Messages in this thread | | | Date | Thu, 14 Oct 2021 21:46:50 +0300 | From | Ville Syrjälä <> | Subject | Re: [regression] commit d298b03506d3 ("x86/fpu: Restore the masking out of reserved MXCSR bits") |
| |
On Thu, Oct 14, 2021 at 08:01:24PM +0200, Borislav Petkov wrote: > On Thu, Oct 14, 2021 at 08:45:33PM +0300, Ville Syrjälä wrote: > > That ~ was indeed the problem. With it gone the machine is happy again. > > > > I presume you'll turn this into a real patch? > > Actually, you found it and you should be the one to write it and do the > honors. Unless you don't want to - then I can do it.
I figured you can write a reasonably succinct commit message, instead of having me ramble on incoherently. ATM I don't even know what mxcsr is or why clobbering it would cause floating point exceptions with sse specifically.
But I can certainly ramble, if you prefer that.
> > If you do, pls add > > Ser Olmy <ser.olmy@protonmail.com> > > to Cc so that he can test your patch. I *think* it should work for him > too but I don't know anything anymore. :-) > > Thx. > > -- > Regards/Gruss, > Boris. > > https://people.kernel.org/tglx/notes-about-netiquette
-- Ville Syrjälä Intel
| |