From: Chuck Lever <chuck.lever@oracle.com>
To: Trond Myklebust <trond.myklebust@hammerspace.com>,
Anna Schumaker <anna.schumaker@netapp.com>
Cc: Linux NFS Mailing List <linux-nfs@vger.kernel.org>
Subject: Re: [PATCH v1 00/22] SUNRPC: Replace dprintk calls with tracepoints
Date: Mon, 17 Aug 2020 13:25:38 -0400 [thread overview]
Message-ID: <242E6FB6-6C7F-4529-A4B4-B8248152F448@oracle.com> (raw)
In-Reply-To: <20200708200121.22129.92375.stgit@manet.1015granger.net>
> On Jul 8, 2020, at 4:08 PM, Chuck Lever <chuck.lever@oracle.com> wrote:
>
> Hi-
>
> This series replaces many client-side RPC dprintk call sites with
> tracepoints. The goals of this series are:
>
> - Replace chatty dprintk call sites with tracepoints, which can
> handle a higher event rate, and won't get rate-limited.
>
> - At some later point, expand the 0-64K range of RPC task IDs.
> Task IDs would be displayed only by tracepoints as 32-bit unsigned
> integers.
>
> - Eliminate redundant tracepoints in the transport implementations.
None of the patches in this series made it into v5.9. I don't recall
seeing any feedback on these either. Can they be considered for v5.10?
> ---
>
> Chuck Lever (22):
> SUNRPC: Remove trace_xprt_complete_rqst()
> SUNRPC: Hoist trace_xprtrdma_op_allocate into generic code
> SUNRPC: Remove debugging instrumentation from xprt_release
> SUNRPC: Update debugging instrumentation in xprt_do_reserve()
> SUNRPC: Replace dprintk() call site in xprt_prepare_transmit
> SUNRPC: Replace dprintk() call site in xs_nospace()
> SUNRPC: Remove the dprint_status() macro
> SUNRPC: Remove dprintk call site in call_start()
> SUNRPC: Replace connect dprintk call sites with a tracepoint
> SUNRPC: Mitigate cond_resched() in xprt_transmit()
> SUNRPC: Add trace_rpc_timeout_status()
> SUNRPC: Trace call_refresh events
> SUNRPC: Remove dprintk call site in call_decode
> SUNRPC: Clean up call_bind_status() observability
> SUNRPC: Remove rpcb_getport_async dprintk call sites
> SUNRPC: Hoist trace_xprtrdma_op_setport into generic code
> SUNRPC: Remove dprintk call sites in rpcbind XDR functions
> SUNRPC: Remove more dprintks in rpcb_clnt.c
> SUNRPC: Replace rpcbind dprintk call sites with tracepoints
> SUNRPC: Clean up RPC scheduler tracepoints
> SUNRPC: Remove dprintk call sites in RPC queuing functions
> SUNRPC: Remove remaining dprintks from sched.c
>
>
> include/trace/events/rpcrdma.h | 63 -------
> include/trace/events/sunrpc.h | 285 ++++++++++++++++++++++++++++----
> net/sunrpc/clnt.c | 75 ++-------
> net/sunrpc/rpcb_clnt.c | 129 +++------------
> net/sunrpc/sched.c | 52 +-----
> net/sunrpc/xprt.c | 22 +--
> net/sunrpc/xprtrdma/transport.c | 7 -
> net/sunrpc/xprtsock.c | 5 +-
> 8 files changed, 304 insertions(+), 334 deletions(-)
>
> --
> Chuck Lever
--
Chuck Lever
prev parent reply other threads:[~2020-08-17 17:26 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-08 20:08 [PATCH v1 00/22] SUNRPC: Replace dprintk calls with tracepoints Chuck Lever
2020-07-08 20:09 ` [PATCH v1 01/22] SUNRPC: Remove trace_xprt_complete_rqst() Chuck Lever
2020-07-08 20:09 ` [PATCH v1 02/22] SUNRPC: Hoist trace_xprtrdma_op_allocate into generic code Chuck Lever
2020-07-08 20:09 ` [PATCH v1 03/22] SUNRPC: Remove debugging instrumentation from xprt_release Chuck Lever
2020-07-08 20:09 ` [PATCH v1 04/22] SUNRPC: Update debugging instrumentation in xprt_do_reserve() Chuck Lever
2020-07-08 20:09 ` [PATCH v1 05/22] SUNRPC: Replace dprintk() call site in xprt_prepare_transmit Chuck Lever
2020-07-08 20:09 ` [PATCH v1 06/22] SUNRPC: Replace dprintk() call site in xs_nospace() Chuck Lever
2020-07-08 20:09 ` [PATCH v1 07/22] SUNRPC: Remove the dprint_status() macro Chuck Lever
2020-07-08 20:09 ` [PATCH v1 08/22] SUNRPC: Remove dprintk call site in call_start() Chuck Lever
2020-07-08 20:09 ` [PATCH v1 09/22] SUNRPC: Replace connect dprintk call sites with a tracepoint Chuck Lever
2020-07-08 20:09 ` [PATCH v1 10/22] SUNRPC: Mitigate cond_resched() in xprt_transmit() Chuck Lever
2020-07-08 20:09 ` [PATCH v1 11/22] SUNRPC: Add trace_rpc_timeout_status() Chuck Lever
2020-07-08 20:10 ` [PATCH v1 12/22] SUNRPC: Trace call_refresh events Chuck Lever
2020-07-08 20:10 ` [PATCH v1 13/22] SUNRPC: Remove dprintk call site in call_decode Chuck Lever
2020-07-08 20:10 ` [PATCH v1 14/22] SUNRPC: Clean up call_bind_status() observability Chuck Lever
2020-07-08 20:10 ` [PATCH v1 15/22] SUNRPC: Remove rpcb_getport_async dprintk call sites Chuck Lever
2020-07-08 20:10 ` [PATCH v1 16/22] SUNRPC: Hoist trace_xprtrdma_op_setport into generic code Chuck Lever
2020-07-08 20:10 ` [PATCH v1 17/22] SUNRPC: Remove dprintk call sites in rpcbind XDR functions Chuck Lever
2020-07-08 20:10 ` [PATCH v1 18/22] SUNRPC: Remove more dprintks in rpcb_clnt.c Chuck Lever
2020-07-08 20:10 ` [PATCH v1 19/22] SUNRPC: Replace rpcbind dprintk call sites with tracepoints Chuck Lever
2020-07-08 20:10 ` [PATCH v1 20/22] SUNRPC: Clean up RPC scheduler tracepoints Chuck Lever
2020-07-08 20:10 ` [PATCH v1 21/22] SUNRPC: Remove dprintk call sites in RPC queuing functions Chuck Lever
2020-07-08 20:10 ` [PATCH v1 22/22] SUNRPC: Remove remaining dprintks from sched.c Chuck Lever
2020-08-17 17:25 ` Chuck Lever [this message]
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=242E6FB6-6C7F-4529-A4B4-B8248152F448@oracle.com \
--to=chuck.lever@oracle.com \
--cc=anna.schumaker@netapp.com \
--cc=linux-nfs@vger.kernel.org \
--cc=trond.myklebust@hammerspace.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).