| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20220327132218.GC32190@xsang-OptiPlex-9020>
Date: Sun, 27 Mar 2022 21:22:18 +0800
From: kernel test robot <oliver.sang@...el.com>
To: Chris Wilson <chris@...is-wilson.co.uk>
Cc: Rodrigo Vivi <rodrigo.vivi@...el.com>,
Jason Ekstrand <jason@...kstrand.net>,
Mika Kuoppala <mika.kuoppala@...ux.intel.com>,
Tvrtko Ursulin <tvrtko.ursulin@...ux.intel.com>,
Joonas Lahtinen <joonas.lahtinen@...ux.intel.com>,
Jason Ekstrand <jason@...sktrand.net>,
LKML <linux-kernel@...r.kernel.org>, lkp@...ts.01.org,
lkp@...el.com, ying.huang@...el.com, feng.tang@...el.com,
zhengjun.xing@...ux.intel.com, fengwei.yin@...el.com,
guobing.chen@...el.com, ming.a.chen@...el.com, frank.du@...el.com,
Shuhua.Fan@...el.com, wangyang.guo@...el.com,
Wenhuan.Huang@...el.com, jessica.ji@...el.com, shan.kang@...el.com,
guangli.li@...el.com, tiejun.li@...el.com, yu.ma@...el.com,
dapeng1.mi@...el.com, jiebin.sun@...el.com, gengxin.xie@...el.com,
fan.zhao@...el.com
Subject: [drm/i915] 58586680ff:
phoronix-test-suite.darktable.Boat.OpenCL.seconds 139.5% regression
Greeting,
FYI, we noticed a 139.5% regression of phoronix-test-suite.darktable.Boat.OpenCL.seconds due to commit:
commit: 58586680ffadc37636120d9f59955aa5f7a32b7d ("drm/i915: Disable atomics in L3 for gen9")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: phoronix-test-suite
on test machine: 12 threads 1 sockets Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz with 16G memory
with following parameters:
test: darktable-1.0.5
option_a: Boat
option_b: OpenCL
cpufreq_governor: performance
ucode: 0xec
test-description: The Phoronix Test Suite is the most comprehensive testing and benchmarking platform available that provides an extensible framework for which new tests can be easily added.
test-url: http://www.phoronix-test-suite.com/
If you fix the issue, kindly add following tag
Reported-by: kernel test robot <oliver.sang@...el.com>
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
sudo bin/lkp install job.yaml # job file is attached in this email
bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
sudo bin/lkp run generated-yaml-file
# if come across any failure that blocks the test,
# please remove ~/.lkp and /lkp dir to run from a clean state.
=========================================================================================
compiler/cpufreq_governor/kconfig/option_a/option_b/rootfs/tbox_group/test/testcase/ucode:
gcc-9/performance/x86_64-rhel-8.3/Boat/OpenCL/debian-x86_64-phoronix/lkp-cfl-d1/darktable-1.0.5/phoronix-test-suite/0xec
commit:
29d88083d6 ("drm/i915/gem: Move freeze/freeze_late next to suspend/suspend_late")
58586680ff ("drm/i915: Disable atomics in L3 for gen9")
29d88083d6ff697c 58586680ffadc37636120d9f599
---------------- ---------------------------
%stddev %change %stddev
\ | \
18.66 ± 2% +139.5% 44.70 ± 2% phoronix-test-suite.darktable.Boat.OpenCL.seconds
89.20 +86.9% 166.70 phoronix-test-suite.time.elapsed_time
89.20 +86.9% 166.70 phoronix-test-suite.time.elapsed_time.max
26101 ± 3% -43.2% 14822 ± 5% phoronix-test-suite.time.involuntary_context_switches
103.33 -46.9% 54.83 phoronix-test-suite.time.percent_of_cpu_this_job_got
9.658e+08 ± 3% +92.2% 1.857e+09 cpuidle..time
2922129 ± 12% +54.4% 4510501 ± 7% cpuidle..usage
103.64 ± 2% +83.5% 190.18 ± 11% uptime.boot
1032 ± 2% +83.3% 1891 ± 14% uptime.idle
4.64 ± 3% +5.3 9.93 ± 3% mpstat.cpu.all.iowait%
2.07 ± 4% -0.3 1.81 ± 14% mpstat.cpu.all.irq%
3.13 ± 3% -1.4 1.77 ± 2% mpstat.cpu.all.sys%
7.14 ± 3% -2.7 4.48 ± 7% mpstat.cpu.all.usr%
422306 +20.5% 508672 meminfo.Active
233866 ± 3% +33.4% 312003 meminfo.Active(anon)
725925 ± 4% +11.9% 811980 meminfo.AnonHugePages
2612386 +11.6% 2916376 meminfo.Cached
3273053 +11.3% 3641931 meminfo.Committed_AS
1324227 +22.3% 1619169 meminfo.Shmem
422032 ± 4% +69.3% 714319 meminfo.Unevictable
4.00 +133.3% 9.33 ± 5% vmstat.cpu.wa
1843 ± 2% -43.3% 1046 ± 8% vmstat.io.bi
2677322 +10.9% 2970380 vmstat.memory.cache
0.00 +1e+102% 1.00 vmstat.procs.b
1.00 -100.0% 0.00 vmstat.procs.r
18596 ± 3% -42.4% 10714 ± 2% vmstat.system.cs
31709 -12.3% 27820 ± 5% vmstat.system.in
77.32 -37.9% 48.03 turbostat.Any%C0
538.00 -40.2% 321.50 ± 2% turbostat.Avg_MHz
12.80 -4.8 8.00 ± 2% turbostat.Busy%
44.08 ± 2% -12.6 31.46 ± 3% turbostat.CPUGFX%
25.58 ± 2% -26.1% 18.90 ± 6% turbostat.CorWatt
63.17 ± 8% -19.0% 51.17 ± 4% turbostat.CoreTmp
53.56 ± 3% +41.1% 75.58 turbostat.GFX%C0
46.26 ± 3% -47.7% 24.20 turbostat.GFX%rc6
2987864 ± 2% +59.0% 4750693 ± 7% turbostat.IRQ
35.14 ± 2% -18.7% 28.58 ± 4% turbostat.PkgWatt
2.03 -32.5% 1.37 turbostat.RAMWatt
126.64 -35.7% 81.45 ± 3% turbostat.Totl%C0
79717 ± 22% +47.1% 117280 ± 17% sched_debug.cfs_rq:/.load.avg
90122 ± 47% +110.3% 189564 ± 31% sched_debug.cfs_rq:/.load.stddev
401.67 ± 28% -40.9% 237.39 ± 38% sched_debug.cfs_rq:/.runnable_avg.min
769.84 ± 29% -39.1% 468.52 ± 16% sched_debug.cfs_rq:/.util_avg.avg
343.92 ± 32% -47.0% 182.19 ± 42% sched_debug.cfs_rq:/.util_avg.min
362716 ± 50% +69.7% 615398 ± 6% sched_debug.cpu.avg_idle.avg
33737 ± 81% +418.8% 175028 ± 69% sched_debug.cpu.avg_idle.min
25768 ± 59% +157.1% 66256 ± 46% sched_debug.cpu.clock.avg
25769 ± 59% +157.1% 66257 ± 46% sched_debug.cpu.clock.max
25768 ± 59% +157.1% 66256 ± 46% sched_debug.cpu.clock.min
25412 ± 59% +157.5% 65432 ± 46% sched_debug.cpu.clock_task.avg
25493 ± 59% +157.0% 65529 ± 46% sched_debug.cpu.clock_task.max
25210 ± 59% +158.7% 65225 ± 46% sched_debug.cpu.clock_task.min
524.29 ± 21% +33.1% 697.73 ± 9% sched_debug.cpu.curr->pid.stddev
39852 ± 88% +124.7% 89528 ± 15% sched_debug.cpu.nr_switches.avg
90571 ± 91% +130.1% 208407 ± 19% sched_debug.cpu.nr_switches.max
23831 ± 96% +143.6% 58062 ± 15% sched_debug.cpu.nr_switches.stddev
25768 ± 59% +157.1% 66256 ± 46% sched_debug.cpu_clk
25675 ± 59% +157.7% 66162 ± 46% sched_debug.ktime
25777 ± 59% +157.1% 66270 ± 46% sched_debug.sched_clk
58793 ± 4% +32.5% 77897 proc-vmstat.nr_active_anon
47110 +4.4% 49167 proc-vmstat.nr_active_file
334594 -2.2% 327272 proc-vmstat.nr_dirty_background_threshold
670008 -2.2% 655345 proc-vmstat.nr_dirty_threshold
653776 +11.6% 729500 proc-vmstat.nr_file_pages
867002 ± 3% +10.7% 960133 proc-vmstat.nr_foll_pin_acquired
802295 ± 3% +15.1% 923766 proc-vmstat.nr_foll_pin_released
3070445 -2.5% 2994848 proc-vmstat.nr_free_pages
429213 -3.2% 415420 proc-vmstat.nr_inactive_anon
29814 +4.4% 31115 proc-vmstat.nr_mapped
331210 +22.2% 404672 proc-vmstat.nr_shmem
17641 +1.4% 17891 proc-vmstat.nr_slab_reclaimable
105399 ± 4% +69.2% 178325 proc-vmstat.nr_unevictable
58793 ± 4% +32.5% 77897 proc-vmstat.nr_zone_active_anon
47110 +4.4% 49167 proc-vmstat.nr_zone_active_file
429213 -3.2% 415420 proc-vmstat.nr_zone_inactive_anon
105399 ± 4% +69.2% 178325 proc-vmstat.nr_zone_unevictable
8880631 +1.3% 8998111 proc-vmstat.numa_hit
8880631 +1.3% 8998111 proc-vmstat.numa_local
13881506 +1.5% 14095193 proc-vmstat.pgalloc_normal
648319 ± 2% +24.5% 807099 ± 3% proc-vmstat.pgfault
13805211 +1.6% 14028593 proc-vmstat.pgfree
48449 +50.0% 72681 proc-vmstat.pgreuse
36.56 ± 16% +181.6% 102.95 ± 5% perf-stat.i.MPKI
7.516e+08 ± 2% -42.1% 4.354e+08 ± 2% perf-stat.i.branch-instructions
2.64 ± 11% +2.4 5.06 ± 9% perf-stat.i.branch-miss-rate%
17282700 ± 7% -33.2% 11545195 ± 4% perf-stat.i.branch-misses
33.38 ± 8% -13.2 20.21 ± 5% perf-stat.i.cache-miss-rate%
28095351 ± 2% -43.1% 15991999 perf-stat.i.cache-misses
93205045 ± 3% -30.8% 64537379 perf-stat.i.cache-references
19079 ± 3% -43.0% 10882 ± 2% perf-stat.i.context-switches
2.19 ± 5% +57.6% 3.45 ± 6% perf-stat.i.cpi
6.282e+09 -40.0% 3.768e+09 perf-stat.i.cpu-cycles
16.44 ± 5% -39.6% 9.93 ± 16% perf-stat.i.cpu-migrations
455.29 +33.2% 606.29 ± 2% perf-stat.i.cycles-between-cache-misses
0.32 ± 12% +0.5 0.83 ± 14% perf-stat.i.dTLB-load-miss-rate%
1562757 ± 8% -28.1% 1123527 ± 8% perf-stat.i.dTLB-load-misses
1.151e+09 -43.3% 6.524e+08 perf-stat.i.dTLB-loads
0.04 ± 31% +0.1 0.14 ± 22% perf-stat.i.dTLB-store-miss-rate%
5.394e+08 -43.3% 3.056e+08 perf-stat.i.dTLB-stores
1108690 ± 2% -35.6% 713796 ± 9% perf-stat.i.iTLB-load-misses
1761170 ± 3% -43.5% 995856 ± 6% perf-stat.i.iTLB-loads
4.451e+09 ± 2% -42.7% 2.55e+09 ± 2% perf-stat.i.instructions
12650 ± 5% -33.0% 8470 ± 9% perf-stat.i.instructions-per-iTLB-miss
0.63 ± 2% -27.0% 0.46 ± 2% perf-stat.i.ipc
21.06 ± 5% -45.4% 11.50 ± 5% perf-stat.i.major-faults
523546 -40.0% 314024 perf-stat.i.metric.GHz
2.125e+08 -42.5% 1.222e+08 perf-stat.i.metric.M/sec
6074 ± 5% -30.6% 4217 ± 2% perf-stat.i.minor-faults
5.53 ± 10% -35.4% 3.57 ± 36% perf-stat.i.node-load-misses
1100446 ± 2% -43.9% 617311 ± 2% perf-stat.i.node-loads
11366839 ± 3% -45.3% 6221416 ± 2% perf-stat.i.node-stores
6095 ± 5% -30.6% 4228 ± 2% perf-stat.i.page-faults
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://01.org/lkp
View attachment "config-5.11.0-rc5-01307-g58586680ffad" of type "text/plain" (156341 bytes)
View attachment "job-script" of type "text/plain" (7418 bytes)
View attachment "job.yaml" of type "text/plain" (4922 bytes)
View attachment "reproduce" of type "text/plain" (301 bytes)
Powered by blists - more mailing lists