From: kernel test robot <oliver.sang@intel.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: <oe-lkp@lists.linux.dev>, <lkp@intel.com>,
<linux-kernel@vger.kernel.org>, <x86@kernel.org>,
Thomas Gleixner <tglx@kernel.org>,
kernel test robot <oliver.sang@intel.com>
Subject: [tip:sched/hrtick] [hrtimer] 92f7ee408c: stress-ng.timermix.ops_per_sec 23.8% improvement
Date: Mon, 23 Mar 2026 15:16:35 +0800 [thread overview]
Message-ID: <202603231407.332d483b-lkp@intel.com> (raw)
Hello,
kernel test robot noticed a 23.8% improvement of stress-ng.timermix.ops_per_sec on:
commit: 92f7ee408c23fcc074e4952fb6ffacdf3eb86b56 ("hrtimer: Less agressive interrupt 'hang' handling")
https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git sched/hrtick
testcase: stress-ng
config: x86_64-rhel-9.4
compiler: gcc-14
test machine: 192 threads 2 sockets Intel(R) Xeon(R) 6740E CPU @ 2.4GHz (Sierra Forest) with 256G memory
parameters:
nr_threads: 100%
testtime: 60s
test: timermix
cpufreq_governor: performance
Details are as below:
-------------------------------------------------------------------------------------------------->
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20260323/202603231407.332d483b-lkp@intel.com
=========================================================================================
compiler/cpufreq_governor/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime:
gcc-14/performance/x86_64-rhel-9.4/100%/debian-13-x86_64-20250902.cgz/lkp-srf-2sp3/timermix/stress-ng/60s
commit:
eef9f648fb ("sched/hrtick: Mark hrtick_clear() as always used")
92f7ee408c ("hrtimer: Less agressive interrupt 'hang' handling")
eef9f648fb0e9261 92f7ee408c23fcc074e4952fb6f
---------------- ---------------------------
%stddev %change %stddev
\ | \
55191 +1.9% 56222 stress-ng.time.minor_page_faults
13705 +24.9% 17115 stress-ng.time.percent_of_cpu_this_job_got
7596 +24.7% 9475 stress-ng.time.system_time
643.57 +26.4% 813.59 stress-ng.time.user_time
7013218 ± 3% -83.0% 1195013 ± 7% stress-ng.time.voluntary_context_switches
93747 +25.6% 117771 stress-ng.timermix.CLOCK_BOOTTIME_ticks_per_sec
93745 +25.7% 117867 stress-ng.timermix.CLOCK_MONOTONIC_ticks_per_sec
784.14 +27.0% 995.58 stress-ng.timermix.CLOCK_PROCESS_CPUTIME_ID_ticks_per_sec
93972 +25.7% 118146 stress-ng.timermix.CLOCK_REALTIME_ticks_per_sec
93713 +25.6% 117697 stress-ng.timermix.CLOCK_TAI_ticks_per_sec
784.22 +27.0% 995.69 stress-ng.timermix.CLOCK_THREAD_CPUTIME_ID_ticks_per_sec
7.84 ± 10% +85.5% 14.54 ± 17% stress-ng.timermix.ITIMER_VIRTUAL_ticks_per_sec
4.4e+09 +23.8% 5.449e+09 stress-ng.timermix.ops
73363779 +23.8% 90846415 stress-ng.timermix.ops_per_sec
2.694e+09 ± 2% -84.1% 4.282e+08 ± 15% cpuidle..time
10885817 ± 3% -92.5% 821275 ± 11% cpuidle..usage
24.41 ± 2% -75.1% 6.08 ± 9% vmstat.cpu.id
146.27 +24.8% 182.61 vmstat.procs.r
225376 ± 3% -82.1% 40249 ± 7% vmstat.system.cs
69523712 +22.2% 84939521 vmstat.system.in
21.93 ± 3% -18.6 3.28 mpstat.cpu.all.idle%
7.69 +2.1 9.81 mpstat.cpu.all.irq%
64.54 +15.1 79.68 mpstat.cpu.all.sys%
5.83 +1.4 7.22 mpstat.cpu.all.usr%
34.33 ± 16% -85.4% 5.00 mpstat.max_utilization.seconds
2538 +20.7% 3065 turbostat.Avg_MHz
79.48 +16.3 95.74 turbostat.Busy%
20.37 ± 2% -16.3 4.06 ± 21% turbostat.C1%
20.32 ± 3% -80.2% 4.03 ± 21% turbostat.CPU%c1
4.38e+09 +23.8% 5.422e+09 turbostat.IRQ
2.71 ± 6% -2.7 0.00 turbostat.POLL%
344.10 +15.6% 397.93 turbostat.PkgWatt
0.01 ± 29% +675.0% 0.08 ± 13% perf-sched.sch_delay.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
0.01 ± 29% +675.0% 0.08 ± 13% perf-sched.total_sch_delay.average.ms
7.77 ± 18% +2551.8% 206.08 ± 2% perf-sched.total_wait_and_delay.average.ms
152368 ± 17% -96.6% 5191 perf-sched.total_wait_and_delay.count.ms
2375 ± 6% +60.7% 3815 ± 15% perf-sched.total_wait_and_delay.max.ms
7.76 ± 18% +2554.5% 205.99 ± 2% perf-sched.total_wait_time.average.ms
2374 ± 6% +60.7% 3815 ± 15% perf-sched.total_wait_time.max.ms
7.77 ± 18% +2551.8% 206.08 ± 2% perf-sched.wait_and_delay.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
152368 ± 17% -96.6% 5191 perf-sched.wait_and_delay.count.[unknown].[unknown].[unknown].[unknown].[unknown]
2375 ± 6% +60.7% 3815 ± 15% perf-sched.wait_and_delay.max.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
7.76 ± 18% +2554.5% 205.99 ± 2% perf-sched.wait_time.avg.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
2374 ± 6% +60.7% 3815 ± 15% perf-sched.wait_time.max.ms.[unknown].[unknown].[unknown].[unknown].[unknown]
0.02 ± 4% -31.3% 0.02 ± 3% perf-stat.i.MPKI
5.401e+10 +27.7% 6.9e+10 perf-stat.i.branch-instructions
0.25 -0.0 0.22 perf-stat.i.branch-miss-rate%
1.197e+08 +24.3% 1.487e+08 perf-stat.i.branch-misses
11654768 ± 5% +32.5% 15440506 ± 3% perf-stat.i.cache-references
273890 ± 3% -83.3% 45691 ± 7% perf-stat.i.context-switches
1.79 -6.3% 1.68 perf-stat.i.cpi
4.788e+11 +25.7% 6.018e+11 perf-stat.i.cpu-cycles
305.34 ± 2% -12.7% 266.67 perf-stat.i.cpu-migrations
290485 ± 3% +357.1% 1327861 ± 81% perf-stat.i.cycles-between-cache-misses
2.844e+11 +28.4% 3.651e+11 perf-stat.i.instructions
0.57 +6.2% 0.61 perf-stat.i.ipc
1.37 ± 3% -83.7% 0.22 ± 9% perf-stat.i.metric.K/sec
2721 ± 3% +22.5% 3332 ± 4% perf-stat.i.minor-faults
2721 ± 3% +22.5% 3332 ± 4% perf-stat.i.page-faults
0.01 -15.5% 0.01 ± 5% perf-stat.overall.MPKI
0.22 -0.0 0.21 perf-stat.overall.branch-miss-rate%
21.40 ± 6% -3.2 18.23 ± 8% perf-stat.overall.cache-miss-rate%
1.68 -1.6% 1.65 perf-stat.overall.cpi
169048 +16.9% 197558 ± 5% perf-stat.overall.cycles-between-cache-misses
0.60 +1.7% 0.61 perf-stat.overall.ipc
5.409e+10 +24.0% 6.706e+10 perf-stat.ps.branch-instructions
1.186e+08 +21.5% 1.44e+08 perf-stat.ps.branch-misses
13265941 ± 6% +23.2% 16341982 ± 3% perf-stat.ps.cache-references
237189 ± 4% -82.2% 42112 ± 6% perf-stat.ps.context-switches
4.777e+11 +22.5% 5.85e+11 perf-stat.ps.cpu-cycles
279.70 ± 2% -11.0% 248.81 perf-stat.ps.cpu-migrations
2.85e+11 +24.5% 3.548e+11 perf-stat.ps.instructions
2443 ± 3% +21.5% 2968 ± 5% perf-stat.ps.minor-faults
2443 ± 3% +21.5% 2968 ± 5% perf-stat.ps.page-faults
1.701e+13 +25.1% 2.128e+13 perf-stat.total.instructions
4.12 ± 3% -0.3 3.80 ± 4% perf-profile.calltrace.cycles-pp.save_xstate_epilog.get_sigframe.x64_setup_rt_frame.arch_do_signal_or_restart.exit_to_user_mode_loop
12.30 ± 14% +2.7 15.03 ± 10% perf-profile.calltrace.cycles-pp.exit_to_user_mode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe.sigpending
4.90 ± 3% -0.5 4.40 perf-profile.children.cycles-pp.save_xstate_epilog
2.04 ± 2% -0.2 1.81 perf-profile.children.cycles-pp.__put_user_nocheck_4
2.15 ± 2% -0.2 1.99 ± 5% perf-profile.children.cycles-pp.__put_user_nocheck_8
2.50 -0.1 2.42 perf-profile.children.cycles-pp.entry_SYSCALL_64
0.77 ± 2% -0.1 0.71 perf-profile.children.cycles-pp.__local_bh_enable_ip
1.53 -0.1 1.48 perf-profile.children.cycles-pp.fpregs_mark_activate
1.76 -0.0 1.72 perf-profile.children.cycles-pp._raw_spin_lock_irq
0.89 ± 2% -0.0 0.85 ± 2% perf-profile.children.cycles-pp.signal_setup_done
0.16 ± 8% -0.0 0.12 ± 6% perf-profile.children.cycles-pp.get_xsave_addr_user
0.86 -0.0 0.83 perf-profile.children.cycles-pp.arch_exit_to_user_mode_prepare
0.46 -0.0 0.44 perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack
0.00 +0.1 0.11 ± 20% perf-profile.children.cycles-pp.add_callchain_ip
0.00 +0.1 0.12 ± 19% perf-profile.children.cycles-pp.thread__resolve_callchain_sample
0.00 +0.1 0.12 ± 5% perf-profile.children.cycles-pp.ordered_events__queue
0.00 +0.1 0.12 ± 5% perf-profile.children.cycles-pp.process_simple
0.00 +0.1 0.12 ± 5% perf-profile.children.cycles-pp.queue_event
0.01 ±223% +0.1 0.14 ± 21% perf-profile.children.cycles-pp.__thread__resolve_callchain
0.01 ±223% +0.1 0.14 ± 21% perf-profile.children.cycles-pp.sample__for_each_callchain_node
0.01 ±223% +0.1 0.15 ± 20% perf-profile.children.cycles-pp.build_id__mark_dso_hit
0.01 ±223% +0.2 0.17 ± 22% perf-profile.children.cycles-pp.perf_session__deliver_event
0.01 ±223% +0.2 0.17 ± 22% perf-profile.children.cycles-pp.__ordered_events__flush
6.06 +0.2 6.23 perf-profile.children.cycles-pp.posixtimer_deliver_signal
6.71 +0.2 6.88 perf-profile.children.cycles-pp.dequeue_signal
0.02 ±223% +0.2 0.22 ± 17% perf-profile.children.cycles-pp.reader__read_event
0.05 ± 86% +0.3 0.33 ± 12% perf-profile.children.cycles-pp.cmd_record
0.02 ±223% +0.3 0.29 ± 13% perf-profile.children.cycles-pp.perf_session__process_events
0.02 ±223% +0.3 0.29 ± 13% perf-profile.children.cycles-pp.record__finish_output
1.59 ± 6% -0.1 1.50 perf-profile.self.cycles-pp.save_xstate_epilog
1.73 -0.1 1.66 perf-profile.self.cycles-pp.entry_SYSCALL_64
1.78 -0.1 1.72 perf-profile.self.cycles-pp.__put_user_nocheck_4
1.81 -0.1 1.75 perf-profile.self.cycles-pp.__put_user_nocheck_8
1.74 -0.0 1.70 perf-profile.self.cycles-pp._raw_spin_lock_irq
1.00 -0.0 0.96 perf-profile.self.cycles-pp.fpregs_mark_activate
0.73 ± 2% -0.0 0.70 perf-profile.self.cycles-pp.__local_bh_enable_ip
0.54 -0.0 0.52 ± 2% perf-profile.self.cycles-pp.do_syscall_64
0.24 -0.0 0.22 ± 2% perf-profile.self.cycles-pp.sysvec_apic_timer_interrupt
0.22 +0.0 0.23 perf-profile.self.cycles-pp._raw_spin_unlock_irqrestore
0.49 +0.0 0.51 perf-profile.self.cycles-pp.posixtimer_send_sigqueue
0.07 ± 7% +0.0 0.09 ± 8% perf-profile.self.cycles-pp.common_hrtimer_rearm
0.00 +0.1 0.12 ± 6% perf-profile.self.cycles-pp.queue_event
Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
reply other threads:[~2026-03-23 7:16 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=202603231407.332d483b-lkp@intel.com \
--to=oliver.sang@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lkp@intel.com \
--cc=oe-lkp@lists.linux.dev \
--cc=peterz@infradead.org \
--cc=tglx@kernel.org \
--cc=x86@kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox