From 0afd4a21ba7d75e93fa79cf05d7a21774e149c0f Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 2 Feb 2009 13:27:44 -0800 Subject: net: Fix userland breakage wrt. linux/if_tunnel.h Reported by Andrew Walrond Changeset c19e654ddbe3831252f61e76a74d661e1a755530 ("gre: Add netlink interface") added an include of linux/ip.h to linux/if_tunnel.h We can't really let that get exposed to userspace because this conflicts with types defined in netinet/ip.h which userland is almost certainly going to have included either explicitly or implicitly. So guard this include with a __KERNEL__ ifdef. Signed-off-by: David S. Miller --- include/linux/Kbuild | 2 +- include/linux/if_tunnel.h | 3 +++ 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 12e9a2957caf..6a9bb9753235 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -89,7 +89,6 @@ header-y += if_ppp.h header-y += if_slip.h header-y += if_strip.h header-y += if_tun.h -header-y += if_tunnel.h header-y += in_route.h header-y += ioctl.h header-y += ip6_tunnel.h @@ -235,6 +234,7 @@ unifdef-y += if_phonet.h unifdef-y += if_pppol2tp.h unifdef-y += if_pppox.h unifdef-y += if_tr.h +unifdef-y += if_tunnel.h unifdef-y += if_vlan.h unifdef-y += igmp.h unifdef-y += inet_diag.h diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h index aeab2cb32a9c..82c43624c067 100644 --- a/include/linux/if_tunnel.h +++ b/include/linux/if_tunnel.h @@ -2,7 +2,10 @@ #define _IF_TUNNEL_H_ #include + +#ifdef __KERNEL__ #include +#endif #define SIOCGETTUNNEL (SIOCDEVPRIVATE + 0) #define SIOCADDTUNNEL (SIOCDEVPRIVATE + 1) -- cgit v1.2.3 From 2999b58b795ad81f10e34bdbbfd2742172f247e4 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sun, 1 Feb 2009 20:46:39 +0400 Subject: ide/libata: fix ata_id_is_cfa() (take 4) When checking for the CFA feature set support, ata_id_is_cfa() tests bit 2 in word 82 of the identify data instead the word 83; it also checks the ATA/PI version support in the word 80 (which the CompactFlash specifications have as reserved), this having no slightest chance to work on the modern CF cards that don't have 0x848A in the word 0... Signed-off-by: Sergei Shtylyov Signed-off-by: Jeff Garzik --- include/linux/ata.h | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ata.h b/include/linux/ata.h index a53318b8cbd0..08a86d5cdf1b 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -731,12 +731,17 @@ static inline int ata_id_current_chs_valid(const u16 *id) static inline int ata_id_is_cfa(const u16 *id) { - if (id[ATA_ID_CONFIG] == 0x848A) /* Standard CF */ + if (id[ATA_ID_CONFIG] == 0x848A) /* Traditional CF */ return 1; - /* Could be CF hiding as standard ATA */ - if (ata_id_major_version(id) >= 3 && - id[ATA_ID_COMMAND_SET_1] != 0xFFFF && - (id[ATA_ID_COMMAND_SET_1] & (1 << 2))) + /* + * CF specs don't require specific value in the word 0 anymore and yet + * they forbid to report the ATA version in the word 80 and require the + * CFA feature set support to be indicated in the word 83 in this case. + * Unfortunately, some cards only follow either of this requirements, + * and while those that don't indicate CFA feature support need some + * sort of quirk list, it seems impractical for the ones that do... + */ + if ((id[ATA_ID_COMMAND_SET_2] & 0xC004) == 0x4004) return 1; return 0; } -- cgit v1.2.3 From 99cf610aa4840d822cdc67d194b23b55010ca9bd Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 29 Jan 2009 20:31:32 +0900 Subject: libata: clear dev->ering in smarter way dev->ering used to be cleared together with the rest of ata_device in ata_dev_init() which is called whenever a probing event occurs. dev->ering is about to be used to track probing failures so it needs to remain persistent over multiple porbing events. This patch achieves this by doing the following. * Instead of CLEAR_OFFSET, define CLEAR_BEGIN and CLEAR_END and only clear between BEGIN and END. ering is moved after END. The split of persistent area is to allow hotter items remain at the head. * ering is explicitly cleared on ata_dev_disable() and when device attach succeeds. So, ering is persistent throug a device's life time (unless explicitly cleared of course) and also through periods inbetween disablement of an attached device and successful detection of the next one. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 4 ++-- drivers/ata/libata-eh.c | 7 +++++++ include/linux/libata.h | 18 ++++++++++-------- 3 files changed, 19 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index d006e5c4768c..564c03c4ebb3 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -5404,8 +5404,8 @@ void ata_dev_init(struct ata_device *dev) dev->horkage = 0; spin_unlock_irqrestore(ap->lock, flags); - memset((void *)dev + ATA_DEVICE_CLEAR_OFFSET, 0, - sizeof(*dev) - ATA_DEVICE_CLEAR_OFFSET); + memset((void *)dev + ATA_DEVICE_CLEAR_BEGIN, 0, + ATA_DEVICE_CLEAR_END - ATA_DEVICE_CLEAR_BEGIN); dev->pio_mask = UINT_MAX; dev->mwdma_mask = UINT_MAX; dev->udma_mask = UINT_MAX; diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index aafe82bf5e2e..90092c1aae53 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -1194,6 +1194,11 @@ void ata_dev_disable(struct ata_device *dev) ata_acpi_on_disable(dev); ata_down_xfermask_limit(dev, ATA_DNXFER_FORCE_PIO0 | ATA_DNXFER_QUIET); dev->class++; + + /* From now till the next successful probe, ering is used to + * track probe failures. Clear accumulated device error info. + */ + ata_ering_clear(&dev->ering); } /** @@ -2765,6 +2770,8 @@ static int ata_eh_revalidate_and_attach(struct ata_link *link, readid_flags, dev->id); switch (rc) { case 0: + /* clear error info accumulated during probe */ + ata_ering_clear(&dev->ering); new_mask |= 1 << dev->devno; break; case -ENOENT: diff --git a/include/linux/libata.h b/include/linux/libata.h index bca3ba25f52a..9dcefee5843c 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -580,7 +580,7 @@ struct ata_device { acpi_handle acpi_handle; union acpi_object *gtf_cache; #endif - /* n_sector is used as CLEAR_OFFSET, read comment above CLEAR_OFFSET */ + /* n_sector is CLEAR_BEGIN, read comment above CLEAR_BEGIN */ u64 n_sectors; /* size of device, if ATA */ unsigned int class; /* ATA_DEV_xxx */ unsigned long unpark_deadline; @@ -605,20 +605,22 @@ struct ata_device { u16 heads; /* Number of heads */ u16 sectors; /* Number of sectors per track */ - /* error history */ - int spdn_cnt; - struct ata_ering ering; - union { u16 id[ATA_ID_WORDS]; /* IDENTIFY xxx DEVICE data */ u32 gscr[SATA_PMP_GSCR_DWORDS]; /* PMP GSCR block */ }; + + /* error history */ + int spdn_cnt; + /* ering is CLEAR_END, read comment above CLEAR_END */ + struct ata_ering ering; }; -/* Offset into struct ata_device. Fields above it are maintained - * acress device init. Fields below are zeroed. +/* Fields between ATA_DEVICE_CLEAR_BEGIN and ATA_DEVICE_CLEAR_END are + * cleared to zero on ata_dev_init(). */ -#define ATA_DEVICE_CLEAR_OFFSET offsetof(struct ata_device, n_sectors) +#define ATA_DEVICE_CLEAR_BEGIN offsetof(struct ata_device, n_sectors) +#define ATA_DEVICE_CLEAR_END offsetof(struct ata_device, ering) struct ata_eh_info { struct ata_device *dev; /* offending device */ -- cgit v1.2.3 From 9062712fa9ed13b531dfc2228086650b8bd6a255 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 29 Jan 2009 20:31:36 +0900 Subject: libata: implement HORKAGE_1_5_GBPS and apply it to WD My Book 3Gbps is often much more prone to transmission failures. It's usually okay to let EH handle speed down after transmission failures but some WD My Book drives completely shutdown after certain transmission failures and after it only power cycling can revive them. Combined with the fact that external drives often end up with cable assembly which is longer than usual and more likely to have intervening gender, this makes these drives very likely to shutdown under certain configurations virtually rendering them unusable. This patch implements HOARKGE_1_5_GBPS and applies it to WD My Book such that 1.5Gbps is forced once the device is identified. Please take a look at the following bz for related reports. http://bugzilla.kernel.org/show_bug.cgi?id=9913 Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 41 +++++++++++++++++++++++++++++++++++++++++ include/linux/libata.h | 1 + 2 files changed, 42 insertions(+) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 3c5965d56c47..9fbf0595f3d4 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -2232,6 +2232,40 @@ retry: return rc; } +static int ata_do_link_spd_horkage(struct ata_device *dev) +{ + struct ata_link *plink = ata_dev_phys_link(dev); + u32 target, target_limit; + + if (!sata_scr_valid(plink)) + return 0; + + if (dev->horkage & ATA_HORKAGE_1_5_GBPS) + target = 1; + else + return 0; + + target_limit = (1 << target) - 1; + + /* if already on stricter limit, no need to push further */ + if (plink->sata_spd_limit <= target_limit) + return 0; + + plink->sata_spd_limit = target_limit; + + /* Request another EH round by returning -EAGAIN if link is + * going faster than the target speed. Forward progress is + * guaranteed by setting sata_spd_limit to target_limit above. + */ + if (plink->sata_spd > target) { + ata_dev_printk(dev, KERN_INFO, + "applying link speed limit horkage to %s\n", + sata_spd_string(target)); + return -EAGAIN; + } + return 0; +} + static inline u8 ata_dev_knobble(struct ata_device *dev) { struct ata_port *ap = dev->link->ap; @@ -2322,6 +2356,10 @@ int ata_dev_configure(struct ata_device *dev) return 0; } + rc = ata_do_link_spd_horkage(dev); + if (rc) + return rc; + /* let ACPI work its magic */ rc = ata_acpi_on_devcfg(dev); if (rc) @@ -4223,6 +4261,9 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { /* Devices that do not need bridging limits applied */ { "MTRON MSP-SATA*", NULL, ATA_HORKAGE_BRIDGE_OK, }, + /* Devices which aren't very happy with higher link speeds */ + { "WD My Book", NULL, ATA_HORKAGE_1_5_GBPS, }, + /* End Marker */ { } }; diff --git a/include/linux/libata.h b/include/linux/libata.h index 9dcefee5843c..5d87bc09a1f5 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -380,6 +380,7 @@ enum { ATA_HORKAGE_ATAPI_MOD16_DMA = (1 << 11), /* use ATAPI DMA for commands not multiple of 16 bytes */ ATA_HORKAGE_FIRMWARE_WARN = (1 << 12), /* firwmare update warning */ + ATA_HORKAGE_1_5_GBPS = (1 << 13), /* force 1.5 Gbps */ /* DMA mask for user DMA control: User visible values; DO NOT renumber */ -- cgit v1.2.3 From 35626129abcd6a7547e84c817ef5b6eff7a8758b Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 2 Feb 2009 16:00:29 -0800 Subject: sched: add missing kernel-doc in sched.h Add kernel-doc notation for @lock: include/linux/sched.h:457: No description found for parameter 'lock' Signed-off-by: Randy Dunlap Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 5a7c76388731..2127e959e0f4 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -443,6 +443,7 @@ struct pacct_struct { * @utime: time spent in user mode, in &cputime_t units * @stime: time spent in kernel mode, in &cputime_t units * @sum_exec_runtime: total time spent on the CPU, in nanoseconds + * @lock: lock for fields in this struct * * This structure groups together three kinds of CPU time that are * tracked for threads and thread groups. Most things considering -- cgit v1.2.3 From 97c44836cdec1ea713a15d84098a1a908157e68f Mon Sep 17 00:00:00 2001 From: "Timothy S. Nelson" Date: Fri, 30 Jan 2009 06:12:47 +1100 Subject: PCI: return error on failure to read PCI ROMs This patch makes the ROM reading code return an error to user space if the size of the ROM read is equal to 0. The patch also emits a warnings if the contents of the ROM are invalid, and documents the effects of the "enable" file on ROM reading. Signed-off-by: Timothy S. Nelson Acked-by: Alex Villacis-Lasso Signed-off-by: Jesse Barnes --- Documentation/filesystems/sysfs-pci.txt | 13 ++++++++++++- arch/ia64/sn/kernel/io_acpi_init.c | 2 +- arch/ia64/sn/kernel/io_init.c | 2 +- drivers/pci/pci-sysfs.c | 4 ++-- drivers/pci/rom.c | 8 +++++--- include/linux/pci.h | 2 +- 6 files changed, 22 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/Documentation/filesystems/sysfs-pci.txt b/Documentation/filesystems/sysfs-pci.txt index 68ef48839c04..9f8740ca3f3b 100644 --- a/Documentation/filesystems/sysfs-pci.txt +++ b/Documentation/filesystems/sysfs-pci.txt @@ -9,6 +9,7 @@ that support it. For example, a given bus might look like this: | |-- class | |-- config | |-- device + | |-- enable | |-- irq | |-- local_cpus | |-- resource @@ -32,6 +33,7 @@ files, each with their own function. class PCI class (ascii, ro) config PCI config space (binary, rw) device PCI device (ascii, ro) + enable Whether the device is enabled (ascii, rw) irq IRQ number (ascii, ro) local_cpus nearby CPU mask (cpumask, ro) resource PCI resource host addresses (ascii, ro) @@ -57,10 +59,19 @@ used to do actual device programming from userspace. Note that some platforms don't support mmapping of certain resources, so be sure to check the return value from any attempted mmap. +The 'enable' file provides a counter that indicates how many times the device +has been enabled. If the 'enable' file currently returns '4', and a '1' is +echoed into it, it will then return '5'. Echoing a '0' into it will decrease +the count. Even when it returns to 0, though, some of the initialisation +may not be reversed. + The 'rom' file is special in that it provides read-only access to the device's ROM file, if available. It's disabled by default, however, so applications should write the string "1" to the file to enable it before attempting a read -call, and disable it following the access by writing "0" to the file. +call, and disable it following the access by writing "0" to the file. Note +that the device must be enabled for a rom read to return data succesfully. +In the event a driver is not bound to the device, it can be enabled using the +'enable' file, documented above. Accessing legacy resources through sysfs ---------------------------------------- diff --git a/arch/ia64/sn/kernel/io_acpi_init.c b/arch/ia64/sn/kernel/io_acpi_init.c index c5a214026a77..d0223abbbbd4 100644 --- a/arch/ia64/sn/kernel/io_acpi_init.c +++ b/arch/ia64/sn/kernel/io_acpi_init.c @@ -443,7 +443,7 @@ sn_acpi_slot_fixup(struct pci_dev *dev) size = pci_resource_len(dev, PCI_ROM_RESOURCE); addr = ioremap(pcidev_info->pdi_pio_mapped_addr[PCI_ROM_RESOURCE], size); - image_size = pci_get_rom_size(addr, size); + image_size = pci_get_rom_size(dev, addr, size); dev->resource[PCI_ROM_RESOURCE].start = (unsigned long) addr; dev->resource[PCI_ROM_RESOURCE].end = (unsigned long) addr + image_size - 1; diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c index 4e1801bad83a..e2eb2da60f96 100644 --- a/arch/ia64/sn/kernel/io_init.c +++ b/arch/ia64/sn/kernel/io_init.c @@ -269,7 +269,7 @@ sn_io_slot_fixup(struct pci_dev *dev) rom = ioremap(pci_resource_start(dev, PCI_ROM_RESOURCE), size + 1); - image_size = pci_get_rom_size(rom, size + 1); + image_size = pci_get_rom_size(dev, rom, size + 1); dev->resource[PCI_ROM_RESOURCE].end = dev->resource[PCI_ROM_RESOURCE].start + image_size - 1; diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index db7ec14fa719..dfc4e0ddf241 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -768,8 +768,8 @@ pci_read_rom(struct kobject *kobj, struct bin_attribute *bin_attr, return -EINVAL; rom = pci_map_rom(pdev, &size); /* size starts out as PCI window size */ - if (!rom) - return 0; + if (!rom || !size) + return -EIO; if (off >= size) count = 0; diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c index 132a78159b60..29cbe47f219f 100644 --- a/drivers/pci/rom.c +++ b/drivers/pci/rom.c @@ -63,7 +63,7 @@ void pci_disable_rom(struct pci_dev *pdev) * The PCI window size could be much larger than the * actual image size. */ -size_t pci_get_rom_size(void __iomem *rom, size_t size) +size_t pci_get_rom_size(struct pci_dev *pdev, void __iomem *rom, size_t size) { void __iomem *image; int last_image; @@ -72,8 +72,10 @@ size_t pci_get_rom_size(void __iomem *rom, size_t size) do { void __iomem *pds; /* Standard PCI ROMs start out with these bytes 55 AA */ - if (readb(image) != 0x55) + if (readb(image) != 0x55) { + dev_err(&pdev->dev, "Invalid ROM contents\n"); break; + } if (readb(image + 1) != 0xAA) break; /* get the PCI data structure and check its signature */ @@ -159,7 +161,7 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size) * size is much larger than the actual size of the ROM. * True size is important if the ROM is going to be copied. */ - *size = pci_get_rom_size(rom, *size); + *size = pci_get_rom_size(pdev, rom, *size); return rom; } diff --git a/include/linux/pci.h b/include/linux/pci.h index 48890cf3f96e..7bd624bfdcfd 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -684,7 +684,7 @@ int pci_enable_rom(struct pci_dev *pdev); void pci_disable_rom(struct pci_dev *pdev); void __iomem __must_check *pci_map_rom(struct pci_dev *pdev, size_t *size); void pci_unmap_rom(struct pci_dev *pdev, void __iomem *rom); -size_t pci_get_rom_size(void __iomem *rom, size_t size); +size_t pci_get_rom_size(struct pci_dev *pdev, void __iomem *rom, size_t size); /* Power management related routines */ int pci_save_state(struct pci_dev *dev); -- cgit v1.2.3 From 7b2cd92adc5430b0c1adeb120971852b4ea1ab08 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 5 Feb 2009 16:48:24 +1100 Subject: crypto: api - Fix zeroing on free Geert Uytterhoeven pointed out that we're not zeroing all the memory when freeing a transform. This patch fixes it by calling ksize to ensure that we zero everything in sight. Reported-by: Geert Uytterhoeven Signed-off-by: Herbert Xu --- crypto/api.c | 20 ++++++++++---------- include/linux/crypto.h | 7 ++++++- 2 files changed, 16 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/crypto/api.c b/crypto/api.c index 9975a7bd246c..efe77df6863f 100644 --- a/crypto/api.c +++ b/crypto/api.c @@ -557,34 +557,34 @@ err: return ERR_PTR(err); } EXPORT_SYMBOL_GPL(crypto_alloc_tfm); - + /* - * crypto_free_tfm - Free crypto transform + * crypto_destroy_tfm - Free crypto transform + * @mem: Start of tfm slab * @tfm: Transform to free * - * crypto_free_tfm() frees up the transform and any associated resources, + * This function frees up the transform and any associated resources, * then drops the refcount on the associated algorithm. */ -void crypto_free_tfm(struct crypto_tfm *tfm) +void crypto_destroy_tfm(void *mem, struct crypto_tfm *tfm) { struct crypto_alg *alg; int size; - if (unlikely(!tfm)) + if (unlikely(!mem)) return; alg = tfm->__crt_alg; - size = sizeof(*tfm) + alg->cra_ctxsize; + size = ksize(mem); if (!tfm->exit && alg->cra_exit) alg->cra_exit(tfm); crypto_exit_ops(tfm); crypto_mod_put(alg); - memset(tfm, 0, size); - kfree(tfm); + memset(mem, 0, size); + kfree(mem); } - -EXPORT_SYMBOL_GPL(crypto_free_tfm); +EXPORT_SYMBOL_GPL(crypto_destroy_tfm); int crypto_has_alg(const char *name, u32 type, u32 mask) { diff --git a/include/linux/crypto.h b/include/linux/crypto.h index 3bacd71509fb..1f2e9020acc6 100644 --- a/include/linux/crypto.h +++ b/include/linux/crypto.h @@ -552,7 +552,12 @@ struct crypto_tfm *crypto_alloc_tfm(const char *alg_name, const struct crypto_type *frontend, u32 type, u32 mask); struct crypto_tfm *crypto_alloc_base(const char *alg_name, u32 type, u32 mask); -void crypto_free_tfm(struct crypto_tfm *tfm); +void crypto_destroy_tfm(void *mem, struct crypto_tfm *tfm); + +static inline void crypto_free_tfm(struct crypto_tfm *tfm) +{ + return crypto_destroy_tfm(tfm, tfm); +} int alg_test(const char *driver, const char *alg, u32 type, u32 mask); -- cgit v1.2.3 From 32bd671d6cbeda60dc73be77fa2b9037d9a9bfa0 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 5 Feb 2009 12:24:15 +0100 Subject: signal: re-add dead task accumulation stats. We're going to split the process wide cpu accounting into two parts: - clocks; which can take all the time they want since they run from user context. - timers; which need constant time tracing but can affort the overhead because they're default off -- and rare. The clock readout will go back to a full sum of the thread group, for this we need to re-add the exit stats that were removed in the initial itimer rework (f06febc9: timers: fix itimer/many thread hang). Furthermore, since that full sum can be rather slow for large thread groups and we have the complete dead task stats, revert the do_notify_parent time computation. Signed-off-by: Peter Zijlstra Reviewed-by: Ingo Molnar Signed-off-by: Ingo Molnar --- include/linux/sched.h | 10 +++++++++- kernel/exit.c | 3 +++ kernel/fork.c | 3 ++- kernel/signal.c | 8 ++++---- 4 files changed, 18 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 2127e959e0f4..2e0646a30314 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -559,7 +559,7 @@ struct signal_struct { * Live threads maintain their own counters and add to these * in __exit_signal, except for the group leader. */ - cputime_t cutime, cstime; + cputime_t utime, stime, cutime, cstime; cputime_t gtime; cputime_t cgtime; unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw; @@ -567,6 +567,14 @@ struct signal_struct { unsigned long inblock, oublock, cinblock, coublock; struct task_io_accounting ioac; + /* + * Cumulative ns of schedule CPU time fo dead threads in the + * group, not including a zombie group leader, (This only differs + * from jiffies_to_ns(utime + stime) if sched_clock uses something + * other than jiffies.) + */ + unsigned long long sum_sched_runtime; + /* * We don't bother to synchronize most readers of this at all, * because there is no reader checking a limit that actually needs diff --git a/kernel/exit.c b/kernel/exit.c index f80dec3f1875..efd30ccf3858 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -118,6 +118,8 @@ static void __exit_signal(struct task_struct *tsk) * We won't ever get here for the group leader, since it * will have been the last reference on the signal_struct. */ + sig->utime = cputime_add(sig->utime, task_utime(tsk)); + sig->stime = cputime_add(sig->stime, task_stime(tsk)); sig->gtime = cputime_add(sig->gtime, task_gtime(tsk)); sig->min_flt += tsk->min_flt; sig->maj_flt += tsk->maj_flt; @@ -126,6 +128,7 @@ static void __exit_signal(struct task_struct *tsk) sig->inblock += task_io_get_inblock(tsk); sig->oublock += task_io_get_oublock(tsk); task_io_accounting_add(&sig->ioac, &tsk->ioac); + sig->sum_sched_runtime += tsk->se.sum_exec_runtime; sig = NULL; /* Marker for below. */ } diff --git a/kernel/fork.c b/kernel/fork.c index 242a706e7721..e8e854a04ad2 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -851,13 +851,14 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) sig->tty_old_pgrp = NULL; sig->tty = NULL; - sig->cutime = sig->cstime = cputime_zero; + sig->utime = sig->stime = sig->cutime = sig->cstime = cputime_zero; sig->gtime = cputime_zero; sig->cgtime = cputime_zero; sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0; sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0; sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0; task_io_accounting_init(&sig->ioac); + sig->sum_sched_runtime = 0; taskstats_tgid_init(sig); task_lock(current->group_leader); diff --git a/kernel/signal.c b/kernel/signal.c index b6b36768b758..2a74fe87c0dd 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1367,7 +1367,6 @@ int do_notify_parent(struct task_struct *tsk, int sig) struct siginfo info; unsigned long flags; struct sighand_struct *psig; - struct task_cputime cputime; int ret = sig; BUG_ON(sig == -1); @@ -1397,9 +1396,10 @@ int do_notify_parent(struct task_struct *tsk, int sig) info.si_uid = __task_cred(tsk)->uid; rcu_read_unlock(); - thread_group_cputime(tsk, &cputime); - info.si_utime = cputime_to_jiffies(cputime.utime); - info.si_stime = cputime_to_jiffies(cputime.stime); + info.si_utime = cputime_to_clock_t(cputime_add(tsk->utime, + tsk->signal->utime)); + info.si_stime = cputime_to_clock_t(cputime_add(tsk->stime, + tsk->signal->stime)); info.si_status = tsk->exit_code & 0x7f; if (tsk->exit_code & 0x80) -- cgit v1.2.3 From 4cd4c1b40d40447fb5e7ba80746c6d7ba91d7a53 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 5 Feb 2009 12:24:16 +0100 Subject: timers: split process wide cpu clocks/timers Change the process wide cpu timers/clocks so that we: 1) don't mess up the kernel with too many threads, 2) don't have a per-cpu allocation for each process, 3) have no impact when not used. In order to accomplish this we're going to split it into two parts: - clocks; which can take all the time they want since they run from user context -- ie. sys_clock_gettime(CLOCK_PROCESS_CPUTIME_ID) - timers; which need constant time sampling but since they're explicity used, the user can pay the overhead. The clock readout will go back to a full sum of the thread group, while the timers will run of a global 'clock' that only runs when needed, so only programs that make use of the facility pay the price. Signed-off-by: Peter Zijlstra Reviewed-by: Ingo Molnar Signed-off-by: Ingo Molnar --- include/linux/init_task.h | 11 +++--- include/linux/sched.h | 54 +++++++++++++++------------ kernel/itimer.c | 4 +- kernel/posix-cpu-timers.c | 95 +++++++++++++++++++++++++++++++++++++++++++++-- kernel/sched_stats.h | 45 ++++++++++++---------- 5 files changed, 155 insertions(+), 54 deletions(-) (limited to 'include/linux') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index ea0ea1a4c36f..e752d973fa21 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -48,12 +48,11 @@ extern struct fs_struct init_fs; .posix_timers = LIST_HEAD_INIT(sig.posix_timers), \ .cpu_timers = INIT_CPU_TIMERS(sig.cpu_timers), \ .rlim = INIT_RLIMITS, \ - .cputime = { .totals = { \ - .utime = cputime_zero, \ - .stime = cputime_zero, \ - .sum_exec_runtime = 0, \ - .lock = __SPIN_LOCK_UNLOCKED(sig.cputime.totals.lock), \ - }, }, \ + .cputimer = { \ + .cputime = INIT_CPUTIME, \ + .running = 0, \ + .lock = __SPIN_LOCK_UNLOCKED(sig.cputimer.lock), \ + }, \ } extern struct nsproxy init_nsproxy; diff --git a/include/linux/sched.h b/include/linux/sched.h index 2e0646a30314..082d7619b3a1 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -443,7 +443,6 @@ struct pacct_struct { * @utime: time spent in user mode, in &cputime_t units * @stime: time spent in kernel mode, in &cputime_t units * @sum_exec_runtime: total time spent on the CPU, in nanoseconds - * @lock: lock for fields in this struct * * This structure groups together three kinds of CPU time that are * tracked for threads and thread groups. Most things considering @@ -454,23 +453,33 @@ struct task_cputime { cputime_t utime; cputime_t stime; unsigned long long sum_exec_runtime; - spinlock_t lock; }; /* Alternate field names when used to cache expirations. */ #define prof_exp stime #define virt_exp utime #define sched_exp sum_exec_runtime +#define INIT_CPUTIME \ + (struct task_cputime) { \ + .utime = cputime_zero, \ + .stime = cputime_zero, \ + .sum_exec_runtime = 0, \ + } + /** - * struct thread_group_cputime - thread group interval timer counts - * @totals: thread group interval timers; substructure for - * uniprocessor kernel, per-cpu for SMP kernel. + * struct thread_group_cputimer - thread group interval timer counts + * @cputime: thread group interval timers. + * @running: non-zero when there are timers running and + * @cputime receives updates. + * @lock: lock for fields in this struct. * * This structure contains the version of task_cputime, above, that is - * used for thread group CPU clock calculations. + * used for thread group CPU timer calculations. */ -struct thread_group_cputime { - struct task_cputime totals; +struct thread_group_cputimer { + struct task_cputime cputime; + int running; + spinlock_t lock; }; /* @@ -519,10 +528,10 @@ struct signal_struct { cputime_t it_prof_incr, it_virt_incr; /* - * Thread group totals for process CPU clocks. - * See thread_group_cputime(), et al, for details. + * Thread group totals for process CPU timers. + * See thread_group_cputimer(), et al, for details. */ - struct thread_group_cputime cputime; + struct thread_group_cputimer cputimer; /* Earliest-expiration cache. */ struct task_cputime cputime_expires; @@ -2191,27 +2200,26 @@ static inline int spin_needbreak(spinlock_t *lock) /* * Thread group CPU time accounting. */ +void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times); static inline -void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) +void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) { - struct task_cputime *totals = &tsk->signal->cputime.totals; + struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; unsigned long flags; - spin_lock_irqsave(&totals->lock, flags); - *times = *totals; - spin_unlock_irqrestore(&totals->lock, flags); + WARN_ON(!cputimer->running); + + spin_lock_irqsave(&cputimer->lock, flags); + *times = cputimer->cputime; + spin_unlock_irqrestore(&cputimer->lock, flags); } static inline void thread_group_cputime_init(struct signal_struct *sig) { - sig->cputime.totals = (struct task_cputime){ - .utime = cputime_zero, - .stime = cputime_zero, - .sum_exec_runtime = 0, - }; - - spin_lock_init(&sig->cputime.totals.lock); + sig->cputimer.cputime = INIT_CPUTIME; + spin_lock_init(&sig->cputimer.lock); + sig->cputimer.running = 0; } static inline void thread_group_cputime_free(struct signal_struct *sig) diff --git a/kernel/itimer.c b/kernel/itimer.c index 6a5fe93dd8bd..58762f7077ec 100644 --- a/kernel/itimer.c +++ b/kernel/itimer.c @@ -62,7 +62,7 @@ int do_getitimer(int which, struct itimerval *value) struct task_cputime cputime; cputime_t utime; - thread_group_cputime(tsk, &cputime); + thread_group_cputimer(tsk, &cputime); utime = cputime.utime; if (cputime_le(cval, utime)) { /* about to fire */ cval = jiffies_to_cputime(1); @@ -82,7 +82,7 @@ int do_getitimer(int which, struct itimerval *value) struct task_cputime times; cputime_t ptime; - thread_group_cputime(tsk, ×); + thread_group_cputimer(tsk, ×); ptime = cputime_add(times.utime, times.stime); if (cputime_le(cval, ptime)) { /* about to fire */ cval = jiffies_to_cputime(1); diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index fa07da94d7be..db107c9bbc05 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -230,6 +230,37 @@ static int cpu_clock_sample(const clockid_t which_clock, struct task_struct *p, return 0; } +void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) +{ + struct sighand_struct *sighand; + struct signal_struct *sig; + struct task_struct *t; + + *times = INIT_CPUTIME; + + rcu_read_lock(); + sighand = rcu_dereference(tsk->sighand); + if (!sighand) + goto out; + + sig = tsk->signal; + + t = tsk; + do { + times->utime = cputime_add(times->utime, t->utime); + times->stime = cputime_add(times->stime, t->stime); + times->sum_exec_runtime += t->se.sum_exec_runtime; + + t = next_thread(t); + } while (t != tsk); + + times->utime = cputime_add(times->utime, sig->utime); + times->stime = cputime_add(times->stime, sig->stime); + times->sum_exec_runtime += sig->sum_sched_runtime; +out: + rcu_read_unlock(); +} + /* * Sample a process (thread group) clock for the given group_leader task. * Must be called with tasklist_lock held for reading. @@ -475,6 +506,29 @@ static void clear_dead_task(struct k_itimer *timer, union cpu_time_count now) now); } +/* + * Enable the process wide cpu timer accounting. + * + * serialized using ->sighand->siglock + */ +static void start_process_timers(struct task_struct *tsk) +{ + tsk->signal->cputimer.running = 1; + barrier(); +} + +/* + * Release the process wide timer accounting -- timer stops ticking when + * nobody cares about it. + * + * serialized using ->sighand->siglock + */ +static void stop_process_timers(struct task_struct *tsk) +{ + tsk->signal->cputimer.running = 0; + barrier(); +} + /* * Insert the timer on the appropriate list before any timers that * expire later. This must be called with the tasklist_lock held @@ -495,6 +549,9 @@ static void arm_timer(struct k_itimer *timer, union cpu_time_count now) BUG_ON(!irqs_disabled()); spin_lock(&p->sighand->siglock); + if (!CPUCLOCK_PERTHREAD(timer->it_clock)) + start_process_timers(p); + listpos = head; if (CPUCLOCK_WHICH(timer->it_clock) == CPUCLOCK_SCHED) { list_for_each_entry(next, head, entry) { @@ -987,13 +1044,15 @@ static void check_process_timers(struct task_struct *tsk, sig->rlim[RLIMIT_CPU].rlim_cur == RLIM_INFINITY && list_empty(&timers[CPUCLOCK_VIRT]) && cputime_eq(sig->it_virt_expires, cputime_zero) && - list_empty(&timers[CPUCLOCK_SCHED])) + list_empty(&timers[CPUCLOCK_SCHED])) { + stop_process_timers(tsk); return; + } /* * Collect the current process totals. */ - thread_group_cputime(tsk, &cputime); + thread_group_cputimer(tsk, &cputime); utime = cputime.utime; ptime = cputime_add(utime, cputime.stime); sum_sched_runtime = cputime.sum_exec_runtime; @@ -1259,7 +1318,7 @@ static inline int fastpath_timer_check(struct task_struct *tsk) if (!task_cputime_zero(&sig->cputime_expires)) { struct task_cputime group_sample; - thread_group_cputime(tsk, &group_sample); + thread_group_cputimer(tsk, &group_sample); if (task_cputime_expired(&group_sample, &sig->cputime_expires)) return 1; } @@ -1328,6 +1387,33 @@ void run_posix_cpu_timers(struct task_struct *tsk) } } +/* + * Sample a process (thread group) timer for the given group_leader task. + * Must be called with tasklist_lock held for reading. + */ +static int cpu_timer_sample_group(const clockid_t which_clock, + struct task_struct *p, + union cpu_time_count *cpu) +{ + struct task_cputime cputime; + + thread_group_cputimer(p, &cputime); + switch (CPUCLOCK_WHICH(which_clock)) { + default: + return -EINVAL; + case CPUCLOCK_PROF: + cpu->cpu = cputime_add(cputime.utime, cputime.stime); + break; + case CPUCLOCK_VIRT: + cpu->cpu = cputime.utime; + break; + case CPUCLOCK_SCHED: + cpu->sched = cputime.sum_exec_runtime + task_delta_exec(p); + break; + } + return 0; +} + /* * Set one of the process-wide special case CPU timers. * The tsk->sighand->siglock must be held by the caller. @@ -1341,7 +1427,8 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, struct list_head *head; BUG_ON(clock_idx == CPUCLOCK_SCHED); - cpu_clock_sample_group(clock_idx, tsk, &now); + start_process_timers(tsk); + cpu_timer_sample_group(clock_idx, tsk, &now); if (oldval) { if (!cputime_eq(*oldval, cputime_zero)) { diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h index 8ab0cef8ecab..a8f93dd374e1 100644 --- a/kernel/sched_stats.h +++ b/kernel/sched_stats.h @@ -296,19 +296,21 @@ sched_info_switch(struct task_struct *prev, struct task_struct *next) static inline void account_group_user_time(struct task_struct *tsk, cputime_t cputime) { - struct task_cputime *times; - struct signal_struct *sig; + struct thread_group_cputimer *cputimer; /* tsk == current, ensure it is safe to use ->signal */ if (unlikely(tsk->exit_state)) return; - sig = tsk->signal; - times = &sig->cputime.totals; + cputimer = &tsk->signal->cputimer; - spin_lock(×->lock); - times->utime = cputime_add(times->utime, cputime); - spin_unlock(×->lock); + if (!cputimer->running) + return; + + spin_lock(&cputimer->lock); + cputimer->cputime.utime = + cputime_add(cputimer->cputime.utime, cputime); + spin_unlock(&cputimer->lock); } /** @@ -324,19 +326,21 @@ static inline void account_group_user_time(struct task_struct *tsk, static inline void account_group_system_time(struct task_struct *tsk, cputime_t cputime) { - struct task_cputime *times; - struct signal_struct *sig; + struct thread_group_cputimer *cputimer; /* tsk == current, ensure it is safe to use ->signal */ if (unlikely(tsk->exit_state)) return; - sig = tsk->signal; - times = &sig->cputime.totals; + cputimer = &tsk->signal->cputimer; + + if (!cputimer->running) + return; - spin_lock(×->lock); - times->stime = cputime_add(times->stime, cputime); - spin_unlock(×->lock); + spin_lock(&cputimer->lock); + cputimer->cputime.stime = + cputime_add(cputimer->cputime.stime, cputime); + spin_unlock(&cputimer->lock); } /** @@ -352,7 +356,7 @@ static inline void account_group_system_time(struct task_struct *tsk, static inline void account_group_exec_runtime(struct task_struct *tsk, unsigned long long ns) { - struct task_cputime *times; + struct thread_group_cputimer *cputimer; struct signal_struct *sig; sig = tsk->signal; @@ -361,9 +365,12 @@ static inline void account_group_exec_runtime(struct task_struct *tsk, if (unlikely(!sig)) return; - times = &sig->cputime.totals; + cputimer = &sig->cputimer; + + if (!cputimer->running) + return; - spin_lock(×->lock); - times->sum_exec_runtime += ns; - spin_unlock(×->lock); + spin_lock(&cputimer->lock); + cputimer->cputime.sum_exec_runtime += ns; + spin_unlock(&cputimer->lock); } -- cgit v1.2.3 From ac7b9004909d03d67016368093e81d37cae72895 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 4 Feb 2009 15:11:59 -0800 Subject: generic swap(): don't return a value from swap() The swap() macro is accidentally retuning the value of its first argument. Change it into a doesn't-return-anything macro before someone goes and relies upon this behaviour. Signed-off-by: Peter Zijlstra Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 343df9ef2412..7fa371898e3e 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -480,7 +480,8 @@ static inline char *pack_hex_byte(char *buf, u8 byte) /* * swap - swap value of @a and @b */ -#define swap(a, b) ({ typeof(a) __tmp = (a); (a) = (b); (b) = __tmp; }) +#define swap(a, b) \ + do { typeof(a) __tmp = (a); (a) = (b); (b) = __tmp; } while (0) /** * container_of - cast a member of a structure out to the containing structure -- cgit v1.2.3 From 1f5e31d7e55ac7fbd4ec5e5b20c8868b0e4564c9 Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Wed, 4 Feb 2009 15:12:03 -0800 Subject: fbmem: don't call copy_from/to_user() with mutex held Avoid calling copy_from/to_user() with fb_info->lock mutex held in fbmem ioctl(). fb_mmap() is called under mm->mmap_sem (A) held, that also acquires fb_info->lock (B); fb_ioctl() takes fb_info->lock (B) and does copy_from/to_user() that might acquire mm->mmap_sem (A), causing a deadlock. NOTE: it doesn't push down the fb_info->lock in each own driver's fb_ioctl(), so there are still potential deadlocks elsewhere. Signed-off-by: Andrea Righi Cc: Dave Jones Cc: "Rafael J. Wysocki" Cc: Johannes Weiner Cc: Krzysztof Helt Cc: Harvey Harrison Cc: Stefan Richter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/fbcmap.c | 20 +++++--- drivers/video/fbmem.c | 135 +++++++++++++++++++++++++------------------------ include/linux/fb.h | 15 ++++++ 3 files changed, 98 insertions(+), 72 deletions(-) (limited to 'include/linux') diff --git a/drivers/video/fbcmap.c b/drivers/video/fbcmap.c index 91b78e691505..f53b9f1d6aba 100644 --- a/drivers/video/fbcmap.c +++ b/drivers/video/fbcmap.c @@ -250,10 +250,6 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info) int rc, size = cmap->len * sizeof(u16); struct fb_cmap umap; - if (cmap->start < 0 || (!info->fbops->fb_setcolreg && - !info->fbops->fb_setcmap)) - return -EINVAL; - memset(&umap, 0, sizeof(struct fb_cmap)); rc = fb_alloc_cmap(&umap, cmap->len, cmap->transp != NULL); if (rc) @@ -262,11 +258,23 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info) copy_from_user(umap.green, cmap->green, size) || copy_from_user(umap.blue, cmap->blue, size) || (cmap->transp && copy_from_user(umap.transp, cmap->transp, size))) { - fb_dealloc_cmap(&umap); - return -EFAULT; + rc = -EFAULT; + goto out; } umap.start = cmap->start; + if (!lock_fb_info(info)) { + rc = -ENODEV; + goto out; + } + if (cmap->start < 0 || (!info->fbops->fb_setcolreg && + !info->fbops->fb_setcmap)) { + rc = -EINVAL; + goto out1; + } rc = fb_set_cmap(&umap, info); +out1: + unlock_fb_info(info); +out: fb_dealloc_cmap(&umap); return rc; } diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 756efeb91abc..cfd9dce1ce0b 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -1013,132 +1013,139 @@ static long do_fb_ioctl(struct fb_info *info, unsigned int cmd, struct fb_var_screeninfo var; struct fb_fix_screeninfo fix; struct fb_con2fbmap con2fb; + struct fb_cmap cmap_from; struct fb_cmap_user cmap; struct fb_event event; void __user *argp = (void __user *)arg; long ret = 0; - fb = info->fbops; - if (!fb) - return -ENODEV; - switch (cmd) { case FBIOGET_VSCREENINFO: - ret = copy_to_user(argp, &info->var, - sizeof(var)) ? -EFAULT : 0; + if (!lock_fb_info(info)) + return -ENODEV; + var = info->var; + unlock_fb_info(info); + + ret = copy_to_user(argp, &var, sizeof(var)) ? -EFAULT : 0; break; case FBIOPUT_VSCREENINFO: - if (copy_from_user(&var, argp, sizeof(var))) { - ret = -EFAULT; - break; - } + if (copy_from_user(&var, argp, sizeof(var))) + return -EFAULT; + if (!lock_fb_info(info)) + return -ENODEV; acquire_console_sem(); info->flags |= FBINFO_MISC_USEREVENT; ret = fb_set_var(info, &var); info->flags &= ~FBINFO_MISC_USEREVENT; release_console_sem(); - if (ret == 0 && copy_to_user(argp, &var, sizeof(var))) + unlock_fb_info(info); + if (!ret && copy_to_user(argp, &var, sizeof(var))) ret = -EFAULT; break; case FBIOGET_FSCREENINFO: - ret = copy_to_user(argp, &info->fix, - sizeof(fix)) ? -EFAULT : 0; + if (!lock_fb_info(info)) + return -ENODEV; + fix = info->fix; + unlock_fb_info(info); + + ret = copy_to_user(argp, &fix, sizeof(fix)) ? -EFAULT : 0; break; case FBIOPUTCMAP: if (copy_from_user(&cmap, argp, sizeof(cmap))) - ret = -EFAULT; - else - ret = fb_set_user_cmap(&cmap, info); + return -EFAULT; + ret = fb_set_user_cmap(&cmap, info); break; case FBIOGETCMAP: if (copy_from_user(&cmap, argp, sizeof(cmap))) - ret = -EFAULT; - else - ret = fb_cmap_to_user(&info->cmap, &cmap); + return -EFAULT; + if (!lock_fb_info(info)) + return -ENODEV; + cmap_from = info->cmap; + unlock_fb_info(info); + ret = fb_cmap_to_user(&cmap_from, &cmap); break; case FBIOPAN_DISPLAY: - if (copy_from_user(&var, argp, sizeof(var))) { - ret = -EFAULT; - break; - } + if (copy_from_user(&var, argp, sizeof(var))) + return -EFAULT; + if (!lock_fb_info(info)) + return -ENODEV; acquire_console_sem(); ret = fb_pan_display(info, &var); release_console_sem(); + unlock_fb_info(info); if (ret == 0 && copy_to_user(argp, &var, sizeof(var))) - ret = -EFAULT; + return -EFAULT; break; case FBIO_CURSOR: ret = -EINVAL; break; case FBIOGET_CON2FBMAP: if (copy_from_user(&con2fb, argp, sizeof(con2fb))) - ret = -EFAULT; - else if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) - ret = -EINVAL; - else { - con2fb.framebuffer = -1; - event.info = info; - event.data = &con2fb; - fb_notifier_call_chain(FB_EVENT_GET_CONSOLE_MAP, - &event); - ret = copy_to_user(argp, &con2fb, - sizeof(con2fb)) ? -EFAULT : 0; - } + return -EFAULT; + if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) + return -EINVAL; + con2fb.framebuffer = -1; + event.data = &con2fb; + + if (!lock_fb_info(info)) + return -ENODEV; + event.info = info; + fb_notifier_call_chain(FB_EVENT_GET_CONSOLE_MAP, &event); + unlock_fb_info(info); + + ret = copy_to_user(argp, &con2fb, sizeof(con2fb)) ? -EFAULT : 0; break; case FBIOPUT_CON2FBMAP: - if (copy_from_user(&con2fb, argp, sizeof(con2fb))) { - ret = -EFAULT; - break; - } - if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) { - ret = -EINVAL; - break; - } - if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX) { - ret = -EINVAL; - break; - } + if (copy_from_user(&con2fb, argp, sizeof(con2fb))) + return -EFAULT; + if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) + return -EINVAL; + if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX) + return -EINVAL; if (!registered_fb[con2fb.framebuffer]) request_module("fb%d", con2fb.framebuffer); if (!registered_fb[con2fb.framebuffer]) { ret = -EINVAL; break; } - event.info = info; event.data = &con2fb; + if (!lock_fb_info(info)) + return -ENODEV; + event.info = info; ret = fb_notifier_call_chain(FB_EVENT_SET_CONSOLE_MAP, &event); + unlock_fb_info(info); break; case FBIOBLANK: + if (!lock_fb_info(info)) + return -ENODEV; acquire_console_sem(); info->flags |= FBINFO_MISC_USEREVENT; ret = fb_blank(info, arg); info->flags &= ~FBINFO_MISC_USEREVENT; release_console_sem(); - break;; + unlock_fb_info(info); + break; default: - if (fb->fb_ioctl == NULL) - ret = -ENOTTY; - else + if (!lock_fb_info(info)) + return -ENODEV; + fb = info->fbops; + if (fb->fb_ioctl) ret = fb->fb_ioctl(info, cmd, arg); + else + ret = -ENOTTY; + unlock_fb_info(info); } return ret; } static long fb_ioctl(struct file *file, unsigned int cmd, unsigned long arg) -__acquires(&info->lock) -__releases(&info->lock) { struct inode *inode = file->f_path.dentry->d_inode; int fbidx = iminor(inode); - struct fb_info *info; - long ret; + struct fb_info *info = registered_fb[fbidx]; - info = registered_fb[fbidx]; - mutex_lock(&info->lock); - ret = do_fb_ioctl(info, cmd, arg); - mutex_unlock(&info->lock); - return ret; + return do_fb_ioctl(info, cmd, arg); } #ifdef CONFIG_COMPAT @@ -1257,8 +1264,6 @@ static int fb_get_fscreeninfo(struct fb_info *info, unsigned int cmd, static long fb_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) -__acquires(&info->lock) -__releases(&info->lock) { struct inode *inode = file->f_path.dentry->d_inode; int fbidx = iminor(inode); @@ -1266,7 +1271,6 @@ __releases(&info->lock) struct fb_ops *fb = info->fbops; long ret = -ENOIOCTLCMD; - mutex_lock(&info->lock); switch(cmd) { case FBIOGET_VSCREENINFO: case FBIOPUT_VSCREENINFO: @@ -1292,7 +1296,6 @@ __releases(&info->lock) ret = fb->fb_compat_ioctl(info, cmd, arg); break; } - mutex_unlock(&info->lock); return ret; } #endif diff --git a/include/linux/fb.h b/include/linux/fb.h index 818fe21257e8..31527e17076b 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -960,6 +960,21 @@ extern struct fb_info *registered_fb[FB_MAX]; extern int num_registered_fb; extern struct class *fb_class; +static inline int lock_fb_info(struct fb_info *info) +{ + mutex_lock(&info->lock); + if (!info->fbops) { + mutex_unlock(&info->lock); + return 0; + } + return 1; +} + +static inline void unlock_fb_info(struct fb_info *info) +{ + mutex_unlock(&info->lock); +} + static inline void __fb_pad_aligned_buffer(u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, u32 height) { -- cgit v1.2.3 From 777c6c5f1f6e757ae49ecca2ed72d6b1f523c007 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 4 Feb 2009 15:12:14 -0800 Subject: wait: prevent exclusive waiter starvation With exclusive waiters, every process woken up through the wait queue must ensure that the next waiter down the line is woken when it has finished. Interruptible waiters don't do that when aborting due to a signal. And if an aborting waiter is concurrently woken up through the waitqueue, noone will ever wake up the next waiter. This has been observed with __wait_on_bit_lock() used by lock_page_killable(): the first contender on the queue was aborting when the actual lock holder woke it up concurrently. The aborted contender didn't acquire the lock and therefor never did an unlock followed by waking up the next waiter. Add abort_exclusive_wait() which removes the process' wait descriptor from the waitqueue, iff still queued, or wakes up the next waiter otherwise. It does so under the waitqueue lock. Racing with a wake up means the aborting process is either already woken (removed from the queue) and will wake up the next waiter, or it will remove itself from the queue and the concurrent wake up will apply to the next waiter after it. Use abort_exclusive_wait() in __wait_event_interruptible_exclusive() and __wait_on_bit_lock() when they were interrupted by other means than a wake up through the queue. [akpm@linux-foundation.org: coding-style fixes] Reported-by: Chris Mason Signed-off-by: Johannes Weiner Mentored-by: Oleg Nesterov Cc: Peter Zijlstra Cc: Matthew Wilcox Cc: Chuck Lever Cc: Nick Piggin Cc: Ingo Molnar Cc: ["after some testing"] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/wait.h | 11 ++++++++-- kernel/sched.c | 4 ++-- kernel/wait.c | 59 +++++++++++++++++++++++++++++++++++++++++++++------- 3 files changed, 63 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/wait.h b/include/linux/wait.h index ef609f842fac..a210ede73b56 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -132,6 +132,8 @@ static inline void __remove_wait_queue(wait_queue_head_t *head, list_del(&old->task_list); } +void __wake_up_common(wait_queue_head_t *q, unsigned int mode, + int nr_exclusive, int sync, void *key); void __wake_up(wait_queue_head_t *q, unsigned int mode, int nr, void *key); extern void __wake_up_locked(wait_queue_head_t *q, unsigned int mode); extern void __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr); @@ -333,16 +335,19 @@ do { \ for (;;) { \ prepare_to_wait_exclusive(&wq, &__wait, \ TASK_INTERRUPTIBLE); \ - if (condition) \ + if (condition) { \ + finish_wait(&wq, &__wait); \ break; \ + } \ if (!signal_pending(current)) { \ schedule(); \ continue; \ } \ ret = -ERESTARTSYS; \ + abort_exclusive_wait(&wq, &__wait, \ + TASK_INTERRUPTIBLE, NULL); \ break; \ } \ - finish_wait(&wq, &__wait); \ } while (0) #define wait_event_interruptible_exclusive(wq, condition) \ @@ -431,6 +436,8 @@ extern long interruptible_sleep_on_timeout(wait_queue_head_t *q, void prepare_to_wait(wait_queue_head_t *q, wait_queue_t *wait, int state); void prepare_to_wait_exclusive(wait_queue_head_t *q, wait_queue_t *wait, int state); void finish_wait(wait_queue_head_t *q, wait_queue_t *wait); +void abort_exclusive_wait(wait_queue_head_t *q, wait_queue_t *wait, + unsigned int mode, void *key); int autoremove_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key); int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key); diff --git a/kernel/sched.c b/kernel/sched.c index 242d0d47a70d..8ee437a5ec1d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4697,8 +4697,8 @@ EXPORT_SYMBOL(default_wake_function); * started to run but is not in state TASK_RUNNING. try_to_wake_up() returns * zero in this (rare) case, and we handle it by continuing to scan the queue. */ -static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, - int nr_exclusive, int sync, void *key) +void __wake_up_common(wait_queue_head_t *q, unsigned int mode, + int nr_exclusive, int sync, void *key) { wait_queue_t *curr, *next; diff --git a/kernel/wait.c b/kernel/wait.c index cd87131f2fc2..42a2dbc181c8 100644 --- a/kernel/wait.c +++ b/kernel/wait.c @@ -91,6 +91,15 @@ prepare_to_wait_exclusive(wait_queue_head_t *q, wait_queue_t *wait, int state) } EXPORT_SYMBOL(prepare_to_wait_exclusive); +/* + * finish_wait - clean up after waiting in a queue + * @q: waitqueue waited on + * @wait: wait descriptor + * + * Sets current thread back to running state and removes + * the wait descriptor from the given waitqueue if still + * queued. + */ void finish_wait(wait_queue_head_t *q, wait_queue_t *wait) { unsigned long flags; @@ -117,6 +126,39 @@ void finish_wait(wait_queue_head_t *q, wait_queue_t *wait) } EXPORT_SYMBOL(finish_wait); +/* + * abort_exclusive_wait - abort exclusive waiting in a queue + * @q: waitqueue waited on + * @wait: wait descriptor + * @state: runstate of the waiter to be woken + * @key: key to identify a wait bit queue or %NULL + * + * Sets current thread back to running state and removes + * the wait descriptor from the given waitqueue if still + * queued. + * + * Wakes up the next waiter if the caller is concurrently + * woken up through the queue. + * + * This prevents waiter starvation where an exclusive waiter + * aborts and is woken up concurrently and noone wakes up + * the next waiter. + */ +void abort_exclusive_wait(wait_queue_head_t *q, wait_queue_t *wait, + unsigned int mode, void *key) +{ + unsigned long flags; + + __set_current_state(TASK_RUNNING); + spin_lock_irqsave(&q->lock, flags); + if (!list_empty(&wait->task_list)) + list_del_init(&wait->task_list); + else if (waitqueue_active(q)) + __wake_up_common(q, mode, 1, 0, key); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL(abort_exclusive_wait); + int autoremove_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key) { int ret = default_wake_function(wait, mode, sync, key); @@ -177,17 +219,20 @@ int __sched __wait_on_bit_lock(wait_queue_head_t *wq, struct wait_bit_queue *q, int (*action)(void *), unsigned mode) { - int ret = 0; - do { + int ret; + prepare_to_wait_exclusive(wq, &q->wait, mode); - if (test_bit(q->key.bit_nr, q->key.flags)) { - if ((ret = (*action)(q->key.flags))) - break; - } + if (!test_bit(q->key.bit_nr, q->key.flags)) + continue; + ret = action(q->key.flags); + if (!ret) + continue; + abort_exclusive_wait(wq, &q->wait, mode, &q->key); + return ret; } while (test_and_set_bit(q->key.bit_nr, q->key.flags)); finish_wait(wq, &q->wait); - return ret; + return 0; } EXPORT_SYMBOL(__wait_on_bit_lock); -- cgit v1.2.3 From 7d8e23df69820e6be42bcc41d441f4860e8c76f7 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 6 Feb 2009 14:57:51 +0100 Subject: timers: split process wide cpu clocks/timers, remove spurious warning Mike Galbraith reported that the new warning in thread_group_cputimer() triggers en masse with Amarok running. Oleg Nesterov observed: Can't fastpath_timer_check()->thread_group_cputimer() have the false warning too? Suppose we had the timer, then posix_cpu_timer_del() removes this timer, but task_cputime_zero(&sig->cputime_expires) still not true. Remove the spurious debug warning. Reported-by: Mike Galbraith Explained-by: Oleg Nesterov Signed-off-by: Ingo Molnar --- include/linux/sched.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 082d7619b3a1..79392916d6c9 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2208,8 +2208,6 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; unsigned long flags; - WARN_ON(!cputimer->running); - spin_lock_irqsave(&cputimer->lock, flags); *times = cputimer->cputime; spin_unlock_irqrestore(&cputimer->lock, flags); -- cgit v1.2.3 From b4bd07c20ba0c1fa7ad09ba257e0a5cfc2bf6bb3 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 6 Feb 2009 22:06:43 -0800 Subject: net_dma: call dmaengine_get only if NET_DMA enabled Based upon a patch from Atsushi Nemoto -------------------- The commit 649274d993212e7c23c0cb734572c2311c200872 ("net_dma: acquire/release dma channels on ifup/ifdown") added unconditional call of dmaengine_get() to net_dma. The API should be called only if NET_DMA was enabled. -------------------- Signed-off-by: David S. Miller Acked-by: Dan Williams --- include/linux/dmaengine.h | 12 ++++++++++++ net/core/dev.c | 4 ++-- 2 files changed, 14 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index 3e0f64c335c8..3e68469c1885 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -282,6 +282,18 @@ static inline void dmaengine_put(void) } #endif +#ifdef CONFIG_NET_DMA +#define net_dmaengine_get() dmaengine_get() +#define net_dmaengine_put() dmaengine_put() +#else +static inline void net_dmaengine_get(void) +{ +} +static inline void net_dmaengine_put(void) +{ +} +#endif + dma_cookie_t dma_async_memcpy_buf_to_buf(struct dma_chan *chan, void *dest, void *src, size_t len); dma_cookie_t dma_async_memcpy_buf_to_pg(struct dma_chan *chan, diff --git a/net/core/dev.c b/net/core/dev.c index 5379b0c1190a..a17e00662363 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1090,7 +1090,7 @@ int dev_open(struct net_device *dev) /* * Enable NET_DMA */ - dmaengine_get(); + net_dmaengine_get(); /* * Initialize multicasting status @@ -1172,7 +1172,7 @@ int dev_close(struct net_device *dev) /* * Shutdown NET_DMA */ - dmaengine_put(); + net_dmaengine_put(); return 0; } -- cgit v1.2.3 From 7f9a50a5b89b87f8e754f59ae9968da28be618a5 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 7 Feb 2009 18:15:56 +1030 Subject: module: remove over-zealous check in __module_get() Impact: fix spurious BUG_ON() triggered under load module_refcount() isn't reliable outside stop_machine(), as demonstrated by Karsten Keil , networking can trigger it under load (an inc on one cpu and dec on another while module_refcount() is tallying can give false results, for example). Almost noone should be using __module_get, but that's another issue. Cc: Karsten Keil Signed-off-by: Rusty Russell Signed-off-by: Linus Torvalds --- include/linux/module.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/module.h b/include/linux/module.h index f3b8329eb5b8..145a75528cc1 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -407,7 +407,6 @@ static inline local_t *__module_ref_addr(struct module *mod, int cpu) static inline void __module_get(struct module *module) { if (module) { - BUG_ON(module_refcount(module) == 0); local_inc(__module_ref_addr(module, get_cpu())); put_cpu(); } -- cgit v1.2.3 From 766ccb9ed406c230d13c145def08ebea1b932982 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Tue, 20 Jan 2009 15:31:31 +0100 Subject: async: Rename _special -> _domain for clarity. Rename the async_*_special() functions to async_*_domain(), which describes the purpose of these functions much better. [Broke up long lines to silence checkpatch] Signed-off-by: Cornelia Huck Signed-off-by: Arjan van de Ven --- fs/super.c | 4 ++-- include/linux/async.h | 8 +++++--- kernel/async.c | 41 ++++++++++++++++++++++------------------- 3 files changed, 29 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/fs/super.c b/fs/super.c index 645e5403f2a0..61dce001dd57 100644 --- a/fs/super.c +++ b/fs/super.c @@ -301,7 +301,7 @@ void generic_shutdown_super(struct super_block *sb) /* * wait for asynchronous fs operations to finish before going further */ - async_synchronize_full_special(&sb->s_async_list); + async_synchronize_full_domain(&sb->s_async_list); /* bad name - it should be evict_inodes() */ invalidate_inodes(sb); @@ -470,7 +470,7 @@ restart: sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); - async_synchronize_full_special(&sb->s_async_list); + async_synchronize_full_domain(&sb->s_async_list); if (sb->s_root && (wait || sb->s_dirt)) sb->s_op->sync_fs(sb, wait); up_read(&sb->s_umount); diff --git a/include/linux/async.h b/include/linux/async.h index c4ecacd0b327..68a9530196f2 100644 --- a/include/linux/async.h +++ b/include/linux/async.h @@ -17,9 +17,11 @@ typedef u64 async_cookie_t; typedef void (async_func_ptr) (void *data, async_cookie_t cookie); extern async_cookie_t async_schedule(async_func_ptr *ptr, void *data); -extern async_cookie_t async_schedule_special(async_func_ptr *ptr, void *data, struct list_head *list); +extern async_cookie_t async_schedule_domain(async_func_ptr *ptr, void *data, + struct list_head *list); extern void async_synchronize_full(void); -extern void async_synchronize_full_special(struct list_head *list); +extern void async_synchronize_full_domain(struct list_head *list); extern void async_synchronize_cookie(async_cookie_t cookie); -extern void async_synchronize_cookie_special(async_cookie_t cookie, struct list_head *list); +extern void async_synchronize_cookie_domain(async_cookie_t cookie, + struct list_head *list); diff --git a/kernel/async.c b/kernel/async.c index b5f0d4b94937..e23399d88bac 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -224,22 +224,23 @@ async_cookie_t async_schedule(async_func_ptr *ptr, void *data) EXPORT_SYMBOL_GPL(async_schedule); /** - * async_schedule_special - schedule a function for asynchronous execution with a special running queue + * async_schedule_domain - schedule a function for asynchronous execution within a certain domain * @ptr: function to execute asynchronously * @data: data pointer to pass to the function - * @running: list head to add to while running + * @running: running list for the domain * * Returns an async_cookie_t that may be used for checkpointing later. - * @running may be used in the async_synchronize_*_special() functions - * to wait on a special running queue rather than on the global running - * queue. + * @running may be used in the async_synchronize_*_domain() functions + * to wait within a certain synchronization domain rather than globally. + * A synchronization domain is specified via the running queue @running to use. * Note: This function may be called from atomic or non-atomic contexts. */ -async_cookie_t async_schedule_special(async_func_ptr *ptr, void *data, struct list_head *running) +async_cookie_t async_schedule_domain(async_func_ptr *ptr, void *data, + struct list_head *running) { return __async_schedule(ptr, data, running); } -EXPORT_SYMBOL_GPL(async_schedule_special); +EXPORT_SYMBOL_GPL(async_schedule_domain); /** * async_synchronize_full - synchronize all asynchronous function calls @@ -255,27 +256,29 @@ void async_synchronize_full(void) EXPORT_SYMBOL_GPL(async_synchronize_full); /** - * async_synchronize_full_special - synchronize all asynchronous function calls for a running list + * async_synchronize_full_domain - synchronize all asynchronous function within a certain domain * @list: running list to synchronize on * - * This function waits until all asynchronous function calls for the running - * list @list have been done. + * This function waits until all asynchronous function calls for the + * synchronization domain specified by the running list @list have been done. */ -void async_synchronize_full_special(struct list_head *list) +void async_synchronize_full_domain(struct list_head *list) { - async_synchronize_cookie_special(next_cookie, list); + async_synchronize_cookie_domain(next_cookie, list); } -EXPORT_SYMBOL_GPL(async_synchronize_full_special); +EXPORT_SYMBOL_GPL(async_synchronize_full_domain); /** - * async_synchronize_cookie_special - synchronize asynchronous function calls on a running list with cookie checkpointing + * async_synchronize_cookie_domain - synchronize asynchronous function calls within a certain domain with cookie checkpointing * @cookie: async_cookie_t to use as checkpoint * @running: running list to synchronize on * - * This function waits until all asynchronous function calls for the running - * list @list submitted prior to @cookie have been done. + * This function waits until all asynchronous function calls for the + * synchronization domain specified by the running list @list submitted + * prior to @cookie have been done. */ -void async_synchronize_cookie_special(async_cookie_t cookie, struct list_head *running) +void async_synchronize_cookie_domain(async_cookie_t cookie, + struct list_head *running) { ktime_t starttime, delta, endtime; @@ -295,7 +298,7 @@ void async_synchronize_cookie_special(async_cookie_t cookie, struct list_head *r (long long)ktime_to_ns(delta) >> 10); } } -EXPORT_SYMBOL_GPL(async_synchronize_cookie_special); +EXPORT_SYMBOL_GPL(async_synchronize_cookie_domain); /** * async_synchronize_cookie - synchronize asynchronous function calls with cookie checkpointing @@ -306,7 +309,7 @@ EXPORT_SYMBOL_GPL(async_synchronize_cookie_special); */ void async_synchronize_cookie(async_cookie_t cookie) { - async_synchronize_cookie_special(cookie, &async_running); + async_synchronize_cookie_domain(cookie, &async_running); } EXPORT_SYMBOL_GPL(async_synchronize_cookie); -- cgit v1.2.3 From a5ef7ca0e2636bad0ccd07b996d775348ae2b65e Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Sun, 8 Feb 2009 17:39:58 -0500 Subject: x86: spinlocks: define dummy __raw_spin_is_contended Architectures other than mips and x86 are not using ticket spinlocks. Therefore, the contention on the lock is meaningless, since there is nobody known to be waiting on it (arguably /fairly/ unfair locks). Dummy it out to return 0 on other architectures. Signed-off-by: Kyle McMartin Acked-by: Ralf Baechle Acked-by: Ingo Molnar Signed-off-by: Linus Torvalds --- arch/mips/include/asm/spinlock.h | 1 + arch/x86/include/asm/paravirt.h | 1 + arch/x86/include/asm/spinlock.h | 1 + include/linux/spinlock.h | 5 +++++ 4 files changed, 8 insertions(+) (limited to 'include/linux') diff --git a/arch/mips/include/asm/spinlock.h b/arch/mips/include/asm/spinlock.h index 1a1f320c30d8..0884947ebe27 100644 --- a/arch/mips/include/asm/spinlock.h +++ b/arch/mips/include/asm/spinlock.h @@ -51,6 +51,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock) return (((counters >> 14) - counters) & 0x1fff) > 1; } +#define __raw_spin_is_contended __raw_spin_is_contended static inline void __raw_spin_lock(raw_spinlock_t *lock) { diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index ba3e2ff6aedc..c09a14127584 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -1402,6 +1402,7 @@ static inline int __raw_spin_is_contended(struct raw_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_is_contended, lock); } +#define __raw_spin_is_contended __raw_spin_is_contended static __always_inline void __raw_spin_lock(struct raw_spinlock *lock) { diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index d17c91981da2..8247e94ac6b1 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -245,6 +245,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock) { return __ticket_spin_is_contended(lock); } +#define __raw_spin_is_contended __raw_spin_is_contended static __always_inline void __raw_spin_lock(raw_spinlock_t *lock) { diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index e0c0fccced46..a0c66a2e00ad 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -124,7 +124,12 @@ do { \ #ifdef CONFIG_GENERIC_LOCKBREAK #define spin_is_contended(lock) ((lock)->break_lock) #else + +#ifdef __raw_spin_is_contended #define spin_is_contended(lock) __raw_spin_is_contended(&(lock)->raw_lock) +#else +#define spin_is_contended(lock) (((void)(lock), 0)) +#endif /*__raw_spin_is_contended*/ #endif /** -- cgit v1.2.3 From 5a6fe125950676015f5108fb71b2a67441755003 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Tue, 10 Feb 2009 14:02:27 +0000 Subject: Do not account for the address space used by hugetlbfs using VM_ACCOUNT When overcommit is disabled, the core VM accounts for pages used by anonymous shared, private mappings and special mappings. It keeps track of VMAs that should be accounted for with VM_ACCOUNT and VMAs that never had a reserve with VM_NORESERVE. Overcommit for hugetlbfs is much riskier than overcommit for base pages due to contiguity requirements. It avoids overcommiting on both shared and private mappings using reservation counters that are checked and updated during mmap(). This ensures (within limits) that hugepages exist in the future when faults occurs or it is too easy to applications to be SIGKILLed. As hugetlbfs makes its own reservations of a different unit to the base page size, VM_ACCOUNT should never be set. Even if the units were correct, we would double account for the usage in the core VM and hugetlbfs. VM_NORESERVE may be set because an application can request no reserves be made for hugetlbfs at the risk of getting killed later. With commit fc8744adc870a8d4366908221508bb113d8b72ee, VM_NORESERVE and VM_ACCOUNT are getting unconditionally set for hugetlbfs-backed mappings. This breaks the accounting for both the core VM and hugetlbfs, can trigger an OOM storm when hugepage pools are too small lockups and corrupted counters otherwise are used. This patch brings hugetlbfs more in line with how the core VM treats VM_NORESERVE but prevents VM_ACCOUNT being set. Signed-off-by: Mel Gorman Signed-off-by: Linus Torvalds --- fs/hugetlbfs/inode.c | 8 +++++--- include/linux/hugetlb.h | 5 +++-- include/linux/mm.h | 3 +-- ipc/shm.c | 8 +++++--- mm/fremap.c | 2 +- mm/hugetlb.c | 39 +++++++++++++++++++++++++-------------- mm/mmap.c | 38 ++++++++++++++++++++++---------------- mm/mprotect.c | 5 +++-- 8 files changed, 65 insertions(+), 43 deletions(-) (limited to 'include/linux') diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 6903d37af037..9b800d97a687 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -108,7 +108,8 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) if (hugetlb_reserve_pages(inode, vma->vm_pgoff >> huge_page_order(h), - len >> huge_page_shift(h), vma)) + len >> huge_page_shift(h), vma, + vma->vm_flags)) goto out; ret = 0; @@ -947,7 +948,7 @@ static int can_do_hugetlb_shm(void) can_do_mlock()); } -struct file *hugetlb_file_setup(const char *name, size_t size) +struct file *hugetlb_file_setup(const char *name, size_t size, int acctflag) { int error = -ENOMEM; struct file *file; @@ -981,7 +982,8 @@ struct file *hugetlb_file_setup(const char *name, size_t size) error = -ENOMEM; if (hugetlb_reserve_pages(inode, 0, - size >> huge_page_shift(hstate_inode(inode)), NULL)) + size >> huge_page_shift(hstate_inode(inode)), NULL, + acctflag)) goto out_inode; d_instantiate(dentry, inode); diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index f1d2fba19ea0..af09660001c7 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -33,7 +33,8 @@ unsigned long hugetlb_total_pages(void); int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, int write_access); int hugetlb_reserve_pages(struct inode *inode, long from, long to, - struct vm_area_struct *vma); + struct vm_area_struct *vma, + int acctflags); void hugetlb_unreserve_pages(struct inode *inode, long offset, long freed); extern unsigned long hugepages_treat_as_movable; @@ -138,7 +139,7 @@ static inline struct hugetlbfs_sb_info *HUGETLBFS_SB(struct super_block *sb) extern const struct file_operations hugetlbfs_file_operations; extern struct vm_operations_struct hugetlb_vm_ops; -struct file *hugetlb_file_setup(const char *name, size_t); +struct file *hugetlb_file_setup(const char *name, size_t, int); int hugetlb_get_quota(struct address_space *mapping, long delta); void hugetlb_put_quota(struct address_space *mapping, long delta); diff --git a/include/linux/mm.h b/include/linux/mm.h index e8ddc98b8405..323561582c10 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1129,8 +1129,7 @@ extern unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, unsigned long flag, unsigned long pgoff); extern unsigned long mmap_region(struct file *file, unsigned long addr, unsigned long len, unsigned long flags, - unsigned int vm_flags, unsigned long pgoff, - int accountable); + unsigned int vm_flags, unsigned long pgoff); static inline unsigned long do_mmap(struct file *file, unsigned long addr, unsigned long len, unsigned long prot, diff --git a/ipc/shm.c b/ipc/shm.c index f8f69fad3a27..05d51d2a792c 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -340,6 +340,7 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) struct file * file; char name[13]; int id; + int acctflag = 0; if (size < SHMMIN || size > ns->shm_ctlmax) return -EINVAL; @@ -364,11 +365,12 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) sprintf (name, "SYSV%08x", key); if (shmflg & SHM_HUGETLB) { - /* hugetlb_file_setup takes care of mlock user accounting */ - file = hugetlb_file_setup(name, size); + /* hugetlb_file_setup applies strict accounting */ + if (shmflg & SHM_NORESERVE) + acctflag = VM_NORESERVE; + file = hugetlb_file_setup(name, size, acctflag); shp->mlock_user = current_user(); } else { - int acctflag = 0; /* * Do not allow no accounting for OVERCOMMIT_NEVER, even * if it's asked for. diff --git a/mm/fremap.c b/mm/fremap.c index 736ba7f3306a..b6ec85abbb39 100644 --- a/mm/fremap.c +++ b/mm/fremap.c @@ -198,7 +198,7 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size, flags &= MAP_NONBLOCK; get_file(file); addr = mmap_region(file, start, size, - flags, vma->vm_flags, pgoff, 1); + flags, vma->vm_flags, pgoff); fput(file); if (IS_ERR_VALUE(addr)) { err = addr; diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 618e98304080..207464209546 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -2269,14 +2269,12 @@ void hugetlb_change_protection(struct vm_area_struct *vma, int hugetlb_reserve_pages(struct inode *inode, long from, long to, - struct vm_area_struct *vma) + struct vm_area_struct *vma, + int acctflag) { - long ret, chg; + long ret = 0, chg; struct hstate *h = hstate_inode(inode); - if (vma && vma->vm_flags & VM_NORESERVE) - return 0; - /* * Shared mappings base their reservation on the number of pages that * are already allocated on behalf of the file. Private mappings need @@ -2285,22 +2283,25 @@ int hugetlb_reserve_pages(struct inode *inode, */ if (!vma || vma->vm_flags & VM_SHARED) chg = region_chg(&inode->i_mapping->private_list, from, to); - else { - struct resv_map *resv_map = resv_map_alloc(); - if (!resv_map) - return -ENOMEM; - + else chg = to - from; - set_vma_resv_map(vma, resv_map); - set_vma_resv_flags(vma, HPAGE_RESV_OWNER); - } - if (chg < 0) return chg; if (hugetlb_get_quota(inode->i_mapping, chg)) return -ENOSPC; + + /* + * Only apply hugepage reservation if asked. We still have to + * take the filesystem quota because it is an upper limit + * defined for the mount and not necessarily memory as a whole + */ + if (acctflag & VM_NORESERVE) { + reset_vma_resv_huge_pages(vma); + return 0; + } + ret = hugetlb_acct_memory(h, chg); if (ret < 0) { hugetlb_put_quota(inode->i_mapping, chg); @@ -2308,6 +2309,16 @@ int hugetlb_reserve_pages(struct inode *inode, } if (!vma || vma->vm_flags & VM_SHARED) region_add(&inode->i_mapping->private_list, from, to); + else { + struct resv_map *resv_map = resv_map_alloc(); + + if (!resv_map) + return -ENOMEM; + + set_vma_resv_map(vma, resv_map); + set_vma_resv_flags(vma, HPAGE_RESV_OWNER); + } + return 0; } diff --git a/mm/mmap.c b/mm/mmap.c index 214b6a258eeb..eb1270bebe67 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -918,7 +918,6 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, struct inode *inode; unsigned int vm_flags; int error; - int accountable = 1; unsigned long reqprot = prot; /* @@ -1019,8 +1018,6 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, return -EPERM; vm_flags &= ~VM_MAYEXEC; } - if (is_file_hugepages(file)) - accountable = 0; if (!file->f_op || !file->f_op->mmap) return -ENODEV; @@ -1053,8 +1050,7 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, if (error) return error; - return mmap_region(file, addr, len, flags, vm_flags, pgoff, - accountable); + return mmap_region(file, addr, len, flags, vm_flags, pgoff); } EXPORT_SYMBOL(do_mmap_pgoff); @@ -1092,17 +1088,23 @@ int vma_wants_writenotify(struct vm_area_struct *vma) /* * We account for memory if it's a private writeable mapping, - * and VM_NORESERVE wasn't set. + * not hugepages and VM_NORESERVE wasn't set. */ -static inline int accountable_mapping(unsigned int vm_flags) +static inline int accountable_mapping(struct file *file, unsigned int vm_flags) { + /* + * hugetlb has its own accounting separate from the core VM + * VM_HUGETLB may not be set yet so we cannot check for that flag. + */ + if (file && is_file_hugepages(file)) + return 0; + return (vm_flags & (VM_NORESERVE | VM_SHARED | VM_WRITE)) == VM_WRITE; } unsigned long mmap_region(struct file *file, unsigned long addr, unsigned long len, unsigned long flags, - unsigned int vm_flags, unsigned long pgoff, - int accountable) + unsigned int vm_flags, unsigned long pgoff) { struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev; @@ -1128,18 +1130,22 @@ munmap_back: /* * Set 'VM_NORESERVE' if we should not account for the - * memory use of this mapping. We only honor MAP_NORESERVE - * if we're allowed to overcommit memory. + * memory use of this mapping. */ - if ((flags & MAP_NORESERVE) && sysctl_overcommit_memory != OVERCOMMIT_NEVER) - vm_flags |= VM_NORESERVE; - if (!accountable) - vm_flags |= VM_NORESERVE; + if ((flags & MAP_NORESERVE)) { + /* We honor MAP_NORESERVE if allowed to overcommit */ + if (sysctl_overcommit_memory != OVERCOMMIT_NEVER) + vm_flags |= VM_NORESERVE; + + /* hugetlb applies strict overcommit unless MAP_NORESERVE */ + if (file && is_file_hugepages(file)) + vm_flags |= VM_NORESERVE; + } /* * Private writable mapping: check memory availability */ - if (accountable_mapping(vm_flags)) { + if (accountable_mapping(file, vm_flags)) { charged = len >> PAGE_SHIFT; if (security_vm_enough_memory(charged)) return -ENOMEM; diff --git a/mm/mprotect.c b/mm/mprotect.c index abe2694e13f4..258197b76fb4 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -151,10 +151,11 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev, /* * If we make a private mapping writable we increase our commit; * but (without finer accounting) cannot reduce our commit if we - * make it unwritable again. + * make it unwritable again. hugetlb mapping were accounted for + * even if read-only so there is no need to account for them here */ if (newflags & VM_WRITE) { - if (!(oldflags & (VM_ACCOUNT|VM_WRITE| + if (!(oldflags & (VM_ACCOUNT|VM_WRITE|VM_HUGETLB| VM_SHARED|VM_NORESERVE))) { charged = nrpages; if (security_vm_enough_memory(charged)) -- cgit v1.2.3 From 1db8508cf483dc1ecf66141f90a7c03659d69512 Mon Sep 17 00:00:00 2001 From: Stefan Richter Date: Tue, 10 Feb 2009 23:27:32 +0100 Subject: hugetlbfs: fix build failure with !CONFIG_HUGETLBFS Fix regression due to 5a6fe125950676015f5108fb71b2a67441755003, "Do not account for the address space used by hugetlbfs using VM_ACCOUNT" which added an argument to the function hugetlb_file_setup() but not to the macro hugetlb_file_setup(). Reported-by: Chris Clayton Signed-off-by: Stefan Richter Acked-by: Mel Gorman Signed-off-by: Linus Torvalds --- include/linux/hugetlb.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index af09660001c7..03be7f29ca01 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -159,9 +159,9 @@ static inline void set_file_hugepages(struct file *file) } #else /* !CONFIG_HUGETLBFS */ -#define is_file_hugepages(file) 0 -#define set_file_hugepages(file) BUG() -#define hugetlb_file_setup(name,size) ERR_PTR(-ENOSYS) +#define is_file_hugepages(file) 0 +#define set_file_hugepages(file) BUG() +#define hugetlb_file_setup(name,size,acctflag) ERR_PTR(-ENOSYS) #endif /* !CONFIG_HUGETLBFS */ -- cgit v1.2.3 From 3fccfd67df79c6351a156eb25a7a514e5f39c4d9 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 10 Feb 2009 16:37:31 +0100 Subject: timers: split process wide cpu clocks/timers, fix To decrease the chance of a missed enable, always enable the timer when we sample it, we'll always disable it when we find that there are no active timers in the jiffy tick. This fixes a flood of warnings reported by Mike Galbraith. Reported-by: Mike Galbraith Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 + kernel/posix-cpu-timers.c | 42 ++++++++++++++---------------------------- 2 files changed, 15 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 79392916d6c9..5d10fa0b6002 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2209,6 +2209,7 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) unsigned long flags; spin_lock_irqsave(&cputimer->lock, flags); + cputimer->running = 1; *times = cputimer->cputime; spin_unlock_irqrestore(&cputimer->lock, flags); } diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index db107c9bbc05..e5d7bfdfa7d4 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -488,7 +488,7 @@ void posix_cpu_timers_exit_group(struct task_struct *tsk) { struct task_cputime cputime; - thread_group_cputime(tsk, &cputime); + thread_group_cputimer(tsk, &cputime); cleanup_timers(tsk->signal->cpu_timers, cputime.utime, cputime.stime, cputime.sum_exec_runtime); } @@ -506,29 +506,6 @@ static void clear_dead_task(struct k_itimer *timer, union cpu_time_count now) now); } -/* - * Enable the process wide cpu timer accounting. - * - * serialized using ->sighand->siglock - */ -static void start_process_timers(struct task_struct *tsk) -{ - tsk->signal->cputimer.running = 1; - barrier(); -} - -/* - * Release the process wide timer accounting -- timer stops ticking when - * nobody cares about it. - * - * serialized using ->sighand->siglock - */ -static void stop_process_timers(struct task_struct *tsk) -{ - tsk->signal->cputimer.running = 0; - barrier(); -} - /* * Insert the timer on the appropriate list before any timers that * expire later. This must be called with the tasklist_lock held @@ -549,9 +526,6 @@ static void arm_timer(struct k_itimer *timer, union cpu_time_count now) BUG_ON(!irqs_disabled()); spin_lock(&p->sighand->siglock); - if (!CPUCLOCK_PERTHREAD(timer->it_clock)) - start_process_timers(p); - listpos = head; if (CPUCLOCK_WHICH(timer->it_clock) == CPUCLOCK_SCHED) { list_for_each_entry(next, head, entry) { @@ -1021,6 +995,19 @@ static void check_thread_timers(struct task_struct *tsk, } } +static void stop_process_timers(struct task_struct *tsk) +{ + struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; + unsigned long flags; + + if (!cputimer->running) + return; + + spin_lock_irqsave(&cputimer->lock, flags); + cputimer->running = 0; + spin_unlock_irqrestore(&cputimer->lock, flags); +} + /* * Check for any per-thread CPU timers that have fired and move them * off the tsk->*_timers list onto the firing list. Per-thread timers @@ -1427,7 +1414,6 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, struct list_head *head; BUG_ON(clock_idx == CPUCLOCK_SCHED); - start_process_timers(tsk); cpu_timer_sample_group(clock_idx, tsk, &now); if (oldval) { -- cgit v1.2.3 From 4da94d49b2ecb0a26e716a8811c3ecc542c2a65d Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 11 Feb 2009 11:30:27 +0100 Subject: timers: fix TIMER_ABSTIME for process wide cpu timers The POSIX timer interface allows for absolute time expiry values through the TIMER_ABSTIME flag, therefore we have to synchronize the timer to the clock every time we start it. Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- include/linux/sched.h | 13 +------------ kernel/posix-cpu-timers.c | 34 ++++++++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 5d10fa0b6002..8981e52c714f 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2201,18 +2201,7 @@ static inline int spin_needbreak(spinlock_t *lock) * Thread group CPU time accounting. */ void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times); - -static inline -void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) -{ - struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; - unsigned long flags; - - spin_lock_irqsave(&cputimer->lock, flags); - cputimer->running = 1; - *times = cputimer->cputime; - spin_unlock_irqrestore(&cputimer->lock, flags); -} +void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times); static inline void thread_group_cputime_init(struct signal_struct *sig) { diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index e5d7bfdfa7d4..2313a4cc14ea 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -261,6 +261,40 @@ out: rcu_read_unlock(); } +static void update_gt_cputime(struct task_cputime *a, struct task_cputime *b) +{ + if (cputime_gt(b->utime, a->utime)) + a->utime = b->utime; + + if (cputime_gt(b->stime, a->stime)) + a->stime = b->stime; + + if (b->sum_exec_runtime > a->sum_exec_runtime) + a->sum_exec_runtime = b->sum_exec_runtime; +} + +void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) +{ + struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; + struct task_cputime sum; + unsigned long flags; + + spin_lock_irqsave(&cputimer->lock, flags); + if (!cputimer->running) { + cputimer->running = 1; + /* + * The POSIX timer interface allows for absolute time expiry + * values through the TIMER_ABSTIME flag, therefore we have + * to synchronize the timer to the clock every time we start + * it. + */ + thread_group_cputime(tsk, &sum); + update_gt_cputime(&cputimer->cputime, &sum); + } + *times = cputimer->cputime; + spin_unlock_irqrestore(&cputimer->lock, flags); +} + /* * Sample a process (thread group) clock for the given group_leader task. * Must be called with tasklist_lock held for reading. -- cgit v1.2.3 From cfebe563bd0a3ff97e1bc167123120d59c7a84db Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 11 Feb 2009 13:04:36 -0800 Subject: cgroups: fix lockdep subclasses overflow I enabled all cgroup subsystems when compiling kernel, and then: # mount -t cgroup -o net_cls xxx /mnt # mkdir /mnt/0 This showed up immediately: BUG: MAX_LOCKDEP_SUBCLASSES too low! turning off the locking correctness validator. It's caused by the cgroup hierarchy lock: for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; if (ss->root == root) mutex_lock_nested(&ss->hierarchy_mutex, i); } Now we have 9 cgroup subsystems, and the above 'i' for net_cls is 8, but MAX_LOCKDEP_SUBCLASSES is 8. This patch uses different lockdep keys for different subsystems. Signed-off-by: Li Zefan Acked-by: Paul Menage Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cgroup.h | 1 + kernel/cgroup.c | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index e4e8e117d27d..499900d0cee7 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -378,6 +378,7 @@ struct cgroup_subsys { * - initiating hotplug events */ struct mutex hierarchy_mutex; + struct lock_class_key subsys_key; /* * Link to parent, and list entry in parent's children. diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 5a54ff42874e..e14db9c089b9 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2351,7 +2351,7 @@ static void cgroup_lock_hierarchy(struct cgroupfs_root *root) for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; if (ss->root == root) - mutex_lock_nested(&ss->hierarchy_mutex, i); + mutex_lock(&ss->hierarchy_mutex); } } @@ -2637,6 +2637,7 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss) BUG_ON(!list_empty(&init_task.tasks)); mutex_init(&ss->hierarchy_mutex); + lockdep_set_class(&ss->hierarchy_mutex, &ss->subsys_key); ss->active = 1; } -- cgit v1.2.3 From 6c5979631b4b03c9288776562c18036765e398c1 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 11 Feb 2009 13:04:38 -0800 Subject: syscall define: fix uml compile bug With the new system call defines we get this on uml: arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x308): undefined reference to `sys_sigprocmask' Reason for this is that uml passes the preprocessor option -Dsigprocmask=kernel_sigprocmask to gcc when compiling the kernel. This causes SYSCALL_DEFINE3(sigprocmask, ...) to be expanded to SYSCALL_DEFINEx(3, kernel_sigprocmask, ...) and finally to a system call named sys_kernel_sigprocmask. However sys_sigprocmask is missing because of this. To avoid macro expansion for the system call name just concatenate the name at first define instead of carrying it through severel levels. This was pointed out by Al Viro. Signed-off-by: Heiko Carstens Cc: Geert Uytterhoeven Cc: Al Viro Reviewed-by: WANG Cong Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/syscalls.h | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 0eda02ff2414..f9f900cfd066 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -95,13 +95,13 @@ struct old_linux_dirent; #define __SC_TEST5(t5, a5, ...) __SC_TEST(t5); __SC_TEST4(__VA_ARGS__) #define __SC_TEST6(t6, a6, ...) __SC_TEST(t6); __SC_TEST5(__VA_ARGS__) -#define SYSCALL_DEFINE0(name) asmlinkage long sys_##name(void) -#define SYSCALL_DEFINE1(...) SYSCALL_DEFINEx(1, __VA_ARGS__) -#define SYSCALL_DEFINE2(...) SYSCALL_DEFINEx(2, __VA_ARGS__) -#define SYSCALL_DEFINE3(...) SYSCALL_DEFINEx(3, __VA_ARGS__) -#define SYSCALL_DEFINE4(...) SYSCALL_DEFINEx(4, __VA_ARGS__) -#define SYSCALL_DEFINE5(...) SYSCALL_DEFINEx(5, __VA_ARGS__) -#define SYSCALL_DEFINE6(...) SYSCALL_DEFINEx(6, __VA_ARGS__) +#define SYSCALL_DEFINE0(name) asmlinkage long sys_##name(void) +#define SYSCALL_DEFINE1(name, ...) SYSCALL_DEFINEx(1, _##name, __VA_ARGS__) +#define SYSCALL_DEFINE2(name, ...) SYSCALL_DEFINEx(2, _##name, __VA_ARGS__) +#define SYSCALL_DEFINE3(name, ...) SYSCALL_DEFINEx(3, _##name, __VA_ARGS__) +#define SYSCALL_DEFINE4(name, ...) SYSCALL_DEFINEx(4, _##name, __VA_ARGS__) +#define SYSCALL_DEFINE5(name, ...) SYSCALL_DEFINEx(5, _##name, __VA_ARGS__) +#define SYSCALL_DEFINE6(name, ...) SYSCALL_DEFINEx(6, _##name, __VA_ARGS__) #ifdef CONFIG_PPC64 #define SYSCALL_ALIAS(alias, name) \ @@ -121,21 +121,21 @@ struct old_linux_dirent; #define SYSCALL_DEFINE(name) static inline long SYSC_##name #define SYSCALL_DEFINEx(x, name, ...) \ - asmlinkage long sys_##name(__SC_DECL##x(__VA_ARGS__)); \ - static inline long SYSC_##name(__SC_DECL##x(__VA_ARGS__)); \ - asmlinkage long SyS_##name(__SC_LONG##x(__VA_ARGS__)) \ + asmlinkage long sys##name(__SC_DECL##x(__VA_ARGS__)); \ + static inline long SYSC##name(__SC_DECL##x(__VA_ARGS__)); \ + asmlinkage long SyS##name(__SC_LONG##x(__VA_ARGS__)) \ { \ __SC_TEST##x(__VA_ARGS__); \ - return (long) SYSC_##name(__SC_CAST##x(__VA_ARGS__)); \ + return (long) SYSC##name(__SC_CAST##x(__VA_ARGS__)); \ } \ - SYSCALL_ALIAS(sys_##name, SyS_##name); \ - static inline long SYSC_##name(__SC_DECL##x(__VA_ARGS__)) + SYSCALL_ALIAS(sys##name, SyS##name); \ + static inline long SYSC##name(__SC_DECL##x(__VA_ARGS__)) #else /* CONFIG_HAVE_SYSCALL_WRAPPERS */ #define SYSCALL_DEFINE(name) asmlinkage long sys_##name #define SYSCALL_DEFINEx(x, name, ...) \ - asmlinkage long sys_##name(__SC_DECL##x(__VA_ARGS__)) + asmlinkage long sys##name(__SC_DECL##x(__VA_ARGS__)) #endif /* CONFIG_HAVE_SYSCALL_WRAPPERS */ -- cgit v1.2.3