linux-trace-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com>,
	Jiri Olsa <olsajiri@gmail.com>,  Oleg Nesterov <oleg@redhat.com>,
	Andrii Nakryiko <andrii@kernel.org>, bpf <bpf@vger.kernel.org>,
	 Song Liu <songliubraving@fb.com>, Yonghong Song <yhs@fb.com>,
	 John Fastabend <john.fastabend@gmail.com>,
	Hao Luo <haoluo@google.com>,
	 Steven Rostedt <rostedt@goodmis.org>,
	Masami Hiramatsu <mhiramat@kernel.org>,
	 Alan Maguire <alan.maguire@oracle.com>,
	LKML <linux-kernel@vger.kernel.org>,
	 linux-trace-kernel <linux-trace-kernel@vger.kernel.org>,
	"H. Peter Anvin" <hpa@zytor.com>
Subject: Re: [RFC perf/core 05/11] uprobes: Add mapping for optimized uprobe trampolines
Date: Thu, 21 Nov 2024 08:02:12 -0800	[thread overview]
Message-ID: <CAADnVQJJ0WS=Y1EudjiFD8fn4zHCz6x1auaEEHaYHsP15Vks2Q@mail.gmail.com> (raw)
In-Reply-To: <20241121115353.GJ24774@noisy.programming.kicks-ass.net>

On Thu, Nov 21, 2024 at 4:17 AM Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Wed, Nov 20, 2024 at 04:07:38PM -0800, Andrii Nakryiko wrote:
>
> > USDTs are meant to be "transparent" to the surrounding code and they
> > don't mark any clobbered registers. Technically it could be added, but
> > I'm not a fan of this.
>
> Sure. Anyway, another thing to consider is FRED, will all of this still
> matter once that lands? If FRED gets us INT3 performance close to what
> SYSCALL has, then all this work will go unused.

afaik not a single cpu in the datacenter supports FRED while
uprobe overhead is real.
imo it's worth improving performance today for existing cpus.
I suspect arm64 might benefit too. Even if arm hw does the same
amount of work for trap vs syscall the sw overhead of handling
trap is different.
I suspect that equation will apply to future FRED cpus too.

  reply	other threads:[~2024-11-21 16:02 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-11-05 13:33 [RFC 00/11] uprobes: Add support to optimize usdt probes on x86_64 Jiri Olsa
2024-11-05 13:33 ` [RFC perf/core 01/11] uprobes: Rename arch_uretprobe_trampoline function Jiri Olsa
2024-11-05 13:33 ` [RFC perf/core 02/11] uprobes: Make copy_from_page global Jiri Olsa
2024-11-14 23:40   ` Andrii Nakryiko
2024-11-16 21:41     ` Jiri Olsa
2024-11-05 13:33 ` [RFC perf/core 03/11] uprobes: Add len argument to uprobe_write_opcode Jiri Olsa
2024-11-14 23:41   ` Andrii Nakryiko
2024-11-16 21:41     ` Jiri Olsa
2024-11-05 13:33 ` [RFC perf/core 04/11] uprobes: Add data argument to uprobe_write_opcode function Jiri Olsa
2024-11-14 23:41   ` Andrii Nakryiko
2024-11-16 21:43     ` Jiri Olsa
2024-11-05 13:33 ` [RFC perf/core 05/11] uprobes: Add mapping for optimized uprobe trampolines Jiri Olsa
2024-11-05 14:23   ` Peter Zijlstra
2024-11-05 16:33     ` Jiri Olsa
2024-11-14 23:44       ` Andrii Nakryiko
2024-11-16 21:44         ` Jiri Olsa
2024-11-19  6:06           ` Andrii Nakryiko
2024-11-19  9:13             ` Peter Zijlstra
2024-11-19 15:15               ` Jiri Olsa
2024-11-21  0:07               ` Andrii Nakryiko
2024-11-21 11:53                 ` Peter Zijlstra
2024-11-21 16:02                   ` Alexei Starovoitov [this message]
2024-11-21 16:34                     ` Peter Zijlstra
2024-11-21 16:47                       ` Alexei Starovoitov
2024-11-21 19:38                         ` Mark Rutland
2024-11-14 23:44   ` Andrii Nakryiko
2024-11-16 21:44     ` Jiri Olsa
2024-11-19  6:05       ` Andrii Nakryiko
2024-11-19 15:14         ` Jiri Olsa
2024-11-21  0:10           ` Andrii Nakryiko
2024-11-05 13:34 ` [RFC perf/core 06/11] uprobes: Add uprobe syscall to speed up uprobe Jiri Olsa
2024-11-05 13:34 ` [RFC perf/core 07/11] uprobes/x86: Add support to optimize uprobes Jiri Olsa
2024-11-14 23:44   ` Andrii Nakryiko
2024-11-16 21:44     ` Jiri Olsa
2024-11-18  8:18   ` Masami Hiramatsu
2024-11-18  9:39     ` Jiri Olsa
2024-11-05 13:34 ` [RFC bpf-next 08/11] selftests/bpf: Use 5-byte nop for x86 usdt probes Jiri Olsa
2024-11-05 13:34 ` [RFC bpf-next 09/11] selftests/bpf: Add usdt trigger bench Jiri Olsa
2024-11-14 23:40   ` Andrii Nakryiko
2024-11-16 21:45     ` Jiri Olsa
2024-11-19  6:08       ` Andrii Nakryiko
2024-11-05 13:34 ` [RFC bpf-next 10/11] selftests/bpf: Add uprobe/usdt optimized test Jiri Olsa
2024-11-05 13:34 ` [RFC bpf-next 11/11] selftests/bpf: Add hit/attach/detach race optimized uprobe test Jiri Olsa
2024-11-17 11:49 ` [RFC 00/11] uprobes: Add support to optimize usdt probes on x86_64 Peter Zijlstra
2024-11-18  9:29   ` Jiri Olsa
2024-11-18 10:06   ` Mark Rutland
2024-11-19  6:13     ` Andrii Nakryiko
2024-11-21 18:18       ` Mark Rutland
2024-11-26 19:13         ` Andrii Nakryiko
2024-11-18  8:04 ` Masami Hiramatsu
2024-11-18  9:52   ` Jiri Olsa

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='CAADnVQJJ0WS=Y1EudjiFD8fn4zHCz6x1auaEEHaYHsP15Vks2Q@mail.gmail.com' \
    --to=alexei.starovoitov@gmail.com \
    --cc=alan.maguire@oracle.com \
    --cc=andrii.nakryiko@gmail.com \
    --cc=andrii@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=haoluo@google.com \
    --cc=hpa@zytor.com \
    --cc=john.fastabend@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=mhiramat@kernel.org \
    --cc=oleg@redhat.com \
    --cc=olsajiri@gmail.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=songliubraving@fb.com \
    --cc=yhs@fb.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;
as well as URLs for NNTP newsgroup(s).