[libc-commits] [PATCH] D147676: [libc][math] Update range reduction step for log10f and reduce its latency.

[Tue Ly via Phabricator via libc-commits]

lntue added a comment.

In D147676#4248087 <https://reviews.llvm.org/D147676#4248087>, @zimmermann6 wrote:

> I get a failure for rounding down:
>
>   zimmerma at biscotte:~/svn/core-math$ CORE_MATH_CHECK_STD=true LIBM=/localdisk/zimmerma/llvm-project/build/projects/libc/lib/libllvmlibc.a ./check.sh log10f
>   Running exhaustive check in --rndn mode...
>   all ok
>   Running exhaustive check in --rndz mode...
>   all ok
>   Running exhaustive check in --rndu mode...
>   all ok
>   Running exhaustive check in --rndd mode...
>   FAIL x=0x1p+0 ref=0x0p+0 y=-0x0p+0
>
> Also, on an AMD EPYC 7282 I get a regression in speed. With master:
>
>   zimmerma at biscotte:~/svn/core-math$ LIBM=/localdisk/zimmerma/llvm-project/build/projects/libc/lib/libllvmlibc.a ./perf.sh log10f
>   GNU libc version: 2.36
>   GNU libc release: stable
>   [####################] 100 %
>   Ntrial = 20 ; Min = 10.531 + 0.273 clc/call; Median-Min = 0.281 clc/call; Max = 13.047 clc/call;
>   [####################] 100 %
>   Ntrial = 20 ; Min = 18.529 + 0.342 clc/call; Median-Min = 0.309 clc/call; Max = 19.811 clc/call;
>   [####################] 100 %
>   Ntrial = 20 ; Min = 13.059 + 0.526 clc/call; Median-Min = 0.290 clc/call; Max = 15.586 clc/call;
>
> With this patch:
>
>   zimmerma at biscotte:~/svn/core-math$ LIBM=/localdisk/zimmerma/llvm-project/build/projects/libc/lib/libllvmlibc.a ./perf.sh log10f
>   GNU libc version: 2.36
>   GNU libc release: stable
>   [####################] 100 %
>   Ntrial = 20 ; Min = 10.534 + 0.297 clc/call; Median-Min = 0.303 clc/call; Max = 11.415 clc/call;
>   [####################] 100 %
>   Ntrial = 20 ; Min = 18.529 + 0.561 clc/call; Median-Min = 0.327 clc/call; Max = 20.729 clc/call;
>   [####################] 100 %
>   Ntrial = 20 ; Min = 19.791 + 0.313 clc/call; Median-Min = 0.338 clc/call; Max = 22.809 clc/call;

I've fixed the issue with signed zero, and here are the current performance numbers I got on Ryzen 5900X:

  $ ./perf.sh log10f --path2
  LIBC-location: /home/lnt/experiment/llvm/llvm-project/build/projects/libc/lib/libllvmlibc.a
  GNU libc version: 2.35
  GNU libc release: stable
  -- CORE-MATH reciprocal throughput --
  [####################] 100 %
  Ntrial = 20 ; Min = 10.004 + 0.159 clc/call; Median-Min = 0.198 clc/call; Max = 10.279 clc/call;
  -- System LIBC reciprocal throughput --
  [####################] 100 %
  Ntrial = 20 ; Min = 16.652 + 0.222 clc/call; Median-Min = 0.264 clc/call; Max = 16.999 clc/call;
  -- LIBC reciprocal throughput --
  [####################] 100 %
  Ntrial = 20 ; Min = 8.586 + 0.194 clc/call; Median-Min = 0.204 clc/call; Max = 8.921 clc/call;

And for latency:

  $ ./perf.sh log10f --path2 --latency
  LIBC-location: /home/lnt/experiment/llvm/llvm-project/build/projects/libc/lib/libllvmlibc.a
  GNU libc version: 2.35
  GNU libc release: stable
  -- CORE-MATH reciprocal throughput --
  [####################] 100 %
  Ntrial = 20 ; Min = 35.298 + 0.314 clc/call; Median-Min = 0.312 clc/call; Max = 35.782 clc/call;
  -- System LIBC reciprocal throughput --
  [####################] 100 %
  Ntrial = 20 ; Min = 48.766 + 0.755 clc/call; Median-Min = 0.917 clc/call; Max = 49.771 clc/call;
  -- LIBC reciprocal throughput --
  [####################] 100 %
  Ntrial = 20 ; Min = 34.878 + 0.203 clc/call; Median-Min = 0.246 clc/call; Max = 35.256 clc/call;


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D147676/new/

https://reviews.llvm.org/D147676