Linux Will End Up Disabling x86 PTI For AMD Processors - Update: Now Disabled

[kbios]

You're right. I hope the amd exception gets merged before distros pick up the fix, it would be an unnecessary pain otherwise...

[eydee]

They should not toggle the patch based on the vendor string. Let's not forget the 3DNow fiasco, shall we? There has to be a proper blacklist of affected products.

[gutigone]

So...

Does it affect gaming performance?

I would also really like to know if it affects game performance, Michael, please test some big well known titles.

[mhalano]

How can I find out if my CPU is affected?

[mhalano]

Where I can find a tool to tell me if my CPU is affected?

[Noee]

Does this mean that any hyper-visor running on the bugged Intel h/w is going to be affected? Meaning, without the patch, the vulnerability exists - with the patch, there will be reduced performance?

[Michael]

I would also really like to know if it affects game performance, Michael, please test some big well known titles.

Already have some tests running with the Steam Linux games... On those two systems with the Xonotic/Tesseract type open-source titles, no impact.

[StefanBruens]

Michael Why aren't you using the pti/nopti kernel commandline switch to rule out differences by any other changes?

[Naib]

How can I find out if my CPU is affected?

Where I can find a tool to tell me if my CPU is affected?

grep -i intel /proc/cpuinfo && echo "FSCK I am affected"

[Naib]

Michael Why aren't you using the pti/nopti kernel commandline switch to rule out differences by any other changes?

Good point, I was gearing my ryzen system up to see the impact on a large matlab model. ill make use of that rather than comparing 4.14 to 4.15