[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240526115834.4080389-1-acme@kernel.org>
Date: Sun, 26 May 2024 08:58:34 -0300
From: Arnaldo Carvalho de Melo <acme@...nel.org>
To: Linus Torvalds <torvalds@...ux-foundation.org>,
Ian Rogers <irogers@...gle.com>
Cc: Ingo Molnar <mingo@...nel.org>,
Thomas Gleixner <tglx@...utronix.de>,
Jiri Olsa <jolsa@...nel.org>,
Namhyung Kim <namhyung@...nel.org>,
Adrian Hunter <adrian.hunter@...el.com>,
Clark Williams <williams@...hat.com>,
Kate Carcia <kcarcia@...hat.com>,
linux-kernel@...r.kernel.org,
linux-perf-users@...r.kernel.org,
Arnaldo Carvalho de Melo <acme@...nel.org>,
Leo Yan <leo.yan@...ux.dev>,
Bhaskar Chowdhury <unixbhaskar@...il.com>,
Ethan Adams <j.ethan.adams@...il.com>,
James Clark <james.clark@....com>,
Kan Liang <kan.liang@...ux.intel.com>,
Thomas Richter <tmricht@...ux.ibm.com>,
Tycho Andersen <tycho@...ho.pizza>,
Yang Jihong <yangjihong@...edance.com>,
Arnaldo Carvalho de Melo <acme@...hat.com>
Subject: [GIT PULL] perf tools fix for v6.10
Hi Linus,
Please consider pulling, with the revert no 'perf test'
regressions were noticed on this Intel hybrid system:
root@x1:~# grep -m1 'model name' /proc/cpuinfo
model name : 13th Gen Intel(R) Core(TM) i7-1365U
root@x1:~# perf stat -a sleep 1
Performance counter stats for 'system wide':
12,054.90 msec cpu-clock # 11.998 CPUs utilized
1,865 context-switches # 154.709 /sec
108 cpu-migrations # 8.959 /sec
111 page-faults # 9.208 /sec
418,070,471 cpu_atom/cycles/ # 0.035 GHz (74.91%)
55,680,862 cpu_core/cycles/ # 0.005 GHz
95,014,672 cpu_atom/instructions/ # 0.23 insn per cycle (87.46%)
33,161,707 cpu_core/instructions/ # 0.08 insn per cycle
18,525,790 cpu_atom/branches/ # 1.537 M/sec (87.49%)
5,792,527 cpu_core/branches/ # 480.512 K/sec
1,939,067 cpu_atom/branch-misses/ # 10.47% of all branches (87.43%)
238,479 cpu_core/branch-misses/ # 1.29% of all branches
TopdownL1 (cpu_core) # 33.3 % tma_backend_bound
# 6.5 % tma_bad_speculation
# 49.0 % tma_frontend_bound
# 11.2 % tma_retiring
TopdownL1 (cpu_atom) # 16.5 % tma_bad_speculation
# 6.5 % tma_retiring (87.56%)
# 41.1 % tma_backend_bound
# 41.1 % tma_backend_bound_aux
# 35.8 % tma_frontend_bound (87.56%)
1.004776564 seconds time elapsed
root@x1:~#
root@x1:~# perf stat -a -e cycles:P sleep 1
Performance counter stats for 'system wide':
273,063,480 cpu_atom/cycles:P/
156,166,111 cpu_core/cycles:P/
1.005749149 seconds time elapsed
root@x1:~#
Best regards,
- Arnaldo
Arnaldo Carvalho de Melo (1):
Revert "perf parse-events: Prefer sysfs/JSON hardware events over legacy"
tools/perf/util/parse-events.c | 31 ++++----------
tools/perf/util/parse-events.h | 2 +-
tools/perf/util/parse-events.l | 76 +++++++++++++++++-----------------
tools/perf/util/parse-events.y | 62 ++++++++++-----------------
4 files changed, 68 insertions(+), 103 deletions(-)
--
2.45.1
The following changes since commit 29c73fc794c83505066ee6db893b2a83ac5fac63:
Merge tag 'perf-tools-for-v6.10-1-2024-05-21' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools (2024-05-21 15:45:14 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools.git tags/perf-tools-fixes-for-v6.10-1-2024-05-26
for you to fetch changes up to 4f1b067359ac8364cdb7f9fda41085fa85789d0f:
Revert "perf parse-events: Prefer sysfs/JSON hardware events over legacy" (2024-05-26 08:41:34 -0300)
----------------------------------------------------------------
Revert a patch causing a regression as described in the cset:
"This made a simple 'perf record -e cycles:pp make -j199' stop working on
the Ampere ARM64 system Linus uses to test ARM64 kernels, as discussed
at length in the threads in the Link tags below.
The fix provided by Ian wasn't acceptable and work to fix this will take
time we don't have at this point, so lets revert this and work on it on
the next devel cycle."
Signed-off-by: Arnaldo Carvalho de Melo <acme@...hat.com>
----------------------------------------------------------------
Arnaldo Carvalho de Melo (1):
Revert "perf parse-events: Prefer sysfs/JSON hardware events over legacy"
tools/perf/util/parse-events.c | 31 +++++------------
tools/perf/util/parse-events.h | 2 +-
tools/perf/util/parse-events.l | 76 +++++++++++++++++++++---------------------
tools/perf/util/parse-events.y | 62 ++++++++++++----------------------
4 files changed, 68 insertions(+), 103 deletions(-)
Powered by blists - more mailing lists