Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1495954

Re: [PATCH 2/9] x86/fpu: Hard-disable lazy fpu mode

Show all headers | View raw

On 05/10/2016 17:59, Andy Lutomirski wrote:
> I actually benchmarked the underlying instructions quite a bit on
> Intel.  (Not on AMD, but I doubt the results are very different.)
> Writes to CR0.TS are *incredibly* slow, as are device-not-available
> exceptions.  Keep in mind that, while there's a (slow) CLTS
> instruction, there is no corresponding STTS instruction, so we're left
> with a fully serializing, slowly microcoded move to CR0.  On SVM, I
> think it's worse, because IIRC SVM doesn't have fancy execution
> controls that let MOV to CR0 avoid exiting.

SVM lets you choose whether to trap on TS and MP; update_cr0_intercept
is where KVM does that (the "selective CR0 write" intercept is always
on, while the "CR0 write" intercept is toggled in that function).

> We're talking a couple
> hundred cycles best case for a TS set/clear pair, and thousands of
> cycles if we actually take a fault.
> 
> In contrast, an unconditional XSAVE + XRSTOR was considerably faster.

Did you also do a comparison against FXSAVE/FXRSTOR (on either pre- or
post-SandyBridge processors)?

But yeah, it's possible that the lack of STTS screws the whole plan,
despite the fpu.preload optimization in switch_fpu_prepare.

Paolo

> This leads to the counterintuitive result that, if we switch from task
> A to B and back and task A is heavily using the FPU, then it's faster
> to unconditoinally save and restore the full state both ways than it
> is to set and clear TS so we can avoid it.

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 2/9] x86/fpu: Hard-disable lazy fpu mode riel@redhat.com - 2016-10-05 02:40 +0200
  Re: [PATCH 2/9] x86/fpu: Hard-disable lazy fpu mode Paolo Bonzini <pbonzini@redhat.com> - 2016-10-05 09:20 +0200
    Re: [PATCH 2/9] x86/fpu: Hard-disable lazy fpu mode Rik van Riel <riel@redhat.com> - 2016-10-05 16:00 +0200
      Re: [PATCH 2/9] x86/fpu: Hard-disable lazy fpu mode Paolo Bonzini <pbonzini@redhat.com> - 2016-10-05 16:10 +0200
        Re: [PATCH 2/9] x86/fpu: Hard-disable lazy fpu mode Andy Lutomirski <luto@amacapital.net> - 2016-10-05 18:10 +0200
          Re: [PATCH 2/9] x86/fpu: Hard-disable lazy fpu mode Paolo Bonzini <pbonzini@redhat.com> - 2016-10-05 18:20 +0200
  [tip:x86/fpu] x86/fpu: Hard-disable lazy FPU mode tip-bot for Andy Lutomirski <tipbot@zytor.com> - 2016-10-07 11:50 +0200

csiph-web