OK , thanks - appreciate all the hard work of making performance regression
analysis & continued improvement to reduce false positives!
On Tue, Dec 7, 2021 at 4:03 AM Maxim Kuvyrkov
wrote:
> Hi David,
>
> This is a false positive, sorry for the noise. Our CI bisects performance
> regressions do
Hi David,
This is a false positive, sorry for the noise. Our CI bisects performance
regressions down to a single commit, and notifies patch authors about
significant regressions.
Since this is benchmarking CI, some noise is expected. Apparently, 433.milc
started to show bi-modal performance,
Seems... unlikely this change had a performance impact.
Also is this email meant to be sent to public contributors like myself, or
only intended for some Linaro folks?
On Sun, Dec 5, 2021 at 6:18 AM wrote:
> After llvm commit bd4c6a476fd037fb07a1c484f75d93ee40713d3d
> Author: David Blaikie
>
>
After llvm commit bd4c6a476fd037fb07a1c484f75d93ee40713d3d
Author: David Blaikie
Add missing header
the following benchmarks slowed down by more than 2%:
- 433.milc slowed down by 4% from 12427 to 12916 perf samples
Below reproducer instructions can be used to re-build both "first_bad" and
