From: Roman Gushchin <guro@fb.com>
To: Hao Luo <haoluo@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@redhat.com>,
Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Andrii Nakryiko <andrii@kernel.org>,
Rik van Riel <riel@surriel.com>,
Mel Gorman <mgorman@techsingularity.net>,
Vincent Guittot <vincent.guittot@linaro.org>,
Valentin Schneider <Valentin.Schneider@arm.com>,
Will Deacon <will@kernel.org>,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Barret Rhoden <brho@google.com>, <bpf@vger.kernel.org>,
<linux-kernel@vger.kernel.org>
Subject: Re: [PATCH rfc 0/6] Scheduler BPF
Date: Wed, 15 Sep 2021 18:42:32 -0700 [thread overview]
Message-ID: <YUKhCHarl+Ng1xxR@carbon.dhcp.thefacebook.com> (raw)
In-Reply-To: <CA+khW7i460ey-UFzpMSJ8AP9QeD8ufa4FzLA4PQckNP00ShQSw@mail.gmail.com>
On Wed, Sep 15, 2021 at 05:19:03PM -0700, Hao Luo wrote:
> Hi Roman,
>
> On Wed, Sep 15, 2021 at 3:04 PM Roman Gushchin <guro@fb.com> wrote:
> >
Hi Hao!
>
> Thanks for initiating the effort of bringing BPF to sched. I've been
> looking at the potential applications of BPF in sched for some time
> and I'm very excited about this work!
>
> My current focus has been using BPF for profiling performance and
> exporting sched related stats. I think BPF can provide a great help
> there. We have many users in Google that want the kernel to export
> various scheduling metrics to userspace. I think BPF is a good fit for
> such a task. So one of my recent attempts is to use BPF to account for
> the forced idle time caused by core scheduling [1]. This is one of the
> topics I want to discuss in my upcoming LPC BPF talk [2].
I guess for profiling we don't necessarily need a dedicated program type
etc, but it might be convenient, and some helpers can be useful too.
Unfortunately I won't be able to attend your talk, but hopefully I can
see it in a record later. I'm very interested.
>
> Looking forward, I agree that BPF has a great potential in customizing
> policies in the scheduler. It has the advantage of quick
> experimentation and deployment. One of the use cases I'm thinking of
> is to customize load balancing policies. For example, allow using BPF
> to influence whether a task can migrate (can_migrate_task). This is
> currently only an idea.
>
> > Our very first experiments with using BPF in CFS look very promising. We're
> > at a very early stage, however already have seen a nice latency and ~1% RPS
> > wins for our (Facebook's) main web workload.
> >
> > As I know, Google is working on a more radical approach [2]: they aim to move
> > the scheduling code into userspace. It seems that their core motivation is
> > somewhat similar: to make the scheduler changes easier to develop, validate
> > and deploy. Even though their approach is different, they also use BPF for
> > speeding up some hot paths. I think the suggested infrastructure can serve
> > their purpose too.
>
> Yes. Barret can talk more about this, but I think it summarized the
> work of ghOSt [3] and the use of BPF in ghOSt well.
I took a brief look over how you use BPF in ghOSt and I think what I suggest
will work for you as well. I'd appreciate any comments/feedback whether it's
definitely true.
Thank you!
Roman
next prev parent reply other threads:[~2021-09-16 1:43 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20210915213550.3696532-1-guro@fb.com>
2021-09-16 0:19 ` [PATCH rfc 0/6] Scheduler BPF Hao Luo
2021-09-16 1:42 ` Roman Gushchin [this message]
2021-09-16 16:24 ` Roman Gushchin
2021-09-16 16:24 ` [PATCH rfc 1/6] bpf: sched: basic infrastructure for scheduler bpf Roman Gushchin
2021-09-16 16:24 ` [PATCH rfc 2/6] bpf: sched: add convenient helpers to identify sched entities Roman Gushchin
2021-11-25 6:09 ` Yafang Shao
2021-11-26 19:50 ` Roman Gushchin
2021-09-16 16:24 ` [PATCH rfc 3/6] bpf: sched: introduce bpf_sched_enable() Roman Gushchin
2021-09-16 16:24 ` [PATCH rfc 4/6] sched: cfs: add bpf hooks to control wakeup and tick preemption Roman Gushchin
2021-10-01 3:35 ` Barry Song
2021-10-02 0:13 ` Roman Gushchin
2021-09-16 16:24 ` [PATCH rfc 5/6] libbpf: add support for scheduler bpf programs Roman Gushchin
2021-09-16 16:24 ` [PATCH rfc 6/6] bpftool: recognize scheduler programs Roman Gushchin
2021-09-16 16:36 ` [PATCH rfc 0/6] Scheduler BPF Roman Gushchin
2021-10-06 16:39 ` Qais Yousef
2021-10-06 18:50 ` Roman Gushchin
2021-10-11 16:38 ` Qais Yousef
2021-10-11 18:09 ` Roman Gushchin
2021-10-12 10:16 ` Qais Yousef
[not found] ` <52EC1E80-4C89-43AD-8A59-8ACA184EAE53@gmail.com>
2021-11-25 6:00 ` Yafang Shao
2021-11-26 19:46 ` Roman Gushchin
2022-01-15 8:29 ` Huichun Feng
2022-01-18 22:54 ` Roman Gushchin
2022-07-19 13:05 ` Ren Zhijie
2022-07-19 13:17 ` Ren Zhijie
2022-07-19 23:21 ` Roman Gushchin
2021-11-20 16:41 Hui-Chun Feng
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=YUKhCHarl+Ng1xxR@carbon.dhcp.thefacebook.com \
--to=guro@fb.com \
--cc=Valentin.Schneider@arm.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=brho@google.com \
--cc=daniel@iogearbox.net \
--cc=dietmar.eggemann@arm.com \
--cc=haoluo@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@techsingularity.net \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=riel@surriel.com \
--cc=vincent.guittot@linaro.org \
--cc=will@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).