I've ran the same test on an Intel system, to ensure we aren't
introducing any regressions there. Besides basic smoke tests, the
benchmarks from the description showed that the performance on Intel is
not significantly affected by this patch.
halves@rotom:~$ head -n5 /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 63
model name : Intel(R) Xeon(R) CPU E5-2683 v3 @ 2.00GHz
halves@rotom:~$ .cargo/bin/hyperfine -n focal-2.31-0ubuntu9.2 'lxc exec
halves-focal ./test_memcpy64 32' -n focal-patched 'lxc exec
halves-focal-patched ./test_memcpy64 32'
Benchmark #1: focal-2.31-0ubuntu9.2
Time (mean ± σ): 2.662 s ± 0.058 s [User: 53.4 ms, System: 79.6 ms]
Range (min … max): 2.559 s … 2.718 s 10 runs
Benchmark #2: focal-patched
Time (mean ± σ): 2.650 s ± 0.074 s [User: 61.5 ms, System: 76.1 ms]
Range (min … max): 2.558 s … 2.759 s 10 runs
Summary
'focal-patched' ran
1.00 ± 0.04 times faster than 'focal-2.31-0ubuntu9.2'
halves@rotom:~$ .cargo/bin/hyperfine -n groovy-2.32-0ubuntu3 'lxc exec
halves-groovy ./test_memcpy64 32' -n groovy-patched 'lxc exec
halves-groovy-patched ./test_memcpy64 32'
Benchmark #1: groovy-2.32-0ubuntu3
Time (mean ± σ): 2.643 s ± 0.044 s [User: 52.4 ms, System: 76.0 ms]
Range (min … max): 2.575 s … 2.746 s 10 runs
Benchmark #2: groovy-patched
Time (mean ± σ): 2.626 s ± 0.036 s [User: 63.1 ms, System: 79.7 ms]
Range (min … max): 2.590 s … 2.701 s 10 runs
Summary
'groovy-patched' ran
1.01 ± 0.02 times faster than 'groovy-2.32-0ubuntu3'
--
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
