[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20200124092808.921441399@linuxfoundation.org>
Date: Fri, 24 Jan 2020 10:30:19 +0100
From: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To: linux-kernel@...r.kernel.org
Cc: Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
stable@...r.kernel.org, Chuck Lever <chuck.lever@...cle.com>,
"J. Bruce Fields" <bfields@...hat.com>
Subject: [PATCH 5.4 018/102] SUNRPC: Fix backchannel latency metrics
From: Chuck Lever <chuck.lever@...cle.com>
commit 8729aaba74626c4ebce3abf1b9e96bb62d2958ca upstream.
I noticed that for callback requests, the reported backlog latency
is always zero, and the rtt value is crazy big. The problem was that
rqst->rq_xtime is never set for backchannel requests.
Fixes: 78215759e20d ("SUNRPC: Make RTT measurement more ... ")
Signed-off-by: Chuck Lever <chuck.lever@...cle.com>
Signed-off-by: J. Bruce Fields <bfields@...hat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
---
net/sunrpc/xprtrdma/svc_rdma_backchannel.c | 1 +
net/sunrpc/xprtsock.c | 3 ++-
2 files changed, 3 insertions(+), 1 deletion(-)
--- a/net/sunrpc/xprtrdma/svc_rdma_backchannel.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_backchannel.c
@@ -195,6 +195,7 @@ rpcrdma_bc_send_request(struct svcxprt_r
pr_info("%s: %*ph\n", __func__, 64, rqst->rq_buffer);
#endif
+ rqst->rq_xtime = ktime_get();
rc = svc_rdma_bc_sendto(rdma, rqst, ctxt);
if (rc) {
svc_rdma_send_ctxt_put(rdma, ctxt);
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -2659,6 +2659,8 @@ static int bc_sendto(struct rpc_rqst *re
.iov_len = sizeof(marker),
};
+ req->rq_xtime = ktime_get();
+
len = kernel_sendmsg(transport->sock, &msg, &iov, 1, iov.iov_len);
if (len != iov.iov_len)
return -EAGAIN;
@@ -2684,7 +2686,6 @@ static int bc_send_request(struct rpc_rq
struct svc_xprt *xprt;
int len;
- dprintk("sending request with xid: %08x\n", ntohl(req->rq_xid));
/*
* Get the server socket associated with this callback xprt
*/
Powered by blists - more mailing lists