bpf.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Menglong Dong <menglong8.dong@gmail.com>
To: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
	Menglong Dong <menglong.dong@linux.dev>,
	 Steven Rostedt <rostedt@goodmis.org>,
	Jiri Olsa <jolsa@kernel.org>, bpf <bpf@vger.kernel.org>,
	 Martin KaFai Lau <martin.lau@linux.dev>,
	Eduard Zingerman <eddyz87@gmail.com>,
	 LKML <linux-kernel@vger.kernel.org>,
	Network Development <netdev@vger.kernel.org>
Subject: Re: Inlining migrate_disable/enable. Was: [PATCH bpf-next v2 02/18] x86,bpf: add bpf_global_caller for global trampoline
Date: Fri, 8 Aug 2025 14:32:20 +0800	[thread overview]
Message-ID: <CADxym3ZeuoSXDS3j+Sv-Au-FbJzKjkh7TPmM619gRODuRCKzJw@mail.gmail.com> (raw)
In-Reply-To: <CAADnVQL_tMbi-xh_znjGwvvY1GkMTUm6EvOUU1x7rNPx53eePQ@mail.gmail.com>

On Fri, Aug 8, 2025 at 8:58 AM Alexei Starovoitov
<alexei.starovoitov@gmail.com> wrote:
>
[......]
> > +{
> > +    DEFINE(RQ_nr_pinned, offsetof(struct rq, nr_pinned));
>
> This part looks nice and sweet. Not sure what you were concerned about.
>
> Respin it as a proper patch targeting tip tree.
>
> And explain the motivation in commit log with detailed
> 'perf report' before/after along with 111M/s to 121M/s speed up,
>
> I suspect with my other __set_cpus_allowed_ptr() suggestion
> the speed up should be even bigger.

Much better.

Before:
fentry         :  113.030 ± 0.149M/s
fentry         :  112.501 ± 0.187M/s
fentry         :  112.828 ± 0.267M/s
fentry         :  115.287 ± 0.241M/s

After:
fentry         :  143.644 ± 0.670M/s
fentry         :  149.764 ± 0.362M/s
fentry         :  149.642 ± 0.156M/s
fentry         :  145.263 ± 0.221M/s
fentry         :  145.558 ± 0.145M/s

>
> > +    return 0;
> > +}

  parent reply	other threads:[~2025-08-08  6:32 UTC|newest]

Thread overview: 86+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-03 12:15 [PATCH bpf-next v2 00/18] bpf: tracing multi-link support Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 01/18] bpf: add function hash table for tracing-multi Menglong Dong
2025-07-04 16:07   ` kernel test robot
2025-07-15  1:55   ` Alexei Starovoitov
2025-07-15  2:37     ` Menglong Dong
2025-07-15  2:49       ` Alexei Starovoitov
2025-07-15  3:13         ` Menglong Dong
2025-07-15  9:06           ` Menglong Dong
2025-07-15 16:22             ` Alexei Starovoitov
2025-07-03 12:15 ` [PATCH bpf-next v2 02/18] x86,bpf: add bpf_global_caller for global trampoline Menglong Dong
2025-07-15  2:25   ` Alexei Starovoitov
2025-07-15  8:36     ` Menglong Dong
2025-07-15  9:30       ` Menglong Dong
2025-07-16 16:56         ` Inlining migrate_disable/enable. Was: " Alexei Starovoitov
2025-07-16 18:24           ` Peter Zijlstra
2025-07-16 22:35             ` Alexei Starovoitov
2025-07-16 22:49               ` Steven Rostedt
2025-07-16 22:50                 ` Steven Rostedt
2025-07-28  9:20               ` Menglong Dong
2025-07-31 16:15                 ` Alexei Starovoitov
2025-08-01  1:42                   ` Menglong Dong
2025-08-06  8:44                   ` Menglong Dong
2025-08-08  0:58                     ` Alexei Starovoitov
2025-08-08  5:48                       ` Menglong Dong
2025-08-08  6:32                       ` Menglong Dong [this message]
2025-08-08 15:47                         ` Alexei Starovoitov
2025-07-15 16:35       ` Alexei Starovoitov
2025-07-16 13:05         ` Menglong Dong
2025-07-17  0:59           ` multi-fentry proposal. Was: " Alexei Starovoitov
2025-07-17  1:50             ` Menglong Dong
2025-07-17  2:13               ` Alexei Starovoitov
2025-07-17  2:37                 ` Menglong Dong
2025-07-16 14:40         ` Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 03/18] ftrace: factor out ftrace_direct_update from register_ftrace_direct Menglong Dong
2025-07-05  2:41   ` kernel test robot
2025-07-03 12:15 ` [PATCH bpf-next v2 04/18] ftrace: add reset_ftrace_direct_ips Menglong Dong
2025-07-03 15:30   ` Steven Rostedt
2025-07-04  1:54     ` Menglong Dong
2025-07-07 18:52       ` Steven Rostedt
2025-07-08  1:26         ` Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 05/18] bpf: introduce bpf_gtramp_link Menglong Dong
2025-07-04  7:00   ` kernel test robot
2025-07-04  7:52   ` kernel test robot
2025-07-03 12:15 ` [PATCH bpf-next v2 06/18] bpf: tracing: add support to record and check the accessed args Menglong Dong
2025-07-14 22:07   ` Andrii Nakryiko
2025-07-14 23:45     ` Menglong Dong
2025-07-15 17:11       ` Andrii Nakryiko
2025-07-16 12:50         ` Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 07/18] bpf: refactor the modules_array to ptr_array Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 08/18] bpf: verifier: add btf to the function args of bpf_check_attach_target Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 09/18] bpf: verifier: move btf_id_deny to bpf_check_attach_target Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 10/18] x86,bpf: factor out arch_bpf_get_regs_nr Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 11/18] bpf: tracing: add multi-link support Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 12/18] libbpf: don't free btf if tracing_multi progs existing Menglong Dong
2025-07-14 22:07   ` Andrii Nakryiko
2025-07-15  1:15     ` Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 13/18] libbpf: support tracing_multi Menglong Dong
2025-07-14 22:07   ` Andrii Nakryiko
2025-07-15  1:58     ` Menglong Dong
2025-07-15 17:20       ` Andrii Nakryiko
2025-07-16 12:43         ` Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 14/18] libbpf: add btf type hash lookup support Menglong Dong
2025-07-14 22:07   ` Andrii Nakryiko
2025-07-15  4:40     ` Menglong Dong
2025-07-15 17:20       ` Andrii Nakryiko
2025-07-16 11:53         ` Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 15/18] libbpf: add skip_invalid and attach_tracing for tracing_multi Menglong Dong
2025-07-14 22:07   ` Andrii Nakryiko
2025-07-15  5:48     ` Menglong Dong
2025-07-15 17:23       ` Andrii Nakryiko
2025-07-16 11:46         ` Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 16/18] selftests/bpf: move get_ksyms and get_addrs to trace_helpers.c Menglong Dong
2025-07-03 12:15 ` [PATCH bpf-next v2 17/18] selftests/bpf: add basic testcases for tracing_multi Menglong Dong
2025-07-08 20:07   ` Alexei Starovoitov
2025-07-09  1:33     ` Menglong Dong
2025-07-14 23:49     ` Ihor Solodrai
2025-07-16  0:26       ` Ihor Solodrai
2025-07-16  0:31         ` Alexei Starovoitov
2025-07-16  0:34           ` Ihor Solodrai
2025-07-03 12:15 ` [PATCH bpf-next v2 18/18] selftests/bpf: add bench tests " Menglong Dong
2025-07-04  8:47 ` [PATCH bpf-next v2 00/18] bpf: tracing multi-link support Jiri Olsa
2025-07-04  8:52   ` Menglong Dong
2025-07-04  8:58     ` Menglong Dong
2025-07-04  9:12       ` Jiri Olsa
2025-07-15  2:31 ` Alexei Starovoitov
2025-07-15  2:44   ` Menglong Dong

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=CADxym3ZeuoSXDS3j+Sv-Au-FbJzKjkh7TPmM619gRODuRCKzJw@mail.gmail.com \
    --to=menglong8.dong@gmail.com \
    --cc=alexei.starovoitov@gmail.com \
    --cc=bpf@vger.kernel.org \
    --cc=eddyz87@gmail.com \
    --cc=jolsa@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=martin.lau@linux.dev \
    --cc=menglong.dong@linux.dev \
    --cc=netdev@vger.kernel.org \
    --cc=peterz@infradead.org \
    --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;
as well as URLs for NNTP newsgroup(s).