From: "Wanghui (John)" <john.wanghui@huawei.com>
To: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>,
<Intel-gfx@lists.freedesktop.org>
Cc: <dri-devel@lists.freedesktop.org>, <linux-kernel@vger.kernel.org>,
Tvrtko Ursulin <tvrtko.ursulin@intel.com>,
Ingo Molnar <mingo@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Juri Lelli <juri.lelli@redhat.com>,
Vincent Guittot <vincent.guittot@linaro.org>
Subject: Re: [RFC 1/8] sched: Add nice value change notifier
Date: Wed, 6 Oct 2021 12:10:19 +0800 [thread overview]
Message-ID: <562d45e1-4a27-3252-f615-3ab1ef531f2b@huawei.com> (raw)
In-Reply-To: <20211004143650.699120-2-tvrtko.ursulin@linux.intel.com>
HI Tvrtko
On 2021/10/4 22:36, Tvrtko Ursulin wrote:
> void set_user_nice(struct task_struct *p, long nice)
> {
> bool queued, running;
> - int old_prio;
> + int old_prio, ret;
> struct rq_flags rf;
> struct rq *rq;
>
> @@ -6915,6 +6947,9 @@ void set_user_nice(struct task_struct *p, long nice)
>
> out_unlock:
> task_rq_unlock(rq, p, &rf);
> +
> + ret = atomic_notifier_call_chain(&user_nice_notifier_list, nice, p);
> + WARN_ON_ONCE(ret != NOTIFY_DONE);
> }
How about adding a new "io_nice" to task_struct，and move the call chain to
sched_setattr/getattr, there are two benefits:
1. Decoupled with fair scheduelr. In our use case, high priority tasks often
use rt scheduler.
2. The range of value don't need to be bound to -20~19 or 0~139
next prev parent reply other threads:[~2021-10-06 4:10 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-04 14:36 [RFC v2 0/8] CPU + GPU synchronised priority scheduling Tvrtko Ursulin
2021-10-04 14:36 ` [RFC 1/8] sched: Add nice value change notifier Tvrtko Ursulin
2021-10-06 4:10 ` Wanghui (John) [this message]
2021-10-06 7:58 ` Barry Song
2021-10-06 13:44 ` Tvrtko Ursulin
2021-10-06 20:21 ` Barry Song
2021-10-07 8:50 ` Tvrtko Ursulin
2021-10-07 9:09 ` Tvrtko Ursulin
2021-10-07 10:00 ` Barry Song
2021-10-04 14:36 ` [RFC 2/8] drm/i915: Explicitly track DRM clients Tvrtko Ursulin
2021-10-04 14:36 ` [RFC 3/8] drm/i915: Make GEM contexts " Tvrtko Ursulin
2021-10-04 14:36 ` [RFC 4/8] drm/i915: Track all user contexts per client Tvrtko Ursulin
2021-10-04 14:36 ` [RFC 5/8] drm/i915: Keep track of registered clients indexed by task struct Tvrtko Ursulin
2021-10-04 14:36 ` [RFC 6/8] drm/i915: Make some recently added vfuncs use full scheduling attribute Tvrtko Ursulin
2021-10-06 17:12 ` Matthew Brost
2021-10-06 19:06 ` Tvrtko Ursulin
2021-10-13 12:01 ` [Intel-gfx] " Daniel Vetter
2021-10-13 15:50 ` Tvrtko Ursulin
2021-10-04 14:36 ` [RFC 7/8] drm/i915: Inherit process nice for context scheduling priority Tvrtko Ursulin
2021-10-06 17:16 ` [Intel-gfx] " Matthew Brost
2021-10-06 17:24 ` Matthew Brost
2021-10-06 18:42 ` Tvrtko Ursulin
2021-10-04 14:36 ` [RFC 8/8] drm/i915: Connect with the process nice change notifier Tvrtko Ursulin
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=562d45e1-4a27-3252-f615-3ab1ef531f2b@huawei.com \
--to=john.wanghui@huawei.com \
--cc=Intel-gfx@lists.freedesktop.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=tvrtko.ursulin@intel.com \
--cc=tvrtko.ursulin@linux.intel.com \
--cc=vincent.guittot@linaro.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).