6.6-stable review patch. If anyone has any objections, please let me know.
------------------
From: Trond Myklebust trond.myklebust@hammerspace.com
[ Upstream commit fcf857ee1958e9247298251f7615d0c76f1e9b38 ]
While it is uncommon for delegations to be held while O_DIRECT writes are in progress, it is possible. The xfstests generic/647 and generic/729 both end up triggering that state, and end up failing due to the fact that the file size is not adjusted.
Reported-by: Chuck Lever chuck.lever@oracle.com Link: https://bugzilla.kernel.org/show_bug.cgi?id=219738 Cc: stable@vger.kernel.org Signed-off-by: Trond Myklebust trond.myklebust@hammerspace.com Signed-off-by: Anna Schumaker anna.schumaker@oracle.com Signed-off-by: Sasha Levin sashal@kernel.org --- fs/nfs/direct.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 258521d5125ed..a1ff4a4f5380e 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -130,6 +130,20 @@ static void nfs_direct_truncate_request(struct nfs_direct_req *dreq, dreq->count = req_start; }
+static void nfs_direct_file_adjust_size_locked(struct inode *inode, + loff_t offset, size_t count) +{ + loff_t newsize = offset + (loff_t)count; + loff_t oldsize = i_size_read(inode); + + if (newsize > oldsize) { + i_size_write(inode, newsize); + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE; + trace_nfs_size_grow(inode, newsize); + nfs_inc_stats(inode, NFSIOS_EXTENDWRITE); + } +} + /** * nfs_swap_rw - NFS address space operation for swap I/O * @iocb: target I/O control block @@ -728,6 +742,7 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) struct nfs_direct_req *dreq = hdr->dreq; struct nfs_commit_info cinfo; struct nfs_page *req = nfs_list_entry(hdr->pages.next); + struct inode *inode = dreq->inode; int flags = NFS_ODIRECT_DONE;
trace_nfs_direct_write_completion(dreq); @@ -749,6 +764,10 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) } spin_unlock(&dreq->lock);
+ spin_lock(&inode->i_lock); + nfs_direct_file_adjust_size_locked(inode, dreq->io_start, dreq->count); + spin_unlock(&inode->i_lock); + while (!list_empty(&hdr->pages)) {
req = nfs_list_entry(hdr->pages.next);