diff options
| author | Chuck Lever <chuck.lever@oracle.com> | 2018-12-19 10:58:40 -0500 |
|---|---|---|
| committer | Anna Schumaker <Anna.Schumaker@Netapp.com> | 2019-01-02 12:05:16 -0500 |
| commit | 0c0829bcf51aef713806e49b8ea2bac7962f54e2 (patch) | |
| tree | 88b2ad9aab6ed364301bc5c2dbcf804de550423c /net/sunrpc/xprtrdma/backchannel.c | |
| parent | 3d433ad812baad45fa697f1af45a651147360712 (diff) | |
xprtrdma: Don't wake pending tasks until disconnect is done
Transport disconnect processing does a "wake pending tasks" at
various points.
Suppose an RPC Reply is being processed. The RPC task that Reply
goes with is waiting on the pending queue. If a disconnect wake-up
happens before reply processing is done, that reply, even if it is
good, is thrown away, and the RPC has to be sent again.
This window apparently does not exist for socket transports because
there is a lock held while a reply is being received which prevents
the wake-up call until after reply processing is done.
To resolve this, all RPC replies being processed on an RPC-over-RDMA
transport have to complete before pending tasks are awoken due to a
transport disconnect.
Callers that already hold the transport write lock may invoke
->ops->close directly. Others use a generic helper that schedules
a close when the write lock can be taken safely.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Diffstat (limited to 'net/sunrpc/xprtrdma/backchannel.c')
| -rw-r--r-- | net/sunrpc/xprtrdma/backchannel.c | 13 |
1 files changed, 7 insertions, 6 deletions
diff --git a/net/sunrpc/xprtrdma/backchannel.c b/net/sunrpc/xprtrdma/backchannel.c index 2cb07a313b3d..79a55fc540a6 100644 --- a/net/sunrpc/xprtrdma/backchannel.c +++ b/net/sunrpc/xprtrdma/backchannel.c @@ -193,14 +193,15 @@ static int rpcrdma_bc_marshal_reply(struct rpc_rqst *rqst) */ int xprt_rdma_bc_send_reply(struct rpc_rqst *rqst) { - struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_xprt); + struct rpc_xprt *xprt = rqst->rq_xprt; + struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt); struct rpcrdma_req *req = rpcr_to_rdmar(rqst); int rc; - if (!xprt_connected(rqst->rq_xprt)) - goto drop_connection; + if (!xprt_connected(xprt)) + return -ENOTCONN; - if (!xprt_request_get_cong(rqst->rq_xprt, rqst)) + if (!xprt_request_get_cong(xprt, rqst)) return -EBADSLT; rc = rpcrdma_bc_marshal_reply(rqst); @@ -215,7 +216,7 @@ failed_marshal: if (rc != -ENOTCONN) return rc; drop_connection: - xprt_disconnect_done(rqst->rq_xprt); + xprt_rdma_close(xprt); return -ENOTCONN; } @@ -338,7 +339,7 @@ void rpcrdma_bc_receive_call(struct rpcrdma_xprt *r_xprt, out_overflow: pr_warn("RPC/RDMA backchannel overflow\n"); - xprt_disconnect_done(xprt); + xprt_force_disconnect(xprt); /* This receive buffer gets reposted automatically * when the connection is re-established. */ |