[Bug 1928508] Re: Performance regression on memcpy() calls for AMD Zen

Tue May 18 17:57:30 UTC 2021

** Description changed:

  [Impact]
  On AMD Zen systems, memcpy() calls see a heavy performance regression in Focal and Groovy, due to the way __x86_non_temporal_threshold is calculated.

- Before 'glibc-2.33~455', cache values were calculated taking into consideration the number of hardware threads in the CPU. On AMD Ryzen and EPYC systems, this can be counter-productive if the number of threads is high enough for the last-level caches to "overrun" each other and cause cache line flushes. The solution is to reduce the allocated size for these non_temporal stores, removing
- the number of threads from the equation.
+ Before 'glibc-2.33~455', cache values were calculated taking into
+ consideration the number of hardware threads in the CPU. On AMD Ryzen
+ and EPYC systems, this can be counter-productive if the number of
+ threads is high enough for the last-level caches to "overrun" each other
+ and cause cache line flushes. The solution is to reduce the allocated
+ size for these non_temporal stores, removing the number of threads from
+ the equation.

  [Test Plan]
  Attached to this bug is a short C program that exercises memcpy() calls in buffers of variable length. This has been obtained from a similar bug report for Red Hat, and is publicly available at [0].
  This test program was compiled with gcc 10.2.0, using the following flags:
  $ gcc -mtune=generic -march=x86_64 -g -03 test_memcpy.c -o test_memcpy64

  Tests were performed with the following criteria:
  - use 32Mb buffers ("./test_memcpy64 32")
  - benchmark with the hyperfine tool [1], as it calculates relevant statistics automatically
  - benchmark with at least 10 runs in the same environment, to minimize variance
  - measure on AMD Zen (3700X) and on Intel Xeon (E5-2683), to ensure we don't penalize one x86 vendor in favor of the other

  Below is a comparison between two Focal containers, leveraging LXD to
  make use of different libc versions on the same host:

  $ hyperfine -n libc-2.31-0ubuntu9.2 'lxc exec focal ./test_memcpy64 32' -n libc-patched 'lxc exec focal-patched ./test_memcpy64 32'
  Benchmark #1: libc-2.31-0ubuntu9.2
    Time (mean ± σ):      2.723 s ±  0.013 s    [User: 4.7 ms, System: 5.1 ms]
    Range (min … max):    2.693 s …  2.735 s    10 runs

  Benchmark #2: libc-patched
    Time (mean ± σ):      1.522 s ±  0.004 s    [User: 3.9 ms, System: 5.6 ms]
    Range (min … max):    1.515 s …  1.528 s    10 runs

  Summary
    'libc-patched' ran
      1.79 ± 0.01 times faster than 'libc-2.31-0ubuntu9.2'
  $ head -n5 /proc/cpuinfo
  processor       : 0
  vendor_id       : AuthenticAMD
  cpu family      : 23
  model           : 113
  model name      : AMD Ryzen 7 3700X 8-Core Processor

  [0] https://bugzilla.redhat.com/show_bug.cgi?id=1880670
  [1] https://github.com/sharkdp/hyperfine/

  [Where problems could occur]
  Since we're messing with the cacheinfo for x86 in general, we need to be careful not to introduce further performance regressions on memory-heavy workloads. Even though initial results might reveal improvement on AMD Ryzen and EPYC hardware, we should also validate different configurations (e.g. Intel, different buffer sizes, etc) to make sure we won't hurt performance in other non-AMD environments.

  [Other Info]
  This has been fixed by the following upstream commit:
  - d3c57027470b (Reversing calculation of __x86_shared_non_temporal_threshold)

  $ git describe --contains d3c57027470b
  glibc-2.33~455
  $ rmadison glibc -s focal,focal-updates,groovy,groovy-proposed,hirsute
   glibc | 2.31-0ubuntu9   | focal           | source
   glibc | 2.31-0ubuntu9.2 | focal-updates   | source
   glibc | 2.32-0ubuntu3   | groovy          | source
   glibc | 2.32-0ubuntu3.2 | groovy-proposed | source
   glibc | 2.33-0ubuntu5   | hirsute         | source

  Affected releases include Ubuntu Focal and Groovy. Bionic is not
  affected, and releases starting with Hirsute already ship the upstream
  patch to fix this regression.

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to glibc in Ubuntu.
https://bugs.launchpad.net/bugs/1928508

Title:
  Performance regression on memcpy() calls for AMD Zen

Status in glibc package in Ubuntu:
  Fix Released
Status in glibc source package in Focal:
  Confirmed
Status in glibc source package in Groovy:
  Confirmed

Bug description:
  [Impact]
  On AMD Zen systems, memcpy() calls see a heavy performance regression in Focal and Groovy, due to the way __x86_non_temporal_threshold is calculated.

  Before 'glibc-2.33~455', cache values were calculated taking into
  consideration the number of hardware threads in the CPU. On AMD Ryzen
  and EPYC systems, this can be counter-productive if the number of
  threads is high enough for the last-level caches to "overrun" each
  other and cause cache line flushes. The solution is to reduce the
  allocated size for these non_temporal stores, removing the number of
  threads from the equation.

  [Test Plan]
  Attached to this bug is a short C program that exercises memcpy() calls in buffers of variable length. This has been obtained from a similar bug report for Red Hat, and is publicly available at [0].
  This test program was compiled with gcc 10.2.0, using the following flags:
  $ gcc -mtune=generic -march=x86_64 -g -03 test_memcpy.c -o test_memcpy64

  Tests were performed with the following criteria:
  - use 32Mb buffers ("./test_memcpy64 32")
  - benchmark with the hyperfine tool [1], as it calculates relevant statistics automatically
  - benchmark with at least 10 runs in the same environment, to minimize variance
  - measure on AMD Zen (3700X) and on Intel Xeon (E5-2683), to ensure we don't penalize one x86 vendor in favor of the other

  Below is a comparison between two Focal containers, leveraging LXD to
  make use of different libc versions on the same host:

  $ hyperfine -n libc-2.31-0ubuntu9.2 'lxc exec focal ./test_memcpy64 32' -n libc-patched 'lxc exec focal-patched ./test_memcpy64 32'
  Benchmark #1: libc-2.31-0ubuntu9.2
    Time (mean ± σ):      2.723 s ±  0.013 s    [User: 4.7 ms, System: 5.1 ms]
    Range (min … max):    2.693 s …  2.735 s    10 runs

  Benchmark #2: libc-patched
    Time (mean ± σ):      1.522 s ±  0.004 s    [User: 3.9 ms, System: 5.6 ms]
    Range (min … max):    1.515 s …  1.528 s    10 runs

  Summary
    'libc-patched' ran
      1.79 ± 0.01 times faster than 'libc-2.31-0ubuntu9.2'
  $ head -n5 /proc/cpuinfo
  processor       : 0
  vendor_id       : AuthenticAMD
  cpu family      : 23
  model           : 113
  model name      : AMD Ryzen 7 3700X 8-Core Processor

  [0] https://bugzilla.redhat.com/show_bug.cgi?id=1880670
  [1] https://github.com/sharkdp/hyperfine/

  [Where problems could occur]
  Since we're messing with the cacheinfo for x86 in general, we need to be careful not to introduce further performance regressions on memory-heavy workloads. Even though initial results might reveal improvement on AMD Ryzen and EPYC hardware, we should also validate different configurations (e.g. Intel, different buffer sizes, etc) to make sure we won't hurt performance in other non-AMD environments.

  [Other Info]
  This has been fixed by the following upstream commit:
  - d3c57027470b (Reversing calculation of __x86_shared_non_temporal_threshold)

  $ git describe --contains d3c57027470b
  glibc-2.33~455
  $ rmadison glibc -s focal,focal-updates,groovy,groovy-proposed,hirsute
   glibc | 2.31-0ubuntu9   | focal           | source
   glibc | 2.31-0ubuntu9.2 | focal-updates   | source
   glibc | 2.32-0ubuntu3   | groovy          | source
   glibc | 2.32-0ubuntu3.2 | groovy-proposed | source
   glibc | 2.33-0ubuntu5   | hirsute         | source

  Affected releases include Ubuntu Focal and Groovy. Bionic is not
  affected, and releases starting with Hirsute already ship the upstream
  patch to fix this regression.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/glibc/+bug/1928508/+subscriptions