]> git.baikalelectronics.ru Git - kernel.git/commit
mm: get rid of vmacache_flush_all() entirely
authorLinus Torvalds <torvalds@linux-foundation.org>
Thu, 13 Sep 2018 09:57:48 +0000 (23:57 -1000)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 14 Sep 2018 01:18:04 +0000 (15:18 -1000)
commit7832c1de3bdcc36ae8d86522fc948f6ca3b7938e
treede9aa025a0d421ee3ae3d1973e63419fff8545e5
parent80191f0b8d22aa66a194f677035f4d54d7e3f59e
mm: get rid of vmacache_flush_all() entirely

Jann Horn points out that the vmacache_flush_all() function is not only
potentially expensive, it's buggy too.  It also happens to be entirely
unnecessary, because the sequence number overflow case can be avoided by
simply making the sequence number be 64-bit.  That doesn't even grow the
data structures in question, because the other adjacent fields are
already 64-bit.

So simplify the whole thing by just making the sequence number overflow
case go away entirely, which gets rid of all the complications and makes
the code faster too.  Win-win.

[ Oleg Nesterov points out that the VMACACHE_FULL_FLUSHES statistics
  also just goes away entirely with this ]

Reported-by: Jann Horn <jannh@google.com>
Suggested-by: Will Deacon <will.deacon@arm.com>
Acked-by: Davidlohr Bueso <dave@stgolabs.net>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
include/linux/mm_types.h
include/linux/mm_types_task.h
include/linux/vm_event_item.h
include/linux/vmacache.h
mm/debug.c
mm/vmacache.c