diff options
| author | Peter Zijlstra <peterz@infradead.org> | 2020-03-04 11:28:31 +0100 |
|---|---|---|
| committer | Peter Zijlstra <peterz@infradead.org> | 2020-03-06 11:06:15 +0100 |
| commit | 8019ad13ef7f64be44d4f892af9c840179009254 (patch) | |
| tree | 889b754b07fa844410d6782bc1b4473007a9bc99 /fs/inode.c | |
| parent | 98d54f81e36ba3bf92172791eba5ca5bd813989b (diff) | |
futex: Fix inode life-time issue
As reported by Jann, ihold() does not in fact guarantee inode
persistence. And instead of making it so, replace the usage of inode
pointers with a per boot, machine wide, unique inode identifier.
This sequence number is global, but shared (file backed) futexes are
rare enough that this should not become a performance issue.
Reported-by: Jann Horn <jannh@google.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Diffstat (limited to 'fs/inode.c')
| -rw-r--r-- | fs/inode.c | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/fs/inode.c b/fs/inode.c index 7d57068b6b7a..93d9252a00ab 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -138,6 +138,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode) inode->i_sb = sb; inode->i_blkbits = sb->s_blocksize_bits; inode->i_flags = 0; + atomic64_set(&inode->i_sequence, 0); atomic_set(&inode->i_count, 1); inode->i_op = &empty_iops; inode->i_fop = &no_open_fops; |