linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Trond Myklebust <trond.myklebust@primarydata.com>
To: linux-nfs@vger.kernel.org
Subject: [PATCH v3 2/5] SUNRPC: Don't hold the transport lock when receiving backchannel data
Date: Wed, 16 Aug 2017 19:00:05 -0400	[thread overview]
Message-ID: <20170816230008.20006-3-trond.myklebust@primarydata.com> (raw)
In-Reply-To: <20170816230008.20006-2-trond.myklebust@primarydata.com>

The backchannel request has no associated task, so it is going nowhere
until we call xprt_complete_bc_request().

Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/backchannel_rqst.c | 4 ++--
 net/sunrpc/xprtsock.c         | 5 +----
 2 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c
index ac701c28f44f..c2c68a15b59d 100644
--- a/net/sunrpc/backchannel_rqst.c
+++ b/net/sunrpc/backchannel_rqst.c
@@ -171,10 +171,10 @@ int xprt_setup_bc(struct rpc_xprt *xprt, unsigned int min_reqs)
 	/*
 	 * Add the temporary list to the backchannel preallocation list
 	 */
-	spin_lock_bh(&xprt->bc_pa_lock);
+	spin_lock(&xprt->bc_pa_lock);
 	list_splice(&tmp_list, &xprt->bc_pa_list);
 	xprt_inc_alloc_count(xprt, min_reqs);
-	spin_unlock_bh(&xprt->bc_pa_lock);
+	spin_unlock(&xprt->bc_pa_lock);
 
 	dprintk("RPC:       setup backchannel transport done\n");
 	return 0;
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 04dbc7027712..a2312f14beb4 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -1389,11 +1389,9 @@ static int xs_tcp_read_callback(struct rpc_xprt *xprt,
 				container_of(xprt, struct sock_xprt, xprt);
 	struct rpc_rqst *req;
 
-	/* Look up and lock the request corresponding to the given XID */
-	spin_lock_bh(&xprt->transport_lock);
+	/* Look up the request corresponding to the given XID */
 	req = xprt_lookup_bc_request(xprt, transport->tcp_xid);
 	if (req == NULL) {
-		spin_unlock_bh(&xprt->transport_lock);
 		printk(KERN_WARNING "Callback slot table overflowed\n");
 		xprt_force_disconnect(xprt);
 		return -1;
@@ -1404,7 +1402,6 @@ static int xs_tcp_read_callback(struct rpc_xprt *xprt,
 
 	if (!(transport->tcp_flags & TCP_RCV_COPY_DATA))
 		xprt_complete_bc_request(req, transport->tcp_copied);
-	spin_unlock_bh(&xprt->transport_lock);
 
 	return 0;
 }
-- 
2.13.5


  reply	other threads:[~2017-08-16 23:00 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-08-16 23:00 [PATCH v3 0/5] RPC client latency fixes Trond Myklebust
2017-08-16 23:00 ` [PATCH v3 1/5] SUNRPC: Don't hold the transport lock across socket copy operations Trond Myklebust
2017-08-16 23:00   ` Trond Myklebust [this message]
2017-08-16 23:00     ` [PATCH v3 3/5] SUNRPC: Don't loop forever in xs_tcp_data_receive() Trond Myklebust
2017-08-16 23:00       ` [PATCH v3 4/5] SUNRPC: Cleanup xs_tcp_read_common() Trond Myklebust
2017-08-16 23:00         ` [PATCH v3 5/5] SUNRPC: Add a separate spinlock to protect the RPC request receive list Trond Myklebust
2017-08-17  1:52           ` Chuck Lever
2017-08-18 18:11           ` Chuck Lever
2017-08-18 18:26             ` Trond Myklebust

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=20170816230008.20006-3-trond.myklebust@primarydata.com \
    --to=trond.myklebust@primarydata.com \
    --cc=linux-nfs@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).