Re: [RFC PATCH] x86: Kill notsc

2015-10-21 Thread Peter Zijlstra
On Wed, Oct 21, 2015 at 07:58:03PM +0200, Borislav Petkov wrote: > > +#ifdef CONFIG_X86_TSC > > + /* return the value in ns */ > > + return cycles_2_ns(rdtsc()); > > +#else > > Ok, we have a problem here: > > See the splat below. On the init path > > start_kernel > |-> sched_init > |-> i

Re: [RFC PATCH] x86: Kill notsc

2015-10-21 Thread Borislav Petkov
On Sun, Oct 18, 2015 at 04:20:07PM +0200, Borislav Petkov wrote: > Ok, > > let's try this and see where it takes us. Patch has been only lightly > tested in kvm - I'll hammer on it for real once we agree about the > general form. > > Aanyway, this patch is something Peter and I have been talking

Re: [RFC PATCH] x86: Kill notsc

2015-10-19 Thread Borislav Petkov
On Sun, Oct 18, 2015 at 09:47:21PM -0700, Andy Lutomirski wrote: > I'm a bit confused. Is it currently the case that, if you boot a > normal kernel on a TSC-less machine without notsc, it fails? No, it shouldn't fail because we're checking CPUID flags. The point of this patch is purely cleaning

Re: [RFC PATCH] x86: Kill notsc

2015-10-18 Thread Andy Lutomirski
On Sun, Oct 18, 2015 at 7:20 AM, Borislav Petkov wrote: > Ok, > > let's try this and see where it takes us. Patch has been only lightly > tested in kvm - I'll hammer on it for real once we agree about the > general form. > > Aanyway, this patch is something Peter and I have been talking about on >

[RFC PATCH] x86: Kill notsc

2015-10-18 Thread Borislav Petkov
Ok, let's try this and see where it takes us. Patch has been only lightly tested in kvm - I'll hammer on it for real once we agree about the general form. Aanyway, this patch is something Peter and I have been talking about on IRC a couple of times already. I finally found some free time to poke