Re: [lkp] [selinux] 759b31c1d4: unixbench.score -9.7% regression
From: Paul Moore
Date: Tue Apr 19 2016 - 17:10:50 EST
On Wed, Mar 30, 2016 at 10:38 PM, kernel test robot
<xiaolong.ye@xxxxxxxxx> wrote:
> FYI, we noticed that unixbench.score -9.7% regression on
>
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
> commit 759b31c1d4295a1dab34b4669ee838fb3131cd91 ("selinux: simply inode label states to INVALID and INITIALIZED")
Thanks for reporting this; it took a while, but I believe we have
resolved this in the SELinux next branch with three patches that were
merged today.
> =========================================================================================
> compiler/cpufreq_governor/kconfig/nr_task/rootfs/tbox_group/test/testcase:
> gcc-4.9/performance/x86_64-rhel/1/debian-x86_64-2015-02-07.cgz/ivb42/pipe/unixbench
>
> commit:
> 48c2353e841141683e13771ecd6684350004ccac
> 759b31c1d4295a1dab34b4669ee838fb3131cd91
>
> 48c2353e84114168 759b31c1d4295a1dab34b4669e
> ---------------- --------------------------
> %stddev %change %stddev
> \ | \
> 1972 Ä 0% -9.7% 1781 Ä 0% unixbench.score
> 16298 Ä 5% -12.4% 14272 Ä 5% cpuidle.C1-IVT.usage
> 10890 Ä 19% +84.8% 20129 Ä 28% cpuidle.POLL.usage
> 2394 Ä153% +93.8% 4641 Ä 90% numa-meminfo.node1.Inactive(anon)
> 2545 Ä146% +89.6% 4827 Ä 86% numa-meminfo.node1.Shmem
> 598.25 Ä153% +93.9% 1160 Ä 90% numa-vmstat.node1.nr_inactive_anon
> 636.00 Ä146% +89.7% 1206 Ä 86% numa-vmstat.node1.nr_shmem
> 19716 Ä 8% -17.5% 16267 Ä 12% sched_debug.cfs_rq:/.min_vruntime.stddev
> 19716 Ä 8% -17.5% 16267 Ä 12% sched_debug.cfs_rq:/.spread0.stddev
>
> lkp-wsx02: Westmere-EX
> Memory: 128G
>
>
>
> unixbench.score
>
> 2000 *+**-*--*-*-----------------*-**-*-**-**-*-**-*-**-------------------+
> | * **.**. + : *.*.**.**. |
> | *.**.*.** : : *.**.*
> 1950 ++ *.* |
> | |
> | |
> 1900 ++ |
> | |
> 1850 ++ |
> | |
> | |
> 1800 ++ O |
> | OO OO O O O OO O OO O |
> O OO O OO O |
> 1750 ++-------------------------------------------------------------------+
>
>
> [*] bisect-good sample
> [O] bisect-bad sample
>
> To reproduce:
>
> git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
> cd lkp-tests
> bin/lkp install job.yaml # job file is attached in this email
> bin/lkp run job.yaml
>
>
> 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,
> Xiaolong Ye
--
paul moore
www.paul-moore.com