From: Chuck Lever <chuck.lever-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org> To: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org Subject: [PATCH v4 10/16] xprtrdma: Don't provide a reply chunk when expecting a short reply Date: Mon, 03 Aug 2015 13:04:08 -0400 [thread overview] Message-ID: <20150803170408.9115.58172.stgit@manet.1015granger.net> (raw) In-Reply-To: <20150803165807.9115.23842.stgit-FYjufvaPoItvLzlybtyyYzGyq/o6K9yX@public.gmane.org> Currently Linux always offers a reply chunk, even when the reply can be sent inline (ie. is smaller than 1KB). On the client, registering a memory region can be expensive. A server may choose not to use the reply chunk, wasting the cost of the registration. This is a change only for RPC replies smaller than 1KB which the server constructs in the RPC reply send buffer. Because the elements of the reply must be XDR encoded, a copy-free data transfer has no benefit in this case. Signed-off-by: Chuck Lever <chuck.lever-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org> Reviewed-by: Sagi Grimberg <sagig-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org> Tested-by: Devesh Sharma <devesh.sharma-1wcpHE2jlwO1Z/+hSey0Gg@public.gmane.org> --- net/sunrpc/xprtrdma/rpc_rdma.c | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-) diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c index e7cf976..62150ae 100644 --- a/net/sunrpc/xprtrdma/rpc_rdma.c +++ b/net/sunrpc/xprtrdma/rpc_rdma.c @@ -420,7 +420,7 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst) * * 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 (but see later). + * return as inline. * o Large non-read ops return as a single reply chunk. */ if (rqst->rq_rcv_buf.flags & XDRBUF_READ) @@ -476,17 +476,6 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst) headerp->rm_body.rm_nochunks.rm_empty[2] = xdr_zero; /* new length after pullup */ rpclen = rqst->rq_svec[0].iov_len; - /* Currently we try to not actually use read inline. - * Reply chunks have the desirable property that - * they land, packed, directly in the target buffers - * without headers, so they require no fixup. The - * additional RDMA Write op sends the same amount - * of data, streams on-the-wire and adds no overhead - * on receive. Therefore, we request a reply chunk - * for non-writes wherever feasible and efficient. - */ - if (wtype == rpcrdma_noch) - wtype = rpcrdma_replych; } if (rtype != rpcrdma_noch) { -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html
WARNING: multiple messages have this Message-ID (diff)
From: Chuck Lever <chuck.lever@oracle.com> To: linux-rdma@vger.kernel.org, linux-nfs@vger.kernel.org Subject: [PATCH v4 10/16] xprtrdma: Don't provide a reply chunk when expecting a short reply Date: Mon, 03 Aug 2015 13:04:08 -0400 [thread overview] Message-ID: <20150803170408.9115.58172.stgit@manet.1015granger.net> (raw) In-Reply-To: <20150803165807.9115.23842.stgit@manet.1015granger.net> Currently Linux always offers a reply chunk, even when the reply can be sent inline (ie. is smaller than 1KB). On the client, registering a memory region can be expensive. A server may choose not to use the reply chunk, wasting the cost of the registration. This is a change only for RPC replies smaller than 1KB which the server constructs in the RPC reply send buffer. Because the elements of the reply must be XDR encoded, a copy-free data transfer has no benefit in this case. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Reviewed-by: Sagi Grimberg <sagig@mellanox.com> Tested-by: Devesh Sharma <devesh.sharma@avagotech.com> --- net/sunrpc/xprtrdma/rpc_rdma.c | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-) diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c index e7cf976..62150ae 100644 --- a/net/sunrpc/xprtrdma/rpc_rdma.c +++ b/net/sunrpc/xprtrdma/rpc_rdma.c @@ -420,7 +420,7 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst) * * 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 (but see later). + * return as inline. * o Large non-read ops return as a single reply chunk. */ if (rqst->rq_rcv_buf.flags & XDRBUF_READ) @@ -476,17 +476,6 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst) headerp->rm_body.rm_nochunks.rm_empty[2] = xdr_zero; /* new length after pullup */ rpclen = rqst->rq_svec[0].iov_len; - /* Currently we try to not actually use read inline. - * Reply chunks have the desirable property that - * they land, packed, directly in the target buffers - * without headers, so they require no fixup. The - * additional RDMA Write op sends the same amount - * of data, streams on-the-wire and adds no overhead - * on receive. Therefore, we request a reply chunk - * for non-writes wherever feasible and efficient. - */ - if (wtype == rpcrdma_noch) - wtype = rpcrdma_replych; } if (rtype != rpcrdma_noch) {
next prev parent reply other threads:[~2015-08-03 17:04 UTC|newest] Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top 2015-08-03 17:02 [PATCH v4 00/16] NFS/RDMA client side for Linux 4.3 Chuck Lever 2015-08-03 17:02 ` Chuck Lever [not found] ` <20150803165807.9115.23842.stgit-FYjufvaPoItvLzlybtyyYzGyq/o6K9yX@public.gmane.org> 2015-08-03 17:02 ` [PATCH v4 01/16] xprtrdma: Make xprt_setup_rdma() agnostic to family of server address Chuck Lever 2015-08-03 17:02 ` Chuck Lever 2015-08-03 17:02 ` [PATCH v4 02/16] xprtrdma: Raise maximum payload size to one megabyte Chuck Lever 2015-08-03 17:02 ` Chuck Lever 2015-08-03 17:02 ` [PATCH v4 03/16] xprtrdma: Increase default credit limit Chuck Lever 2015-08-03 17:02 ` Chuck Lever 2015-08-03 17:03 ` [PATCH v4 04/16] xprtrdma: Don't fall back to PHYSICAL memory registration Chuck Lever 2015-08-03 17:03 ` Chuck Lever 2015-08-03 17:03 ` [PATCH v4 05/16] xprtrdma: Remove last ib_reg_phys_mr() call site Chuck Lever 2015-08-03 17:03 ` Chuck Lever 2015-08-03 17:03 ` [PATCH v4 06/16] xprtrdma: Clean up rpcrdma_ia_open() Chuck Lever 2015-08-03 17:03 ` Chuck Lever 2015-08-03 17:03 ` [PATCH v4 07/16] xprtrdma: Remove logic that constructs RDMA_MSGP type calls Chuck Lever 2015-08-03 17:03 ` Chuck Lever 2015-08-03 17:03 ` [PATCH v4 08/16] xprtrdma: Account for RPC/RDMA header size when deciding to inline Chuck Lever 2015-08-03 17:03 ` Chuck Lever 2015-08-03 17:03 ` [PATCH v4 09/16] xprtrdma: Always provide a write list when sending NFS READ Chuck Lever 2015-08-03 17:03 ` Chuck Lever 2015-08-03 17:04 ` Chuck Lever [this message] 2015-08-03 17:04 ` [PATCH v4 10/16] xprtrdma: Don't provide a reply chunk when expecting a short reply Chuck Lever 2015-08-03 17:04 ` [PATCH v4 11/16] xprtrdma: Fix XDR tail buffer marshalling Chuck Lever 2015-08-03 17:04 ` Chuck Lever 2015-08-03 17:04 ` [PATCH v4 12/16] xprtrdma: Fix large NFS SYMLINK calls Chuck Lever 2015-08-03 17:04 ` Chuck Lever 2015-08-03 17:04 ` [PATCH v4 13/16] xprtrdma: Clean up xprt_rdma_print_stats() Chuck Lever 2015-08-03 17:04 ` Chuck Lever 2015-08-03 17:04 ` [PATCH v4 14/16] xprtrdma: Count RDMA_NOMSG type calls Chuck Lever 2015-08-03 17:04 ` Chuck Lever 2015-08-03 17:04 ` [PATCH v4 15/16] core: Remove the ib_reg_phys_mr() and ib_rereg_phys_mr() verbs Chuck Lever 2015-08-03 17:04 ` Chuck Lever 2015-08-03 17:05 ` [PATCH v4 16/16] xprtrdma: take HCA driver refcount at client Chuck Lever 2015-08-03 17:05 ` Chuck Lever [not found] ` <20150803170504.9115.48166.stgit-FYjufvaPoItvLzlybtyyYzGyq/o6K9yX@public.gmane.org> 2015-08-03 17:07 ` Chuck Lever 2015-08-03 17:07 ` Chuck Lever 2015-08-06 14:28 ` [PATCH v4 00/16] NFS/RDMA client side for Linux 4.3 Anna Schumaker 2015-08-06 14:28 ` Anna Schumaker
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20150803170408.9115.58172.stgit@manet.1015granger.net \ --to=chuck.lever-qhclzuegtsvqt0dzr+alfa@public.gmane.org \ --cc=linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \ --cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.