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


Groups > linux.kernel > #1584758

Re: RFC: Getting rid of LTR in VMX

From Andy Lutomirski <luto@amacapital.net>
Newsgroups linux.kernel
Subject Re: RFC: Getting rid of LTR in VMX
Date 2017-02-20 18:10 +0100
Message-ID <tcZDk-7to-1@gated-at.bofh.it> (permalink)
References <tc42l-4Wt-5@gated-at.bofh.it> <tcU0W-3Tt-17@gated-at.bofh.it> <tcZjY-77l-13@gated-at.bofh.it> <tcZtE-7aS-17@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Mon, Feb 20, 2017 at 8:51 AM, Paolo Bonzini <pbonzini@redhat.com> wrote:
>
>
> On 20/02/2017 17:46, Andy Lutomirski wrote:
>> On Mon, Feb 20, 2017 at 3:05 AM, Paolo Bonzini <pbonzini@redhat.com> wrote:
>>>
>>>
>>> On 18/02/2017 04:29, Andy Lutomirski wrote:
>>>> There's no code here because the patch is trivial, but I want to run
>>>> the idea by you all first to see if there are any issues.
>>>>
>>>> VMX is silly and forces the TSS limit to the minimum on VM exits.  KVM
>>>> wastes lots of cycles bumping it back up to accomodate the io bitmap.
>>>
>>> Actually looked at the code now...
>>>
>>> reload_tss is only invoked for userspace exits, so it is a nice-to-have
>>> but it wouldn't show on most workloads.  Still it does save 150-200
>>> clock cycles to remove it (I just commented out reload_tss() from
>>> __vmx_load_host_state to test).
>>
>> That's for anything involving userspace or preemption, right?
>
> Yes.  But 150-200 clock cycles are nothing compared to the cache misses
> you get from preemption, so I'd ignore that.  Saving 300 clock cycles on
> userspace exits from TR+GSBASE would be about 5% on my Haswell.

That's still 5% :)

>
>> That being said, vmx_save_host_state() is, um, poorly optimized.
>
> Yeah, but again it doesn't run that often in practical cases.

But it will if someone ever tries to upstream Google's approach of not
emulating anything in the kernel. :)

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


Thread

RFC: Getting rid of LTR in VMX Andy Lutomirski <luto@amacapital.net> - 2017-02-18 04:40 +0100
  Re: RFC: Getting rid of LTR in VMX Paolo Bonzini <pbonzini@redhat.com> - 2017-02-18 16:20 +0100
  Re: RFC: Getting rid of LTR in VMX Paolo Bonzini <pbonzini@redhat.com> - 2017-02-20 12:10 +0100
    Re: RFC: Getting rid of LTR in VMX Andy Lutomirski <luto@amacapital.net> - 2017-02-20 17:50 +0100
      Re: RFC: Getting rid of LTR in VMX Paolo Bonzini <pbonzini@redhat.com> - 2017-02-20 18:00 +0100
        Re: RFC: Getting rid of LTR in VMX Andy Lutomirski <luto@amacapital.net> - 2017-02-20 18:10 +0100
          Re: RFC: Getting rid of LTR in VMX Paolo Bonzini <pbonzini@redhat.com> - 2017-02-20 23:10 +0100
            Re: RFC: Getting rid of LTR in VMX hpa@zytor.com - 2017-02-20 23:30 +0100

csiph-web