Linux Perf Users
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: "Mi, Dapeng" <dapeng1.mi@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Namhyung Kim <namhyung@kernel.org>,
	Ian Rogers <irogers@google.com>,
	Adrian Hunter <adrian.hunter@intel.com>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Andi Kleen <ak@linux.intel.com>,
	Eranian Stephane <eranian@google.com>,
	linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
	Dapeng Mi <dapeng1.mi@intel.com>, Zide Chen <zide.chen@intel.com>,
	Falcon Thomas <thomas.falcon@intel.com>,
	Xudong Hao <xudong.hao@intel.com>,
	Gennady Kupava <gennady.kupava@gmail.com>,
	Ravi Bangoria <ravi.bangoria@amd.com>
Subject: Re: [PATCH 2/2] perf/x86: Disable precise sampling for PERF_SAMPLE_STACK_USER
Date: Tue, 8 Sep 2026 12:19:14 +0200	[thread overview]
Message-ID: <20260908101914.GQ4121339@noisy.programming.kicks-ass.net> (raw)
In-Reply-To: <1691a05c-49a6-4b16-8bad-cb3c004ed07c@linux.intel.com>

On Tue, Sep 08, 2026 at 04:56:22PM +0800, Mi, Dapeng wrote:
> 
> On 9/8/2026 4:49 PM, Peter Zijlstra wrote:
> > On Tue, Sep 08, 2026 at 03:51:02PM +0800, Dapeng Mi wrote:
> >> PERF_SAMPLE_STACK_USER needs to return the user stack and user registers
> >> to user space when the PMI exits. Since the skid from the PEBS/IBS sample
> >> and PMI delivery, the PEBS/IBS register snapshot (especially IP/SP/BP)
> >> can diverge from the user stack at PMI return. That mismatch breaks DWARF
> >> unwinding.
> >>
> >> Precise sampling provides no benefit in this case, so disable PEBS/IBS
> >> precise sampling and allow only PMI-based sampling when
> >> PERF_SAMPLE_STACK_USER is requested.
> >>
> >> Reported-by: Gennady Kupava <gennady.kupava@gmail.com>
> >> Closes: https://lore.kernel.org/all/CAPu-DQqF0aF6=GS8Z6KKWeeX_V5LiXeKU_rJQZC+uGg8zuTPNw@mail.gmail.com/
> >> Cc: Ravi Bangoria <ravi.bangoria@amd.com>
> >> Fixes: c5ebcedb566e ("perf: Add ability to attach user stack dump to sample")
> >> Signed-off-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
> > This breaks long standing existing behaviour.
> 
> Yeah, but it seems there is no better way to fix this issue.

Breaking things that worked before isn't fixing.. people get upset.

> An alternative way to fix this issue is still to return the PMI
> context register state rather than the PEBS precise registers for user
> stack sampling, but this actually falls back the imprecise PMI-based
> sampling. 

That's what we already do, no? I have distinct memories of making the
stack unwind use the NMI regs rather then the PEBS regs.

> In my opinion, it could even make the thing worse. User
> requires to get precise samplings, but perf silently returns imprecise
> records, this would mislead user. 

Mostly just the unwind might be off a little, the rest is accurate. This
has been the case 'forever'. Performance analysis isn't for silly
people, if they can't deal with a little fuzz then perhaps they're in
the wrong business.

  reply	other threads:[~2026-09-08 10:19 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-08  7:51 [PATCH 1/2] perf/x86/intel: Correct pt_regs->flags update for PEBS path Dapeng Mi
2026-09-08  7:51 ` [PATCH 2/2] perf/x86: Disable precise sampling for PERF_SAMPLE_STACK_USER Dapeng Mi
2026-09-08  8:15   ` sashiko-bot
2026-09-08  8:47     ` Mi, Dapeng
2026-09-08  8:49   ` Peter Zijlstra
2026-09-08  8:56     ` Mi, Dapeng
2026-09-08 10:19       ` Peter Zijlstra [this message]
2026-09-08 15:10         ` Andi Kleen
2026-09-08 20:56           ` Ian Rogers
2026-09-09  0:59             ` Mi, Dapeng
2026-09-09  1:28               ` Ravi Bangoria
2026-09-09  1:59                 ` Mi, Dapeng
2026-09-09  8:11             ` Peter Zijlstra
2026-09-09  9:36               ` Mi, Dapeng
2026-09-09 19:30                 ` Namhyung Kim
2026-09-10  0:12                   ` Mi, Dapeng
2026-09-09 14:14             ` Namhyung Kim
2026-09-09  1:20         ` Mi, Dapeng
2026-09-08  8:49   ` Mi, Dapeng
2026-09-08  8:05 ` [PATCH 1/2] perf/x86/intel: Correct pt_regs->flags update for PEBS path sashiko-bot

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=20260908101914.GQ4121339@noisy.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=acme@kernel.org \
    --cc=adrian.hunter@intel.com \
    --cc=ak@linux.intel.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=dapeng1.mi@intel.com \
    --cc=dapeng1.mi@linux.intel.com \
    --cc=eranian@google.com \
    --cc=gennady.kupava@gmail.com \
    --cc=irogers@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=ravi.bangoria@amd.com \
    --cc=thomas.falcon@intel.com \
    --cc=xudong.hao@intel.com \
    --cc=zide.chen@intel.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