From: Vivek Goyal <vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
Cc: axboe-tSWWG44O7X1aa/9Udqfwiw@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
avanzini.arianna-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org,
kernel-team-b10kYP2dOMg@public.gmane.org,
hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org,
lizefan-hv44wF8Li93QT0dZR+AlfA@public.gmane.org,
cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCHSET block/for-4.3] blkcg: implement interface for the unified hierarchy
Date: Mon, 27 Jul 2015 12:12:09 -0400 [thread overview]
Message-ID: <20150727161209.GB16064@redhat.com> (raw)
In-Reply-To: <1437763434-28699-1-git-send-email-tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
On Fri, Jul 24, 2015 at 02:43:44PM -0400, Tejun Heo wrote:
> Hello,
>
> blkcg interface grew to be the biggest of all controllers and
> unfortunately most inconsistent too. The interface files are
> inconsistent with a number of cloes duplicates. Some files have
> recursive variants while others don't. There's distinction between
> normal and leaf weights which isn't intuitive and there are a lot of
> stat knobs which don't make much sense outside of debugging and expose
> too much implementation details to userland.
>
> In the unified hierarchy, everything is always hierarchical and
> internal nodes can't have tasks rendering the two structural issues
> twisting the current interface. The interface has to be updated in a
> significant anyway and this is a good chance to revamp it as a whole.
> This patchset implements blkcg interface for the unified hierarchy.
>
> * blkcg is identified by "io" instead of "blkio" on the unified
> hierarchy. Given that the whole interface is updated anyway, the
> rename shouldn't carry noticeable conversion overhead.
>
> * The original interface consisted of 27 files is replaced with the
> following three files.
>
> blkio.stat : per-blkcg stats
> blkio.weight : per-cgroup and per-cgroup-queue weight settings
> blkio.max : per-cgroup-queue bps and iops max limits
>
Hi Tejun,
I browsed though the details of above knobs and it sounds great. It is
clean and much less number of knobs and files. You got rid of all the
debug CFQ knobs which is good. I was not happy with these either. Glad
to see that all the magic about leaf weight is gone. That was really
mind bending. Knob for reset stats is gone and instead of mutiple files
for configuration now we are using single file for R/W BPS/IOPS
configuration.
I will do some basic testing and see if something pops up.
Userspace will need to understand these new files but that's the
understanding anyway that unified hierarchy is different and needs
to be handled differently.
Thanks
Vivek
WARNING: multiple messages have this Message-ID (diff)
From: Vivek Goyal <vgoyal@redhat.com>
To: Tejun Heo <tj@kernel.org>
Cc: axboe@kernel.dk, linux-kernel@vger.kernel.org,
avanzini.arianna@gmail.com, kernel-team@fb.com,
hannes@cmpxchg.org, lizefan@huawei.com, cgroups@vger.kernel.org
Subject: Re: [PATCHSET block/for-4.3] blkcg: implement interface for the unified hierarchy
Date: Mon, 27 Jul 2015 12:12:09 -0400 [thread overview]
Message-ID: <20150727161209.GB16064@redhat.com> (raw)
In-Reply-To: <1437763434-28699-1-git-send-email-tj@kernel.org>
On Fri, Jul 24, 2015 at 02:43:44PM -0400, Tejun Heo wrote:
> Hello,
>
> blkcg interface grew to be the biggest of all controllers and
> unfortunately most inconsistent too. The interface files are
> inconsistent with a number of cloes duplicates. Some files have
> recursive variants while others don't. There's distinction between
> normal and leaf weights which isn't intuitive and there are a lot of
> stat knobs which don't make much sense outside of debugging and expose
> too much implementation details to userland.
>
> In the unified hierarchy, everything is always hierarchical and
> internal nodes can't have tasks rendering the two structural issues
> twisting the current interface. The interface has to be updated in a
> significant anyway and this is a good chance to revamp it as a whole.
> This patchset implements blkcg interface for the unified hierarchy.
>
> * blkcg is identified by "io" instead of "blkio" on the unified
> hierarchy. Given that the whole interface is updated anyway, the
> rename shouldn't carry noticeable conversion overhead.
>
> * The original interface consisted of 27 files is replaced with the
> following three files.
>
> blkio.stat : per-blkcg stats
> blkio.weight : per-cgroup and per-cgroup-queue weight settings
> blkio.max : per-cgroup-queue bps and iops max limits
>
Hi Tejun,
I browsed though the details of above knobs and it sounds great. It is
clean and much less number of knobs and files. You got rid of all the
debug CFQ knobs which is good. I was not happy with these either. Glad
to see that all the magic about leaf weight is gone. That was really
mind bending. Knob for reset stats is gone and instead of mutiple files
for configuration now we are using single file for R/W BPS/IOPS
configuration.
I will do some basic testing and see if something pops up.
Userspace will need to understand these new files but that's the
understanding anyway that unified hierarchy is different and needs
to be handled differently.
Thanks
Vivek
next prev parent reply other threads:[~2015-07-27 16:12 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-24 18:43 [PATCHSET block/for-4.3] blkcg: implement interface for the unified hierarchy Tejun Heo
2015-07-24 18:43 ` Tejun Heo
[not found] ` <1437763434-28699-1-git-send-email-tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2015-07-24 18:43 ` [PATCH 01/10] cgroup: don't print subsystems for the default hierarchy Tejun Heo
2015-07-24 18:43 ` Tejun Heo
[not found] ` <1437763434-28699-2-git-send-email-tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2015-07-29 1:18 ` Zefan Li
2015-07-29 1:18 ` Zefan Li
2015-08-18 21:01 ` Tejun Heo
2015-08-18 21:01 ` Tejun Heo
2015-07-24 18:43 ` [PATCH 05/10] blkcg: rename subsystem name from blkio to io Tejun Heo
2015-07-24 18:43 ` Tejun Heo
2015-07-24 18:43 ` [PATCH 06/10] blkcg: mark existing cftypes as legacy Tejun Heo
2015-07-24 18:43 ` Tejun Heo
2015-07-27 16:12 ` Vivek Goyal [this message]
2015-07-27 16:12 ` [PATCHSET block/for-4.3] blkcg: implement interface for the unified hierarchy Vivek Goyal
[not found] ` <20150727161209.GB16064-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2015-07-27 18:13 ` Tejun Heo
2015-07-27 18:13 ` Tejun Heo
2015-07-24 18:43 ` [PATCH 02/10] cgroup: introduce cgroup_subsys->legacy_name Tejun Heo
[not found] ` <1437763434-28699-3-git-send-email-tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2015-07-28 6:39 ` Zefan Li
2015-07-28 6:39 ` Zefan Li
2015-07-28 15:23 ` Tejun Heo
2015-07-28 17:56 ` [PATCH v2 " Tejun Heo
2015-08-18 21:01 ` Tejun Heo
2015-07-29 1:18 ` [PATCH " Zefan Li
2015-07-29 1:18 ` Zefan Li
2015-07-24 18:43 ` [PATCH 03/10] blkcg: remove unnecessary NULL checks from __cfqg_set_weight_device() Tejun Heo
2015-07-24 18:43 ` [PATCH 04/10] blkcg: refine error codes returned during blkcg configuration Tejun Heo
2015-07-24 18:43 ` [PATCH 07/10] blkcg: move body parsing from blkg_conf_prep() to its callers Tejun Heo
2015-07-24 18:43 ` [PATCH 08/10] blkcg: separate out tg_conf_updated() from tg_set_conf() Tejun Heo
2015-07-24 18:43 ` [PATCH 09/10] blkcg: misc preparations for unified hierarchy interface Tejun Heo
2015-07-24 18:43 ` [PATCH 10/10] blkcg: implement interface for the unified hierarchy Tejun Heo
[not found] ` <1437763434-28699-11-git-send-email-tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2015-07-30 22:58 ` [PATCH v2 " Tejun Heo
2015-07-30 22:58 ` Tejun Heo
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=20150727161209.GB16064@redhat.com \
--to=vgoyal-h+wxahxf7alqt0dzr+alfa@public.gmane.org \
--cc=avanzini.arianna-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=axboe-tSWWG44O7X1aa/9Udqfwiw@public.gmane.org \
--cc=cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org \
--cc=kernel-team-b10kYP2dOMg@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=lizefan-hv44wF8Li93QT0dZR+AlfA@public.gmane.org \
--cc=tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.