AMD Has A Nice Performance Optimization Coming With Linux 6.8

[phoronix]

Phoronix: AMD Has A Nice Performance Optimization Coming With Linux 6.8

Queued up into tip/tip.git's x86/cpu branch ahead of the Linux 6.8 merge window opening in a month is an optimization that should prove helpful in cloud/VM scenarios...

https://www.phoronix.com/news/Linux-6.8-AMD-No-Serialize-MSR

[Kjell]

I'm very happy to see bottlenecks as such getting fixed!

[Eirikr1848]

Should this also apply to older K8/K10 Opterons; Bulldozer, Vishera, etc or Zen 1 the chosen “modern era cut off point”?

[skeevy420]

Should this also apply to older K8/K10 Opterons; Bulldozer, Vishera, etc or Zen 1 the chosen “modern era cut off point”?

All of them. Every single 64-bit processor from AMD.

[Eirikr1848]

All of them. Every single 64-bit processor from AMD.

Oh I see for some reason I read the change as coming in Linux 6.8 only for Zen 1/Hygon.

[TNZfr]

Can't wait to test this on my 5950x ...

[skeevy420]

Oh I see for some reason I read the change as coming in Linux 6.8 only for Zen 1/Hygon.

I thought that, too, but, fortunately for my non-programming, Bash-friendly brain, the C was easy enough to read.

[schmidtbag]

Makes me curious what kinds of workloads would benefit from this change, and how much it scales with cores and frequency.

[npwx]

It appears this is just for IPIs, so I don't think any reasonable workload will noticeably benefit from this as any workload excessively using IPIs is probably quite slow already.