[PATCH v2] nfs/pnfs: fix nfs_direct_req ref leak when i/o falls back to the mds

[Scott Mayhew <smayhew@redhat.com>]

Currently when falling back to doing I/O through the MDS (via
pnfs_{read|write}_through_mds), the client frees the nfs_pgio_header
without releasing the reference taken on the dreq
via pnfs_generic_pg_{read|write}pages -> nfs_pgheader_init ->
nfs_direct_pgio_init.  It then takes another reference on the dreq via
nfs_generic_pg_pgios -> nfs_pgheader_init -> nfs_direct_pgio_init and
as a result the requester will become stuck in inode_dio_wait.  Once
that happens, other processes accessing the inode will become stuck as
well.

Ensure that pnfs_read_through_mds() and pnfs_write_through_mds() clean
up correctly by calling hdr->completion_ops->completion() instead of
calling hdr->release() directly.

This can be reproduced (sometimes) by performing "storage failover
takeover" commands on NetApp filer while doing direct I/O from a client.

This can also be reproduced using SystemTap to simulate a failure while
doing direct I/O from a client (from Dave Wysochanski
<dwysocha@redhat.com>):

stap -v -g -e 'probe module("nfs_layout_nfsv41_files").function("nfs4_fl_prepare_ds").return { $return=NULL; exit(); }'

Suggested-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Scott Mayhew <smayhew@redhat.com>
---
 fs/nfs/pnfs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index d602fe9..eb098cc 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -2255,7 +2255,7 @@ pnfs_write_through_mds(struct nfs_pageio_descriptor *desc,
 		nfs_pageio_reset_write_mds(desc);
 		mirror->pg_recoalesce = 1;
 	}
-	hdr->release(hdr);
+	hdr->completion_ops->completion(hdr);
 }
 
 static enum pnfs_try_status
@@ -2378,7 +2378,7 @@ pnfs_read_through_mds(struct nfs_pageio_descriptor *desc,
 		nfs_pageio_reset_read_mds(desc);
 		mirror->pg_recoalesce = 1;
 	}
-	hdr->release(hdr);
+	hdr->completion_ops->completion(hdr);
 }
 
 /*
-- 
2.9.5