[v2,0/2] hugetlbfs: use i_mmap_rwsem for better synchronization

Message

Mike Kravetz Dec. 18, 2018, 10:35 p.m. UTC

There are two primary issues addressed here:
1) For shared pmds, huge PTE pointers returned by huge_pte_alloc can become
   invalid via a call to huge_pmd_unshare by another thread.
2) hugetlbfs page faults can race with truncation causing invalid global
   reserve counts and state.
Both issues are addressed by expanding the use of i_mmap_rwsem.

These issues have existed for a long time.  They can be recreated with a
test program that causes page fault/truncation races.  For simple mappings,
this results in a negative HugePages_Rsvd count.  If racing with mappings
that contain shared pmds, we can hit "BUG at fs/hugetlbfs/inode.c:444!" or
Oops! as the result of an invalid memory reference.

v1 -> v2
  Combined patches 2 and 3 of v1 series as suggested by Aneesh.  No other
  changes were made.
Patches are a follow up to the RFC,
  http://lkml.kernel.org/r/[e-mail]
  Comments made by Naoya were addressed.

Mike Kravetz (2):
  hugetlbfs: use i_mmap_rwsem for more pmd sharing synchronization
  hugetlbfs: Use i_mmap_rwsem to fix page fault/truncate race

 fs/hugetlbfs/inode.c | 50 +++++++++----------------
 mm/hugetlb.c         | 87 +++++++++++++++++++++++++++++++-------------
 mm/memory-failure.c  | 14 ++++++-
 mm/migrate.c         | 13 ++++++-
 mm/rmap.c            |  3 ++
 mm/userfaultfd.c     | 11 +++++-
 6 files changed, 116 insertions(+), 62 deletions(-)

Comments

Andrew Morton Dec. 20, 2018, 9:06 p.m. UTC | #1

On Tue, 18 Dec 2018 14:35:55 -0800 Mike Kravetz <[e-mail]> wrote:

> There are two primary issues addressed here:
> 1) For shared pmds, huge PTE pointers returned by huge_pte_alloc can become
>    invalid via a call to huge_pmd_unshare by another thread.
> 2) hugetlbfs page faults can race with truncation causing invalid global
>    reserve counts and state.
> Both issues are addressed by expanding the use of i_mmap_rwsem.
> 
> These issues have existed for a long time.  They can be recreated with a
> test program that causes page fault/truncation races.  For simple mappings,
> this results in a negative HugePages_Rsvd count.  If racing with mappings
> that contain shared pmds, we can hit "BUG at fs/hugetlbfs/inode.c:444!" or
> Oops! as the result of an invalid memory reference.

Still no reviewers or ackers :(

I'll queue these for 4.21-rc1.  The Fixes: commits are over a decade
old so I assume things aren't super-urgent and the cc:stable will do
its work.