Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math

From	Peter Zijlstra <peterz@infradead.org>
Newsgroups	linux.kernel
Subject	Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math
Date	2016-12-09 10:20 +0100
Message-ID	<sMpvs-1sl-9@gated-at.bofh.it> (permalink)
References	<sMdXk-2rQ-13@gated-at.bofh.it> <sMdXl-2rQ-43@gated-at.bofh.it> <sMlUR-7HU-3@gated-at.bofh.it> <sMn0C-8kb-3@gated-at.bofh.it> <sMoSK-Zv-11@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Fri, Dec 09, 2016 at 09:30:11AM +0100, Peter Zijlstra wrote:

> > > Just for giggles, on tilegx the branch is actually slower than doing the
> > > mult unconditionally.
> > > 
> > > The problem is that the two multiplies would otherwise completely
> > > pipeline, whereas with the conditional you serialize them.

> Only when using x86_64 instructions, once I fixed the i386 variant it
> was slower, probably due to register pressure and the like.

OK, maybe I messed up on i386, although I've yet to try running that on
an actual 32bit machine. I also need to dig up a small core, who knows
what atoms do.

Results are in cycles, average over 1e6 loops. I think the 128 results
are around 1 cycle, measurements are maybe a tad wobbly because I
compare against an empty loop to correct measurement overhead.


root@ivb-ep:~/tmp# for i in -m64 -m32 -mx32; do echo $i ; gcc -O3 $i -o mult mult.c -lm; ./mult ; done

-m64
cond: avg: 5.487738 +- 0.004152
uncond: avg: 4.495690 +- 0.006009
128: avg: 0.634496 +- 0.004795

-m32
cond: avg: 14.807630 +- 0.006890
uncond: avg: 11.601985 +- 0.009722

-mx32
cond: avg: 5.027696 +- 0.005766
uncond: avg: 4.038013 +- 0.008069
128: avg: 0.009928 +- 0.005730


root@hsw:~/tmp# for i in -m64 -m32 -mx32; do echo $i ; gcc -O3 $i -o mult mult.c -lm; ./mult ; done

-m64
cond: avg: 1.998718 +- 0.008775
uncond: avg: 2.004795 +- 0.009865
128: avg: 0.991947 +- 0.007607

-m32
cond: avg: 12.981868 +- 0.011239
uncond: avg: 13.000566 +- 0.011668

-mx32
cond: avg: 2.005437 +- 0.006840
uncond: avg: 3.001631 +- 0.004786
128: avg: 1.990425 +- 0.003880

Thread

[patch 0/6] timekeeping: Cure the signed/unsigned wreckage Thomas Gleixner <tglx@linutronix.de> - 2016-12-08 22:00 +0100
  [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Thomas Gleixner <tglx@linutronix.de> - 2016-12-08 22:00 +0100
    Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Ingo Molnar <mingo@kernel.org> - 2016-12-09 05:10 +0100
      Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Ingo Molnar <mingo@kernel.org> - 2016-12-09 05:30 +0100
        Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math John Stultz <john.stultz@linaro.org> - 2016-12-09 05:50 +0100
      Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Peter Zijlstra <peterz@infradead.org> - 2016-12-09 05:50 +0100
        Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Ingo Molnar <mingo@kernel.org> - 2016-12-09 06:30 +0100
          Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Peter Zijlstra <peterz@infradead.org> - 2016-12-09 06:50 +0100
    Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Peter Zijlstra <peterz@infradead.org> - 2016-12-09 06:20 +0100
      Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Peter Zijlstra <peterz@infradead.org> - 2016-12-09 07:10 +0100
    Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Peter Zijlstra <peterz@infradead.org> - 2016-12-09 06:30 +0100
      Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Peter Zijlstra <peterz@infradead.org> - 2016-12-09 07:40 +0100
        Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Peter Zijlstra <peterz@infradead.org> - 2016-12-09 09:40 +0100
          Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Peter Zijlstra <peterz@infradead.org> - 2016-12-09 10:20 +0100
          Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Peter Zijlstra <peterz@infradead.org> - 2016-12-09 11:10 +0100
        Re: [patch 5/6] [RFD] timekeeping: Provide optional 128bit math Peter Zijlstra <peterz@infradead.org> - 2016-12-09 11:20 +0100
  [patch 3/6] timekeeping: Get rid of pointless typecasts Thomas Gleixner <tglx@linutronix.de> - 2016-12-08 22:00 +0100
    Re: [patch 3/6] timekeeping: Get rid of pointless typecasts David Gibson <david@gibson.dropbear.id.au> - 2016-12-09 00:50 +0100
    [tip:timers/core] timekeeping: Get rid of pointless typecasts tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-12-09 12:20 +0100
  Re: [patch 0/6] timekeeping: Cure the signed/unsigned wreckage John Stultz <john.stultz@linaro.org> - 2016-12-09 06:00 +0100
  Re: [patch 0/6] timekeeping: Cure the signed/unsigned wreckage Peter Zijlstra <peterz@infradead.org> - 2016-12-09 06:40 +0100

csiph-web