xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

SUNRPC: Move xpt_mutex into socket xpo_sendto methods

Browse Source 
It appears that the RPC/RDMA transport does not need serialization
of calls to its xpo_sendto method. Move the mutex into the socket
methods that still need that serialization.

Tail latencies are unambiguously better with this patch applied.
fio randrw 8KB 70/30 on NFSv3, smaller numbers are better:

    clat percentiles (usec):

With xpt_mutex:
r    | 99.99th=[ 8848]
w    | 99.99th=[ 9634]

Without xpt_mutex:
r    | 99.99th=[ 8586]
w    | 99.99th=[ 8979]

Serializing the construction of RPC/RDMA transport headers is not
really necessary at this point, because the Linux NFS server
implementation never changes its credit grant on a connection. If
that should change, then svc_rdma_sendto will need to serialize
access to the transport's credit grant fields.

Reported-by: kbuild test robot <lkp@intel.com>
[ cel: fix uninitialized variable warning ]
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
This commit is contained in:
Chuck Lever
2020-05-02 10:37:44 -04:00
parent b9bbe6ed63
commit ca4faf543a
6 changed files with 64 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
net/sunrpc/svcsock.c
View File

@@ -506,6 +506,9 @@ out_free:
* svc_udp_sendto - Send out a reply on a UDP socket
* @rqstp: completed svc_rqst
*
* xpt_mutex ensures @rqstp's whole message is written to the socket
* without interruption.
*
* Returns the number of bytes sent, or a negative errno.
*/
static int svc_udp_sendto(struct svc_rqst *rqstp)
@@ -531,6 +534,11 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
svc_set_cmsg_data(rqstp, cmh);
mutex_lock(&xprt->xpt_mutex);
if (svc_xprt_is_dead(xprt))
goto out_notconn;
err = xprt_sock_sendmsg(svsk->sk_sock, &msg, xdr, 0, 0, &sent);
xdr_free_bvec(xdr);
if (err == -ECONNREFUSED) {
@@ -538,9 +546,15 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
err = xprt_sock_sendmsg(svsk->sk_sock, &msg, xdr, 0, 0, &sent);
xdr_free_bvec(xdr);
}
mutex_unlock(&xprt->xpt_mutex);
if (err < 0)
return err;
return sent;
out_notconn:
mutex_unlock(&xprt->xpt_mutex);
return -ENOTCONN;
}
static int svc_udp_has_wspace(struct svc_xprt *xprt)
@@ -1063,6 +1077,9 @@ err_noclose:
* svc_tcp_sendto - Send out a reply on a TCP socket
* @rqstp: completed svc_rqst
*
* xpt_mutex ensures @rqstp's whole message is written to the socket
* without interruption.
*
* Returns the number of bytes sent, or a negative errno.
*/
static int svc_tcp_sendto(struct svc_rqst *rqstp)
@@ -1080,12 +1097,19 @@ static int svc_tcp_sendto(struct svc_rqst *rqstp)
svc_release_skb(rqstp);
mutex_lock(&xprt->xpt_mutex);
if (svc_xprt_is_dead(xprt))
goto out_notconn;
err = xprt_sock_sendmsg(svsk->sk_sock, &msg, xdr, 0, marker, &sent);
xdr_free_bvec(xdr);
if (err < 0 || sent != (xdr->len + sizeof(marker)))
goto out_close;
mutex_unlock(&xprt->xpt_mutex);
return sent;
out_notconn:
mutex_unlock(&xprt->xpt_mutex);
return -ENOTCONN;
out_close:
pr_notice("rpc-srv/tcp: %s: %s %d when sending %d bytes - shutting down socket\n",
xprt->xpt_server->sv_name,
@@ -1093,6 +1117,7 @@ out_close:
(err < 0) ? err : sent, xdr->len);
set_bit(XPT_CLOSE, &xprt->xpt_flags);
svc_xprt_enqueue(xprt);
mutex_unlock(&xprt->xpt_mutex);
return -EAGAIN;
}