[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 02 Feb 2018 21:14:02 +0000
From: David Woodhouse <dwmw2@...radead.org>
To: Paolo Bonzini <pbonzini@...hat.com>,
Filippo Sironi <sironi@...zon.de>
Cc: tglx@...utronix.de, KarimAllah Raslan <karahmed@...zon.de>,
x86@...nel.org, KVM list <kvm@...r.kernel.org>,
torvalds@...ux-foundation.org, linux-kernel@...r.kernel.org,
bp@...en8.de, peterz@...radead.org
Subject: Re: [PATCH] KVM: x86: Reduce retpoline performance impact in
slot_handle_level_range()
On Fri, 2018-02-02 at 16:10 -0500, Paolo Bonzini wrote:
> > > On 2. Feb 2018, at 15:59, David Woodhouse <dwmw@...zon.co.uk> wrote:
> > > With retpoline, tight loops of "call this function for every XXX" are
> > > very much pessimised by taking a prediction miss *every* time.
> > >
> > > This one showed up very high in our early testing, and it only has five
> > > things it'll ever call so make it take an 'op' enum instead of a
> > > function pointer and let's see how that works out...
> > >
> > > Signed-off-by: David Woodhouse <dwmw@...zon.co.uk>
>
> What about __always_inline instead?
Yeah, Linus suggested that and it looks like it generates sane (in fact
better) code. Will do some more testing and send it out for real
probably on Monday. Thanks.
Download attachment "smime.p7s" of type "application/x-pkcs7-signature" (5213 bytes)
Powered by blists - more mailing lists