linux/include/asm-generic
Linus Torvalds 2b047252d0 Fix TLB gather virtual address range invalidation corner cases
Ben Tebulin reported:

 "Since v3.7.2 on two independent machines a very specific Git
  repository fails in 9/10 cases on git-fsck due to an SHA1/memory
  failures.  This only occurs on a very specific repository and can be
  reproduced stably on two independent laptops.  Git mailing list ran
  out of ideas and for me this looks like some very exotic kernel issue"

and bisected the failure to the backport of commit 53a59fc67f ("mm:
limit mmu_gather batching to fix soft lockups on !CONFIG_PREEMPT").

That commit itself is not actually buggy, but what it does is to make it
much more likely to hit the partial TLB invalidation case, since it
introduces a new case in tlb_next_batch() that previously only ever
happened when running out of memory.

The real bug is that the TLB gather virtual memory range setup is subtly
buggered.  It was introduced in commit 597e1c3580 ("mm/mmu_gather:
enable tlb flush range in generic mmu_gather"), and the range handling
was already fixed at least once in commit e6c495a96c ("mm: fix the TLB
range flushed when __tlb_remove_page() runs out of slots"), but that fix
was not complete.

The problem with the TLB gather virtual address range is that it isn't
set up by the initial tlb_gather_mmu() initialization (which didn't get
the TLB range information), but it is set up ad-hoc later by the
functions that actually flush the TLB.  And so any such case that forgot
to update the TLB range entries would potentially miss TLB invalidates.

Rather than try to figure out exactly which particular ad-hoc range
setup was missing (I personally suspect it's the hugetlb case in
zap_huge_pmd(), which didn't have the same logic as zap_pte_range()
did), this patch just gets rid of the problem at the source: make the
TLB range information available to tlb_gather_mmu(), and initialize it
when initializing all the other tlb gather fields.

This makes the patch larger, but conceptually much simpler.  And the end
result is much more understandable; even if you want to play games with
partial ranges when invalidating the TLB contents in chunks, now the
range information is always there, and anybody who doesn't want to
bother with it won't introduce subtle bugs.

Ben verified that this fixes his problem.

Reported-bisected-and-tested-by: Ben Tebulin <tebulin@googlemail.com>
Build-testing-by: Stephen Rothwell <sfr@canb.auug.org.au>
Build-testing-by: Richard Weinberger <richard.weinberger@gmail.com>
Reviewed-by: Michal Hocko <mhocko@suse.cz>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-08-16 08:52:46 -07:00
..
bitops Merge branch 'modules-next' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux 2012-10-14 13:39:34 -07:00
4level-fixup.h
atomic64.h
atomic-long.h
atomic.h asm-generic: move cmpxchg*_local defs to cmpxchg.h 2013-03-13 06:11:05 +01:00
audit_change_attr.h
audit_dir_write.h
audit_read.h
audit_signal.h
audit_write.h
barrier.h Create asm-generic/barrier.h 2012-03-28 18:30:03 +01:00
bitops.h
bitsperlong.h UAPI: (Scripted) Disintegrate include/asm-generic 2012-10-04 18:20:15 +01:00
bug.h bug.h: Fix up CONFIG_BUG=n implicit function declarations. 2012-06-25 10:32:49 -07:00
bugs.h
cache.h
cacheflush.h
checksum.h asm-generic headers: Allow yet more arch overrides in checksum.h 2013-02-11 20:00:33 +05:30
clkdev.h asm-generic: Add default clkdev.h 2012-10-03 21:33:53 +02:00
cmpxchg-local.h generic: Use raw local irq variant for generic cmpxchg 2013-02-19 08:43:37 +01:00
cmpxchg.h asm-generic: move cmpxchg*_local defs to cmpxchg.h 2013-03-13 06:11:05 +01:00
cputime_jiffies.h cputime: Move default nsecs_to_cputime() to jiffies based cputime file 2013-01-27 19:23:25 +01:00
cputime_nsecs.h cputime_nsecs: use math64.h for nsec resolution conversion helpers 2013-04-26 18:58:12 +02:00
cputime.h cputime: Generic on-demand virtual cputime accounting 2013-01-27 19:23:27 +01:00
current.h
delay.h
device.h
div64.h
dma-coherent.h common: dma-mapping: add support for generic dma_mmap_* calls 2012-07-30 12:25:46 +02:00
dma-contiguous.h mm: cma: fix condition check when setting global cma area 2012-07-06 12:02:04 +02:00
dma-mapping-broken.h asm-generic/dma-mapping-broken.h: Provide dma_alloc_attrs()/dma_free_attrs() 2012-12-25 20:14:54 +01:00
dma-mapping-common.h common: dma-mapping: introduce dma_get_sgtable() function 2012-07-30 12:25:46 +02:00
dma.h
emergency-restart.h
exec.h Split arch_align_stack() out from asm-generic/system.h 2012-03-28 18:30:03 +01:00
fb.h
ftrace.h
futex.h
getorder.h bitops: Add missing parentheses to new get_order macro 2012-02-24 10:39:27 -08:00
gpio.h gpiolib: let gpio_chip reference its descriptors 2013-02-11 22:21:28 +00:00
hardirq.h
hugetlb.h mm/hugetlb: add more arch-defined huge_pte functions 2013-04-29 15:54:33 -07:00
hw_irq.h
ide_iops.h
int-l64.h UAPI: (Scripted) Disintegrate include/asm-generic 2012-10-04 18:20:15 +01:00
int-ll64.h UAPI: (Scripted) Disintegrate include/asm-generic 2012-10-04 18:20:15 +01:00
io-64-nonatomic-hi-lo.h asm-generic: architecture independent readq/writeq for 32bit environment 2012-02-21 16:47:28 -08:00
io-64-nonatomic-lo-hi.h asm-generic: architecture independent readq/writeq for 32bit environment 2012-02-21 16:47:28 -08:00
io.h kernel: Fix s390 absolute memory access for /dev/mem 2013-05-22 09:45:57 +02:00
ioctl.h UAPI: (Scripted) Disintegrate include/asm-generic 2012-10-04 18:20:15 +01:00
iomap.h [PARISC] fix compile break caused by iomap: make IOPORT/PCI mapping functions conditional 2012-02-27 09:43:30 -06:00
irq_regs.h
irq.h
irqflags.h
Kbuild.asm UAPI: Set up uapi/asm/Kbuild.asm 2012-10-02 18:01:56 +01:00
kdebug.h
kmap_types.h asm-generic: remove km_type definitions 2012-07-24 15:27:30 +08:00
kvm_para.h KVM: add kvm_para_available to asm-generic/kvm_para.h 2013-06-05 13:21:29 +03:00
libata-portmap.h
linkage.h
local64.h
local.h
memory_model.h
mm_hooks.h
mmu_context.h
mmu.h asm-generic/mmu.h: Add support for FDPIC 2012-12-09 23:14:14 +01:00
module.h Make most arch asm/module.h files use asm-generic/module.h 2012-09-28 14:31:03 +09:30
mutex-dec.h arch: Make __mutex_fastpath_lock_retval return whether fastpath succeeded or not 2013-06-26 12:10:55 +02:00
mutex-null.h arch: Make __mutex_fastpath_lock_retval return whether fastpath succeeded or not 2013-06-26 12:10:55 +02:00
mutex-xchg.h arch: Make __mutex_fastpath_lock_retval return whether fastpath succeeded or not 2013-06-26 12:10:55 +02:00
mutex.h
page.h
param.h UAPI: (Scripted) Disintegrate include/asm-generic 2012-10-04 18:20:15 +01:00
parport.h include: remove __dev* attributes. 2013-01-03 15:57:16 -08:00
pci_iomap.h [PARISC] fix compile break caused by iomap: make IOPORT/PCI mapping functions conditional 2012-02-27 09:43:30 -06:00
pci-bridge.h PCI: work around Stratus ftServer broken PCIe hierarchy 2012-04-30 15:21:02 -06:00
pci-dma-compat.h
pci.h PCI: collapse pcibios_resource_to_bus 2012-02-23 20:19:04 -07:00
percpu.h
pgalloc.h
pgtable-nopmd.h
pgtable-nopud.h
pgtable.h mm: save soft-dirty bits on file pages 2013-08-13 17:57:48 -07:00
ptrace.h
resource.h UAPI: (Scripted) Disintegrate include/asm-generic 2012-10-04 18:20:15 +01:00
rtc.h
rwsem.h
scatterlist.h
sections.h vmlinux.lds: add comments for global variables and clean up useless declarations 2013-07-03 16:07:34 -07:00
segment.h
serial.h
siginfo.h UAPI: (Scripted) Disintegrate include/asm-generic 2012-10-04 18:20:15 +01:00
signal.h unify default ptrace_signal_deliver 2012-11-29 00:01:23 -05:00
sizes.h ARM: 7430/1: sizes.h: move from asm-generic to <linux/sizes.h> 2012-06-28 17:14:34 +01:00
spinlock.h
statfs.h UAPI: (Scripted) Disintegrate include/asm-generic 2012-10-04 18:20:15 +01:00
string.h
switch_to.h Split the switch_to() wrapper out of asm-generic/system.h 2012-03-28 18:30:03 +01:00
syscall.h asm/syscall.h: add syscall_get_arch 2012-04-14 11:13:19 +10:00
syscalls.h burying unused conditionals 2013-02-14 09:21:15 -05:00
termios-base.h
termios.h UAPI: (Scripted) Disintegrate include/asm-generic 2012-10-04 18:20:15 +01:00
timex.h
tlb.h Fix TLB gather virtual address range invalidation corner cases 2013-08-16 08:52:46 -07:00
tlbflush.h BUG: headers with BUG/BUG_ON etc. need linux/bug.h 2012-03-04 17:54:34 -05:00
topology.h
trace_clock.h tracing,x86: Add a TSC trace_clock 2012-11-13 15:48:27 -05:00
uaccess-unaligned.h
uaccess.h asm-generic: uaccess s/might_sleep/might_fault/ 2013-05-28 09:41:05 +02:00
unaligned.h
unistd.h We get rid of the general module prefix confusion with a binary config option, 2013-05-05 10:58:06 -07:00
user.h
vga.h
vmlinux.lds.h Merge branch 'cpuinit-delete' of git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux 2013-07-07 11:01:19 -07:00
word-at-a-time.h word-at-a-time: make the interfaces truly generic 2012-05-26 11:33:40 -07:00
xor.h asm-generic: xor: mark static functions as __maybe_unused 2012-10-03 21:21:06 +02:00