From: Chuck Lever <chucklever@gmail.com>
To: Trond Myklebust <trondmy@gmail.com>
Cc: Linux NFS Mailing List <linux-nfs@vger.kernel.org>
Subject: Re: [PATCH 05/27] SUNRPC: Avoid holding locks across the XDR encoding of the RPC message
Date: Mon, 3 Sep 2018 13:11:07 -0400 [thread overview]
Message-ID: <3CB8AF72-4F0E-4273-B91E-FB4A28124E29@gmail.com> (raw)
In-Reply-To: <20180903152936.24325-6-trond.myklebust@hammerspace.com>
> On Sep 3, 2018, at 11:29 AM, Trond Myklebust <trondmy@gmail.com> wrote:
>
> Currently, we grab the socket bit lock before we allow the message
> to be XDR encoded. That significantly slows down the transmission
> rate, since we serialise on a potentially blocking operation.
Which operation blocks, and how often?
> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
> ---
> net/sunrpc/clnt.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
> index e5ac35e803ad..66ec61347716 100644
> --- a/net/sunrpc/clnt.c
> +++ b/net/sunrpc/clnt.c
> @@ -1949,9 +1949,6 @@ call_transmit(struct rpc_task *task)
> task->tk_action = call_status;
> if (task->tk_status < 0)
> return;
> - if (!xprt_prepare_transmit(task))
> - return;
> - task->tk_action = call_transmit_status;
> /* Encode here so that rpcsec_gss can use correct sequence number. */
> if (rpc_task_need_encode(task)) {
> rpc_xdr_encode(task);
> @@ -1965,8 +1962,11 @@ call_transmit(struct rpc_task *task)
> return;
> }
> }
> + if (!xprt_prepare_transmit(task))
> + return;
> + task->tk_action = call_transmit_status;
> xprt_transmit(task);
> - if (task->tk_status < 0)
> + if (task->tk_status < 0) {
The added curly bracket seems incorrect.
> return;
> if (is_retrans)
> task->tk_client->cl_stats->rpcretrans++;
> --
> 2.17.1
>
--
Chuck Lever
chucklever@gmail.com
next prev parent reply other threads:[~2018-09-03 21:32 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-09-03 15:29 [PATCH 00/27] Convert RPC client transmission to a queued model Trond Myklebust
2018-09-03 15:29 ` [PATCH 01/27] SUNRPC: Clean up initialisation of the struct rpc_rqst Trond Myklebust
2018-09-03 15:29 ` [PATCH 02/27] SUNRPC: If there is no reply expected, bail early from call_decode Trond Myklebust
2018-09-03 15:29 ` [PATCH 03/27] SUNRPC: The transmitted message must lie in the RPCSEC window of validity Trond Myklebust
2018-09-03 15:29 ` [PATCH 04/27] SUNRPC: Simplify identification of when the message send/receive is complete Trond Myklebust
2018-09-03 15:29 ` [PATCH 05/27] SUNRPC: Avoid holding locks across the XDR encoding of the RPC message Trond Myklebust
2018-09-03 15:29 ` [PATCH 06/27] SUNRPC: Rename TCP receive-specific state variables Trond Myklebust
2018-09-03 15:29 ` [PATCH 07/27] SUNRPC: Move reset of TCP state variables into the reconnect code Trond Myklebust
2018-09-03 15:29 ` [PATCH 08/27] SUNRPC: Add socket transmit queue offset tracking Trond Myklebust
2018-09-03 15:29 ` [PATCH 09/27] SUNRPC: Simplify dealing with aborted partially transmitted messages Trond Myklebust
2018-09-03 15:29 ` [PATCH 10/27] SUNRPC: Refactor the transport request pinning Trond Myklebust
2018-09-03 15:29 ` [PATCH 11/27] SUNRPC: Add a helper to wake up a sleeping rpc_task and set its status Trond Myklebust
2018-09-03 15:29 ` [PATCH 12/27] SUNRPC: Don't wake queued RPC calls multiple times in xprt_transmit Trond Myklebust
2018-09-03 15:29 ` [PATCH 13/27] SUNRPC: Rename xprt->recv_lock to xprt->queue_lock Trond Myklebust
2018-09-03 15:29 ` [PATCH 14/27] SUNRPC: Refactor xprt_transmit() to remove the reply queue code Trond Myklebust
2018-09-03 15:29 ` [PATCH 15/27] SUNRPC: Refactor xprt_transmit() to remove wait for reply code Trond Myklebust
2018-09-03 15:29 ` [PATCH 16/27] SUNRPC: Minor cleanup for call_transmit() Trond Myklebust
2018-09-03 15:29 ` [PATCH 17/27] SUNRPC: Distinguish between the slot allocation list and receive queue Trond Myklebust
2018-09-03 15:29 ` [PATCH 18/27] NFS: Add a transmission queue for RPC requests Trond Myklebust
2018-09-03 15:29 ` [PATCH 19/27] SUNRPC: Refactor RPC call encoding Trond Myklebust
2018-09-03 15:29 ` [PATCH 20/27] SUNRPC: Treat the task and request as separate in the xprt_ops->send_request() Trond Myklebust
2018-09-03 15:29 ` [PATCH 21/27] SUNRPC: Don't reset the request 'bytes_sent' counter when releasing XPRT_LOCK Trond Myklebust
2018-09-03 15:29 ` [PATCH 22/27] SUNRPC: Simplify xprt_prepare_transmit() Trond Myklebust
2018-09-03 15:29 ` [PATCH 23/27] SUNRPC: Move RPC retransmission stat counter to xprt_transmit() Trond Myklebust
2018-09-03 15:29 ` [PATCH 24/27] SUNRPC: Fix up the back channel transmit Trond Myklebust
2018-09-03 15:29 ` [PATCH 25/27] SUNRPC: Allow calls to xprt_transmit() to drain the entire transmit queue Trond Myklebust
2018-09-03 15:29 ` [PATCH 26/27] SUNRPC: Queue the request for transmission immediately after encoding Trond Myklebust
2018-09-03 15:29 ` [PATCH 27/27] SUNRPC: Convert the xprt->sending queue back to an ordinary wait queue Trond Myklebust
2018-09-03 17:31 ` [PATCH 23/27] SUNRPC: Move RPC retransmission stat counter to xprt_transmit() Chuck Lever
2018-09-03 17:49 ` Trond Myklebust
2018-09-03 17:28 ` [PATCH 20/27] SUNRPC: Treat the task and request as separate in the xprt_ops->send_request() Chuck Lever
2018-09-03 17:47 ` Trond Myklebust
2018-09-03 17:49 ` Chuck Lever
2018-09-03 17:11 ` Chuck Lever [this message]
2018-09-03 17:40 ` [PATCH 05/27] SUNRPC: Avoid holding locks across the XDR encoding of the RPC message Trond Myklebust
2018-09-03 17:15 ` [PATCH 04/27] SUNRPC: Simplify identification of when the message send/receive is complete Chuck Lever
2018-09-04 15:46 ` [PATCH 03/27] SUNRPC: The transmitted message must lie in the RPCSEC window of validity J. Bruce Fields
2018-09-03 17:41 ` [PATCH 00/27] Convert RPC client transmission to a queued model Chuck Lever
2018-09-03 17:55 ` 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=3CB8AF72-4F0E-4273-B91E-FB4A28124E29@gmail.com \
--to=chucklever@gmail.com \
--cc=linux-nfs@vger.kernel.org \
--cc=trondmy@gmail.com \
/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).