Hello,
kernel test robot noticed a 4.0% improvement of unixbench.throughput on:
commit: 7b8a74bc42016173ddb4872d113f132c8ca59fc1 ("drm/ast: Program mode for AST DP in atomic_mode_set")
git://anongit.freedesktop.org/drm/drm-misc for-linux-next
testcase: unixbench
test machine: 224 threads 2 sockets Intel(R) Xeon(R) Platinum 8480CTDX (Sapphire Rapids) with 256G memory
parameters:
runtime: 300s
nr_task: 1
test: 2d-window
cpufreq_governor: performance
Details are as below:
-------------------------------------------------------------------------------------------------->
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20240708/202407081414.8fc2a8f-oliver.sang@xxxxxxxxx
=========================================================================================
compiler/cpufreq_governor/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase:
gcc-13/performance/x86_64-rhel-8.3/1/debian-12-x86_64-20240206.cgz/300s/lkp-spr-r02/2d-window/unixbench
commit:
2a7e5f4531 ("drm/ast: Implement atomic enable/disable for encoders")
7b8a74bc42 ("drm/ast: Program mode for AST DP in atomic_mode_set")
2a7e5f45314cfd99 7b8a74bc42016173ddb4872d113
---------------- ---------------------------
%stddev %change %stddev
\ | \
0.18 -0.0 0.16 mpstat.cpu.all.sys%
2094 +4.0% 2177 unixbench.score
3141 +4.0% 3266 unixbench.throughput
215.57 ± 2% -14.9% 183.46 ± 2% unixbench.time.system_time
318.37 +10.8% 352.81 unixbench.time.user_time
1753586 +4.3% 1829782 unixbench.workload
2.037e+09 +12.4% 2.289e+09 perf-stat.i.branch-instructions
0.56 ± 2% -0.0 0.54 perf-stat.i.branch-miss-rate%
11300458 +2.9% 11628854 perf-stat.i.branch-misses
0.65 -7.8% 0.60 perf-stat.i.cpi
9.653e+09 +12.0% 1.081e+10 perf-stat.i.instructions
2.19 +12.3% 2.46 perf-stat.i.ipc
0.03 ± 4% -11.6% 0.03 ± 5% perf-stat.overall.MPKI
0.56 ± 2% -0.0 0.51 perf-stat.overall.branch-miss-rate%
0.47 -10.5% 0.42 perf-stat.overall.cpi
2.12 +11.7% 2.37 perf-stat.overall.ipc
3086774 +7.8% 3326418 perf-stat.overall.path-length
2.031e+09 +12.4% 2.283e+09 perf-stat.ps.branch-instructions
11277892 +2.9% 11606260 perf-stat.ps.branch-misses
9.625e+09 +12.0% 1.078e+10 perf-stat.ps.instructions
5.413e+12 ± 2% +12.4% 6.086e+12 perf-stat.total.instructions
0.24 ± 43% -0.2 0.00 perf-profile.children.cycles-pp.ast_dp_set_mode
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://github.com/intel/lkp-tests/wiki
