From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============3232176220756546994==" MIME-Version: 1.0 From: Ye Xiaolong To: lkp@lists.01.org Subject: Re: [x86/copy_user] adb402cd14: will-it-scale.per_process_ops -12.7% regression Date: Mon, 14 Nov 2016 22:19:31 +0800 Message-ID: <20161114141931.GA27318@yexl-desktop> In-Reply-To: <20161115100101.nzvhd5mlcb747n2h@pd.tnic> List-Id: --===============3232176220756546994== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On 11/15, Borislav Petkov wrote: >On Mon, Nov 14, 2016 at 10:37:11PM +0800, Ye Xiaolong wrote: >> Below cmd should help on it: > >Right, before I go do this, can you please retest without > >CONFIG_DEBUG_ATOMIC_SLEEP=3Dy > >? Sorry for the late, here is the comparison, disable CONFIG_DEBUG_ATOMIC_SLEEP helps to restore the performance back according to the result. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D commit/compiler/cpufreq_governor/rootfs/tbox_group/test/testcase: adb402cd1461eef6e1a21db4532a3b9e6a6be853/gcc-6/performance/debian-x86_64-= 2016-08-31.cgz/xps/poll1/will-it-scale kconfig: x86_64-rhel-7.2 x86_64-rhel-7.2-CONFIG_DEBUG_ATOMIC_SLEEP x86_64-rhel-7.2 x86_64-rhel-7.2-CONFIG_DEB ---------------- -------------------------- %stddev %change %stddev \ | \ 5899566 =C2=B1 2% +28.5% 7583094 =C2=B1 0% will-it-scale.per_p= rocess_ops 5121156 =C2=B1 0% +21.1% 6203340 =C2=B1 0% will-it-scale.per_t= hread_ops 49191 =C2=B1127% -79.8% 9938 =C2=B1 3% will-it-scale.time.= involuntary_context_switches 588.63 =C2=B1 0% -4.1% 564.27 =C2=B1 0% will-it-scale.time.= system_time 132.53 =C2=B1 0% +18.7% 157.26 =C2=B1 0% will-it-scale.time.= user_time 23063 =C2=B1 0% -92.0% 1846 =C2=B1 0% interrupts.CAL:Func= tion_call_interrupts 10336 =C2=B1 28% -66.1% 3499 =C2=B1 5% softirqs.NET_RX 444.00 =C2=B1 0% -24.3% 336.00 =C2=B1 0% vmstat.memory.buff 49191 =C2=B1127% -79.8% 9938 =C2=B1 3% time.involuntary_co= ntext_switches 132.53 =C2=B1 0% +18.7% 157.26 =C2=B1 0% time.user_time 4823 =C2=B1 1% -9.9% 4344 =C2=B1 0% proc-vmstat.nr_mapp= ed 4994 =C2=B1 2% -13.2% 4336 =C2=B1 0% proc-vmstat.nr_shmem 2265 =C2=B1 7% -71.6% 642.67 =C2=B1 4% proc-vmstat.pgactiv= ate 61.33 =C2=B1 0% -1.6% 60.33 =C2=B1 0% turbostat.%Busy 12.34 =C2=B1 7% -98.8% 0.15 =C2=B1 17% turbostat.CPU%c3 1.29 =C2=B1 6% +1004.7% 14.22 =C2=B1 5% turbostat.CPU%c6 113064 =C2=B1 1% -9.4% 102395 =C2=B1 2% meminfo.Active 196087 =C2=B1 0% -16.4% 163999 =C2=B1 0% meminfo.Committed_AS 19288 =C2=B1 1% -9.9% 17372 =C2=B1 0% meminfo.Mapped 19983 =C2=B1 2% -13.2% 17348 =C2=B1 0% meminfo.Shmem 15975 =C2=B1 27% -74.7% 4043 =C2=B1 10% latency_stats.avg.m= ax Thanks, Xiaolong > >Thanks. > >-- = >Regards/Gruss, > Boris. > >Good mailing practices for 400: avoid top-posting and trim the reply. --===============3232176220756546994==--