From: Li Zefan <lizf@cn.fujitsu.com>
To: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>, Jens Axboe <jens.axboe@oracle.com>,
Steven Rostedt <rostedt@goodmis.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 3/7] blktrace: remove blk_probe_mutex
Date: Sun, 22 Mar 2009 14:04:14 +0800 [thread overview]
Message-ID: <49C5D4DE.30002@cn.fujitsu.com> (raw)
In-Reply-To: <20090320130309.GJ30407@ghostprotocols.net>
Arnaldo Carvalho de Melo wrote:
> Em Fri, Mar 20, 2009 at 09:48:26AM +0800, Li Zefan escreveu:
>> blk_register_tracepoints() always returns 0, so make it return void, thus
>> we don't need to use blk_probe_mutex to protect blk_probes_ref.
>>
>> Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
>
> Historic reasons, when I first worked on this I found all those WARN_ONs
> in blk_register_tracepoints a bogosity that should be later fixed, but
> to reduce the patch size I never got around to submit the proper patch,
> so I don't think that the fix is what you suggests, here it is what I
> had before I removed the non-essential parts of the patch:
>
I though register_trace_xxx() will fail only when there is a name collision,
in this case WARN_ON() and return 0 should be reasonable. But I just dig into
the code and found it may allocate memory and thus might return ENOMEM, so
I guess it's better to check the return value, though it will hardly happen
in real-life..
> +static void blk_tracer_error(const char *name)
> +{
> + pr_info("blk trace: Couldn't activate tracepoint "
> + "probe to block_%s\n", name);
> +}
> +
> +#define register_trace_block(tpoint) \
> + ret = register_trace_block_##tpoint(blk_add_trace_##tpoint); \
> + if (ret) { \
> + blk_tracer_error(#tpoint); \
> + goto *exit_point; \
> + } else \
> + exit_point = &&fail_deprobe_##tpoint;
> +
> +
> +#define fail_trace_block(tpoint) \
> + fail_deprobe_##tpoint: \
> + unregister_trace_block_##tpoint(blk_add_trace_##tpoint)
> +
> +static int tracing_blk_register(void)
> +{
> + int ret;
> + void *exit_point = &&error;
> +
> + register_trace_block(rq_abort);
> + register_trace_block(rq_insert);
> + register_trace_block(rq_issue);
> + register_trace_block(rq_requeue);
> + register_trace_block(rq_complete);
> + register_trace_block(bio_bounce);
> + register_trace_block(bio_backmerge);
> + register_trace_block(bio_frontmerge);
> + register_trace_block(bio_queue);
> + register_trace_block(getrq);
> + register_trace_block(sleeprq);
> + register_trace_block(plug);
> + register_trace_block(unplug_timer);
> + register_trace_block(unplug_io);
> + register_trace_block(split);
> + register_trace_block(remap);
> +
> + return 0;
> +
> + fail_trace_block(remap);
> + fail_trace_block(split);
> + fail_trace_block(unplug_io);
> + fail_trace_block(unplug_timer);
> + fail_trace_block(plug);
> + fail_trace_block(sleeprq);
> + fail_trace_block(getrq);
> + fail_trace_block(bio_queue);
> + fail_trace_block(bio_frontmerge);
> + fail_trace_block(bio_backmerge);
> + fail_trace_block(bio_bounce);
> + fail_trace_block(rq_complete);
> + fail_trace_block(rq_requeue);
> + fail_trace_block(rq_issue);
> + fail_trace_block(rq_insert);
> + fail_trace_block(rq_abort);
> +error:
> + return ret;
> +}
> +
> +static void tracing_blk_unregister(void)
> +{
> + unregister_trace_block_remap(blk_add_trace_remap);
> + unregister_trace_block_split(blk_add_trace_split);
> + unregister_trace_block_unplug_io(blk_add_trace_unplug_io);
> + unregister_trace_block_unplug_timer(blk_add_trace_unplug_timer);
> + unregister_trace_block_plug(blk_add_trace_plug);
> + unregister_trace_block_sleeprq(blk_add_trace_sleeprq);
> + unregister_trace_block_getrq(blk_add_trace_getrq);
> + unregister_trace_block_bio_queue(blk_add_trace_bio_queue);
> + unregister_trace_block_bio_frontmerge(blk_add_trace_bio_frontmerge);
> + unregister_trace_block_bio_backmerge(blk_add_trace_bio_backmerge);
> + unregister_trace_block_bio_complete(blk_add_trace_bio_complete);
> + unregister_trace_block_bio_bounce(blk_add_trace_bio_bounce);
> + unregister_trace_block_rq_complete(blk_add_trace_rq_complete);
> + unregister_trace_block_rq_requeue(blk_add_trace_rq_requeue);
> + unregister_trace_block_rq_issue(blk_add_trace_rq_issue);
> + unregister_trace_block_rq_insert(blk_add_trace_rq_insert);
> + unregister_trace_block_rq_abort(blk_add_trace_rq_abort);
> + tracepoint_synchronize_unregister();
> +}
>
> Regards,
>
> - Arnaldo
>
>
next prev parent reply other threads:[~2009-03-22 6:04 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-03-20 1:47 [PATCH 0/7] blktrace: various cleanups and fixes Li Zefan
2009-03-20 1:47 ` [PATCH 1/7] blktrace: fix possible memory leak Li Zefan
2009-03-20 10:25 ` [tip:tracing/blktrace] " Li Zefan
2009-03-20 12:53 ` [PATCH 1/7] " Arnaldo Carvalho de Melo
2009-03-21 15:18 ` [tip:tracing/blktrace] " Li Zefan
2009-03-20 1:48 ` [PATCH 2/7] blktrace: make blk_tracer_enabled a bool flag Li Zefan
2009-03-20 8:47 ` Frederic Weisbecker
2009-03-20 10:26 ` [tip:tracing/blktrace] " Li Zefan
2009-03-20 12:58 ` [PATCH 2/7] " Arnaldo Carvalho de Melo
2009-03-21 15:18 ` [tip:tracing/blktrace] " Li Zefan
2009-03-20 1:48 ` [PATCH 3/7] blktrace: remove blk_probe_mutex Li Zefan
2009-03-20 8:50 ` Frederic Weisbecker
2009-03-20 10:26 ` [tip:tracing/blktrace] " Li Zefan
2009-03-20 13:03 ` [PATCH 3/7] " Arnaldo Carvalho de Melo
2009-03-22 6:04 ` Li Zefan [this message]
2009-03-22 15:09 ` Arnaldo Carvalho de Melo
2009-03-21 15:18 ` [tip:tracing/blktrace] " Li Zefan
2009-03-20 1:48 ` [PATCH 4/7] blktrace: don't increase blk_probes_ref if failed to setup blk trace Li Zefan
2009-03-20 10:26 ` [tip:tracing/blktrace] " Li Zefan
2009-03-20 13:04 ` [PATCH 4/7] " Arnaldo Carvalho de Melo
2009-03-21 15:18 ` [tip:tracing/blktrace] " Li Zefan
2009-03-20 1:49 ` [PATCH 5/7] blktrace: report EBUSY correctly Li Zefan
2009-03-20 10:26 ` [tip:tracing/blktrace] " Li Zefan
2009-03-20 13:10 ` [PATCH 5/7] " Arnaldo Carvalho de Melo
2009-03-21 15:18 ` Ingo Molnar
2009-03-21 15:19 ` [tip:tracing/blktrace] " Li Zefan
2009-03-20 1:49 ` [PATCH 6/7] blktrace: remove sysfs_blk_trace_enable_show/store() Li Zefan
2009-03-20 3:33 ` Li Zefan
2009-03-20 10:26 ` [tip:tracing/blktrace] " Li Zefan
2009-03-21 15:19 ` Li Zefan
2009-03-20 13:07 ` [PATCH 6/7] " Arnaldo Carvalho de Melo
2009-03-20 1:49 ` [PATCH 7/7] blktrace: avoid accessing NULL bdev->bd_disk Li Zefan
2009-03-20 2:34 ` Li Zefan
2009-03-20 10:26 ` [tip:tracing/blktrace] " Li Zefan
2009-03-21 15:19 ` Li Zefan
2009-03-20 13:08 ` [PATCH 7/7] " Arnaldo Carvalho de Melo
2009-03-20 10:20 ` [PATCH 0/7] blktrace: various cleanups and fixes Ingo Molnar
2009-03-20 11:09 ` Ingo Molnar
2009-03-23 14:48 ` Steven Rostedt
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=49C5D4DE.30002@cn.fujitsu.com \
--to=lizf@cn.fujitsu.com \
--cc=acme@redhat.com \
--cc=fweisbec@gmail.com \
--cc=jens.axboe@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=rostedt@goodmis.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