Return-Path: Received: from mail-ig0-f196.google.com ([209.85.213.196]:33425 "EHLO mail-ig0-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933412AbcDYTVb (ORCPT ); Mon, 25 Apr 2016 15:21:31 -0400 Subject: [PATCH v2 05/18] xprtrdma: Avoid using Write list for small NFS READ requests From: Chuck Lever To: linux-rdma@vger.kernel.org, linux-nfs@vger.kernel.org Date: Mon, 25 Apr 2016 15:21:29 -0400 Message-ID: <20160425192129.3566.16808.stgit@manet.1015granger.net> In-Reply-To: <20160425185956.3566.64142.stgit@manet.1015granger.net> References: <20160425185956.3566.64142.stgit@manet.1015granger.net> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Sender: linux-nfs-owner@vger.kernel.org List-ID: Avoid the latency and interrupt overhead of registering a Write chunk when handling NFS READ requests of a few hundred bytes or less. This change does not interoperate with Linux NFS/RDMA servers that do not have commit 9d11b51ce7c1 ('svcrdma: Fix send_reply() scatter/gather set-up'). Commit 9d11b51ce7c1 was introduced in v4.3, and is included in 4.2.y, 4.1.y, and 3.18.y. Oracle bug 22925946 has been filed to request that the above fix be included in the Oracle Linux UEK4 NFS/RDMA server. Red Hat bugzillas 1327280 and 1327554 have been filed to request that RHEL NFS/RDMA server backports include the above fix. Workaround: Replace the "proto=rdma,port=20049" mount options with "proto=tcp" until commit 9d11b51ce7c1 is applied to your NFS server. Signed-off-by: Chuck Lever --- net/sunrpc/xprtrdma/rpc_rdma.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c index 205b81b..0105e65 100644 --- a/net/sunrpc/xprtrdma/rpc_rdma.c +++ b/net/sunrpc/xprtrdma/rpc_rdma.c @@ -539,15 +539,16 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst) /* * Chunks needed for results? * - * o Read ops return data as write chunk(s), header as inline. * o If the expected result is under the inline threshold, all ops * return as inline. + * o Large read ops return data as write chunk(s), header as + * inline. * o Large non-read ops return as a single reply chunk. */ - if (rqst->rq_rcv_buf.flags & XDRBUF_READ) - wtype = rpcrdma_writech; - else if (rpcrdma_results_inline(r_xprt, rqst)) + if (rpcrdma_results_inline(r_xprt, rqst)) wtype = rpcrdma_noch; + else if (rqst->rq_rcv_buf.flags & XDRBUF_READ) + wtype = rpcrdma_writech; else wtype = rpcrdma_replych;