| 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
| ||
|
Date: Tue, 30 Jun 2020 17:26:05 +0800
From: kernel test robot <rong.a.chen@...el.com>
To: Sayali Lokhande <sayalil@...eaurora.org>
Cc: Jaegeuk Kim <jaegeuk@...nel.org>, Chao Yu <yuchao0@...wei.com>,
LKML <linux-kernel@...r.kernel.org>,
linux-f2fs-devel@...ts.sourceforge.net, lkp@...ts.01.org
Subject: [f2fs] 2913aa304c: filebench.sum_bytes_mb/s -41.7% regression
Greeting,
FYI, we noticed a -41.7% regression of filebench.sum_bytes_mb/s due to commit:
commit: 2913aa304cefe4033228a315bc497983b3f3be08 ("f2fs: Avoid double lock for cp_rwsem during checkpoint")
https://git.kernel.org/cgit/linux/kernel/git/jaegeuk/f2fs.git dev
in testcase: filebench
on test machine: 8 threads Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz with 8G memory
with following parameters:
disk: 1HDD
fs: f2fs
test: createfiles.f
cpufreq_governor: performance
ucode: 0x27
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
=========================================================================================
compiler/cpufreq_governor/disk/fs/kconfig/rootfs/tbox_group/test/testcase/ucode:
gcc-7/performance/1HDD/f2fs/x86_64-rhel-7.6/debian-x86_64-20191114.cgz/lkp-hsw-d01/createfiles.f/filebench/0x27
commit:
435cbab95e ("f2fs: fix quota_sync failure due to f2fs_lock_op")
2913aa304c ("f2fs: Avoid double lock for cp_rwsem during checkpoint")
435cbab95e3966cd 2913aa304cefe4033228a315bc4
---------------- ---------------------------
fail:runs %reproduction fail:runs
| | |
:4 25% 1:4 dmesg.RIP:drm_wait_one_vblank[drm]
:4 25% 1:4 dmesg.WARNING:at_drivers/gpu/drm/drm_vblank.c:#drm_wait_one_vblank[drm]
:4 25% 1:4 kmsg.drm:drm_atomic_helper_wait_for_dependencies[drm_kms_helper]]*ERROR*[CRTC:#:pipe_A]flip_done_timed_out
:4 25% 1:4 kmsg.drm:drm_atomic_helper_wait_for_dependencies[drm_kms_helper]]*ERROR*[PLANE:#:primary_A]flip_done_timed_out
:4 25% 1:4 kmsg.drm:drm_atomic_helper_wait_for_flip_done[drm_kms_helper]]*ERROR*[CRTC:#:pipe_A]flip_done_timed_out
%stddev %change %stddev
\ | \
390.50 -41.7% 227.77 ± 24% filebench.sum_bytes_mb/s
74981 -41.7% 43735 ± 24% filebench.sum_operations/s
24991 -41.7% 14578 ± 24% filebench.sum_writes/s
11.18 ± 2% +18.5% 13.25 ± 7% filebench.time.elapsed_time
11.18 ± 2% +18.5% 13.25 ± 7% filebench.time.elapsed_time.max
2595080 +6.5% 2762632 ± 2% filebench.time.file_system_outputs
1631 ± 2% +13.4% 1849 filebench.time.involuntary_context_switches
19660 ± 8% +16.5% 22896 ± 4% filebench.time.voluntary_context_switches
5.38 ± 48% -2.2 3.22 ± 9% mpstat.cpu.all.usr%
8409 ± 15% -27.5% 6094 ± 8% softirqs.CPU0.RCU
10405 ± 22% +31.4% 13675 ± 18% softirqs.CPU6.TIMER
14597804 ± 21% +27.5% 18612441 ± 14% cpuidle.C1E.time
211107 ± 17% +24.2% 262299 ± 10% cpuidle.C1E.usage
6265 ± 12% +29.8% 8133 ± 10% cpuidle.POLL.usage
14527 -15.7% 12244 ± 8% vmstat.io.bi
20825 +56.8% 32658 ± 11% vmstat.io.bo
1385202 +13.4% 1570761 ± 3% vmstat.memory.cache
32357 +3.0% 33343 vmstat.system.cs
4644 ± 8% +16.6% 5415 interrupts.19:IR-IO-APIC.19-fasteoi.ata_piix
135.25 ± 35% +137.5% 321.25 ± 17% interrupts.CPU0.RES:Rescheduling_interrupts
85.50 ± 41% +86.5% 159.50 ± 10% interrupts.CPU1.RES:Rescheduling_interrupts
35730 ± 20% +76.0% 62883 ± 14% interrupts.CPU3.LOC:Local_timer_interrupts
368385 ± 2% +24.7% 459371 ± 11% interrupts.LOC:Local_timer_interrupts
668162 ± 64% -70.7% 196041 ± 17% sched_debug.cfs_rq:/.load.max
198442 ± 61% -72.2% 55082 ± 25% sched_debug.cfs_rq:/.load.stddev
67.25 ± 36% +53.2% 103.00 ± 12% sched_debug.cfs_rq:/.load_avg.min
143.22 ± 30% -55.7% 63.38 ± 48% sched_debug.cfs_rq:/.util_est_enqueued.avg
1.51 ± 24% -47.2% 0.79 ± 23% sched_debug.cpu.nr_running.stddev
7.24 ± 2% +19.9% 8.68 ± 11% iostat.cpu.iowait
6.13 ± 44% -35.5% 3.96 ± 9% iostat.cpu.system
5.31 ± 48% -38.3% 3.28 ± 8% iostat.cpu.user
1.24 +7.4% 1.33 ± 5% iostat.sda.avgqu-sz
17.35 +182.7% 49.05 ± 60% iostat.sda.await.max
14563 -15.7% 12271 ± 8% iostat.sda.rkB/s
3335 -16.1% 2798 ± 8% iostat.sda.rrqm/s
8.68 +408.0% 44.11 ± 77% iostat.sda.svctm.max
34.87 ± 2% +81.2% 63.18 ± 6% iostat.sda.w/s
17.35 +183.9% 49.25 ± 59% iostat.sda.w_await.max
22803 +50.9% 34402 ± 10% iostat.sda.wkB/s
323829 +19.0% 385387 ± 6% meminfo.Active
38849 ± 6% +96.3% 76253 ± 14% meminfo.Active(file)
1337341 +13.1% 1512358 ± 3% meminfo.Cached
142980 ± 4% +11.1% 158852 ± 5% meminfo.DirectMap4k
158412 ± 5% +86.1% 294859 ± 10% meminfo.Dirty
200125 ± 5% +73.8% 347833 ± 13% meminfo.Inactive
153146 ± 6% +96.4% 300834 ± 14% meminfo.Inactive(file)
63813 +30.7% 83421 ± 6% meminfo.KReclaimable
1822868 +12.5% 2050046 ± 3% meminfo.Memused
63813 +30.7% 83421 ± 6% meminfo.SReclaimable
98413 +19.8% 117935 ± 4% meminfo.Slab
219853 -16.8% 183008 ± 8% meminfo.max_used_kB
9692 ± 6% +96.7% 19065 ± 14% proc-vmstat.nr_active_file
322980 ± 3% +18.4% 382341 ± 4% proc-vmstat.nr_dirtied
39504 ± 5% +86.3% 73614 ± 10% proc-vmstat.nr_dirty
334308 +13.1% 378141 ± 3% proc-vmstat.nr_file_pages
1557882 -3.7% 1500980 proc-vmstat.nr_free_pages
38201 ± 6% +96.9% 75215 ± 14% proc-vmstat.nr_inactive_file
15941 +30.7% 20831 ± 6% proc-vmstat.nr_slab_reclaimable
115.25 ± 22% +78.1% 205.25 ± 10% proc-vmstat.nr_writeback
71036 ± 2% +92.0% 136396 ± 16% proc-vmstat.nr_written
9689 ± 6% +96.8% 19065 ± 14% proc-vmstat.nr_zone_active_file
38198 ± 6% +96.9% 75212 ± 14% proc-vmstat.nr_zone_inactive_file
39620 ± 5% +86.3% 73815 ± 10% proc-vmstat.nr_zone_write_pending
59065 +1.4% 59908 proc-vmstat.pgactivate
284557 ± 2% +91.7% 545541 ± 16% proc-vmstat.pgpgout
6.681e+08 ± 18% -24.9% 5.02e+08 ± 6% perf-stat.i.branch-instructions
24545186 ± 2% -19.8% 19685027 ± 9% perf-stat.i.branch-misses
12.59 ± 8% +2.2 14.83 ± 6% perf-stat.i.cache-miss-rate%
4.189e+09 ± 36% -32.2% 2.84e+09 ± 7% perf-stat.i.cpu-cycles
3044 ± 55% -43.8% 1711 ± 16% perf-stat.i.cycles-between-cache-misses
7.31e+08 ± 22% -26.0% 5.411e+08 ± 4% perf-stat.i.dTLB-loads
3.809e+08 ± 40% -31.3% 2.618e+08 ± 3% perf-stat.i.dTLB-stores
503093 ± 6% -9.0% 457752 ± 7% perf-stat.i.iTLB-loads
3.198e+09 ± 17% -24.6% 2.41e+09 ± 6% perf-stat.i.instructions
0.52 ± 36% -32.2% 0.36 ± 7% perf-stat.i.metric.GHz
225.24 ± 24% -26.5% 165.45 ± 4% perf-stat.i.metric.M/sec
353470 ±171% -99.3% 2473 ± 4% perf-stat.i.minor-faults
1038357 ± 4% +19.4% 1239750 ± 3% perf-stat.i.node-loads
353470 ±171% -99.3% 2473 ± 4% perf-stat.i.page-faults
5.47 ± 23% +20.8% 6.61 ± 4% perf-stat.overall.MPKI
10.32 ± 7% +1.6 11.89 ± 8% perf-stat.overall.cache-miss-rate%
2483 ± 42% -38.8% 1519 ± 12% perf-stat.overall.cycles-between-cache-misses
6.111e+08 ± 17% -23.5% 4.673e+08 ± 5% perf-stat.ps.branch-instructions
22458139 ± 2% -18.5% 18301777 ± 8% perf-stat.ps.branch-misses
31948 ± 2% +4.2% 33293 ± 2% perf-stat.ps.context-switches
7312 +1.8% 7447 perf-stat.ps.cpu-clock
3.829e+09 ± 35% -30.9% 2.646e+09 ± 6% perf-stat.ps.cpu-cycles
6.683e+08 ± 22% -24.6% 5.037e+08 ± 4% perf-stat.ps.dTLB-loads
3.481e+08 ± 40% -29.9% 2.439e+08 ± 2% perf-stat.ps.dTLB-stores
459741 ± 6% -7.4% 425929 ± 6% perf-stat.ps.iTLB-loads
2.925e+09 ± 17% -23.3% 2.243e+09 ± 5% perf-stat.ps.instructions
321419 ±171% -99.3% 2324 ± 3% perf-stat.ps.minor-faults
949112 ± 4% +21.5% 1153505 ± 3% perf-stat.ps.node-loads
321419 ±171% -99.3% 2324 ± 3% perf-stat.ps.page-faults
7312 +1.8% 7447 perf-stat.ps.task-clock
5959 ± 14% +28.2% 7639 ± 6% slabinfo.anon_vma_chain.active_objs
6690 ± 8% +17.7% 7875 ± 2% slabinfo.anon_vma_chain.num_objs
51921 +17.1% 60797 ± 4% slabinfo.dentry.active_objs
2522 +16.3% 2932 ± 4% slabinfo.dentry.active_slabs
52971 +16.3% 61595 ± 4% slabinfo.dentry.num_objs
2522 +16.3% 2932 ± 4% slabinfo.dentry.num_slabs
11001 ± 11% +80.0% 19802 ± 14% slabinfo.f2fs_extent_tree.active_objs
219.00 ± 11% +78.3% 390.50 ± 13% slabinfo.f2fs_extent_tree.active_slabs
11190 ± 11% +78.2% 19935 ± 13% slabinfo.f2fs_extent_tree.num_objs
219.00 ± 11% +78.3% 390.50 ± 13% slabinfo.f2fs_extent_tree.num_slabs
43658 ± 2% +17.9% 51468 ± 6% slabinfo.f2fs_free_nid.active_objs
44212 ± 2% +17.2% 51811 ± 6% slabinfo.f2fs_free_nid.num_objs
10353 ± 6% +88.6% 19521 ± 14% slabinfo.f2fs_inode_cache.active_objs
371.50 ± 6% +88.0% 698.50 ± 14% slabinfo.f2fs_inode_cache.active_slabs
10416 ± 6% +87.9% 19569 ± 14% slabinfo.f2fs_inode_cache.num_objs
371.50 ± 6% +88.0% 698.50 ± 14% slabinfo.f2fs_inode_cache.num_slabs
44085 ± 2% +37.6% 60678 ± 9% slabinfo.f2fs_nat_entry.active_objs
344.50 ± 2% +38.2% 476.25 ± 9% slabinfo.f2fs_nat_entry.active_slabs
44164 ± 2% +38.2% 61034 ± 9% slabinfo.f2fs_nat_entry.num_objs
344.50 ± 2% +38.2% 476.25 ± 9% slabinfo.f2fs_nat_entry.num_slabs
607.75 +29.1% 784.75 ± 11% slabinfo.f2fs_nat_entry_set.active_objs
607.75 +29.1% 784.75 ± 11% slabinfo.f2fs_nat_entry_set.num_objs
1090 ± 2% -11.6% 964.00 ± 2% slabinfo.kmalloc-2k.active_objs
1090 ± 2% -11.5% 964.50 slabinfo.kmalloc-2k.num_objs
23236 ± 2% +40.5% 32656 ± 9% slabinfo.radix_tree_node.active_objs
859.00 +43.0% 1228 ± 7% slabinfo.radix_tree_node.active_slabs
24065 +42.9% 34395 ± 7% slabinfo.radix_tree_node.num_objs
859.00 +43.0% 1228 ± 7% slabinfo.radix_tree_node.num_slabs
519.00 ± 10% -100.0% 0.00 slabinfo.skbuff_ext_cache.active_objs
572.25 ± 5% -100.0% 0.00 slabinfo.skbuff_ext_cache.num_objs
filebench.sum_operations_s
80000 +-------------------------------------------------------------------+
| ++.+.+.++ + +.+.+.++ +.+ ++.+ ++ + +.++.+.+.++.|
70000 |-: : : : : : : : : :: : : |
60000 |-: : : : : : : : : :: : : |
| : : : : : : : : : :: : : |
50000 |:OO O : : : : : : :O O O: : :OO: O :: : |
|: : : : : : : : : : : : :: : |
40000 |:+ :O: :O: O OO:O:O : : : : : :: : |
|: : : :: : : : : : : : : : : |
30000 |:+ O: : :: : : : : : : : : : : |
20000 |:+ O : : :: : : : :O O: : :O : : : |
| : : : : : : : : : |
10000 |-+ : : : : : : : : : |
| : : : : : : : : : |
0 +-------------------------------------------------------------------+
filebench.sum_writes_s
25000 +-------------------------------------------------------------------+
| : : : : : : : : : :: : : |
| : : : : : : : : : :: : : |
20000 |-: : : : : : : : : :: : : |
|: : : : : : : : : : : : :: : |
|:OO O : : : : : : :O O O: : :OO: O :: : |
15000 |:+ : : : : : : : : : : : :: : |
|: :O: :O: O OO:O:O : : : : : :: : |
10000 |:+ O: : :: : : : : : : : : : : |
|: : : :: : : : :O O: : :O : : : |
|: O : : :: : : : : : : : : : : |
5000 |-+ : : : : : : : : : |
| : : : : : : : : : |
| : : : : : : : : : |
0 +-------------------------------------------------------------------+
filebench.sum_bytes_mb_s
400 +---------------------------------------------------------------------+
| : : : : : : : : : : : : : |
350 |-: : : : : : : : : : : : : |
300 |-: : : : : : : : : : : : : |
|: : :: : : : : : : : : : : : |
250 |:O O O : :: : : : :O O O: : :O : O : : : |
|: : :: : : : : : : : : : : : |
200 |:+ :O::O:O O O O:O:O : : : : : : : : |
|: :: : : : : : : : : : : :: |
150 |:+ O:: : : : : : :O O: : : : :: |
100 |:+ O :: : : : : : : : : : : :: |
| : : : : : : : : : |
50 |-+ : : : : : : : : : |
| : : : : : : : : : |
0 +---------------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
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.
Thanks,
Rong Chen
View attachment "config-5.6.0-11868-g2913aa304cefe" of type "text/plain" (202591 bytes)
View attachment "job-script" of type "text/plain" (7812 bytes)
View attachment "job.yaml" of type "text/plain" (5428 bytes)
View attachment "reproduce" of type "text/plain" (572 bytes)
Powered by blists - more mailing lists