From: "J. Bruce Fields" <bfields@fieldses.org>
To: Chuck Lever <chuck.lever@oracle.com>
Cc: linux-nfs@vger.kernel.org
Subject: Re: [PATCH 15/15] nfsd: Trace NFSv4 COMPOUND execution
Date: Wed, 14 Mar 2018 16:55:15 -0400 [thread overview]
Message-ID: <20180314205515.GD7241@fieldses.org> (raw)
In-Reply-To: <20180313154459.21531.606.stgit@klimt.1015granger.net>
On Tue, Mar 13, 2018 at 11:44:59AM -0400, Chuck Lever wrote:
> This helps record the identity and timing of the ops in each NFSv4
> COMPOUND, replacing dprintk calls that did much the same thing.
Would it be better to leave it to userspace to translate an operation
number to a name?
--b.
>
> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
> ---
> fs/nfsd/nfs4proc.c | 9 +++------
> fs/nfsd/trace.h | 39 +++++++++++++++++++++++++++++++++++++++
> 2 files changed, 42 insertions(+), 6 deletions(-)
>
> diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
> index 39016b6..0df37e0 100644
> --- a/fs/nfsd/nfs4proc.c
> +++ b/fs/nfsd/nfs4proc.c
> @@ -1722,12 +1722,10 @@ static void svcxdr_init_encode(struct svc_rqst *rqstp,
> goto encode_op;
> }
>
> + trace_nfsd_compound(rqstp, args->opcnt);
> while (!status && resp->opcnt < args->opcnt) {
> op = &args->ops[resp->opcnt++];
>
> - dprintk("nfsv4 compound op #%d/%d: %d (%s)\n",
> - resp->opcnt, args->opcnt, op->opnum,
> - nfsd4_op_name(op->opnum));
> /*
> * The XDR decode routines may have pre-set op->status;
> * for example, if there is a miscellaneous XDR error
> @@ -1801,9 +1799,8 @@ static void svcxdr_init_encode(struct svc_rqst *rqstp,
> status = op->status;
> }
>
> - dprintk("nfsv4 compound op %p opcnt %d #%d: %d: status %d\n",
> - args->ops, args->opcnt, resp->opcnt, op->opnum,
> - be32_to_cpu(status));
> + trace_nfsd_compound_status(args->opcnt, resp->opcnt, status,
> + nfsd4_op_name(op->opnum));
>
> nfsd4_cstate_clear_replay(cstate);
> nfsd4_increment_op_stats(op->opnum);
> diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h
> index a8bbd9d..80933e4 100644
> --- a/fs/nfsd/trace.h
> +++ b/fs/nfsd/trace.h
> @@ -11,6 +11,45 @@
> #include <linux/tracepoint.h>
> #include "nfsfh.h"
>
> +TRACE_EVENT(nfsd_compound,
> + TP_PROTO(const struct svc_rqst *rqst,
> + u32 args_opcnt),
> + TP_ARGS(rqst, args_opcnt),
> + TP_STRUCT__entry(
> + __field(u32, xid)
> + __field(u32, args_opcnt)
> + ),
> + TP_fast_assign(
> + __entry->xid = be32_to_cpu(rqst->rq_xid);
> + __entry->args_opcnt = args_opcnt;
> + ),
> + TP_printk("xid=0x%08x opcnt=%u",
> + __entry->xid, __entry->args_opcnt)
> +)
> +
> +TRACE_EVENT(nfsd_compound_status,
> + TP_PROTO(u32 args_opcnt,
> + u32 resp_opcnt,
> + __be32 status,
> + const char *name),
> + TP_ARGS(args_opcnt, resp_opcnt, status, name),
> + TP_STRUCT__entry(
> + __field(u32, args_opcnt)
> + __field(u32, resp_opcnt)
> + __field(int, status)
> + __string(name, name)
> + ),
> + TP_fast_assign(
> + __entry->args_opcnt = args_opcnt;
> + __entry->resp_opcnt = resp_opcnt;
> + __entry->status = be32_to_cpu(status);
> + __assign_str(name, name);
> + ),
> + TP_printk("op=%u/%u %s status=%d",
> + __entry->resp_opcnt, __entry->args_opcnt,
> + __get_str(name), __entry->status)
> +)
> +
> DECLARE_EVENT_CLASS(nfsd_io_class,
> TP_PROTO(struct svc_rqst *rqstp,
> struct svc_fh *fhp,
next prev parent reply other threads:[~2018-03-14 20:55 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-13 15:43 [PATCH 00/15] NFSD trace point clean up Chuck Lever
2018-03-13 15:43 ` [PATCH 01/15] sunrpc: Remove unneeded pointer dereference Chuck Lever
2018-03-13 15:43 ` [PATCH 02/15] svc: Simplify ->xpo_secure_port Chuck Lever
2018-03-13 15:43 ` [PATCH 03/15] sunrpc: Update show_svc_xprt_flags() to include recently added flags Chuck Lever
2018-03-13 15:44 ` [PATCH 04/15] sunrpc: Move trace_svc_xprt_dequeue() Chuck Lever
2018-03-13 15:44 ` [PATCH 05/15] sunrpc: Simplify do_enqueue tracing Chuck Lever
2018-03-13 15:44 ` [PATCH 06/15] sunrpc: Simplify trace_svc_recv Chuck Lever
2018-03-13 15:44 ` [PATCH 07/15] sunrpc: Save remote presentation address in svc_xprt for trace events Chuck Lever
2018-03-14 20:35 ` J. Bruce Fields
2018-03-15 0:54 ` Chuck Lever
2018-03-15 1:43 ` Bruce Fields
2018-03-13 15:44 ` [PATCH 08/15] sunrpc: Re-purpose trace_svc_process Chuck Lever
2018-03-13 15:44 ` [PATCH 09/15] sunrpc: Report per-RPC execution stats Chuck Lever
2018-03-14 20:50 ` J. Bruce Fields
2018-03-15 1:11 ` Chuck Lever
2018-03-22 20:32 ` Chuck Lever
2018-03-22 21:01 ` Bruce Fields
2018-03-22 21:18 ` Chuck Lever
2018-03-13 15:44 ` [PATCH 10/15] nfsd: Fix NFSD trace points Chuck Lever
2018-03-13 15:44 ` [PATCH 11/15] nfsd: Record request byte count, not count of vectors Chuck Lever
2018-03-14 20:52 ` J. Bruce Fields
2018-03-15 1:18 ` Chuck Lever
2018-03-13 15:44 ` [PATCH 12/15] nfsd: Add "nfsd_" to trace point names Chuck Lever
2018-03-13 15:44 ` [PATCH 13/15] nfsd: Add I/O trace points in the NFSv4 write path Chuck Lever
2018-03-13 15:44 ` [PATCH 14/15] nfsd: Add I/O trace points in the NFSv4 read proc Chuck Lever
2018-03-13 15:44 ` [PATCH 15/15] nfsd: Trace NFSv4 COMPOUND execution Chuck Lever
2018-03-14 20:55 ` J. Bruce Fields [this message]
2018-03-15 1:29 ` Chuck Lever
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=20180314205515.GD7241@fieldses.org \
--to=bfields@fieldses.org \
--cc=chuck.lever@oracle.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).