Re: [sched/fair] 38ac256d1c: stress-ng.vm-segv.ops_per_sec -13.8% regression

[Peter Zijlstra]

On Wed, Apr 21, 2021 at 11:27:49AM +0100, Valentin Schneider wrote:
> o 2-socket Xeon E5-2690 (x86, 40 cores)
> 
> and found at worse a -0.3% regression and at best a 2% improvement. I know
> that x86 box is somewhat ancient, but it's been my go-to "have I broken
> x86?" test victim for a while :-)

It happens that my main test-box is an E5-2680. I've got a slightly more
modern one too, but that boots like treacle so I end up using the IVB-EP
most :-)