| 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: <20220301143017.GB18164@xsang-OptiPlex-9020>
Date: Tue, 1 Mar 2022 22:30:17 +0800
From: kernel test robot <oliver.sang@...el.com>
To: Christophe Leroy <christophe.leroy@...roup.eu>
Cc: Steven Rostedt <rostedt@...dmis.org>,
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
Subject: [tracing] bc82c38a69: will-it-scale.per_process_ops 6.1%
improvement
Greeting,
FYI, we noticed a 6.1% improvement of will-it-scale.per_process_ops due to commit:
commit: bc82c38a6933aab308387d4aca47e0a05de7b553 ("tracing: Uninline trace_trigger_soft_disabled() partly")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: will-it-scale
on test machine: 88 threads 2 sockets Intel(R) Xeon(R) Gold 6238M CPU @ 2.10GHz with 128G memory
with following parameters:
nr_task: 50%
mode: process
test: open2
cpufreq_governor: performance
ucode: 0x500320a
test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
test-url: https://github.com/antonblanchard/will-it-scale
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/mode/nr_task/rootfs/tbox_group/test/testcase/ucode:
gcc-9/performance/x86_64-rhel-8.3/process/50%/debian-10.4-x86_64-20200603.cgz/lkp-csl-2sp9/open2/will-it-scale/0x500320a
commit:
b61edd5774 ("eprobes: Remove redundant event type information")
bc82c38a69 ("tracing: Uninline trace_trigger_soft_disabled() partly")
b61edd57740de589 bc82c38a6933aab308387d4aca4
---------------- ---------------------------
%stddev %change %stddev
\ | \
2514844 +6.1% 2668143 will-it-scale.44.processes
57155 +6.1% 60639 will-it-scale.per_process_ops
2514844 +6.1% 2668143 will-it-scale.workload
0.77 ± 2% +0.1 0.88 ± 2% mpstat.cpu.all.soft%
5930655 ± 7% +9.6% 6498445 ± 4% numa-vmstat.node0.numa_local
34.79 ± 42% -18.8 15.97 ± 75% turbostat.C1E%
1787 +3.2% 1845 vmstat.system.cs
4498 ± 20% +87.5% 8435 ± 34% proc-vmstat.numa_hint_faults_local
23027207 +5.9% 24396202 proc-vmstat.numa_hit
22948903 +6.0% 24318218 proc-vmstat.numa_local
23030185 +6.0% 24402075 proc-vmstat.pgalloc_normal
23043723 +6.0% 24416779 proc-vmstat.pgfree
5.75 -2.5% 5.60 perf-stat.i.MPKI
3.255e+09 +5.9% 3.446e+09 perf-stat.i.branch-instructions
0.44 +0.1 0.55 perf-stat.i.branch-miss-rate%
14404199 +32.9% 19146006 perf-stat.i.branch-misses
38.37 +0.8 39.20 perf-stat.i.cache-miss-rate%
36874960 +5.4% 38874899 perf-stat.i.cache-misses
96030811 +3.2% 99084523 perf-stat.i.cache-references
1701 +3.3% 1757 perf-stat.i.context-switches
7.46 -5.6% 7.05 perf-stat.i.cpi
132.55 +6.5% 141.18 ± 2% perf-stat.i.cpu-migrations
3386 -5.1% 3213 perf-stat.i.cycles-between-cache-misses
4.887e+09 +5.9% 5.176e+09 perf-stat.i.dTLB-loads
2.756e+09 +5.9% 2.918e+09 perf-stat.i.dTLB-stores
88.91 +3.2 92.14 perf-stat.i.iTLB-load-miss-rate%
12033536 +42.4% 17140878 perf-stat.i.iTLB-load-misses
1496993 -2.7% 1457072 perf-stat.i.iTLB-loads
1.671e+10 +5.9% 1.769e+10 perf-stat.i.instructions
1391 -25.6% 1034 perf-stat.i.instructions-per-iTLB-miss
0.13 +5.9% 0.14 perf-stat.i.ipc
124.91 +5.9% 132.25 perf-stat.i.metric.M/sec
62.97 -3.0 59.95 perf-stat.i.node-store-miss-rate%
2487432 ± 3% +13.6% 2826245 ± 3% perf-stat.i.node-stores
5.75 -2.5% 5.60 perf-stat.overall.MPKI
0.45 +0.1 0.56 perf-stat.overall.branch-miss-rate%
38.42 +0.8 39.25 perf-stat.overall.cache-miss-rate%
7.45 -5.5% 7.04 perf-stat.overall.cpi
3376 -5.1% 3204 perf-stat.overall.cycles-between-cache-misses
88.94 +3.2 92.17 perf-stat.overall.iTLB-load-miss-rate%
1389 -25.7% 1032 perf-stat.overall.instructions-per-iTLB-miss
0.13 +5.9% 0.14 perf-stat.overall.ipc
62.92 -3.0 59.91 perf-stat.overall.node-store-miss-rate%
3.246e+09 +5.9% 3.436e+09 perf-stat.ps.branch-instructions
14474992 +32.4% 19164228 perf-stat.ps.branch-misses
36784615 +5.4% 38766783 perf-stat.ps.cache-misses
95735715 +3.2% 98772627 perf-stat.ps.cache-references
1695 +3.3% 1750 perf-stat.ps.context-switches
132.26 +6.5% 140.89 ± 2% perf-stat.ps.cpu-migrations
4.872e+09 +5.9% 5.16e+09 perf-stat.ps.dTLB-loads
2.747e+09 +5.9% 2.909e+09 perf-stat.ps.dTLB-stores
11994113 +42.4% 17083388 perf-stat.ps.iTLB-load-misses
1491471 -2.6% 1451966 perf-stat.ps.iTLB-loads
1.666e+10 +5.9% 1.764e+10 perf-stat.ps.instructions
2482902 ± 3% +13.5% 2819135 ± 3% perf-stat.ps.node-stores
5.034e+12 +5.9% 5.33e+12 perf-stat.total.instructions
0.85 ± 11% +0.1 0.97 perf-profile.calltrace.cycles-pp.kmem_cache_alloc.__alloc_file.alloc_empty_file.path_openat.do_filp_open
0.36 ± 70% +0.2 0.58 perf-profile.calltrace.cycles-pp.link_path_walk.path_openat.do_filp_open.do_sys_openat2.do_sys_open
9.00 ± 10% +1.1 10.13 perf-profile.calltrace.cycles-pp.security_file_free.__fput.task_work_run.exit_to_user_mode_prepare.syscall_exit_to_user_mode
8.98 ± 10% +1.1 10.12 perf-profile.calltrace.cycles-pp.apparmor_file_free_security.security_file_free.__fput.task_work_run.exit_to_user_mode_prepare
9.30 ± 11% +1.2 10.48 perf-profile.calltrace.cycles-pp.__fput.task_work_run.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64
9.54 ± 10% +1.2 10.74 perf-profile.calltrace.cycles-pp.task_work_run.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
9.62 ± 10% +1.2 10.83 perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__close
9.64 ± 10% +1.2 10.84 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.__close
9.91 ± 10% +1.2 11.15 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__close
9.93 ± 10% +1.2 11.18 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__close
10.17 ± 10% +1.3 11.44 perf-profile.calltrace.cycles-pp.__close
9.50 ± 10% +1.3 10.79 perf-profile.calltrace.cycles-pp.apparmor_file_alloc_security.security_file_alloc.__alloc_file.alloc_empty_file.path_openat
9.64 ± 10% +1.3 10.95 perf-profile.calltrace.cycles-pp.security_file_alloc.__alloc_file.alloc_empty_file.path_openat.do_filp_open
10.73 ± 11% +1.5 12.20 perf-profile.calltrace.cycles-pp.__alloc_file.alloc_empty_file.path_openat.do_filp_open.do_sys_openat2
10.76 ± 11% +1.5 12.23 perf-profile.calltrace.cycles-pp.alloc_empty_file.path_openat.do_filp_open.do_sys_openat2.do_sys_open
18.80 ± 10% +2.5 21.26 perf-profile.calltrace.cycles-pp.apparmor_current_getsecid_subj.security_current_getsecid_subj.ima_file_check.do_open.path_openat
18.83 ± 10% +2.5 21.30 perf-profile.calltrace.cycles-pp.security_current_getsecid_subj.ima_file_check.do_open.path_openat.do_filp_open
18.86 ± 10% +2.5 21.33 perf-profile.calltrace.cycles-pp.ima_file_check.do_open.path_openat.do_filp_open.do_sys_openat2
18.67 ± 10% +2.6 21.30 perf-profile.calltrace.cycles-pp.apparmor_file_open.security_file_open.do_dentry_open.do_open.path_openat
18.70 ± 10% +2.6 21.34 perf-profile.calltrace.cycles-pp.security_file_open.do_dentry_open.do_open.path_openat.do_filp_open
0.18 ± 9% +0.0 0.20 ± 4% perf-profile.children.cycles-pp.__might_resched
0.18 ± 10% +0.0 0.21 ± 2% perf-profile.children.cycles-pp.page_counter_charge
0.11 ± 13% +0.0 0.15 ± 10% perf-profile.children.cycles-pp.__slab_free
0.11 ± 14% +0.0 0.15 ± 7% perf-profile.children.cycles-pp.file_free_rcu
0.30 ± 9% +0.0 0.33 ± 2% perf-profile.children.cycles-pp.strncpy_from_user
0.22 ± 10% +0.0 0.26 ± 3% perf-profile.children.cycles-pp.obj_cgroup_charge
0.20 ± 11% +0.0 0.24 ± 3% perf-profile.children.cycles-pp.obj_cgroup_charge_pages
0.16 ± 13% +0.0 0.20 ± 5% perf-profile.children.cycles-pp.propagate_protected_usage
0.53 ± 10% +0.1 0.60 ± 2% perf-profile.children.cycles-pp.link_path_walk
1.03 ± 9% +0.1 1.16 ± 2% perf-profile.children.cycles-pp.irq_exit_rcu
1.08 ± 11% +0.1 1.22 perf-profile.children.cycles-pp.kmem_cache_alloc
0.41 ± 10% +0.2 0.56 ± 6% perf-profile.children.cycles-pp.page_counter_cancel
0.16 ± 22% +0.2 0.33 ± 30% perf-profile.children.cycles-pp.ret_from_fork
0.16 ± 22% +0.2 0.33 ± 30% perf-profile.children.cycles-pp.kthread
0.13 ± 23% +0.2 0.31 ± 32% perf-profile.children.cycles-pp.run_ksoftirqd
0.13 ± 22% +0.2 0.31 ± 32% perf-profile.children.cycles-pp.smpboot_thread_fn
0.56 ± 9% +0.2 0.76 ± 6% perf-profile.children.cycles-pp.obj_cgroup_uncharge_pages
0.55 ± 10% +0.2 0.74 ± 6% perf-profile.children.cycles-pp.page_counter_uncharge
0.83 ± 10% +0.2 1.06 ± 5% perf-profile.children.cycles-pp.kmem_cache_free
1.14 ± 10% +0.3 1.46 ± 5% perf-profile.children.cycles-pp.__softirqentry_text_start
1.14 ± 10% +0.3 1.46 ± 5% perf-profile.children.cycles-pp.rcu_do_batch
1.14 ± 10% +0.3 1.46 ± 5% perf-profile.children.cycles-pp.rcu_core
9.00 ± 10% +1.1 10.13 perf-profile.children.cycles-pp.apparmor_file_free_security
9.00 ± 10% +1.1 10.14 perf-profile.children.cycles-pp.security_file_free
9.32 ± 10% +1.2 10.50 perf-profile.children.cycles-pp.__fput
9.56 ± 10% +1.2 10.76 perf-profile.children.cycles-pp.task_work_run
9.68 ± 10% +1.2 10.89 perf-profile.children.cycles-pp.syscall_exit_to_user_mode
9.65 ± 10% +1.2 10.86 perf-profile.children.cycles-pp.exit_to_user_mode_prepare
10.23 ± 10% +1.3 11.50 perf-profile.children.cycles-pp.__close
9.52 ± 10% +1.3 10.80 perf-profile.children.cycles-pp.apparmor_file_alloc_security
9.65 ± 10% +1.3 10.96 perf-profile.children.cycles-pp.security_file_alloc
10.74 ± 11% +1.5 12.21 perf-profile.children.cycles-pp.__alloc_file
10.77 ± 11% +1.5 12.24 perf-profile.children.cycles-pp.alloc_empty_file
18.84 ± 10% +2.5 21.31 perf-profile.children.cycles-pp.security_current_getsecid_subj
18.81 ± 10% +2.5 21.28 perf-profile.children.cycles-pp.apparmor_current_getsecid_subj
18.86 ± 10% +2.5 21.34 perf-profile.children.cycles-pp.ima_file_check
18.68 ± 10% +2.6 21.32 perf-profile.children.cycles-pp.apparmor_file_open
18.70 ± 10% +2.6 21.34 perf-profile.children.cycles-pp.security_file_open
6.94 ± 11% -3.3 3.64 ± 3% perf-profile.self.cycles-pp.do_dentry_open
0.06 ± 11% +0.0 0.08 perf-profile.self.cycles-pp.do_sys_openat2
0.12 ± 9% +0.0 0.14 ± 3% perf-profile.self.cycles-pp.strncpy_from_user
0.11 ± 14% +0.0 0.14 ± 11% perf-profile.self.cycles-pp.__slab_free
0.11 ± 14% +0.0 0.15 ± 7% perf-profile.self.cycles-pp.file_free_rcu
0.22 ± 11% +0.0 0.26 ± 4% perf-profile.self.cycles-pp.__alloc_file
0.16 ± 12% +0.0 0.20 ± 6% perf-profile.self.cycles-pp.propagate_protected_usage
0.41 ± 10% +0.2 0.56 ± 6% perf-profile.self.cycles-pp.page_counter_cancel
8.82 ± 10% +1.1 9.92 perf-profile.self.cycles-pp.apparmor_file_free_security
9.30 ± 10% +1.2 10.55 perf-profile.self.cycles-pp.apparmor_file_alloc_security
18.44 ± 10% +2.4 20.85 perf-profile.self.cycles-pp.apparmor_current_getsecid_subj
18.32 ± 10% +2.6 20.88 perf-profile.self.cycles-pp.apparmor_file_open
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.
---
0DAY/LKP+ Test Infrastructure Open Source Technology Center
https://lists.01.org/hyperkitty/list/lkp@lists.01.org Intel Corporation
Thanks,
Oliver Sang
View attachment "config-5.17.0-rc1-00021-gbc82c38a6933" of type "text/plain" (162048 bytes)
View attachment "job-script" of type "text/plain" (7989 bytes)
View attachment "job.yaml" of type "text/plain" (5316 bytes)
View attachment "reproduce" of type "text/plain" (341 bytes)
Powered by blists - more mailing lists