Pure nfs client performance using odirect.

When an application opens a file with O_DIRECT flag, if the size of
the data that is written is equal to wsize, the client sends a
WRITE RPC with stable flag set to UNSTABLE followed by a single
COMMIT RPC rather than sending a single WRITE RPC with the stable
flag set to FILE_SYNC. This a bug.

Patch to fix this.

Signed-off-by: Arun R Bharadwaj <arun@linux.vnet.ibm.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
This commit is contained in:
Arun Bharadwaj 2010-11-18 10:36:43 +00:00 committed by Trond Myklebust
parent 5fc43978a7
commit b47d19de2c

View file

@ -867,7 +867,7 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, const struct iovec *iov,
goto out;
nfs_alloc_commit_data(dreq);
if (dreq->commit_data == NULL || count < wsize)
if (dreq->commit_data == NULL || count <= wsize)
sync = NFS_FILE_SYNC;
dreq->inode = inode;