NFSv4: Fix a potential CLOSE race
authorTrond Myklebust <Trond.Myklebust@netapp.com>
Tue, 18 Oct 2005 21:20:12 +0000 (14:20 -0700)
committerTrond Myklebust <Trond.Myklebust@netapp.com>
Tue, 18 Oct 2005 21:20:12 +0000 (14:20 -0700)
commit9512135df14f8293b9bc5e8fb22d4279dee5ff66
tree83fb778ed53ba10c46734968f538aa0a57d376a1
parentcee54fc944422c44e476736c045a9e8053cb0644
NFSv4: Fix a potential CLOSE race

 Once the state_owner and lock_owner semaphores get removed, it will be
 possible for other OPEN requests to reopen the same file if they have
 lower sequence ids than our CLOSE call.
 This patch ensures that we recheck the file state once
 nfs_wait_on_sequence() has completed waiting.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
fs/nfs/nfs4proc.c
fs/nfs/nfs4state.c
fs/nfs/nfs4xdr.c
include/linux/nfs_xdr.h