Public bug reported: [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. ** Affects: glibc (Ubuntu) Importance: High Assignee: Heitor Alves de Siqueira (halves) Status: Fix Released ** Affects: glibc (Ubuntu Focal) Importance: High Assignee: Heitor Alves de Siqueira (halves) Status: Confirmed ** Affects: glibc (Ubuntu Groovy) Importance: High Assignee: Heitor Alves de Siqueira (halves) Status: Confirmed ** Tags: sts ** Also affects: glibc (Ubuntu Groovy) Importance: Undecided Status: New ** Also affects: glibc (Ubuntu Focal) Importance: Undecided Status: New ** Changed in: glibc (Ubuntu Focal) Importance: Undecided => High ** Changed in: glibc (Ubuntu Groovy) Importance: Undecided => High ** Changed in: glibc (Ubuntu Focal) Status: New => Confirmed ** Changed in: glibc (Ubuntu Groovy) Status: New => Won't Fix ** Changed in: glibc (Ubuntu Groovy) Status: Won't Fix => Confirmed ** Changed in: glibc (Ubuntu) Status: New => Fix Released ** Changed in: glibc (Ubuntu Focal) Assignee: (unassigned) => Heitor Alves de Siqueira (halves) ** Changed in: glibc (Ubuntu Groovy) Assignee: (unassigned) => Heitor Alves de Siqueira (halves) ** 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. [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 + 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 + 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' + 'libc-patched' ran + 1.79 ± 0.01 times faster than 'libc-2.31-0ubuntu9.2' [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. ** 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. [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 + 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 Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1928508 Title: Performance regression on memcpy() calls for AMD Zen To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/glibc/+bug/1928508/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs