Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1222996

Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig

From Morten Rasmussen <morten.rasmussen@arm.com>
Newsgroups linux.kernel
Subject Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig
Date 2015-09-11 19:20 +0200
Message-ID <q7A9s-89P-3@gated-at.bofh.it> (permalink)
References (5 earlier) <q6qFd-4GE-21@gated-at.bofh.it> <q6s4j-6PU-29@gated-at.bofh.it> <q6ufN-1tS-23@gated-at.bofh.it> <q6KaS-7Lb-13@gated-at.bofh.it> <q6Lqh-1gs-7@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Wed, Sep 09, 2015 at 12:13:10PM +0100, Morten Rasmussen wrote:
> On Wed, Sep 09, 2015 at 11:43:05AM +0200, Peter Zijlstra wrote:
> > Sadly that makes the code worse; I get 14 mul instructions where
> > previously I had 11.
> > 
> > What happens is that GCC gets confused and cannot constant propagate the
> > new variables, so what used to be shifts now end up being actual
> > multiplications.
> > 
> > With this, I get back to 11. Can you see what happens on ARM where you
> > have both functions defined to non constants?
> 
> We repeated the experiment on arm and arm64 but still with functions
> defined to constant to compare with your results. The mul instruction
> count seems to be somewhat compiler version dependent, but consistently
> show no effect of the patch:
> 
> arm	before	after
> gcc4.9	12	12
> gcc4.8	10	10
> 
> arm64	before	after
> gcc4.9	11	11
> 
> I will get numbers with the arch-functions implemented as well and do
> hackbench runs to see what happens in terms of performance.

I have done some runs with the proposed fixes added:

1. PeterZ's util_sum shift fix (change util_sum).
2. Morten's scaling of weight instead of time (reduce bit loss).
3. PeterZ's unconditional calls to arch*() functions (compiler opt).

To be clear: 2 includes 1, and 3 includes 1 and 2.

Runs where done with the default (#define) implementation of the
arch-functions and with arch specific implementation for ARM.

I realized that just looking for 'mul' instructions in
update_blocked_averages() is probably not a fair comparison on ARM as it
turned out that it has quite a few multiply-accumulate instructions. So
I have included the total count including those too.


Test platforms:

ARM TC2 (A7x3 only)
perf stat --null --repeat 10 -- perf bench sched messaging -g 50 -l 200
#mul: grep -e mul (in update_blocked_averages())
#mul_all: grep -e mul -e mla -e mls -e mia (in update_blocked_averages())
gcc: 4.9.3

Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz
perf stat --null --repeat 10 -- perf bench sched messaging -g 50 -l 15000
#mul: grep -e mul (in update_blocked_averages())
gcc: 4.9.2


Results:

perf numbers are average of three (x10) runs. Raw data is available
further down.

ARM TC2		#mul		#mul_all	perf bench
arch*()		default	arm	default	arm	default	arm

1 shift_fix	10	16	22	36	13.401	13.288
2 scaled_weight	12	14	30	32	13.282	13.238
3 unconditional	12	14	26	32	13.296	13.427

Intel E5-2690	#mul		#mul_all	perf bench
arch*()		default		default		default

1 shift_fix	13				14.786
2 scaled_weight	18				15.078
3 unconditional	14				15.195


Overall it appears that fewer 'mul' instructions doesn't necessarily
mean better perf bench score. For ARM, 2 seems the best choice overall.
While 1 is better for Intel. If we want to try avoid the bit loss by
scaling weight instead of time, 2 is best for both. However, all that
said, looking at the raw numbers there is a significant difference
between runs of perf --repeat, so we can't really draw any strong
conclusions. It all appears to be in the noise.

I suggest that I spin a v2 of this series and go with scaled_weight to
reduce bit loss. Any objections?

While at it, should I include Yuyang's patch redefining the SCALE/SHIFT
mess?


Raw numbers:

ARM TC2

shift_fix	default_arch
gcc4.9.3
#mul 10
#mul+mla+mls+mia 22
13.384416727 seconds time elapsed ( +-  0.17% )
13.431014702 seconds time elapsed ( +-  0.18% )
13.387434890 seconds time elapsed ( +-  0.15% )

shift_fix	arm_arch
gcc4.9.3
#mul 16
#mul+mla+mls+mia 36
13.271044081 seconds time elapsed ( +-  0.11% )
13.310189123 seconds time elapsed ( +-  0.19% )
13.283594740 seconds time elapsed ( +-  0.12% )

scaled_weight	default_arch
gcc4.9.3
#mul 12
#mul+mla+mls+mia 30
13.295649553 seconds time elapsed ( +-  0.20% )
13.271634654 seconds time elapsed ( +-  0.19% )
13.280081329 seconds time elapsed ( +-  0.14% )

scaled_weight	arm_arch
gcc4.9.3
#mul 14
#mul+mla+mls+mia 32
13.230659223 seconds time elapsed ( +-  0.15% )
13.222276527 seconds time elapsed ( +-  0.15% )
13.260275081 seconds time elapsed ( +-  0.21% )

unconditional	default_arch
gcc4.9.3
#mul 12
#mul+mla+mls+mia 26
13.274904460 seconds time elapsed ( +-  0.13% )
13.307853511 seconds time elapsed ( +-  0.15% )
13.304084844 seconds time elapsed ( +-  0.22% )

unconditional	arm_arch
gcc4.9.3
#mul 14
#mul+mla+mls+mia 32
13.432878577 seconds time elapsed ( +-  0.13% )
13.417950552 seconds time elapsed ( +-  0.12% )
13.431682719 seconds time elapsed ( +-  0.18% )


Intel

shift_fix	default_arch
gcc4.9.2
#mul 13
14.905815416 seconds time elapsed ( +-  0.61% )
14.811113694 seconds time elapsed ( +-  0.84% )
14.639739309 seconds time elapsed ( +-  0.76% )

scaled_weight	default_arch
gcc4.9.2
#mul 18
15.113275474 seconds time elapsed ( +-  0.64% )
15.056777680 seconds time elapsed ( +-  0.44% )
15.064074416 seconds time elapsed ( +-  0.71% )

unconditional	default_arch
gcc4.9.2
#mul 14
15.105152500 seconds time elapsed ( +-  0.71% )
15.346405473 seconds time elapsed ( +-  0.81% )
15.132933523 seconds time elapsed ( +-  0.82% )
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Dietmar Eggemann <dietmar.eggemann@arm.com> - 2015-09-07 17:40 +0200
  Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig Vincent Guittot <vincent.guittot@linaro.org> - 2015-09-07 18:30 +0200
    Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Dietmar Eggemann <dietmar.eggemann@arm.com> - 2015-09-07 21:00 +0200
      Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-07 21:50 +0200
        Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Dietmar Eggemann <dietmar.eggemann@arm.com> - 2015-09-08 14:50 +0200
      Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig Vincent Guittot <vincent.guittot@linaro.org> - 2015-09-08 09:30 +0200
        Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-08 14:30 +0200
          Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-08 15:00 +0200
            Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig Vincent Guittot <vincent.guittot@linaro.org> - 2015-09-08 16:10 +0200
              Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-08 16:40 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig Vincent Guittot <vincent.guittot@linaro.org> - 2015-09-08 16:50 +0200
            Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-08 16:30 +0200
              Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-08 17:40 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig bsegall@google.com - 2015-09-10 00:30 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-10 13:10 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig Vincent Guittot <vincent.guittot@linaro.org> - 2015-09-10 13:20 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-10 14:10 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Yuyang Du <yuyang.du@intel.com> - 2015-09-11 10:40 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig bsegall@google.com - 2015-09-10 19:30 +0200
              Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-08 18:50 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-09 11:50 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-09 11:50 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-09 13:10 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-11 19:20 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-17 12:00 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-17 12:50 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Yuyang Du <yuyang.du@intel.com> - 2015-09-21 11:10 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig bsegall@google.com - 2015-09-21 19:40 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Yuyang Du <yuyang.du@intel.com> - 2015-09-22 09:30 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Leo Yan <leo.yan@linaro.org> - 2015-09-11 09:50 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-11 12:00 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Leo Yan <leo.yan@linaro.org> - 2015-09-11 16:20 +0200
            Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Yuyang Du <yuyang.du@intel.com> - 2015-09-10 05:00 +0200
              Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-10 12:10 +0200
          Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig Vincent Guittot <vincent.guittot@linaro.org> - 2015-09-08 15:50 +0200
            Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-08 16:20 +0200
              Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig Vincent Guittot <vincent.guittot@linaro.org> - 2015-09-08 17:20 +0200
        Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Dietmar Eggemann <dietmar.eggemann@arm.com> - 2015-09-08 15:00 +0200
          Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig Vincent Guittot <vincent.guittot@linaro.org> - 2015-09-08 16:10 +0200
            Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Dietmar Eggemann <dietmar.eggemann@arm.com> - 2015-09-08 16:30 +0200
          Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Yuyang Du <yuyang.du@intel.com> - 2015-09-10 06:10 +0200
            Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-10 12:10 +0200
              Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Yuyang Du <yuyang.du@intel.com> - 2015-09-11 10:20 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-11 12:30 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig bsegall@google.com - 2015-09-11 19:10 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Yuyang Du <yuyang.du@intel.com> - 2015-09-12 04:20 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig bsegall@google.com - 2015-09-14 19:40 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-14 15:00 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig bsegall@google.com - 2015-09-14 19:40 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Yuyang Du <yuyang.du@intel.com> - 2015-09-15 08:50 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig bsegall@google.com - 2015-09-15 19:20 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Yuyang Du <yuyang.du@intel.com> - 2015-09-16 04:30 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by capacity_orig bsegall@google.com - 2015-09-16 19:10 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Yuyang Du <yuyang.du@intel.com> - 2015-09-17 12:30 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Morten Rasmussen <morten.rasmussen@arm.com> - 2015-09-15 10:40 +0200
                Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-16 17:50 +0200
      Re: [PATCH 5/6] sched/fair: Get rid of scaling utilization by  capacity_orig Peter Zijlstra <peterz@infradead.org> - 2015-09-08 13:50 +0200
  [tip:sched/core] sched/fair:   Get rid of scaling utilization by capacity_orig tip-bot for Dietmar Eggemann <tipbot@zytor.com> - 2015-09-13 13:10 +0200

csiph-web