From: Chuck Lever <chuck.lever@oracle.com>
To: linux-rdma@vger.kernel.org, linux-nfs@vger.kernel.org
Subject: [PATCH v1 08/17] svcrdma: Don't account for Receive queue "starvation"
Date: Tue, 30 May 2017 14:15:06 -0400 [thread overview]
Message-ID: <20170530181506.2992.9237.stgit@klimt.1015granger.net> (raw)
In-Reply-To: <20170530175808.2992.60365.stgit@klimt.1015granger.net>
>From what I can tell, calling ->recvfrom when there is no work to do
is a normal part of operation. This is the only way svc_recv can
tell when there is no more data ready to receive on the transport.
Neither the TCP nor the UDP transport implementations have a
"starve" metric.
The cost of receive starvation accounting is bumping an atomic, which
results in extra (IMO unnecessary) bus traffic between CPU sockets,
while holding a spin lock.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 21 ++++++---------------
1 file changed, 6 insertions(+), 15 deletions(-)
diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
index b480893..1452bd0 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
@@ -844,9 +844,9 @@ int svc_rdma_recvfrom(struct svc_rqst *rqstp)
struct svc_xprt *xprt = rqstp->rq_xprt;
struct svcxprt_rdma *rdma_xprt =
container_of(xprt, struct svcxprt_rdma, sc_xprt);
- struct svc_rdma_op_ctxt *ctxt = NULL;
+ struct svc_rdma_op_ctxt *ctxt;
struct rpcrdma_msg *rmsgp;
- int ret = 0;
+ int ret;
dprintk("svcrdma: rqstp=%p\n", rqstp);
@@ -863,21 +863,13 @@ int svc_rdma_recvfrom(struct svc_rqst *rqstp)
struct svc_rdma_op_ctxt, list);
list_del(&ctxt->list);
} else {
- atomic_inc(&rdma_stat_rq_starve);
+ /* No new incoming requests, terminate the loop */
clear_bit(XPT_DATA, &xprt->xpt_flags);
- ctxt = NULL;
+ spin_unlock(&rdma_xprt->sc_rq_dto_lock);
+ return 0;
}
spin_unlock(&rdma_xprt->sc_rq_dto_lock);
- if (!ctxt) {
- /* This is the EAGAIN path. The svc_recv routine will
- * return -EAGAIN, the nfsd thread will go to call into
- * svc_recv again and we shouldn't be on the active
- * transport list
- */
- if (test_bit(XPT_CLOSE, &xprt->xpt_flags))
- goto defer;
- goto out;
- }
+
dprintk("svcrdma: processing ctxt=%p on xprt=%p, rqstp=%p\n",
ctxt, rdma_xprt, rqstp);
atomic_inc(&rdma_stat_recv);
@@ -920,7 +912,6 @@ int svc_rdma_recvfrom(struct svc_rqst *rqstp)
+ rqstp->rq_arg.page_len
+ rqstp->rq_arg.tail[0].iov_len;
svc_rdma_put_context(ctxt, 0);
- out:
dprintk("svcrdma: ret=%d, rq_arg.len=%u, "
"rq_arg.head[0].iov_base=%p, rq_arg.head[0].iov_len=%zd\n",
ret, rqstp->rq_arg.len,
next prev parent reply other threads:[~2017-05-30 18:15 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-30 18:14 [PATCH v1 00/17] Server-side NFS/RDMA changes proposed for v4.13 Chuck Lever
2017-05-30 18:14 ` [PATCH v1 01/17] ib_core: Enable and expose force_mr module parameter Chuck Lever
2017-05-31 4:36 ` Leon Romanovsky
2017-05-30 18:14 ` [PATCH v1 02/17] svcrdma: Squelch disconnection messages Chuck Lever
2017-05-30 18:14 ` [PATCH v1 03/17] svcrdma: Avoid Send Queue overflow Chuck Lever
2017-05-30 18:14 ` [PATCH v1 04/17] svcrdma: Remove svc_rdma_marshal.c Chuck Lever
2017-05-30 18:14 ` [PATCH v1 05/17] svcrdma: Improve Read chunk sanity checking Chuck Lever
2017-05-30 18:14 ` [PATCH v1 06/17] svcrdma: Improve Write " Chuck Lever
2017-05-30 18:14 ` [PATCH v1 07/17] svcrdma: Improve Reply " Chuck Lever
2017-05-30 18:15 ` Chuck Lever [this message]
2017-05-30 18:15 ` [PATCH v1 09/17] sunrpc: Allocate one more page per svc_rqst Chuck Lever
2017-05-30 18:15 ` [PATCH v1 10/17] svcrdma: Add recvfrom helpers to svc_rdma_rw.c Chuck Lever
2017-05-30 18:15 ` [PATCH v1 11/17] svcrdma: Use generic RDMA R/W API in RPC Call path Chuck Lever
2017-05-30 18:15 ` [PATCH v1 12/17] svcrdma: Properly compute .len and .buflen for received RPC Calls Chuck Lever
2017-05-30 18:15 ` [PATCH v1 13/17] svcrdma: Remove unused Read completion handlers Chuck Lever
2017-05-30 18:15 ` [PATCH v1 14/17] svcrdma: Remove frmr cache Chuck Lever
2017-05-30 18:16 ` [PATCH v1 15/17] svcrdma: Clean-up svc_rdma_unmap_dma Chuck Lever
2017-05-30 18:16 ` [PATCH v1 16/17] svcrdma: Clean up after converting svc_rdma_recvfrom to rdma_rw API Chuck Lever
2017-05-30 18:16 ` [PATCH v1 17/17] svcrdma: use offset_in_page() macro Chuck Lever
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=20170530181506.2992.9237.stgit@klimt.1015granger.net \
--to=chuck.lever@oracle.com \
--cc=linux-nfs@vger.kernel.org \
--cc=linux-rdma@vger.kernel.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: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).