All of lore.kernel.org
 help / color / mirror / Atom feed
From: Will Deacon <will@kernel.org>
To: torvalds@linux-foundation.org
Cc: catalin.marinas@arm.com, linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, kernel-team@android.com
Subject: [GIT PULL] arm64 perf/PMU regression fixes for 6.2
Date: Sat, 18 Feb 2023 11:11:22 +0000	[thread overview]
Message-ID: <20230218111100.GA23413@willie-the-truck> (raw)

Hi Linus,

Apologies for the _extremely_ late pull request here, but we had a 'perf'
(i.e. CPU PMU) regression on the Apple M1 reported on Wednesday [1] which
was introduced by bd2756811766 ("perf: Rewrite core context handling")
during the merge window.

Mark and I looked into this and noticed an additional problem caused by
the same patch, where the 'CHAIN' event (used to combine two adjacent
32-bit counters into a single 64-bit counter) was not being filtered
correctly. Mark posted a series on Thursday [2] which addresses both of
these regressions and I queued it the same day.

Unfortunately, the timing has meant that these two patches are yet to
land in linux-next. We narrowly missed the 20230217 release and now
we're into the weekend. The changes are small, self-contained and have
been confirmed to fix the original regression, so I'd really appreciate
it if you could please pull them in for 6.2 despite the caveat above.

Thanks, and sorry again for the unfortunate timing.

Will

[1] https://lore.kernel.org/asahi/20230215-arm_pmu_m1_regression-v1-1-f5a266577c8d@jannau.net/
[2] https://lore.kernel.org/linux-arm-kernel/20230216141240.3833272-1-mark.rutland@arm.com/

--->8

The following changes since commit a428eb4b99ab80454f06ad256b25e930fe8a4954:

  Partially revert "perf/arm-cmn: Optimise DTC counter accesses" (2023-01-26 13:55:38 +0000)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux.git tags/arm64-fixes

for you to fetch changes up to 853e2dac25c15f7431dfe59805de1bada34c96e9:

  arm64: perf: reject CHAIN events at creation time (2023-02-16 21:23:52 +0000)

----------------------------------------------------------------
arm64 regression fix for 6.2

- Fix 'perf' regression for non-standard CPU PMU hardware (i.e. Apple M1)

----------------------------------------------------------------
Mark Rutland (2):
      arm_pmu: fix event CPU filtering
      arm64: perf: reject CHAIN events at creation time

 arch/arm64/kernel/perf_event.c | 15 ++++++++-------
 drivers/perf/arm_pmu.c         |  8 +-------
 include/linux/perf/arm_pmu.h   |  1 -
 3 files changed, 9 insertions(+), 15 deletions(-)

WARNING: multiple messages have this Message-ID (diff)
From: Will Deacon <will@kernel.org>
To: torvalds@linux-foundation.org
Cc: catalin.marinas@arm.com, linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, kernel-team@android.com
Subject: [GIT PULL] arm64 perf/PMU regression fixes for 6.2
Date: Sat, 18 Feb 2023 11:11:22 +0000	[thread overview]
Message-ID: <20230218111100.GA23413@willie-the-truck> (raw)

Hi Linus,

Apologies for the _extremely_ late pull request here, but we had a 'perf'
(i.e. CPU PMU) regression on the Apple M1 reported on Wednesday [1] which
was introduced by bd2756811766 ("perf: Rewrite core context handling")
during the merge window.

Mark and I looked into this and noticed an additional problem caused by
the same patch, where the 'CHAIN' event (used to combine two adjacent
32-bit counters into a single 64-bit counter) was not being filtered
correctly. Mark posted a series on Thursday [2] which addresses both of
these regressions and I queued it the same day.

Unfortunately, the timing has meant that these two patches are yet to
land in linux-next. We narrowly missed the 20230217 release and now
we're into the weekend. The changes are small, self-contained and have
been confirmed to fix the original regression, so I'd really appreciate
it if you could please pull them in for 6.2 despite the caveat above.

Thanks, and sorry again for the unfortunate timing.

Will

[1] https://lore.kernel.org/asahi/20230215-arm_pmu_m1_regression-v1-1-f5a266577c8d@jannau.net/
[2] https://lore.kernel.org/linux-arm-kernel/20230216141240.3833272-1-mark.rutland@arm.com/

--->8

The following changes since commit a428eb4b99ab80454f06ad256b25e930fe8a4954:

  Partially revert "perf/arm-cmn: Optimise DTC counter accesses" (2023-01-26 13:55:38 +0000)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux.git tags/arm64-fixes

for you to fetch changes up to 853e2dac25c15f7431dfe59805de1bada34c96e9:

  arm64: perf: reject CHAIN events at creation time (2023-02-16 21:23:52 +0000)

----------------------------------------------------------------
arm64 regression fix for 6.2

- Fix 'perf' regression for non-standard CPU PMU hardware (i.e. Apple M1)

----------------------------------------------------------------
Mark Rutland (2):
      arm_pmu: fix event CPU filtering
      arm64: perf: reject CHAIN events at creation time

 arch/arm64/kernel/perf_event.c | 15 ++++++++-------
 drivers/perf/arm_pmu.c         |  8 +-------
 include/linux/perf/arm_pmu.h   |  1 -
 3 files changed, 9 insertions(+), 15 deletions(-)

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

             reply	other threads:[~2023-02-18 11:11 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-02-18 11:11 Will Deacon [this message]
2023-02-18 11:11 ` [GIT PULL] arm64 perf/PMU regression fixes for 6.2 Will Deacon
2023-02-18 19:15 ` pr-tracker-bot
2023-02-18 19:15   ` pr-tracker-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=20230218111100.GA23413@willie-the-truck \
    --to=will@kernel.org \
    --cc=catalin.marinas@arm.com \
    --cc=kernel-team@android.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@linux-foundation.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.