diff options
| author | Weston Andros Adamson <dros@primarydata.com> | 2014-11-12 18:08:00 +0100 |
|---|---|---|
| committer | Trond Myklebust <trond.myklebust@primarydata.com> | 2014-11-24 23:00:42 +0100 |
| commit | cb1410c71e0b6b2eba8c1890645a76ff86169d24 (patch) | |
| tree | 22fb6770fed66df7d51676e25e5a525bdd352abe /fs/nfs/callback_proc.c | |
| parent | pnfs/blocklayout: fix end calculation in pnfs_num_cont_bytes (diff) | |
| download | linux-cb1410c71e0b6b2eba8c1890645a76ff86169d24.tar.xz linux-cb1410c71e0b6b2eba8c1890645a76ff86169d24.zip | |
NFS: fix subtle change in COMMIT behavior
Recent work in the pgio layer made it possible for there to be more than one
request per page. This caused a subtle change in commit behavior, because
write.c:nfs_commit_unstable_pages compares the number of *pages* waiting for
writeback against the number of requests on a commit list to choose when to
send a COMMIT in a non-blocking flush.
This is probably hard to hit in normal operation - you have to be using
rsize/wsize < PAGE_SIZE, or pnfs with lots of boundaries that are not page
aligned to have a noticeable change in behavior.
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Diffstat (limited to 'fs/nfs/callback_proc.c')
| -rw-r--r-- | fs/nfs/callback_proc.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 73466b934090..e36a9d78ea49 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -49,7 +49,7 @@ __be32 nfs4_callback_getattr(struct cb_getattrargs *args, goto out_iput; res->size = i_size_read(inode); res->change_attr = delegation->change_attr; - if (nfsi->npages != 0) + if (nfsi->nrequests != 0) res->change_attr++; res->ctime = inode->i_ctime; res->mtime = inode->i_mtime; |