From: Tejun Heo <tj@kernel.org>
To: axboe@kernel.dk, mingo@redhat.com, rostedt@goodmis.org,
fweisbec@gmail.com, teravest@google.com, slavapestov@google.com,
ctalbott@google.com, dhsharp@google.com
Cc: linux-kernel@vger.kernel.org, winget@google.com,
namhyung@gmail.com, Chanho Park <chanho61.park@samsung.com>
Subject: Re: [PATCH RESEND 9/9] block, trace: implement ioblame - IO tracer with origin tracking
Date: Wed, 11 Jan 2012 10:08:17 -0800 [thread overview]
Message-ID: <20120111180817.GJ26832@google.com> (raw)
In-Reply-To: <20120111013212.GA6843@dhcp-172-17-108-109.mtv.corp.google.com>
On Tue, Jan 10, 2012 at 05:32:12PM -0800, Tejun Heo wrote:
> Implement ioblame, which can attribute each IO to its origin and
> export the information using a tracepoint.
>
> Operations which may eventually cause IOs and IO operations themselves
> are identified and tracked primarily by their stack traces along with
> the task and the target file (dev:ino:gen). On each IO completion,
> ioblame knows why that specific IO happened and exports the
> information via ioblame:ioblame_io tracepoint.
>
> While ioblame adds fields to a few fs and block layer objects, all
> logic is well insulated inside ioblame proper and all hooking goes
> through well defined tracepoints and doesn't add any significant
> maintenance overhead.
>
> For details, please read Documentation/trace/ioblame.txt.
>
> -v2: Namhyung pointed out that all the information available at IO
> completion can be exported via tracepoint and letting userland do
> whatever it wants to do with that would be better. Stripped out
> in-kernel statistics gathering.
>
> Now that everything is exported through tracepoint, iolog and
> counters_pipe[_pipe] are unnecessary. Removed. intents_bin too
> is removed.
>
> As data collection no longer requires polling, ioblame/intents is
> updated to generate inotify IN_MODIFY event after a new intent is
> created.
One planned update is exporting issuer and dirtier separately.
Currently dirtier, if exists, simply overrides issuer as it wasn't
useful for in-kernel statistics anyway. With that gone, I think it
makes much more sense to expose both of them.
Thanks.
--
tejun
next prev parent reply other threads:[~2012-01-11 18:08 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-01-10 18:28 [RFC PATCHSET take#2] ioblame: IO tracer with origin tracking Tejun Heo
2012-01-10 18:28 ` [PATCH 1/9] block: abstract disk iteration into disk_iter Tejun Heo
2012-01-10 18:28 ` [PATCH 2/9] block: block_bio_complete tracepoint was missing Tejun Heo
2012-01-11 17:25 ` Steven Rostedt
2012-01-11 17:30 ` Tejun Heo
2012-01-12 0:24 ` Namhyung Kim
2012-01-10 18:28 ` [PATCH 3/9] block: add @req to bio_{front|back}_merge tracepoints Tejun Heo
2012-01-10 18:28 ` [PATCH 4/9] writeback: move struct wb_writeback_work to writeback.h Tejun Heo
2012-01-10 18:28 ` [PATCH 5/9] writeback: add more tracepoints Tejun Heo
2012-01-10 18:28 ` [PATCH 6/9] block: add block_touch_buffer tracepoint Tejun Heo
2012-01-11 17:42 ` Steven Rostedt
2012-01-11 17:58 ` Tejun Heo
2012-01-10 18:28 ` [PATCH 7/9] vfs: add fcheck tracepoint Tejun Heo
2012-01-10 18:28 ` [PATCH 8/9] stacktrace: implement save_stack_trace_quick() Tejun Heo
2012-01-11 16:26 ` Frederic Weisbecker
2012-01-11 16:38 ` Tejun Heo
2012-01-11 17:37 ` Tejun Heo
2012-01-17 2:22 ` Frederic Weisbecker
2012-01-10 18:28 ` [PATCH 9/9] block, trace: implement ioblame - IO tracer with origin tracking Tejun Heo
2012-01-11 0:25 ` Chanho Park
2012-01-11 1:04 ` Tejun Heo
2012-01-11 1:32 ` [PATCH RESEND " Tejun Heo
2012-01-11 6:15 ` Namhyung Kim
2012-01-11 17:06 ` Tejun Heo
2012-01-12 1:05 ` Namhyung Kim
2012-01-12 1:14 ` Tejun Heo
2012-01-12 1:35 ` Namhyung Kim
2012-01-12 1:37 ` Tejun Heo
2012-01-12 1:40 ` Namhyung Kim
2012-01-12 1:41 ` Namhyung Kim
2012-01-12 1:44 ` Tejun Heo
2012-01-12 2:19 ` Namhyung Kim
2012-01-12 2:24 ` Tejun Heo
2012-01-11 18:08 ` Tejun Heo [this message]
2012-01-11 14:40 ` [RFC PATCHSET take#2] ioblame: " Frederic Weisbecker
2012-01-11 17:02 ` Tejun Heo
2012-01-11 22:45 ` David Sharp
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=20120111180817.GJ26832@google.com \
--to=tj@kernel.org \
--cc=axboe@kernel.dk \
--cc=chanho61.park@samsung.com \
--cc=ctalbott@google.com \
--cc=dhsharp@google.com \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=namhyung@gmail.com \
--cc=rostedt@goodmis.org \
--cc=slavapestov@google.com \
--cc=teravest@google.com \
--cc=winget@google.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