More Benchmarks Of The Performance Pullback In Linux 4.20

[phoronix]

Phoronix: More Benchmarks Of The Performance Pullback In Linux 4.20

Last night I published some benchmarks after finding Linux 4.20 is regressing in several workloads compared to Linux 4.18/4.19 and at least was affecting Intel Core i9 "HEDT" boxes. Here are more affected workloads regressing on Linux 4.20 and it's not just limited to high-end hardware...

http://www.phoronix.com/scan.php?page=news_item&px=Linux-4.20-Performance-Pullback

[Hi-Angel]

It would be nice if titles in this and the previous articles have been using "4.20rc2" instead of just "4.20", so to not confuse people who can come here from a search engine later in cycle, when the regression is fixed.

[davidbepo]

thanks for doing this, hopefully you can get to the exact commit and report it

[treba]

Great stuff, Michael, thanks! That's why I'm happy to be a phoronix premium member

[Michael]

Great stuff, Michael, thanks! That's why I'm happy to be a phoronix premium member

Thanks for your support!

[miabrahams]

The first thing I wondered seeing your headline was "is it possible to bisect this?" and turns out you're already on it. Thanks Michael!

[mbello]

Great work Michael!

[varikonniemi]

It's weird so little official automated testing happens that Phoronix can find general regressions like this. I think Michael should get some sponsorship from the linux foundation for all the testing he provides.

[Hibbelharry]

Maybe the devs are patching some new spectre and meltdown holes? There are new reports and new attack mechanisms.

[lxusr]

Make Linux Floppy Again
make eve drop apple