From: sashiko-bot@kernel.org
To: "Dapeng Mi" <dapeng1.mi@linux.intel.com>
Cc: linux-perf-users@vger.kernel.org
Subject: Re: [Patch v6 12/20] perf/x86/intel: Use a dedicated LBR stack for BPF snapshotting
Date: Fri, 06 Nov 2026 05:17:17 +0000 [thread overview]
Message-ID: <sashiko-outbox-200808@kernel.org> (raw)
In-Reply-To: <20261106043257.842864-13-dapeng1.mi@linux.intel.com>
> BPF LBR snapshotting currently reuses the default LBR stack
> (cpuc->__lbr_stack) when storing sampled data. This is unsafe when the
> BPF helper runs in PMI context via bpf_overflow_handler(): it can
> overwrite the cached LBR state for active PEBS LBR events, causing
> incorrect LBR data to be reported to userspace. When the PEBS and BPF
> stacks have different depths, this can also lead to out-of-bounds reads
> when the LBR stack is later exported.
>
> Use a dedicated LBR stack for BPF snapshotting so the BPF path no
> longer clobbers the shared LBR state used by other perf events.
>
> Reported-by: Sashiko <sashiko-bot@kernel.org>
> Fixes: 425507fa5f32 ("perf/x86/intel/lbr: Optimize v4 LBR unfreezing")
> Signed-off-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
Sashiko has reviewed this patch and found no issues.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261106043257.842864-1-dapeng1.mi@linux.intel.com?part=12
next prev parent reply other threads:[~2026-11-06 5:17 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-06 4:32 [Patch v6 00/20] perf/x86: Fix multiple LBR sampling issues Dapeng Mi
2026-11-06 4:32 ` [Patch v6 01/20] perf/x86: Disable LBR if x86_pmu.lbr_nr exceeds MAX_LBR_ENTRIES Dapeng Mi
2026-11-06 5:01 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 02/20] perf/x86/intel: Validate LBR XSAVE state before updating LBR stack Dapeng Mi
2026-11-06 5:05 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 03/20] perf/x86/amd/lbr: Limit branch snapshots to valid entries Dapeng Mi
2026-11-06 4:50 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 04/20] perf/x86/intel/lbr: " Dapeng Mi
2026-11-06 4:52 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 05/20] perf/x86: Add lbr_enabled to track LBR state Dapeng Mi
2026-11-06 5:10 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 06/20] perf/x86/amd: Avoid re-enabling PMU and LBR for BPF when disabled Dapeng Mi
2026-11-06 5:01 ` sashiko-bot
2026-11-06 7:07 ` Mi, Dapeng
2026-11-06 4:32 ` [Patch v6 07/20] perf/x86/intel: Add helper to enable PMU without LBR Dapeng Mi
2026-11-06 4:45 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 08/20] perf/x86/intel: Avoid re-enabling PMU and LBR for BPF when disabled Dapeng Mi
2026-11-06 5:06 ` sashiko-bot
2026-11-06 7:09 ` Mi, Dapeng
2026-11-06 4:32 ` [Patch v6 09/20] perf/x86/intel: Keep LBR frozen throughout PMI handling Dapeng Mi
2026-11-06 4:58 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 10/20] perf: Add counted-bounds tracking for perf_branch_stack Dapeng Mi
2026-11-06 5:00 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 11/20] perf/x86/intel: Allow LBR read helpers to operate on explicit stacks Dapeng Mi
2026-11-06 5:09 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 12/20] perf/x86/intel: Use a dedicated LBR stack for BPF snapshotting Dapeng Mi
2026-11-06 5:17 ` sashiko-bot [this message]
2026-11-06 4:32 ` [Patch v6 13/20] perf/x86/intel: Always snapshot live hardware LBR state for BPF Dapeng Mi
2026-11-06 5:13 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 14/20] perf/x86/amd: Refresh LBR stack in PMI only when LBR users are active Dapeng Mi
2026-11-06 5:09 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 15/20] perf/x86/intel: Refresh LBR stack for non-PEBS LBR events when needed Dapeng Mi
2026-11-06 5:58 ` sashiko-bot
2026-11-06 6:54 ` Mi, Dapeng
2026-11-06 4:32 ` [Patch v6 16/20] perf/x86/intel: Remove dead lbr_pebs_users Dapeng Mi
2026-11-06 5:08 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 17/20] perf/x86/intel: Clear raw LBR counters in the LBR stack Dapeng Mi
2026-11-06 5:17 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 18/20] perf/x86/intel: Refill LBR stack for LBR counter events in PMI handler Dapeng Mi
2026-11-06 5:17 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 19/20] perf/x86/intel: Pass LBR stack size into store_lbr() Dapeng Mi
2026-11-06 5:12 ` sashiko-bot
2026-11-06 4:32 ` [Patch v6 20/20] perf/core: Don't overwrite shared LBR stack depth Dapeng Mi
2026-11-06 5:17 ` sashiko-bot
2026-11-06 7:23 ` [Patch v6 00/20] perf/x86: Fix multiple LBR sampling issues Mi, Dapeng
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=sashiko-outbox-200808@kernel.org \
--to=sashiko-bot@kernel.org \
--cc=dapeng1.mi@linux.intel.com \
--cc=linux-perf-users@vger.kernel.org \
--cc=sashiko-reviews@lists.linux.dev \
/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