NFS: Reclaim writes via writepage are opportunistic
authorTrond Myklebust <trond.myklebust@primarydata.com>
Mon, 16 May 2016 21:42:45 +0000 (17:42 -0400)
committerAnna Schumaker <Anna.Schumaker@Netapp.com>
Tue, 17 May 2016 19:48:07 +0000 (15:48 -0400)
No need to make them a priority any more, or to make them succeed.

Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
fs/nfs/write.c

index e18a97aab275854169e1b78c6bdbd97efec6edf3..4dac51ba1f7ea858eead8a1d002dce1d5bfaff39 100644 (file)
@@ -245,8 +245,7 @@ static void nfs_mark_uptodate(struct nfs_page *req)
 static int wb_priority(struct writeback_control *wbc)
 {
        int ret = 0;
-       if (wbc->for_reclaim)
-               return FLUSH_HIGHPRI | FLUSH_COND_STABLE;
+
        if (wbc->sync_mode == WB_SYNC_ALL)
                ret = FLUSH_COND_STABLE;
        return ret;