Skip to content

fs: inode: shmem: Reduce risk of inum overflow

Chris Down linux-fsdevel lkml oe-lkp 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest3 patches 4 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/3fs: inode: Recycle volatile inode numbers from private slabsfs/hugetlbfs/inode.c fs/inode.c mm/shmem.c+11 -20 replies
2/3fs: inode: Add API to retrieve global next ino using full ino_t widthfs/inode.c include/linux/fs.h+38 -41 reply
3/3shmem: Add support for using full width of ino_tinclude/linux/shmem_fs.h mm/shmem.c+40 -23 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [PATCH 0/3] fs: inode: shmem: Reduce risk of inum overflow

In Facebook production we are seeing heavy i_ino wraparounds on tmpfs. On affected tiers, in excess of 10% of hosts show multiple files with different content and the same inode number, with some servers even having as many as 150 duplicated inode numbers with differing file content.

This causes actual, tangible problems in production. For example, we have complaints from those working on remote caches that their application is reporting cache corruptions because it uses (device, inodenum) to establish the identity of a particular cache object, but because it's not unique any more, the application refuses to continue and reports cache corruption. Even worse, sometimes applications may not even detect the corruption but may continue anyway, causing phantom and hard to debug behaviour.

In general, userspace applications expect that (device, inodenum) should be enough to be uniquely point to one inode, which seems fair enough. One might also need to check the generation, but in this case:

1. That's not currently exposed to userspace (ioctl(...FS_IOC_GETVERSION...) returns ENOTTY on tmpfs);
2. Even with generation, there shouldn't be two live inodes with the same inode number on one device.

In order to mitigate this, we take a two-pronged approach:

1. A mitigation that works both for 32- and 64-bit inodes: we reuse inode numbers from recycled slabs where possible (ie. where the filesystem uses their own private inode slabs instead of shared inode slabs), allowing us to significantly reduce the risk of 32 bit wraparound.
2. A fix that works on machines with 64-bit ino_t only: we allow users to mount tmpfs with a new inode64 option that uses the full width of ino_t. Other filesystems can also use get_next_ino_full to get similar behaviour as desired.

Chris Down (3):
fs: inode: Recycle volatile inode numbers from private slabs
fs: inode: Add API to retrieve global next ino using full ino_t width
shmem: Add support for using full width of ino_t

 fs/hugetlbfs/inode.c     |  4 +++-
 fs/inode.c               | 44 +++++++++++++++++++++++++++++++++++++---
 include/linux/fs.h       |  1 +
 include/linux/shmem_fs.h |  1 +
 mm/shmem.c               | 41 ++++++++++++++++++++++++++++++++++++-
 5 files changed, 86 insertions(+), 5 deletions(-)
Signature · 1 line
2.24.1