]> git.proxmox.com Git - mirror_ubuntu-kernels.git/commit
mm,hwpoison: unmap poisoned page before invalidation
authorRik van Riel <riel@surriel.com>
Fri, 1 Apr 2022 18:28:42 +0000 (11:28 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 1 Apr 2022 18:46:09 +0000 (11:46 -0700)
commit3149c79f3cb0e2e3bafb7cfadacec090cbd250d3
tree88be8de834faea2903d15acb52163b52965d5817
parent4f1f9698d77734a90cae1bb9a50188f44f15e305
mm,hwpoison: unmap poisoned page before invalidation

In some cases it appears the invalidation of a hwpoisoned page fails
because the page is still mapped in another process.  This can cause a
program to be continuously restarted and die when it page faults on the
page that was not invalidated.  Avoid that problem by unmapping the
hwpoisoned page when we find it.

Another issue is that sometimes we end up oopsing in finish_fault, if
the code tries to do something with the now-NULL vmf->page.  I did not
hit this error when submitting the previous patch because there are
several opportunities for alloc_set_pte to bail out before accessing
vmf->page, and that apparently happened on those systems, and most of
the time on other systems, too.

However, across several million systems that error does occur a handful
of times a day.  It can be avoided by returning VM_FAULT_NOPAGE which
will cause do_read_fault to return before calling finish_fault.

Link: https://lkml.kernel.org/r/20220325161428.5068d97e@imladris.surriel.com
Fixes: e53ac7374e64 ("mm: invalidate hwpoison page cache page in fault path")
Signed-off-by: Rik van Riel <riel@surriel.com>
Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
Tested-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/memory.c