nfs: stat(2) fails during cthon04 basic test5 on NFSv4.0

When running the Connectathon basic tests against a Solaris NFS
server over NFSv4.0, test5 reports that stat(2) returns a file size
of zero instead of 1MB.

On success, nfs_commit_inode() can return a positive result; see
other call sites such as nfs_file_fsync_commit() and
nfs_commit_unstable_pages().

The call site recently added in nfs_wb_all() does not prevent that
positive return value from leaking to its callers. If it leaks
through nfs_sync_inode() back to nfs_getattr(), that causes stat(2)
to return a positive return value to user space while also not
filling in the passed-in struct stat.

Additional clean up: the new logic in nfs_wb_all() is rewritten in
bfields-normal form.

Fixes: 5bb89b4702e2 ("NFSv4.1/pnfs: Separate out metadata . . .")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

1 files changed, 8 insertions, 5 deletions

diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index d12a4be613a5..dfc19f1575a1 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1845,12 +1845,15 @@ int nfs_wb_all(struct inode *inode)
         trace_nfs_writeback_inode_enter(inode);
 
         ret = filemap_write_and_wait(inode->i_mapping);
-        if (!ret) {
-                ret = nfs_commit_inode(inode, FLUSH_SYNC);
-                if (!ret)
-                        pnfs_sync_inode(inode, true);
-        }
+        if (ret)
+                goto out;
+        ret = nfs_commit_inode(inode, FLUSH_SYNC);
+        if (ret < 0)
+                goto out;
+        pnfs_sync_inode(inode, true);
+        ret = 0;
 
+out:
         trace_nfs_writeback_inode_exit(inode, ret);
         return ret;
 }