From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============4754889347820234460==" MIME-Version: 1.0 From: Philip Li To: lkp@lists.01.org Subject: Re: [x86/irq] 82ba4faca1: [No primary change] [unixbench] Date: Wed, 24 Aug 2016 12:34:02 +0800 Message-ID: <20160824043402.GG3980@intel.com> In-Reply-To: List-Id: --===============4754889347820234460== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Wed, Aug 24, 2016 at 11:13:59AM +0800, Aaron Lu wrote: > On 08/24/2016 11:04 AM, kernel test robot wrote: > > = > > FYI, we noticed interrupts.CAL:Function_call_interrupts +771.1% improve= ment due to commit: to Xiaolong, i guess more interrupts can't be called as improvement, or it = doesn't matter. > = > This is expected. > = > CAL:Function_call_interrupts now contains TLB shootdowns while it > doesn't previously. > = > Thanks, > Aaron > = > > = > > commit 82ba4faca1bffad429f15c90c980ffd010366c25 ("x86/irq: Do not subst= ract irq_tlb_count from irq_call_count") > > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git mast= er > > = > > in testcase: unixbench > > on test machine: 8 threads Nehalem with 6G memory > > with following parameters: > > = > > runtime: 300s > > nr_task: 1 > > test: shell8 > > = > > iN addition to that, the commit also has significant impact (interrupts= .CAL:Function_call_interrupts +1822.5%) > > on the following tests: > > = > > +------------------+----------------------------------+ > > | testcase | unixbench | > > | test machine | 8 threads Nehalem with 6G memory | > > | test parameters | nr_task=3D100% | > > | | runtime=3D300s | > > | | test=3Dexecl | > > +------------------+----------------------------------+ > > = > > = > > Disclaimer: > > Results have been estimated based on internal Intel analysis and are pr= ovided > > for informational purposes only. Any difference in system hardware or s= oftware > > design or configuration may affect actual performance. > > = > > Details are as below: > > -----------------------------------------------------------------------= ---------------------------> > > = > > = > > 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 > > = > > =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 > > compiler/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase: > > gcc-6/x86_64-rhel/1/debian-x86_64-2015-02-07.cgz/300s/nhm-white/shell= 8/unixbench > > = > > commit: = > > ace7fab7a6 ("x86/mm: Fix swap entry comment and macro") > > 82ba4faca1 ("x86/irq: Do not substract irq_tlb_count from irq_call_co= unt") > > = > > ace7fab7a6cdd363 82ba4faca1bffad429f15c90c9 = > > ---------------- -------------------------- = > > %stddev %change %stddev > > \ | \ = > > 48379 =C2=B1 0% +771.1% 421448 =C2=B1 0% interrupts.CAL:= Function_call_interrupts > > 19934 =C2=B1 8% -10.3% 17882 =C2=B1 4% meminfo.AnonHug= ePages > > 177575 =C2=B1 8% -11.9% 156428 =C2=B1 7% sched_debug.cpu= .avg_idle.min > > 0.36 =C2=B1 35% +431.2% 1.93 =C2=B1 55% sched_debug.cpu= .cpu_load[0].min > > 25.84 =C2=B1 1% +11.3% 28.77 =C2=B1 4% sched_debug.cpu= .cpu_load[1].min > > 5.38 =C2=B1 3% -13.9% 4.63 =C2=B1 5% sched_debug.cpu= .cpu_load[1].stddev > > = > > = > > = > > = > > interrupts.CAL:Function_call_interrupts > > = > > 450000 ++------------------------------------------------------------= -----+ > > O O OO O O OO O O OO O O OO O O O OO = | > > 400000 ++ = | > > 350000 ++ = | > > | = | > > 300000 ++ = | > > 250000 ++ = | > > | = | > > 200000 ++ = | > > 150000 ++ = | > > | = | > > 100000 ++ = | > > 50000 *+ .**.*.*.* .**.*.*. *.*.*.*.* .*. *.*.*.** .* .*.= * .*.* > > | * :.*.* * :.* * + .* *.* = * | > > 0 ++----------*----------------------*---------------*----------= -----+ > > = > > = > > [*] bisect-good sample > > [O] bisect-bad sample > > = > > ***********************************************************************= **************************** > > nhm-white: 8 threads Nehalem with 6G memory > > =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 > > compiler/kconfig/nr_task/rootfs/runtime/tbox_group/test/testcase: > > gcc-6/x86_64-rhel/100%/debian-x86_64-2015-02-07.cgz/300s/nhm-white/ex= ecl/unixbench > > = > > commit: = > > ace7fab7a6 ("x86/mm: Fix swap entry comment and macro") > > 82ba4faca1 ("x86/irq: Do not substract irq_tlb_count from irq_call_co= unt") > > = > > ace7fab7a6cdd363 82ba4faca1bffad429f15c90c9 = > > ---------------- -------------------------- = > > fail:runs %reproduction fail:runs > > | | | = > > %stddev %change %stddev > > \ | \ = > > 25890 =C2=B1 1% +1822.5% 497748 =C2=B1 1% interrupts.CAL:= Function_call_interrupts > > 15522 =C2=B1 80% +354.6% 70567 =C2=B1146% latency_stats.a= vg.max > > 0.28 =C2=B1 11% +23.9% 0.34 =C2=B1 4% sched_debug.cfs= _rq:/.nr_running.stddev > > 535.29 =C2=B1 0% +2071.2% 11622 =C2=B1 55% sched_debug.cfs= _rq:/.runnable_load_avg.max > > 18.12 =C2=B1 35% -66.9% 6.00 =C2=B1 85% sched_debug.cfs= _rq:/.runnable_load_avg.min > > 159.21 =C2=B1 0% +2305.2% 3829 =C2=B1 55% sched_debug.cfs= _rq:/.runnable_load_avg.stddev > > 109.46 =C2=B1 3% +1268.4% 1497 =C2=B1 53% sched_debug.cpu= .cpu_load[0].avg > > 535.25 =C2=B1 0% +2071.2% 11621 =C2=B1 55% sched_debug.cpu= .cpu_load[0].max > > 536.29 =C2=B1 0% +2067.3% 11623 =C2=B1 55% sched_debug.cpu= .cpu_load[1].max > > 158.13 =C2=B1 0% +2317.9% 3823 =C2=B1 55% sched_debug.cpu= .cpu_load[1].stddev > > 535.92 =C2=B1 0% +2068.9% 11623 =C2=B1 55% sched_debug.cpu= .cpu_load[2].max > > 158.04 =C2=B1 0% +2319.1% 3823 =C2=B1 55% sched_debug.cpu= .cpu_load[2].stddev > > 535.00 =C2=B1 0% +2072.4% 11622 =C2=B1 55% sched_debug.cpu= .cpu_load[3].max > > 157.68 =C2=B1 0% +2324.3% 3822 =C2=B1 55% sched_debug.cpu= .cpu_load[3].stddev > > 533.17 =C2=B1 0% +2079.7% 11621 =C2=B1 55% sched_debug.cpu= .cpu_load[4].max > > 157.11 =C2=B1 0% +2332.9% 3822 =C2=B1 55% sched_debug.cpu= .cpu_load[4].stddev > > 168510 =C2=B1 2% +840.5% 1584778 =C2=B1 51% sched_debug.cpu= .load.avg > > 600190 =C2=B1 1% +1904.9% 12033432 =C2=B1 54% sched_debug.cpu= .load.max > > 197074 =C2=B1 1% +1909.8% 3960765 =C2=B1 54% sched_debug.cpu= .load.stddev > > -1804 =C2=B1-12% -28.0% -1299 =C2=B1-13% sched_debug.cpu= .nr_uninterruptible.min > > 1327 =C2=B1 19% -24.6% 1000 =C2=B1 19% sched_debug.cpu= .nr_uninterruptible.stddev > > = > > = > > = > > = > > = > > = > > = > > Thanks, > > Xiaolong > > = > = > _______________________________________________ > LKP mailing list > LKP(a)lists.01.org > https://lists.01.org/mailman/listinfo/lkp --===============4754889347820234460==--