| 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: <4b5d1c96-44c5-e8bb-01d8-9f9c72621d0d@intel.com>
Date: Tue, 30 Aug 2022 14:32:10 +0800
From: Yujie Liu <yujie.liu@...el.com>
To: Filipe Manana <fdmanana@...e.com>
CC: <lkp@...ts.01.org>, <lkp@...el.com>,
David Sterba <dsterba@...e.com>,
Anand Jain <anand.jain@...cle.com>,
Nikolay Borisov <nborisov@...e.com>,
<linux-kernel@...r.kernel.org>, <linux-btrfs@...r.kernel.org>,
<fengwei.yin@...el.com>, <regressions@...ts.linux.dev>
Subject: Re: [LKP] [btrfs] ca6dee6b79:
fxmark.ssd_btrfs_MWRM_72_bufferedio.works/sec -8.4% regression
Hi Filipe,
We noticed that this case was reported when the patch was in linux-next.
Thanks for your comment that it is an expected result due to heavy rename.
https://lore.kernel.org/all/Ysb4T7Z8hKgdvPRk@xsang-OptiPlex-9020/
This report is due to the patch being merged into mainline, if it is still
the same case, please ignore this duplicate report. Sorry for the inconvenience.
--
Thanks,
Yujie
On 8/30/2022 11:17, kernel test robot wrote:
> Greeting,
>
> FYI, we noticed a -8.4% regression of fxmark.ssd_btrfs_MWRM_72_bufferedio.works/sec due to commit:
>
>
> commit: ca6dee6b7946794fa340a7290ca399a50b61705f ("btrfs: balance btree dirty pages and delayed items after a rename")
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
>
> in testcase: fxmark
> on test machine: 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake) with 128G memory
> with following parameters:
>
> disk: 1SSD
> media: ssd
> test: MWRM
> fstype: btrfs
> directio: bufferedio
> cpufreq_governor: performance
> ucode: 0xd000363
>
> test-description: FxMark is a filesystem benchmark that test multicore scalability.
> test-url: https://github.com/sslab-gatech/fxmark
>
>
> =========================================================================================
> compiler/cpufreq_governor/directio/disk/fstype/kconfig/media/rootfs/tbox_group/test/testcase/ucode:
> gcc-11/performance/bufferedio/1SSD/btrfs/x86_64-rhel-8.3/ssd/debian-11.1-x86_64-20220510.cgz/lkp-icl-2sp5/MWRM/fxmark/0xd000363
>
> commit:
> b8bea09a45 ("btrfs: add trace event for submitted RAID56 bio")
> ca6dee6b79 ("btrfs: balance btree dirty pages and delayed items after a rename")
>
> b8bea09a456fc31a ca6dee6b7946794fa340a7290ca
> ---------------- ---------------------------
> %stddev %change %stddev
> \ | \
> 1821853 -13.9% 1568247 ± 3% fxmark.ssd_btrfs_MWRM_36_bufferedio.works
> 36436 -13.9% 31362 ± 3% fxmark.ssd_btrfs_MWRM_36_bufferedio.works/sec
> 1675102 -14.0% 1439994 ± 7% fxmark.ssd_btrfs_MWRM_54_bufferedio.works
> 33497 -14.0% 28796 ± 7% fxmark.ssd_btrfs_MWRM_54_bufferedio.works/sec
> 1572332 -8.4% 1440600 ± 6% fxmark.ssd_btrfs_MWRM_72_bufferedio.works
> 31445 -8.4% 28809 ± 6% fxmark.ssd_btrfs_MWRM_72_bufferedio.works/sec
> 356010 +80.0% 640832 fxmark.time.involuntary_context_switches
> 68.50 -24.1% 52.00 fxmark.time.percent_of_cpu_this_job_got
> 630.47 -24.0% 479.23 fxmark.time.system_time
> 1.335e+10 +49.8% 2e+10 cpuidle..time
> 1045 ± 4% +11.8% 1168 uptime.idle
> 31.54 +50.2% 47.37 iostat.cpu.idle
> 64.16 -24.7% 48.29 iostat.cpu.system
> 31.17 +50.3% 46.83 vmstat.cpu.id
> 12.83 ± 5% -55.8% 5.67 ± 8% vmstat.procs.r
> 32.13 +15.8 47.95 mpstat.cpu.all.idle%
> 0.47 ± 7% +0.1 0.53 ± 3% mpstat.cpu.all.iowait%
> 63.37 -16.1 47.31 mpstat.cpu.all.sys%
> 10.04 ± 3% +13.5% 11.39 ± 3% perf-stat.i.metric.K/sec
> 869.81 ± 10% -16.2% 728.74 ± 15% perf-stat.i.node-loads
> 871.23 ± 10% -16.2% 730.49 ± 15% perf-stat.ps.node-loads
> 3004 ± 8% -52.1% 1440 ± 6% numa-meminfo.node0.Active(anon)
> 1218568 -10.8% 1086453 numa-meminfo.node0.Inactive
> 351812 ± 3% -29.0% 249640 ± 12% numa-meminfo.node0.Inactive(anon)
> 120150 -79.3% 24861 ± 3% numa-meminfo.node0.Shmem
> 3489 ± 8% -45.0% 1919 ± 2% meminfo.Active(anon)
> 492107 -19.0% 398809 meminfo.Committed_AS
> 382253 -24.6% 288151 meminfo.Inactive(anon)
> 124727 -76.8% 28886 ± 2% meminfo.Shmem
> 2050 ± 4% -10.5% 1834 ± 5% meminfo.Writeback
> 750.83 ± 8% -52.1% 360.00 ± 6% numa-vmstat.node0.nr_active_anon
> 87951 ± 3% -29.0% 62408 ± 12% numa-vmstat.node0.nr_inactive_anon
> 30038 -79.3% 6216 ± 3% numa-vmstat.node0.nr_shmem
> 750.83 ± 8% -52.1% 360.00 ± 6% numa-vmstat.node0.nr_zone_active_anon
> 87951 ± 3% -29.0% 62408 ± 12% numa-vmstat.node0.nr_zone_inactive_anon
> 7554028 ± 3% -71.2% 2174126 ± 5% sched_debug.cfs_rq:/.min_vruntime.avg
> 7640393 ± 3% -70.5% 2254050 ± 5% sched_debug.cfs_rq:/.min_vruntime.max
> 7291209 ± 3% -73.6% 1926973 ± 5% sched_debug.cfs_rq:/.min_vruntime.min
> 873.62 ± 7% -19.2% 705.68 ± 10% sched_debug.cfs_rq:/.runnable_avg.avg
> 790.32 ± 7% -21.4% 621.34 ± 12% sched_debug.cfs_rq:/.runnable_avg.min
> 747.11 ± 3% -22.7% 577.37 ± 3% sched_debug.cfs_rq:/.util_avg.avg
> 670.92 ± 5% -25.2% 501.70 ± 2% sched_debug.cfs_rq:/.util_avg.min
> 409.44 ± 9% -35.1% 265.80 ± 21% sched_debug.cfs_rq:/.util_est_enqueued.avg
> 789.44 ± 3% -20.1% 630.53 ± 5% sched_debug.cfs_rq:/.util_est_enqueued.max
> 0.00 ± 13% -67.3% 0.00 ± 22% sched_debug.cpu.next_balance.stddev
> 872.67 ± 8% -45.0% 479.83 ± 2% proc-vmstat.nr_active_anon
> 1801345 -1.7% 1771330 proc-vmstat.nr_file_pages
> 95550 -24.6% 72037 proc-vmstat.nr_inactive_anon
> 8752 -3.7% 8426 proc-vmstat.nr_mapped
> 31169 -76.8% 7221 ± 2% proc-vmstat.nr_shmem
> 872.67 ± 8% -45.0% 479.83 ± 2% proc-vmstat.nr_zone_active_anon
> 95550 -24.6% 72037 proc-vmstat.nr_zone_inactive_anon
> 9553 ± 10% -16.8% 7950 ± 3% proc-vmstat.numa_hint_faults
> 18886391 -3.6% 18207624 proc-vmstat.numa_hit
> 18770999 -3.6% 18091363 proc-vmstat.numa_local
> 7398756 -4.0% 7105675 proc-vmstat.pgactivate
> 18885154 -3.6% 18206666 proc-vmstat.pgalloc_normal
> 7248262 -4.3% 6933915 ± 2% proc-vmstat.pgdeactivate
> 18894473 -3.4% 18243898 proc-vmstat.pgfree
> 7829962 -3.0% 7596447 ± 2% proc-vmstat.pgrotated
>
>
> If you fix the issue, kindly add following tag
> Reported-by: kernel test robot <yujie.liu@...el.com>
>
>
> 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.
>
>
> 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.
>
>
> #regzbot introduced: ca6dee6b79
>
>
>
> _______________________________________________
> LKP mailing list -- lkp@...ts.01.org
> To unsubscribe send an email to lkp-leave@...ts.01.org
Powered by blists - more mailing lists