp_hook_exp_gre); +EXPORT_SYMBOL(ip_nat_pptp_hook_expectfn); diff --git a/net/ipv4/netfilter/ip_conntrack_proto_gre.c b/net/ipv4/netfilter/ip_conntrack_proto_gre.c new file mode 100644 index 000000000000..de3cb9db6f85 --- /dev/null +++ b/net/ipv4/netfilter/ip_conntrack_proto_gre.c @@ -0,0 +1,327 @@ +/* + * ip_conntrack_proto_gre.c - Version 3.0 + * + * Connection tracking protocol helper module for GRE. + * + * GRE is a generic encapsulation protocol, which is generally not very + * suited for NAT, as it has no protocol-specific part as port numbers. + * + * It has an optional key field, which may help us distinguishing two + * connections between the same two hosts. + * + * GRE is defined in RFC 1701 and RFC 1702, as well as RFC 2784 + * + * PPTP is built on top of a modified version of GRE, and has a mandatory + * field called "CallID", which serves us for the same purpose as the key + * field in plain GRE. + * + * Documentation about PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +static DEFINE_RWLOCK(ip_ct_gre_lock); +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) + +#include +#include +#include +#include + +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("netfilter connection tracking protocol helper for GRE"); + +/* shamelessly stolen from ip_conntrack_proto_udp.c */ +#define GRE_TIMEOUT (30*HZ) +#define GRE_STREAM_TIMEOUT (180*HZ) + +#if 0 +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, __FUNCTION__, ## args) +#define DUMP_TUPLE_GRE(x) printk("%u.%u.%u.%u:0x%x -> %u.%u.%u.%u:0x%x\n", \ + NIPQUAD((x)->src.ip), ntohs((x)->src.u.gre.key), \ + NIPQUAD((x)->dst.ip), ntohs((x)->dst.u.gre.key)) +#else +#define DEBUGP(x, args...) +#define DUMP_TUPLE_GRE(x) +#endif + +/* GRE KEYMAP HANDLING FUNCTIONS */ +static LIST_HEAD(gre_keymap_list); + +static inline int gre_key_cmpfn(const struct ip_ct_gre_keymap *km, + const struct ip_conntrack_tuple *t) +{ + return ((km->tuple.src.ip == t->src.ip) && + (km->tuple.dst.ip == t->dst.ip) && + (km->tuple.dst.protonum == t->dst.protonum) && + (km->tuple.dst.u.all == t->dst.u.all)); +} + +/* look up the source key for a given tuple */ +static u_int32_t gre_keymap_lookup(struct ip_conntrack_tuple *t) +{ + struct ip_ct_gre_keymap *km; + u_int32_t key = 0; + + read_lock_bh(&ip_ct_gre_lock); + km = LIST_FIND(&gre_keymap_list, gre_key_cmpfn, + struct ip_ct_gre_keymap *, t); + if (km) + key = km->tuple.src.u.gre.key; + read_unlock_bh(&ip_ct_gre_lock); + + DEBUGP("lookup src key 0x%x up key for ", key); + DUMP_TUPLE_GRE(t); + + return key; +} + +/* add a single keymap entry, associate with specified master ct */ +int +ip_ct_gre_keymap_add(struct ip_conntrack *ct, + struct ip_conntrack_tuple *t, int reply) +{ + struct ip_ct_gre_keymap **exist_km, *km, *old; + + if (!ct->helper || strcmp(ct->helper->name, "pptp")) { + DEBUGP("refusing to add GRE keymap to non-pptp session\n"); + return -1; + } + + if (!reply) + exist_km = &ct->help.ct_pptp_info.keymap_orig; + else + exist_km = &ct->help.ct_pptp_info.keymap_reply; + + if (*exist_km) { + /* check whether it's a retransmission */ + old = LIST_FIND(&gre_keymap_list, gre_key_cmpfn, + struct ip_ct_gre_keymap *, t); + if (old == *exist_km) { + DEBUGP("retransmission\n"); + return 0; + } + + DEBUGP("trying to override keymap_%s for ct %p\n", + reply? "reply":"orig", ct); + return -EEXIST; + } + + km = kmalloc(sizeof(*km), GFP_ATOMIC); + if (!km) + return -ENOMEM; + + memcpy(&km->tuple, t, sizeof(*t)); + *exist_km = km; + + DEBUGP("adding new entry %p: ", km); + DUMP_TUPLE_GRE(&km->tuple); + + write_lock_bh(&ip_ct_gre_lock); + list_append(&gre_keymap_list, km); + write_unlock_bh(&ip_ct_gre_lock); + + return 0; +} + +/* destroy the keymap entries associated with specified master ct */ +void ip_ct_gre_keymap_destroy(struct ip_conntrack *ct) +{ + DEBUGP("entering for ct %p\n", ct); + + if (!ct->helper || strcmp(ct->helper->name, "pptp")) { + DEBUGP("refusing to destroy GRE keymap to non-pptp session\n"); + return; + } + + write_lock_bh(&ip_ct_gre_lock); + if (ct->help.ct_pptp_info.keymap_orig) { + DEBUGP("removing %p from list\n", + ct->help.ct_pptp_info.keymap_orig); + list_del(&ct->help.ct_pptp_info.keymap_orig->list); + kfree(ct->help.ct_pptp_info.keymap_orig); + ct->help.ct_pptp_info.keymap_orig = NULL; + } + if (ct->help.ct_pptp_info.keymap_reply) { + DEBUGP("removing %p from list\n", + ct->help.ct_pptp_info.keymap_reply); + list_del(&ct->help.ct_pptp_info.keymap_reply->list); + kfree(ct->help.ct_pptp_info.keymap_reply); + ct->help.ct_pptp_info.keymap_reply = NULL; + } + write_unlock_bh(&ip_ct_gre_lock); +} + + +/* PUBLIC CONNTRACK PROTO HELPER FUNCTIONS */ + +/* invert gre part of tuple */ +static int gre_invert_tuple(struct ip_conntrack_tuple *tuple, + const struct ip_conntrack_tuple *orig) +{ + tuple->dst.u.gre.key = orig->src.u.gre.key; + tuple->src.u.gre.key = orig->dst.u.gre.key; + + return 1; +} + +/* gre hdr info to tuple */ +static int gre_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct ip_conntrack_tuple *tuple) +{ + struct gre_hdr_pptp _pgrehdr, *pgrehdr; + u_int32_t srckey; + struct gre_hdr _grehdr, *grehdr; + + /* first only delinearize old RFC1701 GRE header */ + grehdr = skb_header_pointer(skb, dataoff, sizeof(_grehdr), &_grehdr); + if (!grehdr || grehdr->version != GRE_VERSION_PPTP) { + /* try to behave like "ip_conntrack_proto_generic" */ + tuple->src.u.all = 0; + tuple->dst.u.all = 0; + return 1; + } + + /* PPTP header is variable length, only need up to the call_id field */ + pgrehdr = skb_header_pointer(skb, dataoff, 8, &_pgrehdr); + if (!pgrehdr) + return 1; + + if (ntohs(grehdr->protocol) != GRE_PROTOCOL_PPTP) { + DEBUGP("GRE_VERSION_PPTP but unknown proto\n"); + return 0; + } + + tuple->dst.u.gre.key = pgrehdr->call_id; + srckey = gre_keymap_lookup(tuple); + tuple->src.u.gre.key = srckey; + + return 1; +} + +/* print gre part of tuple */ +static int gre_print_tuple(struct seq_file *s, + const struct ip_conntrack_tuple *tuple) +{ + return seq_printf(s, "srckey=0x%x dstkey=0x%x ", + ntohs(tuple->src.u.gre.key), + ntohs(tuple->dst.u.gre.key)); +} + +/* print private data for conntrack */ +static int gre_print_conntrack(struct seq_file *s, + const struct ip_conntrack *ct) +{ + return seq_printf(s, "timeout=%u, stream_timeout=%u ", + (ct->proto.gre.timeout / HZ), + (ct->proto.gre.stream_timeout / HZ)); +} + +/* Returns verdict for packet, and may modify conntrack */ +static int gre_packet(struct ip_conntrack *ct, + const struct sk_buff *skb, + enum ip_conntrack_info conntrackinfo) +{ + /* If we've seen traffic both ways, this is a GRE connection. + * Extend timeout. */ + if (ct->status & IPS_SEEN_REPLY) { + ip_ct_refresh_acct(ct, conntrackinfo, skb, + ct->proto.gre.stream_timeout); + /* Also, more likely to be important, and not a probe. */ + set_bit(IPS_ASSURED_BIT, &ct->status); + } else + ip_ct_refresh_acct(ct, conntrackinfo, skb, + ct->proto.gre.timeout); + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int gre_new(struct ip_conntrack *ct, + const struct sk_buff *skb) +{ + DEBUGP(": "); + DUMP_TUPLE_GRE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + + /* initialize to sane value. Ideally a conntrack helper + * (e.g. in case of pptp) is increasing them */ + ct->proto.gre.stream_timeout = GRE_STREAM_TIMEOUT; + ct->proto.gre.timeout = GRE_TIMEOUT; + + return 1; +} + +/* Called when a conntrack entry has already been removed from the hashes + * and is about to be deleted from memory */ +static void gre_destroy(struct ip_conntrack *ct) +{ + struct ip_conntrack *master = ct->master; + DEBUGP(" entering\n"); + + if (!master) + DEBUGP("no master !?!\n"); + else + ip_ct_gre_keymap_destroy(master); +} + +/* protocol helper struct */ +static struct ip_conntrack_protocol gre = { + .proto = IPPROTO_GRE, + .name = "gre", + .pkt_to_tuple = gre_pkt_to_tuple, + .invert_tuple = gre_invert_tuple, + .print_tuple = gre_print_tuple, + .print_conntrack = gre_print_conntrack, + .packet = gre_packet, + .new = gre_new, + .destroy = gre_destroy, + .me = THIS_MODULE, +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) + .tuple_to_nfattr = ip_ct_port_tuple_to_nfattr, + .nfattr_to_tuple = ip_ct_port_nfattr_to_tuple, +#endif +}; + +/* ip_conntrack_proto_gre initialization */ +int __init ip_ct_proto_gre_init(void) +{ + return ip_conntrack_protocol_register(&gre); +} + +void __exit ip_ct_proto_gre_fini(void) +{ + struct list_head *pos, *n; + + /* delete all keymap entries */ + write_lock_bh(&ip_ct_gre_lock); + list_for_each_safe(pos, n, &gre_keymap_list) { + DEBUGP("deleting keymap %p at module unload time\n", pos); + list_del(pos); + kfree(pos); + } + write_unlock_bh(&ip_ct_gre_lock); + + ip_conntrack_protocol_unregister(&gre); +} + +EXPORT_SYMBOL(ip_ct_gre_keymap_add); +EXPORT_SYMBOL(ip_ct_gre_keymap_destroy); diff --git a/net/ipv4/netfilter/ip_nat_helper_pptp.c b/net/ipv4/netfilter/ip_nat_helper_pptp.c new file mode 100644 index 000000000000..3cdd0684d30d --- /dev/null +++ b/net/ipv4/netfilter/ip_nat_helper_pptp.c @@ -0,0 +1,401 @@ +/* + * ip_nat_pptp.c - Version 3.0 + * + * NAT support for PPTP (Point to Point Tunneling Protocol). + * PPTP is a a protocol for creating virtual private networks. + * It is a specification defined by Microsoft and some vendors + * working with Microsoft. PPTP is built on top of a modified + * version of the Internet Generic Routing Encapsulation Protocol. + * GRE is defined in RFC 1701 and RFC 1702. Documentation of + * PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + * TODO: - NAT to a unique tuple, not to TCP source port + * (needs netfilter tuple reservation) + * + * Changes: + * 2002-02-10 - Version 1.3 + * - Use ip_nat_mangle_tcp_packet() because of cloned skb's + * in local connections (Philip Craig ) + * - add checks for magicCookie and pptp version + * - make argument list of pptp_{out,in}bound_packet() shorter + * - move to C99 style initializers + * - print version number at module loadtime + * 2003-09-22 - Version 1.5 + * - use SNATed tcp sourceport as callid, since we get called before + * TCP header is mangled (Philip Craig ) + * 2004-10-22 - Version 2.0 + * - kernel 2.6.x version + * 2005-06-10 - Version 3.0 + * - kernel >= 2.6.11 version, + * funded by Oxcoda NetBox Blue (http://www.netboxblue.com/) + * + */ + +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + +#define IP_NAT_PPTP_VERSION "3.0" + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("Netfilter NAT helper module for PPTP"); + + +#if 0 +extern const char *pptp_msg_name[]; +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, \ + __FUNCTION__, ## args) +#else +#define DEBUGP(format, args...) +#endif + +static void pptp_nat_expected(struct ip_conntrack *ct, + struct ip_conntrack_expect *exp) +{ + struct ip_conntrack *master = ct->master; + struct ip_conntrack_expect *other_exp; + struct ip_conntrack_tuple t; + struct ip_ct_pptp_master *ct_pptp_info; + struct ip_nat_pptp *nat_pptp_info; + + ct_pptp_info = &master->help.ct_pptp_info; + nat_pptp_info = &master->nat.help.nat_pptp_info; + + /* And here goes the grand finale of corrosion... */ + + if (exp->dir == IP_CT_DIR_ORIGINAL) { + DEBUGP("we are PNS->PAC\n"); + /* therefore, build tuple for PAC->PNS */ + t.src.ip = master->tuplehash[IP_CT_DIR_REPLY].tuple.src.ip; + t.src.u.gre.key = htons(master->help.ct_pptp_info.pac_call_id); + t.dst.ip = master->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip; + t.dst.u.gre.key = htons(master->help.ct_pptp_info.pns_call_id); + t.dst.protonum = IPPROTO_GRE; + } else { + DEBUGP("we are PAC->PNS\n"); + /* build tuple for PNS->PAC */ + t.src.ip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip; + t.src.u.gre.key = + htons(master->nat.help.nat_pptp_info.pns_call_id); + t.dst.ip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip; + t.dst.u.gre.key = + htons(master->nat.help.nat_pptp_info.pac_call_id); + t.dst.protonum = IPPROTO_GRE; + } + + DEBUGP("trying to unexpect other dir: "); + DUMP_TUPLE(&t); + other_exp = ip_conntrack_expect_find(&t); + if (other_exp) { + ip_conntrack_unexpect_related(other_exp); + ip_conntrack_expect_put(other_exp); + DEBUGP("success\n"); + } else { + DEBUGP("not found!\n"); + } + + ip_nat_follow_master(ct, exp); +} + +/* outbound packets == from PNS to PAC */ +static int +pptp_outbound_pkt(struct sk_buff **pskb, + struct ip_conntrack *ct, + enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq) + +{ + struct ip_ct_pptp_master *ct_pptp_info = &ct->help.ct_pptp_info; + struct ip_nat_pptp *nat_pptp_info = &ct->nat.help.nat_pptp_info; + + u_int16_t msg, *cid = NULL, new_callid; + + new_callid = htons(ct_pptp_info->pns_call_id); + + switch (msg = ntohs(ctlh->messageType)) { + case PPTP_OUT_CALL_REQUEST: + cid = &pptpReq->ocreq.callID; + /* FIXME: ideally we would want to reserve a call ID + * here. current netfilter NAT core is not able to do + * this :( For now we use TCP source port. This breaks + * multiple calls within one control session */ + + /* save original call ID in nat_info */ + nat_pptp_info->pns_call_id = ct_pptp_info->pns_call_id; + + /* don't use tcph->source since we are at a DSTmanip + * hook (e.g. PREROUTING) and pkt is not mangled yet */ + new_callid = ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.tcp.port; + + /* save new call ID in ct info */ + ct_pptp_info->pns_call_id = ntohs(new_callid); + break; + case PPTP_IN_CALL_REPLY: + cid = &pptpReq->icreq.callID; + break; + case PPTP_CALL_CLEAR_REQUEST: + cid = &pptpReq->clrreq.callID; + break; + default: + DEBUGP("unknown outbound packet 0x%04x:%s\n", msg, + (msg <= PPTP_MSG_MAX)? + pptp_msg_name[msg]:pptp_msg_name[0]); + /* fall through */ + + case PPTP_SET_LINK_INFO: + /* only need to NAT in case PAC is behind NAT box */ + case PPTP_START_SESSION_REQUEST: + case PPTP_START_SESSION_REPLY: + case PPTP_STOP_SESSION_REQUEST: + case PPTP_STOP_SESSION_REPLY: + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* no need to alter packet */ + return NF_ACCEPT; + } + + /* only OUT_CALL_REQUEST, IN_CALL_REPLY, CALL_CLEAR_REQUEST pass + * down to here */ + + IP_NF_ASSERT(cid); + + DEBUGP("altering call id from 0x%04x to 0x%04x\n", + ntohs(*cid), ntohs(new_callid)); + + /* mangle packet */ + if (ip_nat_mangle_tcp_packet(pskb, ct, ctinfo, + (void *)cid - ((void *)ctlh - sizeof(struct pptp_pkt_hdr)), + sizeof(new_callid), + (char *)&new_callid, + sizeof(new_callid)) == 0) + return NF_DROP; + + return NF_ACCEPT; +} + +static int +pptp_exp_gre(struct ip_conntrack_expect *expect_orig, + struct ip_conntrack_expect *expect_reply) +{ + struct ip_ct_pptp_master *ct_pptp_info = + &expect_orig->master->help.ct_pptp_info; + struct ip_nat_pptp *nat_pptp_info = + &expect_orig->master->nat.help.nat_pptp_info; + + struct ip_conntrack *ct = expect_orig->master; + + struct ip_conntrack_tuple inv_t; + struct ip_conntrack_tuple *orig_t, *reply_t; + + /* save original PAC call ID in nat_info */ + nat_pptp_info->pac_call_id = ct_pptp_info->pac_call_id; + + /* alter expectation */ + orig_t = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + reply_t = &ct->tuplehash[IP_CT_DIR_REPLY].tuple; + + /* alter expectation for PNS->PAC direction */ + invert_tuplepr(&inv_t, &expect_orig->tuple); + expect_orig->saved_proto.gre.key = htons(nat_pptp_info->pac_call_id); + expect_orig->tuple.src.u.gre.key = htons(nat_pptp_info->pns_call_id); + expect_orig->tuple.dst.u.gre.key = htons(ct_pptp_info->pac_call_id); + inv_t.src.ip = reply_t->src.ip; + inv_t.dst.ip = reply_t->dst.ip; + inv_t.src.u.gre.key = htons(nat_pptp_info->pac_call_id); + inv_t.dst.u.gre.key = htons(ct_pptp_info->pns_call_id); + + if (!ip_conntrack_expect_related(expect_orig)) { + DEBUGP("successfully registered expect\n"); + } else { + DEBUGP("can't expect_related(expect_orig)\n"); + return 1; + } + + /* alter expectation for PAC->PNS direction */ + invert_tuplepr(&inv_t, &expect_reply->tuple); + expect_reply->saved_proto.gre.key = htons(nat_pptp_info->pns_call_id); + expect_reply->tuple.src.u.gre.key = htons(nat_pptp_info->pac_call_id); + expect_reply->tuple.dst.u.gre.key = htons(ct_pptp_info->pns_call_id); + inv_t.src.ip = orig_t->src.ip; + inv_t.dst.ip = orig_t->dst.ip; + inv_t.src.u.gre.key = htons(nat_pptp_info->pns_call_id); + inv_t.dst.u.gre.key = htons(ct_pptp_info->pac_call_id); + + if (!ip_conntrack_expect_related(expect_reply)) { + DEBUGP("successfully registered expect\n"); + } else { + DEBUGP("can't expect_related(expect_reply)\n"); + ip_conntrack_unexpect_related(expect_orig); + return 1; + } + + if (ip_ct_gre_keymap_add(ct, &expect_reply->tuple, 0) < 0) { + DEBUGP("can't register original keymap\n"); + ip_conntrack_unexpect_related(expect_orig); + ip_conntrack_unexpect_related(expect_reply); + return 1; + } + + if (ip_ct_gre_keymap_add(ct, &inv_t, 1) < 0) { + DEBUGP("can't register reply keymap\n"); + ip_conntrack_unexpect_related(expect_orig); + ip_conntrack_unexpect_related(expect_reply); + ip_ct_gre_keymap_destroy(ct); + return 1; + } + + return 0; +} + +/* inbound packets == from PAC to PNS */ +static int +pptp_inbound_pkt(struct sk_buff **pskb, + struct ip_conntrack *ct, + enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq) +{ + struct ip_nat_pptp *nat_pptp_info = &ct->nat.help.nat_pptp_info; + u_int16_t msg, new_cid = 0, new_pcid, *pcid = NULL, *cid = NULL; + + int ret = NF_ACCEPT, rv; + + new_pcid = htons(nat_pptp_info->pns_call_id); + + switch (msg = ntohs(ctlh->messageType)) { + case PPTP_OUT_CALL_REPLY: + pcid = &pptpReq->ocack.peersCallID; + cid = &pptpReq->ocack.callID; + break; + case PPTP_IN_CALL_CONNECT: + pcid = &pptpReq->iccon.peersCallID; + break; + case PPTP_IN_CALL_REQUEST: + /* only need to nat in case PAC is behind NAT box */ + break; + case PPTP_WAN_ERROR_NOTIFY: + pcid = &pptpReq->wanerr.peersCallID; + break; + case PPTP_CALL_DISCONNECT_NOTIFY: + pcid = &pptpReq->disc.callID; + break; + case PPTP_SET_LINK_INFO: + pcid = &pptpReq->setlink.peersCallID; + break; + + default: + DEBUGP("unknown inbound packet %s\n", (msg <= PPTP_MSG_MAX)? + pptp_msg_name[msg]:pptp_msg_name[0]); + /* fall through */ + + case PPTP_START_SESSION_REQUEST: + case PPTP_START_SESSION_REPLY: + case PPTP_STOP_SESSION_REQUEST: + case PPTP_STOP_SESSION_REPLY: + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* no need to alter packet */ + return NF_ACCEPT; + } + + /* only OUT_CALL_REPLY, IN_CALL_CONNECT, IN_CALL_REQUEST, + * WAN_ERROR_NOTIFY, CALL_DISCONNECT_NOTIFY pass down here */ + + /* mangle packet */ + IP_NF_ASSERT(pcid); + DEBUGP("altering peer call id from 0x%04x to 0x%04x\n", + ntohs(*pcid), ntohs(new_pcid)); + + rv = ip_nat_mangle_tcp_packet(pskb, ct, ctinfo, + (void *)pcid - ((void *)ctlh - sizeof(struct pptp_pkt_hdr)), + sizeof(new_pcid), (char *)&new_pcid, + sizeof(new_pcid)); + if (rv != NF_ACCEPT) + return rv; + + if (new_cid) { + IP_NF_ASSERT(cid); + DEBUGP("altering call id from 0x%04x to 0x%04x\n", + ntohs(*cid), ntohs(new_cid)); + rv = ip_nat_mangle_tcp_packet(pskb, ct, ctinfo, + (void *)cid - ((void *)ctlh - sizeof(struct pptp_pkt_hdr)), + sizeof(new_cid), + (char *)&new_cid, + sizeof(new_cid)); + if (rv != NF_ACCEPT) + return rv; + } + + /* check for earlier return value of 'switch' above */ + if (ret != NF_ACCEPT) + return ret; + + /* great, at least we don't need to resize packets */ + return NF_ACCEPT; +} + + +extern int __init ip_nat_proto_gre_init(void); +extern void __exit ip_nat_proto_gre_fini(void); + +static int __init init(void) +{ + int ret; + + DEBUGP("%s: registering NAT helper\n", __FILE__); + + ret = ip_nat_proto_gre_init(); + if (ret < 0) + return ret; + + BUG_ON(ip_nat_pptp_hook_outbound); + ip_nat_pptp_hook_outbound = &pptp_outbound_pkt; + + BUG_ON(ip_nat_pptp_hook_inbound); + ip_nat_pptp_hook_inbound = &pptp_inbound_pkt; + + BUG_ON(ip_nat_pptp_hook_exp_gre); + ip_nat_pptp_hook_exp_gre = &pptp_exp_gre; + + BUG_ON(ip_nat_pptp_hook_expectfn); + ip_nat_pptp_hook_expectfn = &pptp_nat_expected; + + printk("ip_nat_pptp version %s loaded\n", IP_NAT_PPTP_VERSION); + return 0; +} + +static void __exit fini(void) +{ + DEBUGP("cleanup_module\n" ); + + ip_nat_pptp_hook_expectfn = NULL; + ip_nat_pptp_hook_exp_gre = NULL; + ip_nat_pptp_hook_inbound = NULL; + ip_nat_pptp_hook_outbound = NULL; + + ip_nat_proto_gre_fini(); + /* Make sure noone calls it, meanwhile */ + synchronize_net(); + + printk("ip_nat_pptp version %s unloaded\n", IP_NAT_PPTP_VERSION); +} + +module_init(init); +module_exit(fini); diff --git a/net/ipv4/netfilter/ip_nat_proto_gre.c b/net/ipv4/netfilter/ip_nat_proto_gre.c new file mode 100644 index 000000000000..7c1285401672 --- /dev/null +++ b/net/ipv4/netfilter/ip_nat_proto_gre.c @@ -0,0 +1,214 @@ +/* + * ip_nat_proto_gre.c - Version 2.0 + * + * NAT protocol helper module for GRE. + * + * GRE is a generic encapsulation protocol, which is generally not very + * suited for NAT, as it has no protocol-specific part as port numbers. + * + * It has an optional key field, which may help us distinguishing two + * connections between the same two hosts. + * + * GRE is defined in RFC 1701 and RFC 1702, as well as RFC 2784 + * + * PPTP is built on top of a modified version of GRE, and has a mandatory + * field called "CallID", which serves us for the same purpose as the key + * field in plain GRE. + * + * Documentation about PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + */ + +#include +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("Netfilter NAT protocol helper module for GRE"); + +#if 0 +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, \ + __FUNCTION__, ## args) +#else +#define DEBUGP(x, args...) +#endif + +/* is key in given range between min and max */ +static int +gre_in_range(const struct ip_conntrack_tuple *tuple, + enum ip_nat_manip_type maniptype, + const union ip_conntrack_manip_proto *min, + const union ip_conntrack_manip_proto *max) +{ + u_int32_t key; + + if (maniptype == IP_NAT_MANIP_SRC) + key = tuple->src.u.gre.key; + else + key = tuple->dst.u.gre.key; + + return ntohl(key) >= ntohl(min->gre.key) + && ntohl(key) <= ntohl(max->gre.key); +} + +/* generate unique tuple ... */ +static int +gre_unique_tuple(struct ip_conntrack_tuple *tuple, + const struct ip_nat_range *range, + enum ip_nat_manip_type maniptype, + const struct ip_conntrack *conntrack) +{ + static u_int16_t key; + u_int16_t *keyptr; + unsigned int min, i, range_size; + + if (maniptype == IP_NAT_MANIP_SRC) + keyptr = &tuple->src.u.gre.key; + else + keyptr = &tuple->dst.u.gre.key; + + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { + DEBUGP("%p: NATing GRE PPTP\n", conntrack); + min = 1; + range_size = 0xffff; + } else { + min = ntohl(range->min.gre.key); + range_size = ntohl(range->max.gre.key) - min + 1; + } + + DEBUGP("min = %u, range_size = %u\n", min, range_size); + + for (i = 0; i < range_size; i++, key++) { + *keyptr = htonl(min + key % range_size); + if (!ip_nat_used_tuple(tuple, conntrack)) + return 1; + } + + DEBUGP("%p: no NAT mapping\n", conntrack); + + return 0; +} + +/* manipulate a GRE packet according to maniptype */ +static int +gre_manip_pkt(struct sk_buff **pskb, + unsigned int iphdroff, + const struct ip_conntrack_tuple *tuple, + enum ip_nat_manip_type maniptype) +{ + struct gre_hdr *greh; + struct gre_hdr_pptp *pgreh; + struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff); + unsigned int hdroff = iphdroff + iph->ihl*4; + + /* pgreh includes two optional 32bit fields which are not required + * to be there. That's where the magic '8' comes from */ + if (!skb_make_writable(pskb, hdroff + sizeof(*pgreh)-8)) + return 0; + + greh = (void *)(*pskb)->data + hdroff; + pgreh = (struct gre_hdr_pptp *) greh; + + /* we only have destination manip of a packet, since 'source key' + * is not present in the packet itself */ + if (maniptype == IP_NAT_MANIP_DST) { + /* key manipulation is always dest */ + switch (greh->version) { + case 0: + if (!greh->key) { + DEBUGP("can't nat GRE w/o key\n"); + break; + } + if (greh->csum) { + /* FIXME: Never tested this code... */ + *(gre_csum(greh)) = + ip_nat_cheat_check(~*(gre_key(greh)), + tuple->dst.u.gre.key, + *(gre_csum(greh))); + } + *(gre_key(greh)) = tuple->dst.u.gre.key; + break; + case GRE_VERSION_PPTP: + DEBUGP("call_id -> 0x%04x\n", + ntohl(tuple->dst.u.gre.key)); + pgreh->call_id = htons(ntohl(tuple->dst.u.gre.key)); + break; + default: + DEBUGP("can't nat unknown GRE version\n"); + return 0; + break; + } + } + return 1; +} + +/* print out a nat tuple */ +static unsigned int +gre_print(char *buffer, + const struct ip_conntrack_tuple *match, + const struct ip_conntrack_tuple *mask) +{ + unsigned int len = 0; + + if (mask->src.u.gre.key) + len += sprintf(buffer + len, "srckey=0x%x ", + ntohl(match->src.u.gre.key)); + + if (mask->dst.u.gre.key) + len += sprintf(buffer + len, "dstkey=0x%x ", + ntohl(match->src.u.gre.key)); + + return len; +} + +/* print a range of keys */ +static unsigned int +gre_print_range(char *buffer, const struct ip_nat_range *range) +{ + if (range->min.gre.key != 0 + || range->max.gre.key != 0xFFFF) { + if (range->min.gre.key == range->max.gre.key) + return sprintf(buffer, "key 0x%x ", + ntohl(range->min.gre.key)); + else + return sprintf(buffer, "keys 0x%u-0x%u ", + ntohl(range->min.gre.key), + ntohl(range->max.gre.key)); + } else + return 0; +} + +/* nat helper struct */ +static struct ip_nat_protocol gre = { + .name = "GRE", + .protonum = IPPROTO_GRE, + .manip_pkt = gre_manip_pkt, + .in_range = gre_in_range, + .unique_tuple = gre_unique_tuple, + .print = gre_print, + .print_range = gre_print_range, +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) + .range_to_nfattr = ip_nat_port_range_to_nfattr, + .nfattr_to_range = ip_nat_port_nfattr_to_range, +#endif +}; + +int __init ip_nat_proto_gre_init(void) +{ + return ip_nat_protocol_register(&gre); +} + +void __exit ip_nat_proto_gre_fini(void) +{ + ip_nat_protocol_unregister(&gre); +} From e674d0f38de6109b59dbe30fba8b296a03229b8e Mon Sep 17 00:00:00 2001 From: Yasuyuki Kozakai Date: Mon, 19 Sep 2005 15:34:40 -0700 Subject: [PATCH 179/240] [NETFILTER] ip6tables: remove duplicate code Some IPv6 matches have very similar loops to find IPv6 extension header and we can unify them. This patch introduces ipv6_find_hdr() to do it. I just checked that it can find the target headers in the packet which has dst,hbh,rt,frag,ah,esp headers. Signed-off-by: Yasuyuki Kozakai Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- include/linux/netfilter_ipv6/ip6_tables.h | 3 + net/ipv6/netfilter/ip6_tables.c | 52 ++++++++++++++ net/ipv6/netfilter/ip6t_ah.c | 81 ++------------------- net/ipv6/netfilter/ip6t_dst.c | 88 ++--------------------- net/ipv6/netfilter/ip6t_esp.c | 73 ++----------------- net/ipv6/netfilter/ip6t_frag.c | 88 +++-------------------- net/ipv6/netfilter/ip6t_hbh.c | 88 ++--------------------- net/ipv6/netfilter/ip6t_rt.c | 83 ++------------------- 8 files changed, 93 insertions(+), 463 deletions(-) diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h index 58c72a52dc65..59f70b34e029 100644 --- a/include/linux/netfilter_ipv6/ip6_tables.h +++ b/include/linux/netfilter_ipv6/ip6_tables.h @@ -455,6 +455,9 @@ extern unsigned int ip6t_do_table(struct sk_buff **pskb, /* Check for an extension */ extern int ip6t_ext_hdr(u8 nexthdr); +/* find specified header and get offset to it */ +extern int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, + u8 target); #define IP6T_ALIGN(s) (((s) + (__alignof__(struct ip6t_entry)-1)) & ~(__alignof__(struct ip6t_entry)-1)) diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 1cb8adb2787f..2da514b16d95 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -1955,6 +1955,57 @@ static void __exit fini(void) #endif } +/* + * find specified header up to transport protocol header. + * If found target header, the offset to the header is set to *offset + * and return 0. otherwise, return -1. + * + * Notes: - non-1st Fragment Header isn't skipped. + * - ESP header isn't skipped. + * - The target header may be trancated. + */ +int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, u8 target) +{ + unsigned int start = (u8*)(skb->nh.ipv6h + 1) - skb->data; + u8 nexthdr = skb->nh.ipv6h->nexthdr; + unsigned int len = skb->len - start; + + while (nexthdr != target) { + struct ipv6_opt_hdr _hdr, *hp; + unsigned int hdrlen; + + if ((!ipv6_ext_hdr(nexthdr)) || nexthdr == NEXTHDR_NONE) + return -1; + hp = skb_header_pointer(skb, start, sizeof(_hdr), &_hdr); + if (hp == NULL) + return -1; + if (nexthdr == NEXTHDR_FRAGMENT) { + unsigned short _frag_off, *fp; + fp = skb_header_pointer(skb, + start+offsetof(struct frag_hdr, + frag_off), + sizeof(_frag_off), + &_frag_off); + if (fp == NULL) + return -1; + + if (ntohs(*fp) & ~0x7) + return -1; + hdrlen = 8; + } else if (nexthdr == NEXTHDR_AUTH) + hdrlen = (hp->hdrlen + 2) << 2; + else + hdrlen = ipv6_optlen(hp); + + nexthdr = hp->nexthdr; + len -= hdrlen; + start += hdrlen; + } + + *offset = start; + return 0; +} + EXPORT_SYMBOL(ip6t_register_table); EXPORT_SYMBOL(ip6t_unregister_table); EXPORT_SYMBOL(ip6t_do_table); @@ -1963,6 +2014,7 @@ EXPORT_SYMBOL(ip6t_unregister_match); EXPORT_SYMBOL(ip6t_register_target); EXPORT_SYMBOL(ip6t_unregister_target); EXPORT_SYMBOL(ip6t_ext_hdr); +EXPORT_SYMBOL(ipv6_find_hdr); module_init(init); module_exit(fini); diff --git a/net/ipv6/netfilter/ip6t_ah.c b/net/ipv6/netfilter/ip6t_ah.c index d5b94f142bba..dde37793d20b 100644 --- a/net/ipv6/netfilter/ip6t_ah.c +++ b/net/ipv6/netfilter/ip6t_ah.c @@ -48,92 +48,21 @@ match(const struct sk_buff *skb, unsigned int protoff, int *hotdrop) { - struct ip_auth_hdr *ah = NULL, _ah; + struct ip_auth_hdr *ah, _ah; const struct ip6t_ah *ahinfo = matchinfo; - unsigned int temp; - int len; - u8 nexthdr; unsigned int ptr; unsigned int hdrlen = 0; - /*DEBUGP("IPv6 AH entered\n");*/ - /* if (opt->auth == 0) return 0; - * It does not filled on output */ - - /* type of the 1st exthdr */ - nexthdr = skb->nh.ipv6h->nexthdr; - /* pointer to the 1st exthdr */ - ptr = sizeof(struct ipv6hdr); - /* available length */ - len = skb->len - ptr; - temp = 0; - - while (ip6t_ext_hdr(nexthdr)) { - struct ipv6_opt_hdr _hdr, *hp; - - DEBUGP("ipv6_ah header iteration \n"); - - /* Is there enough space for the next ext header? */ - if (len < sizeof(struct ipv6_opt_hdr)) - return 0; - /* No more exthdr -> evaluate */ - if (nexthdr == NEXTHDR_NONE) - break; - /* ESP -> evaluate */ - if (nexthdr == NEXTHDR_ESP) - break; - - hp = skb_header_pointer(skb, ptr, sizeof(_hdr), &_hdr); - BUG_ON(hp == NULL); - - /* Calculate the header length */ - if (nexthdr == NEXTHDR_FRAGMENT) - hdrlen = 8; - else if (nexthdr == NEXTHDR_AUTH) - hdrlen = (hp->hdrlen+2)<<2; - else - hdrlen = ipv6_optlen(hp); - - /* AH -> evaluate */ - if (nexthdr == NEXTHDR_AUTH) { - temp |= MASK_AH; - break; - } - - - /* set the flag */ - switch (nexthdr) { - case NEXTHDR_HOP: - case NEXTHDR_ROUTING: - case NEXTHDR_FRAGMENT: - case NEXTHDR_AUTH: - case NEXTHDR_DEST: - break; - default: - DEBUGP("ipv6_ah match: unknown nextheader %u\n",nexthdr); - return 0; - } - - nexthdr = hp->nexthdr; - len -= hdrlen; - ptr += hdrlen; - if (ptr > skb->len) { - DEBUGP("ipv6_ah: new pointer too large! \n"); - break; - } - } - - /* AH header not found */ - if (temp != MASK_AH) + if (ipv6_find_hdr(skb, &ptr, NEXTHDR_AUTH) < 0) return 0; - if (len < sizeof(struct ip_auth_hdr)){ + ah = skb_header_pointer(skb, ptr, sizeof(_ah), &_ah); + if (ah == NULL) { *hotdrop = 1; return 0; } - ah = skb_header_pointer(skb, ptr, sizeof(_ah), &_ah); - BUG_ON(ah == NULL); + hdrlen = (ah->hdrlen + 2) << 2; DEBUGP("IPv6 AH LEN %u %u ", hdrlen, ah->hdrlen); DEBUGP("RES %04X ", ah->reserved); diff --git a/net/ipv6/netfilter/ip6t_dst.c b/net/ipv6/netfilter/ip6t_dst.c index 540925e4a7a8..c450a635e54b 100644 --- a/net/ipv6/netfilter/ip6t_dst.c +++ b/net/ipv6/netfilter/ip6t_dst.c @@ -63,8 +63,6 @@ match(const struct sk_buff *skb, struct ipv6_opt_hdr _optsh, *oh; const struct ip6t_opts *optinfo = matchinfo; unsigned int temp; - unsigned int len; - u8 nexthdr; unsigned int ptr; unsigned int hdrlen = 0; unsigned int ret = 0; @@ -72,97 +70,25 @@ match(const struct sk_buff *skb, u8 _optlen, *lp = NULL; unsigned int optlen; - /* type of the 1st exthdr */ - nexthdr = skb->nh.ipv6h->nexthdr; - /* pointer to the 1st exthdr */ - ptr = sizeof(struct ipv6hdr); - /* available length */ - len = skb->len - ptr; - temp = 0; - - while (ip6t_ext_hdr(nexthdr)) { - struct ipv6_opt_hdr _hdr, *hp; - - DEBUGP("ipv6_opts header iteration \n"); - - /* Is there enough space for the next ext header? */ - if (len < (int)sizeof(struct ipv6_opt_hdr)) - return 0; - /* No more exthdr -> evaluate */ - if (nexthdr == NEXTHDR_NONE) { - break; - } - /* ESP -> evaluate */ - if (nexthdr == NEXTHDR_ESP) { - break; - } - - hp = skb_header_pointer(skb, ptr, sizeof(_hdr), &_hdr); - BUG_ON(hp == NULL); - - /* Calculate the header length */ - if (nexthdr == NEXTHDR_FRAGMENT) { - hdrlen = 8; - } else if (nexthdr == NEXTHDR_AUTH) - hdrlen = (hp->hdrlen+2)<<2; - else - hdrlen = ipv6_optlen(hp); - - /* OPTS -> evaluate */ #if HOPBYHOP - if (nexthdr == NEXTHDR_HOP) { - temp |= MASK_HOPOPTS; + if (ipv6_find_hdr(skb, &ptr, NEXTHDR_HOP) < 0) #else - if (nexthdr == NEXTHDR_DEST) { - temp |= MASK_DSTOPTS; + if (ipv6_find_hdr(skb, &ptr, NEXTHDR_DEST) < 0) #endif - break; - } + return 0; - - /* set the flag */ - switch (nexthdr){ - case NEXTHDR_HOP: - case NEXTHDR_ROUTING: - case NEXTHDR_FRAGMENT: - case NEXTHDR_AUTH: - case NEXTHDR_DEST: - break; - default: - DEBUGP("ipv6_opts match: unknown nextheader %u\n",nexthdr); - return 0; - break; - } - - nexthdr = hp->nexthdr; - len -= hdrlen; - ptr += hdrlen; - if ( ptr > skb->len ) { - DEBUGP("ipv6_opts: new pointer is too large! \n"); - break; - } - } - - /* OPTIONS header not found */ -#if HOPBYHOP - if ( temp != MASK_HOPOPTS ) return 0; -#else - if ( temp != MASK_DSTOPTS ) return 0; -#endif - - if (len < (int)sizeof(struct ipv6_opt_hdr)){ + oh = skb_header_pointer(skb, ptr, sizeof(_optsh), &_optsh); + if (oh == NULL){ *hotdrop = 1; return 0; } - if (len < hdrlen){ + hdrlen = ipv6_optlen(oh); + if (skb->len - ptr < hdrlen){ /* Packet smaller than it's length field */ return 0; } - oh = skb_header_pointer(skb, ptr, sizeof(_optsh), &_optsh); - BUG_ON(oh == NULL); - DEBUGP("IPv6 OPTS LEN %u %u ", hdrlen, oh->hdrlen); DEBUGP("len %02X %04X %02X ", diff --git a/net/ipv6/netfilter/ip6t_esp.c b/net/ipv6/netfilter/ip6t_esp.c index e39dd236fd8e..24bc0cde43a1 100644 --- a/net/ipv6/netfilter/ip6t_esp.c +++ b/net/ipv6/netfilter/ip6t_esp.c @@ -48,87 +48,22 @@ match(const struct sk_buff *skb, unsigned int protoff, int *hotdrop) { - struct ip_esp_hdr _esp, *eh = NULL; + struct ip_esp_hdr _esp, *eh; const struct ip6t_esp *espinfo = matchinfo; - unsigned int temp; - int len; - u8 nexthdr; unsigned int ptr; /* Make sure this isn't an evil packet */ /*DEBUGP("ipv6_esp entered \n");*/ - /* type of the 1st exthdr */ - nexthdr = skb->nh.ipv6h->nexthdr; - /* pointer to the 1st exthdr */ - ptr = sizeof(struct ipv6hdr); - /* available length */ - len = skb->len - ptr; - temp = 0; - - while (ip6t_ext_hdr(nexthdr)) { - struct ipv6_opt_hdr _hdr, *hp; - int hdrlen; - - DEBUGP("ipv6_esp header iteration \n"); - - /* Is there enough space for the next ext header? */ - if (len < sizeof(struct ipv6_opt_hdr)) - return 0; - /* No more exthdr -> evaluate */ - if (nexthdr == NEXTHDR_NONE) - break; - /* ESP -> evaluate */ - if (nexthdr == NEXTHDR_ESP) { - temp |= MASK_ESP; - break; - } - - hp = skb_header_pointer(skb, ptr, sizeof(_hdr), &_hdr); - BUG_ON(hp == NULL); - - /* Calculate the header length */ - if (nexthdr == NEXTHDR_FRAGMENT) - hdrlen = 8; - else if (nexthdr == NEXTHDR_AUTH) - hdrlen = (hp->hdrlen+2)<<2; - else - hdrlen = ipv6_optlen(hp); - - /* set the flag */ - switch (nexthdr) { - case NEXTHDR_HOP: - case NEXTHDR_ROUTING: - case NEXTHDR_FRAGMENT: - case NEXTHDR_AUTH: - case NEXTHDR_DEST: - break; - default: - DEBUGP("ipv6_esp match: unknown nextheader %u\n",nexthdr); - return 0; - } - - nexthdr = hp->nexthdr; - len -= hdrlen; - ptr += hdrlen; - if (ptr > skb->len) { - DEBUGP("ipv6_esp: new pointer too large! \n"); - break; - } - } - - /* ESP header not found */ - if (temp != MASK_ESP) + if (ipv6_find_hdr(skb, &ptr, NEXTHDR_ESP) < 0) return 0; - if (len < sizeof(struct ip_esp_hdr)) { + eh = skb_header_pointer(skb, ptr, sizeof(_esp), &_esp); + if (eh == NULL) { *hotdrop = 1; return 0; } - eh = skb_header_pointer(skb, ptr, sizeof(_esp), &_esp); - BUG_ON(eh == NULL); - DEBUGP("IPv6 ESP SPI %u %08X\n", ntohl(eh->spi), ntohl(eh->spi)); return (eh != NULL) diff --git a/net/ipv6/netfilter/ip6t_frag.c b/net/ipv6/netfilter/ip6t_frag.c index 4bfa30a9bc80..085d5f8eea29 100644 --- a/net/ipv6/netfilter/ip6t_frag.c +++ b/net/ipv6/netfilter/ip6t_frag.c @@ -48,90 +48,18 @@ match(const struct sk_buff *skb, unsigned int protoff, int *hotdrop) { - struct frag_hdr _frag, *fh = NULL; + struct frag_hdr _frag, *fh; const struct ip6t_frag *fraginfo = matchinfo; - unsigned int temp; - int len; - u8 nexthdr; unsigned int ptr; - unsigned int hdrlen = 0; - /* type of the 1st exthdr */ - nexthdr = skb->nh.ipv6h->nexthdr; - /* pointer to the 1st exthdr */ - ptr = sizeof(struct ipv6hdr); - /* available length */ - len = skb->len - ptr; - temp = 0; + if (ipv6_find_hdr(skb, &ptr, NEXTHDR_FRAGMENT) < 0) + return 0; - while (ip6t_ext_hdr(nexthdr)) { - struct ipv6_opt_hdr _hdr, *hp; - - DEBUGP("ipv6_frag header iteration \n"); - - /* Is there enough space for the next ext header? */ - if (len < (int)sizeof(struct ipv6_opt_hdr)) - return 0; - /* No more exthdr -> evaluate */ - if (nexthdr == NEXTHDR_NONE) { - break; - } - /* ESP -> evaluate */ - if (nexthdr == NEXTHDR_ESP) { - break; - } - - hp = skb_header_pointer(skb, ptr, sizeof(_hdr), &_hdr); - BUG_ON(hp == NULL); - - /* Calculate the header length */ - if (nexthdr == NEXTHDR_FRAGMENT) { - hdrlen = 8; - } else if (nexthdr == NEXTHDR_AUTH) - hdrlen = (hp->hdrlen+2)<<2; - else - hdrlen = ipv6_optlen(hp); - - /* FRAG -> evaluate */ - if (nexthdr == NEXTHDR_FRAGMENT) { - temp |= MASK_FRAGMENT; - break; - } - - - /* set the flag */ - switch (nexthdr){ - case NEXTHDR_HOP: - case NEXTHDR_ROUTING: - case NEXTHDR_FRAGMENT: - case NEXTHDR_AUTH: - case NEXTHDR_DEST: - break; - default: - DEBUGP("ipv6_frag match: unknown nextheader %u\n",nexthdr); - return 0; - break; - } - - nexthdr = hp->nexthdr; - len -= hdrlen; - ptr += hdrlen; - if ( ptr > skb->len ) { - DEBUGP("ipv6_frag: new pointer too large! \n"); - break; - } - } - - /* FRAG header not found */ - if ( temp != MASK_FRAGMENT ) return 0; - - if (len < sizeof(struct frag_hdr)){ - *hotdrop = 1; - return 0; - } - - fh = skb_header_pointer(skb, ptr, sizeof(_frag), &_frag); - BUG_ON(fh == NULL); + fh = skb_header_pointer(skb, ptr, sizeof(_frag), &_frag); + if (fh == NULL){ + *hotdrop = 1; + return 0; + } DEBUGP("INFO %04X ", fh->frag_off); DEBUGP("OFFSET %04X ", ntohs(fh->frag_off) & ~0x7); diff --git a/net/ipv6/netfilter/ip6t_hbh.c b/net/ipv6/netfilter/ip6t_hbh.c index 27f3650d127e..1d09485111d0 100644 --- a/net/ipv6/netfilter/ip6t_hbh.c +++ b/net/ipv6/netfilter/ip6t_hbh.c @@ -63,8 +63,6 @@ match(const struct sk_buff *skb, struct ipv6_opt_hdr _optsh, *oh; const struct ip6t_opts *optinfo = matchinfo; unsigned int temp; - unsigned int len; - u8 nexthdr; unsigned int ptr; unsigned int hdrlen = 0; unsigned int ret = 0; @@ -72,97 +70,25 @@ match(const struct sk_buff *skb, u8 _optlen, *lp = NULL; unsigned int optlen; - /* type of the 1st exthdr */ - nexthdr = skb->nh.ipv6h->nexthdr; - /* pointer to the 1st exthdr */ - ptr = sizeof(struct ipv6hdr); - /* available length */ - len = skb->len - ptr; - temp = 0; - - while (ip6t_ext_hdr(nexthdr)) { - struct ipv6_opt_hdr _hdr, *hp; - - DEBUGP("ipv6_opts header iteration \n"); - - /* Is there enough space for the next ext header? */ - if (len < (int)sizeof(struct ipv6_opt_hdr)) - return 0; - /* No more exthdr -> evaluate */ - if (nexthdr == NEXTHDR_NONE) { - break; - } - /* ESP -> evaluate */ - if (nexthdr == NEXTHDR_ESP) { - break; - } - - hp = skb_header_pointer(skb, ptr, sizeof(_hdr), &_hdr); - BUG_ON(hp == NULL); - - /* Calculate the header length */ - if (nexthdr == NEXTHDR_FRAGMENT) { - hdrlen = 8; - } else if (nexthdr == NEXTHDR_AUTH) - hdrlen = (hp->hdrlen+2)<<2; - else - hdrlen = ipv6_optlen(hp); - - /* OPTS -> evaluate */ #if HOPBYHOP - if (nexthdr == NEXTHDR_HOP) { - temp |= MASK_HOPOPTS; + if (ipv6_find_hdr(skb, &ptr, NEXTHDR_HOP) < 0) #else - if (nexthdr == NEXTHDR_DEST) { - temp |= MASK_DSTOPTS; + if (ipv6_find_hdr(skb, &ptr, NEXTHDR_DEST) < 0) #endif - break; - } + return 0; - - /* set the flag */ - switch (nexthdr){ - case NEXTHDR_HOP: - case NEXTHDR_ROUTING: - case NEXTHDR_FRAGMENT: - case NEXTHDR_AUTH: - case NEXTHDR_DEST: - break; - default: - DEBUGP("ipv6_opts match: unknown nextheader %u\n",nexthdr); - return 0; - break; - } - - nexthdr = hp->nexthdr; - len -= hdrlen; - ptr += hdrlen; - if ( ptr > skb->len ) { - DEBUGP("ipv6_opts: new pointer is too large! \n"); - break; - } - } - - /* OPTIONS header not found */ -#if HOPBYHOP - if ( temp != MASK_HOPOPTS ) return 0; -#else - if ( temp != MASK_DSTOPTS ) return 0; -#endif - - if (len < (int)sizeof(struct ipv6_opt_hdr)){ + oh = skb_header_pointer(skb, ptr, sizeof(_optsh), &_optsh); + if (oh == NULL){ *hotdrop = 1; return 0; } - if (len < hdrlen){ + hdrlen = ipv6_optlen(oh); + if (skb->len - ptr < hdrlen){ /* Packet smaller than it's length field */ return 0; } - oh = skb_header_pointer(skb, ptr, sizeof(_optsh), &_optsh); - BUG_ON(oh == NULL); - DEBUGP("IPv6 OPTS LEN %u %u ", hdrlen, oh->hdrlen); DEBUGP("len %02X %04X %02X ", diff --git a/net/ipv6/netfilter/ip6t_rt.c b/net/ipv6/netfilter/ip6t_rt.c index 2bb670037df3..beb2fd5cebbb 100644 --- a/net/ipv6/netfilter/ip6t_rt.c +++ b/net/ipv6/netfilter/ip6t_rt.c @@ -50,98 +50,29 @@ match(const struct sk_buff *skb, unsigned int protoff, int *hotdrop) { - struct ipv6_rt_hdr _route, *rh = NULL; + struct ipv6_rt_hdr _route, *rh; const struct ip6t_rt *rtinfo = matchinfo; unsigned int temp; - unsigned int len; - u8 nexthdr; unsigned int ptr; unsigned int hdrlen = 0; unsigned int ret = 0; struct in6_addr *ap, _addr; - /* type of the 1st exthdr */ - nexthdr = skb->nh.ipv6h->nexthdr; - /* pointer to the 1st exthdr */ - ptr = sizeof(struct ipv6hdr); - /* available length */ - len = skb->len - ptr; - temp = 0; + if (ipv6_find_hdr(skb, &ptr, NEXTHDR_ROUTING) < 0) + return 0; - while (ip6t_ext_hdr(nexthdr)) { - struct ipv6_opt_hdr _hdr, *hp; - - DEBUGP("ipv6_rt header iteration \n"); - - /* Is there enough space for the next ext header? */ - if (len < (int)sizeof(struct ipv6_opt_hdr)) - return 0; - /* No more exthdr -> evaluate */ - if (nexthdr == NEXTHDR_NONE) { - break; - } - /* ESP -> evaluate */ - if (nexthdr == NEXTHDR_ESP) { - break; - } - - hp = skb_header_pointer(skb, ptr, sizeof(_hdr), &_hdr); - BUG_ON(hp == NULL); - - /* Calculate the header length */ - if (nexthdr == NEXTHDR_FRAGMENT) { - hdrlen = 8; - } else if (nexthdr == NEXTHDR_AUTH) - hdrlen = (hp->hdrlen+2)<<2; - else - hdrlen = ipv6_optlen(hp); - - /* ROUTING -> evaluate */ - if (nexthdr == NEXTHDR_ROUTING) { - temp |= MASK_ROUTING; - break; - } - - - /* set the flag */ - switch (nexthdr){ - case NEXTHDR_HOP: - case NEXTHDR_ROUTING: - case NEXTHDR_FRAGMENT: - case NEXTHDR_AUTH: - case NEXTHDR_DEST: - break; - default: - DEBUGP("ipv6_rt match: unknown nextheader %u\n",nexthdr); - return 0; - break; - } - - nexthdr = hp->nexthdr; - len -= hdrlen; - ptr += hdrlen; - if ( ptr > skb->len ) { - DEBUGP("ipv6_rt: new pointer is too large! \n"); - break; - } - } - - /* ROUTING header not found */ - if ( temp != MASK_ROUTING ) return 0; - - if (len < (int)sizeof(struct ipv6_rt_hdr)){ + rh = skb_header_pointer(skb, ptr, sizeof(_route), &_route); + if (rh == NULL){ *hotdrop = 1; return 0; } - if (len < hdrlen){ + hdrlen = ipv6_optlen(rh); + if (skb->len - ptr < hdrlen){ /* Pcket smaller than its length field */ return 0; } - rh = skb_header_pointer(skb, ptr, sizeof(_route), &_route); - BUG_ON(rh == NULL); - DEBUGP("IPv6 RT LEN %u %u ", hdrlen, rh->hdrlen); DEBUGP("TYPE %04X ", rh->type); DEBUGP("SGS_LEFT %u %02X\n", rh->segments_left, rh->segments_left); From a41bc00234a0a2ccaa99a194341ae108ae17ddc8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 19 Sep 2005 15:35:31 -0700 Subject: [PATCH 180/240] [NETFILTER]: Rename misnamed function Both __ip_conntrack_expect_find and ip_conntrack_expect_find_get take a reference to the expectation, the difference is that callers of __ip_conntrack_expect_find must hold ip_conntrack_lock. Signed-off-by: Patrick McHardy Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ip_conntrack.h | 2 +- net/ipv4/netfilter/ip_conntrack_core.c | 2 +- net/ipv4/netfilter/ip_conntrack_netlink.c | 4 ++-- net/ipv4/netfilter/ip_conntrack_standalone.c | 4 ++-- 4 files changed, 6 insertions(+), 6 deletions(-) diff --git a/include/linux/netfilter_ipv4/ip_conntrack.h b/include/linux/netfilter_ipv4/ip_conntrack.h index 2df446c952ef..bace72a76cc4 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack.h +++ b/include/linux/netfilter_ipv4/ip_conntrack.h @@ -384,7 +384,7 @@ extern struct ip_conntrack_expect * __ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple); extern struct ip_conntrack_expect * -ip_conntrack_expect_find_get(const struct ip_conntrack_tuple *tuple); +ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple); extern struct ip_conntrack_tuple_hash * __ip_conntrack_find(const struct ip_conntrack_tuple *tuple, diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index f8cd8e42961e..c1f82e0c81cf 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -233,7 +233,7 @@ __ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple) /* Just find a expectation corresponding to a tuple. */ struct ip_conntrack_expect * -ip_conntrack_expect_find_get(const struct ip_conntrack_tuple *tuple) +ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple) { struct ip_conntrack_expect *i; diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c index 15aef3564742..b08a432efcf8 100644 --- a/net/ipv4/netfilter/ip_conntrack_netlink.c +++ b/net/ipv4/netfilter/ip_conntrack_netlink.c @@ -1270,7 +1270,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - exp = ip_conntrack_expect_find_get(&tuple); + exp = ip_conntrack_expect_find(&tuple); if (!exp) return -ENOENT; @@ -1318,7 +1318,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, return err; /* bump usage count to 2 */ - exp = ip_conntrack_expect_find_get(&tuple); + exp = ip_conntrack_expect_find(&tuple); if (!exp) return -ENOENT; diff --git a/net/ipv4/netfilter/ip_conntrack_standalone.c b/net/ipv4/netfilter/ip_conntrack_standalone.c index ae3e3e655db5..d3c7808010ec 100644 --- a/net/ipv4/netfilter/ip_conntrack_standalone.c +++ b/net/ipv4/netfilter/ip_conntrack_standalone.c @@ -993,11 +993,11 @@ EXPORT_SYMBOL(ip_ct_refresh_acct); EXPORT_SYMBOL(ip_conntrack_expect_alloc); EXPORT_SYMBOL(ip_conntrack_expect_put); -EXPORT_SYMBOL_GPL(ip_conntrack_expect_find_get); +EXPORT_SYMBOL_GPL(__ip_conntrack_expect_find); +EXPORT_SYMBOL_GPL(ip_conntrack_expect_find); EXPORT_SYMBOL(ip_conntrack_expect_related); EXPORT_SYMBOL(ip_conntrack_unexpect_related); EXPORT_SYMBOL_GPL(ip_conntrack_expect_list); -EXPORT_SYMBOL_GPL(__ip_conntrack_expect_find); EXPORT_SYMBOL_GPL(ip_ct_unlink_expect); EXPORT_SYMBOL(ip_conntrack_tuple_taken); From 8922bc93aa78a202c2b68b2de19b71329cb321e1 Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Mon, 19 Sep 2005 15:35:57 -0700 Subject: [PATCH 181/240] [NETFILTER]: Export ip_nat_port_{nfattr_to_range,range_to_nfattr} Those exports are needed by the PPTP helper following in the next couple of changes. Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_nat_core.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c index 1adedb743f60..c3ea891d38e7 100644 --- a/net/ipv4/netfilter/ip_nat_core.c +++ b/net/ipv4/netfilter/ip_nat_core.c @@ -578,6 +578,8 @@ ip_nat_port_nfattr_to_range(struct nfattr *tb[], struct ip_nat_range *range) return ret; } +EXPORT_SYMBOL_GPL(ip_nat_port_nfattr_to_range); +EXPORT_SYMBOL_GPL(ip_nat_port_range_to_nfattr); #endif int __init ip_nat_init(void) From 2cf655cd65888e9fed0803d77e9e4f7d1db674cc Mon Sep 17 00:00:00 2001 From: Krzysztof Halasa Date: Mon, 19 Sep 2005 15:39:32 -0700 Subject: [PATCH 182/240] [WAN] hdlc_cisco: Fix regression introduced by skb->tail changes. The following commit breaks cisco mode with my WAN drivers: author David S. Miller Tue, 28 Jun 2005 22:25:31 +0000 (15:25 -0700) commit 689be43945e9ca7dd704522e55af1b8a73a994d3 "[NET]: Remove gratuitous use of skb->tail in network drivers." The following patch fixes it - please apply (cisco_hard_header does skb_push(4 bytes)). Signed-off-by: David S. Miller --- drivers/net/wan/hdlc_cisco.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wan/hdlc_cisco.c b/drivers/net/wan/hdlc_cisco.c index 48c03c11cd9a..a01efa6d5c62 100644 --- a/drivers/net/wan/hdlc_cisco.c +++ b/drivers/net/wan/hdlc_cisco.c @@ -72,7 +72,7 @@ static void cisco_keepalive_send(struct net_device *dev, u32 type, } skb_reserve(skb, 4); cisco_hard_header(skb, dev, CISCO_KEEPALIVE, NULL, NULL, 0); - data = (cisco_packet*)skb->data; + data = (cisco_packet*)(skb->data + 4); data->type = htonl(type); data->par1 = htonl(par1); From 3c3f8f25c177e4f9e4e00bcc1b90b28b1be37937 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 19 Sep 2005 15:41:28 -0700 Subject: [PATCH 183/240] [8021Q]: Add endian annotations. Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/linux/if_vlan.h | 8 ++++---- net/8021q/vlan_dev.c | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 17d0c0d40b0e..eef0876d8307 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -42,8 +42,8 @@ struct hlist_node; struct vlan_ethhdr { unsigned char h_dest[ETH_ALEN]; /* destination eth addr */ unsigned char h_source[ETH_ALEN]; /* source ether addr */ - unsigned short h_vlan_proto; /* Should always be 0x8100 */ - unsigned short h_vlan_TCI; /* Encapsulates priority and VLAN ID */ + __be16 h_vlan_proto; /* Should always be 0x8100 */ + __be16 h_vlan_TCI; /* Encapsulates priority and VLAN ID */ unsigned short h_vlan_encapsulated_proto; /* packet type ID field (or len) */ }; @@ -55,8 +55,8 @@ static inline struct vlan_ethhdr *vlan_eth_hdr(const struct sk_buff *skb) } struct vlan_hdr { - unsigned short h_vlan_TCI; /* Encapsulates priority and VLAN ID */ - unsigned short h_vlan_encapsulated_proto; /* packet type ID field (or len) */ + __be16 h_vlan_TCI; /* Encapsulates priority and VLAN ID */ + __be16 h_vlan_encapsulated_proto; /* packet type ID field (or len) */ }; #define VLAN_VID_MASK 0xfff diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 145f5cde96cf..b74864889670 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -120,7 +120,7 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev, unsigned short vid; struct net_device_stats *stats; unsigned short vlan_TCI; - unsigned short proto; + __be16 proto; /* vlan_TCI = ntohs(get_unaligned(&vhdr->h_vlan_TCI)); */ vlan_TCI = ntohs(vhdr->h_vlan_TCI); From e14c3caf605dfd29bd1aac3097e39db94afc9f07 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 19 Sep 2005 18:18:38 -0700 Subject: [PATCH 184/240] [TCP]: Handle SACK'd packets properly in tcp_fragment(). The problem is that we're now calling tcp_fragment() in a context where the packets might be marked as SACKED_ACKED or SACKED_RETRANS. This was not possible before as you never retransmitted packets that are so marked. Because of this, we need to adjust sacked_out and retrans_out in tcp_fragment(). This is exactly what the following patch does. We also need to preserve the SACKED_ACKED/SACKED_RETRANS marking if they exist. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/tcp_output.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index b018e31b6530..5dd6dd7d091e 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -461,9 +461,7 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss flags = TCP_SKB_CB(skb)->flags; TCP_SKB_CB(skb)->flags = flags & ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH); TCP_SKB_CB(buff)->flags = flags; - TCP_SKB_CB(buff)->sacked = - (TCP_SKB_CB(skb)->sacked & - (TCPCB_LOST | TCPCB_EVER_RETRANS | TCPCB_AT_TAIL)); + TCP_SKB_CB(buff)->sacked = TCP_SKB_CB(skb)->sacked; TCP_SKB_CB(skb)->sacked &= ~TCPCB_AT_TAIL; if (!skb_shinfo(skb)->nr_frags && skb->ip_summed != CHECKSUM_HW) { @@ -501,6 +499,12 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss tcp_skb_pcount(buff); tp->packets_out -= diff; + + if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) + tp->sacked_out -= diff; + if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS) + tp->retrans_out -= diff; + if (TCP_SKB_CB(skb)->sacked & TCPCB_LOST) { tp->lost_out -= diff; tp->left_out -= diff; From 6d1cfe3f1752f17e297df60c8bcc6cd6e0a58449 Mon Sep 17 00:00:00 2001 From: Mark J Cox Date: Mon, 19 Sep 2005 17:55:30 -0700 Subject: [PATCH 185/240] [PATCH] raw_sendmsg DoS on 2.6 Fix unchecked __get_user that could be tricked into generating a memory read on an arbitrary address. The result of the read is not returned directly but you may be able to divine some information about it, or use the read to cause a crash on some architectures by reading hardware state. CAN-2004-2492. Fix from Al Viro, ack from Dave Miller. Signed-off-by: Linus Torvalds --- net/ipv4/raw.c | 2 +- net/ipv6/raw.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 304bb0a1d4f0..4b0d7e4d6269 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -361,7 +361,7 @@ static void raw_probe_proto_opt(struct flowi *fl, struct msghdr *msg) if (type && code) { get_user(fl->fl_icmp_type, type); - __get_user(fl->fl_icmp_code, code); + get_user(fl->fl_icmp_code, code); probed = 1; } break; diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 5aa3691c578d..a1265a320b11 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -627,7 +627,7 @@ static void rawv6_probe_proto_opt(struct flowi *fl, struct msghdr *msg) if (type && code) { get_user(fl->fl_icmp_type, type); - __get_user(fl->fl_icmp_code, code); + get_user(fl->fl_icmp_code, code); probed = 1; } break; From f805fbdaacf4367ce566743a665622387768ac0d Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 19 Sep 2005 19:54:29 -0700 Subject: [PATCH 186/240] Make fsnotify possibly work better for the inode removal case Checking i_nlink is dubious, but the alternatives look even less appetizing. Signed-off-by: Linus Torvalds --- fs/dcache.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/dcache.c b/fs/dcache.c index 7376b61269fb..fb10386c59be 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -102,7 +102,8 @@ static inline void dentry_iput(struct dentry * dentry) list_del_init(&dentry->d_alias); spin_unlock(&dentry->d_lock); spin_unlock(&dcache_lock); - fsnotify_inoderemove(inode); + if (!inode->i_nlink) + fsnotify_inoderemove(inode); if (dentry->d_op && dentry->d_op->d_iput) dentry->d_op->d_iput(dentry, inode); else From ff171d8f66a7fe1a000e610e9de11224749f9a22 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 19 Sep 2005 19:56:06 -0700 Subject: [PATCH 187/240] [SPARC64]: Handle little-endian unaligned loads/stores correctly. Because we use byte loads/stores to cons up the value in and out of registers, we can't expect the ASI endianness setting to take care of this for us. So do it by hand. This case is triggered by drivers/block/aoe/aoecmd.c in the ataid_complete() function where it goes: /* word 100: number lba48 sectors */ ssize = le64_to_cpup((__le64 *) &id[100<<1]); This &id[100<<1] address is 4 byte, rather than 8 byte aligned, thus triggering the unaligned exception. Signed-off-by: David S. Miller --- arch/sparc64/kernel/una_asm.S | 2 +- arch/sparc64/kernel/unaligned.c | 64 +++++++++++++++++++++++++++++---- 2 files changed, 58 insertions(+), 8 deletions(-) diff --git a/arch/sparc64/kernel/una_asm.S b/arch/sparc64/kernel/una_asm.S index cbb40585253c..da48400bcc95 100644 --- a/arch/sparc64/kernel/una_asm.S +++ b/arch/sparc64/kernel/una_asm.S @@ -17,7 +17,7 @@ kernel_unaligned_trap_fault: __do_int_store: rd %asi, %o4 wr %o3, 0, %asi - ldx [%o2], %g3 + mov %o2, %g3 cmp %o1, 2 be,pn %icc, 2f cmp %o1, 4 diff --git a/arch/sparc64/kernel/unaligned.c b/arch/sparc64/kernel/unaligned.c index da9739f0d437..42718f6a7d36 100644 --- a/arch/sparc64/kernel/unaligned.c +++ b/arch/sparc64/kernel/unaligned.c @@ -184,13 +184,14 @@ extern void do_int_load(unsigned long *dest_reg, int size, unsigned long *saddr, int is_signed, int asi); extern void __do_int_store(unsigned long *dst_addr, int size, - unsigned long *src_val, int asi); + unsigned long src_val, int asi); static inline void do_int_store(int reg_num, int size, unsigned long *dst_addr, - struct pt_regs *regs, int asi) + struct pt_regs *regs, int asi, int orig_asi) { unsigned long zero = 0; - unsigned long *src_val = &zero; + unsigned long *src_val_p = &zero; + unsigned long src_val; if (size == 16) { size = 8; @@ -198,7 +199,25 @@ static inline void do_int_store(int reg_num, int size, unsigned long *dst_addr, (unsigned)fetch_reg(reg_num, regs) : 0)) << 32) | (unsigned)fetch_reg(reg_num + 1, regs); } else if (reg_num) { - src_val = fetch_reg_addr(reg_num, regs); + src_val_p = fetch_reg_addr(reg_num, regs); + } + src_val = *src_val_p; + if (unlikely(asi != orig_asi)) { + switch (size) { + case 2: + src_val = swab16(src_val); + break; + case 4: + src_val = swab32(src_val); + break; + case 8: + src_val = swab64(src_val); + break; + case 16: + default: + BUG(); + break; + }; } __do_int_store(dst_addr, size, src_val, asi); } @@ -276,6 +295,7 @@ asmlinkage void kernel_unaligned_trap(struct pt_regs *regs, unsigned int insn, u kernel_mna_trap_fault(); } else { unsigned long addr; + int orig_asi, asi; addr = compute_effective_address(regs, insn, ((insn >> 25) & 0x1f)); @@ -285,18 +305,48 @@ asmlinkage void kernel_unaligned_trap(struct pt_regs *regs, unsigned int insn, u regs->tpc, dirstrings[dir], addr, size, regs->u_regs[UREG_RETPC]); #endif + orig_asi = asi = decode_asi(insn, regs); + switch (asi) { + case ASI_NL: + case ASI_AIUPL: + case ASI_AIUSL: + case ASI_PL: + case ASI_SL: + case ASI_PNFL: + case ASI_SNFL: + asi &= ~0x08; + break; + }; switch (dir) { case load: do_int_load(fetch_reg_addr(((insn>>25)&0x1f), regs), size, (unsigned long *) addr, - decode_signedness(insn), - decode_asi(insn, regs)); + decode_signedness(insn), asi); + if (unlikely(asi != orig_asi)) { + unsigned long val_in = *(unsigned long *) addr; + switch (size) { + case 2: + val_in = swab16(val_in); + break; + case 4: + val_in = swab32(val_in); + break; + case 8: + val_in = swab64(val_in); + break; + case 16: + default: + BUG(); + break; + }; + *(unsigned long *) addr = val_in; + } break; case store: do_int_store(((insn>>25)&0x1f), size, (unsigned long *) addr, regs, - decode_asi(insn, regs)); + asi, orig_asi); break; default: From e0487992ce1dd7ae7da9c6aabdb19570bb95432b Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Mon, 19 Sep 2005 19:57:36 -0700 Subject: [PATCH 188/240] [BYTEORDER]: Document alignment and byteorder macros This patch comments the fact that although passing le64_to_cpup et al. is within the intended use of the byteorder macros, using get_unaligned is the recommended way to go. Signed-off-by: Ed L. Cashin Signed-off-by: David S. Miller --- include/linux/byteorder/generic.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/include/linux/byteorder/generic.h b/include/linux/byteorder/generic.h index 5fde6f4d6c1e..04bd756efc67 100644 --- a/include/linux/byteorder/generic.h +++ b/include/linux/byteorder/generic.h @@ -5,6 +5,10 @@ * linux/byteorder_generic.h * Generic Byte-reordering support * + * The "... p" macros, like le64_to_cpup, can be used with pointers + * to unaligned data, but there will be a performance penalty on + * some architectures. Use get_unaligned for unaligned data. + * * Francois-Rene Rideau 19970707 * gathered all the good ideas from all asm-foo/byteorder.h into one file, * cleaned them up. From 676d55ae30ea3b688f0386f70553489f25f24d55 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 19 Sep 2005 20:00:41 -0700 Subject: [PATCH 189/240] Linux v2.6.14-rc2 Avast, ye scurvy land-lubbers! Time to try out a new release. Arrr! --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 4e0d7c68d223..8cf6becf68dc 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 14 -EXTRAVERSION =-rc1 +EXTRAVERSION =-rc2 NAME=Affluent Albatross # *DOCUMENTATION* From 6a9b490d5fd7f23c5bcd75f970e01633ad3136e3 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 19 Sep 2005 20:11:57 -0700 Subject: [PATCH 190/240] [SPARC64]: Move DCACHE_ALIASING_POSSIBLE define to asm/page.h This showed that arch/sparc64/kernel/ptrace.c was not getting the define properly, and thus the code protected by this ifdef was never actually compiled before. So fix that too. Signed-off-by: David S. Miller --- arch/sparc64/kernel/ptrace.c | 7 ++++--- include/asm-sparc64/cacheflush.h | 7 ------- include/asm-sparc64/ide.h | 1 + include/asm-sparc64/page.h | 7 +++++++ include/asm-sparc64/pgalloc.h | 1 + 5 files changed, 13 insertions(+), 10 deletions(-) diff --git a/arch/sparc64/kernel/ptrace.c b/arch/sparc64/kernel/ptrace.c index 23ad839d113f..5efbff90d668 100644 --- a/arch/sparc64/kernel/ptrace.c +++ b/arch/sparc64/kernel/ptrace.c @@ -30,6 +30,7 @@ #include #include #include +#include /* Returning from ptrace is a bit tricky because the syscall return * low level code assumes any value returned which is negative and @@ -128,20 +129,20 @@ void flush_ptrace_access(struct vm_area_struct *vma, struct page *page, * is mapped to in the user's address space, we can skip the * D-cache flush. */ - if ((uaddr ^ kaddr) & (1UL << 13)) { + if ((uaddr ^ (unsigned long) kaddr) & (1UL << 13)) { unsigned long start = __pa(kaddr); unsigned long end = start + len; if (tlb_type == spitfire) { for (; start < end; start += 32) - spitfire_put_dcache_tag(va & 0x3fe0, 0x0); + spitfire_put_dcache_tag(start & 0x3fe0, 0x0); } else { for (; start < end; start += 32) __asm__ __volatile__( "stxa %%g0, [%0] %1\n\t" "membar #Sync" : /* no outputs */ - : "r" (va), + : "r" (start), "i" (ASI_DCACHE_INVALIDATE)); } } diff --git a/include/asm-sparc64/cacheflush.h b/include/asm-sparc64/cacheflush.h index 51b26e81d828..ededd2659eab 100644 --- a/include/asm-sparc64/cacheflush.h +++ b/include/asm-sparc64/cacheflush.h @@ -4,13 +4,6 @@ #include #include -/* Flushing for D-cache alias handling is only needed if - * the page size is smaller than 16K. - */ -#if PAGE_SHIFT < 14 -#define DCACHE_ALIASING_POSSIBLE -#endif - #ifndef __ASSEMBLY__ #include diff --git a/include/asm-sparc64/ide.h b/include/asm-sparc64/ide.h index 4c1098474c73..c393f815b0be 100644 --- a/include/asm-sparc64/ide.h +++ b/include/asm-sparc64/ide.h @@ -15,6 +15,7 @@ #include #include #include +#include #ifndef MAX_HWIFS # ifdef CONFIG_BLK_DEV_IDEPCI diff --git a/include/asm-sparc64/page.h b/include/asm-sparc64/page.h index c9f8ef208ea5..7f8d764abc47 100644 --- a/include/asm-sparc64/page.h +++ b/include/asm-sparc64/page.h @@ -21,6 +21,13 @@ #define PAGE_SIZE (_AC(1,UL) << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE-1)) +/* Flushing for D-cache alias handling is only needed if + * the page size is smaller than 16K. + */ +#if PAGE_SHIFT < 14 +#define DCACHE_ALIASING_POSSIBLE +#endif + #ifdef __KERNEL__ #ifndef __ASSEMBLY__ diff --git a/include/asm-sparc64/pgalloc.h b/include/asm-sparc64/pgalloc.h index b9b1914aae63..a96067cca963 100644 --- a/include/asm-sparc64/pgalloc.h +++ b/include/asm-sparc64/pgalloc.h @@ -10,6 +10,7 @@ #include #include #include +#include /* Page table allocation/freeing. */ #ifdef CONFIG_SMP From eed8b2dee7cff46dd4bf5b82dc53465d229162ba Mon Sep 17 00:00:00 2001 From: Anton Altaparmakov Date: Tue, 20 Sep 2005 14:19:30 +0100 Subject: [PATCH 191/240] NTFS: More runlist handling fixes from Richard Russon and myself. Signed-off-by: Anton Altaparmakov --- fs/ntfs/runlist.c | 55 ++++++++++++++++++++++++++++------------------- 1 file changed, 33 insertions(+), 22 deletions(-) diff --git a/fs/ntfs/runlist.c b/fs/ntfs/runlist.c index e2665d011d72..061b5ff6b73c 100644 --- a/fs/ntfs/runlist.c +++ b/fs/ntfs/runlist.c @@ -158,17 +158,21 @@ static inline BOOL ntfs_are_rl_mergeable(runlist_element *dst, BUG_ON(!dst); BUG_ON(!src); - if ((dst->lcn < 0) || (src->lcn < 0)) { /* Are we merging holes? */ - if (dst->lcn == LCN_HOLE && src->lcn == LCN_HOLE) - return TRUE; + /* We can merge unmapped regions even if they are misaligned. */ + if ((dst->lcn == LCN_RL_NOT_MAPPED) && (src->lcn == LCN_RL_NOT_MAPPED)) + return TRUE; + /* If the runs are misaligned, we cannot merge them. */ + if ((dst->vcn + dst->length) != src->vcn) return FALSE; - } - if ((dst->lcn + dst->length) != src->lcn) /* Are the runs contiguous? */ - return FALSE; - if ((dst->vcn + dst->length) != src->vcn) /* Are the runs misaligned? */ - return FALSE; - - return TRUE; + /* If both runs are non-sparse and contiguous, we can merge them. */ + if ((dst->lcn >= 0) && (src->lcn >= 0) && + ((dst->lcn + dst->length) == src->lcn)) + return TRUE; + /* If we are merging two holes, we can merge them. */ + if ((dst->lcn == LCN_HOLE) && (src->lcn == LCN_HOLE)) + return TRUE; + /* Cannot merge. */ + return FALSE; } /** @@ -214,14 +218,15 @@ static inline void __ntfs_rl_merge(runlist_element *dst, runlist_element *src) static inline runlist_element *ntfs_rl_append(runlist_element *dst, int dsize, runlist_element *src, int ssize, int loc) { - BOOL right; /* Right end of @src needs merging. */ - int marker; /* End of the inserted runs. */ + BOOL right = FALSE; /* Right end of @src needs merging. */ + int marker; /* End of the inserted runs. */ BUG_ON(!dst); BUG_ON(!src); /* First, check if the right hand end needs merging. */ - right = ntfs_are_rl_mergeable(src + ssize - 1, dst + loc + 1); + if ((loc + 1) < dsize) + right = ntfs_are_rl_mergeable(src + ssize - 1, dst + loc + 1); /* Space required: @dst size + @src size, less one if we merged. */ dst = ntfs_rl_realloc(dst, dsize, dsize + ssize - right); @@ -377,20 +382,21 @@ static inline runlist_element *ntfs_rl_replace(runlist_element *dst, int dsize, runlist_element *src, int ssize, int loc) { BOOL left = FALSE; /* Left end of @src needs merging. */ - BOOL right; /* Right end of @src needs merging. */ + BOOL right = FALSE; /* Right end of @src needs merging. */ int tail; /* Start of tail of @dst. */ int marker; /* End of the inserted runs. */ BUG_ON(!dst); BUG_ON(!src); - /* First, merge the left and right ends, if necessary. */ - right = ntfs_are_rl_mergeable(src + ssize - 1, dst + loc + 1); + /* First, see if the left and right ends need merging. */ + if ((loc + 1) < dsize) + right = ntfs_are_rl_mergeable(src + ssize - 1, dst + loc + 1); if (loc > 0) left = ntfs_are_rl_mergeable(dst + loc - 1, src); /* * Allocate some space. We will need less if the left, right, or both - * ends were merged. + * ends get merged. */ dst = ntfs_rl_realloc(dst, dsize, dsize + ssize - left - right); if (IS_ERR(dst)) @@ -399,21 +405,26 @@ static inline runlist_element *ntfs_rl_replace(runlist_element *dst, * We are guaranteed to succeed from here so can start modifying the * original runlists. */ + + /* First, merge the left and right ends, if necessary. */ if (right) __ntfs_rl_merge(src + ssize - 1, dst + loc + 1); if (left) __ntfs_rl_merge(dst + loc - 1, src); /* - * First run of @dst that needs to be moved out of the way to make - * space for the runs to be copied from @src, i.e. the first run of the - * tail of @dst. + * Offset of the tail of @dst. This needs to be moved out of the way + * to make space for the runs to be copied from @src, i.e. the first + * run of the tail of @dst. + * Nominally, @tail equals @loc + 1, i.e. location, skipping the + * replaced run. However, if @right, then one of @dst's runs is + * already merged into @src. */ tail = loc + right + 1; /* * First run after the @src runs that have been inserted, i.e. where * the tail of @dst needs to be moved to. - * Nominally, marker equals @loc + @ssize, i.e. location + number of - * runs in @src). However, if @left, then the first run in @src has + * Nominally, @marker equals @loc + @ssize, i.e. location + number of + * runs in @src. However, if @left, then the first run in @src has * been merged with one in @dst. */ marker = loc + ssize - left; From 9506057fca54464f3291b62156e6cd907c4cbc95 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Tue, 20 Sep 2005 16:20:49 +0100 Subject: [PATCH 192/240] [ARM] 2924/3: taglist - postfix section with .init for `make buildcheck` Patch from Ben Dooks The `make buildcheck` is erroneously reporting that the taglist is referencing items in the .init section as it is not itself postfixed with .init Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/kernel/vmlinux.lds.S | 2 +- include/asm-arm/setup.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index ad2d66c93a5c..350b53b41e5b 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -29,7 +29,7 @@ SECTIONS *(.arch.info) __arch_info_end = .; __tagtable_begin = .; - *(.taglist) + *(.taglist.init) __tagtable_end = .; . = ALIGN(16); __setup_start = .; diff --git a/include/asm-arm/setup.h b/include/asm-arm/setup.h index adcbd79762bf..a4b3ebf9bdd1 100644 --- a/include/asm-arm/setup.h +++ b/include/asm-arm/setup.h @@ -171,7 +171,7 @@ struct tagtable { int (*parse)(const struct tag *); }; -#define __tag __attribute_used__ __attribute__((__section__(".taglist"))) +#define __tag __attribute_used__ __attribute__((__section__(".taglist.init"))) #define __tagtable(tag, fn) \ static struct tagtable __tagtable_##fn __tag = { tag, fn } From 58dd48a6575d3ba86ba659eb8c6fc23246783fba Mon Sep 17 00:00:00 2001 From: Vincent Sanders Date: Tue, 20 Sep 2005 16:21:42 +0100 Subject: [PATCH 193/240] [ARM] 2922/1: compile fix for shark Patch from Vincent Sanders Shark platform fails to build with gcc 4 because of a bad lvalue assignement Signed-off-by: Vincent Sanders Signed-off-by: Russell King --- arch/arm/boot/compressed/ofw-shark.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/compressed/ofw-shark.c b/arch/arm/boot/compressed/ofw-shark.c index 7f6f5db0d060..465c54b6b128 100644 --- a/arch/arm/boot/compressed/ofw-shark.c +++ b/arch/arm/boot/compressed/ofw-shark.c @@ -256,5 +256,5 @@ asmlinkage void ofw_init(ofw_handle_t o, int *nomr, int *pointer) temp[11]='\0'; mem_len = OF_getproplen(o,phandle, temp); OF_getprop(o,phandle, temp, buffer, mem_len); - (unsigned char) pointer[32] = ((unsigned char *) buffer)[mem_len-2]; + * ((unsigned char *) &pointer[32]) = ((unsigned char *) buffer)[mem_len-2]; } From bfe6815e0465035d013b2b676444376fe2b3716e Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Tue, 20 Sep 2005 16:25:12 +0100 Subject: [PATCH 194/240] [ARM] 2925/3: earlyparam - postfix section with .init for `make buildcheck` Patch from Ben Dooks The `make buildcheck` is erroneously reporting that the earlyparam list is referencing items in the .init section as it is not itself postfixed with .init Also, as per rmk's suggestion, rename the __early_param to .early_param to bring it into line with everything else Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/kernel/vmlinux.lds.S | 2 +- include/asm-arm/setup.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 350b53b41e5b..47423aaf9d77 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -36,7 +36,7 @@ SECTIONS *(.init.setup) __setup_end = .; __early_begin = .; - *(__early_param) + *(.early_param.init) __early_end = .; __initcall_start = .; *(.initcall1.init) diff --git a/include/asm-arm/setup.h b/include/asm-arm/setup.h index a4b3ebf9bdd1..ea3ed2465233 100644 --- a/include/asm-arm/setup.h +++ b/include/asm-arm/setup.h @@ -213,6 +213,6 @@ struct early_params { #define __early_param(name,fn) \ static struct early_params __early_##fn __attribute_used__ \ -__attribute__((__section__("__early_param"))) = { name, fn } +__attribute__((__section__(".early_param.init"))) = { name, fn } #endif From 02b7dd1244aab9267ae4078e1ad6a2fdaabeb6ed Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Tue, 20 Sep 2005 16:35:03 +0100 Subject: [PATCH 195/240] [ARM] 2926/1: .proc.info - postfix section with .init for `make buildcheck` Patch from Ben Dooks The `make buildcheck` is erroneously reporting that the .proc.info list is referencing items in the .init section as it is not itself postfixed with .init Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/kernel/vmlinux.lds.S | 2 +- arch/arm/mm/proc-arm1020.S | 2 +- arch/arm/mm/proc-arm1020e.S | 2 +- arch/arm/mm/proc-arm1022.S | 2 +- arch/arm/mm/proc-arm1026.S | 2 +- arch/arm/mm/proc-arm6_7.S | 2 +- arch/arm/mm/proc-arm720.S | 2 +- arch/arm/mm/proc-arm920.S | 2 +- arch/arm/mm/proc-arm922.S | 2 +- arch/arm/mm/proc-arm925.S | 2 +- arch/arm/mm/proc-arm926.S | 2 +- arch/arm/mm/proc-sa110.S | 2 +- arch/arm/mm/proc-sa1100.S | 2 +- arch/arm/mm/proc-v6.S | 2 +- arch/arm/mm/proc-xscale.S | 2 +- 15 files changed, 15 insertions(+), 15 deletions(-) diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 47423aaf9d77..420b8d09a383 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -23,7 +23,7 @@ SECTIONS *(.init.text) _einittext = .; __proc_info_begin = .; - *(.proc.info) + *(.proc.info.init) __proc_info_end = .; __arch_info_begin = .; *(.arch.info) diff --git a/arch/arm/mm/proc-arm1020.S b/arch/arm/mm/proc-arm1020.S index 1d739d282a45..82ec954e45b6 100644 --- a/arch/arm/mm/proc-arm1020.S +++ b/arch/arm/mm/proc-arm1020.S @@ -509,7 +509,7 @@ cpu_arm1020_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __arm1020_proc_info,#object __arm1020_proc_info: diff --git a/arch/arm/mm/proc-arm1020e.S b/arch/arm/mm/proc-arm1020e.S index 9b725665b5c7..7375fe930f72 100644 --- a/arch/arm/mm/proc-arm1020e.S +++ b/arch/arm/mm/proc-arm1020e.S @@ -491,7 +491,7 @@ cpu_arm1020e_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __arm1020e_proc_info,#object __arm1020e_proc_info: diff --git a/arch/arm/mm/proc-arm1022.S b/arch/arm/mm/proc-arm1022.S index 37b70fa21c76..6ca639094d6f 100644 --- a/arch/arm/mm/proc-arm1022.S +++ b/arch/arm/mm/proc-arm1022.S @@ -473,7 +473,7 @@ cpu_arm1022_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __arm1022_proc_info,#object __arm1022_proc_info: diff --git a/arch/arm/mm/proc-arm1026.S b/arch/arm/mm/proc-arm1026.S index 931b690d1be2..10317e4f55d2 100644 --- a/arch/arm/mm/proc-arm1026.S +++ b/arch/arm/mm/proc-arm1026.S @@ -469,7 +469,7 @@ cpu_arm1026_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __arm1026_proc_info,#object __arm1026_proc_info: diff --git a/arch/arm/mm/proc-arm6_7.S b/arch/arm/mm/proc-arm6_7.S index d0f1bbb48f6c..8e7e1e70ab05 100644 --- a/arch/arm/mm/proc-arm6_7.S +++ b/arch/arm/mm/proc-arm6_7.S @@ -332,7 +332,7 @@ cpu_arm710_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __arm6_proc_info, #object __arm6_proc_info: diff --git a/arch/arm/mm/proc-arm720.S b/arch/arm/mm/proc-arm720.S index c69c9de32391..a13e0184d343 100644 --- a/arch/arm/mm/proc-arm720.S +++ b/arch/arm/mm/proc-arm720.S @@ -222,7 +222,7 @@ cpu_arm720_name: * See linux/include/asm-arm/procinfo.h for a definition of this structure. */ - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __arm710_proc_info, #object __arm710_proc_info: diff --git a/arch/arm/mm/proc-arm920.S b/arch/arm/mm/proc-arm920.S index 0f490a0fcb71..d16513899999 100644 --- a/arch/arm/mm/proc-arm920.S +++ b/arch/arm/mm/proc-arm920.S @@ -452,7 +452,7 @@ cpu_arm920_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __arm920_proc_info,#object __arm920_proc_info: diff --git a/arch/arm/mm/proc-arm922.S b/arch/arm/mm/proc-arm922.S index 62bc34a139ee..23b8ed97f4e3 100644 --- a/arch/arm/mm/proc-arm922.S +++ b/arch/arm/mm/proc-arm922.S @@ -456,7 +456,7 @@ cpu_arm922_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __arm922_proc_info,#object __arm922_proc_info: diff --git a/arch/arm/mm/proc-arm925.S b/arch/arm/mm/proc-arm925.S index ee49aa2ca781..ee95c52db513 100644 --- a/arch/arm/mm/proc-arm925.S +++ b/arch/arm/mm/proc-arm925.S @@ -521,7 +521,7 @@ cpu_arm925_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __arm925_proc_info,#object __arm925_proc_info: diff --git a/arch/arm/mm/proc-arm926.S b/arch/arm/mm/proc-arm926.S index bb95cc9fed03..7d042dc20c47 100644 --- a/arch/arm/mm/proc-arm926.S +++ b/arch/arm/mm/proc-arm926.S @@ -471,7 +471,7 @@ cpu_arm926_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __arm926_proc_info,#object __arm926_proc_info: diff --git a/arch/arm/mm/proc-sa110.S b/arch/arm/mm/proc-sa110.S index 34f7e7d3f419..bd330c4075a1 100644 --- a/arch/arm/mm/proc-sa110.S +++ b/arch/arm/mm/proc-sa110.S @@ -249,7 +249,7 @@ cpu_sa110_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __sa110_proc_info,#object __sa110_proc_info: diff --git a/arch/arm/mm/proc-sa1100.S b/arch/arm/mm/proc-sa1100.S index ca14f80d5ab4..91b89124c0d7 100644 --- a/arch/arm/mm/proc-sa1100.S +++ b/arch/arm/mm/proc-sa1100.S @@ -280,7 +280,7 @@ cpu_sa1110_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __sa1100_proc_info,#object __sa1100_proc_info: diff --git a/arch/arm/mm/proc-v6.S b/arch/arm/mm/proc-v6.S index eb34823c9dbf..caf3b19b167f 100644 --- a/arch/arm/mm/proc-v6.S +++ b/arch/arm/mm/proc-v6.S @@ -240,7 +240,7 @@ cpu_elf_name: .size cpu_elf_name, . - cpu_elf_name .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr /* * Match any ARMv6 processor core. diff --git a/arch/arm/mm/proc-xscale.S b/arch/arm/mm/proc-xscale.S index b88de2700146..861b35947280 100644 --- a/arch/arm/mm/proc-xscale.S +++ b/arch/arm/mm/proc-xscale.S @@ -578,7 +578,7 @@ cpu_pxa270_name: .align - .section ".proc.info", #alloc, #execinstr + .section ".proc.info.init", #alloc, #execinstr .type __80200_proc_info,#object __80200_proc_info: From 9d0fd1eb8a3c19f3ede5418540b3c9f64fac4b86 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Tue, 20 Sep 2005 16:45:20 +0100 Subject: [PATCH 196/240] [ARM] 2927/1: .arch.info - postfix section with .init for `make buildcheck` Patch from Ben Dooks The `make buildcheck` is erroneously reporting that the .arch.info list is referencing items in the .init section as it is not itself postfixed with .init Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/kernel/vmlinux.lds.S | 2 +- include/asm-arm/mach/arch.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 420b8d09a383..08e58ecd44be 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -26,7 +26,7 @@ SECTIONS *(.proc.info.init) __proc_info_end = .; __arch_info_begin = .; - *(.arch.info) + *(.arch.info.init) __arch_info_end = .; __tagtable_begin = .; *(.taglist.init) diff --git a/include/asm-arm/mach/arch.h b/include/asm-arm/mach/arch.h index 56c6bf4ab0c3..4fa95084a8c0 100644 --- a/include/asm-arm/mach/arch.h +++ b/include/asm-arm/mach/arch.h @@ -50,7 +50,7 @@ struct machine_desc { */ #define MACHINE_START(_type,_name) \ const struct machine_desc __mach_desc_##_type \ - __attribute__((__section__(".arch.info"))) = { \ + __attribute__((__section__(".arch.info.init"))) = { \ .nr = MACH_TYPE_##_type, \ .name = _name, From 5fe10ab19046d84f3fd243436cbd5fa01019e809 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Tue, 20 Sep 2005 17:24:33 +0100 Subject: [PATCH 197/240] [ARM] 2928/1: S3C2410 - make machine init code static Patch from Ben Dooks This code is not being exported, declare it static Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/mach-s3c2410/mach-anubis.c | 3 ++- arch/arm/mach-s3c2410/mach-bast.c | 3 ++- arch/arm/mach-s3c2410/mach-h1940.c | 7 ++++--- arch/arm/mach-s3c2410/mach-n30.c | 6 +++--- arch/arm/mach-s3c2410/mach-nexcoder.c | 2 +- arch/arm/mach-s3c2410/mach-otom.c | 2 +- arch/arm/mach-s3c2410/mach-rx3715.c | 5 +++-- arch/arm/mach-s3c2410/mach-smdk2410.c | 5 +++-- arch/arm/mach-s3c2410/mach-smdk2440.c | 5 +++-- arch/arm/mach-s3c2410/mach-vr1000.c | 3 ++- 10 files changed, 24 insertions(+), 17 deletions(-) diff --git a/arch/arm/mach-s3c2410/mach-anubis.c b/arch/arm/mach-s3c2410/mach-anubis.c index f87aa0b669ad..7c05f27fe1d6 100644 --- a/arch/arm/mach-s3c2410/mach-anubis.c +++ b/arch/arm/mach-s3c2410/mach-anubis.c @@ -12,6 +12,7 @@ * * Modifications: * 02-May-2005 BJD Copied from mach-bast.c + * 20-Sep-2005 BJD Added static to non-exported items */ #include @@ -232,7 +233,7 @@ static struct s3c24xx_board anubis_board __initdata = { .clocks_count = ARRAY_SIZE(anubis_clocks) }; -void __init anubis_map_io(void) +static void __init anubis_map_io(void) { /* initialise the clocks */ diff --git a/arch/arm/mach-s3c2410/mach-bast.c b/arch/arm/mach-s3c2410/mach-bast.c index 1a3367da6408..ed1f07d7252f 100644 --- a/arch/arm/mach-s3c2410/mach-bast.c +++ b/arch/arm/mach-s3c2410/mach-bast.c @@ -31,6 +31,7 @@ * 17-Jul-2005 BJD Changed to platform device for SuperIO 16550s * 25-Jul-2005 BJD Removed ASIX static mappings * 27-Jul-2005 BJD Ensure maximum frequency of i2c bus + * 20-Sep-2005 BJD Added static to non-exported items */ #include @@ -428,7 +429,7 @@ static struct s3c24xx_board bast_board __initdata = { .clocks_count = ARRAY_SIZE(bast_clocks) }; -void __init bast_map_io(void) +static void __init bast_map_io(void) { /* initialise the clocks */ diff --git a/arch/arm/mach-s3c2410/mach-h1940.c b/arch/arm/mach-s3c2410/mach-h1940.c index 6ff1889fbd21..fb3cb01266e5 100644 --- a/arch/arm/mach-s3c2410/mach-h1940.c +++ b/arch/arm/mach-s3c2410/mach-h1940.c @@ -24,6 +24,7 @@ * 10-Jan-2005 BJD Removed include of s3c2410.h * 14-Jan-2005 BJD Added clock init * 10-Mar-2005 LCVR Changed S3C2410_VA to S3C24XX_VA + * 20-Sep-2005 BJD Added static to non-exported items */ #include @@ -147,7 +148,7 @@ static struct s3c24xx_board h1940_board __initdata = { .devices_count = ARRAY_SIZE(h1940_devices) }; -void __init h1940_map_io(void) +static void __init h1940_map_io(void) { s3c24xx_init_io(h1940_iodesc, ARRAY_SIZE(h1940_iodesc)); s3c24xx_init_clocks(0); @@ -155,13 +156,13 @@ void __init h1940_map_io(void) s3c24xx_set_board(&h1940_board); } -void __init h1940_init_irq(void) +static void __init h1940_init_irq(void) { s3c24xx_init_irq(); } -void __init h1940_init(void) +static void __init h1940_init(void) { set_s3c2410fb_info(&h1940_lcdcfg); } diff --git a/arch/arm/mach-s3c2410/mach-n30.c b/arch/arm/mach-s3c2410/mach-n30.c index 66bf5bb2b3db..5c0f2b091f95 100644 --- a/arch/arm/mach-s3c2410/mach-n30.c +++ b/arch/arm/mach-s3c2410/mach-n30.c @@ -97,7 +97,7 @@ static struct s3c24xx_board n30_board __initdata = { .devices_count = ARRAY_SIZE(n30_devices) }; -void __init n30_map_io(void) +static void __init n30_map_io(void) { s3c24xx_init_io(n30_iodesc, ARRAY_SIZE(n30_iodesc)); s3c24xx_init_clocks(0); @@ -105,14 +105,14 @@ void __init n30_map_io(void) s3c24xx_set_board(&n30_board); } -void __init n30_init_irq(void) +static void __init n30_init_irq(void) { s3c24xx_init_irq(); } /* GPB3 is the line that controls the pull-up for the USB D+ line */ -void __init n30_init(void) +static void __init n30_init(void) { s3c_device_i2c.dev.platform_data = &n30_i2ccfg; diff --git a/arch/arm/mach-s3c2410/mach-nexcoder.c b/arch/arm/mach-s3c2410/mach-nexcoder.c index d24c242414ca..c22f8216032d 100644 --- a/arch/arm/mach-s3c2410/mach-nexcoder.c +++ b/arch/arm/mach-s3c2410/mach-nexcoder.c @@ -136,7 +136,7 @@ static void __init nexcoder_sensorboard_init(void) s3c2410_gpio_cfgpin(S3C2410_GPF2, S3C2410_GPF2_OUTP); // CAM_GPIO6 => CAM_PWRDN } -void __init nexcoder_map_io(void) +static void __init nexcoder_map_io(void) { s3c24xx_init_io(nexcoder_iodesc, ARRAY_SIZE(nexcoder_iodesc)); s3c24xx_init_clocks(0); diff --git a/arch/arm/mach-s3c2410/mach-otom.c b/arch/arm/mach-s3c2410/mach-otom.c index d901ed492ff5..ad1459e402e2 100644 --- a/arch/arm/mach-s3c2410/mach-otom.c +++ b/arch/arm/mach-s3c2410/mach-otom.c @@ -105,7 +105,7 @@ static struct s3c24xx_board otom11_board __initdata = { }; -void __init otom11_map_io(void) +static void __init otom11_map_io(void) { s3c24xx_init_io(otom11_iodesc, ARRAY_SIZE(otom11_iodesc)); s3c24xx_init_clocks(0); diff --git a/arch/arm/mach-s3c2410/mach-rx3715.c b/arch/arm/mach-s3c2410/mach-rx3715.c index a73d61c1de46..22d9e070fd68 100644 --- a/arch/arm/mach-s3c2410/mach-rx3715.c +++ b/arch/arm/mach-s3c2410/mach-rx3715.c @@ -16,6 +16,7 @@ * 14-Jan-2005 BJD Added new clock init * 10-Mar-2005 LCVR Changed S3C2410_VA to S3C24XX_VA * 14-Mar-2005 BJD Fixed __iomem warnings + * 20-Sep-2005 BJD Added static to non-exported items */ #include @@ -108,7 +109,7 @@ static struct s3c24xx_board rx3715_board __initdata = { .devices_count = ARRAY_SIZE(rx3715_devices) }; -void __init rx3715_map_io(void) +static void __init rx3715_map_io(void) { s3c24xx_init_io(rx3715_iodesc, ARRAY_SIZE(rx3715_iodesc)); s3c24xx_init_clocks(16934000); @@ -116,7 +117,7 @@ void __init rx3715_map_io(void) s3c24xx_set_board(&rx3715_board); } -void __init rx3715_init_irq(void) +static void __init rx3715_init_irq(void) { s3c24xx_init_irq(); } diff --git a/arch/arm/mach-s3c2410/mach-smdk2410.c b/arch/arm/mach-s3c2410/mach-smdk2410.c index 67e903a700d3..2eda55a6b678 100644 --- a/arch/arm/mach-s3c2410/mach-smdk2410.c +++ b/arch/arm/mach-s3c2410/mach-smdk2410.c @@ -28,6 +28,7 @@ * Ben Dooks * * 10-Mar-2005 LCVR Changed S3C2410_VA to S3C24XX_VA + * 20-Sep-2005 BJD Added static to non-exported items * ***********************************************************************/ @@ -97,7 +98,7 @@ static struct s3c24xx_board smdk2410_board __initdata = { .devices_count = ARRAY_SIZE(smdk2410_devices) }; -void __init smdk2410_map_io(void) +static void __init smdk2410_map_io(void) { s3c24xx_init_io(smdk2410_iodesc, ARRAY_SIZE(smdk2410_iodesc)); s3c24xx_init_clocks(0); @@ -105,7 +106,7 @@ void __init smdk2410_map_io(void) s3c24xx_set_board(&smdk2410_board); } -void __init smdk2410_init_irq(void) +static void __init smdk2410_init_irq(void) { s3c24xx_init_irq(); } diff --git a/arch/arm/mach-s3c2410/mach-smdk2440.c b/arch/arm/mach-s3c2410/mach-smdk2440.c index 357522106f68..722ef46b630a 100644 --- a/arch/arm/mach-s3c2410/mach-smdk2440.c +++ b/arch/arm/mach-s3c2410/mach-smdk2440.c @@ -18,6 +18,7 @@ * 22-Feb-2005 BJD Updated for 2.6.11-rc5 relesa * 10-Mar-2005 LCVR Replaced S3C2410_VA by S3C24XX_VA * 14-Mar-2005 BJD void __iomem fixes + * 20-Sep-2005 BJD Added static to non-exported items */ #include @@ -98,7 +99,7 @@ static struct s3c24xx_board smdk2440_board __initdata = { .devices_count = ARRAY_SIZE(smdk2440_devices) }; -void __init smdk2440_map_io(void) +static void __init smdk2440_map_io(void) { s3c24xx_init_io(smdk2440_iodesc, ARRAY_SIZE(smdk2440_iodesc)); s3c24xx_init_clocks(16934400); @@ -106,7 +107,7 @@ void __init smdk2440_map_io(void) s3c24xx_set_board(&smdk2440_board); } -void __init smdk2440_machine_init(void) +static void __init smdk2440_machine_init(void) { /* Configure the LEDs (even if we have no LED support)*/ diff --git a/arch/arm/mach-s3c2410/mach-vr1000.c b/arch/arm/mach-s3c2410/mach-vr1000.c index 8f9ab2893df4..663a7f98fc0b 100644 --- a/arch/arm/mach-s3c2410/mach-vr1000.c +++ b/arch/arm/mach-s3c2410/mach-vr1000.c @@ -28,6 +28,7 @@ * 10-Mar-2005 LCVR Changed S3C2410_VA to S3C24XX_VA * 14-Mar-2006 BJD void __iomem fixes * 22-Jun-2006 BJD Added DM9000 platform information + * 20-Sep-2005 BJD Added static to non-exported items */ #include @@ -347,7 +348,7 @@ static void vr1000_power_off(void) s3c2410_gpio_setpin(S3C2410_GPB9, 1); } -void __init vr1000_map_io(void) +static void __init vr1000_map_io(void) { /* initialise clock sources */ From 840ff6a4f6174d7fe19c206b5f36ff64123a2f45 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 20 Sep 2005 17:52:13 +0100 Subject: [PATCH 198/240] [ARM] Prevent deadlock in page fault handler As per x86, we may deadlock while trying to get the mmap semaphore. Implement the same fix, which allows (eg) recursive faults to cause an oops instead of deadlocking. Signed-off-by: Russell King --- arch/arm/mm/fault.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c index 0b6c4db44e08..4a884baf3b9c 100644 --- a/arch/arm/mm/fault.c +++ b/arch/arm/mm/fault.c @@ -233,7 +233,17 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs) if (in_interrupt() || !mm) goto no_context; - down_read(&mm->mmap_sem); + /* + * As per x86, we may deadlock here. However, since the kernel only + * validly references user space from well defined areas of the code, + * we can bug out early if this is from code which shouldn't. + */ + if (!down_read_trylock(&mm->mmap_sem)) { + if (!user_mode(regs) && !search_exception_tables(regs->ARM_pc)) + goto no_context; + down_read(&mm->mmap_sem); + } + fault = __do_page_fault(mm, addr, fsr, tsk); up_read(&mm->mmap_sem); From 8d2cae0651502028bf64844508ab18528bbd65c2 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Tue, 20 Sep 2005 10:52:04 -0700 Subject: [PATCH 199/240] [PATCH] IPoIB: Don't flush workqueue from within workqueue ipoib_mcast_restart_task() is always called from within the single-threaded IPoIB workqueue, so flushing the workqueue from within the function can lead to a recursion overflow. But since we're running in a single-threaded workqueue, we're already synchronized against other items in the workqueue, so just get rid of the flush in ipoib_mcast_restart_task(). Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/ipoib/ipoib.h | 2 +- drivers/infiniband/ulp/ipoib/ipoib_ib.c | 4 ++-- drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 7 ++++--- 3 files changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index bea960b8191f..4ea1c1ca85bc 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -257,7 +257,7 @@ void ipoib_mcast_send(struct net_device *dev, union ib_gid *mgid, void ipoib_mcast_restart_task(void *dev_ptr); int ipoib_mcast_start_thread(struct net_device *dev); -int ipoib_mcast_stop_thread(struct net_device *dev); +int ipoib_mcast_stop_thread(struct net_device *dev, int flush); void ipoib_mcast_dev_down(struct net_device *dev); void ipoib_mcast_dev_flush(struct net_device *dev); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index ef0e3894863c..f7440096b5ed 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -432,7 +432,7 @@ int ipoib_ib_dev_down(struct net_device *dev) flush_workqueue(ipoib_workqueue); } - ipoib_mcast_stop_thread(dev); + ipoib_mcast_stop_thread(dev, 1); /* * Flush the multicast groups first so we stop any multicast joins. The @@ -599,7 +599,7 @@ void ipoib_ib_dev_cleanup(struct net_device *dev) ipoib_dbg(priv, "cleaning up ib_dev\n"); - ipoib_mcast_stop_thread(dev); + ipoib_mcast_stop_thread(dev, 1); /* Delete the broadcast address and the local address */ ipoib_mcast_dev_down(dev); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 3aab4d85abfc..36ce29836bf2 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -598,7 +598,7 @@ int ipoib_mcast_start_thread(struct net_device *dev) return 0; } -int ipoib_mcast_stop_thread(struct net_device *dev) +int ipoib_mcast_stop_thread(struct net_device *dev, int flush) { struct ipoib_dev_priv *priv = netdev_priv(dev); struct ipoib_mcast *mcast; @@ -610,7 +610,8 @@ int ipoib_mcast_stop_thread(struct net_device *dev) cancel_delayed_work(&priv->mcast_task); up(&mcast_mutex); - flush_workqueue(ipoib_workqueue); + if (flush) + flush_workqueue(ipoib_workqueue); if (priv->broadcast && priv->broadcast->query) { ib_sa_cancel_query(priv->broadcast->query_id, priv->broadcast->query); @@ -832,7 +833,7 @@ void ipoib_mcast_restart_task(void *dev_ptr) ipoib_dbg_mcast(priv, "restarting multicast task\n"); - ipoib_mcast_stop_thread(dev); + ipoib_mcast_stop_thread(dev, 0); spin_lock_irqsave(&priv->lock, flags); From 71eea47d853bb0ce0c6befe11b3e08111263170f Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Tue, 20 Sep 2005 10:54:48 -0700 Subject: [PATCH 200/240] [PATCH] IB/mthca: Fix device removal memory leak Clean up QP table array on device removal. Signed-off-by: Michael S. Tsirkin Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mthca/mthca_qp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c index eaa521857630..5fa00669f9b8 100644 --- a/drivers/infiniband/hw/mthca/mthca_qp.c +++ b/drivers/infiniband/hw/mthca/mthca_qp.c @@ -2123,5 +2123,6 @@ void __devexit mthca_cleanup_qp_table(struct mthca_dev *dev) for (i = 0; i < 2; ++i) mthca_CONF_SPECIAL_QP(dev, i, 0, &status); + mthca_array_cleanup(&dev->qp_table.qp, dev->limits.num_qps); mthca_alloc_cleanup(&dev->qp_table.alloc); } From 729b4f7de68191478b20fab19a6d0c6b8c4380c9 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 20 Sep 2005 12:18:38 -0700 Subject: [PATCH 201/240] [SPARC64]: Verify vmalloc TLB misses more strictly. Arrange the modules, OBP, and vmalloc areas such that a range verification can be done quite minimally. Signed-off-by: David S. Miller --- arch/sparc64/kernel/entry.S | 41 +++++++++++++++++------------------ include/asm-sparc64/pgtable.h | 20 +++++++++-------- 2 files changed, 31 insertions(+), 30 deletions(-) diff --git a/arch/sparc64/kernel/entry.S b/arch/sparc64/kernel/entry.S index 3e0badb820c5..b48349527853 100644 --- a/arch/sparc64/kernel/entry.S +++ b/arch/sparc64/kernel/entry.S @@ -42,19 +42,15 @@ * executing (see inherit_locked_prom_mappings() rant). */ sparc64_vpte_nucleus: - /* Load 0xf0000000, which is LOW_OBP_ADDRESS. */ - mov 0xf, %g5 - sllx %g5, 28, %g5 - - /* Is addr >= LOW_OBP_ADDRESS? */ + /* Note that kvmap below has verified that the address is + * in the range MODULES_VADDR --> VMALLOC_END already. So + * here we need only check if it is an OBP address or not. + */ + sethi %hi(LOW_OBP_ADDRESS), %g5 cmp %g4, %g5 blu,pn %xcc, sparc64_vpte_patchme1 mov 0x1, %g5 - - /* Load 0x100000000, which is HI_OBP_ADDRESS. */ sllx %g5, 32, %g5 - - /* Is addr < HI_OBP_ADDRESS? */ cmp %g4, %g5 blu,pn %xcc, obp_iaddr_patch nop @@ -156,26 +152,29 @@ obp_daddr_patch: * rather, use information saved during inherit_prom_mappings() using 8k * pagesize. */ + .align 32 kvmap: - /* Load 0xf0000000, which is LOW_OBP_ADDRESS. */ - mov 0xf, %g5 - sllx %g5, 28, %g5 - - /* Is addr >= LOW_OBP_ADDRESS? */ + sethi %hi(MODULES_VADDR), %g5 cmp %g4, %g5 - blu,pn %xcc, vmalloc_addr + blu,pn %xcc, longpath + mov (VMALLOC_END >> 24), %g5 + sllx %g5, 24, %g5 + cmp %g4, %g5 + bgeu,pn %xcc, longpath + nop + +kvmap_check_obp: + sethi %hi(LOW_OBP_ADDRESS), %g5 + cmp %g4, %g5 + blu,pn %xcc, kvmap_vmalloc_addr mov 0x1, %g5 - - /* Load 0x100000000, which is HI_OBP_ADDRESS. */ sllx %g5, 32, %g5 - - /* Is addr < HI_OBP_ADDRESS? */ cmp %g4, %g5 blu,pn %xcc, obp_daddr_patch nop -vmalloc_addr: - /* If we get here, a vmalloc addr accessed, load kernel VPTE. */ +kvmap_vmalloc_addr: + /* If we get here, a vmalloc addr was accessed, load kernel VPTE. */ ldxa [%g3 + %g6] ASI_N, %g5 brgez,pn %g5, longpath nop diff --git a/include/asm-sparc64/pgtable.h b/include/asm-sparc64/pgtable.h index a2b4f5ed4625..a297f6144f0f 100644 --- a/include/asm-sparc64/pgtable.h +++ b/include/asm-sparc64/pgtable.h @@ -24,21 +24,23 @@ #include #include -/* The kernel image occupies 0x4000000 to 0x1000000 (4MB --> 16MB). - * The page copy blockops use 0x1000000 to 0x18000000 (16MB --> 24MB). +/* The kernel image occupies 0x4000000 to 0x1000000 (4MB --> 32MB). + * The page copy blockops can use 0x2000000 to 0x10000000. * The PROM resides in an area spanning 0xf0000000 to 0x100000000. - * The vmalloc area spans 0x140000000 to 0x200000000. + * The vmalloc area spans 0x100000000 to 0x200000000. + * Since modules need to be in the lowest 32-bits of the address space, + * we place them right before the OBP area from 0x10000000 to 0xf0000000. * There is a single static kernel PMD which maps from 0x0 to address * 0x400000000. */ -#define TLBTEMP_BASE _AC(0x0000000001000000,UL) -#define MODULES_VADDR _AC(0x0000000002000000,UL) -#define MODULES_LEN _AC(0x000000007e000000,UL) -#define MODULES_END _AC(0x0000000080000000,UL) -#define VMALLOC_START _AC(0x0000000140000000,UL) -#define VMALLOC_END _AC(0x0000000200000000,UL) +#define TLBTEMP_BASE _AC(0x0000000002000000,UL) +#define MODULES_VADDR _AC(0x0000000010000000,UL) +#define MODULES_LEN _AC(0x00000000e0000000,UL) +#define MODULES_END _AC(0x00000000f0000000,UL) #define LOW_OBP_ADDRESS _AC(0x00000000f0000000,UL) #define HI_OBP_ADDRESS _AC(0x0000000100000000,UL) +#define VMALLOC_START _AC(0x0000000100000000,UL) +#define VMALLOC_END _AC(0x0000000200000000,UL) /* XXX All of this needs to be rethought so we can take advantage * XXX cheetah's full 64-bit virtual address space, ie. no more hole From a9f7baf4d695dbe3028d40ca374c1d11499eaee6 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 20 Sep 2005 21:01:13 +0100 Subject: [PATCH 202/240] [ARM] Fix pcf8583 to build Seems that the Acorn RTC driver missed an update. Fix it. Signed-off-by: Russell King --- drivers/acorn/char/pcf8583.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/acorn/char/pcf8583.c b/drivers/acorn/char/pcf8583.c index 141b4c237a50..2b850e5860a0 100644 --- a/drivers/acorn/char/pcf8583.c +++ b/drivers/acorn/char/pcf8583.c @@ -23,12 +23,13 @@ static struct i2c_driver pcf8583_driver; static unsigned short ignore[] = { I2C_CLIENT_END }; static unsigned short normal_addr[] = { 0x50, I2C_CLIENT_END }; +static unsigned short *forces[] = { NULL }; static struct i2c_client_address_data addr_data = { .normal_i2c = normal_addr, .probe = ignore, .ignore = ignore, - .force = ignore, + .forces = forces, }; #define DAT(x) ((unsigned int)(x->dev.driver_data)) From 7a482425f45e695a2e2821b32a887865f016fbf9 Mon Sep 17 00:00:00 2001 From: "Antonino A. Daplas" Date: Wed, 21 Sep 2005 07:30:21 +0800 Subject: [PATCH 203/240] [PATCH] nvidiafb: Fix absence of cursor in nvidiafb A recent change in nvidiafb caused nvidiafb_cursor to always return -ENXIO instead of using the soft_cursor. This will happen if the parameter "hwcur" is not set, which happens to be the default. Signed-off-by: Antonino Daplas Signed-off-by: Linus Torvalds --- drivers/video/Kconfig | 1 + drivers/video/nvidia/nvidia.c | 5 ++++- 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig index 31ee13eef7af..773ae11b4a19 100644 --- a/drivers/video/Kconfig +++ b/drivers/video/Kconfig @@ -650,6 +650,7 @@ config FB_NVIDIA select FB_CFB_FILLRECT select FB_CFB_COPYAREA select FB_CFB_IMAGEBLIT + select FB_SOFT_CURSOR help This driver supports graphics boards with the nVidia chips, TNT and newer. For very old chipsets, such as the RIVA128, then use diff --git a/drivers/video/nvidia/nvidia.c b/drivers/video/nvidia/nvidia.c index 3620de0f252e..a7f020ada630 100644 --- a/drivers/video/nvidia/nvidia.c +++ b/drivers/video/nvidia/nvidia.c @@ -893,7 +893,7 @@ static int nvidiafb_cursor(struct fb_info *info, struct fb_cursor *cursor) int i, set = cursor->set; u16 fg, bg; - if (!hwcur || cursor->image.width > MAX_CURS || cursor->image.height > MAX_CURS) + if (cursor->image.width > MAX_CURS || cursor->image.height > MAX_CURS) return -ENXIO; NVShowHideCursor(par, 0); @@ -1356,6 +1356,9 @@ static int __devinit nvidia_set_fbinfo(struct fb_info *info) info->pixmap.size = 8 * 1024; info->pixmap.flags = FB_PIXMAP_SYSTEM; + if (!hwcur) + info->fbops->fb_cursor = soft_cursor; + info->var.accel_flags = (!noaccel); switch (par->Architecture) { From 6d67e34de5a378de2f461137944c5d931283d557 Mon Sep 17 00:00:00 2001 From: "Antonino A. Daplas" Date: Wed, 21 Sep 2005 07:32:49 +0800 Subject: [PATCH 204/240] [PATCH] fbdev: Fix reversed back and front porches In fbdev perspective, the frontporch is the lower/right margin and the backporch is the upper/left margin. Correct. Signed-off-by: Antonino Daplas Signed-off-by: Linus Torvalds --- drivers/video/fbcvt.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/video/fbcvt.c b/drivers/video/fbcvt.c index cfa61b512de0..0b6af00d197e 100644 --- a/drivers/video/fbcvt.c +++ b/drivers/video/fbcvt.c @@ -272,11 +272,11 @@ static void fb_cvt_convert_to_mode(struct fb_cvt_data *cvt, { mode->refresh = cvt->f_refresh; mode->pixclock = KHZ2PICOS(cvt->pixclock/1000); - mode->left_margin = cvt->h_front_porch; - mode->right_margin = cvt->h_back_porch; + mode->left_margin = cvt->h_back_porch; + mode->right_margin = cvt->h_front_porch; mode->hsync_len = cvt->hsync; - mode->upper_margin = cvt->v_front_porch; - mode->lower_margin = cvt->v_back_porch; + mode->upper_margin = cvt->v_back_porch; + mode->lower_margin = cvt->v_front_porch; mode->vsync_len = cvt->vsync; mode->sync &= ~(FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT); From 0fb375fb9b93b7d822debc6a734052337ccfdb1f Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 21 Sep 2005 00:11:37 -0700 Subject: [PATCH 205/240] [AF_PACKET]: Allow for > 8 byte hardware addresses. The convention is that longer addresses will simply extend the hardeware address byte arrays at the end of sockaddr_ll and packet_mreq. In making this change a small information leak was also closed. The code only initializes the hardware address bytes that are used, but all of struct sockaddr_ll was copied to userspace. Now we just copy sockaddr_ll to the last byte of the hardware address used. For error checking larger structures than our internal maximums continue to be allowed but an error is signaled if we can not fit the hardware address into our internal structure. Signed-off-by: Eric W. Biederman Signed-off-by: David S. Miller --- net/packet/af_packet.c | 65 +++++++++++++++++++++++++++++++----------- 1 file changed, 48 insertions(+), 17 deletions(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 8690f171c1ef..ee865d88183b 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -36,6 +36,11 @@ * Michal Ostrowski : Module initialization cleanup. * Ulises Alonso : Frame number limit removal and * packet_set_ring memory leak. + * Eric Biederman : Allow for > 8 byte hardware addresses. + * The convention is that longer addresses + * will simply extend the hardware address + * byte arrays at the end of sockaddr_ll + * and packet_mreq. * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -161,7 +166,17 @@ struct packet_mclist int count; unsigned short type; unsigned short alen; - unsigned char addr[8]; + unsigned char addr[MAX_ADDR_LEN]; +}; +/* identical to struct packet_mreq except it has + * a longer address field. + */ +struct packet_mreq_max +{ + int mr_ifindex; + unsigned short mr_type; + unsigned short mr_alen; + unsigned char mr_address[MAX_ADDR_LEN]; }; #endif #ifdef CONFIG_PACKET_MMAP @@ -716,6 +731,8 @@ static int packet_sendmsg(struct kiocb *iocb, struct socket *sock, err = -EINVAL; if (msg->msg_namelen < sizeof(struct sockaddr_ll)) goto out; + if (msg->msg_namelen < (saddr->sll_halen + offsetof(struct sockaddr_ll, sll_addr))) + goto out; ifindex = saddr->sll_ifindex; proto = saddr->sll_protocol; addr = saddr->sll_addr; @@ -744,6 +761,12 @@ static int packet_sendmsg(struct kiocb *iocb, struct socket *sock, if (dev->hard_header) { int res; err = -EINVAL; + if (saddr) { + if (saddr->sll_halen != dev->addr_len) + goto out_free; + if (saddr->sll_hatype != dev->type) + goto out_free; + } res = dev->hard_header(skb, dev, ntohs(proto), addr, NULL, len); if (sock->type != SOCK_DGRAM) { skb->tail = skb->data; @@ -1045,6 +1068,7 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock, struct sock *sk = sock->sk; struct sk_buff *skb; int copied, err; + struct sockaddr_ll *sll; err = -EINVAL; if (flags & ~(MSG_PEEK|MSG_DONTWAIT|MSG_TRUNC|MSG_CMSG_COMPAT)) @@ -1056,16 +1080,6 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock, return -ENODEV; #endif - /* - * If the address length field is there to be filled in, we fill - * it in now. - */ - - if (sock->type == SOCK_PACKET) - msg->msg_namelen = sizeof(struct sockaddr_pkt); - else - msg->msg_namelen = sizeof(struct sockaddr_ll); - /* * Call the generic datagram receiver. This handles all sorts * of horrible races and re-entrancy so we can forget about it @@ -1086,6 +1100,17 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock, if(skb==NULL) goto out; + /* + * If the address length field is there to be filled in, we fill + * it in now. + */ + + sll = (struct sockaddr_ll*)skb->cb; + if (sock->type == SOCK_PACKET) + msg->msg_namelen = sizeof(struct sockaddr_pkt); + else + msg->msg_namelen = sll->sll_halen + offsetof(struct sockaddr_ll, sll_addr); + /* * You lose any data beyond the buffer you gave. If it worries a * user program they can ask the device for its MTU anyway. @@ -1166,7 +1191,7 @@ static int packet_getname(struct socket *sock, struct sockaddr *uaddr, sll->sll_hatype = 0; /* Bad: we have no ARPHRD_UNSPEC */ sll->sll_halen = 0; } - *uaddr_len = sizeof(*sll); + *uaddr_len = offsetof(struct sockaddr_ll, sll_addr) + sll->sll_halen; return 0; } @@ -1199,7 +1224,7 @@ static void packet_dev_mclist(struct net_device *dev, struct packet_mclist *i, i } } -static int packet_mc_add(struct sock *sk, struct packet_mreq *mreq) +static int packet_mc_add(struct sock *sk, struct packet_mreq_max *mreq) { struct packet_sock *po = pkt_sk(sk); struct packet_mclist *ml, *i; @@ -1249,7 +1274,7 @@ done: return err; } -static int packet_mc_drop(struct sock *sk, struct packet_mreq *mreq) +static int packet_mc_drop(struct sock *sk, struct packet_mreq_max *mreq) { struct packet_mclist *ml, **mlp; @@ -1315,11 +1340,17 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv case PACKET_ADD_MEMBERSHIP: case PACKET_DROP_MEMBERSHIP: { - struct packet_mreq mreq; - if (optlen sizeof(mreq)) + len = sizeof(mreq); + if (copy_from_user(&mreq,optval,len)) return -EFAULT; + if (len < (mreq.mr_alen + offsetof(struct packet_mreq, mr_address))) + return -EINVAL; if (optname == PACKET_ADD_MEMBERSHIP) ret = packet_mc_add(sk, &mreq); else From 78c6671a88313fd3c4364dc46e8c8186612616b8 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 21 Sep 2005 00:15:39 -0700 Subject: [PATCH 206/240] [FIB_TRIE]: message cleanup Cleanup the printk's in fib_trie: * Convert a couple of places in the dump code to BUG_ON * Put log level's on each message The version message really needed the message since it leaks out on the pretty Fedora bootup. Signed-off-by: Stephen Hemminger Acked-by: Robert Olsson , Signed-off-by: David S. Miller --- net/ipv4/fib_trie.c | 15 +++------------ 1 file changed, 3 insertions(+), 12 deletions(-) diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 90ae70870a10..50c0519cd70d 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -1086,7 +1086,7 @@ fib_insert_node(struct trie *t, int *err, u32 key, int plen) } if (tp && tp->pos + tp->bits > 32) - printk("ERROR tp=%p pos=%d, bits=%d, key=%0x plen=%d\n", + printk(KERN_WARNING "fib_trie tp=%p pos=%d, bits=%d, key=%0x plen=%d\n", tp, tp->pos, tp->bits, key, plen); /* Rebalance the trie */ @@ -1832,16 +1832,7 @@ static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah, struct fi i++; continue; } - if (fa->fa_info->fib_nh == NULL) { - printk("Trie error _fib_nh=NULL in fa[%d] k=%08x plen=%d\n", i, key, plen); - i++; - continue; - } - if (fa->fa_info == NULL) { - printk("Trie error fa_info=NULL in fa[%d] k=%08x plen=%d\n", i, key, plen); - i++; - continue; - } + BUG_ON(!fa->fa_info); if (fib_dump_info(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, @@ -1964,7 +1955,7 @@ struct fib_table * __init fib_hash_init(int id) trie_main = t; if (id == RT_TABLE_LOCAL) - printk("IPv4 FIB: Using LC-trie version %s\n", VERSION); + printk(KERN_INFO "IPv4 FIB: Using LC-trie version %s\n", VERSION); return tb; } From 7957aed72b104f7528da484f292442c6f3a131a1 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 21 Sep 2005 00:19:46 -0700 Subject: [PATCH 207/240] [TCP]: Set default congestion control correctly for incoming connections. Patch from Joel Sing to fix the default congestion control algorithm for incoming connections. If a new congestion control handler is added (via module), it should become the default for new connections. Instead, the incoming connections use reno. The cause is incorrect initialisation causes the tcp_init_congestion_control() function to return after the initial if test fails. Signed-off-by: Stephen Hemminger Acked-by: Ian McDonald Signed-off-by: David S. Miller --- net/ipv4/tcp_minisocks.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index a88db28b0af7..b1a63b2c6b4a 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -384,7 +384,7 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, newtp->frto_counter = 0; newtp->frto_highmark = 0; - newicsk->icsk_ca_ops = &tcp_reno; + newicsk->icsk_ca_ops = &tcp_init_congestion_ops; tcp_set_ca_state(newsk, TCP_CA_Open); tcp_init_xmit_timers(newsk); From 1a3e64aa480101d11d8cac94ca57fd8bbe2536b7 Mon Sep 17 00:00:00 2001 From: Tom 'spot' Callaway Date: Wed, 21 Sep 2005 00:29:16 -0700 Subject: [PATCH 208/240] [ATYFB]: Fix build with CONFIG_FB_ATY_GENERIC_LCD disabled. Signed-off-by: Tom 'spot' Callaway Signed-off-by: David S. Miller --- drivers/video/aty/xlinit.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/video/aty/xlinit.c b/drivers/video/aty/xlinit.c index 92643af12581..0bea0d8d7821 100644 --- a/drivers/video/aty/xlinit.c +++ b/drivers/video/aty/xlinit.c @@ -174,7 +174,7 @@ int atyfb_xl_init(struct fb_info *info) const struct xl_card_cfg_t * card = &card_cfg[xl_card]; struct atyfb_par *par = (struct atyfb_par *) info->par; union aty_pll pll; - int i, err; + int err; u32 temp; aty_st_8(CONFIG_STAT0, 0x85, par); @@ -252,9 +252,12 @@ int atyfb_xl_init(struct fb_info *info) aty_st_le32(0xEC, 0x00000000, par); aty_st_le32(0xFC, 0x00000000, par); +#if defined (CONFIG_FB_ATY_GENERIC_LCD) + int i; for (i=0; i Date: Wed, 21 Sep 2005 00:41:22 -0700 Subject: [PATCH 209/240] [SPARC] cs4231: Fix SBUS support in this driver. From: Christopher Zimmermann This patch enables SBus support for the cs4231 sound driver. It is tested on an Ultra2. Capture and playback both work. I experienced lags and crashes using certain threaded players like ogg123 and mp3blaster, while the former is lagging far more. This behavior may be specific to SMP systems. It is reproducable using the dummy sound card driver. Sox works flawlessly. Setting up the calculation of ptr in snd_cs4231_playback_pointer was a bit strange. I got it to work by not incrementing the [pc]_periods_sent counter when starting DMA the first time in cs4231_dma_trigger. Therefore this dummy thing. [ I did some minor cleanups -DaveM ] Signed-off-by: David S. Miller --- sound/sparc/cs4231.c | 323 ++++++++++++++++++++++--------------------- 1 file changed, 166 insertions(+), 157 deletions(-) diff --git a/sound/sparc/cs4231.c b/sound/sparc/cs4231.c index 2fb27c4e951f..f4361c518e46 100644 --- a/sound/sparc/cs4231.c +++ b/sound/sparc/cs4231.c @@ -173,7 +173,7 @@ static cs4231_t *cs4231_list; #define CS4231_GLOBALIRQ 0x01 /* IRQ is active */ -/* definitions for codec irq status */ +/* definitions for codec irq status - CS4231_IRQ_STATUS */ #define CS4231_PLAYBACK_IRQ 0x10 #define CS4231_RECORD_IRQ 0x20 @@ -402,7 +402,7 @@ static void snd_cs4231_outm(cs4231_t *chip, unsigned char reg, udelay(100); #ifdef CONFIG_SND_DEBUG if (__cs4231_readb(chip, CS4231P(chip, REGSEL)) & CS4231_INIT) - snd_printk("outm: auto calibration time out - reg = 0x%x, value = 0x%x\n", reg, value); + snd_printdd("outm: auto calibration time out - reg = 0x%x, value = 0x%x\n", reg, value); #endif if (chip->calibrate_mute) { chip->image[reg] &= mask; @@ -425,6 +425,10 @@ static void snd_cs4231_dout(cs4231_t *chip, unsigned char reg, unsigned char val timeout > 0 && (__cs4231_readb(chip, CS4231P(chip, REGSEL)) & CS4231_INIT); timeout--) udelay(100); +#ifdef CONFIG_SND_DEBUG + if (__cs4231_readb(chip, CS4231P(chip, REGSEL)) & CS4231_INIT) + snd_printdd("out: auto calibration time out - reg = 0x%x, value = 0x%x\n", reg, value); +#endif __cs4231_writeb(chip, chip->mce_bit | reg, CS4231P(chip, REGSEL)); __cs4231_writeb(chip, value, CS4231P(chip, REG)); mb(); @@ -440,15 +444,12 @@ static void snd_cs4231_out(cs4231_t *chip, unsigned char reg, unsigned char valu udelay(100); #ifdef CONFIG_SND_DEBUG if (__cs4231_readb(chip, CS4231P(chip, REGSEL)) & CS4231_INIT) - snd_printk("out: auto calibration time out - reg = 0x%x, value = 0x%x\n", reg, value); + snd_printdd("out: auto calibration time out - reg = 0x%x, value = 0x%x\n", reg, value); #endif __cs4231_writeb(chip, chip->mce_bit | reg, CS4231P(chip, REGSEL)); __cs4231_writeb(chip, value, CS4231P(chip, REG)); chip->image[reg] = value; mb(); -#if 0 - printk("codec out - reg 0x%x = 0x%x\n", chip->mce_bit | reg, value); -#endif } static unsigned char snd_cs4231_in(cs4231_t *chip, unsigned char reg) @@ -462,61 +463,14 @@ static unsigned char snd_cs4231_in(cs4231_t *chip, unsigned char reg) udelay(100); #ifdef CONFIG_SND_DEBUG if (__cs4231_readb(chip, CS4231P(chip, REGSEL)) & CS4231_INIT) - snd_printk("in: auto calibration time out - reg = 0x%x\n", reg); + snd_printdd("in: auto calibration time out - reg = 0x%x\n", reg); #endif __cs4231_writeb(chip, chip->mce_bit | reg, CS4231P(chip, REGSEL)); mb(); ret = __cs4231_readb(chip, CS4231P(chip, REG)); -#if 0 - printk("codec in - reg 0x%x = 0x%x\n", chip->mce_bit | reg, ret); -#endif return ret; } -#if 0 - -static void snd_cs4231_debug(cs4231_t *chip) -{ - printk("CS4231 REGS: INDEX = 0x%02x ", - __cs4231_readb(chip, CS4231P(chip, REGSEL))); - printk(" STATUS = 0x%02x\n", - __cs4231_readb(chip, CS4231P(chip, STATUS))); - printk(" 0x00: left input = 0x%02x ", snd_cs4231_in(chip, 0x00)); - printk(" 0x10: alt 1 (CFIG 2) = 0x%02x\n", snd_cs4231_in(chip, 0x10)); - printk(" 0x01: right input = 0x%02x ", snd_cs4231_in(chip, 0x01)); - printk(" 0x11: alt 2 (CFIG 3) = 0x%02x\n", snd_cs4231_in(chip, 0x11)); - printk(" 0x02: GF1 left input = 0x%02x ", snd_cs4231_in(chip, 0x02)); - printk(" 0x12: left line in = 0x%02x\n", snd_cs4231_in(chip, 0x12)); - printk(" 0x03: GF1 right input = 0x%02x ", snd_cs4231_in(chip, 0x03)); - printk(" 0x13: right line in = 0x%02x\n", snd_cs4231_in(chip, 0x13)); - printk(" 0x04: CD left input = 0x%02x ", snd_cs4231_in(chip, 0x04)); - printk(" 0x14: timer low = 0x%02x\n", snd_cs4231_in(chip, 0x14)); - printk(" 0x05: CD right input = 0x%02x ", snd_cs4231_in(chip, 0x05)); - printk(" 0x15: timer high = 0x%02x\n", snd_cs4231_in(chip, 0x15)); - printk(" 0x06: left output = 0x%02x ", snd_cs4231_in(chip, 0x06)); - printk(" 0x16: left MIC (PnP) = 0x%02x\n", snd_cs4231_in(chip, 0x16)); - printk(" 0x07: right output = 0x%02x ", snd_cs4231_in(chip, 0x07)); - printk(" 0x17: right MIC (PnP) = 0x%02x\n", snd_cs4231_in(chip, 0x17)); - printk(" 0x08: playback format = 0x%02x ", snd_cs4231_in(chip, 0x08)); - printk(" 0x18: IRQ status = 0x%02x\n", snd_cs4231_in(chip, 0x18)); - printk(" 0x09: iface (CFIG 1) = 0x%02x ", snd_cs4231_in(chip, 0x09)); - printk(" 0x19: left line out = 0x%02x\n", snd_cs4231_in(chip, 0x19)); - printk(" 0x0a: pin control = 0x%02x ", snd_cs4231_in(chip, 0x0a)); - printk(" 0x1a: mono control = 0x%02x\n", snd_cs4231_in(chip, 0x1a)); - printk(" 0x0b: init & status = 0x%02x ", snd_cs4231_in(chip, 0x0b)); - printk(" 0x1b: right line out = 0x%02x\n", snd_cs4231_in(chip, 0x1b)); - printk(" 0x0c: revision & mode = 0x%02x ", snd_cs4231_in(chip, 0x0c)); - printk(" 0x1c: record format = 0x%02x\n", snd_cs4231_in(chip, 0x1c)); - printk(" 0x0d: loopback = 0x%02x ", snd_cs4231_in(chip, 0x0d)); - printk(" 0x1d: var freq (PnP) = 0x%02x\n", snd_cs4231_in(chip, 0x1d)); - printk(" 0x0e: ply upr count = 0x%02x ", snd_cs4231_in(chip, 0x0e)); - printk(" 0x1e: rec upr count = 0x%02x\n", snd_cs4231_in(chip, 0x1e)); - printk(" 0x0f: ply lwr count = 0x%02x ", snd_cs4231_in(chip, 0x0f)); - printk(" 0x1f: rec lwr count = 0x%02x\n", snd_cs4231_in(chip, 0x1f)); -} - -#endif - /* * CS4231 detection / MCE routines */ @@ -528,11 +482,12 @@ static void snd_cs4231_busy_wait(cs4231_t *chip) /* huh.. looks like this sequence is proper for CS4231A chip (GUS MAX) */ for (timeout = 5; timeout > 0; timeout--) __cs4231_readb(chip, CS4231P(chip, REGSEL)); + /* end of cleanup sequence */ - for (timeout = 250; + for (timeout = 500; timeout > 0 && (__cs4231_readb(chip, CS4231P(chip, REGSEL)) & CS4231_INIT); timeout--) - udelay(100); + udelay(1000); } static void snd_cs4231_mce_up(cs4231_t *chip) @@ -545,12 +500,12 @@ static void snd_cs4231_mce_up(cs4231_t *chip) udelay(100); #ifdef CONFIG_SND_DEBUG if (__cs4231_readb(chip, CS4231P(chip, REGSEL)) & CS4231_INIT) - snd_printk("mce_up - auto calibration time out (0)\n"); + snd_printdd("mce_up - auto calibration time out (0)\n"); #endif chip->mce_bit |= CS4231_MCE; timeout = __cs4231_readb(chip, CS4231P(chip, REGSEL)); if (timeout == 0x80) - snd_printk("mce_up [%p]: serious init problem - codec still busy\n", chip->port); + snd_printdd("mce_up [%p]: serious init problem - codec still busy\n", chip->port); if (!(timeout & CS4231_MCE)) __cs4231_writeb(chip, chip->mce_bit | (timeout & 0x1f), CS4231P(chip, REGSEL)); spin_unlock_irqrestore(&chip->lock, flags); @@ -563,18 +518,15 @@ static void snd_cs4231_mce_down(cs4231_t *chip) spin_lock_irqsave(&chip->lock, flags); snd_cs4231_busy_wait(chip); -#if 0 - printk("(1) timeout = %i\n", timeout); -#endif #ifdef CONFIG_SND_DEBUG if (__cs4231_readb(chip, CS4231P(chip, REGSEL)) & CS4231_INIT) - snd_printk("mce_down [%p] - auto calibration time out (0)\n", CS4231P(chip, REGSEL)); + snd_printdd("mce_down [%p] - auto calibration time out (0)\n", CS4231P(chip, REGSEL)); #endif chip->mce_bit &= ~CS4231_MCE; timeout = __cs4231_readb(chip, CS4231P(chip, REGSEL)); __cs4231_writeb(chip, chip->mce_bit | (timeout & 0x1f), CS4231P(chip, REGSEL)); if (timeout == 0x80) - snd_printk("mce_down [%p]: serious init problem - codec still busy\n", chip->port); + snd_printdd("mce_down [%p]: serious init problem - codec still busy\n", chip->port); if ((timeout & CS4231_MCE) == 0) { spin_unlock_irqrestore(&chip->lock, flags); return; @@ -590,9 +542,7 @@ static void snd_cs4231_mce_down(cs4231_t *chip) spin_unlock_irqrestore(&chip->lock, flags); return; } -#if 0 - printk("(2) timeout = %i, jiffies = %li\n", timeout, jiffies); -#endif + /* in 10ms increments, check condition, up to 250ms */ timeout = 25; while (snd_cs4231_in(chip, CS4231_TEST_INIT) & CS4231_CALIB_IN_PROGRESS) { @@ -604,9 +554,7 @@ static void snd_cs4231_mce_down(cs4231_t *chip) msleep(10); spin_lock_irqsave(&chip->lock, flags); } -#if 0 - printk("(3) jiffies = %li\n", jiffies); -#endif + /* in 10ms increments, check condition, up to 100ms */ timeout = 10; while (__cs4231_readb(chip, CS4231P(chip, REGSEL)) & CS4231_INIT) { @@ -619,54 +567,58 @@ static void snd_cs4231_mce_down(cs4231_t *chip) spin_lock_irqsave(&chip->lock, flags); } spin_unlock_irqrestore(&chip->lock, flags); -#if 0 - printk("(4) jiffies = %li\n", jiffies); - snd_printk("mce_down - exit = 0x%x\n", __cs4231_readb(chip, CS4231P(chip, REGSEL))); -#endif } -#if 0 /* Unused for now... */ -static unsigned int snd_cs4231_get_count(unsigned char format, unsigned int size) -{ - switch (format & 0xe0) { - case CS4231_LINEAR_16: - case CS4231_LINEAR_16_BIG: - size >>= 1; - break; - case CS4231_ADPCM_16: - return size >> 2; - } - if (format & CS4231_STEREO) - size >>= 1; - return size; -} -#endif - #ifdef EBUS_SUPPORT static void snd_cs4231_ebus_advance_dma(struct ebus_dma_info *p, snd_pcm_substream_t *substream, unsigned int *periods_sent) { snd_pcm_runtime_t *runtime = substream->runtime; while (1) { - unsigned int dma_size = snd_pcm_lib_period_bytes(substream); - unsigned int offset = dma_size * (*periods_sent); + unsigned int period_size = snd_pcm_lib_period_bytes(substream); + unsigned int offset = period_size * (*periods_sent); - if (dma_size >= (1 << 24)) + if (period_size >= (1 << 24)) BUG(); - if (ebus_dma_request(p, runtime->dma_addr + offset, dma_size)) + if (ebus_dma_request(p, runtime->dma_addr + offset, period_size)) return; -#if 0 - printk("ebus_advance: Sent period %u (size[%x] offset[%x])\n", - (*periods_sent), dma_size, offset); -#endif (*periods_sent) = ((*periods_sent) + 1) % runtime->periods; } } #endif -static void cs4231_dma_trigger(cs4231_t *chip, unsigned int what, int on) +#ifdef SBUS_SUPPORT +static void snd_cs4231_sbus_advance_dma(snd_pcm_substream_t *substream, unsigned int *periods_sent) { + cs4231_t *chip = snd_pcm_substream_chip(substream); + snd_pcm_runtime_t *runtime = substream->runtime; + + unsigned int period_size = snd_pcm_lib_period_bytes(substream); + unsigned int offset = period_size * (*periods_sent % runtime->periods); + + if (runtime->period_size > 0xffff + 1) + BUG(); + + switch (substream->stream) { + case SNDRV_PCM_STREAM_PLAYBACK: + sbus_writel(runtime->dma_addr + offset, chip->port + APCPNVA); + sbus_writel(period_size, chip->port + APCPNC); + break; + case SNDRV_PCM_STREAM_CAPTURE: + sbus_writel(runtime->dma_addr + offset, chip->port + APCCNVA); + sbus_writel(period_size, chip->port + APCCNC); + break; + } + + (*periods_sent) = (*periods_sent + 1) % runtime->periods; +} +#endif + +static void cs4231_dma_trigger(snd_pcm_substream_t *substream, unsigned int what, int on) +{ + cs4231_t *chip = snd_pcm_substream_chip(substream); + #ifdef EBUS_SUPPORT if (chip->flags & CS4231_FLAG_EBUS) { if (what & CS4231_PLAYBACK_ENABLE) { @@ -694,6 +646,60 @@ static void cs4231_dma_trigger(cs4231_t *chip, unsigned int what, int on) } else { #endif #ifdef SBUS_SUPPORT + u32 csr = sbus_readl(chip->port + APCCSR); + /* I don't know why, but on sbus the period counter must + * only start counting after the first period is sent. + * Therefore this dummy thing. + */ + unsigned int dummy = 0; + + switch (what) { + case CS4231_PLAYBACK_ENABLE: + if (on) { + csr &= ~APC_XINT_PLAY; + sbus_writel(csr, chip->port + APCCSR); + + csr &= ~APC_PPAUSE; + sbus_writel(csr, chip->port + APCCSR); + + snd_cs4231_sbus_advance_dma(substream, &dummy); + + csr |= APC_GENL_INT | APC_PLAY_INT | APC_XINT_ENA | + APC_XINT_PLAY | APC_XINT_EMPT | APC_XINT_GENL | + APC_XINT_PENA | APC_PDMA_READY; + sbus_writel(csr, chip->port + APCCSR); + } else { + csr |= APC_PPAUSE; + sbus_writel(csr, chip->port + APCCSR); + + csr &= ~APC_PDMA_READY; + sbus_writel(csr, chip->port + APCCSR); + } + break; + case CS4231_RECORD_ENABLE: + if (on) { + csr &= ~APC_XINT_CAPT; + sbus_writel(csr, chip->port + APCCSR); + + csr &= ~APC_CPAUSE; + sbus_writel(csr, chip->port + APCCSR); + + snd_cs4231_sbus_advance_dma(substream, &dummy); + + csr |= APC_GENL_INT | APC_CAPT_INT | APC_XINT_ENA | + APC_XINT_CAPT | APC_XINT_CEMP | APC_XINT_GENL | + APC_CDMA_READY; + + sbus_writel(csr, chip->port + APCCSR); + } else { + csr |= APC_CPAUSE; + sbus_writel(csr, chip->port + APCCSR); + + csr &= ~APC_CDMA_READY; + sbus_writel(csr, chip->port + APCCSR); + } + break; + } #endif #ifdef EBUS_SUPPORT } @@ -725,25 +731,12 @@ static int snd_cs4231_trigger(snd_pcm_substream_t *substream, int cmd) } } -#if 0 - printk("TRIGGER: what[%x] on(%d)\n", - what, (cmd == SNDRV_PCM_TRIGGER_START)); -#endif - spin_lock_irqsave(&chip->lock, flags); if (cmd == SNDRV_PCM_TRIGGER_START) { - cs4231_dma_trigger(chip, what, 1); + cs4231_dma_trigger(substream, what, 1); chip->image[CS4231_IFACE_CTRL] |= what; - if (what & CS4231_PLAYBACK_ENABLE) { - snd_cs4231_out(chip, CS4231_PLY_LWR_CNT, 0xff); - snd_cs4231_out(chip, CS4231_PLY_UPR_CNT, 0xff); - } - if (what & CS4231_RECORD_ENABLE) { - snd_cs4231_out(chip, CS4231_REC_LWR_CNT, 0xff); - snd_cs4231_out(chip, CS4231_REC_UPR_CNT, 0xff); - } } else { - cs4231_dma_trigger(chip, what, 0); + cs4231_dma_trigger(substream, what, 0); chip->image[CS4231_IFACE_CTRL] &= ~what; } snd_cs4231_out(chip, CS4231_IFACE_CTRL, @@ -755,9 +748,7 @@ static int snd_cs4231_trigger(snd_pcm_substream_t *substream, int cmd) result = -EINVAL; break; } -#if 0 - snd_cs4231_debug(chip); -#endif + return result; } @@ -790,9 +781,6 @@ static unsigned char snd_cs4231_get_format(cs4231_t *chip, int format, int chann } if (channels > 1) rformat |= CS4231_STEREO; -#if 0 - snd_printk("get_format: 0x%x (mode=0x%x)\n", format, mode); -#endif return rformat; } @@ -944,7 +932,7 @@ static void snd_cs4231_init(cs4231_t *chip) snd_cs4231_mce_down(chip); #ifdef SNDRV_DEBUG_MCE - snd_printk("init: (1)\n"); + snd_printdd("init: (1)\n"); #endif snd_cs4231_mce_up(chip); spin_lock_irqsave(&chip->lock, flags); @@ -957,7 +945,7 @@ static void snd_cs4231_init(cs4231_t *chip) snd_cs4231_mce_down(chip); #ifdef SNDRV_DEBUG_MCE - snd_printk("init: (2)\n"); + snd_printdd("init: (2)\n"); #endif snd_cs4231_mce_up(chip); @@ -967,7 +955,7 @@ static void snd_cs4231_init(cs4231_t *chip) snd_cs4231_mce_down(chip); #ifdef SNDRV_DEBUG_MCE - snd_printk("init: (3) - afei = 0x%x\n", chip->image[CS4231_ALT_FEATURE_1]); + snd_printdd("init: (3) - afei = 0x%x\n", chip->image[CS4231_ALT_FEATURE_1]); #endif spin_lock_irqsave(&chip->lock, flags); @@ -981,7 +969,7 @@ static void snd_cs4231_init(cs4231_t *chip) snd_cs4231_mce_down(chip); #ifdef SNDRV_DEBUG_MCE - snd_printk("init: (4)\n"); + snd_printdd("init: (4)\n"); #endif snd_cs4231_mce_up(chip); @@ -991,7 +979,7 @@ static void snd_cs4231_init(cs4231_t *chip) snd_cs4231_mce_down(chip); #ifdef SNDRV_DEBUG_MCE - snd_printk("init: (5)\n"); + snd_printdd("init: (5)\n"); #endif } @@ -1022,6 +1010,7 @@ static int snd_cs4231_open(cs4231_t *chip, unsigned int mode) CS4231_RECORD_IRQ | CS4231_TIMER_IRQ); snd_cs4231_out(chip, CS4231_IRQ_STATUS, 0); + spin_unlock_irqrestore(&chip->lock, flags); chip->mode = mode; @@ -1136,11 +1125,21 @@ static int snd_cs4231_playback_hw_free(snd_pcm_substream_t *substream) static int snd_cs4231_playback_prepare(snd_pcm_substream_t *substream) { cs4231_t *chip = snd_pcm_substream_chip(substream); + snd_pcm_runtime_t *runtime = substream->runtime; unsigned long flags; spin_lock_irqsave(&chip->lock, flags); + chip->image[CS4231_IFACE_CTRL] &= ~(CS4231_PLAYBACK_ENABLE | CS4231_PLAYBACK_PIO); + + if (runtime->period_size > 0xffff + 1) + BUG(); + + snd_cs4231_out(chip, CS4231_PLY_LWR_CNT, (runtime->period_size - 1) & 0x00ff); + snd_cs4231_out(chip, CS4231_PLY_UPR_CNT, (runtime->period_size - 1) >> 8 & 0x00ff); + chip->p_periods_sent = 0; + spin_unlock_irqrestore(&chip->lock, flags); return 0; @@ -1172,12 +1171,16 @@ static int snd_cs4231_capture_hw_free(snd_pcm_substream_t *substream) static int snd_cs4231_capture_prepare(snd_pcm_substream_t *substream) { cs4231_t *chip = snd_pcm_substream_chip(substream); + snd_pcm_runtime_t *runtime = substream->runtime; unsigned long flags; spin_lock_irqsave(&chip->lock, flags); chip->image[CS4231_IFACE_CTRL] &= ~(CS4231_RECORD_ENABLE | CS4231_RECORD_PIO); + snd_cs4231_out(chip, CS4231_REC_LWR_CNT, (runtime->period_size - 1) & 0x00ff); + snd_cs4231_out(chip, CS4231_REC_LWR_CNT, (runtime->period_size - 1) >> 8 & 0x00ff); + spin_unlock_irqrestore(&chip->lock, flags); return 0; @@ -1196,53 +1199,61 @@ static void snd_cs4231_overrange(cs4231_t *chip) chip->capture_substream->runtime->overrange++; } -static void snd_cs4231_generic_interrupt(cs4231_t *chip) +static irqreturn_t snd_cs4231_generic_interrupt(cs4231_t *chip) { unsigned long flags; unsigned char status; + /*This is IRQ is not raised by the cs4231*/ + if (!(__cs4231_readb(chip, CS4231P(chip, STATUS)) & CS4231_GLOBALIRQ)) + return IRQ_NONE; + status = snd_cs4231_in(chip, CS4231_IRQ_STATUS); - if (!status) - return; if (status & CS4231_TIMER_IRQ) { if (chip->timer) snd_timer_interrupt(chip->timer, chip->timer->sticks); } - if (status & CS4231_PLAYBACK_IRQ) - snd_pcm_period_elapsed(chip->playback_substream); - if (status & CS4231_RECORD_IRQ) { + + if (status & CS4231_RECORD_IRQ) snd_cs4231_overrange(chip); - snd_pcm_period_elapsed(chip->capture_substream); - } /* ACK the CS4231 interrupt. */ spin_lock_irqsave(&chip->lock, flags); snd_cs4231_outm(chip, CS4231_IRQ_STATUS, ~CS4231_ALL_IRQS | ~status, 0); spin_unlock_irqrestore(&chip->lock, flags); + + return 0; } #ifdef SBUS_SUPPORT static irqreturn_t snd_cs4231_sbus_interrupt(int irq, void *dev_id, struct pt_regs *regs) { cs4231_t *chip = dev_id; - u32 csr; - - csr = sbus_readl(chip->port + APCCSR); - if (!(csr & (APC_INT_PENDING | - APC_PLAY_INT | - APC_CAPT_INT | - APC_GENL_INT | - APC_XINT_PEMP | - APC_XINT_CEMP))) - return IRQ_NONE; /* ACK the APC interrupt. */ + u32 csr = sbus_readl(chip->port + APCCSR); + sbus_writel(csr, chip->port + APCCSR); - snd_cs4231_generic_interrupt(chip); + if ((chip->image[CS4231_IFACE_CTRL] & CS4231_PLAYBACK_ENABLE) && + (csr & APC_PLAY_INT) && + (csr & APC_XINT_PNVA) && + !(csr & APC_XINT_EMPT)) { + snd_cs4231_sbus_advance_dma(chip->playback_substream, + &chip->p_periods_sent); + snd_pcm_period_elapsed(chip->playback_substream); + } - return IRQ_HANDLED; + if ((chip->image[CS4231_IFACE_CTRL] & CS4231_RECORD_ENABLE) && + (csr & APC_CAPT_INT) && + (csr & APC_XINT_CNVA)) { + snd_cs4231_sbus_advance_dma(chip->capture_substream, + &chip->c_periods_sent); + snd_pcm_period_elapsed(chip->capture_substream); + } + + return snd_cs4231_generic_interrupt(chip); } #endif @@ -1290,7 +1301,8 @@ static snd_pcm_uframes_t snd_cs4231_playback_pointer(snd_pcm_substream_t *substr #ifdef EBUS_SUPPORT } #endif - ptr += (period_bytes - residue); + ptr += period_bytes - residue; + return bytes_to_frames(substream->runtime, ptr); } @@ -1314,7 +1326,7 @@ static snd_pcm_uframes_t snd_cs4231_capture_pointer(snd_pcm_substream_t * substr #ifdef EBUS_SUPPORT } #endif - ptr += (period_bytes - residue); + ptr += period_bytes - residue; return bytes_to_frames(substream->runtime, ptr); } @@ -1328,9 +1340,6 @@ static int snd_cs4231_probe(cs4231_t *chip) int i, id, vers; unsigned char *ptr; -#if 0 - snd_cs4231_debug(chip); -#endif id = vers = 0; for (i = 0; i < 50; i++) { mb(); @@ -1985,13 +1994,13 @@ static int __init snd_cs4231_sbus_create(snd_card_t *card, chip->port = sbus_ioremap(&sdev->resource[0], 0, chip->regs_size, "cs4231"); if (!chip->port) { - snd_printk("cs4231-%d: Unable to map chip registers.\n", dev); + snd_printdd("cs4231-%d: Unable to map chip registers.\n", dev); return -EIO; } if (request_irq(sdev->irqs[0], snd_cs4231_sbus_interrupt, SA_SHIRQ, "cs4231", chip)) { - snd_printk("cs4231-%d: Unable to grab SBUS IRQ %s\n", + snd_printdd("cs4231-%d: Unable to grab SBUS IRQ %s\n", dev, __irq_itoa(sdev->irqs[0])); snd_cs4231_sbus_free(chip); @@ -2113,29 +2122,29 @@ static int __init snd_cs4231_ebus_create(snd_card_t *card, chip->eb2c.regs = ioremap(edev->resource[2].start, 0x10); if (!chip->port || !chip->eb2p.regs || !chip->eb2c.regs) { snd_cs4231_ebus_free(chip); - snd_printk("cs4231-%d: Unable to map chip registers.\n", dev); + snd_printdd("cs4231-%d: Unable to map chip registers.\n", dev); return -EIO; } if (ebus_dma_register(&chip->eb2c)) { snd_cs4231_ebus_free(chip); - snd_printk("cs4231-%d: Unable to register EBUS capture DMA\n", dev); + snd_printdd("cs4231-%d: Unable to register EBUS capture DMA\n", dev); return -EBUSY; } if (ebus_dma_irq_enable(&chip->eb2c, 1)) { snd_cs4231_ebus_free(chip); - snd_printk("cs4231-%d: Unable to enable EBUS capture IRQ\n", dev); + snd_printdd("cs4231-%d: Unable to enable EBUS capture IRQ\n", dev); return -EBUSY; } if (ebus_dma_register(&chip->eb2p)) { snd_cs4231_ebus_free(chip); - snd_printk("cs4231-%d: Unable to register EBUS play DMA\n", dev); + snd_printdd("cs4231-%d: Unable to register EBUS play DMA\n", dev); return -EBUSY; } if (ebus_dma_irq_enable(&chip->eb2p, 1)) { snd_cs4231_ebus_free(chip); - snd_printk("cs4231-%d: Unable to enable EBUS play IRQ\n", dev); + snd_printdd("cs4231-%d: Unable to enable EBUS play IRQ\n", dev); return -EBUSY; } From d79e743e9fcf03f521300a970eb1ab69641910ba Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Wed, 21 Sep 2005 14:14:22 +1000 Subject: [PATCH 210/240] [PATCH] ppc64: Fix PCI flags when using OF device tree My code to set up the PCI tree from the Open Firmware device tree was setting IORESOURCE_* flags on the resources for the devices, but not the PCI_BASE_ADDRESS_* flags. This meant that some drivers misbehaved, and /proc/pci showed the wrong types for the resources. This fixes it. Signed-off-by: Paul Mackerras Signed-off-by: Linus Torvalds --- arch/ppc64/kernel/pci.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/arch/ppc64/kernel/pci.c b/arch/ppc64/kernel/pci.c index 861138ad092c..ff4be1da69d5 100644 --- a/arch/ppc64/kernel/pci.c +++ b/arch/ppc64/kernel/pci.c @@ -246,11 +246,14 @@ static unsigned int pci_parse_of_flags(u32 addr0) unsigned int flags = 0; if (addr0 & 0x02000000) { - flags |= IORESOURCE_MEM; + flags = IORESOURCE_MEM | PCI_BASE_ADDRESS_SPACE_MEMORY; + flags |= (addr0 >> 22) & PCI_BASE_ADDRESS_MEM_TYPE_64; + flags |= (addr0 >> 28) & PCI_BASE_ADDRESS_MEM_TYPE_1M; if (addr0 & 0x40000000) - flags |= IORESOURCE_PREFETCH; + flags |= IORESOURCE_PREFETCH + | PCI_BASE_ADDRESS_MEM_PREFETCH; } else if (addr0 & 0x01000000) - flags |= IORESOURCE_IO; + flags = IORESOURCE_IO | PCI_BASE_ADDRESS_SPACE_IO; return flags; } From 66dcca062847bcd261ebb3ac96d51101f31a8630 Mon Sep 17 00:00:00 2001 From: Sripathi Kodi Date: Mon, 19 Sep 2005 18:26:12 -0500 Subject: [PATCH 211/240] [PATCH] Fix invisible threads problem When the main thread of a thread group has done pthread_exit() and died, the other threads are still happily running, but will not be visible under /proc because their leader is no longer accessible. This fixes the access control so that we can see the sub-threads again. Signed-off-by: Sripathi Kodi Acked-by: Al Viro Signed-off-by: Linus Torvalds --- fs/proc/base.c | 84 +++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 77 insertions(+), 7 deletions(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index 23db452ab428..fb34f88a4a74 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -340,6 +340,52 @@ static int proc_root_link(struct inode *inode, struct dentry **dentry, struct vf return result; } + +/* Same as proc_root_link, but this addionally tries to get fs from other + * threads in the group */ +static int proc_task_root_link(struct inode *inode, struct dentry **dentry, struct vfsmount **mnt) +{ + struct fs_struct *fs; + int result = -ENOENT; + struct task_struct *leader = proc_task(inode); + + task_lock(leader); + fs = leader->fs; + if (fs) { + atomic_inc(&fs->count); + task_unlock(leader); + } else { + /* Try to get fs from other threads */ + task_unlock(leader); + struct task_struct *task = leader; + read_lock(&tasklist_lock); + if (pid_alive(task)) { + while ((task = next_thread(task)) != leader) { + task_lock(task); + fs = task->fs; + if (fs) { + atomic_inc(&fs->count); + task_unlock(task); + break; + } + task_unlock(task); + } + } + read_unlock(&tasklist_lock); + } + + if (fs) { + read_lock(&fs->lock); + *mnt = mntget(fs->rootmnt); + *dentry = dget(fs->root); + read_unlock(&fs->lock); + result = 0; + put_fs_struct(fs); + } + return result; +} + + #define MAY_PTRACE(task) \ (task == current || \ (task->parent == current && \ @@ -471,14 +517,14 @@ static int proc_oom_score(struct task_struct *task, char *buffer) /* permission checks */ -static int proc_check_root(struct inode *inode) +/* If the process being read is separated by chroot from the reading process, + * don't let the reader access the threads. + */ +static int proc_check_chroot(struct dentry *root, struct vfsmount *vfsmnt) { - struct dentry *de, *base, *root; - struct vfsmount *our_vfsmnt, *vfsmnt, *mnt; + struct dentry *de, *base; + struct vfsmount *our_vfsmnt, *mnt; int res = 0; - - if (proc_root_link(inode, &root, &vfsmnt)) /* Ewww... */ - return -ENOENT; read_lock(¤t->fs->lock); our_vfsmnt = mntget(current->fs->rootmnt); base = dget(current->fs->root); @@ -511,6 +557,16 @@ out: goto exit; } +static int proc_check_root(struct inode *inode) +{ + struct dentry *root; + struct vfsmount *vfsmnt; + + if (proc_root_link(inode, &root, &vfsmnt)) /* Ewww... */ + return -ENOENT; + return proc_check_chroot(root, vfsmnt); +} + static int proc_permission(struct inode *inode, int mask, struct nameidata *nd) { if (generic_permission(inode, mask, NULL) != 0) @@ -518,6 +574,20 @@ static int proc_permission(struct inode *inode, int mask, struct nameidata *nd) return proc_check_root(inode); } +static int proc_task_permission(struct inode *inode, int mask, struct nameidata *nd) +{ + struct dentry *root; + struct vfsmount *vfsmnt; + + if (generic_permission(inode, mask, NULL) != 0) + return -EACCES; + + if (proc_task_root_link(inode, &root, &vfsmnt)) + return -ENOENT; + + return proc_check_chroot(root, vfsmnt); +} + extern struct seq_operations proc_pid_maps_op; static int maps_open(struct inode *inode, struct file *file) { @@ -1419,7 +1489,7 @@ static struct inode_operations proc_fd_inode_operations = { static struct inode_operations proc_task_inode_operations = { .lookup = proc_task_lookup, - .permission = proc_permission, + .permission = proc_task_permission, }; #ifdef CONFIG_SECURITY From 31f6d9d628739c097964b8dbae939ea997da94a3 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 21 Sep 2005 09:55:30 -0700 Subject: [PATCH 212/240] [PATCH] Add printk_clock() ia64's sched_clock() accesses per-cpu data which isn't set up at boot time. Hence ia64 cannot use printk timestamping, because printk() will crash in sched_clock(). So make printk() use printk_clock(), defaulting to sched_clock(), overrideable by the architecture via attribute(weak). Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/printk.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/kernel/printk.c b/kernel/printk.c index a967605bc2e3..4b8f0f9230a4 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -488,6 +488,11 @@ static int __init printk_time_setup(char *str) __setup("time", printk_time_setup); +__attribute__((weak)) unsigned long long printk_clock(void) +{ + return sched_clock(); +} + /* * This is printk. It can be called from any context. We want it to work. * @@ -565,7 +570,7 @@ asmlinkage int vprintk(const char *fmt, va_list args) loglev_char = default_message_loglevel + '0'; } - t = sched_clock(); + t = printk_clock(); nanosec_rem = do_div(t, 1000000000); tlen = sprintf(tbuf, "<%c>[%5lu.%06lu] ", From 3c2822ccb1f8cc96fc006aa82e68e1944290014a Mon Sep 17 00:00:00 2001 From: Olof Johansson Date: Wed, 21 Sep 2005 09:55:31 -0700 Subject: [PATCH 213/240] [PATCH] PPC64: Fix boot for some pre-POWER4 systems Some RS64 systems (such as F80) have non-python host bridges with EADS. However, they have two EADS with 4 buses each under them, so the old logic that assumed no more than 7 busses per PHB failed miserably. Big thanks to Olaf Hering for helping me test this, he's got one of the few machines that broke from the previous logic. Also, to be a bit smarter at detecting the need for a PHB-level IOMMU table by checking for the presence of an ISA bus. Only PHBs with ISA bridges should need the PHB-level table. Signed-off-by: Olof Johansson Cc: Anton Blanchard Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc64/kernel/pSeries_iommu.c | 171 ++++++++++++++++-------------- 1 file changed, 90 insertions(+), 81 deletions(-) diff --git a/arch/ppc64/kernel/pSeries_iommu.c b/arch/ppc64/kernel/pSeries_iommu.c index f0fd7fbd6531..8c6313e7e145 100644 --- a/arch/ppc64/kernel/pSeries_iommu.c +++ b/arch/ppc64/kernel/pSeries_iommu.c @@ -265,8 +265,10 @@ static void iommu_table_setparms(struct pci_controller *phb, tbl->it_offset = phb->dma_window_base_cur >> PAGE_SHIFT; /* Test if we are going over 2GB of DMA space */ - if (phb->dma_window_base_cur + phb->dma_window_size > (1L << 31)) + if (phb->dma_window_base_cur + phb->dma_window_size > 0x80000000ul) { + udbg_printf("PCI_DMA: Unexpected number of IOAs under this PHB.\n"); panic("PCI_DMA: Unexpected number of IOAs under this PHB.\n"); + } phb->dma_window_base_cur += phb->dma_window_size; @@ -310,92 +312,84 @@ static void iommu_table_setparms_lpar(struct pci_controller *phb, static void iommu_bus_setup_pSeries(struct pci_bus *bus) { - struct device_node *dn, *pdn; - struct pci_dn *pci; + struct device_node *dn; struct iommu_table *tbl; + struct device_node *isa_dn, *isa_dn_orig; + struct device_node *tmp; + struct pci_dn *pci; + int children; DBG("iommu_bus_setup_pSeries, bus %p, bus->self %p\n", bus, bus->self); - /* For each (root) bus, we carve up the available DMA space in 256MB - * pieces. Since each piece is used by one (sub) bus/device, that would - * give a maximum of 7 devices per PHB. In most cases, this is plenty. + dn = pci_bus_to_OF_node(bus); + pci = PCI_DN(dn); + + if (bus->self) { + /* This is not a root bus, any setup will be done for the + * device-side of the bridge in iommu_dev_setup_pSeries(). + */ + return; + } + + /* Check if the ISA bus on the system is under + * this PHB. + */ + isa_dn = isa_dn_orig = of_find_node_by_type(NULL, "isa"); + + while (isa_dn && isa_dn != dn) + isa_dn = isa_dn->parent; + + if (isa_dn_orig) + of_node_put(isa_dn_orig); + + /* Count number of direct PCI children of the PHB. + * All PCI device nodes have class-code property, so it's + * an easy way to find them. + */ + for (children = 0, tmp = dn->child; tmp; tmp = tmp->sibling) + if (get_property(tmp, "class-code", NULL)) + children++; + + DBG("Children: %d\n", children); + + /* Calculate amount of DMA window per slot. Each window must be + * a power of two (due to pci_alloc_consistent requirements). * - * The exception is on Python PHBs (pre-POWER4). Here we don't have EADS - * bridges below the PHB to allocate the sectioned tables to, so instead - * we allocate a 1GB table at the PHB level. + * Keep 256MB aside for PHBs with ISA. */ - dn = pci_bus_to_OF_node(bus); - pci = dn->data; + if (!isa_dn) { + /* No ISA/IDE - just set window size and return */ + pci->phb->dma_window_size = 0x80000000ul; /* To be divided */ - if (!bus->self) { - /* Root bus */ - if (is_python(dn)) { - unsigned int *iohole; + while (pci->phb->dma_window_size * children > 0x80000000ul) + pci->phb->dma_window_size >>= 1; + DBG("No ISA/IDE, window size is %x\n", pci->phb->dma_window_size); + pci->phb->dma_window_base_cur = 0; - DBG("Python root bus %s\n", bus->name); - - iohole = (unsigned int *)get_property(dn, "io-hole", 0); - - if (iohole) { - /* On first bus we need to leave room for the - * ISA address space. Just skip the first 256MB - * alltogether. This leaves 768MB for the window. - */ - DBG("PHB has io-hole, reserving 256MB\n"); - pci->phb->dma_window_size = 3 << 28; - pci->phb->dma_window_base_cur = 1 << 28; - } else { - /* 1GB window by default */ - pci->phb->dma_window_size = 1 << 30; - pci->phb->dma_window_base_cur = 0; - } - - tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL); - - iommu_table_setparms(pci->phb, dn, tbl); - pci->iommu_table = iommu_init_table(tbl); - } else { - /* Do a 128MB table at root. This is used for the IDE - * controller on some SMP-mode POWER4 machines. It - * doesn't hurt to allocate it on other machines - * -- it'll just be unused since new tables are - * allocated on the EADS level. - * - * Allocate at offset 128MB to avoid having to deal - * with ISA holes; 128MB table for IDE is plenty. - */ - pci->phb->dma_window_size = 1 << 27; - pci->phb->dma_window_base_cur = 1 << 27; - - tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL); - - iommu_table_setparms(pci->phb, dn, tbl); - pci->iommu_table = iommu_init_table(tbl); - - /* All child buses have 256MB tables */ - pci->phb->dma_window_size = 1 << 28; - } - } else { - pdn = pci_bus_to_OF_node(bus->parent); - - if (!bus->parent->self && !is_python(pdn)) { - struct iommu_table *tbl; - /* First child and not python means this is the EADS - * level. Allocate new table for this slot with 256MB - * window. - */ - - tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL); - - iommu_table_setparms(pci->phb, dn, tbl); - - pci->iommu_table = iommu_init_table(tbl); - } else { - /* Lower than first child or under python, use parent table */ - pci->iommu_table = PCI_DN(pdn)->iommu_table; - } + return; } + + /* If we have ISA, then we probably have an IDE + * controller too. Allocate a 128MB table but + * skip the first 128MB to avoid stepping on ISA + * space. + */ + pci->phb->dma_window_size = 0x8000000ul; + pci->phb->dma_window_base_cur = 0x8000000ul; + + tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL); + + iommu_table_setparms(pci->phb, dn, tbl); + pci->iommu_table = iommu_init_table(tbl); + + /* Divide the rest (1.75GB) among the children */ + pci->phb->dma_window_size = 0x80000000ul; + while (pci->phb->dma_window_size * children > 0x70000000ul) + pci->phb->dma_window_size >>= 1; + + DBG("ISA/IDE, window size is %x\n", pci->phb->dma_window_size); + } @@ -446,14 +440,29 @@ static void iommu_bus_setup_pSeriesLP(struct pci_bus *bus) static void iommu_dev_setup_pSeries(struct pci_dev *dev) { struct device_node *dn, *mydn; + struct iommu_table *tbl; DBG("iommu_dev_setup_pSeries, dev %p (%s)\n", dev, dev->pretty_name); - /* Now copy the iommu_table ptr from the bus device down to the - * pci device_node. This means get_iommu_table() won't need to search - * up the device tree to find it. - */ + mydn = dn = pci_device_to_OF_node(dev); + /* If we're the direct child of a root bus, then we need to allocate + * an iommu table ourselves. The bus setup code should have setup + * the window sizes already. + */ + if (!dev->bus->self) { + DBG(" --> first child, no bridge. Allocating iommu table.\n"); + tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL); + iommu_table_setparms(PCI_DN(dn)->phb, dn, tbl); + PCI_DN(mydn)->iommu_table = iommu_init_table(tbl); + + return; + } + + /* If this device is further down the bus tree, search upwards until + * an already allocated iommu table is found and use that. + */ + while (dn && dn->data && PCI_DN(dn)->iommu_table == NULL) dn = dn->parent; From d6dbf42ee36d4f9ffc2a93877c19937c81974124 Mon Sep 17 00:00:00 2001 From: Mike Miller Date: Wed, 21 Sep 2005 09:55:32 -0700 Subject: [PATCH 214/240] [PATCH] cciss: busy_initializing bug fix This patch fixes the problem Bjorn reported. The busy_initializing flag should have cleared before going into the for loop. Signed-off-by: Mike Miller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/cciss.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index c56f995aadad..486b6e1c7dfb 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -483,9 +483,6 @@ static int cciss_open(struct inode *inode, struct file *filep) printk(KERN_DEBUG "cciss_open %s\n", inode->i_bdev->bd_disk->disk_name); #endif /* CCISS_DEBUG */ - if (host->busy_initializing) - return -EBUSY; - if (host->busy_initializing || drv->busy_configuring) return -EBUSY; /* @@ -2991,6 +2988,7 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, hba[i]->access.set_intr_mask(hba[i], CCISS_INTR_ON); cciss_procinit(i); + hba[i]->busy_initializing = 0; for(j=0; j < NWD; j++) { /* mfm */ drive_info_struct *drv = &(hba[i]->drv[j]); @@ -3033,7 +3031,6 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, add_disk(disk); } - hba[i]->busy_initializing = 0; return(1); clean4: From 1bc2a3bb86688ffca691ba2ad30a2d5dcba774cf Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Wed, 21 Sep 2005 09:55:33 -0700 Subject: [PATCH 215/240] [PATCH] ppc64: Fix issue with non zero boot cpu The new version of the flattened device tree passes the boot cpuid in the header instead of via a linux,boot-cpu property. We need to update the in kernel OF parsing code to do this, otherwise machines with a non zero boot cpuid fail to come up. Signed-off-by: Anton Blanchard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc64/kernel/prom_init.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/ppc64/kernel/prom_init.c b/arch/ppc64/kernel/prom_init.c index 9979919cdf92..f252670874a4 100644 --- a/arch/ppc64/kernel/prom_init.c +++ b/arch/ppc64/kernel/prom_init.c @@ -1711,6 +1711,7 @@ static void __init flatten_device_tree(void) unsigned long offset = reloc_offset(); unsigned long mem_start, mem_end, room; struct boot_param_header *hdr; + struct prom_t *_prom = PTRRELOC(&prom); char *namep; u64 *rsvmap; @@ -1765,6 +1766,7 @@ static void __init flatten_device_tree(void) RELOC(dt_struct_end) = PAGE_ALIGN(mem_start); /* Finish header */ + hdr->boot_cpuid_phys = _prom->cpu; hdr->magic = OF_DT_HEADER; hdr->totalsize = RELOC(dt_struct_end) - RELOC(dt_header_start); hdr->off_dt_struct = RELOC(dt_struct_start) - RELOC(dt_header_start); @@ -1854,7 +1856,6 @@ static void __init prom_find_boot_cpu(void) cpu_pkg = call_prom("instance-to-package", 1, 1, prom_cpu); - prom_setprop(cpu_pkg, "linux,boot-cpu", NULL, 0); prom_getprop(cpu_pkg, "reg", &getprop_rval, sizeof(getprop_rval)); _prom->cpu = getprop_rval; From bbcc6bee9e5bf45ec35e87a4e0755dff09fe25d4 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 21 Sep 2005 09:55:33 -0700 Subject: [PATCH 216/240] [PATCH] remove blkdev_scsi_issue_flush_fn again This function was removed a while ago, but crept in again via a recent scsi merge. Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/ll_rw_blk.c | 38 -------------------------------------- 1 file changed, 38 deletions(-) diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c index 483d71b10cf9..baedac522945 100644 --- a/drivers/block/ll_rw_blk.c +++ b/drivers/block/ll_rw_blk.c @@ -2373,44 +2373,6 @@ int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector) EXPORT_SYMBOL(blkdev_issue_flush); -/** - * blkdev_scsi_issue_flush_fn - issue flush for SCSI devices - * @q: device queue - * @disk: gendisk - * @error_sector: error offset - * - * Description: - * Devices understanding the SCSI command set, can use this function as - * a helper for issuing a cache flush. Note: driver is required to store - * the error offset (in case of error flushing) in ->sector of struct - * request. - */ -int blkdev_scsi_issue_flush_fn(request_queue_t *q, struct gendisk *disk, - sector_t *error_sector) -{ - struct request *rq = blk_get_request(q, WRITE, __GFP_WAIT); - int ret; - - rq->flags |= REQ_BLOCK_PC | REQ_SOFTBARRIER; - rq->sector = 0; - memset(rq->cmd, 0, sizeof(rq->cmd)); - rq->cmd[0] = 0x35; - rq->cmd_len = 12; - rq->data = NULL; - rq->data_len = 0; - rq->timeout = 60 * HZ; - - ret = blk_execute_rq(q, disk, rq, 0); - - if (ret && error_sector) - *error_sector = rq->sector; - - blk_put_request(rq); - return ret; -} - -EXPORT_SYMBOL(blkdev_scsi_issue_flush_fn); - static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io) { int rw = rq_data_dir(rq); From 40da47e1159b89f2df5c718d2d5e269ced72b541 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 21 Sep 2005 09:55:34 -0700 Subject: [PATCH 217/240] [PATCH] ppc64: Build zImage.vmode for G5 zImage.vmode was recently added. It's a version of zImage in which the ELF note section used by open firmware indicates that it requires a virtual mode instance of OF instead of real mode. This allows it to work with Apple OF, and thus is directly bootable (or netbootable) from OF command line. (Unfortunately, pSeries OF sort-of requires real mode and Apple OF sort-of requires virtual mode, and both tend to be unhappy if no notes section specifies the mode at all). However, we forgot to add zImage.vmode to the default G5 build. This fixes it. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc64/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/ppc64/Makefile b/arch/ppc64/Makefile index 17d2c1eac3b8..521c2a5a2862 100644 --- a/arch/ppc64/Makefile +++ b/arch/ppc64/Makefile @@ -107,7 +107,7 @@ install: vmlinux $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(BOOTIMAGE) $@ defaultimage-$(CONFIG_PPC_PSERIES) := zImage -defaultimage-$(CONFIG_PPC_PMAC) := vmlinux +defaultimage-$(CONFIG_PPC_PMAC) := zImage.vmode defaultimage-$(CONFIG_PPC_MAPLE) := zImage defaultimage-$(CONFIG_PPC_ISERIES) := vmlinux KBUILD_IMAGE := $(defaultimage-y) From 8b20f6da20d874a58becc7a08b130c4382ea7325 Mon Sep 17 00:00:00 2001 From: Deepak Saxena Date: Wed, 21 Sep 2005 09:55:35 -0700 Subject: [PATCH 218/240] [PATCH] Fix I2O config-osm init to return proper error We currently unregister the config-osm driver if initialization of the legacy ioctl() handlers failed but still return success. We should be returning -EBUSY in this case. Signed-off-by: Deepak Saxena Cc: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/i2o/config-osm.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/message/i2o/config-osm.c b/drivers/message/i2o/config-osm.c index af32ab4e90cd..10432f665201 100644 --- a/drivers/message/i2o/config-osm.c +++ b/drivers/message/i2o/config-osm.c @@ -56,8 +56,11 @@ static int __init i2o_config_init(void) return -EBUSY; } #ifdef CONFIG_I2O_CONFIG_OLD_IOCTL - if (i2o_config_old_init()) + if (i2o_config_old_init()) { + osm_err("old config handler initialization failed\n"); i2o_driver_unregister(&i2o_config_driver); + return -EBUSY; + } #endif return 0; From f10df6860498c7f52db4f00ff00a1ba334bcfea0 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 09:55:37 -0700 Subject: [PATCH 219/240] [PATCH] fix locking comment in unmap_region() That comment is plain wrong (we even take the pagetable lock inside unmap_region()). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Acked-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/mmap.c b/mm/mmap.c index 8b8e05f07cdb..fa11d91242e8 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1640,7 +1640,7 @@ static void unmap_vma_list(struct mm_struct *mm, struct vm_area_struct *vma) /* * Get rid of page table information in the indicated region. * - * Called with the page table lock held. + * Called with the mm semaphore held. */ static void unmap_region(struct mm_struct *mm, struct vm_area_struct *vma, struct vm_area_struct *prev, From f875a1a6650edce859bd21051a22e1c27ac7ea63 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 09:55:37 -0700 Subject: [PATCH 220/240] [PATCH] README update from the stone age We have no options which the user can set in the Makefile. Only the EXTRAVERSION, which is also useful in place of the "backup modules" suggestion. We don't have configuration options in the top Makefile. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- README | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/README b/README index 76dd780d88ed..2b5844d8cfa0 100644 --- a/README +++ b/README @@ -149,6 +149,9 @@ CONFIGURING the kernel: "make gconfig" X windows (Gtk) based configuration tool. "make oldconfig" Default all questions based on the contents of your existing ./.config file. + "make silentoldconfig" + Like above, but avoids cluttering the screen + with question already answered. NOTES on "make config": - having unnecessary drivers will make the kernel bigger, and can @@ -169,9 +172,6 @@ CONFIGURING the kernel: should probably answer 'n' to the questions for "development", "experimental", or "debugging" features. - - Check the top Makefile for further site-dependent configuration - (default SVGA mode etc). - COMPILING the kernel: - Make sure you have gcc 2.95.3 available. @@ -199,6 +199,9 @@ COMPILING the kernel: are installing a new kernel with the same version number as your working kernel, make a backup of your modules directory before you do a "make modules_install". + In alternative, before compiling, edit your Makefile and change the + "EXTRAVERSION" line - its content is appended to the regular kernel + version. - In order to boot your new kernel, you'll need to copy the kernel image (e.g. .../linux/arch/i386/boot/bzImage after compilation) From 7e871b6c8f1f4fda41e51ef86147facecac3be9f Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 09:55:38 -0700 Subject: [PATCH 221/240] [PATCH] mm: update stale comment for removal of page->list Update comment for the 2.6.6-rc1 conversion from page->list and address_space->{clean,dirty,locked}_pages to radix tree tagging and ->lru. I've mostly avoided to mention page lists (at least I've shortened the comment). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Acked-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/include/linux/mm.h b/include/linux/mm.h index 82d7024f0765..0d94c94d9d81 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -350,7 +350,8 @@ static inline void put_page(struct page *page) * only one copy in memory, at most, normally. * * For the non-reserved pages, page_count(page) denotes a reference count. - * page_count() == 0 means the page is free. + * page_count() == 0 means the page is free. page->lru is then used for + * freelist management in the buddy allocator. * page_count() == 1 means the page is used for exactly one purpose * (e.g. a private data page of one process). * @@ -376,10 +377,8 @@ static inline void put_page(struct page *page) * attaches, plus 1 if `private' contains something, plus one for * the page cache itself. * - * All pages belonging to an inode are in these doubly linked lists: - * mapping->clean_pages, mapping->dirty_pages and mapping->locked_pages; - * using the page->list list_head. These fields are also used for - * freelist managemet (when page_count()==0). + * Instead of keeping dirty/clean pages in per address-space lists, we instead + * now tag pages as dirty/under writeback in the radix tree. * * There is also a per-mapping radix tree mapping index to the page * in memory if present. The tree is rooted at mapping->root. From 7e2cff42cfac27c25202648c5c89f9171e5bc085 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 09:55:39 -0700 Subject: [PATCH 222/240] [PATCH] mm: add a note about partially hardcoded VM_* flags Hugh made me note this line for permission checking in mprotect(): if ((newflags & ~(newflags >> 4)) & 0xf) { after figuring out what's that about, I decided it's nasty enough. Btw Hugh itself didn't like the 0xf. We can safely change it to VM_READ|VM_WRITE|VM_EXEC because we never change VM_SHARED, so no need to check that. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Acked-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 1 + mm/mprotect.c | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/include/linux/mm.h b/include/linux/mm.h index 0d94c94d9d81..097b3a3c693d 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -136,6 +136,7 @@ extern unsigned int kobjsize(const void *objp); #define VM_EXEC 0x00000004 #define VM_SHARED 0x00000008 +/* mprotect() hardcodes VM_MAYREAD >> 4 == VM_READ, and so for r/w/x bits. */ #define VM_MAYREAD 0x00000010 /* limits for mprotect() etc */ #define VM_MAYWRITE 0x00000020 #define VM_MAYEXEC 0x00000040 diff --git a/mm/mprotect.c b/mm/mprotect.c index e9fbd013ad9a..57577f63b305 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -248,7 +248,8 @@ sys_mprotect(unsigned long start, size_t len, unsigned long prot) newflags = vm_flags | (vma->vm_flags & ~(VM_READ | VM_WRITE | VM_EXEC)); - if ((newflags & ~(newflags >> 4)) & 0xf) { + /* newflags >> 4 shift VM_MAY% in place of VM_% */ + if ((newflags & ~(newflags >> 4)) & (VM_READ | VM_WRITE | VM_EXEC)) { error = -EACCES; goto out; } From 8f91648dcb0685d58aa046b25c69ce0d5f284f8c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 21 Sep 2005 09:55:40 -0700 Subject: [PATCH 223/240] [PATCH] fixup Documentation/DocBook/kernel-hacking.tmpl __FUNCTION__ is the prefered kernel idiom, __func__ is not supported by gcc 2.95 (we actually map __FUNCTION__ to __func__ for more recent compilers, but it should never be used directly) Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/DocBook/kernel-hacking.tmpl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/DocBook/kernel-hacking.tmpl b/Documentation/DocBook/kernel-hacking.tmpl index 6367bba32d22..582032eea872 100644 --- a/Documentation/DocBook/kernel-hacking.tmpl +++ b/Documentation/DocBook/kernel-hacking.tmpl @@ -1105,7 +1105,7 @@ static struct block_device_operations opt_fops = { - Function names as strings (__func__). + Function names as strings (__FUNCTION__). From 4b1ac9ab15c1fc1e0b73d887a0c0cafe92991f97 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 21 Sep 2005 09:55:41 -0700 Subject: [PATCH 224/240] [PATCH] skge: add maintainer Add MAINTAINER record for Andrew ;-) Signed-off-by: Stephen Hemminger Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index dc8f3babcabd..ade7415d2467 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2266,6 +2266,12 @@ M: kristen.c.accardi@intel.com L: pcihpd-discuss@lists.sourceforge.net S: Maintained +SKGE, SKY2 10/100/1000 GIGABIT ETHERNET DRIVERS +P: Stephen Hemminger +M: shemminger@osdl.org +L: netdev@vger.kernel.org +S: Maintained + SPARC (sparc32): P: William L. Irwin M: wli@holomorphy.com From 7980cbbb30bf044e6f40912a3f6456204ddfc27e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 21 Sep 2005 09:55:43 -0700 Subject: [PATCH 225/240] [PATCH] Adds sys_set_mempolicy() in include/linux/syscalls.h Signed-off-by: Eric Dumazet Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/syscalls.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 425f58c8ea4a..a6f03e473737 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -508,5 +508,7 @@ asmlinkage long sys_keyctl(int cmd, unsigned long arg2, unsigned long arg3, asmlinkage long sys_ioprio_set(int which, int who, int ioprio); asmlinkage long sys_ioprio_get(int which, int who); +asmlinkage long sys_set_mempolicy(int mode, unsigned long __user *nmask, + unsigned long maxnode); #endif From 89d7cbf73e1a85241eb42339f6cb7a429fec178c Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 21 Sep 2005 09:55:44 -0700 Subject: [PATCH 226/240] [PATCH] update URL for HPET spec. Correct URL for HPET spec. Signed-off-by: Randy Dunlap Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 0969d570f3b5..21afa69a086d 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -308,7 +308,7 @@ config HPET_TIMER present. The HPET provides a stable time base on SMP systems, unlike the TSC, but it is more expensive to access, as it is off-chip. You can find the HPET spec at - . + . config X86_PM_TIMER bool "PM timer" From 62a36c43c898d45efcfe3376ea1da6a9a182e1ad Mon Sep 17 00:00:00 2001 From: Stephane Kardas Date: Wed, 21 Sep 2005 09:55:45 -0700 Subject: [PATCH 227/240] [PATCH] fat: fix adate During a forensic analysis on the fat file system, I found than the result for the last access date on this file system was different between the stat command and the istat command (package tct-utils). The istat command display a true date (the right windows date) but the stat primitive (so stat, find, ls command) displays a wrong date. Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/inode.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 51b1d15d9d5c..e2effe2dc9b2 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -300,9 +300,9 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de) inode->i_blksize = sbi->cluster_size; inode->i_blocks = ((inode->i_size + (sbi->cluster_size - 1)) & ~((loff_t)sbi->cluster_size - 1)) >> 9; - inode->i_mtime.tv_sec = inode->i_atime.tv_sec = + inode->i_mtime.tv_sec = date_dos2unix(le16_to_cpu(de->time), le16_to_cpu(de->date)); - inode->i_mtime.tv_nsec = inode->i_atime.tv_nsec = 0; + inode->i_mtime.tv_nsec = 0; if (sbi->options.isvfat) { int secs = de->ctime_cs / 100; int csecs = de->ctime_cs % 100; @@ -310,8 +310,11 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de) date_dos2unix(le16_to_cpu(de->ctime), le16_to_cpu(de->cdate)) + secs; inode->i_ctime.tv_nsec = csecs * 10000000; + inode->i_atime.tv_sec = + date_dos2unix(le16_to_cpu(0), le16_to_cpu(de->adate)); + inode->i_atime.tv_nsec = 0; } else - inode->i_ctime = inode->i_mtime; + inode->i_ctime = inode->i_atime = inode->i_mtime; return 0; } @@ -513,7 +516,9 @@ retry: raw_entry->starthi = cpu_to_le16(MSDOS_I(inode)->i_logstart >> 16); fat_date_unix2dos(inode->i_mtime.tv_sec, &raw_entry->time, &raw_entry->date); if (sbi->options.isvfat) { + __le16 atime; fat_date_unix2dos(inode->i_ctime.tv_sec,&raw_entry->ctime,&raw_entry->cdate); + fat_date_unix2dos(inode->i_atime.tv_sec,&atime,&raw_entry->adate); raw_entry->ctime_cs = (inode->i_ctime.tv_sec & 1) * 100 + inode->i_ctime.tv_nsec / 10000000; } From 972d512a17c1bb7c4b784a9da2ca75745fcc6989 Mon Sep 17 00:00:00 2001 From: Sean Hefty Date: Wed, 21 Sep 2005 12:31:26 -0700 Subject: [PATCH 228/240] [IB] Add MAD data field size definitions Clean up code by using enums instead of hard-coded magic numbers. Signed-off-by: Sean Hefty Signed-off-by: Roland Dreier --- drivers/infiniband/core/mad_rmpp.c | 15 ++++++--------- include/rdma/ib_mad.h | 15 +++++++++++---- 2 files changed, 17 insertions(+), 13 deletions(-) diff --git a/drivers/infiniband/core/mad_rmpp.c b/drivers/infiniband/core/mad_rmpp.c index 2bd8b1cc57c4..8f7cef0812f6 100644 --- a/drivers/infiniband/core/mad_rmpp.c +++ b/drivers/infiniband/core/mad_rmpp.c @@ -583,6 +583,7 @@ static int send_next_seg(struct ib_mad_send_wr_private *mad_send_wr) { struct ib_rmpp_mad *rmpp_mad; int timeout; + u32 paylen; rmpp_mad = (struct ib_rmpp_mad *)mad_send_wr->send_wr.wr.ud.mad_hdr; ib_set_rmpp_flags(&rmpp_mad->rmpp_hdr, IB_MGMT_RMPP_FLAG_ACTIVE); @@ -590,11 +591,9 @@ static int send_next_seg(struct ib_mad_send_wr_private *mad_send_wr) if (mad_send_wr->seg_num == 1) { rmpp_mad->rmpp_hdr.rmpp_rtime_flags |= IB_MGMT_RMPP_FLAG_FIRST; - rmpp_mad->rmpp_hdr.paylen_newwin = - cpu_to_be32(mad_send_wr->total_seg * - (sizeof(struct ib_rmpp_mad) - - offsetof(struct ib_rmpp_mad, data)) - - mad_send_wr->pad); + paylen = mad_send_wr->total_seg * IB_MGMT_RMPP_DATA - + mad_send_wr->pad; + rmpp_mad->rmpp_hdr.paylen_newwin = cpu_to_be32(paylen); mad_send_wr->sg_list[0].length = sizeof(struct ib_rmpp_mad); } else { mad_send_wr->send_wr.num_sge = 2; @@ -608,10 +607,8 @@ static int send_next_seg(struct ib_mad_send_wr_private *mad_send_wr) if (mad_send_wr->seg_num == mad_send_wr->total_seg) { rmpp_mad->rmpp_hdr.rmpp_rtime_flags |= IB_MGMT_RMPP_FLAG_LAST; - rmpp_mad->rmpp_hdr.paylen_newwin = - cpu_to_be32(sizeof(struct ib_rmpp_mad) - - offsetof(struct ib_rmpp_mad, data) - - mad_send_wr->pad); + paylen = IB_MGMT_RMPP_DATA - mad_send_wr->pad; + rmpp_mad->rmpp_hdr.paylen_newwin = cpu_to_be32(paylen); } /* 2 seconds for an ACK until we can find the packet lifetime */ diff --git a/include/rdma/ib_mad.h b/include/rdma/ib_mad.h index 53184a38fdf6..0e293fe733b0 100644 --- a/include/rdma/ib_mad.h +++ b/include/rdma/ib_mad.h @@ -108,6 +108,13 @@ #define IB_QP1_QKEY 0x80010000 #define IB_QP_SET_QKEY 0x80000000 +enum { + IB_MGMT_MAD_DATA = 232, + IB_MGMT_RMPP_DATA = 220, + IB_MGMT_VENDOR_DATA = 216, + IB_MGMT_SA_DATA = 200 +}; + struct ib_mad_hdr { u8 base_version; u8 mgmt_class; @@ -149,20 +156,20 @@ struct ib_sa_hdr { struct ib_mad { struct ib_mad_hdr mad_hdr; - u8 data[232]; + u8 data[IB_MGMT_MAD_DATA]; }; struct ib_rmpp_mad { struct ib_mad_hdr mad_hdr; struct ib_rmpp_hdr rmpp_hdr; - u8 data[220]; + u8 data[IB_MGMT_RMPP_DATA]; }; struct ib_sa_mad { struct ib_mad_hdr mad_hdr; struct ib_rmpp_hdr rmpp_hdr; struct ib_sa_hdr sa_hdr; - u8 data[200]; + u8 data[IB_MGMT_SA_DATA]; } __attribute__ ((packed)); struct ib_vendor_mad { @@ -170,7 +177,7 @@ struct ib_vendor_mad { struct ib_rmpp_hdr rmpp_hdr; u8 reserved; u8 oui[3]; - u8 data[216]; + u8 data[IB_MGMT_VENDOR_DATA]; }; struct ib_class_port_info From f2065e4242e5078d53c521b973c5cd4eae91eca2 Mon Sep 17 00:00:00 2001 From: Hal Rosenstock Date: Wed, 21 Sep 2005 12:33:09 -0700 Subject: [PATCH 229/240] [IB] Fix RMPP receive length calculation Based on simplification idea from Sean Hefty Signed-off-by: Hal Rosenstock Signed-off-by: Roland Dreier --- drivers/infiniband/core/mad_rmpp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/core/mad_rmpp.c b/drivers/infiniband/core/mad_rmpp.c index 8f7cef0812f6..e23836d0e21b 100644 --- a/drivers/infiniband/core/mad_rmpp.c +++ b/drivers/infiniband/core/mad_rmpp.c @@ -412,8 +412,8 @@ static inline int get_mad_len(struct mad_rmpp_recv *rmpp_recv) hdr_size = data_offset(rmpp_mad->mad_hdr.mgmt_class); data_size = sizeof(struct ib_rmpp_mad) - hdr_size; - pad = data_size - be32_to_cpu(rmpp_mad->rmpp_hdr.paylen_newwin); - if (pad > data_size || pad < 0) + pad = IB_MGMT_RMPP_DATA - be32_to_cpu(rmpp_mad->rmpp_hdr.paylen_newwin); + if (pad > IB_MGMT_RMPP_DATA || pad < 0) pad = 0; return hdr_size + rmpp_recv->seg_num * data_size - pad; From 6a1ced59b3ac5d1088f597fd0613a724cdf19e2c Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Wed, 21 Sep 2005 22:14:05 +0100 Subject: [PATCH 230/240] [ARM] 2932/1: Avoid the "noreturn" warning in arch/arm/kernel/traps.c Patch from Catalin Marinas This patch prevents the "noreturn function does return" warning in the __bug() function in arch/arm/kernel/traps.c Signed-off-by: Catalin Marinas Signed-off-by: Russell King --- arch/arm/kernel/traps.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index 4554c961251c..e7d22dbcb691 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -624,6 +624,9 @@ void __attribute__((noreturn)) __bug(const char *file, int line, void *data) printk(" - extra data = %p", data); printk("\n"); *(int *)0 = 0; + + /* Avoid "noreturn function does return" */ + for (;;); } EXPORT_SYMBOL(__bug); From c51179fb0c77ad91df5825f8f7eb670da97e137e Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 18:37:14 +0200 Subject: [PATCH 231/240] [PATCH] uml: adapt asm/futex.h to our arch Follow up to 4732efbeb997189d9f9b04708dc26bf8613ed721 - uml must just reuse as-is the backing architecture support. There is a micro-fixup is needed for the included file, which won't affect i386 behaviour at all. I've not tested compilation on x86_64, only on x86, but the code is almost the same except the culprit test, so everything should be ok on x86_64 too. Cc: Jakub Jelinek Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- include/asm-i386/futex.h | 2 +- include/asm-um/futex.h | 51 ++++------------------------------------ 2 files changed, 6 insertions(+), 47 deletions(-) diff --git a/include/asm-i386/futex.h b/include/asm-i386/futex.h index 44b9db806474..e7a271d39309 100644 --- a/include/asm-i386/futex.h +++ b/include/asm-i386/futex.h @@ -61,7 +61,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (op == FUTEX_OP_SET) __futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg); else { -#ifndef CONFIG_X86_BSWAP +#if !defined(CONFIG_X86_BSWAP) && !defined(CONFIG_UML) if (boot_cpu_data.x86 == 3) ret = -ENOSYS; else diff --git a/include/asm-um/futex.h b/include/asm-um/futex.h index 2cac5ecd9d00..142ee2d8e0fd 100644 --- a/include/asm-um/futex.h +++ b/include/asm-um/futex.h @@ -1,53 +1,12 @@ -#ifndef _ASM_FUTEX_H -#define _ASM_FUTEX_H - -#ifdef __KERNEL__ +#ifndef __UM_FUTEX_H +#define __UM_FUTEX_H #include #include +#include +#include #include -static inline int -futex_atomic_op_inuser (int encoded_op, int __user *uaddr) -{ - int op = (encoded_op >> 28) & 7; - int cmp = (encoded_op >> 24) & 15; - int oparg = (encoded_op << 8) >> 20; - int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; - if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) - oparg = 1 << oparg; - - if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) - return -EFAULT; - - inc_preempt_count(); - - switch (op) { - case FUTEX_OP_SET: - case FUTEX_OP_ADD: - case FUTEX_OP_OR: - case FUTEX_OP_ANDN: - case FUTEX_OP_XOR: - default: - ret = -ENOSYS; - } - - dec_preempt_count(); - - if (!ret) { - switch (cmp) { - case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break; - case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break; - case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break; - case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break; - case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break; - case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break; - default: ret = -ENOSYS; - } - } - return ret; -} +#include "asm/arch/futex.h" #endif -#endif From 676067cfeaa16f6f338e067e83ce4733b41c0b24 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 18:38:09 +0200 Subject: [PATCH 232/240] [PATCH] Remove unused var from asm/futex.h As recently done by Russell King for ARM, commit 4732efbeb997189d9f9b04708dc26bf8613ed721 introduces a generic asm/futex.h copied along most arches, which includes a "-ENOSYS support" to be changed if needed. However, it includes an unused var (taken from the "real" version) which GCC warns about. Remove it from all arches having that file version (i.e. same GIT id). $ git-diff-tree -r HEAD and $ git-ls-tree -r HEAD include/|grep 9feff4ce1424bc390608326240be369eb13aa648 may be more interesting than looking at the patch itself, to make sure I've just copied the arm header to all other archs having the original dummy version of this file. Cc: Jakub Jelinek Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- include/asm-alpha/futex.h | 2 +- include/asm-arm26/futex.h | 2 +- include/asm-cris/futex.h | 2 +- include/asm-frv/futex.h | 2 +- include/asm-h8300/futex.h | 2 +- include/asm-ia64/futex.h | 2 +- include/asm-m32r/futex.h | 2 +- include/asm-m68k/futex.h | 2 +- include/asm-m68knommu/futex.h | 2 +- include/asm-parisc/futex.h | 2 +- include/asm-ppc/futex.h | 2 +- include/asm-s390/futex.h | 2 +- include/asm-sh/futex.h | 2 +- include/asm-sh64/futex.h | 2 +- include/asm-sparc/futex.h | 2 +- include/asm-sparc64/futex.h | 2 +- include/asm-v850/futex.h | 2 +- 17 files changed, 17 insertions(+), 17 deletions(-) diff --git a/include/asm-alpha/futex.h b/include/asm-alpha/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-alpha/futex.h +++ b/include/asm-alpha/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-arm26/futex.h b/include/asm-arm26/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-arm26/futex.h +++ b/include/asm-arm26/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-cris/futex.h b/include/asm-cris/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-cris/futex.h +++ b/include/asm-cris/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-frv/futex.h b/include/asm-frv/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-frv/futex.h +++ b/include/asm-frv/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-h8300/futex.h b/include/asm-h8300/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-h8300/futex.h +++ b/include/asm-h8300/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-ia64/futex.h b/include/asm-ia64/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-ia64/futex.h +++ b/include/asm-ia64/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-m32r/futex.h b/include/asm-m32r/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-m32r/futex.h +++ b/include/asm-m32r/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-m68k/futex.h b/include/asm-m68k/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-m68k/futex.h +++ b/include/asm-m68k/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-m68knommu/futex.h b/include/asm-m68knommu/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-m68knommu/futex.h +++ b/include/asm-m68knommu/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-parisc/futex.h b/include/asm-parisc/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-parisc/futex.h +++ b/include/asm-parisc/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-ppc/futex.h b/include/asm-ppc/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-ppc/futex.h +++ b/include/asm-ppc/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-s390/futex.h b/include/asm-s390/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-s390/futex.h +++ b/include/asm-s390/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-sh/futex.h b/include/asm-sh/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-sh/futex.h +++ b/include/asm-sh/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-sh64/futex.h b/include/asm-sh64/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-sh64/futex.h +++ b/include/asm-sh64/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-sparc/futex.h b/include/asm-sparc/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-sparc/futex.h +++ b/include/asm-sparc/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-sparc64/futex.h b/include/asm-sparc64/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-sparc64/futex.h +++ b/include/asm-sparc64/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; diff --git a/include/asm-v850/futex.h b/include/asm-v850/futex.h index 2cac5ecd9d00..9feff4ce1424 100644 --- a/include/asm-v850/futex.h +++ b/include/asm-v850/futex.h @@ -14,7 +14,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; int cmparg = (encoded_op << 20) >> 20; - int oldval = 0, ret, tem; + int oldval = 0, ret; if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; From 76629ac154aacfec8883954c3ff9871905d2fb8e Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 18:38:33 +0200 Subject: [PATCH 233/240] [PATCH] uml: remove verify_area_{tt,skas} When removing verify_area, verify_area_{tt,skas} were forgotten. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- arch/um/kernel/skas/include/uaccess-skas.h | 6 ------ arch/um/kernel/tt/include/uaccess-tt.h | 6 ------ 2 files changed, 12 deletions(-) diff --git a/arch/um/kernel/skas/include/uaccess-skas.h b/arch/um/kernel/skas/include/uaccess-skas.h index 6ee3f3902e68..7da0c2def0ef 100644 --- a/arch/um/kernel/skas/include/uaccess-skas.h +++ b/arch/um/kernel/skas/include/uaccess-skas.h @@ -18,12 +18,6 @@ ((unsigned long) (addr) + (size) <= FIXADDR_USER_END) && \ ((unsigned long) (addr) + (size) >= (unsigned long)(addr)))) -static inline int verify_area_skas(int type, const void __user * addr, - unsigned long size) -{ - return(access_ok_skas(type, addr, size) ? 0 : -EFAULT); -} - extern int copy_from_user_skas(void *to, const void __user *from, int n); extern int copy_to_user_skas(void __user *to, const void *from, int n); extern int strncpy_from_user_skas(char *dst, const char __user *src, int count); diff --git a/arch/um/kernel/tt/include/uaccess-tt.h b/arch/um/kernel/tt/include/uaccess-tt.h index aa6db384af80..dc2ebfa8c54f 100644 --- a/arch/um/kernel/tt/include/uaccess-tt.h +++ b/arch/um/kernel/tt/include/uaccess-tt.h @@ -33,12 +33,6 @@ extern unsigned long uml_physmem; (((unsigned long) (addr) <= ((unsigned long) (addr) + (size))) && \ (under_task_size(addr, size) || is_stack(addr, size)))) -static inline int verify_area_tt(int type, const void __user * addr, - unsigned long size) -{ - return(access_ok_tt(type, addr, size) ? 0 : -EFAULT); -} - extern unsigned long get_fault_addr(void); extern int __do_copy_from_user(void *to, const void *from, int n, From 36decba90b184a25b7f77fb7df1612125cc41ac5 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 18:38:57 +0200 Subject: [PATCH 234/240] [PATCH] uml: fix modify_ldt - missing break in switch I am a lamer :-(. Luckily, Luo Xin performed LTP testing and found this failure. Btw, the fact that the patch in which I introduced this was merged shows that: a) I'm really trusted by people b) sometimes they're wrong about point a). c) lack of time for reviewers. CC: Luo Xin Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- arch/um/sys-i386/ldt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/um/sys-i386/ldt.c b/arch/um/sys-i386/ldt.c index bd3c34aa52e5..36b5c2c13289 100644 --- a/arch/um/sys-i386/ldt.c +++ b/arch/um/sys-i386/ldt.c @@ -83,6 +83,7 @@ int sys_modify_ldt(int func, void __user *ptr, unsigned long bytecount) goto out; } p = buf; + break; default: res = -ENOSYS; goto out; From 8e5cb35a60c8792ffc3ba89ea66ebdd3c2f01cb1 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 18:39:14 +0200 Subject: [PATCH 235/240] [PATCH] uml: fix uname output on 32-bit binary on 64-bit host Translate uname output taken from the host if needed. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- arch/um/kernel/user_util.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/um/kernel/user_util.c b/arch/um/kernel/user_util.c index a25f3ea11fd7..41d17c71511c 100644 --- a/arch/um/kernel/user_util.c +++ b/arch/um/kernel/user_util.c @@ -128,6 +128,12 @@ void setup_machinename(char *machine_out) struct utsname host; uname(&host); +#if defined(UML_CONFIG_UML_X86) && !defined(UML_CONFIG_64BIT) + if (!strcmp(host.machine, "x86_64")) { + strcpy(machine_out, "i686"); + return; + } +#endif strcpy(machine_out, host.machine); } From fd7481047882eaa0a692a545d5f8419612ece1b0 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 18:39:32 +0200 Subject: [PATCH 236/240] [PATCH] uml: Fix conflict between libc and ipv6 gcc is now complaining during link on some hosts - fix it as for other things. Reported by Antoine Martin . Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- arch/um/Makefile | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/um/Makefile b/arch/um/Makefile index ce987266dac6..5b5af95721ab 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -53,9 +53,13 @@ SYS_DIR := $(ARCH_DIR)/include/sysdep-$(SUBARCH) # -Dvmap=kernel_vmap affects everything, and prevents anything from # referencing the libpcap.o symbol so named. +# +# Same things for in6addr_loopback - found in libc. CFLAGS += $(CFLAGS-y) -D__arch_um__ -DSUBARCH=\"$(SUBARCH)\" \ - $(ARCH_INCLUDE) $(MODE_INCLUDE) -Dvmap=kernel_vmap + $(ARCH_INCLUDE) $(MODE_INCLUDE) -Dvmap=kernel_vmap \ + -Din6addr_loopback=kernel_in6addr_loopback + AFLAGS += $(ARCH_INCLUDE) USER_CFLAGS := $(patsubst -I%,,$(CFLAGS)) From 41c79dc3825545d9aa2d5e5b1a99f8faabf51455 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 18:39:47 +0200 Subject: [PATCH 237/240] [PATCH] uml: fix bogus HOST_ELF_CLASS symbol name Even if with a bit of misunderstanding, Al fixed this in commit 95608261dae863bc43292e6fbd946a3abd3aa49f. Well, the symbol was intended to come from userspace (it exists there on normal host), but since some hosts may miss that, using the kernel one is just as fine. However, rename it to be named consistently with the rest. Actually, he missed converting ELFCLASS32 to coming from kernel headers. For consistence, add ELFCLASS64 too. Cc: Al Viro Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- arch/um/include/common-offsets.h | 4 +++- arch/um/os-Linux/elf_aux.c | 3 ++- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/um/include/common-offsets.h b/arch/um/include/common-offsets.h index 0aa620970adb..782ac3a3baf9 100644 --- a/arch/um/include/common-offsets.h +++ b/arch/um/include/common-offsets.h @@ -12,4 +12,6 @@ DEFINE_STR(UM_KERN_WARNING, KERN_WARNING); DEFINE_STR(UM_KERN_NOTICE, KERN_NOTICE); DEFINE_STR(UM_KERN_INFO, KERN_INFO); DEFINE_STR(UM_KERN_DEBUG, KERN_DEBUG); -DEFINE(HOST_ELF_CLASS, ELF_CLASS); +DEFINE(UM_ELF_CLASS, ELF_CLASS); +DEFINE(UM_ELFCLASS32, ELFCLASS32); +DEFINE(UM_ELFCLASS64, ELFCLASS64); diff --git a/arch/um/os-Linux/elf_aux.c b/arch/um/os-Linux/elf_aux.c index e770cb029578..ab33cb3c74ec 100644 --- a/arch/um/os-Linux/elf_aux.c +++ b/arch/um/os-Linux/elf_aux.c @@ -14,7 +14,8 @@ #include "mem_user.h" #include -#if HOST_ELF_CLASS == ELFCLASS32 +/* Use the one from the kernel - the host may miss it, if having old headers. */ +#if UM_ELF_CLASS == UM_ELFCLASS32 typedef Elf32_auxv_t elf_auxv_t; #else typedef Elf64_auxv_t elf_auxv_t; From ae756df8ddd00a30bb3bdf6774b514c096dda4a4 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 18:40:10 +0200 Subject: [PATCH 238/240] [PATCH] uml: readd removed unistd.h inclusion Readd this header (deleted in 60d339f6fe0831060600c62418b71a62ad26c281). A warning is spit out here about undeclared getpgrp(). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- arch/um/os-Linux/process.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c index d32413e4b4ce..d9c52387c4a1 100644 --- a/arch/um/os-Linux/process.c +++ b/arch/um/os-Linux/process.c @@ -3,6 +3,7 @@ * Licensed under the GPL */ +#include #include #include #include From a46c9047642d70f5619e4a14129d74dd1f6ea74f Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 18:40:29 +0200 Subject: [PATCH 239/240] [PATCH] uml: comment about cast build fix Explain why the casting we do to silence this warning is indeed safe. It is because the field we're casting from, though being 64-bit wide, was filled with a pointer in first place by ourselves. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- arch/um/os-Linux/aio.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/um/os-Linux/aio.c b/arch/um/os-Linux/aio.c index 298d5632128b..f6e64026f995 100644 --- a/arch/um/os-Linux/aio.c +++ b/arch/um/os-Linux/aio.c @@ -144,6 +144,7 @@ static int aio_thread(void *arg) "errno = %d\n", errno); } else { + /* This is safe as we've just a pointer here. */ aio = (struct aio_context *) (long) event.data; if(update_aio(aio, event.res)){ do_aio(ctx, aio); From efb0372bbaf5b829ff8c39db372779928af542a7 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 21 Sep 2005 18:40:49 +0200 Subject: [PATCH 240/240] [PATCH] uml: fix compile warning after consolidation patch The header declaring this function wasn't included, so the function declaration was totally bogus wrt. the proto - even if this wasn't going to fail at all. It was so bad that the compile warning I got was "control reaches end of non-void function", i.e. missing return. Actually, this has been there for ages, the consolidation patch just added the warning which was needed to clean it up. Nice. Really. Cc: Allan Graves Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Linus Torvalds --- arch/um/kernel/tt/process_kern.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/um/kernel/tt/process_kern.c b/arch/um/kernel/tt/process_kern.c index 0de05a268b22..cfaa373a6e77 100644 --- a/arch/um/kernel/tt/process_kern.c +++ b/arch/um/kernel/tt/process_kern.c @@ -23,10 +23,11 @@ #include "mem_user.h" #include "tlb.h" #include "mode.h" +#include "mode_kern.h" #include "init.h" #include "tt.h" -int switch_to_tt(void *prev, void *next, void *last) +void switch_to_tt(void *prev, void *next) { struct task_struct *from, *to, *prev_sched; unsigned long flags;