Greeting,
FYI, we noticed a -4.0% regression of will-it-scale.per_thread_ops due to commit:
commit: c7299fea67696db5bd09d924d1f1080d894f92ef ("spi: Fix spi device unregister flow")
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: 100%
mode: thread
test: getppid1
cpufreq_governor: performance
ucode: 0x5003006
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
If you fix the issue, kindly add following tag
Reported-by: kernel test robot <[email protected]>
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 split-job --compatible job.yaml # generate the yaml file for lkp run
bin/lkp run generated-yaml-file
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode:
gcc-9/performance/x86_64-rhel-8.3/thread/100%/debian-10.4-x86_64-20200603.cgz/lkp-csl-2sp9/getppid1/will-it-scale/0x5003006
commit:
86527bcbc8 ("spi: Rename enable1 to activate in spi_set_cs()")
c7299fea67 ("spi: Fix spi device unregister flow")
86527bcbc88922ea c7299fea67696db5bd09d924d1f
---------------- ---------------------------
%stddev %change %stddev
\ | \
7.592e+08 -4.0% 7.289e+08 will-it-scale.88.threads
8626955 -4.0% 8283121 will-it-scale.per_thread_ops
7.592e+08 -4.0% 7.289e+08 will-it-scale.workload
0.49 ? 66% -68.8% 0.15 ? 77% perf-sched.wait_time.avg.ms.wait_for_partner.fifo_open.do_dentry_open.do_open.isra
9743 ? 26% +52.0% 14808 ? 25% proc-vmstat.numa_hint_faults
3.948e+10 -4.0% 3.792e+10 perf-stat.i.branch-instructions
1.897e+08 -4.1% 1.819e+08 perf-stat.i.branch-misses
1.36 +4.0% 1.41 perf-stat.i.cpi
625971 ? 3% -10.0% 563208 ? 7% perf-stat.i.cycles-between-cache-misses
6.138e+10 -3.9% 5.897e+10 perf-stat.i.dTLB-loads
0.00 +0.0 0.00 perf-stat.i.dTLB-store-miss-rate%
4.243e+10 -3.9% 4.076e+10 perf-stat.i.dTLB-stores
1.785e+11 -4.0% 1.714e+11 perf-stat.i.instructions
871.41 ? 3% -8.4% 798.42 perf-stat.i.instructions-per-iTLB-miss
0.74 -3.8% 0.71 perf-stat.i.ipc
1628 -3.9% 1564 perf-stat.i.metric.M/sec
42181 ? 2% +5.0% 44291 perf-stat.i.node-store-misses
0.02 ? 3% +12.2% 0.02 ? 4% perf-stat.overall.MPKI
1.36 +4.0% 1.41 perf-stat.overall.cpi
0.00 +0.0 0.00 perf-stat.overall.dTLB-store-miss-rate%
867.19 ? 4% -8.3% 795.25 perf-stat.overall.instructions-per-iTLB-miss
0.74 -3.8% 0.71 perf-stat.overall.ipc
3.935e+10 -4.0% 3.779e+10 perf-stat.ps.branch-instructions
1.891e+08 -4.1% 1.813e+08 perf-stat.ps.branch-misses
6.117e+10 -3.9% 5.877e+10 perf-stat.ps.dTLB-loads
4.229e+10 -3.9% 4.062e+10 perf-stat.ps.dTLB-stores
1.779e+11 -4.0% 1.709e+11 perf-stat.ps.instructions
42047 ? 2% +5.0% 44151 perf-stat.ps.node-store-misses
5.372e+13 -3.9% 5.161e+13 perf-stat.total.instructions
44.46 -1.7 42.80 perf-profile.calltrace.cycles-pp.__entry_text_start.getppid
9.26 ? 3% -0.6 8.67 perf-profile.calltrace.cycles-pp.testcase
3.08 -0.2 2.90 perf-profile.calltrace.cycles-pp.syscall_enter_from_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.getppid
8.27 -0.2 8.09 perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.getppid
3.11 -0.1 2.98 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_safe_stack.getppid
94.30 +0.4 94.74 perf-profile.calltrace.cycles-pp.getppid
1.59 ? 7% +0.5 2.04 ? 2% perf-profile.calltrace.cycles-pp.rcu_nocb_flush_deferred_wakeup.exit_to_user_mode_prepare.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe.getppid
1.66 ? 7% +1.1 2.81 ? 2% perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode_prepare.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe.getppid
5.75 ? 7% +1.8 7.50 perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe.getppid
37.24 +2.4 39.59 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.getppid
8.99 ? 5% +2.8 11.81 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.entry_SYSCALL_64_after_hwframe.getppid
28.71 -1.0 27.67 perf-profile.children.cycles-pp.__entry_text_start
24.10 -0.8 23.34 perf-profile.children.cycles-pp.syscall_return_via_sysret
5.51 ? 3% -0.4 5.14 ? 2% perf-profile.children.cycles-pp.testcase
3.12 -0.2 2.95 perf-profile.children.cycles-pp.syscall_enter_from_user_mode
3.32 -0.1 3.17 perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack
0.88 -0.0 0.84 perf-profile.children.cycles-pp.getppid@plt
1.04 +0.1 1.11 perf-profile.children.cycles-pp.rcu_read_unlock_strict
1.66 ? 7% +0.5 2.12 perf-profile.children.cycles-pp.rcu_nocb_flush_deferred_wakeup
1.70 ? 7% +1.2 2.86 ? 2% perf-profile.children.cycles-pp.syscall_exit_to_user_mode_prepare
6.20 ? 6% +1.7 7.86 perf-profile.children.cycles-pp.exit_to_user_mode_prepare
37.86 +2.3 40.19 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
10.12 ? 5% +2.8 12.91 perf-profile.children.cycles-pp.syscall_exit_to_user_mode
20.01 -0.9 19.14 perf-profile.self.cycles-pp.getppid
23.90 -0.7 23.15 perf-profile.self.cycles-pp.syscall_return_via_sysret
9.11 -0.7 8.43 perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe
12.94 -0.4 12.53 perf-profile.self.cycles-pp.__entry_text_start
2.67 -0.2 2.52 perf-profile.self.cycles-pp.syscall_enter_from_user_mode
3.31 -0.1 3.16 perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack
0.36 -0.0 0.34 perf-profile.self.cycles-pp.__x86_indirect_thunk_rax
0.71 ? 2% +0.0 0.76 ? 2% perf-profile.self.cycles-pp.rcu_read_unlock_strict
1.32 ? 9% +0.5 1.84 ? 2% perf-profile.self.cycles-pp.rcu_nocb_flush_deferred_wakeup
3.52 +0.5 4.07 perf-profile.self.cycles-pp.__x64_sys_getppid
4.49 ? 6% +1.2 5.64 perf-profile.self.cycles-pp.exit_to_user_mode_prepare
1.32 ? 9% +1.2 2.49 ? 3% perf-profile.self.cycles-pp.syscall_exit_to_user_mode_prepare
will-it-scale.88.threads
7.45e+08 +----------------------------------------------------------------+
|.++.+. .++.+.+ |
7.4e+08 |-+ ++.+ +.+.+ |
| |
7.35e+08 |-+ |
| |
7.3e+08 |-+ O O O O O O O O OO |
| O O O O O |
7.25e+08 |-+ O O O OO O O O O |
| O O O |
7.2e+08 |-+ |
| O |
7.15e+08 |-+O O OO O OO O OO |
| |
7.1e+08 +----------------------------------------------------------------+
will-it-scale.per_thread_ops
8.45e+06 +----------------------------------------------------------------+
| + .+ +. .+ |
8.4e+06 |-+ + + |
| |
8.35e+06 |-+ |
| |
8.3e+06 |-+ O O O |
| O OO OO O OO OO OO |
8.25e+06 |-+ |
| O O O OO O OO O OO |
8.2e+06 |-+ O |
| |
8.15e+06 |-O |
| O O OO O OO O OO |
8.1e+06 +----------------------------------------------------------------+
will-it-scale.workload
7.45e+08 +----------------------------------------------------------------+
|.++.+. .++.+.+ |
7.4e+08 |-+ ++.+ +.+.+ |
| |
7.35e+08 |-+ |
| |
7.3e+08 |-+ O O O O O O O O OO |
| O O O O O |
7.25e+08 |-+ O O O OO O O O O |
| O O O |
7.2e+08 |-+ |
| O |
7.15e+08 |-+O O OO O OO O OO |
| |
7.1e+08 +----------------------------------------------------------------+
[*] 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.
---
0DAY/LKP+ Test Infrastructure Open Source Technology Center
https://lists.01.org/hyperkitty/list/[email protected] Intel Corporation
Thanks,
Oliver Sang