diff options
114 files changed, 1941 insertions, 558 deletions
diff --git a/Documentation/ioctl/ioctl-number.txt b/Documentation/ioctl/ioctl-number.txt index 7bb0d934b6d..dbea4f95fc8 100644 --- a/Documentation/ioctl/ioctl-number.txt +++ b/Documentation/ioctl/ioctl-number.txt @@ -139,6 +139,7 @@ Code Seq# Include File Comments 'm' all linux/synclink.h conflict! 'm' 00-1F net/irda/irmod.h conflict! 'n' 00-7F linux/ncp_fs.h +'n' 80-8F linux/nilfs2_fs.h NILFS2 'n' E0-FF video/matrox.h matroxfb 'o' 00-1F fs/ocfs2/ocfs2_fs.h OCFS2 'o' 00-03 include/mtd/ubi-user.h conflict! (OCFS2 and UBI overlaps) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index dd1a6d4bb74..7936b801fe6 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1115,6 +1115,10 @@ and is between 256 and 4096 characters. It is defined in the file libata.dma=4 Compact Flash DMA only Combinations also work, so libata.dma=3 enables DMA for disks and CDROMs, but not CFs. + + libata.ignore_hpa= [LIBATA] Ignore HPA limit + libata.ignore_hpa=0 keep BIOS limits (default) + libata.ignore_hpa=1 ignore limits, using full disk libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume when set. diff --git a/arch/ia64/Makefile b/arch/ia64/Makefile index 58a7e46affd..e7cbaa02cd0 100644 --- a/arch/ia64/Makefile +++ b/arch/ia64/Makefile @@ -41,11 +41,6 @@ $(error Sorry, you need a newer version of the assember, one that is built from ftp://ftp.hpl.hp.com/pub/linux-ia64/gas-030124.tar.gz) endif -ifeq ($(call cc-version),0304) - cflags-$(CONFIG_ITANIUM) += -mtune=merced - cflags-$(CONFIG_MCKINLEY) += -mtune=mckinley -endif - KBUILD_CFLAGS += $(cflags-y) head-y := arch/ia64/kernel/head.o arch/ia64/kernel/init_task.o diff --git a/arch/ia64/include/asm/bitops.h b/arch/ia64/include/asm/bitops.h index e2ca8003733..57a2787bc9f 100644 --- a/arch/ia64/include/asm/bitops.h +++ b/arch/ia64/include/asm/bitops.h @@ -286,7 +286,7 @@ __test_and_clear_bit(int nr, volatile void * addr) { __u32 *p = (__u32 *) addr + (nr >> 5); __u32 m = 1 << (nr & 31); - int oldbitset = *p & m; + int oldbitset = (*p & m) != 0; *p &= ~m; return oldbitset; diff --git a/arch/ia64/include/asm/pgtable.h b/arch/ia64/include/asm/pgtable.h index 0a9cc73d35c..8840a690d1e 100644 --- a/arch/ia64/include/asm/pgtable.h +++ b/arch/ia64/include/asm/pgtable.h @@ -155,7 +155,6 @@ #include <linux/bitops.h> #include <asm/cacheflush.h> #include <asm/mmu_context.h> -#include <asm/processor.h> /* * Next come the mappings that determine how mmap() protection bits diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c index 2d311864e35..8ebccb589e1 100644 --- a/arch/ia64/kernel/ia64_ksyms.c +++ b/arch/ia64/kernel/ia64_ksyms.c @@ -21,6 +21,7 @@ EXPORT_SYMBOL(csum_ipv6_magic); #include <asm/page.h> EXPORT_SYMBOL(clear_page); +EXPORT_SYMBOL(copy_page); #ifdef CONFIG_VIRTUAL_MEM_MAP #include <linux/bootmem.h> @@ -60,9 +61,6 @@ EXPORT_SYMBOL(__udivdi3); EXPORT_SYMBOL(__moddi3); EXPORT_SYMBOL(__umoddi3); -#include <asm/page.h> -EXPORT_SYMBOL(copy_page); - #if defined(CONFIG_MD_RAID456) || defined(CONFIG_MD_RAID456_MODULE) extern void xor_ia64_2(void); extern void xor_ia64_3(void); diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index c48b03f2b61..dab4d393908 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -1072,6 +1072,10 @@ iosapic_init (unsigned long phys_addr, unsigned int gsi_base) } addr = ioremap(phys_addr, 0); + if (addr == NULL) { + spin_unlock_irqrestore(&iosapic_lock, flags); + return -ENOMEM; + } ver = iosapic_version(addr); if ((err = iosapic_check_gsi_range(gsi_base, ver))) { iounmap(addr); diff --git a/arch/ia64/kernel/pci-dma.c b/arch/ia64/kernel/pci-dma.c index 05695962fe4..f6b1ff0aea7 100644 --- a/arch/ia64/kernel/pci-dma.c +++ b/arch/ia64/kernel/pci-dma.c @@ -69,11 +69,6 @@ iommu_dma_init(void) int iommu_dma_supported(struct device *dev, u64 mask) { - struct dma_map_ops *ops = platform_dma_get_ops(dev); - - if (ops->dma_supported) - return ops->dma_supported(dev, mask); - /* Copied from i386. Doesn't make much sense, because it will only work for pci_alloc_coherent. The caller just has to use GFP_DMA in this case. */ diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index bc80dff1df7..8f060352e12 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -372,6 +372,10 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev) retval = kobject_init_and_add(&all_cpu_cache_info[cpu].kobj, &cache_ktype_percpu_entry, &sys_dev->kobj, "%s", "cache"); + if (unlikely(retval < 0)) { + cpu_cache_sysfs_exit(cpu); + return retval; + } for (i = 0; i < all_cpu_cache_info[cpu].num_cache_leaves; i++) { this_object = LEAF_KOBJECT_PTR(cpu,i); @@ -385,7 +389,7 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev) } kobject_put(&all_cpu_cache_info[cpu].kobj); cpu_cache_sysfs_exit(cpu); - break; + return retval; } kobject_uevent(&(this_object->kobj), KOBJ_ADD); } diff --git a/arch/mn10300/include/asm/pci.h b/arch/mn10300/include/asm/pci.h index 35d2ed6396f..19aecc90f7a 100644 --- a/arch/mn10300/include/asm/pci.h +++ b/arch/mn10300/include/asm/pci.h @@ -59,7 +59,6 @@ void pcibios_penalize_isa_irq(int irq); #include <linux/slab.h> #include <asm/scatterlist.h> #include <linux/string.h> -#include <linux/mm.h> #include <asm/io.h> struct pci_dev; diff --git a/arch/powerpc/kernel/dma.c b/arch/powerpc/kernel/dma.c index 20a60d661ba..ccf129d47d8 100644 --- a/arch/powerpc/kernel/dma.c +++ b/arch/powerpc/kernel/dma.c @@ -7,6 +7,7 @@ #include <linux/device.h> #include <linux/dma-mapping.h> +#include <linux/lmb.h> #include <asm/bug.h> #include <asm/abs_addr.h> @@ -90,11 +91,10 @@ static void dma_direct_unmap_sg(struct device *dev, struct scatterlist *sg, static int dma_direct_dma_supported(struct device *dev, u64 mask) { #ifdef CONFIG_PPC64 - /* Could be improved to check for memory though it better be - * done via some global so platforms can set the limit in case + /* Could be improved so platforms can set the limit in case * they have limited DMA windows */ - return mask >= DMA_BIT_MASK(32); + return mask >= (lmb_end_of_DRAM() - 1); #else return 1; #endif diff --git a/arch/powerpc/kernel/perf_counter.c b/arch/powerpc/kernel/perf_counter.c index 809fdf94b95..70e1f57f7dd 100644 --- a/arch/powerpc/kernel/perf_counter.c +++ b/arch/powerpc/kernel/perf_counter.c @@ -518,6 +518,8 @@ void hw_perf_disable(void) struct cpu_hw_counters *cpuhw; unsigned long flags; + if (!ppmu) + return; local_irq_save(flags); cpuhw = &__get_cpu_var(cpu_hw_counters); @@ -572,6 +574,8 @@ void hw_perf_enable(void) int n_lim; int idx; + if (!ppmu) + return; local_irq_save(flags); cpuhw = &__get_cpu_var(cpu_hw_counters); if (!cpuhw->disabled) { @@ -737,6 +741,8 @@ int hw_perf_group_sched_in(struct perf_counter *group_leader, long i, n, n0; struct perf_counter *sub; + if (!ppmu) + return 0; cpuhw = &__get_cpu_var(cpu_hw_counters); n0 = cpuhw->n_counters; n = collect_events(group_leader, ppmu->n_counter - n0, @@ -1281,6 +1287,8 @@ void hw_perf_counter_setup(int cpu) { struct cpu_hw_counters *cpuhw = &per_cpu(cpu_hw_counters, cpu); + if (!ppmu) + return; memset(cpuhw, 0, sizeof(*cpuhw)); cpuhw->mmcr[0] = MMCR0_FC; } diff --git a/arch/sh/boards/board-ap325rxa.c b/arch/sh/boards/board-ap325rxa.c index 7ffd1b4315b..b9c88cc519e 100644 --- a/arch/sh/boards/board-ap325rxa.c +++ b/arch/sh/boards/board-ap325rxa.c @@ -547,7 +547,7 @@ static int __init ap325rxa_devices_setup(void) return platform_add_devices(ap325rxa_devices, ARRAY_SIZE(ap325rxa_devices)); } -device_initcall(ap325rxa_devices_setup); +arch_initcall(ap325rxa_devices_setup); /* Return the board specific boot mode pin configuration */ static int ap325rxa_mode_pins(void) diff --git a/arch/sh/boards/mach-migor/setup.c b/arch/sh/boards/mach-migor/setup.c index f70f4644deb..f9b2e4df35b 100644 --- a/arch/sh/boards/mach-migor/setup.c +++ b/arch/sh/boards/mach-migor/setup.c @@ -608,7 +608,7 @@ static int __init migor_devices_setup(void) return platform_add_devices(migor_devices, ARRAY_SIZE(migor_devices)); } -__initcall(migor_devices_setup); +arch_initcall(migor_devices_setup); /* Return the board specific boot mode pin configuration */ static int migor_mode_pins(void) diff --git a/arch/sh/kernel/cpu/sh2/setup-sh7619.c b/arch/sh/kernel/cpu/sh2/setup-sh7619.c index 13798733f2d..8555c05e866 100644 --- a/arch/sh/kernel/cpu/sh2/setup-sh7619.c +++ b/arch/sh/kernel/cpu/sh2/setup-sh7619.c @@ -187,7 +187,7 @@ static int __init sh7619_devices_setup(void) return platform_add_devices(sh7619_devices, ARRAY_SIZE(sh7619_devices)); } -__initcall(sh7619_devices_setup); +arch_initcall(sh7619_devices_setup); void __init plat_irq_setup(void) { diff --git a/arch/sh/kernel/cpu/sh2a/setup-mxg.c b/arch/sh/kernel/cpu/sh2a/setup-mxg.c index 869c2da4820..b6737644531 100644 --- a/arch/sh/kernel/cpu/sh2a/setup-mxg.c +++ b/arch/sh/kernel/cpu/sh2a/setup-mxg.c @@ -238,7 +238,7 @@ static int __init mxg_devices_setup(void) return platform_add_devices(mxg_devices, ARRAY_SIZE(mxg_devices)); } -__initcall(mxg_devices_setup); +arch_initcall(mxg_devices_setup); void __init plat_irq_setup(void) { diff --git a/arch/sh/kernel/cpu/sh2a/setup-sh7201.c b/arch/sh/kernel/cpu/sh2a/setup-sh7201.c index d8febe12806..fbde5b75deb 100644 --- a/arch/sh/kernel/cpu/sh2a/setup-sh7201.c +++ b/arch/sh/kernel/cpu/sh2a/setup-sh7201.c @@ -357,7 +357,7 @@ static int __init sh7201_devices_setup(void) return platform_add_devices(sh7201_devices, ARRAY_SIZE(sh7201_devices)); } -__initcall(sh7201_devices_setup); +arch_initcall(sh7201_devices_setup); void __init plat_irq_setup(void) { diff --git a/arch/sh/kernel/cpu/sh2a/setup-sh7203.c b/arch/sh/kernel/cpu/sh2a/setup-sh7203.c index 62e3039d239..d3fd536c9a8 100644 --- a/arch/sh/kernel/cpu/sh2a/setup-sh7203.c +++ b/arch/sh/kernel/cpu/sh2a/setup-sh7203.c @@ -367,7 +367,7 @@ static int __init sh7203_devices_setup(void) return platform_add_devices(sh7203_devices, ARRAY_SIZE(sh7203_devices)); } -__initcall(sh7203_devices_setup); +arch_initcall(sh7203_devices_setup); void __init plat_irq_setup(void) { diff --git a/arch/sh/kernel/cpu/sh2a/setup-sh7206.c b/arch/sh/kernel/cpu/sh2a/setup-sh7206.c index 3e6f3d7a58b..a9ccc5e8d9e 100644 --- a/arch/sh/kernel/cpu/sh2a/setup-sh7206.c +++ b/arch/sh/kernel/cpu/sh2a/setup-sh7206.c @@ -338,7 +338,7 @@ static int __init sh7206_devices_setup(void) return platform_add_devices(sh7206_devices, ARRAY_SIZE(sh7206_devices)); } -__initcall(sh7206_devices_setup); +arch_initcall(sh7206_devices_setup); void __init plat_irq_setup(void) { diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7705.c b/arch/sh/kernel/cpu/sh3/setup-sh7705.c index 88f742fed9e..c2310598387 100644 --- a/arch/sh/kernel/cpu/sh3/setup-sh7705.c +++ b/arch/sh/kernel/cpu/sh3/setup-sh7705.c @@ -222,7 +222,7 @@ static int __init sh7705_devices_setup(void) return platform_add_devices(sh7705_devices, ARRAY_SIZE(sh7705_devices)); } -__initcall(sh7705_devices_setup); +arch_initcall(sh7705_devices_setup); static struct platform_device *sh7705_early_devices[] __initdata = { &tmu0_device, diff --git a/arch/sh/kernel/cpu/sh3/setup-sh770x.c b/arch/sh/kernel/cpu/sh3/setup-sh770x.c index c5630679858..347ab35d069 100644 --- a/arch/sh/kernel/cpu/sh3/setup-sh770x.c +++ b/arch/sh/kernel/cpu/sh3/setup-sh770x.c @@ -250,7 +250,7 @@ static int __init sh770x_devices_setup(void) return platform_add_devices(sh770x_devices, ARRAY_SIZE(sh770x_devices)); } -__initcall(sh770x_devices_setup); +arch_initcall(sh770x_devices_setup); static struct platform_device *sh770x_early_devices[] __initdata = { &tmu0_device, diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7710.c b/arch/sh/kernel/cpu/sh3/setup-sh7710.c index efa76c8148f..717e90ae109 100644 --- a/arch/sh/kernel/cpu/sh3/setup-sh7710.c +++ b/arch/sh/kernel/cpu/sh3/setup-sh7710.c @@ -226,7 +226,7 @@ static int __init sh7710_devices_setup(void) return platform_add_devices(sh7710_devices, ARRAY_SIZE(sh7710_devices)); } -__initcall(sh7710_devices_setup); +arch_initcall(sh7710_devices_setup); static struct platform_device *sh7710_early_devices[] __initdata = { &tmu0_device, diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7720.c b/arch/sh/kernel/cpu/sh3/setup-sh7720.c index 5b2107798ed..74d8baaf8e9 100644 --- a/arch/sh/kernel/cpu/sh3/setup-sh7720.c +++ b/arch/sh/kernel/cpu/sh3/setup-sh7720.c @@ -388,7 +388,7 @@ static int __init sh7720_devices_setup(void) return platform_add_devices(sh7720_devices, ARRAY_SIZE(sh7720_devices)); } -__initcall(sh7720_devices_setup); +arch_initcall(sh7720_devices_setup); static struct platform_device *sh7720_early_devices[] __initdata = { &cmt0_device, diff --git a/arch/sh/kernel/cpu/sh4/setup-sh4-202.c b/arch/sh/kernel/cpu/sh4/setup-sh4-202.c index 6d088d12359..de4827df19a 100644 --- a/arch/sh/kernel/cpu/sh4/setup-sh4-202.c +++ b/arch/sh/kernel/cpu/sh4/setup-sh4-202.c @@ -138,7 +138,7 @@ static int __init sh4202_devices_setup(void) return platform_add_devices(sh4202_devices, ARRAY_SIZE(sh4202_devices)); } -__initcall(sh4202_devices_setup); +arch_initcall(sh4202_devices_setup); static struct platform_device *sh4202_early_devices[] __initdata = { &tmu0_device, diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7750.c b/arch/sh/kernel/cpu/sh4/setup-sh7750.c index 851672d15cf..1b8b122e8f3 100644 --- a/arch/sh/kernel/cpu/sh4/setup-sh7750.c +++ b/arch/sh/kernel/cpu/sh4/setup-sh7750.c @@ -239,7 +239,7 @@ static int __init sh7750_devices_setup(void) return platform_add_devices(sh7750_devices, ARRAY_SIZE(sh7750_devices)); } -__initcall(sh7750_devices_setup); +arch_initcall(sh7750_devices_setup); static struct platform_device *sh7750_early_devices[] __initdata = { &tmu0_device, diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7760.c b/arch/sh/kernel/cpu/sh4/setup-sh7760.c index 5b822519bd9..7fbb7be9284 100644 --- a/arch/sh/kernel/cpu/sh4/setup-sh7760.c +++ b/arch/sh/kernel/cpu/sh4/setup-sh7760.c @@ -265,7 +265,7 @@ static int __init sh7760_devices_setup(void) return platform_add_devices(sh7760_devices, ARRAY_SIZE(sh7760_devices)); } -__initcall(sh7760_devices_setup); +arch_initcall(sh7760_devices_setup); static struct platform_device *sh7760_early_devices[] __initdata = { &tmu0_device, diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7343.c b/arch/sh/kernel/cpu/sh4a/setup-sh7343.c index 6307e087c86..ac4d5672ec1 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7343.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7343.c @@ -325,7 +325,7 @@ static int __init sh7343_devices_setup(void) return platform_add_devices(sh7343_devices, ARRAY_SIZE(sh7343_devices)); } -__initcall(sh7343_devices_setup); +arch_initcall(sh7343_devices_setup); static struct platform_device *sh7343_early_devices[] __initdata = { &cmt_device, diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7366.c b/arch/sh/kernel/cpu/sh4a/setup-sh7366.c index c18f7d09281..1a956b1becc 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7366.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7366.c @@ -318,7 +318,7 @@ static int __init sh7366_devices_setup(void) return platform_add_devices(sh7366_devices, ARRAY_SIZE(sh7366_devices)); } -__initcall(sh7366_devices_setup); +arch_initcall(sh7366_devices_setup); static struct platform_device *sh7366_early_devices[] __initdata = { &cmt_device, diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7722.c b/arch/sh/kernel/cpu/sh4a/setup-sh7722.c index ea524a2da3e..cda76ebf87c 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7722.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7722.c @@ -359,7 +359,7 @@ static int __init sh7722_devices_setup(void) return platform_add_devices(sh7722_devices, ARRAY_SIZE(sh7722_devices)); } -__initcall(sh7722_devices_setup); +arch_initcall(sh7722_devices_setup); static struct platform_device *sh7722_early_devices[] __initdata = { &cmt_device, diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7723.c b/arch/sh/kernel/cpu/sh4a/setup-sh7723.c index e1bb80b2a27..b45dace9539 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7723.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7723.c @@ -473,7 +473,7 @@ static int __init sh7723_devices_setup(void) return platform_add_devices(sh7723_devices, ARRAY_SIZE(sh7723_devices)); } -__initcall(sh7723_devices_setup); +arch_initcall(sh7723_devices_setup); static struct platform_device *sh7723_early_devices[] __initdata = { &cmt_device, diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7724.c b/arch/sh/kernel/cpu/sh4a/setup-sh7724.c index e5ac9eb11c6..a04edaab9a2 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7724.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7724.c @@ -508,7 +508,7 @@ static int __init sh7724_devices_setup(void) return platform_add_devices(sh7724_devices, ARRAY_SIZE(sh7724_devices)); } -device_initcall(sh7724_devices_setup); +arch_initcall(sh7724_devices_setup); static struct platform_device *sh7724_early_devices[] __initdata = { &cmt_device, diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7763.c b/arch/sh/kernel/cpu/sh4a/setup-sh7763.c index f1e0c0d36da..4659fff6b84 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7763.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7763.c @@ -314,7 +314,7 @@ static int __init sh7763_devices_setup(void) return platform_add_devices(sh7763_devices, ARRAY_SIZE(sh7763_devices)); } -__initcall(sh7763_devices_setup); +arch_initcall(sh7763_devices_setup); static struct platform_device *sh7763_early_devices[] __initdata = { &tmu0_device, diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7770.c b/arch/sh/kernel/cpu/sh4a/setup-sh7770.c index 1e86209db28..eead08d89d3 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7770.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7770.c @@ -368,7 +368,7 @@ static int __init sh7770_devices_setup(void) return platform_add_devices(sh7770_devices, ARRAY_SIZE(sh7770_devices)); } -__initcall(sh7770_devices_setup); +arch_initcall(sh7770_devices_setup); static struct platform_device *sh7770_early_devices[] __initdata = { &tmu0_device, diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7780.c b/arch/sh/kernel/cpu/sh4a/setup-sh7780.c index 715e05b431e..2c901f44695 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7780.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7780.c @@ -256,7 +256,7 @@ static int __init sh7780_devices_setup(void) return platform_add_devices(sh7780_devices, ARRAY_SIZE(sh7780_devices)); } -__initcall(sh7780_devices_setup); +arch_initcall(sh7780_devices_setup); static struct platform_device *sh7780_early_devices[] __initdata = { &tmu0_device, diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7785.c b/arch/sh/kernel/cpu/sh4a/setup-sh7785.c index af561402570..7f6c718b6c3 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7785.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7785.c @@ -263,7 +263,7 @@ static int __init sh7785_devices_setup(void) return platform_add_devices(sh7785_devices, ARRAY_SIZE(sh7785_devices)); } -__initcall(sh7785_devices_setup); +arch_initcall(sh7785_devices_setup); static struct platform_device *sh7785_early_devices[] __initdata = { &tmu0_device, diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7786.c b/arch/sh/kernel/cpu/sh4a/setup-sh7786.c index b70049470a0..0104a8ec536 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-sh7786.c +++ b/arch/sh/kernel/cpu/sh4a/setup-sh7786.c @@ -547,7 +547,7 @@ static int __init sh7786_devices_setup(void) return platform_add_devices(sh7786_devices, ARRAY_SIZE(sh7786_devices)); } -device_initcall(sh7786_devices_setup); +arch_initcall(sh7786_devices_setup); void __init plat_early_device_setup(void) { diff --git a/arch/sh/kernel/cpu/sh4a/setup-shx3.c b/arch/sh/kernel/cpu/sh4a/setup-shx3.c index 53c65fd9cce..07f078961c7 100644 --- a/arch/sh/kernel/cpu/sh4a/setup-shx3.c +++ b/arch/sh/kernel/cpu/sh4a/setup-shx3.c @@ -256,7 +256,7 @@ static int __init shx3_devices_setup(void) return platform_add_devices(shx3_devices, ARRAY_SIZE(shx3_devices)); } -__initcall(shx3_devices_setup); +arch_initcall(shx3_devices_setup); void __init plat_early_device_setup(void) { diff --git a/arch/sh/kernel/cpu/sh5/setup-sh5.c b/arch/sh/kernel/cpu/sh5/setup-sh5.c index f5ff1ac57fc..6a0f82f7003 100644 --- a/arch/sh/kernel/cpu/sh5/setup-sh5.c +++ b/arch/sh/kernel/cpu/sh5/setup-sh5.c @@ -186,7 +186,7 @@ static int __init sh5_devices_setup(void) return platform_add_devices(sh5_devices, ARRAY_SIZE(sh5_devices)); } -__initcall(sh5_devices_setup); +arch_initcall(sh5_devices_setup); void __init plat_early_device_setup(void) { diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index e2485b03f1c..63fddcd082c 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -400,6 +400,13 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) level = cpuid_eax(1); if((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58) set_cpu_cap(c, X86_FEATURE_REP_GOOD); + + /* + * Some BIOSes incorrectly force this feature, but only K8 + * revision D (model = 0x14) and later actually support it. + */ + if (c->x86_model < 0x14) + clear_cpu_cap(c, X86_FEATURE_LAHF_LM); } if (c->x86 == 0x10 || c->x86 == 0x11) set_cpu_cap(c, X86_FEATURE_REP_GOOD); diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index f1961c07af9..5ce60a88027 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -59,7 +59,30 @@ void __init setup_cpu_local_masks(void) alloc_bootmem_cpumask_var(&cpu_sibling_setup_mask); } -static const struct cpu_dev *this_cpu __cpuinitdata; +static void __cpuinit default_init(struct cpuinfo_x86 *c) +{ +#ifdef CONFIG_X86_64 + display_cacheinfo(c); +#else + /* Not much we can do here... */ + /* Check if at least it has cpuid */ + if (c->cpuid_level == -1) { + /* No cpuid. It must be an ancient CPU */ + if (c->x86 == 4) + strcpy(c->x86_model_id, "486"); + else if (c->x86 == 3) + strcpy(c->x86_model_id, "386"); + } +#endif +} + +static const struct cpu_dev __cpuinitconst default_cpu = { + .c_init = default_init, + .c_vendor = "Unknown", + .c_x86_vendor = X86_VENDOR_UNKNOWN, +}; + +static const struct cpu_dev *this_cpu __cpuinitdata = &default_cpu; DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { #ifdef CONFIG_X86_64 @@ -332,29 +355,6 @@ void switch_to_new_gdt(int cpu) static const struct cpu_dev *__cpuinitdata cpu_devs[X86_VENDOR_NUM] = {}; -static void __cpuinit default_init(struct cpuinfo_x86 *c) -{ -#ifdef CONFIG_X86_64 - display_cacheinfo(c); -#else - /* Not much we can do here... */ - /* Check if at least it has cpuid */ - if (c->cpuid_level == -1) { - /* No cpuid. It must be an ancient CPU */ - if (c->x86 == 4) - strcpy(c->x86_model_id, "486"); - else if (c->x86 == 3) - strcpy(c->x86_model_id, "386"); - } -#endif -} - -static const struct cpu_dev __cpuinitconst default_cpu = { - .c_init = default_init, - .c_vendor = "Unknown", - .c_x86_vendor = X86_VENDOR_UNKNOWN, -}; - static void __cpuinit get_model_name(struct cpuinfo_x86 *c) { unsigned int *v; diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index bff8dd191dd..8bc64cfbe93 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -36,6 +36,7 @@ static DEFINE_PER_CPU(__u64, next_check) = INITIAL_JIFFIES; static DEFINE_PER_CPU(unsigned long, thermal_throttle_count); +static DEFINE_PER_CPU(bool, thermal_throttle_active); static atomic_t therm_throt_en = ATOMIC_INIT(0); @@ -96,24 +97,27 @@ static int therm_throt_process(int curr) { unsigned int cpu = smp_processor_id(); __u64 tmp_jiffs = get_jiffies_64(); + bool was_throttled = __get_cpu_var(thermal_throttle_active); + bool is_throttled = __get_cpu_var(thermal_throttle_active) = curr; - if (curr) + if (is_throttled) __get_cpu_var(thermal_throttle_count)++; - if (time_before64(tmp_jiffs, __get_cpu_var(next_check))) + if (!(was_throttled ^ is_throttled) && + time_before64(tmp_jiffs, __get_cpu_var(next_check))) return 0; __get_cpu_var(next_check) = tmp_jiffs + CHECK_INTERVAL; /* if we just entered the thermal event */ - if (curr) { + if (is_throttled) { printk(KERN_CRIT "CPU%d: Temperature above threshold, " - "cpu clock throttled (total events = %lu)\n", cpu, - __get_cpu_var(thermal_throttle_count)); + "cpu clock throttled (total events = %lu)\n", + cpu, __get_cpu_var(thermal_throttle_count)); add_taint(TAINT_MACHINE_CHECK); - } else { - printk(KERN_CRIT "CPU%d: Temperature/speed normal\n", cpu); + } else if (was_throttled) { + printk(KERN_INFO "CPU%d: Temperature/speed normal\n", cpu); } return 1; diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 9eb89760370..a06e8d10184 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -418,20 +418,20 @@ static int __init set_pci_reboot(const struct dmi_system_id *d) } static struct dmi_system_id __initdata pci_reboot_dmi_table[] = { - { /* Handle problems with rebooting on Apple MacBook5,2 */ + { /* Handle problems with rebooting on Apple MacBook5 */ .callback = set_pci_reboot, - .ident = "Apple MacBook", + .ident = "Apple MacBook5", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), - DMI_MATCH(DMI_PRODUCT_NAME, "MacBook5,2"), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBook5"), }, }, - { /* Handle problems with rebooting on Apple MacBookPro5,1 */ + { /* Handle problems with rebooting on Apple MacBookPro5 */ .callback = set_pci_reboot, - .ident = "Apple MacBookPro5,1", + .ident = "Apple MacBookPro5", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), - DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro5,1"), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro5"), }, }, { } diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 6e1a368d21d..71f4368b357 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -275,15 +275,20 @@ static unsigned long pit_calibrate_tsc(u32 latch, unsigned long ms, int loopmin) * use the TSC value at the transitions to calculate a pretty * good value for the TSC frequencty. */ +static inline int pit_verify_msb(unsigned char val) +{ + /* Ignore LSB */ + inb(0x42); + return inb(0x42) == val; +} + static inline int pit_expect_msb(unsigned char val, u64 *tscp, unsigned long *deltap) { int count; u64 tsc = 0; for (count = 0; count < 50000; count++) { - /* Ignore LSB */ - inb(0x42); - if (inb(0x42) != val) + if (!pit_verify_msb(val)) break; tsc = get_cycles(); } @@ -336,8 +341,7 @@ static unsigned long quick_pit_calibrate(void) * to do that is to just read back the 16-bit counter * once from the PIT. */ - inb(0x42); - inb(0x42); + pit_verify_msb(0); if (pit_expect_msb(0xff, &tsc, &d1)) { for (i = 1; i <= MAX_QUICK_PIT_ITERATIONS; i++) { @@ -348,8 +352,19 @@ static unsigned long quick_pit_calibrate(void) * Iterate until the error is less than 500 ppm */ delta -= tsc; - if (d1+d2 < delta >> 11) - goto success; + if (d1+d2 >= delta >> 11) + continue; + + /* + * Check the PIT one more time to verify that + * all TSC reads were stable wrt the PIT. + * + * This also guarantees serialization of the + * last cycle read ('d2') in pit_expect_msb. + */ + if (!pit_verify_msb(0xfe - i)) + break; + goto success; } } printk("Fast TSC calibration failed\n"); diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 958c1fa4190..fe3eba5d6b3 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -219,6 +219,8 @@ enum { AHCI_HFLAG_SECT255 = (1 << 8), /* max 255 sectors */ AHCI_HFLAG_YES_NCQ = (1 << 9), /* force NCQ cap on */ AHCI_HFLAG_NO_SUSPEND = (1 << 10), /* don't suspend */ + AHCI_HFLAG_SRST_TOUT_IS_OFFLINE = (1 << 11), /* treat SRST timeout as + link offline */ /* ap->flags bits */ @@ -1663,6 +1665,7 @@ static int ahci_do_softreset(struct ata_link *link, unsigned int *class, int (*check_ready)(struct ata_link *link)) { struct ata_port *ap = link->ap; + struct ahci_host_priv *hpriv = ap->host->private_data; const char *reason = NULL; unsigned long now, msecs; struct ata_taskfile tf; @@ -1701,12 +1704,21 @@ static int ahci_do_softreset(struct ata_link *link, unsigned int *class, /* wait for link to become ready */ rc = ata_wait_after_reset(link, deadline, check_ready); - /* link occupied, -ENODEV too is an error */ - if (rc) { + if (rc == -EBUSY && hpriv->flags & AHCI_HFLAG_SRST_TOUT_IS_OFFLINE) { + /* + * Workaround for cases where link online status can't + * be trusted. Treat device readiness timeout as link + * offline. + */ + ata_link_printk(link, KERN_INFO, + "device not ready, treating as offline\n"); + *class = ATA_DEV_NONE; + } else if (rc) { + /* link occupied, -ENODEV too is an error */ reason = "device not ready"; goto fail; - } - *class = ahci_dev_classify(ap); + } else + *class = ahci_dev_classify(ap); DPRINTK("EXIT, class=%u\n", *class); return 0; @@ -1773,7 +1785,8 @@ static int ahci_sb600_softreset(struct ata_link *link, unsigned int *class, irq_sts = readl(port_mmio + PORT_IRQ_STAT); if (irq_sts & PORT_IRQ_BAD_PMP) { ata_link_printk(link, KERN_WARNING, - "failed due to HW bug, retry pmp=0\n"); + "applying SB600 PMP SRST workaround " + "and retrying\n"); rc = ahci_do_softreset(link, class, 0, deadline, ahci_check_ready); } @@ -2726,6 +2739,56 @@ static bool ahci_broken_suspend(struct pci_dev *pdev) return !ver || strcmp(ver, dmi->driver_data) < 0; } +static bool ahci_broken_online(struct pci_dev *pdev) +{ +#define ENCODE_BUSDEVFN(bus, slot, func) \ + (void *)(unsigned long)(((bus) << 8) | PCI_DEVFN((slot), (func))) + static const struct dmi_system_id sysids[] = { + /* + * There are several gigabyte boards which use + * SIMG5723s configured as hardware RAID. Certain + * 5723 firmware revisions shipped there keep the link + * online but fail to answer properly to SRST or + * IDENTIFY when no device is attached downstream + * causing libata to retry quite a few times leading + * to excessive detection delay. + * + * As these firmwares respond to the second reset try + * with invalid device signature, considering unknown + * sig as offline works around the problem acceptably. + */ + { + .ident = "EP45-DQ6", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, + "Gigabyte Technology Co., Ltd."), + DMI_MATCH(DMI_BOARD_NAME, "EP45-DQ6"), + }, + .driver_data = ENCODE_BUSDEVFN(0x0a, 0x00, 0), + }, + { + .ident = "EP45-DS5", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, + "Gigabyte Technology Co., Ltd."), + DMI_MATCH(DMI_BOARD_NAME, "EP45-DS5"), + }, + .driver_data = ENCODE_BUSDEVFN(0x03, 0x00, 0), + }, + { } /* terminate list */ + }; +#undef ENCODE_BUSDEVFN + const struct dmi_system_id *dmi = dmi_first_match(sysids); + unsigned int val; + + if (!dmi) + return false; + + val = (unsigned long)dmi->driver_data; + + return pdev->bus->number == (val >> 8) && pdev->devfn == (val & 0xff); +} + static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { static int printed_version; @@ -2841,6 +2904,12 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) "BIOS update required for suspend/resume\n"); } + if (ahci_broken_online(pdev)) { + hpriv->flags |= AHCI_HFLAG_SRST_TOUT_IS_OFFLINE; + dev_info(&pdev->dev, + "online status unreliable, applying workaround\n"); + } + /* CAP.NP sometimes indicate the index of the last enabled * port, at other times, that of the last possible port, so * determining the maximum port number requires looking at diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 8ac98ff16d7..072ba5ea138 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4302,6 +4302,9 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { { "WDC WD2500JD-00HBB0", "WD-WMAL71490727", ATA_HORKAGE_BROKEN_HPA }, { "MAXTOR 6L080L4", "A93.0500", ATA_HORKAGE_BROKEN_HPA }, + /* this one allows HPA unlocking but fails IOs on the area */ + { "OCZ-VERTEX", "1.30", ATA_HORKAGE_BROKEN_HPA }, + /* Devices which report 1 sector over size HPA */ { "ST340823A", NULL, ATA_HORKAGE_HPA_SIZE, }, { "ST320413A", NULL, ATA_HORKAGE_HPA_SIZE, }, diff --git a/drivers/ata/pata_at91.c b/drivers/ata/pata_at91.c index 5702affcb32..41c94b1ae49 100644 --- a/drivers/ata/pata_at91.c +++ b/drivers/ata/pata_at91.c @@ -250,7 +250,7 @@ static int __devinit pata_at91_probe(struct platform_device *pdev) ata_port_desc(ap, "no IRQ, using PIO polling"); } - info = kzalloc(sizeof(*info), GFP_KERNEL); + info = devm_kzalloc(dev, sizeof(*info), GFP_KERNEL); if (!info) { dev_err(dev, "failed to allocate memory for private data\n"); @@ -275,7 +275,7 @@ static int __devinit pata_at91_probe(struct platform_device *pdev) if (!info->ide_addr) { dev_err(dev, "failed to map IO base\n"); ret = -ENOMEM; - goto err_ide_ioremap; + goto err_put; } info->alt_addr = devm_ioremap(dev, @@ -284,7 +284,7 @@ static int __devinit pata_at91_probe(struct platform_device *pdev) if (!info->alt_addr) { dev_err(dev, "failed to map CTL base\n"); ret = -ENOMEM; - goto err_alt_ioremap; + goto err_put; } ap->ioaddr.cmd_addr = info->ide_addr; @@ -303,13 +303,8 @@ static int __devinit pata_at91_probe(struct platform_device *pdev) irq ? ata_sff_interrupt : NULL, irq_flags, &pata_at91_sht); -err_alt_ioremap: - devm_iounmap(dev, info->ide_addr); - -err_ide_ioremap: +err_put: clk_put(info->mck); - kfree(info); - return ret; } @@ -317,7 +312,6 @@ static int __devexit pata_at91_remove(struct platform_device *pdev) { struct ata_host *host = dev_get_drvdata(&pdev->dev); struct at91_ide_info *info; - struct device *dev = &pdev->dev; if (!host) return 0; @@ -328,11 +322,8 @@ static int __devexit pata_at91_remove(struct platform_device *pdev) if (!info) return 0; - devm_iounmap(dev, info->ide_addr); - devm_iounmap(dev, info->alt_addr); clk_put(info->mck); - kfree(info); return 0; } diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c index bec0b8ade66..45915566e4e 100644 --- a/drivers/ata/pata_atiixp.c +++ b/drivers/ata/pata_atiixp.c @@ -1,6 +1,7 @@ /* * pata_atiixp.c - ATI PATA for new ATA layer * (C) 2005 Red Hat Inc + * (C) 2009 Bartlomiej Zolnierkiewicz * * Based on * @@ -61,20 +62,19 @@ static void atiixp_set_pio_timing(struct ata_port *ap, struct ata_device *adev, struct pci_dev *pdev = to_pci_dev(ap->host->dev); int dn = 2 * ap->port_no + adev->devno; - - /* Check this is correct - the order is odd in both drivers */ int timing_shift = (16 * ap->port_no) + 8 * (adev->devno ^ 1); - u16 pio_mode_data, pio_timing_data; + u32 pio_timing_data; + u16 pio_mode_data; pci_read_config_word(pdev, ATIIXP_IDE_PIO_MODE, &pio_mode_data); pio_mode_data &= ~(0x7 << (4 * dn)); pio_mode_data |= pio << (4 * dn); pci_write_config_word(pdev, ATIIXP_IDE_PIO_MODE, pio_mode_data); - pci_read_config_word(pdev, ATIIXP_IDE_PIO_TIMING, &pio_timing_data); + pci_read_config_dword(pdev, ATIIXP_IDE_PIO_TIMING, &pio_timing_data); pio_timing_data &= ~(0xFF << timing_shift); pio_timing_data |= (pio_timings[pio] << timing_shift); - pci_write_config_word(pdev, ATIIXP_IDE_PIO_TIMING, pio_timing_data); + pci_write_config_dword(pdev, ATIIXP_IDE_PIO_TIMING, pio_timing_data); } /** @@ -119,16 +119,17 @@ static void atiixp_set_dmamode(struct ata_port *ap, struct ata_device *adev) udma_mode_data |= dma << (4 * dn); pci_write_config_word(pdev, ATIIXP_IDE_UDMA_MODE, udma_mode_data); } else { - u16 mwdma_timing_data; - /* Check this is correct - the order is odd in both drivers */ int timing_shift = (16 * ap->port_no) + 8 * (adev->devno ^ 1); + u32 mwdma_timing_data; dma -= XFER_MW_DMA_0; - pci_read_config_word(pdev, ATIIXP_IDE_MWDMA_TIMING, &mwdma_timing_data); + pci_read_config_dword(pdev, ATIIXP_IDE_MWDMA_TIMING, + &mwdma_timing_data); mwdma_timing_data &= ~(0xFF << timing_shift); mwdma_timing_data |= (mwdma_timings[dma] << timing_shift); - pci_write_config_word(pdev, ATIIXP_IDE_MWDMA_TIMING, mwdma_timing_data); + pci_write_config_dword(pdev, ATIIXP_IDE_MWDMA_TIMING, + mwdma_timing_data); } /* * We must now look at the PIO mode situation. We may need to diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c index b2d11f300c3..86a40582999 100644 --- a/drivers/ata/sata_nv.c +++ b/drivers/ata/sata_nv.c @@ -602,6 +602,7 @@ MODULE_VERSION(DRV_VERSION); static int adma_enabled; static int swncq_enabled = 1; +static int msi_enabled; static void nv_adma_register_mode(struct ata_port *ap) { @@ -2459,6 +2460,11 @@ static int nv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) } else if (type == SWNCQ) nv_swncq_host_init(host); + if (msi_enabled) { + dev_printk(KERN_NOTICE, &pdev->dev, "Using MSI\n"); + pci_enable_msi(pdev); + } + pci_set_master(pdev); return ata_host_activate(host, pdev->irq, ipriv->irq_handler, IRQF_SHARED, ipriv->sht); @@ -2558,4 +2564,6 @@ module_param_named(adma, adma_enabled, bool, 0444); MODULE_PARM_DESC(adma, "Enable use of ADMA (Default: false)"); module_param_named(swncq, swncq_enabled, bool, 0444); MODULE_PARM_DESC(swncq, "Enable use of SWNCQ (Default: true)"); +module_param_named(msi, msi_enabled, bool, 0444); +MODULE_PARM_DESC(msi, "Enable use of MSI (Default: false)"); diff --git a/drivers/base/platform.c b/drivers/base/platform.c index 81cb01bfc35..456594bd97b 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -483,9 +483,6 @@ int platform_driver_register(struct platform_driver *drv) drv->driver.remove = platform_drv_remove; if (drv->shutdown) drv->driver.shutdown = platform_drv_shutdown; - if (drv->suspend || drv->resume) - pr_warning("Platform driver '%s' needs updating - please use " - "dev_pm_ops\n", drv->driver.name); return driver_register(&drv->driver); } diff --git a/drivers/char/pty.c b/drivers/char/pty.c index 6e6942c45f5..d083c73d784 100644 --- a/drivers/char/pty.c +++ b/drivers/char/pty.c @@ -144,6 +144,8 @@ static int pty_write(struct tty_struct *tty, const unsigned char *buf, static int pty_write_room(struct tty_struct *tty) { + if (tty->stopped) + return 0; return pty_space(tty->link); } diff --git a/drivers/md/md.c b/drivers/md/md.c index 5b98bea4ff9..103f2d33fa8 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -359,6 +359,7 @@ static mddev_t * mddev_find(dev_t unit) else new->md_minor = MINOR(unit) >> MdpMinorShift; + mutex_init(&new->open_mutex); mutex_init(&new->reconfig_mutex); INIT_LIST_HEAD(&new->disks); INIT_LIST_HEAD(&new->all_mddevs); @@ -1974,17 +1975,14 @@ repeat: /* otherwise we have to go forward and ... */ mddev->events ++; if (!mddev->in_sync || mddev->recovery_cp != MaxSector) { /* not clean */ - /* .. if the array isn't clean, insist on an odd 'events' */ - if ((mddev->events&1)==0) { - mddev->events++; + /* .. if the array isn't clean, an 'even' event must also go + * to spares. */ + if ((mddev->events&1)==0) nospares = 0; - } } else { - /* otherwise insist on an even 'events' (for clean states) */ - if ((mddev->events&1)) { - mddev->events++; + /* otherwise an 'odd' event must go to spares */ + if ((mddev->events&1)) nospares = 0; - } } } @@ -3601,6 +3599,7 @@ max_sync_store(mddev_t *mddev, const char *buf, size_t len) if (max < mddev->resync_min) return -EINVAL; if (max < mddev->resync_max && + mddev->ro == 0 && test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)) return -EBUSY; @@ -4304,12 +4303,11 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open) struct gendisk *disk = mddev->gendisk; mdk_rdev_t *rdev; + mutex_lock(&mddev->open_mutex); if (atomic_read(&mddev->openers) > is_open) { printk("md: %s still in use.\n",mdname(mddev)); - return -EBUSY; - } - - if (mddev->pers) { + err = -EBUSY; + } else if (mddev->pers) { if (mddev->sync_thread) { set_bit(MD_RECOVERY_FROZEN, &mddev->recovery); @@ -4367,7 +4365,10 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open) set_disk_ro(disk, 1); clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery); } - +out: + mutex_unlock(&mddev->open_mutex); + if (err) + return err; /* * Free resources if final stop */ @@ -4433,7 +4434,6 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open) blk_integrity_unregister(disk); md_new_event(mddev); sysfs_notify_dirent(mddev->sysfs_state); -out: return err; } @@ -5518,12 +5518,12 @@ static int md_open(struct block_device *bdev, fmode_t mode) } BUG_ON(mddev != bdev->bd_disk->private_data); - if ((err = mutex_lock_interruptible_nested(&mddev->reconfig_mutex, 1))) + if ((err = mutex_lock_interruptible(&mddev->open_mutex))) goto out; err = 0; atomic_inc(&mddev->openers); - mddev_unlock(mddev); + mutex_unlock(&mddev->open_mutex); check_disk_change(bdev); out: diff --git a/drivers/md/md.h b/drivers/md/md.h index 78f03168baf..f8fc188bc76 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -223,6 +223,16 @@ struct mddev_s * so we don't loop trying */ int in_sync; /* know to not need resync */ + /* 'open_mutex' avoids races between 'md_open' and 'do_md_stop', so + * that we are never stopping an array while it is open. + * 'reconfig_mutex' protects all other reconfiguration. + * These locks are separate due to conflicting interactions + * with bdev->bd_mutex. + * Lock ordering is: + * reconfig_mutex -> bd_mutex : e.g. do_md_run -> revalidate_disk + * bd_mutex -> open_mutex: e.g. __blkdev_get -> md_open + */ + struct mutex open_mutex; struct mutex reconfig_mutex; atomic_t active; /* general refcount */ atomic_t openers; /* number of active opens */ diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 2b521ee67df..b8a2c5dc67b 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3785,7 +3785,7 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped conf->reshape_progress < raid5_size(mddev, 0, 0)) { sector_nr = raid5_size(mddev, 0, 0) - conf->reshape_progress; - } else if (mddev->delta_disks > 0 && + } else if (mddev->delta_disks >= 0 && conf->reshape_progress > 0) sector_nr = conf->reshape_progress; sector_div(sector_nr, new_data_disks); @@ -4509,7 +4509,26 @@ static int run(mddev_t *mddev) (old_disks-max_degraded)); /* here_old is the first stripe that we might need to read * from */ - if (here_new >= here_old) { + if (mddev->delta_disks == 0) { + /* We cannot be sure it is safe to start an in-place + * reshape. It is only safe if user-space if monitoring + * and taking constant backups. + * mdadm always starts a situation like this in + * readonly mode so it can take control before + * allowing any writes. So just check for that. + */ + if ((here_new * mddev->new_chunk_sectors != + here_old * mddev->chunk_sectors) || + mddev->ro == 0) { + printk(KERN_ERR "raid5: in-place reshape must be started" + " in read-only mode - aborting\n"); + return -EINVAL; + } + } else if (mddev->delta_disks < 0 + ? (here_new * mddev->new_chunk_sectors <= + here_old * mddev->chunk_sectors) + : (here_new * mddev->new_chunk_sectors >= + here_old * mddev->chunk_sectors)) { /* Reading from the same stripe as writing to - bad */ printk(KERN_ERR "raid5: reshape_position too early for " "auto-recovery - aborting.\n"); @@ -5078,8 +5097,15 @@ static void raid5_finish_reshape(mddev_t *mddev) mddev->degraded--; for (d = conf->raid_disks ; d < conf->raid_disks - mddev->delta_disks; - d++) - raid5_remove_disk(mddev, d); + d++) { + mdk_rdev_t *rdev = conf->disks[d].rdev; + if (rdev && raid5_remove_disk(mddev, d) == 0) { + char nm[20]; + sprintf(nm, "rd%d", rdev->raid_disk); + sysfs_remove_link(&mddev->kobj, nm); + rdev->raid_disk = -1; + } + } } mddev->layout = conf->algorithm; mddev->chunk_sectors = conf->chunk_sectors; diff --git a/drivers/mtd/maps/sbc8240.c b/drivers/mtd/maps/sbc8240.c deleted file mode 100644 index e69de29bb2d..00000000000 --- a/drivers/mtd/maps/sbc8240.c +++ /dev/null diff --git a/drivers/pci/hotplug/sgi_hotplug.c b/drivers/pci/hotplug/sgi_hotplug.c index a4494d78e7c..8aebe1e9d3d 100644 --- a/drivers/pci/hotplug/sgi_hotplug.c +++ b/drivers/pci/hotplug/sgi_hotplug.c @@ -90,11 +90,10 @@ static struct hotplug_slot_ops sn_hotplug_slot_ops = { static DEFINE_MUTEX(sn_hotplug_mutex); -static ssize_t path_show (struct hotplug_slot *bss_hotplug_slot, - char *buf) +static ssize_t path_show(struct pci_slot *pci_slot, char *buf) { int retval = -ENOENT; - struct slot *slot = bss_hotplug_slot->private; + struct slot *slot = pci_slot->hotplug->private; if (!slot) return retval; @@ -103,7 +102,7 @@ static ssize_t path_show (struct hotplug_slot *bss_hotplug_slot, return retval; } -static struct hotplug_slot_attribute sn_slot_path_attr = __ATTR_RO(path); +static struct pci_slot_attribute sn_slot_path_attr = __ATTR_RO(path); static int sn_pci_slot_valid(struct pci_bus *pci_bus, int device) { diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 489fc01a320..e4e089a8f29 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -255,7 +255,7 @@ static void nfs_direct_read_release(void *calldata) if (put_dreq(dreq)) nfs_direct_complete(dreq); - nfs_readdata_release(calldata); + nfs_readdata_free(data); } static const struct rpc_call_ops nfs_read_direct_ops = { @@ -314,14 +314,14 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq, data->npages, 1, 0, data->pagevec, NULL); up_read(¤t->mm->mmap_sem); if (result < 0) { - nfs_readdata_release(data); + nfs_readdata_free(data); break; } if ((unsigned)result < data->npages) { bytes = result * PAGE_SIZE; if (bytes <= pgbase) { nfs_direct_release_pages(data->pagevec, result); - nfs_readdata_release(data); + nfs_readdata_free(data); break; } bytes -= pgbase; @@ -334,7 +334,7 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq, data->inode = inode; data->cred = msg.rpc_cred; data->args.fh = NFS_FH(inode); - data->args.context = get_nfs_open_context(ctx); + data->args.context = ctx; data->args.offset = pos; data->args.pgbase = pgbase; data->args.pages = data->pagevec; @@ -441,7 +441,7 @@ static void nfs_direct_free_writedata(struct nfs_direct_req *dreq) struct nfs_write_data *data = list_entry(dreq->rewrite_list.next, struct nfs_write_data, pages); list_del(&data->pages); nfs_direct_release_pages(data->pagevec, data->npages); - nfs_writedata_release(data); + nfs_writedata_free(data); } } @@ -534,7 +534,7 @@ static void nfs_direct_commit_release(void *calldata) dprintk("NFS: %5u commit returned %d\n", data->task.tk_pid, status); nfs_direct_write_complete(dreq, data->inode); - nfs_commitdata_release(calldata); + nfs_commit_free(data); } static const struct rpc_call_ops nfs_commit_direct_ops = { @@ -570,7 +570,7 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq) data->args.fh = NFS_FH(data->inode); data->args.offset = 0; data->args.count = 0; - data->args.context = get_nfs_open_context(dreq->ctx); + data->args.context = dreq->ctx; data->res.count = 0; data->res.fattr = &data->fattr; data->res.verf = &data->verf; @@ -734,14 +734,14 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq, data->npages, 0, 0, data->pagevec, NULL); up_read(¤t->mm->mmap_sem); if (result < 0) { - nfs_writedata_release(data); + nfs_writedata_free(data); break; } if ((unsigned)result < data->npages) { bytes = result * PAGE_SIZE; if (bytes <= pgbase) { nfs_direct_release_pages(data->pagevec, result); - nfs_writedata_release(data); + nfs_writedata_free(data); break; } bytes -= pgbase; @@ -756,7 +756,7 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq, data->inode = inode; data->cred = msg.rpc_cred; data->args.fh = NFS_FH(inode); - data->args.context = get_nfs_open_context(ctx); + data->args.context = ctx; data->args.offset = pos; data->args.pgbase = pgbase; data->args.pages = data->pagevec; diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 73ea5e8d66c..12c9e66d3f1 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -60,17 +60,15 @@ struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount) return p; } -static void nfs_readdata_free(struct nfs_read_data *p) +void nfs_readdata_free(struct nfs_read_data *p) { if (p && (p->pagevec != &p->page_array[0])) kfree(p->pagevec); mempool_free(p, nfs_rdata_mempool); } -void nfs_readdata_release(void *data) +static void nfs_readdata_release(struct nfs_read_data *rdata) { - struct nfs_read_data *rdata = data; - put_nfs_open_context(rdata->args.context); nfs_readdata_free(rdata); } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 0a0a2ff767c..a34fae21fe1 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -87,17 +87,15 @@ struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount) return p; } -static void nfs_writedata_free(struct nfs_write_data *p) +void nfs_writedata_free(struct nfs_write_data *p) { if (p && (p->pagevec != &p->page_array[0])) kfree(p->pagevec); mempool_free(p, nfs_wdata_mempool); } -void nfs_writedata_release(void *data) +static void nfs_writedata_release(struct nfs_write_data *wdata) { - struct nfs_write_data *wdata = data; - put_nfs_open_context(wdata->args.context); nfs_writedata_free(wdata); } diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index 9edcde4974a..f9a3e894266 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -1914,7 +1914,8 @@ static void ocfs2_adjust_adjacent_records(struct ocfs2_extent_rec *left_rec, * immediately to their right. */ left_clusters = le32_to_cpu(right_child_el->l_recs[0].e_cpos); - if (ocfs2_is_empty_extent(&right_child_el->l_recs[0])) { + if (!ocfs2_rec_clusters(right_child_el, &right_child_el->l_recs[0])) { + BUG_ON(right_child_el->l_tree_depth); BUG_ON(le16_to_cpu(right_child_el->l_next_free_rec) <= 1); left_clusters = le32_to_cpu(right_child_el->l_recs[1].e_cpos); } @@ -2476,15 +2477,37 @@ out_ret_path: return ret; } -static void ocfs2_update_edge_lengths(struct inode *inode, handle_t *handle, - struct ocfs2_path *path) +static int ocfs2_update_edge_lengths(struct inode *inode, handle_t *handle, + int subtree_index, struct ocfs2_path *path) { - int i, idx; + int i, idx, ret; struct ocfs2_extent_rec *rec; struct ocfs2_extent_list *el; struct ocfs2_extent_block *eb; u32 range; + /* + * In normal tree rotation process, we will never touch the + * tree branch above subtree_index and ocfs2_extend_rotate_transaction + * doesn't reserve the credits for them either. + * + * But we do have a special case here which will update the rightmost + * records for all the bh in the path. + * So we have to allocate extra credits and access them. + */ + ret = ocfs2_extend_trans(handle, + handle->h_buffer_credits + subtree_index); + if (ret) { + mlog_errno(ret); + goto out; + } + + ret = ocfs2_journal_access_path(inode, handle, path); + if (ret) { + mlog_errno(ret); + goto out; + } + /* Path should always be rightmost. */ eb = (struct ocfs2_extent_block *)path_leaf_bh(path)->b_data; BUG_ON(eb->h_next_leaf_blk != 0ULL); @@ -2505,6 +2528,8 @@ static void ocfs2_update_edge_lengths(struct inode *inode, handle_t *handle, ocfs2_journal_dirty(handle, path->p_node[i].bh); } +out: + return ret; } static void ocfs2_unlink_path(struct inode *inode, handle_t *handle, @@ -2717,7 +2742,12 @@ static int ocfs2_rotate_subtree_left(struct inode *inode, handle_t *handle, if (del_right_subtree) { ocfs2_unlink_subtree(inode, handle, left_path, right_path, subtree_index, dealloc); - ocfs2_update_edge_lengths(inode, handle, left_path); + ret = ocfs2_update_edge_lengths(inode, handle, subtree_index, + left_path); + if (ret) { + mlog_errno(ret); + goto out; + } eb = (struct ocfs2_extent_block *)path_leaf_bh(left_path)->b_data; ocfs2_et_set_last_eb_blk(et, le64_to_cpu(eb->h_blkno)); @@ -3034,7 +3064,12 @@ static int ocfs2_remove_rightmost_path(struct inode *inode, handle_t *handle, ocfs2_unlink_subtree(inode, handle, left_path, path, subtree_index, dealloc); - ocfs2_update_edge_lengths(inode, handle, left_path); + ret = ocfs2_update_edge_lengths(inode, handle, subtree_index, + left_path); + if (ret) { + mlog_errno(ret); + goto out; + } eb = (struct ocfs2_extent_block *)path_leaf_bh(left_path)->b_data; ocfs2_et_set_last_eb_blk(et, le64_to_cpu(eb->h_blkno)); diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index b2c52b3a148..b401654011a 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -193,6 +193,7 @@ static int ocfs2_get_block(struct inode *inode, sector_t iblock, (unsigned long long)OCFS2_I(inode)->ip_blkno); mlog(ML_ERROR, "Size %llu, clusters %u\n", (unsigned long long)i_size_read(inode), OCFS2_I(inode)->ip_clusters); dump_stack(); + goto bail; } past_eof = ocfs2_blocks_for_bytes(inode->i_sb, i_size_read(inode)); @@ -894,18 +895,17 @@ struct ocfs2_write_cluster_desc { */ unsigned c_new; unsigned c_unwritten; + unsigned c_needs_zero; }; -static inline int ocfs2_should_zero_cluster(struct ocfs2_write_cluster_desc *d) -{ - return d->c_new || d->c_unwritten; -} - struct ocfs2_write_ctxt { /* Logical cluster position / len of write */ u32 w_cpos; u32 w_clen; + /* First cluster allocated in a nonsparse extend */ + u32 w_first_new_cpos; + struct ocfs2_write_cluster_desc w_desc[OCFS2_MAX_CLUSTERS_PER_PAGE]; /* @@ -983,6 +983,7 @@ static int ocfs2_alloc_write_ctxt(struct ocfs2_write_ctxt **wcp, return -ENOMEM; wc->w_cpos = pos >> osb->s_clustersize_bits; + wc->w_first_new_cpos = UINT_MAX; cend = (pos + len - 1) >> osb->s_clustersize_bits; wc->w_clen = cend - wc->w_cpos + 1; get_bh(di_bh); @@ -1217,20 +1218,18 @@ out: */ static int ocfs2_write_cluster(struct address_space *mapping, u32 phys, unsigned int unwritten, + unsigned int should_zero, struct ocfs2_alloc_context *data_ac, struct ocfs2_alloc_context *meta_ac, struct ocfs2_write_ctxt *wc, u32 cpos, loff_t user_pos, unsigned user_len) { - int ret, i, new, should_zero = 0; + int ret, i, new; u64 v_blkno, p_blkno; struct inode *inode = mapping->host; struct ocfs2_extent_tree et; new = phys == 0 ? 1 : 0; - if (new || unwritten) - should_zero = 1; - if (new) { u32 tmp_pos; @@ -1301,7 +1300,7 @@ static int ocfs2_write_cluster(struct address_space *mapping, if (tmpret) { mlog_errno(tmpret); if (ret == 0) - tmpret = ret; + ret = tmpret; } } @@ -1341,7 +1340,9 @@ static int ocfs2_write_cluster_by_desc(struct address_space *mapping, local_len = osb->s_clustersize - cluster_off; ret = ocfs2_write_cluster(mapping, desc->c_phys, - desc->c_unwritten, data_ac, meta_ac, + desc->c_unwritten, + desc->c_needs_zero, + data_ac, meta_ac, wc, desc->c_cpos, pos, local_len); if (ret) { mlog_errno(ret); @@ -1391,14 +1392,14 @@ static void ocfs2_set_target_boundaries(struct ocfs2_super *osb, * newly allocated cluster. */ desc = &wc->w_desc[0]; - if (ocfs2_should_zero_cluster(desc)) + if (desc->c_needs_zero) ocfs2_figure_cluster_boundaries(osb, desc->c_cpos, &wc->w_target_from, NULL); desc = &wc->w_desc[wc->w_clen - 1]; - if (ocfs2_should_zero_cluster(desc)) + if (desc->c_needs_zero) ocfs2_figure_cluster_boundaries(osb, desc->c_cpos, NULL, @@ -1466,13 +1467,28 @@ static int ocfs2_populate_write_desc(struct inode *inode, phys++; } + /* + * If w_first_new_cpos is < UINT_MAX, we have a non-sparse + * file that got extended. w_first_new_cpos tells us + * where the newly allocated clusters are so we can + * zero them. + */ + if (desc->c_cpos >= wc->w_first_new_cpos) { + BUG_ON(phys == 0); + desc->c_needs_zero = 1; + } + desc->c_phys = phys; if (phys == 0) { desc->c_new = 1; + desc->c_needs_zero = 1; *clusters_to_alloc = *clusters_to_alloc + 1; } - if (ext_flags & OCFS2_EXT_UNWRITTEN) + + if (ext_flags & OCFS2_EXT_UNWRITTEN) { desc->c_unwritten = 1; + desc->c_needs_zero = 1; + } num_clusters--; } @@ -1632,10 +1648,13 @@ static int ocfs2_expand_nonsparse_inode(struct inode *inode, loff_t pos, if (newsize <= i_size_read(inode)) return 0; - ret = ocfs2_extend_no_holes(inode, newsize, newsize - len); + ret = ocfs2_extend_no_holes(inode, newsize, pos); if (ret) mlog_errno(ret); + wc->w_first_new_cpos = + ocfs2_clusters_for_bytes(inode->i_sb, i_size_read(inode)); + return ret; } @@ -1644,7 +1663,7 @@ int ocfs2_write_begin_nolock(struct address_space *mapping, struct page **pagep, void **fsdata, struct buffer_head *di_bh, struct page *mmap_page) { - int ret, credits = OCFS2_INODE_UPDATE_CREDITS; + int ret, cluster_of_pages, credits = OCFS2_INODE_UPDATE_CREDITS; unsigned int clusters_to_alloc, extents_to_split; struct ocfs2_write_ctxt *wc; struct inode *inode = mapping->host; @@ -1722,8 +1741,19 @@ int ocfs2_write_begin_nolock(struct address_space *mapping, } - ocfs2_set_target_boundaries(osb, wc, pos, len, - clusters_to_alloc + extents_to_split); + /* + * We have to zero sparse allocated clusters, unwritten extent clusters, + * and non-sparse clusters we just extended. For non-sparse writes, + * we know zeros will only be needed in the first and/or last cluster. + */ + if (clusters_to_alloc || extents_to_split || + wc->w_desc[0].c_needs_zero || + wc->w_desc[wc->w_clen - 1].c_needs_zero) + cluster_of_pages = 1; + else + cluster_of_pages = 0; + + ocfs2_set_target_boundaries(osb, wc, pos, len, cluster_of_pages); handle = ocfs2_start_trans(osb, credits); if (IS_ERR(handle)) { @@ -1756,8 +1786,7 @@ int ocfs2_write_begin_nolock(struct address_space *mapping, * extent. */ ret = ocfs2_grab_pages_for_write(mapping, wc, wc->w_cpos, pos, - clusters_to_alloc + extents_to_split, - mmap_page); + cluster_of_pages, mmap_page); if (ret) { mlog_errno(ret); goto out_quota; diff --git a/fs/ocfs2/dcache.c b/fs/ocfs2/dcache.c index b574431a031..2f28b7de2c8 100644 --- a/fs/ocfs2/dcache.c +++ b/fs/ocfs2/dcache.c @@ -310,22 +310,19 @@ out_attach: return ret; } -static DEFINE_SPINLOCK(dentry_list_lock); +DEFINE_SPINLOCK(dentry_list_lock); /* We limit the number of dentry locks to drop in one go. We have * this limit so that we don't starve other users of ocfs2_wq. */ #define DL_INODE_DROP_COUNT 64 /* Drop inode references from dentry locks */ -void ocfs2_drop_dl_inodes(struct work_struct *work) +static void __ocfs2_drop_dl_inodes(struct ocfs2_super *osb, int drop_count) { - struct ocfs2_super *osb = container_of(work, struct ocfs2_super, - dentry_lock_work); struct ocfs2_dentry_lock *dl; - int drop_count = DL_INODE_DROP_COUNT; spin_lock(&dentry_list_lock); - while (osb->dentry_lock_list && drop_count--) { + while (osb->dentry_lock_list && (drop_count < 0 || drop_count--)) { dl = osb->dentry_lock_list; osb->dentry_lock_list = dl->dl_next; spin_unlock(&dentry_list_lock); @@ -333,11 +330,32 @@ void ocfs2_drop_dl_inodes(struct work_struct *work) kfree(dl); spin_lock(&dentry_list_lock); } - if (osb->dentry_lock_list) + spin_unlock(&dentry_list_lock); +} + +void ocfs2_drop_dl_inodes(struct work_struct *work) +{ + struct ocfs2_super *osb = container_of(work, struct ocfs2_super, + dentry_lock_work); + + __ocfs2_drop_dl_inodes(osb, DL_INODE_DROP_COUNT); + /* + * Don't queue dropping if umount is in progress. We flush the + * list in ocfs2_dismount_volume + */ + spin_lock(&dentry_list_lock); + if (osb->dentry_lock_list && + !ocfs2_test_osb_flag(osb, OCFS2_OSB_DROP_DENTRY_LOCK_IMMED)) queue_work(ocfs2_wq, &osb->dentry_lock_work); spin_unlock(&dentry_list_lock); } +/* Flush the whole work queue */ +void ocfs2_drop_all_dl_inodes(struct ocfs2_super *osb) +{ + __ocfs2_drop_dl_inodes(osb, -1); +} + /* * ocfs2_dentry_iput() and friends. * @@ -368,7 +386,8 @@ static void ocfs2_drop_dentry_lock(struct ocfs2_super *osb, /* We leave dropping of inode reference to ocfs2_wq as that can * possibly lead to inode deletion which gets tricky */ spin_lock(&dentry_list_lock); - if (!osb->dentry_lock_list) + if (!osb->dentry_lock_list && + !ocfs2_test_osb_flag(osb, OCFS2_OSB_DROP_DENTRY_LOCK_IMMED)) queue_work(ocfs2_wq, &osb->dentry_lock_work); dl->dl_next = osb->dentry_lock_list; osb->dentry_lock_list = dl; diff --git a/fs/ocfs2/dcache.h b/fs/ocfs2/dcache.h index faa12e75f98..f5dd1789acf 100644 --- a/fs/ocfs2/dcache.h +++ b/fs/ocfs2/dcache.h @@ -49,10 +49,13 @@ struct ocfs2_dentry_lock { int ocfs2_dentry_attach_lock(struct dentry *dentry, struct inode *inode, u64 parent_blkno); +extern spinlock_t dentry_list_lock; + void ocfs2_dentry_lock_put(struct ocfs2_super *osb, struct ocfs2_dentry_lock *dl); void ocfs2_drop_dl_inodes(struct work_struct *work); +void ocfs2_drop_all_dl_inodes(struct ocfs2_super *osb); struct dentry *ocfs2_find_local_alias(struct inode *inode, u64 parent_blkno, int skip_unhashed); diff --git a/fs/ocfs2/dlm/dlmast.c b/fs/ocfs2/dlm/dlmast.c index d07ddbe4b28..81eff8e5832 100644 --- a/fs/ocfs2/dlm/dlmast.c +++ b/fs/ocfs2/dlm/dlmast.c @@ -103,7 +103,6 @@ static void __dlm_queue_ast(struct dlm_ctxt *dlm, struct dlm_lock *lock) lock->ast_pending, lock->ml.type); BUG(); } - BUG_ON(!list_empty(&lock->ast_list)); if (lock->ast_pending) mlog(0, "lock has an ast getting flushed right now\n"); diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index bcb9260c373..43e6e328056 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -1118,7 +1118,7 @@ static int dlm_send_mig_lockres_msg(struct dlm_ctxt *dlm, mlog(0, "%s:%.*s: sending mig lockres (%s) to %u\n", dlm->name, res->lockname.len, res->lockname.name, - orig_flags & DLM_MRES_MIGRATION ? "migrate" : "recovery", + orig_flags & DLM_MRES_MIGRATION ? "migration" : "recovery", send_to); /* send it */ diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 62442e413a0..aa501d3f93f 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1851,6 +1851,7 @@ relock: if (ret) goto out_dio; + count = ocount; ret = generic_write_checks(file, ppos, &count, S_ISBLK(inode->i_mode)); if (ret) @@ -1918,8 +1919,10 @@ out_sems: mutex_unlock(&inode->i_mutex); + if (written) + ret = written; mlog_exit(ret); - return written ? written : ret; + return ret; } static int ocfs2_splice_to_file(struct pipe_inode_info *pipe, diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index f033760ecbe..c48b93ac6b6 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -1954,10 +1954,16 @@ void ocfs2_orphan_scan_init(struct ocfs2_super *osb) os->os_osb = osb; os->os_count = 0; os->os_seqno = 0; - os->os_scantime = CURRENT_TIME; mutex_init(&os->os_lock); INIT_DELAYED_WORK(&os->os_orphan_scan_work, ocfs2_orphan_scan_work); +} +void ocfs2_orphan_scan_start(struct ocfs2_super *osb) +{ + struct ocfs2_orphan_scan *os; + + os = &osb->osb_orphan_scan; + os->os_scantime = CURRENT_TIME; if (ocfs2_is_hard_readonly(osb) || ocfs2_mount_local(osb)) atomic_set(&os->os_state, ORPHAN_SCAN_INACTIVE); else { diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index 5432c7f79cc..2c3222aec62 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -145,6 +145,7 @@ static inline void ocfs2_inode_set_new(struct ocfs2_super *osb, /* Exported only for the journal struct init code in super.c. Do not call. */ void ocfs2_orphan_scan_init(struct ocfs2_super *osb); +void ocfs2_orphan_scan_start(struct ocfs2_super *osb); void ocfs2_orphan_scan_stop(struct ocfs2_super *osb); void ocfs2_orphan_scan_exit(struct ocfs2_super *osb); @@ -329,20 +330,27 @@ int ocfs2_journal_dirty(handle_t *handle, /* extended attribute block update */ #define OCFS2_XATTR_BLOCK_UPDATE_CREDITS 1 +/* Update of a single quota block */ +#define OCFS2_QUOTA_BLOCK_UPDATE_CREDITS 1 + /* global quotafile inode update, data block */ -#define OCFS2_QINFO_WRITE_CREDITS (OCFS2_INODE_UPDATE_CREDITS + 1) +#define OCFS2_QINFO_WRITE_CREDITS (OCFS2_INODE_UPDATE_CREDITS + \ + OCFS2_QUOTA_BLOCK_UPDATE_CREDITS) +#define OCFS2_LOCAL_QINFO_WRITE_CREDITS OCFS2_QUOTA_BLOCK_UPDATE_CREDITS /* * The two writes below can accidentally see global info dirty due * to set_info() quotactl so make them prepared for the writes. */ /* quota data block, global info */ /* Write to local quota file */ -#define OCFS2_QWRITE_CREDITS (OCFS2_QINFO_WRITE_CREDITS + 1) +#define OCFS2_QWRITE_CREDITS (OCFS2_QINFO_WRITE_CREDITS + \ + OCFS2_QUOTA_BLOCK_UPDATE_CREDITS) /* global quota data block, local quota data block, global quota inode, * global quota info */ -#define OCFS2_QSYNC_CREDITS (OCFS2_INODE_UPDATE_CREDITS + 3) +#define OCFS2_QSYNC_CREDITS (OCFS2_QINFO_WRITE_CREDITS + \ + 2 * OCFS2_QUOTA_BLOCK_UPDATE_CREDITS) static inline int ocfs2_quota_trans_credits(struct super_block *sb) { @@ -355,11 +363,6 @@ static inline int ocfs2_quota_trans_credits(struct super_block *sb) return credits; } -/* Number of credits needed for removing quota structure from file */ -int ocfs2_calc_qdel_credits(struct super_block *sb, int type); -/* Number of credits needed for initialization of new quota structure */ -int ocfs2_calc_qinit_credits(struct super_block *sb, int type); - /* group extend. inode update and last group update. */ #define OCFS2_GROUP_EXTEND_CREDITS (OCFS2_INODE_UPDATE_CREDITS + 1) diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index c9345ebb849..39e1d5a3950 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -224,10 +224,12 @@ enum ocfs2_mount_options OCFS2_MOUNT_GRPQUOTA = 1 << 10, /* We support group quotas */ }; -#define OCFS2_OSB_SOFT_RO 0x0001 -#define OCFS2_OSB_HARD_RO 0x0002 -#define OCFS2_OSB_ERROR_FS 0x0004 -#define OCFS2_DEFAULT_ATIME_QUANTUM 60 +#define OCFS2_OSB_SOFT_RO 0x0001 +#define OCFS2_OSB_HARD_RO 0x0002 +#define OCFS2_OSB_ERROR_FS 0x0004 +#define OCFS2_OSB_DROP_DENTRY_LOCK_IMMED 0x0008 + +#define OCFS2_DEFAULT_ATIME_QUANTUM 60 struct ocfs2_journal; struct ocfs2_slot_info; @@ -490,6 +492,18 @@ static inline void ocfs2_set_osb_flag(struct ocfs2_super *osb, spin_unlock(&osb->osb_lock); } + +static inline unsigned long ocfs2_test_osb_flag(struct ocfs2_super *osb, + unsigned long flag) +{ + unsigned long ret; + + spin_lock(&osb->osb_lock); + ret = osb->osb_flags & flag; + spin_unlock(&osb->osb_lock); + return ret; +} + static inline void ocfs2_set_ro_flag(struct ocfs2_super *osb, int hard) { diff --git a/fs/ocfs2/quota.h b/fs/ocfs2/quota.h index 7365e2e0870..3fb96fcd4c8 100644 --- a/fs/ocfs2/quota.h +++ b/fs/ocfs2/quota.h @@ -50,7 +50,6 @@ struct ocfs2_mem_dqinfo { unsigned int dqi_chunks; /* Number of chunks in local quota file */ unsigned int dqi_blocks; /* Number of blocks allocated for local quota file */ unsigned int dqi_syncms; /* How often should we sync with other nodes */ - unsigned int dqi_syncjiff; /* Precomputed dqi_syncms in jiffies */ struct list_head dqi_chunk; /* List of chunks */ struct inode *dqi_gqinode; /* Global quota file inode */ struct ocfs2_lock_res dqi_gqlock; /* Lock protecting quota information structure */ diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c index edfa60cd155..bf7742d0ee3 100644 --- a/fs/ocfs2/quota_global.c +++ b/fs/ocfs2/quota_global.c @@ -69,6 +69,7 @@ static void ocfs2_global_mem2diskdqb(void *dp, struct dquot *dquot) d->dqb_curspace = cpu_to_le64(m->dqb_curspace); d->dqb_btime = cpu_to_le64(m->dqb_btime); d->dqb_itime = cpu_to_le64(m->dqb_itime); + d->dqb_pad1 = d->dqb_pad2 = 0; } static int ocfs2_global_is_id(void *dp, struct dquot *dquot) @@ -211,14 +212,13 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type, mutex_lock_nested(&gqinode->i_mutex, I_MUTEX_QUOTA); if (gqinode->i_size < off + len) { - down_write(&OCFS2_I(gqinode)->ip_alloc_sem); - err = ocfs2_extend_no_holes(gqinode, off + len, off); - up_write(&OCFS2_I(gqinode)->ip_alloc_sem); - if (err < 0) - goto out; + loff_t rounded_end = + ocfs2_align_bytes_to_blocks(sb, off + len); + + /* Space is already allocated in ocfs2_global_read_dquot() */ err = ocfs2_simple_size_update(gqinode, oinfo->dqi_gqi_bh, - off + len); + rounded_end); if (err < 0) goto out; new = 1; @@ -234,7 +234,7 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type, } if (err) { mlog_errno(err); - return err; + goto out; } lock_buffer(bh); if (new) @@ -342,7 +342,6 @@ int ocfs2_global_read_info(struct super_block *sb, int type) info->dqi_bgrace = le32_to_cpu(dinfo.dqi_bgrace); info->dqi_igrace = le32_to_cpu(dinfo.dqi_igrace); oinfo->dqi_syncms = le32_to_cpu(dinfo.dqi_syncms); - oinfo->dqi_syncjiff = msecs_to_jiffies(oinfo->dqi_syncms); oinfo->dqi_gi.dqi_blocks = le32_to_cpu(dinfo.dqi_blocks); oinfo->dqi_gi.dqi_free_blk = le32_to_cpu(dinfo.dqi_free_blk); oinfo->dqi_gi.dqi_free_entry = le32_to_cpu(dinfo.dqi_free_entry); @@ -352,7 +351,7 @@ int ocfs2_global_read_info(struct super_block *sb, int type) oinfo->dqi_gi.dqi_qtree_depth = qtree_depth(&oinfo->dqi_gi); INIT_DELAYED_WORK(&oinfo->dqi_sync_work, qsync_work_fn); queue_delayed_work(ocfs2_quota_wq, &oinfo->dqi_sync_work, - oinfo->dqi_syncjiff); + msecs_to_jiffies(oinfo->dqi_syncms)); out_err: mlog_exit(status); @@ -402,13 +401,36 @@ int ocfs2_global_write_info(struct super_block *sb, int type) return err; } +static int ocfs2_global_qinit_alloc(struct super_block *sb, int type) +{ + struct ocfs2_mem_dqinfo *oinfo = sb_dqinfo(sb, type)->dqi_priv; + + /* + * We may need to allocate tree blocks and a leaf block but not the + * root block + */ + return oinfo->dqi_gi.dqi_qtree_depth; +} + +static int ocfs2_calc_global_qinit_credits(struct super_block *sb, int type) +{ + /* We modify all the allocated blocks, tree root, and info block */ + return (ocfs2_global_qinit_alloc(sb, type) + 2) * + OCFS2_QUOTA_BLOCK_UPDATE_CREDITS; +} + /* Read in information from global quota file and acquire a reference to it. * dquot_acquire() has already started the transaction and locked quota file */ int ocfs2_global_read_dquot(struct dquot *dquot) { int err, err2, ex = 0; - struct ocfs2_mem_dqinfo *info = - sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv; + struct super_block *sb = dquot->dq_sb; + int type = dquot->dq_type; + struct ocfs2_mem_dqinfo *info = sb_dqinfo(sb, type)->dqi_priv; + struct ocfs2_super *osb = OCFS2_SB(sb); + struct inode *gqinode = info->dqi_gqinode; + int need_alloc = ocfs2_global_qinit_alloc(sb, type); + handle_t *handle = NULL; err = ocfs2_qinfo_lock(info, 0); if (err < 0) @@ -419,14 +441,33 @@ int ocfs2_global_read_dquot(struct dquot *dquot) OCFS2_DQUOT(dquot)->dq_use_count++; OCFS2_DQUOT(dquot)->dq_origspace = dquot->dq_dqb.dqb_curspace; OCFS2_DQUOT(dquot)->dq_originodes = dquot->dq_dqb.dqb_curinodes; + ocfs2_qinfo_unlock(info, 0); + if (!dquot->dq_off) { /* No real quota entry? */ - /* Upgrade to exclusive lock for allocation */ - ocfs2_qinfo_unlock(info, 0); - err = ocfs2_qinfo_lock(info, 1); - if (err < 0) - goto out_qlock; ex = 1; + /* + * Add blocks to quota file before we start a transaction since + * locking allocators ranks above a transaction start + */ + WARN_ON(journal_current_handle()); + down_write(&OCFS2_I(gqinode)->ip_alloc_sem); + err = ocfs2_extend_no_holes(gqinode, + gqinode->i_size + (need_alloc << sb->s_blocksize_bits), + gqinode->i_size); + up_write(&OCFS2_I(gqinode)->ip_alloc_sem); + if (err < 0) + goto out; } + + handle = ocfs2_start_trans(osb, + ocfs2_calc_global_qinit_credits(sb, type)); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + goto out; + } + err = ocfs2_qinfo_lock(info, ex); + if (err < 0) + goto out_trans; err = qtree_write_dquot(&info->dqi_gi, dquot); if (ex && info_dirty(sb_dqinfo(dquot->dq_sb, dquot->dq_type))) { err2 = __ocfs2_global_write_info(dquot->dq_sb, dquot->dq_type); @@ -438,6 +479,9 @@ out_qlock: ocfs2_qinfo_unlock(info, 1); else ocfs2_qinfo_unlock(info, 0); +out_trans: + if (handle) + ocfs2_commit_trans(osb, handle); out: if (err < 0) mlog_errno(err); @@ -607,7 +651,7 @@ static void qsync_work_fn(struct work_struct *work) dquot_scan_active(sb, ocfs2_sync_dquot_helper, oinfo->dqi_type); queue_delayed_work(ocfs2_quota_wq, &oinfo->dqi_sync_work, - oinfo->dqi_syncjiff); + msecs_to_jiffies(oinfo->dqi_syncms)); } /* @@ -635,20 +679,18 @@ out: return status; } -int ocfs2_calc_qdel_credits(struct super_block *sb, int type) +static int ocfs2_calc_qdel_credits(struct super_block *sb, int type) { - struct ocfs2_mem_dqinfo *oinfo; - int features[MAXQUOTAS] = { OCFS2_FEATURE_RO_COMPAT_USRQUOTA, - OCFS2_FEATURE_RO_COMPAT_GRPQUOTA }; - - if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, features[type])) - return 0; - - oinfo = sb_dqinfo(sb, type)->dqi_priv; - /* We modify tree, leaf block, global info, local chunk header, - * global and local inode */ - return oinfo->dqi_gi.dqi_qtree_depth + 2 + 1 + - 2 * OCFS2_INODE_UPDATE_CREDITS; + struct ocfs2_mem_dqinfo *oinfo = sb_dqinfo(sb, type)->dqi_priv; + /* + * We modify tree, leaf block, global info, local chunk header, + * global and local inode; OCFS2_QINFO_WRITE_CREDITS already + * accounts for inode update + */ + return (oinfo->dqi_gi.dqi_qtree_depth + 2) * + OCFS2_QUOTA_BLOCK_UPDATE_CREDITS + + OCFS2_QINFO_WRITE_CREDITS + + OCFS2_INODE_UPDATE_CREDITS; } static int ocfs2_release_dquot(struct dquot *dquot) @@ -680,33 +722,10 @@ out: return status; } -int ocfs2_calc_qinit_credits(struct super_block *sb, int type) -{ - struct ocfs2_mem_dqinfo *oinfo; - int features[MAXQUOTAS] = { OCFS2_FEATURE_RO_COMPAT_USRQUOTA, - OCFS2_FEATURE_RO_COMPAT_GRPQUOTA }; - struct ocfs2_dinode *lfe, *gfe; - - if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, features[type])) - return 0; - - oinfo = sb_dqinfo(sb, type)->dqi_priv; - gfe = (struct ocfs2_dinode *)oinfo->dqi_gqi_bh->b_data; - lfe = (struct ocfs2_dinode *)oinfo->dqi_lqi_bh->b_data; - /* We can extend local file + global file. In local file we - * can modify info, chunk header block and dquot block. In - * global file we can modify info, tree and leaf block */ - return ocfs2_calc_extend_credits(sb, &lfe->id2.i_list, 0) + - ocfs2_calc_extend_credits(sb, &gfe->id2.i_list, 0) + - 3 + oinfo->dqi_gi.dqi_qtree_depth + 2; -} - static int ocfs2_acquire_dquot(struct dquot *dquot) { - handle_t *handle; struct ocfs2_mem_dqinfo *oinfo = sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv; - struct ocfs2_super *osb = OCFS2_SB(dquot->dq_sb); int status = 0; mlog_entry("id=%u, type=%d", dquot->dq_id, dquot->dq_type); @@ -715,16 +734,7 @@ static int ocfs2_acquire_dquot(struct dquot *dquot) status = ocfs2_lock_global_qf(oinfo, 1); if (status < 0) goto out; - handle = ocfs2_start_trans(osb, - ocfs2_calc_qinit_credits(dquot->dq_sb, dquot->dq_type)); - if (IS_ERR(handle)) { - status = PTR_ERR(handle); - mlog_errno(status); - goto out_ilock; - } status = dquot_acquire(dquot); - ocfs2_commit_trans(osb, handle); -out_ilock: ocfs2_unlock_global_qf(oinfo, 1); out: mlog_exit(status); diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c index 5a460fa8255..bdb09cb6e1f 100644 --- a/fs/ocfs2/quota_local.c +++ b/fs/ocfs2/quota_local.c @@ -20,6 +20,7 @@ #include "sysfile.h" #include "dlmglue.h" #include "quota.h" +#include "uptodate.h" /* Number of local quota structures per block */ static inline unsigned int ol_quota_entries_per_block(struct super_block *sb) @@ -100,7 +101,8 @@ static int ocfs2_modify_bh(struct inode *inode, struct buffer_head *bh, handle_t *handle; int status; - handle = ocfs2_start_trans(OCFS2_SB(sb), 1); + handle = ocfs2_start_trans(OCFS2_SB(sb), + OCFS2_QUOTA_BLOCK_UPDATE_CREDITS); if (IS_ERR(handle)) { status = PTR_ERR(handle); mlog_errno(status); @@ -610,7 +612,8 @@ int ocfs2_finish_quota_recovery(struct ocfs2_super *osb, goto out_bh; /* Mark quota file as clean if we are recovering quota file of * some other node. */ - handle = ocfs2_start_trans(osb, 1); + handle = ocfs2_start_trans(osb, + OCFS2_LOCAL_QINFO_WRITE_CREDITS); if (IS_ERR(handle)) { status = PTR_ERR(handle); mlog_errno(status); @@ -940,7 +943,7 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk( struct ocfs2_local_disk_chunk *dchunk; int status; handle_t *handle; - struct buffer_head *bh = NULL; + struct buffer_head *bh = NULL, *dbh = NULL; u64 p_blkno; /* We are protected by dqio_sem so no locking needed */ @@ -964,32 +967,35 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk( mlog_errno(status); goto out; } + /* Local quota info and two new blocks we initialize */ + handle = ocfs2_start_trans(OCFS2_SB(sb), + OCFS2_LOCAL_QINFO_WRITE_CREDITS + + 2 * OCFS2_QUOTA_BLOCK_UPDATE_CREDITS); + if (IS_ERR(handle)) { + status = PTR_ERR(handle); + mlog_errno(status); + goto out; + } + /* Initialize chunk header */ down_read(&OCFS2_I(lqinode)->ip_alloc_sem); status = ocfs2_extent_map_get_blocks(lqinode, oinfo->dqi_blocks, &p_blkno, NULL, NULL); up_read(&OCFS2_I(lqinode)->ip_alloc_sem); if (status < 0) { mlog_errno(status); - goto out; + goto out_trans; } bh = sb_getblk(sb, p_blkno); if (!bh) { status = -ENOMEM; mlog_errno(status); - goto out; + goto out_trans; } dchunk = (struct ocfs2_local_disk_chunk *)bh->b_data; - - handle = ocfs2_start_trans(OCFS2_SB(sb), 2); - if (IS_ERR(handle)) { - status = PTR_ERR(handle); - mlog_errno(status); - goto out; - } - + ocfs2_set_new_buffer_uptodate(lqinode, bh); status = ocfs2_journal_access_dq(handle, lqinode, bh, - OCFS2_JOURNAL_ACCESS_WRITE); + OCFS2_JOURNAL_ACCESS_CREATE); if (status < 0) { mlog_errno(status); goto out_trans; @@ -999,7 +1005,6 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk( memset(dchunk->dqc_bitmap, 0, sb->s_blocksize - sizeof(struct ocfs2_local_disk_chunk) - OCFS2_QBLK_RESERVED_SPACE); - set_buffer_uptodate(bh); unlock_buffer(bh); status = ocfs2_journal_dirty(handle, bh); if (status < 0) { @@ -1007,6 +1012,38 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk( goto out_trans; } + /* Initialize new block with structures */ + down_read(&OCFS2_I(lqinode)->ip_alloc_sem); + status = ocfs2_extent_map_get_blocks(lqinode, oinfo->dqi_blocks + 1, + &p_blkno, NULL, NULL); + up_read(&OCFS2_I(lqinode)->ip_alloc_sem); + if (status < 0) { + mlog_errno(status); + goto out_trans; + } + dbh = sb_getblk(sb, p_blkno); + if (!dbh) { + status = -ENOMEM; + mlog_errno(status); + goto out_trans; + } + ocfs2_set_new_buffer_uptodate(lqinode, dbh); + status = ocfs2_journal_access_dq(handle, lqinode, dbh, + OCFS2_JOURNAL_ACCESS_CREATE); + if (status < 0) { + mlog_errno(status); + goto out_trans; + } + lock_buffer(dbh); + memset(dbh->b_data, 0, sb->s_blocksize - OCFS2_QBLK_RESERVED_SPACE); + unlock_buffer(dbh); + status = ocfs2_journal_dirty(handle, dbh); + if (status < 0) { + mlog_errno(status); + goto out_trans; + } + + /* Update local quotafile info */ oinfo->dqi_blocks += 2; oinfo->dqi_chunks++; status = ocfs2_local_write_info(sb, type); @@ -1031,6 +1068,7 @@ out_trans: ocfs2_commit_trans(OCFS2_SB(sb), handle); out: brelse(bh); + brelse(dbh); kmem_cache_free(ocfs2_qf_chunk_cachep, chunk); return ERR_PTR(status); } @@ -1048,6 +1086,8 @@ static struct ocfs2_quota_chunk *ocfs2_extend_local_quota_file( struct ocfs2_local_disk_chunk *dchunk; int epb = ol_quota_entries_per_block(sb); unsigned int chunk_blocks; + struct buffer_head *bh; + u64 p_blkno; int status; handle_t *handle; @@ -1075,12 +1115,49 @@ static struct ocfs2_quota_chunk *ocfs2_extend_local_quota_file( mlog_errno(status); goto out; } - handle = ocfs2_start_trans(OCFS2_SB(sb), 2); + + /* Get buffer from the just added block */ + down_read(&OCFS2_I(lqinode)->ip_alloc_sem); + status = ocfs2_extent_map_get_blocks(lqinode, oinfo->dqi_blocks, + &p_blkno, NULL, NULL); + up_read(&OCFS2_I(lqinode)->ip_alloc_sem); + if (status < 0) { + mlog_errno(status); + goto out; + } + bh = sb_getblk(sb, p_blkno); + if (!bh) { + status = -ENOMEM; + mlog_errno(status); + goto out; + } + ocfs2_set_new_buffer_uptodate(lqinode, bh); + + /* Local quota info, chunk header and the new block we initialize */ + handle = ocfs2_start_trans(OCFS2_SB(sb), + OCFS2_LOCAL_QINFO_WRITE_CREDITS + + 2 * OCFS2_QUOTA_BLOCK_UPDATE_CREDITS); if (IS_ERR(handle)) { status = PTR_ERR(handle); mlog_errno(status); goto out; } + /* Zero created block */ + status = ocfs2_journal_access_dq(handle, lqinode, bh, + OCFS2_JOURNAL_ACCESS_CREATE); + if (status < 0) { + mlog_errno(status); + goto out_trans; + } + lock_buffer(bh); + memset(bh->b_data, 0, sb->s_blocksize); + unlock_buffer(bh); + status = ocfs2_journal_dirty(handle, bh); + if (status < 0) { + mlog_errno(status); + goto out_trans; + } + /* Update chunk header */ status = ocfs2_journal_access_dq(handle, lqinode, chunk->qc_headerbh, OCFS2_JOURNAL_ACCESS_WRITE); if (status < 0) { @@ -1097,6 +1174,7 @@ static struct ocfs2_quota_chunk *ocfs2_extend_local_quota_file( mlog_errno(status); goto out_trans; } + /* Update file header */ oinfo->dqi_blocks++; status = ocfs2_local_write_info(sb, type); if (status < 0) { diff --git a/fs/ocfs2/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c index 3f661376a2d..e49c4105026 100644 --- a/fs/ocfs2/stack_o2cb.c +++ b/fs/ocfs2/stack_o2cb.c @@ -17,6 +17,7 @@ * General Public License for more details. */ +#include <linux/kernel.h> #include <linux/crc32.h> #include <linux/module.h> @@ -153,7 +154,7 @@ static int status_map[] = { static int dlm_status_to_errno(enum dlm_status status) { - BUG_ON(status > (sizeof(status_map) / sizeof(status_map[0]))); + BUG_ON(status < 0 || status >= ARRAY_SIZE(status_map)); return status_map[status]; } diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 7efb349fb9b..b0ee0fdf799 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -777,6 +777,7 @@ static int ocfs2_sb_probe(struct super_block *sb, } di = (struct ocfs2_dinode *) (*bh)->b_data; memset(stats, 0, sizeof(struct ocfs2_blockcheck_stats)); + spin_lock_init(&stats->b_lock); status = ocfs2_verify_volume(di, *bh, blksize, stats); if (status >= 0) goto bail; @@ -1182,7 +1183,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) wake_up(&osb->osb_mount_event); /* Start this when the mount is almost sure of being successful */ - ocfs2_orphan_scan_init(osb); + ocfs2_orphan_scan_start(osb); mlog_exit(status); return status; @@ -1213,14 +1214,27 @@ static int ocfs2_get_sb(struct file_system_type *fs_type, mnt); } +static void ocfs2_kill_sb(struct super_block *sb) +{ + struct ocfs2_super *osb = OCFS2_SB(sb); + + /* Prevent further queueing of inode drop events */ + spin_lock(&dentry_list_lock); + ocfs2_set_osb_flag(osb, OCFS2_OSB_DROP_DENTRY_LOCK_IMMED); + spin_unlock(&dentry_list_lock); + /* Wait for work to finish and/or remove it */ + cancel_work_sync(&osb->dentry_lock_work); + + kill_block_super(sb); +} + static struct file_system_type ocfs2_fs_type = { .owner = THIS_MODULE, .name = "ocfs2", .get_sb = ocfs2_get_sb, /* is this called when we mount * the fs? */ - .kill_sb = kill_block_super, /* set to the generic one - * right now, but do we - * need to change that? */ + .kill_sb = ocfs2_kill_sb, + .fs_flags = FS_REQUIRES_DEV|FS_RENAME_DOES_D_MOVE, .next = NULL }; @@ -1819,6 +1833,12 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err) debugfs_remove(osb->osb_ctxt); + /* + * Flush inode dropping work queue so that deletes are + * performed while the filesystem is still working + */ + ocfs2_drop_all_dl_inodes(osb); + /* Orphan scan should be stopped as early as possible */ ocfs2_orphan_scan_stop(osb); @@ -1981,6 +2001,8 @@ static int ocfs2_initialize_super(struct super_block *sb, snprintf(osb->dev_str, sizeof(osb->dev_str), "%u,%u", MAJOR(osb->sb->s_dev), MINOR(osb->sb->s_dev)); + ocfs2_orphan_scan_init(osb); + status = ocfs2_recovery_init(osb); if (status) { mlog(ML_ERROR, "Unable to initialize recovery state\n"); diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c index ba320e25074..d1a27cda984 100644 --- a/fs/ocfs2/xattr.c +++ b/fs/ocfs2/xattr.c @@ -1052,7 +1052,8 @@ static int ocfs2_xattr_block_get(struct inode *inode, struct ocfs2_xattr_block *xb; struct ocfs2_xattr_value_root *xv; size_t size; - int ret = -ENODATA, name_offset, name_len, block_off, i; + int ret = -ENODATA, name_offset, name_len, i; + int uninitialized_var(block_off); xs->bucket = ocfs2_xattr_bucket_new(inode); if (!xs->bucket) { diff --git a/fs/proc/base.c b/fs/proc/base.c index 3ce5ae9e3d2..175db258942 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -234,23 +234,20 @@ static int check_mem_permission(struct task_struct *task) struct mm_struct *mm_for_maps(struct task_struct *task) { - struct mm_struct *mm = get_task_mm(task); - if (!mm) + struct mm_struct *mm; + + if (mutex_lock_killable(&task->cred_guard_mutex)) return NULL; - down_read(&mm->mmap_sem); - task_lock(task); - if (task->mm != mm) - goto out; - if (task->mm != current->mm && - __ptrace_may_access(task, PTRACE_MODE_READ) < 0) - goto out; - task_unlock(task); + + mm = get_task_mm(task); + if (mm && mm != current->mm && + !ptrace_may_access(task, PTRACE_MODE_READ)) { + mmput(mm); + mm = NULL; + } + mutex_unlock(&task->cred_guard_mutex); + return mm; -out: - task_unlock(task); - up_read(&mm->mmap_sem); - mmput(mm); - return NULL; } static int proc_pid_cmdline(struct task_struct *task, char * buffer) diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 6f61b7cc32e..9bd8be1d235 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -119,6 +119,7 @@ static void *m_start(struct seq_file *m, loff_t *pos) mm = mm_for_maps(priv->task); if (!mm) return NULL; + down_read(&mm->mmap_sem); tail_vma = get_gate_vma(priv->task); priv->tail_vma = tail_vma; diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c index 64a72e2e765..8f5c05d3dbd 100644 --- a/fs/proc/task_nommu.c +++ b/fs/proc/task_nommu.c @@ -189,6 +189,7 @@ static void *m_start(struct seq_file *m, loff_t *pos) priv->task = NULL; return NULL; } + down_read(&mm->mmap_sem); /* start from the Nth VMA */ for (p = rb_first(&mm->mm_rb); p; p = rb_next(p)) diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 0c93c7ef3d1..965df1227d6 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -770,7 +770,7 @@ xfs_buf_associate_memory( bp->b_pages = NULL; bp->b_addr = mem; - rval = _xfs_buf_get_pages(bp, page_count, 0); + rval = _xfs_buf_get_pages(bp, page_count, XBF_DONT_BLOCK); if (rval) return rval; diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index db15feb906f..4ece1906bd4 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -2010,7 +2010,9 @@ xfs_attr_rmtval_get(xfs_da_args_t *args) dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock); blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount); error = xfs_read_buf(mp, mp->m_ddev_targp, dblkno, - blkcnt, XFS_BUF_LOCK, &bp); + blkcnt, + XFS_BUF_LOCK | XBF_DONT_BLOCK, + &bp); if (error) return(error); @@ -2141,8 +2143,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args) dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock), blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount); - bp = xfs_buf_get_flags(mp->m_ddev_targp, dblkno, - blkcnt, XFS_BUF_LOCK); + bp = xfs_buf_get_flags(mp->m_ddev_targp, dblkno, blkcnt, + XFS_BUF_LOCK | XBF_DONT_BLOCK); ASSERT(bp); ASSERT(!XFS_BUF_GETERROR(bp)); diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 7928b9983c1..8ee5b5a76a2 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -6009,7 +6009,7 @@ xfs_getbmap( */ error = ENOMEM; subnex = 16; - map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL); + map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL | KM_NOFS); if (!map) goto out_unlock_ilock; diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c index e9df9957482..26717388acf 100644 --- a/fs/xfs/xfs_btree.c +++ b/fs/xfs/xfs_btree.c @@ -120,8 +120,8 @@ xfs_btree_check_sblock( XFS_RANDOM_BTREE_CHECK_SBLOCK))) { if (bp) xfs_buftrace("SBTREE ERROR", bp); - XFS_ERROR_REPORT("xfs_btree_check_sblock", XFS_ERRLEVEL_LOW, - cur->bc_mp); + XFS_CORRUPTION_ERROR("xfs_btree_check_sblock", + XFS_ERRLEVEL_LOW, cur->bc_mp, block); return XFS_ERROR(EFSCORRUPTED); } return 0; diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index 9ff6e57a507..2847bbc1c53 100644 --- a/fs/xfs/xfs_da_btree.c +++ b/fs/xfs/xfs_da_btree.c @@ -2201,7 +2201,7 @@ kmem_zone_t *xfs_dabuf_zone; /* dabuf zone */ xfs_da_state_t * xfs_da_state_alloc(void) { - return kmem_zone_zalloc(xfs_da_state_zone, KM_SLEEP); + return kmem_zone_zalloc(xfs_da_state_zone, KM_NOFS); } /* @@ -2261,9 +2261,9 @@ xfs_da_buf_make(int nbuf, xfs_buf_t **bps, inst_t *ra) int off; if (nbuf == 1) - dabuf = kmem_zone_alloc(xfs_dabuf_zone, KM_SLEEP); + dabuf = kmem_zone_alloc(xfs_dabuf_zone, KM_NOFS); else - dabuf = kmem_alloc(XFS_DA_BUF_SIZE(nbuf), KM_SLEEP); + dabuf = kmem_alloc(XFS_DA_BUF_SIZE(nbuf), KM_NOFS); dabuf->dirty = 0; #ifdef XFS_DABUF_DEBUG dabuf->ra = ra; diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index c657bec6d95..bb1d58eb398 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -256,7 +256,7 @@ xfs_dir_cilookup_result( !(args->op_flags & XFS_DA_OP_CILOOKUP)) return EEXIST; - args->value = kmem_alloc(len, KM_MAYFAIL); + args->value = kmem_alloc(len, KM_NOFS | KM_MAYFAIL); if (!args->value) return ENOMEM; diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index cbd451bb484..2d0b3e1da9e 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -167,17 +167,25 @@ xfs_growfs_data_private( new = nb - mp->m_sb.sb_dblocks; oagcount = mp->m_sb.sb_agcount; if (nagcount > oagcount) { + void *new_perag, *old_perag; + xfs_filestream_flush(mp); + + new_perag = kmem_zalloc(sizeof(xfs_perag_t) * nagcount, + KM_MAYFAIL); + if (!new_perag) + return XFS_ERROR(ENOMEM); + down_write(&mp->m_peraglock); - mp->m_perag = kmem_realloc(mp->m_perag, - sizeof(xfs_perag_t) * nagcount, - sizeof(xfs_perag_t) * oagcount, - KM_SLEEP); - memset(&mp->m_perag[oagcount], 0, - (nagcount - oagcount) * sizeof(xfs_perag_t)); + memcpy(new_perag, mp->m_perag, sizeof(xfs_perag_t) * oagcount); + old_perag = mp->m_perag; + mp->m_perag = new_perag; + mp->m_flags |= XFS_MOUNT_32BITINODES; nagimax = xfs_initialize_perag(mp, nagcount); up_write(&mp->m_peraglock); + + kmem_free(old_perag); } tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFS); tp->t_flags |= XFS_TRANS_RESERVE; diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 1f22d65fed0..da428b3fe0f 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -343,6 +343,16 @@ xfs_iformat( return XFS_ERROR(EFSCORRUPTED); } + if (unlikely((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) && + !ip->i_mount->m_rtdev_targp)) { + xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount, + "corrupt dinode %Lu, has realtime flag set.", + ip->i_ino); + XFS_CORRUPTION_ERROR("xfs_iformat(realtime)", + XFS_ERRLEVEL_LOW, ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + switch (ip->i_d.di_mode & S_IFMT) { case S_IFIFO: case S_IFCHR: diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 3750f04ede0..9dbdff3ea48 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -3180,7 +3180,7 @@ try_again: STATIC void xlog_state_want_sync(xlog_t *log, xlog_in_core_t *iclog) { - ASSERT(spin_is_locked(&log->l_icloglock)); + assert_spin_locked(&log->l_icloglock); if (iclog->ic_state == XLOG_STATE_ACTIVE) { xlog_state_switch_iclogs(log, iclog, 0); diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index c4eca5ed5da..492d75bae2b 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -538,7 +538,9 @@ xfs_readlink_bmap( d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock); byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount); - bp = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt), 0); + bp = xfs_buf_read_flags(mp->m_ddev_targp, d, BTOBB(byte_cnt), + XBF_LOCK | XBF_MAPPED | + XBF_DONT_BLOCK); error = XFS_BUF_GETERROR(bp); if (error) { xfs_ioerror_alert("xfs_readlink", diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index fdffb413b19..f6b90240dd4 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -473,7 +473,6 @@ extern int nfs_writepages(struct address_space *, struct writeback_control *); extern int nfs_flush_incompatible(struct file *file, struct page *page); extern int nfs_updatepage(struct file *, struct page *, unsigned int, unsigned int); extern int nfs_writeback_done(struct rpc_task *, struct nfs_write_data *); -extern void nfs_writedata_release(void *); /* * Try to write back everything synchronously (but check the @@ -488,7 +487,6 @@ extern int nfs_wb_page_cancel(struct inode *inode, struct page* page); extern int nfs_commit_inode(struct inode *, int); extern struct nfs_write_data *nfs_commitdata_alloc(void); extern void nfs_commit_free(struct nfs_write_data *wdata); -extern void nfs_commitdata_release(void *wdata); #else static inline int nfs_commit_inode(struct inode *inode, int how) @@ -507,6 +505,7 @@ nfs_have_writebacks(struct inode *inode) * Allocate nfs_write_data structures */ extern struct nfs_write_data *nfs_writedata_alloc(unsigned int npages); +extern void nfs_writedata_free(struct nfs_write_data *); /* * linux/fs/nfs/read.c @@ -515,7 +514,6 @@ extern int nfs_readpage(struct file *, struct page *); extern int nfs_readpages(struct file *, struct address_space *, struct list_head *, unsigned); extern int nfs_readpage_result(struct rpc_task *, struct nfs_read_data *); -extern void nfs_readdata_release(void *data); extern int nfs_readpage_async(struct nfs_open_context *, struct inode *, struct page *); @@ -523,6 +521,7 @@ extern int nfs_readpage_async(struct nfs_open_context *, struct inode *, * Allocate nfs_read_data structures */ extern struct nfs_read_data *nfs_readdata_alloc(unsigned int npages); +extern void nfs_readdata_free(struct nfs_read_data *); /* * linux/fs/nfs3proc.c diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h index a67dd5c5b6d..a9d823a93fe 100644 --- a/include/linux/perf_counter.h +++ b/include/linux/perf_counter.h @@ -121,7 +121,7 @@ enum perf_counter_sample_format { PERF_SAMPLE_CPU = 1U << 7, PERF_SAMPLE_PERIOD = 1U << 8, PERF_SAMPLE_STREAM_ID = 1U << 9, - PERF_SAMPLE_TP_RECORD = 1U << 10, + PERF_SAMPLE_RAW = 1U << 10, PERF_SAMPLE_MAX = 1U << 11, /* non-ABI */ }; @@ -369,6 +369,8 @@ enum perf_event_type { * * { u64 nr, * u64 ips[nr]; } && PERF_SAMPLE_CALLCHAIN + * { u32 size; + * char data[size];}&& PERF_SAMPLE_RAW * }; */ PERF_EVENT_SAMPLE = 9, @@ -414,9 +416,9 @@ struct perf_callchain_entry { __u64 ip[PERF_MAX_STACK_DEPTH]; }; -struct perf_tracepoint_record { - int size; - char *record; +struct perf_raw_record { + u32 size; + void *data; }; struct task_struct; @@ -687,7 +689,7 @@ struct perf_sample_data { struct pt_regs *regs; u64 addr; u64 period; - void *private; + struct perf_raw_record *raw; }; extern int perf_counter_overflow(struct perf_counter *counter, int nmi, diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index 7fb16d90e7b..f64fbaae781 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -637,12 +637,20 @@ __attribute__((section("_ftrace_events"))) event_##call = { \ * pc = preempt_count(); * * __data_size = ftrace_get_offsets_<call>(&__data_offsets, args); - * __entry_size = __data_size + sizeof(*entry); + * + * // Below we want to get the aligned size by taking into account + * // the u32 field that will later store the buffer size + * __entry_size = ALIGN(__data_size + sizeof(*entry) + sizeof(u32), + * sizeof(u64)); + * __entry_size -= sizeof(u32); * * do { * char raw_data[__entry_size]; <- allocate our sample in the stack * struct trace_entry *ent; * + * zero dead bytes from alignment to avoid stack leak to userspace: + * + * *(u64 *)(&raw_data[__entry_size - sizeof(u64)]) = 0ULL; * entry = (struct ftrace_raw_<call> *)raw_data; * ent = &entry->ent; * tracing_generic_entry_update(ent, irq_flags, pc); @@ -685,12 +693,15 @@ static void ftrace_profile_##call(proto) \ pc = preempt_count(); \ \ __data_size = ftrace_get_offsets_##call(&__data_offsets, args); \ - __entry_size = ALIGN(__data_size + sizeof(*entry), sizeof(u64));\ + __entry_size = ALIGN(__data_size + sizeof(*entry) + sizeof(u32),\ + sizeof(u64)); \ + __entry_size -= sizeof(u32); \ \ do { \ char raw_data[__entry_size]; \ struct trace_entry *ent; \ \ + *(u64 *)(&raw_data[__entry_size - sizeof(u64)]) = 0ULL; \ entry = (struct ftrace_raw_##call *)raw_data; \ ent = &entry->ent; \ tracing_generic_entry_update(ent, irq_flags, pc); \ diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c index 2f69bee57bf..3fd30197da2 100644 --- a/kernel/irq/numa_migrate.c +++ b/kernel/irq/numa_migrate.c @@ -107,8 +107,8 @@ out_unlock: struct irq_desc *move_irq_desc(struct irq_desc *desc, int node) { - /* those all static, do move them */ - if (desc->irq < NR_IRQS_LEGACY) + /* those static or target node is -1, do not move them */ + if (desc->irq < NR_IRQS_LEGACY || node == -1) return desc; if (desc->node != node) diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c index 868102172aa..b0b20a07f39 100644 --- a/kernel/perf_counter.c +++ b/kernel/perf_counter.c @@ -2646,7 +2646,6 @@ static void perf_counter_output(struct perf_counter *counter, int nmi, u64 counter; } group_entry; struct perf_callchain_entry *callchain = NULL; - struct perf_tracepoint_record *tp; int callchain_size = 0; u64 time; struct { @@ -2715,9 +2714,16 @@ static void perf_counter_output(struct perf_counter *counter, int nmi, header.size += sizeof(u64); } - if (sample_type & PERF_SAMPLE_TP_RECORD) { - tp = data->private; - header.size += tp->size; + if (sample_type & PERF_SAMPLE_RAW) { + int size = sizeof(u32); + + if (data->raw) + size += data->raw->size; + else + size += sizeof(u32); + + WARN_ON_ONCE(size & (sizeof(u64)-1)); + header.size += size; } ret = perf_output_begin(&handle, counter, header.size, nmi, 1); @@ -2783,8 +2789,21 @@ static void perf_counter_output(struct perf_counter *counter, int nmi, } } - if (sample_type & PERF_SAMPLE_TP_RECORD) - perf_output_copy(&handle, tp->record, tp->size); + if (sample_type & PERF_SAMPLE_RAW) { + if (data->raw) { + perf_output_put(&handle, data->raw->size); + perf_output_copy(&handle, data->raw->data, data->raw->size); + } else { + struct { + u32 size; + u32 data; + } raw = { + .size = sizeof(u32), + .data = 0, + }; + perf_output_put(&handle, raw); + } + } perf_output_end(&handle); } @@ -2849,7 +2868,8 @@ perf_counter_read_event(struct perf_counter *counter, */ struct perf_task_event { - struct task_struct *task; + struct task_struct *task; + struct perf_counter_context *task_ctx; struct { struct perf_event_header header; @@ -2909,24 +2929,23 @@ static void perf_counter_task_ctx(struct perf_counter_context *ctx, static void perf_counter_task_event(struct perf_task_event *task_event) { struct perf_cpu_context *cpuctx; - struct perf_counter_context *ctx; + struct perf_counter_context *ctx = task_event->task_ctx; cpuctx = &get_cpu_var(perf_cpu_context); perf_counter_task_ctx(&cpuctx->ctx, task_event); put_cpu_var(perf_cpu_context); rcu_read_lock(); - /* - * doesn't really matter which of the child contexts the - * events ends up in. - */ - ctx = rcu_dereference(current->perf_counter_ctxp); + if (!ctx) + ctx = rcu_dereference(task_event->task->perf_counter_ctxp); if (ctx) perf_counter_task_ctx(ctx, task_event); rcu_read_unlock(); } -static void perf_counter_task(struct task_struct *task, int new) +static void perf_counter_task(struct task_struct *task, + struct perf_counter_context *task_ctx, + int new) { struct perf_task_event task_event; @@ -2936,8 +2955,9 @@ static void perf_counter_task(struct task_struct *task, int new) return; task_event = (struct perf_task_event){ - .task = task, - .event = { + .task = task, + .task_ctx = task_ctx, + .event = { .header = { .type = new ? PERF_EVENT_FORK : PERF_EVENT_EXIT, .misc = 0, @@ -2955,7 +2975,7 @@ static void perf_counter_task(struct task_struct *task, int new) void perf_counter_fork(struct task_struct *task) { - perf_counter_task(task, 1); + perf_counter_task(task, NULL, 1); } /* @@ -3344,87 +3364,81 @@ int perf_counter_overflow(struct perf_counter *counter, int nmi, * Generic software counter infrastructure */ -static void perf_swcounter_update(struct perf_counter *counter) +/* + * We directly increment counter->count and keep a second value in + * counter->hw.period_left to count intervals. This period counter + * is kept in the range [-sample_period, 0] so that we can use the + * sign as trigger. + */ + +static u64 perf_swcounter_set_period(struct perf_counter *counter) { struct hw_perf_counter *hwc = &counter->hw; - u64 prev, now; - s64 delta; + u64 period = hwc->last_period; + u64 nr, offset; + s64 old, val; + + hwc->last_period = hwc->sample_period; again: - prev = atomic64_read(&hwc->prev_count); - now = atomic64_read(&hwc->count); - if (atomic64_cmpxchg(&hwc->prev_count, prev, now) != prev) - goto again; + old = val = atomic64_read(&hwc->period_left); + if (val < 0) + return 0; - delta = now - prev; + nr = div64_u64(period + val, period); + offset = nr * period; + val -= offset; + if (atomic64_cmpxchg(&hwc->period_left, old, val) != old) + goto again; - atomic64_add(delta, &counter->count); - atomic64_sub(delta, &hwc->period_left); + return nr; } -static void perf_swcounter_set_period(struct perf_counter *counter) +static void perf_swcounter_overflow(struct perf_counter *counter, + int nmi, struct perf_sample_data *data) { struct hw_perf_counter *hwc = &counter->hw; - s64 left = atomic64_read(&hwc->period_left); - s64 period = hwc->sample_period; + u64 overflow; - if (unlikely(left <= -period)) { - left = period; - atomic64_set(&hwc->period_left, left); - hwc->last_period = period; - } + data->period = counter->hw.last_period; + overflow = perf_swcounter_set_period(counter); - if (unlikely(left <= 0)) { - left += period; - atomic64_add(period, &hwc->period_left); - hwc->last_period = period; - } + if (hwc->interrupts == MAX_INTERRUPTS) + return; - atomic64_set(&hwc->prev_count, -left); - atomic64_set(&hwc->count, -left); + for (; overflow; overflow--) { + if (perf_counter_overflow(counter, nmi, data)) { + /* + * We inhibit the overflow from happening when + * hwc->interrupts == MAX_INTERRUPTS. + */ + break; + } + } } -static enum hrtimer_restart perf_swcounter_hrtimer(struct hrtimer *hrtimer) +static void perf_swcounter_unthrottle(struct perf_counter *counter) { - enum hrtimer_restart ret = HRTIMER_RESTART; - struct perf_sample_data data; - struct perf_counter *counter; - u64 period; - - counter = container_of(hrtimer, struct perf_counter, hw.hrtimer); - counter->pmu->read(counter); - - data.addr = 0; - data.regs = get_irq_regs(); /* - * In case we exclude kernel IPs or are somehow not in interrupt - * context, provide the next best thing, the user IP. + * Nothing to do, we already reset hwc->interrupts. */ - if ((counter->attr.exclude_kernel || !data.regs) && - !counter->attr.exclude_user) - data.regs = task_pt_regs(current); +} - if (data.regs) { - if (perf_counter_overflow(counter, 0, &data)) - ret = HRTIMER_NORESTART; - } +static void perf_swcounter_add(struct perf_counter *counter, u64 nr, + int nmi, struct perf_sample_data *data) +{ + struct hw_perf_counter *hwc = &counter->hw; - period = max_t(u64, 10000, counter->hw.sample_period); - hrtimer_forward_now(hrtimer, ns_to_ktime(period)); + atomic64_add(nr, &counter->count); - return ret; -} + if (!hwc->sample_period) + return; -static void perf_swcounter_overflow(struct perf_counter *counter, - int nmi, struct perf_sample_data *data) -{ - data->period = counter->hw.last_period; + if (!data->regs) + return; - perf_swcounter_update(counter); - perf_swcounter_set_period(counter); - if (perf_counter_overflow(counter, nmi, data)) - /* soft-disable the counter */ - ; + if (!atomic64_add_negative(nr, &hwc->period_left)) + perf_swcounter_overflow(counter, nmi, data); } static int perf_swcounter_is_counting(struct perf_counter *counter) @@ -3488,15 +3502,6 @@ static int perf_swcounter_match(struct perf_counter *counter, return 1; } -static void perf_swcounter_add(struct perf_counter *counter, u64 nr, - int nmi, struct perf_sample_data *data) -{ - int neg = atomic64_add_negative(nr, &counter->hw.count); - - if (counter->hw.sample_period && !neg && data->regs) - perf_swcounter_overflow(counter, nmi, data); -} - static void perf_swcounter_ctx_event(struct perf_counter_context *ctx, enum perf_type_id type, u32 event, u64 nr, int nmi, @@ -3575,27 +3580,66 @@ void __perf_swcounter_event(u32 event, u64 nr, int nmi, static void perf_swcounter_read(struct perf_counter *counter) { - perf_swcounter_update(counter); } static int perf_swcounter_enable(struct perf_counter *counter) { - perf_swcounter_set_period(counter); + struct hw_perf_counter *hwc = &counter->hw; + + if (hwc->sample_period) { + hwc->last_period = hwc->sample_period; + perf_swcounter_set_period(counter); + } return 0; } static void perf_swcounter_disable(struct perf_counter *counter) { - perf_swcounter_update(counter); } static const struct pmu perf_ops_generic = { .enable = perf_swcounter_enable, .disable = perf_swcounter_disable, .read = perf_swcounter_read, + .unthrottle = perf_swcounter_unthrottle, }; /* + * hrtimer based swcounter callback + */ + +static enum hrtimer_restart perf_swcounter_hrtimer(struct hrtimer *hrtimer) +{ + enum hrtimer_restart ret = HRTIMER_RESTART; + struct perf_sample_data data; + struct perf_counter *counter; + u64 period; + + counter = container_of(hrtimer, struct perf_counter, hw.hrtimer); + counter->pmu->read(counter); + + data.addr = 0; + data.regs = get_irq_regs(); + /* + * In case we exclude kernel IPs or are somehow not in interrupt + * context, provide the next best thing, the user IP. + */ + if ((counter->attr.exclude_kernel || !data.regs) && + !counter->attr.exclude_user) + data.regs = task_pt_regs(current); + + if (data.regs) { + if (perf_counter_overflow(counter, 0, &data)) + ret = HRTIMER_NORESTART; + } + + period = max_t(u64, 10000, counter->hw.sample_period); + hrtimer_forward_now(hrtimer, ns_to_ktime(period)); + + return ret; +} + +/* * Software counter: cpu wall time clock */ @@ -3715,15 +3759,15 @@ static const struct pmu perf_ops_task_clock = { void perf_tpcounter_event(int event_id, u64 addr, u64 count, void *record, int entry_size) { - struct perf_tracepoint_record tp = { + struct perf_raw_record raw = { .size = entry_size, - .record = record, + .data = record, }; struct perf_sample_data data = { .regs = get_irq_regs(), .addr = addr, - .private = &tp, + .raw = &raw, }; if (!data.regs) @@ -3743,6 +3787,14 @@ static void tp_perf_counter_destroy(struct perf_counter *counter) static const struct pmu *tp_perf_counter_init(struct perf_counter *counter) { + /* + * Raw tracepoint data is a severe data leak, only allow root to + * have these. + */ + if ((counter->attr.sample_type & PERF_SAMPLE_RAW) && + !capable(CAP_SYS_ADMIN)) + return ERR_PTR(-EPERM); + if (ftrace_profile_enable(counter->attr.config)) return NULL; @@ -4285,7 +4337,7 @@ void perf_counter_exit_task(struct task_struct *child) unsigned long flags; if (likely(!child->perf_counter_ctxp)) { - perf_counter_task(child, 0); + perf_counter_task(child, NULL, 0); return; } @@ -4305,6 +4357,7 @@ void perf_counter_exit_task(struct task_struct *child) * incremented the context's refcount before we do put_ctx below. */ spin_lock(&child_ctx->lock); + child->perf_counter_ctxp = NULL; /* * If this context is a clone; unclone it so it can't get * swapped to another process while we're removing all @@ -4318,9 +4371,7 @@ void perf_counter_exit_task(struct task_struct *child) * won't get any samples after PERF_EVENT_EXIT. We can however still * get a few PERF_EVENT_READ events. */ - perf_counter_task(child, 0); - - child->perf_counter_ctxp = NULL; + perf_counter_task(child, child_ctx, 0); /* * We can recurse on the same lock type through: diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 1090b0aed9b..7a34cb563fe 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -267,8 +267,8 @@ static void blk_trace_free(struct blk_trace *bt) { debugfs_remove(bt->msg_file); debugfs_remove(bt->dropped_file); - debugfs_remove(bt->dir); relay_close(bt->rchan); + debugfs_remove(bt->dir); free_percpu(bt->sequence); free_percpu(bt->msg_data); kfree(bt); @@ -378,18 +378,8 @@ static int blk_subbuf_start_callback(struct rchan_buf *buf, void *subbuf, static int blk_remove_buf_file_callback(struct dentry *dentry) { - struct dentry *parent = dentry->d_parent; debugfs_remove(dentry); - /* - * this will fail for all but the last file, but that is ok. what we - * care about is the top level buts->name directory going away, when - * the last trace file is gone. Then we don't have to rmdir() that - * manually on trace stop, so it nicely solves the issue with - * force killing of running traces. - */ - - debugfs_remove(parent); return 0; } diff --git a/mm/mempool.c b/mm/mempool.c index a46eb1b4bb6..32e75d40050 100644 --- a/mm/mempool.c +++ b/mm/mempool.c @@ -303,14 +303,14 @@ EXPORT_SYMBOL(mempool_free_slab); */ void *mempool_kmalloc(gfp_t gfp_mask, void *pool_data) { - size_t size = (size_t)(long)pool_data; + size_t size = (size_t)pool_data; return kmalloc(size, gfp_mask); } EXPORT_SYMBOL(mempool_kmalloc); void *mempool_kzalloc(gfp_t gfp_mask, void *pool_data) { - size_t size = (size_t) pool_data; + size_t size = (size_t)pool_data; return kzalloc(size, gfp_mask); } EXPORT_SYMBOL(mempool_kzalloc); diff --git a/net/socket.c b/net/socket.c index 791d71a36a9..6d471655904 100644 --- a/net/socket.c +++ b/net/socket.c @@ -736,7 +736,7 @@ static ssize_t sock_sendpage(struct file *file, struct page *page, if (more) flags |= MSG_MORE; - return sock->ops->sendpage(sock, page, offset, size, flags); + return kernel_sendpage(sock, page, offset, size, flags); } static ssize_t sock_splice_read(struct file *file, loff_t *ppos, diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 15c2a08a66f..1e8cfc4c2ed 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -1285,6 +1285,8 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent rc = inode->i_op->getxattr(dentry, XATTR_NAME_SELINUX, context, len); if (rc == -ERANGE) { + kfree(context); + /* Need a larger buffer. Query for the right size. */ rc = inode->i_op->getxattr(dentry, XATTR_NAME_SELINUX, NULL, 0); @@ -1292,7 +1294,6 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent dput(dentry); goto out_unlock; } - kfree(context); len = rc; context = kmalloc(len+1, GFP_NOFS); if (!context) { diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 51c44fdbc0f..fea976793ae 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -13563,6 +13563,8 @@ static int patch_alc269(struct hda_codec *codec) set_capture_mixer(spec); set_beep_amp(spec, 0x0b, 0x04, HDA_INPUT); + spec->vmaster_nid = 0x02; + codec->patch_ops = alc_patch_ops; if (board_config == ALC269_AUTO) spec->init_hook = alc269_auto_init; @@ -15577,9 +15579,12 @@ static int patch_alc861vd(struct hda_codec *codec) spec->stream_digital_playback = &alc861vd_pcm_digital_playback; spec->stream_digital_capture = &alc861vd_pcm_digital_capture; - spec->adc_nids = alc861vd_adc_nids; - spec->num_adc_nids = ARRAY_SIZE(alc861vd_adc_nids); - spec->capsrc_nids = alc861vd_capsrc_nids; + if (!spec->adc_nids) { + spec->adc_nids = alc861vd_adc_nids; + spec->num_adc_nids = ARRAY_SIZE(alc861vd_adc_nids); + } + if (!spec->capsrc_nids) + spec->capsrc_nids = alc861vd_capsrc_nids; set_capture_mixer(spec); set_beep_amp(spec, 0x0b, 0x05, HDA_INPUT); @@ -17496,9 +17501,12 @@ static int patch_alc662(struct hda_codec *codec) spec->stream_digital_playback = &alc662_pcm_digital_playback; spec->stream_digital_capture = &alc662_pcm_digital_capture; - spec->adc_nids = alc662_adc_nids; - spec->num_adc_nids = ARRAY_SIZE(alc662_adc_nids); - spec->capsrc_nids = alc662_capsrc_nids; + if (!spec->adc_nids) { + spec->adc_nids = alc662_adc_nids; + spec->num_adc_nids = ARRAY_SIZE(alc662_adc_nids); + } + if (!spec->capsrc_nids) + spec->capsrc_nids = alc662_capsrc_nids; if (!spec->cap_mixer) set_capture_mixer(spec); diff --git a/sound/soc/fsl/efika-audio-fabric.c b/sound/soc/fsl/efika-audio-fabric.c index 85b0e756950..3326e2a1e86 100644 --- a/sound/soc/fsl/efika-audio-fabric.c +++ b/sound/soc/fsl/efika-audio-fabric.c @@ -30,6 +30,8 @@ #include "mpc5200_psc_ac97.h" #include "../codecs/stac9766.h" +#define DRV_NAME "efika-audio-fabric" + static struct snd_soc_device device; static struct snd_soc_card card; diff --git a/sound/soc/fsl/pcm030-audio-fabric.c b/sound/soc/fsl/pcm030-audio-fabric.c index 8766f7a3893..b928ef7d28e 100644 --- a/sound/soc/fsl/pcm030-audio-fabric.c +++ b/sound/soc/fsl/pcm030-audio-fabric.c @@ -30,6 +30,8 @@ #include "mpc5200_psc_ac97.h" #include "../codecs/wm9712.h" +#define DRV_NAME "pcm030-audio-fabric" + static struct snd_soc_device device; static struct snd_soc_card card; diff --git a/tools/perf/Documentation/perf-examples.txt b/tools/perf/Documentation/perf-examples.txt new file mode 100644 index 00000000000..8eb6c489fb1 --- /dev/null +++ b/tools/perf/Documentation/perf-examples.txt @@ -0,0 +1,225 @@ + + ------------------------------ + ****** perf by examples ****** + ------------------------------ + +[ From an e-mail by Ingo Molnar, http://lkml.org/lkml/2009/8/4/346 ] + + +First, discovery/enumeration of available counters can be done via +'perf list': + +titan:~> perf list + [...] + kmem:kmalloc [Tracepoint event] + kmem:kmem_cache_alloc [Tracepoint event] + kmem:kmalloc_node [Tracepoint event] + kmem:kmem_cache_alloc_node [Tracepoint event] + kmem:kfree [Tracepoint event] + kmem:kmem_cache_free [Tracepoint event] + kmem:mm_page_free_direct [Tracepoint event] + kmem:mm_pagevec_free [Tracepoint event] + kmem:mm_page_alloc [Tracepoint event] + kmem:mm_page_alloc_zone_locked [Tracepoint event] + kmem:mm_page_pcpu_drain [Tracepoint event] + kmem:mm_page_alloc_extfrag [Tracepoint event] + +Then any (or all) of the above event sources can be activated and +measured. For example the page alloc/free properties of a 'hackbench +run' are: + + titan:~> perf stat -e kmem:mm_page_pcpu_drain -e kmem:mm_page_alloc + -e kmem:mm_pagevec_free -e kmem:mm_page_free_direct ./hackbench 10 + Time: 0.575 + + Performance counter stats for './hackbench 10': + + 13857 kmem:mm_page_pcpu_drain + 27576 kmem:mm_page_alloc + 6025 kmem:mm_pagevec_free + 20934 kmem:mm_page_free_direct + + 0.613972165 seconds time elapsed + +You can observe the statistical properties as well, by using the +'repeat the workload N times' feature of perf stat: + + titan:~> perf stat --repeat 5 -e kmem:mm_page_pcpu_drain -e + kmem:mm_page_alloc -e kmem:mm_pagevec_free -e + kmem:mm_page_free_direct ./hackbench 10 + Time: 0.627 + Time: 0.644 + Time: 0.564 + Time: 0.559 + Time: 0.626 + + Performance counter stats for './hackbench 10' (5 runs): + + 12920 kmem:mm_page_pcpu_drain ( +- 3.359% ) + 25035 kmem:mm_page_alloc ( +- 3.783% ) + 6104 kmem:mm_pagevec_free ( +- 0.934% ) + 18376 kmem:mm_page_free_direct ( +- 4.941% ) + + 0.643954516 seconds time elapsed ( +- 2.363% ) + +Furthermore, these tracepoints can be used to sample the workload as +well. For example the page allocations done by a 'git gc' can be +captured the following way: + + titan:~/git> perf record -f -e kmem:mm_page_alloc -c 1 ./git gc + Counting objects: 1148, done. + Delta compression using up to 2 threads. + Compressing objects: 100% (450/450), done. + Writing objects: 100% (1148/1148), done. + Total 1148 (delta 690), reused 1148 (delta 690) + [ perf record: Captured and wrote 0.267 MB perf.data (~11679 samples) ] + +To check which functions generated page allocations: + + titan:~/git> perf report + # Samples: 10646 + # + # Overhead Command Shared Object + # ........ ............... .......................... + # + 23.57% git-repack /lib64/libc-2.5.so + 21.81% git /lib64/libc-2.5.so + 14.59% git ./git + 11.79% git-repack ./git + 7.12% git /lib64/ld-2.5.so + 3.16% git-repack /lib64/libpthread-2.5.so + 2.09% git-repack /bin/bash + 1.97% rm /lib64/libc-2.5.so + 1.39% mv /lib64/ld-2.5.so + 1.37% mv /lib64/libc-2.5.so + 1.12% git-repack /lib64/ld-2.5.so + 0.95% rm /lib64/ld-2.5.so + 0.90% git-update-serv /lib64/libc-2.5.so + 0.73% git-update-serv /lib64/ld-2.5.so + 0.68% perf /lib64/libpthread-2.5.so + 0.64% git-repack /usr/lib64/libz.so.1.2.3 + +Or to see it on a more finegrained level: + +titan:~/git> perf report --sort comm,dso,symbol +# Samples: 10646 +# +# Overhead Command Shared Object Symbol +# ........ ............... .......................... ...... +# + 9.35% git-repack ./git [.] insert_obj_hash + 9.12% git ./git [.] insert_obj_hash + 7.31% git /lib64/libc-2.5.so [.] memcpy + 6.34% git-repack /lib64/libc-2.5.so [.] _int_malloc + 6.24% git-repack /lib64/libc-2.5.so [.] memcpy + 5.82% git-repack /lib64/libc-2.5.so [.] __GI___fork + 5.47% git /lib64/libc-2.5.so [.] _int_malloc + 2.99% git /lib64/libc-2.5.so [.] memset + +Furthermore, call-graph sampling can be done too, of page +allocations - to see precisely what kind of page allocations there +are: + + titan:~/git> perf record -f -g -e kmem:mm_page_alloc -c 1 ./git gc + Counting objects: 1148, done. + Delta compression using up to 2 threads. + Compressing objects: 100% (450/450), done. + Writing objects: 100% (1148/1148), done. + Total 1148 (delta 690), reused 1148 (delta 690) + [ perf record: Captured and wrote 0.963 MB perf.data (~42069 samples) ] + + titan:~/git> perf report -g + # Samples: 10686 + # + # Overhead Command Shared Object + # ........ ............... .......................... + # + 23.25% git-repack /lib64/libc-2.5.so + | + |--50.00%-- _int_free + | + |--37.50%-- __GI___fork + | make_child + | + |--12.50%-- ptmalloc_unlock_all2 + | make_child + | + --6.25%-- __GI_strcpy + 21.61% git /lib64/libc-2.5.so + | + |--30.00%-- __GI_read + | | + | --83.33%-- git_config_from_file + | git_config + | | + [...] + +Or you can observe the whole system's page allocations for 10 +seconds: + +titan:~/git> perf stat -a -e kmem:mm_page_pcpu_drain -e +kmem:mm_page_alloc -e kmem:mm_pagevec_free -e +kmem:mm_page_free_direct sleep 10 + + Performance counter stats for 'sleep 10': + + 171585 kmem:mm_page_pcpu_drain + 322114 kmem:mm_page_alloc + 73623 kmem:mm_pagevec_free + 254115 kmem:mm_page_free_direct + + 10.000591410 seconds time elapsed + +Or observe how fluctuating the page allocations are, via statistical +analysis done over ten 1-second intervals: + + titan:~/git> perf stat --repeat 10 -a -e kmem:mm_page_pcpu_drain -e + kmem:mm_page_alloc -e kmem:mm_pagevec_free -e + kmem:mm_page_free_direct sleep 1 + + Performance counter stats for 'sleep 1' (10 runs): + + 17254 kmem:mm_page_pcpu_drain ( +- 3.709% ) + 34394 kmem:mm_page_alloc ( +- 4.617% ) + 7509 kmem:mm_pagevec_free ( +- 4.820% ) + 25653 kmem:mm_page_free_direct ( +- 3.672% ) + + 1.058135029 seconds time elapsed ( +- 3.089% ) + +Or you can annotate the recorded 'git gc' run on a per symbol basis +and check which instructions/source-code generated page allocations: + + titan:~/git> perf annotate __GI___fork + ------------------------------------------------ + Percent | Source code & Disassembly of libc-2.5.so + ------------------------------------------------ + : + : + : Disassembly of section .plt: + : Disassembly of section .text: + : + : 00000031a2e95560 <__fork>: + [...] + 0.00 : 31a2e95602: b8 38 00 00 00 mov $0x38,%eax + 0.00 : 31a2e95607: 0f 05 syscall + 83.42 : 31a2e95609: 48 3d 00 f0 ff ff cmp $0xfffffffffffff000,%rax + 0.00 : 31a2e9560f: 0f 87 4d 01 00 00 ja 31a2e95762 <__fork+0x202> + 0.00 : 31a2e95615: 85 c0 test %eax,%eax + +( this shows that 83.42% of __GI___fork's page allocations come from + the 0x38 system call it performs. ) + +etc. etc. - a lot more is possible. I could list a dozen of +other different usecases straight away - neither of which is +possible via /proc/vmstat. + +/proc/vmstat is not in the same league really, in terms of +expressive power of system analysis and performance +analysis. + +All that the above results needed were those new tracepoints +in include/tracing/events/kmem.h. + + Ingo + + diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt index 0d74346d21a..484080dd5b6 100644 --- a/tools/perf/Documentation/perf-stat.txt +++ b/tools/perf/Documentation/perf-stat.txt @@ -40,7 +40,7 @@ OPTIONS -a:: system-wide collection --S:: +-c:: scale counter values EXAMPLES diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt index 539d0128972..4a7d558dc30 100644 --- a/tools/perf/Documentation/perf-top.txt +++ b/tools/perf/Documentation/perf-top.txt @@ -3,36 +3,122 @@ perf-top(1) NAME ---- -perf-top - Run a command and profile it +perf-top - System profiling tool. SYNOPSIS -------- [verse] -'perf top' [-e <EVENT> | --event=EVENT] [-l] [-a] <command> +'perf top' [-e <EVENT> | --event=EVENT] [<options>] DESCRIPTION ----------- -This command runs a command and gathers a performance counter profile -from it. +This command generates and displays a performance counter profile in realtime. OPTIONS ------- -<command>...:: - Any command you can specify in a shell. +-a:: +--all-cpus:: + System-wide collection. (default) + +-c <count>:: +--count=<count>:: + Event period to sample. + +-C <cpu>:: +--CPU=<cpu>:: + CPU to profile. + +-d <seconds>:: +--delay=<seconds>:: + Number of seconds to delay between refreshes. --e:: ---event=:: +-e <event>:: +--event=<event>:: Select the PMU event. Selection can be a symbolic event name (use 'perf list' to list all events) or a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a - hexadecimal event descriptor. + hexadecimal event descriptor. --a:: - system-wide collection +-E <entries>:: +--entries=<entries>:: + Display this many functions. + +-f <count>:: +--count-filter=<count>:: + Only display functions with more events than this. + +-F <freq>:: +--freq=<freq>:: + Profile at this frequency. + +-i:: +--inherit:: + Child tasks inherit counters, only makes sens with -p option. + +-k <path>:: +--vmlinux=<path>:: + Path to vmlinux. Required for annotation functionality. + +-m <pages>:: +--mmap-pages=<pages>:: + Number of mmapped data pages. + +-p <pid>:: +--pid=<pid>:: + Profile events on existing pid. + +-r <priority>:: +--realtime=<priority>:: + Collect data with this RT SCHED_FIFO priority. + +-s <symbol>:: +--sym-annotate=<symbol>:: + Annotate this symbol. Requires -k option. + +-v:: +--verbose:: + Be more verbose (show counter open errors, etc). + +-z:: +--zero:: + Zero history across display updates. + +INTERACTIVE PROMPTING KEYS +-------------------------- + +[d]:: + Display refresh delay. + +[e]:: + Number of entries to display. + +[E]:: + Event to display when multiple counters are active. + +[f]:: + Profile display filter (>= hit count). + +[F]:: + Annotation display filter (>= % of total). + +[s]:: + Annotate symbol. + +[S]:: + Stop annotation, return to full profile display. + +[w]:: + Toggle between weighted sum and individual count[E]r profile. + +[z]:: + Toggle event count zeroing across display updates. + +[qQ]:: + Quit. + +Pressing any unmapped key displays a menu, and prompts for input. --l:: - scale counter values SEE ALSO -------- diff --git a/tools/perf/Makefile b/tools/perf/Makefile index 1916e44b9bb..60411e94113 100644 --- a/tools/perf/Makefile +++ b/tools/perf/Makefile @@ -387,10 +387,14 @@ else has_bfd_iberty := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty > /dev/null 2>&1 && echo y") + has_bfd_iberty_z := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty -lz > /dev/null 2>&1 && echo y") + ifeq ($(has_bfd),y) EXTLIBS += -lbfd else ifeq ($(has_bfd_iberty),y) EXTLIBS += -lbfd -liberty + else ifeq ($(has_bfd_iberty_z),y) + EXTLIBS += -lbfd -liberty -lz else msg := $(warning No bfd.h/libbfd found, install binutils-dev[el] to gain symbol demangling) BASIC_CFLAGS += -DNO_DEMANGLE diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 90c98082af1..0345aad8eba 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -525,10 +525,14 @@ static int __cmd_record(int argc, const char **argv) signal(SIGCHLD, sig_handler); signal(SIGINT, sig_handler); - if (!stat(output_name, &st) && !force && !append_file) { - fprintf(stderr, "Error, output file %s exists, use -A to append or -f to overwrite.\n", - output_name); - exit(-1); + if (!stat(output_name, &st) && st.st_size) { + if (!force && !append_file) { + fprintf(stderr, "Error, output file %s exists, use -A to append or -f to overwrite.\n", + output_name); + exit(-1); + } + } else { + append_file = 0; } flags = O_CREAT|O_RDWR; diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index 8cb58d68a00..99274cec0ad 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -68,7 +68,7 @@ static int callchain; static struct callchain_param callchain_param = { - .mode = CHAIN_GRAPH_ABS, + .mode = CHAIN_GRAPH_REL, .min_percent = 0.5 }; @@ -112,7 +112,9 @@ struct read_event { struct perf_event_header header; u32 pid,tid; u64 value; - u64 format[3]; + u64 time_enabled; + u64 time_running; + u64 id; }; typedef union event_union { @@ -698,7 +700,8 @@ sort__sym_print(FILE *fp, struct hist_entry *self, unsigned int width __used) size_t ret = 0; if (verbose) - ret += repsep_fprintf(fp, "%#018llx ", (u64)self->ip); + ret += repsep_fprintf(fp, "%#018llx %c ", (u64)self->ip, + dso__symtab_origin(self->dso)); ret += repsep_fprintf(fp, "[%c] ", self->level); if (self->sym) { @@ -888,6 +891,21 @@ ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain, int depth, return ret; } +static struct symbol *rem_sq_bracket; +static struct callchain_list rem_hits; + +static void init_rem_hits(void) +{ + rem_sq_bracket = malloc(sizeof(*rem_sq_bracket) + 6); + if (!rem_sq_bracket) { + fprintf(stderr, "Not enough memory to display remaining hits\n"); + return; + } + + strcpy(rem_sq_bracket->name, "[...]"); + rem_hits.sym = rem_sq_bracket; +} + static size_t callchain__fprintf_graph(FILE *fp, struct callchain_node *self, u64 total_samples, int depth, int depth_mask) @@ -897,25 +915,34 @@ callchain__fprintf_graph(FILE *fp, struct callchain_node *self, struct callchain_list *chain; int new_depth_mask = depth_mask; u64 new_total; + u64 remaining; size_t ret = 0; int i; if (callchain_param.mode == CHAIN_GRAPH_REL) - new_total = self->cumul_hit; + new_total = self->children_hit; else new_total = total_samples; + remaining = new_total; + node = rb_first(&self->rb_root); while (node) { + u64 cumul; + child = rb_entry(node, struct callchain_node, rb_node); + cumul = cumul_hits(child); + remaining -= cumul; /* * The depth mask manages the output of pipes that show * the depth. We don't want to keep the pipes of the current - * level for the last child of this depth + * level for the last child of this depth. + * Except if we have remaining filtered hits. They will + * supersede the last child */ next = rb_next(node); - if (!next) + if (!next && (callchain_param.mode != CHAIN_GRAPH_REL || !remaining)) new_depth_mask &= ~(1 << (depth - 1)); /* @@ -930,7 +957,7 @@ callchain__fprintf_graph(FILE *fp, struct callchain_node *self, ret += ipchain__fprintf_graph(fp, chain, depth, new_depth_mask, i++, new_total, - child->cumul_hit); + cumul); } ret += callchain__fprintf_graph(fp, child, new_total, depth + 1, @@ -938,6 +965,19 @@ callchain__fprintf_graph(FILE *fp, struct callchain_node *self, node = next; } + if (callchain_param.mode == CHAIN_GRAPH_REL && + remaining && remaining != new_total) { + + if (!rem_sq_bracket) + return ret; + + new_depth_mask &= ~(1 << (depth - 1)); + + ret += ipchain__fprintf_graph(fp, &rem_hits, depth, + new_depth_mask, 0, new_total, + remaining); + } + return ret; } @@ -1358,6 +1398,8 @@ static size_t output__fprintf(FILE *fp, u64 total_samples) unsigned int width; char *col_width = col_width_list_str; + init_rem_hits(); + fprintf(fp, "# Samples: %Ld\n", (u64)total_samples); fprintf(fp, "#\n"); @@ -1429,6 +1471,8 @@ print_entries: } fprintf(fp, "\n"); + free(rem_sq_bracket); + return ret; } @@ -1690,14 +1734,37 @@ static void trace_event(event_t *event) dprintf(".\n"); } +static struct perf_header *header; + +static struct perf_counter_attr *perf_header__find_attr(u64 id) +{ + int i; + + for (i = 0; i < header->attrs; i++) { + struct perf_header_attr *attr = header->attr[i]; + int j; + + for (j = 0; j < attr->ids; j++) { + if (attr->id[j] == id) + return &attr->attr; + } + } + + return NULL; +} + static int process_read_event(event_t *event, unsigned long offset, unsigned long head) { - dprintf("%p [%p]: PERF_EVENT_READ: %d %d %Lu\n", + struct perf_counter_attr *attr = perf_header__find_attr(event->read.id); + + dprintf("%p [%p]: PERF_EVENT_READ: %d %d %s %Lu\n", (void *)(offset + head), (void *)(long)(event->header.size), event->read.pid, event->read.tid, + attr ? __event_name(attr->type, attr->config) + : "FAIL", event->read.value); return 0; @@ -1743,8 +1810,6 @@ process_event(event_t *event, unsigned long offset, unsigned long head) return 0; } -static struct perf_header *header; - static u64 perf_header__sample_type(void) { u64 sample_type = 0; @@ -1812,6 +1877,13 @@ static int __cmd_report(void) " -g?\n"); exit(-1); } + } else if (callchain_param.mode != CHAIN_NONE && !callchain) { + callchain = 1; + if (register_callchain_param(&callchain_param) < 0) { + fprintf(stderr, "Can't register callchain" + " params\n"); + exit(-1); + } } if (load_kernel() < 0) { @@ -1950,6 +2022,13 @@ parse_callchain_opt(const struct option *opt __used, const char *arg, else if (!strncmp(tok, "fractal", strlen(arg))) callchain_param.mode = CHAIN_GRAPH_REL; + else if (!strncmp(tok, "none", strlen(arg))) { + callchain_param.mode = CHAIN_NONE; + callchain = 0; + + return 0; + } + else return -1; diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index f9510eeeb6c..b4b06c7903e 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -496,7 +496,7 @@ static const struct option options[] = { "stat events on existing pid"), OPT_BOOLEAN('a', "all-cpus", &system_wide, "system-wide collection from all CPUs"), - OPT_BOOLEAN('S', "scale", &scale, + OPT_BOOLEAN('c', "scale", &scale, "scale/normalize counters"), OPT_BOOLEAN('v', "verbose", &verbose, "be more verbose (show counter open errors, etc)"), diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index f139f1ab933..7de28ce9ca2 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -31,6 +31,8 @@ #include <fcntl.h> #include <stdio.h> +#include <termios.h> +#include <unistd.h> #include <errno.h> #include <time.h> @@ -54,7 +56,7 @@ static int system_wide = 0; static int default_interval = 100000; -static u64 count_filter = 5; +static int count_filter = 5; static int print_entries = 15; static int target_pid = -1; @@ -69,15 +71,28 @@ static int freq = 0; static int verbose = 0; static char *vmlinux = NULL; -static char *sym_filter; -static unsigned long filter_start; -static unsigned long filter_end; - static int delay_secs = 2; static int zero; static int dump_symtab; /* + * Source + */ + +struct source_line { + u64 eip; + unsigned long count[MAX_COUNTERS]; + char *line; + struct source_line *next; +}; + +static char *sym_filter = NULL; +struct sym_entry *sym_filter_entry = NULL; +static int sym_pcnt_filter = 5; +static int sym_counter = 0; +static int display_weighted = -1; + +/* * Symbols */ @@ -91,9 +106,237 @@ struct sym_entry { unsigned long snap_count; double weight; int skip; + struct source_line *source; + struct source_line *lines; + struct source_line **lines_tail; + pthread_mutex_t source_lock; }; -struct sym_entry *sym_filter_entry; +/* + * Source functions + */ + +static void parse_source(struct sym_entry *syme) +{ + struct symbol *sym; + struct module *module; + struct section *section = NULL; + FILE *file; + char command[PATH_MAX*2], *path = vmlinux; + u64 start, end, len; + + if (!syme) + return; + + if (syme->lines) { + pthread_mutex_lock(&syme->source_lock); + goto out_assign; + } + + sym = (struct symbol *)(syme + 1); + module = sym->module; + + if (module) + path = module->path; + if (!path) + return; + + start = sym->obj_start; + if (!start) + start = sym->start; + + if (module) { + section = module->sections->find_section(module->sections, ".text"); + if (section) + start -= section->vma; + } + + end = start + sym->end - sym->start + 1; + len = sym->end - sym->start; + + sprintf(command, "objdump --start-address=0x%016Lx --stop-address=0x%016Lx -dS %s", start, end, path); + + file = popen(command, "r"); + if (!file) + return; + + pthread_mutex_lock(&syme->source_lock); + syme->lines_tail = &syme->lines; + while (!feof(file)) { + struct source_line *src; + size_t dummy = 0; + char *c; + + src = malloc(sizeof(struct source_line)); + assert(src != NULL); + memset(src, 0, sizeof(struct source_line)); + + if (getline(&src->line, &dummy, file) < 0) + break; + if (!src->line) + break; + + c = strchr(src->line, '\n'); + if (c) + *c = 0; + + src->next = NULL; + *syme->lines_tail = src; + syme->lines_tail = &src->next; + + if (strlen(src->line)>8 && src->line[8] == ':') { + src->eip = strtoull(src->line, NULL, 16); + if (section) + src->eip += section->vma; + } + if (strlen(src->line)>8 && src->line[16] == ':') { + src->eip = strtoull(src->line, NULL, 16); + if (section) + src->eip += section->vma; + } + } + pclose(file); +out_assign: + sym_filter_entry = syme; + pthread_mutex_unlock(&syme->source_lock); +} + +static void __zero_source_counters(struct sym_entry *syme) +{ + int i; + struct source_line *line; + + line = syme->lines; + while (line) { + for (i = 0; i < nr_counters; i++) + line->count[i] = 0; + line = line->next; + } +} + +static void record_precise_ip(struct sym_entry *syme, int counter, u64 ip) +{ + struct source_line *line; + + if (syme != sym_filter_entry) + return; + + if (pthread_mutex_trylock(&syme->source_lock)) + return; + + if (!syme->source) + goto out_unlock; + + for (line = syme->lines; line; line = line->next) { + if (line->eip == ip) { + line->count[counter]++; + break; + } + if (line->eip > ip) + break; + } +out_unlock: + pthread_mutex_unlock(&syme->source_lock); +} + +static void lookup_sym_source(struct sym_entry *syme) +{ + struct symbol *symbol = (struct symbol *)(syme + 1); + struct source_line *line; + char pattern[PATH_MAX]; + char *idx; + + sprintf(pattern, "<%s>:", symbol->name); + + if (symbol->module) { + idx = strstr(pattern, "\t"); + if (idx) + *idx = 0; + } + + pthread_mutex_lock(&syme->source_lock); + for (line = syme->lines; line; line = line->next) { + if (strstr(line->line, pattern)) { + syme->source = line; + break; + } + } + pthread_mutex_unlock(&syme->source_lock); +} + +static void show_lines(struct source_line *queue, int count, int total) +{ + int i; + struct source_line *line; + + line = queue; + for (i = 0; i < count; i++) { + float pcnt = 100.0*(float)line->count[sym_counter]/(float)total; + + printf("%8li %4.1f%%\t%s\n", line->count[sym_counter], pcnt, line->line); + line = line->next; + } +} + +#define TRACE_COUNT 3 + +static void show_details(struct sym_entry *syme) +{ + struct symbol *symbol; + struct source_line *line; + struct source_line *line_queue = NULL; + int displayed = 0; + int line_queue_count = 0, total = 0, more = 0; + + if (!syme) + return; + + if (!syme->source) + lookup_sym_source(syme); + + if (!syme->source) + return; + + symbol = (struct symbol *)(syme + 1); + printf("Showing %s for %s\n", event_name(sym_counter), symbol->name); + printf(" Events Pcnt (>=%d%%)\n", sym_pcnt_filter); + + pthread_mutex_lock(&syme->source_lock); + line = syme->source; + while (line) { + total += line->count[sym_counter]; + line = line->next; + } + + line = syme->source; + while (line) { + float pcnt = 0.0; + + if (!line_queue_count) + line_queue = line; + line_queue_count++; + + if (line->count[sym_counter]) + pcnt = 100.0 * line->count[sym_counter] / (float)total; + if (pcnt >= (float)sym_pcnt_filter) { + if (displayed <= print_entries) + show_lines(line_queue, line_queue_count, total); + else more++; + displayed += line_queue_count; + line_queue_count = 0; + line_queue = NULL; + } else if (line_queue_count > TRACE_COUNT) { + line_queue = line_queue->next; + line_queue_count--; + } + + line->count[sym_counter] = zero ? 0 : line->count[sym_counter] * 7 / 8; + line = line->next; + } + pthread_mutex_unlock(&syme->source_lock); + if (more) + printf("%d lines not displayed, maybe increase display entries [e]\n", more); +} struct dso *kernel_dso; @@ -112,6 +355,9 @@ static double sym_weight(const struct sym_entry *sym) double weight = sym->snap_count; int counter; + if (!display_weighted) + return weight; + for (counter = 1; counter < nr_counters-1; counter++) weight *= sym->count[counter]; @@ -159,7 +405,7 @@ static void rb_insert_active_sym(struct rb_root *tree, struct sym_entry *se) static void print_sym_table(void) { int printed = 0, j; - int counter; + int counter, snap = !display_weighted ? sym_counter : 0; float samples_per_sec = samples/delay_secs; float ksamples_per_sec = (samples-userspace_samples)/delay_secs; float sum_ksamples = 0.0; @@ -175,7 +421,7 @@ static void print_sym_table(void) pthread_mutex_unlock(&active_symbols_lock); list_for_each_entry_safe_from(syme, n, &active_symbols, node) { - syme->snap_count = syme->count[0]; + syme->snap_count = syme->count[snap]; if (syme->snap_count != 0) { syme->weight = sym_weight(syme); rb_insert_active_sym(&tmp, syme); @@ -195,7 +441,7 @@ static void print_sym_table(void) samples_per_sec, 100.0 - (100.0*((samples_per_sec-ksamples_per_sec)/samples_per_sec))); - if (nr_counters == 1) { + if (nr_counters == 1 || !display_weighted) { printf("%Ld", (u64)attrs[0].sample_period); if (freq) printf("Hz "); @@ -203,7 +449,9 @@ static void print_sym_table(void) printf(" "); } - for (counter = 0; counter < nr_counters; counter++) { + if (!display_weighted) + printf("%s", event_name(sym_counter)); + else for (counter = 0; counter < nr_counters; counter++) { if (counter) printf("/"); @@ -228,6 +476,11 @@ static void print_sym_table(void) printf("------------------------------------------------------------------------------\n\n"); + if (sym_filter_entry) { + show_details(sym_filter_entry); + return; + } + if (nr_counters == 1) printf(" samples pcnt"); else @@ -242,13 +495,13 @@ static void print_sym_table(void) struct symbol *sym = (struct symbol *)(syme + 1); double pcnt; - if (++printed > print_entries || syme->snap_count < count_filter) + if (++printed > print_entries || (int)syme->snap_count < count_filter) continue; pcnt = 100.0 - (100.0 * ((sum_ksamples - syme->snap_count) / sum_ksamples)); - if (nr_counters == 1) + if (nr_counters == 1 || !display_weighted) printf("%20.2f - ", syme->weight); else printf("%9.1f %10ld - ", syme->weight, syme->snap_count); @@ -261,19 +514,250 @@ static void print_sym_table(void) } } +static void prompt_integer(int *target, const char *msg) +{ + char *buf = malloc(0), *p; + size_t dummy = 0; + int tmp; + + fprintf(stdout, "\n%s: ", msg); + if (getline(&buf, &dummy, stdin) < 0) + return; + + p = strchr(buf, '\n'); + if (p) + *p = 0; + + p = buf; + while(*p) { + if (!isdigit(*p)) + goto out_free; + p++; + } + tmp = strtoul(buf, NULL, 10); + *target = tmp; +out_free: + free(buf); +} + +static void prompt_percent(int *target, const char *msg) +{ + int tmp = 0; + + prompt_integer(&tmp, msg); + if (tmp >= 0 && tmp <= 100) + *target = tmp; +} + +static void prompt_symbol(struct sym_entry **target, const char *msg) +{ + char *buf = malloc(0), *p; + struct sym_entry *syme = *target, *n, *found = NULL; + size_t dummy = 0; + + /* zero counters of active symbol */ + if (syme) { + pthread_mutex_lock(&syme->source_lock); + __zero_source_counters(syme); + *target = NULL; + pthread_mutex_unlock(&syme->source_lock); + } + + fprintf(stdout, "\n%s: ", msg); + if (getline(&buf, &dummy, stdin) < 0) + goto out_free; + + p = strchr(buf, '\n'); + if (p) + *p = 0; + + pthread_mutex_lock(&active_symbols_lock); + syme = list_entry(active_symbols.next, struct sym_entry, node); + pthread_mutex_unlock(&active_symbols_lock); + + list_for_each_entry_safe_from(syme, n, &active_symbols, node) { + struct symbol *sym = (struct symbol *)(syme + 1); + + if (!strcmp(buf, sym->name)) { + found = syme; + break; + } + } + + if (!found) { + fprintf(stderr, "Sorry, %s is not active.\n", sym_filter); + sleep(1); + return; + } else + parse_source(found); + +out_free: + free(buf); +} + +static void print_mapped_keys(void) +{ + char *name = NULL; + + if (sym_filter_entry) { + struct symbol *sym = (struct symbol *)(sym_filter_entry+1); + name = sym->name; + } + + fprintf(stdout, "\nMapped keys:\n"); + fprintf(stdout, "\t[d] display refresh delay. \t(%d)\n", delay_secs); + fprintf(stdout, "\t[e] display entries (lines). \t(%d)\n", print_entries); + + if (nr_counters > 1) + fprintf(stdout, "\t[E] active event counter. \t(%s)\n", event_name(sym_counter)); + + fprintf(stdout, "\t[f] profile display filter (count). \t(%d)\n", count_filter); + + if (vmlinux) { + fprintf(stdout, "\t[F] annotate display filter (percent). \t(%d%%)\n", sym_pcnt_filter); + fprintf(stdout, "\t[s] annotate symbol. \t(%s)\n", name?: "NULL"); + fprintf(stdout, "\t[S] stop annotation.\n"); + } + + if (nr_counters > 1) + fprintf(stdout, "\t[w] toggle display weighted/count[E]r. \t(%d)\n", display_weighted ? 1 : 0); + + fprintf(stdout, "\t[z] toggle sample zeroing. \t(%d)\n", zero ? 1 : 0); + fprintf(stdout, "\t[qQ] quit.\n"); +} + +static int key_mapped(int c) +{ + switch (c) { + case 'd': + case 'e': + case 'f': + case 'z': + case 'q': + case 'Q': + return 1; + case 'E': + case 'w': + return nr_counters > 1 ? 1 : 0; + case 'F': + case 's': + case 'S': + return vmlinux ? 1 : 0; + } + + return 0; +} + +static void handle_keypress(int c) +{ + if (!key_mapped(c)) { + struct pollfd stdin_poll = { .fd = 0, .events = POLLIN }; + struct termios tc, save; + + print_mapped_keys(); + fprintf(stdout, "\nEnter selection, or unmapped key to continue: "); + fflush(stdout); + + tcgetattr(0, &save); + tc = save; + tc.c_lflag &= ~(ICANON | ECHO); + tc.c_cc[VMIN] = 0; + tc.c_cc[VTIME] = 0; + tcsetattr(0, TCSANOW, &tc); + + poll(&stdin_poll, 1, -1); + c = getc(stdin); + + tcsetattr(0, TCSAFLUSH, &save); + if (!key_mapped(c)) + return; + } + + switch (c) { + case 'd': + prompt_integer(&delay_secs, "Enter display delay"); + break; + case 'e': + prompt_integer(&print_entries, "Enter display entries (lines)"); + break; + case 'E': + if (nr_counters > 1) { + int i; + + fprintf(stderr, "\nAvailable events:"); + for (i = 0; i < nr_counters; i++) + fprintf(stderr, "\n\t%d %s", i, event_name(i)); + + prompt_integer(&sym_counter, "Enter details event counter"); + + if (sym_counter >= nr_counters) { + fprintf(stderr, "Sorry, no such event, using %s.\n", event_name(0)); + sym_counter = 0; + sleep(1); + } + } else sym_counter = 0; + break; + case 'f': + prompt_integer(&count_filter, "Enter display event count filter"); + break; + case 'F': + prompt_percent(&sym_pcnt_filter, "Enter details display event filter (percent)"); + break; + case 'q': + case 'Q': + printf("exiting.\n"); + exit(0); + case 's': + prompt_symbol(&sym_filter_entry, "Enter details symbol"); + break; + case 'S': + if (!sym_filter_entry) + break; + else { + struct sym_entry *syme = sym_filter_entry; + + pthread_mutex_lock(&syme->source_lock); + sym_filter_entry = NULL; + __zero_source_counters(syme); + pthread_mutex_unlock(&syme->source_lock); + } + break; + case 'w': + display_weighted = ~display_weighted; + break; + case 'z': + zero = ~zero; + break; + } +} + static void *display_thread(void *arg __used) { struct pollfd stdin_poll = { .fd = 0, .events = POLLIN }; - int delay_msecs = delay_secs * 1000; + struct termios tc, save; + int delay_msecs, c; + + tcgetattr(0, &save); + tc = save; + tc.c_lflag &= ~(ICANON | ECHO); + tc.c_cc[VMIN] = 0; + tc.c_cc[VTIME] = 0; - printf("PerfTop refresh period: %d seconds\n", delay_secs); +repeat: + delay_msecs = delay_secs * 1000; + tcsetattr(0, TCSANOW, &tc); + /* trash return*/ + getc(stdin); do { print_sym_table(); } while (!poll(&stdin_poll, 1, delay_msecs) == 1); - printf("key pressed - exiting.\n"); - exit(0); + c = getc(stdin); + tcsetattr(0, TCSAFLUSH, &save); + + handle_keypress(c); + goto repeat; return NULL; } @@ -293,7 +777,6 @@ static const char *skip_symbols[] = { static int symbol_filter(struct dso *self, struct symbol *sym) { - static int filter_match; struct sym_entry *syme; const char *name = sym->name; int i; @@ -315,6 +798,10 @@ static int symbol_filter(struct dso *self, struct symbol *sym) return 1; syme = dso__sym_priv(self, sym); + pthread_mutex_init(&syme->source_lock, NULL); + if (!sym_filter_entry && sym_filter && !strcmp(name, sym_filter)) + sym_filter_entry = syme; + for (i = 0; skip_symbols[i]; i++) { if (!strcmp(skip_symbols[i], name)) { syme->skip = 1; @@ -322,29 +809,6 @@ static int symbol_filter(struct dso *self, struct symbol *sym) } } - if (filter_match == 1) { - filter_end = sym->start; - filter_match = -1; - if (filter_end - filter_start > 10000) { - fprintf(stderr, - "hm, too large filter symbol <%s> - skipping.\n", - sym_filter); - fprintf(stderr, "symbol filter start: %016lx\n", - filter_start); - fprintf(stderr, " end: %016lx\n", - filter_end); - filter_end = filter_start = 0; - sym_filter = NULL; - sleep(1); - } - } - - if (filter_match == 0 && sym_filter && !strcmp(name, sym_filter)) { - filter_match = 1; - filter_start = sym->start; - } - - return 0; } @@ -380,8 +844,6 @@ out_delete_dso: return -1; } -#define TRACE_COUNT 3 - /* * Binary search in the histogram table and record the hit: */ @@ -394,6 +856,7 @@ static void record_ip(u64 ip, int counter) if (!syme->skip) { syme->count[counter]++; + record_precise_ip(syme, counter, ip); pthread_mutex_lock(&active_symbols_lock); if (list_empty(&syme->node) || !syme->node.next) __list_insert_active_sym(syme); @@ -690,8 +1153,8 @@ static const struct option options[] = { "put the counters into a counter group"), OPT_BOOLEAN('i', "inherit", &inherit, "child tasks inherit counters"), - OPT_STRING('s', "sym-filter", &sym_filter, "pattern", - "only display symbols matchig this pattern"), + OPT_STRING('s', "sym-annotate", &sym_filter, "symbol name", + "symbol to annotate - requires -k option"), OPT_BOOLEAN('z', "zero", &zero, "zero history across updates"), OPT_INTEGER('F', "freq", &freq, @@ -734,6 +1197,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __used) delay_secs = 1; parse_symbols(); + parse_source(sym_filter_entry); /* * Fill in the ones not specifically initialized via -c: diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c index 9d3c8141b8c..01147341164 100644 --- a/tools/perf/util/callchain.c +++ b/tools/perf/util/callchain.c @@ -13,6 +13,7 @@ #include <stdio.h> #include <stdbool.h> #include <errno.h> +#include <math.h> #include "callchain.h" @@ -26,10 +27,14 @@ rb_insert_callchain(struct rb_root *root, struct callchain_node *chain, struct rb_node **p = &root->rb_node; struct rb_node *parent = NULL; struct callchain_node *rnode; + u64 chain_cumul = cumul_hits(chain); while (*p) { + u64 rnode_cumul; + parent = *p; rnode = rb_entry(parent, struct callchain_node, rb_node); + rnode_cumul = cumul_hits(rnode); switch (mode) { case CHAIN_FLAT: @@ -40,7 +45,7 @@ rb_insert_callchain(struct rb_root *root, struct callchain_node *chain, break; case CHAIN_GRAPH_ABS: /* Falldown */ case CHAIN_GRAPH_REL: - if (rnode->cumul_hit < chain->cumul_hit) + if (rnode_cumul < chain_cumul) p = &(*p)->rb_left; else p = &(*p)->rb_right; @@ -87,7 +92,7 @@ static void __sort_chain_graph_abs(struct callchain_node *node, chain_for_each_child(child, node) { __sort_chain_graph_abs(child, min_hit); - if (child->cumul_hit >= min_hit) + if (cumul_hits(child) >= min_hit) rb_insert_callchain(&node->rb_root, child, CHAIN_GRAPH_ABS); } @@ -108,11 +113,11 @@ static void __sort_chain_graph_rel(struct callchain_node *node, u64 min_hit; node->rb_root = RB_ROOT; - min_hit = node->cumul_hit * min_percent / 100.0; + min_hit = ceil(node->children_hit * min_percent); chain_for_each_child(child, node) { __sort_chain_graph_rel(child, min_percent); - if (child->cumul_hit >= min_hit) + if (cumul_hits(child) >= min_hit) rb_insert_callchain(&node->rb_root, child, CHAIN_GRAPH_REL); } @@ -122,7 +127,7 @@ static void sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_node *chain_root, u64 min_hit __used, struct callchain_param *param) { - __sort_chain_graph_rel(chain_root, param->min_percent); + __sort_chain_graph_rel(chain_root, param->min_percent / 100.0); rb_root->rb_node = chain_root->rb_root.rb_node; } @@ -211,7 +216,8 @@ add_child(struct callchain_node *parent, struct ip_callchain *chain, new = create_child(parent, false); fill_node(new, chain, start, syms); - new->cumul_hit = new->hit = 1; + new->children_hit = 0; + new->hit = 1; } /* @@ -241,7 +247,8 @@ split_add_child(struct callchain_node *parent, struct ip_callchain *chain, /* split the hits */ new->hit = parent->hit; - new->cumul_hit = parent->cumul_hit; + new->children_hit = parent->children_hit; + parent->children_hit = cumul_hits(new); new->val_nr = parent->val_nr - idx_local; parent->val_nr = idx_local; @@ -249,6 +256,7 @@ split_add_child(struct callchain_node *parent, struct ip_callchain *chain, if (idx_total < chain->nr) { parent->hit = 0; add_child(parent, chain, idx_total, syms); + parent->children_hit++; } else { parent->hit = 1; } @@ -269,13 +277,13 @@ __append_chain_children(struct callchain_node *root, struct ip_callchain *chain, unsigned int ret = __append_chain(rnode, chain, start, syms); if (!ret) - goto cumul; + goto inc_children_hit; } /* nothing in children, add to the current node */ add_child(root, chain, start, syms); -cumul: - root->cumul_hit++; +inc_children_hit: + root->children_hit++; } static int @@ -317,8 +325,6 @@ __append_chain(struct callchain_node *root, struct ip_callchain *chain, /* we match 100% of the path, increment the hit */ if (i - start == root->val_nr && i == chain->nr) { root->hit++; - root->cumul_hit++; - return 0; } @@ -331,5 +337,7 @@ __append_chain(struct callchain_node *root, struct ip_callchain *chain, void append_chain(struct callchain_node *root, struct ip_callchain *chain, struct symbol **syms) { + if (!chain->nr) + return; __append_chain_children(root, chain, syms, 0); } diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h index 7812122bea1..a926ae4f5a1 100644 --- a/tools/perf/util/callchain.h +++ b/tools/perf/util/callchain.h @@ -7,6 +7,7 @@ #include "symbol.h" enum chain_mode { + CHAIN_NONE, CHAIN_FLAT, CHAIN_GRAPH_ABS, CHAIN_GRAPH_REL @@ -21,7 +22,7 @@ struct callchain_node { struct rb_root rb_root; /* sorted tree of children */ unsigned int val_nr; u64 hit; - u64 cumul_hit; /* hit + hits of children */ + u64 children_hit; }; struct callchain_param; @@ -48,6 +49,11 @@ static inline void callchain_init(struct callchain_node *node) INIT_LIST_HEAD(&node->val); } +static inline u64 cumul_hits(struct callchain_node *node) +{ + return node->hit + node->children_hit; +} + int register_callchain_param(struct callchain_param *param); void append_chain(struct callchain_node *root, struct ip_callchain *chain, struct symbol **syms); diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index 450384b3bbe..b92a457ca32 100644 --- a/tools/perf/util/header.c +++ b/tools/perf/util/header.c @@ -185,6 +185,8 @@ static void do_read(int fd, void *buf, size_t size) if (ret < 0) die("failed to read"); + if (ret == 0) + die("failed to read: missing data"); size -= ret; buf += ret; @@ -213,9 +215,10 @@ struct perf_header *perf_header__read(int fd) for (i = 0; i < nr_attrs; i++) { struct perf_header_attr *attr; - off_t tmp = lseek(fd, 0, SEEK_CUR); + off_t tmp; do_read(fd, &f_attr, sizeof(f_attr)); + tmp = lseek(fd, 0, SEEK_CUR); attr = perf_header_attr__new(&f_attr.attr); diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 7bdad8df22a..4858d83b3b6 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -121,13 +121,29 @@ static unsigned long hw_cache_stat[C(MAX)] = { (strcmp(sys_dirent.d_name, ".")) && \ (strcmp(sys_dirent.d_name, ".."))) +static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir) +{ + char evt_path[MAXPATHLEN]; + int fd; + + snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", debugfs_path, + sys_dir->d_name, evt_dir->d_name); + fd = open(evt_path, O_RDONLY); + if (fd < 0) + return -EINVAL; + close(fd); + + return 0; +} + #define for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next, file, st) \ while (!readdir_r(evt_dir, &evt_dirent, &evt_next) && evt_next) \ if (snprintf(file, MAXPATHLEN, "%s/%s/%s", debugfs_path, \ sys_dirent.d_name, evt_dirent.d_name) && \ (!stat(file, &st)) && (S_ISDIR(st.st_mode)) && \ (strcmp(evt_dirent.d_name, ".")) && \ - (strcmp(evt_dirent.d_name, ".."))) + (strcmp(evt_dirent.d_name, "..")) && \ + (!tp_event_has_id(&sys_dirent, &evt_dirent))) #define MAX_EVENT_LENGTH 30 @@ -223,9 +239,15 @@ char *event_name(int counter) { u64 config = attrs[counter].config; int type = attrs[counter].type; + + return __event_name(type, config); +} + +char *__event_name(int type, u64 config) +{ static char buf[32]; - if (attrs[counter].type == PERF_TYPE_RAW) { + if (type == PERF_TYPE_RAW) { sprintf(buf, "raw 0x%llx", config); return buf; } diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h index 1ea5d09b6eb..192a962e3a0 100644 --- a/tools/perf/util/parse-events.h +++ b/tools/perf/util/parse-events.h @@ -10,6 +10,7 @@ extern int nr_counters; extern struct perf_counter_attr attrs[MAX_COUNTERS]; extern char *event_name(int ctr); +extern char *__event_name(int type, u64 config); extern int parse_events(const struct option *opt, const char *str, int unset); diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 16ddca20294..f1dcede1430 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -24,6 +24,16 @@ const char *sym_hist_filter; #define DMGL_ANSI (1 << 1) /* Include const, volatile, etc */ #endif +enum dso_origin { + DSO__ORIG_KERNEL = 0, + DSO__ORIG_JAVA_JIT, + DSO__ORIG_FEDORA, + DSO__ORIG_UBUNTU, + DSO__ORIG_BUILDID, + DSO__ORIG_DSO, + DSO__ORIG_NOT_FOUND, +}; + static struct symbol *symbol__new(u64 start, u64 len, const char *name, unsigned int priv_size, u64 obj_start, int verbose) @@ -81,6 +91,7 @@ struct dso *dso__new(const char *name, unsigned int sym_priv_size) self->sym_priv_size = sym_priv_size; self->find_symbol = dso__find_symbol; self->slen_calculated = 0; + self->origin = DSO__ORIG_NOT_FOUND; } return self; @@ -710,7 +721,7 @@ static char *dso__read_build_id(struct dso *self, int verbose) ++raw; bid += 2; } - if (verbose) + if (verbose >= 2) printf("%s(%s): %s\n", __func__, self->name, build_id); out_elf_end: elf_end(elf); @@ -720,11 +731,26 @@ out: return build_id; } +char dso__symtab_origin(const struct dso *self) +{ + static const char origin[] = { + [DSO__ORIG_KERNEL] = 'k', + [DSO__ORIG_JAVA_JIT] = 'j', + [DSO__ORIG_FEDORA] = 'f', + [DSO__ORIG_UBUNTU] = 'u', + [DSO__ORIG_BUILDID] = 'b', + [DSO__ORIG_DSO] = 'd', + }; + + if (self == NULL || self->origin == DSO__ORIG_NOT_FOUND) + return '!'; + return origin[self->origin]; +} + int dso__load(struct dso *self, symbol_filter_t filter, int verbose) { int size = PATH_MAX; char *name = malloc(size), *build_id = NULL; - int variant = 0; int ret = -1; int fd; @@ -733,19 +759,26 @@ int dso__load(struct dso *self, symbol_filter_t filter, int verbose) self->adjust_symbols = 0; - if (strncmp(self->name, "/tmp/perf-", 10) == 0) - return dso__load_perf_map(self, filter, verbose); + if (strncmp(self->name, "/tmp/perf-", 10) == 0) { + ret = dso__load_perf_map(self, filter, verbose); + self->origin = ret > 0 ? DSO__ORIG_JAVA_JIT : + DSO__ORIG_NOT_FOUND; + return ret; + } + + self->origin = DSO__ORIG_FEDORA - 1; more: do { - switch (variant) { - case 0: /* Fedora */ + self->origin++; + switch (self->origin) { + case DSO__ORIG_FEDORA: snprintf(name, size, "/usr/lib/debug%s.debug", self->name); break; - case 1: /* Ubuntu */ + case DSO__ORIG_UBUNTU: snprintf(name, size, "/usr/lib/debug%s", self->name); break; - case 2: + case DSO__ORIG_BUILDID: build_id = dso__read_build_id(self, verbose); if (build_id != NULL) { snprintf(name, size, @@ -754,16 +787,15 @@ more: free(build_id); break; } - variant++; + self->origin++; /* Fall thru */ - case 3: /* Sane people */ + case DSO__ORIG_DSO: snprintf(name, size, "%s", self->name); break; default: goto out; } - variant++; fd = open(name, O_RDONLY); } while (fd < 0); @@ -899,6 +931,9 @@ int dso__load_kernel(struct dso *self, const char *vmlinux, if (err <= 0) err = dso__load_kallsyms(self, filter, verbose); + if (err > 0) + self->origin = DSO__ORIG_KERNEL; + return err; } diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h index 2f92b21c712..1e003ec2f4b 100644 --- a/tools/perf/util/symbol.h +++ b/tools/perf/util/symbol.h @@ -26,6 +26,7 @@ struct dso { unsigned int sym_priv_size; unsigned char adjust_symbols; unsigned char slen_calculated; + unsigned char origin; char name[0]; }; @@ -49,6 +50,7 @@ int dso__load_modules(struct dso *self, symbol_filter_t filter, int verbose); int dso__load(struct dso *self, symbol_filter_t filter, int verbose); size_t dso__fprintf(struct dso *self, FILE *fp); +char dso__symtab_origin(const struct dso *self); void symbol__init(void); #endif /* _PERF_SYMBOL_ */ |