[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Xen-devel] [PATCH v2] x86/paravirt: Don't make vcpu_is_preempted() a callee-save function
- To: Waiman Long <longman@xxxxxxxxxx>
- From: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
- Date: Mon, 13 Feb 2017 23:31:30 +0100
- Cc: linux-arch@xxxxxxxxxxxxxxx, Juergen Gross <jgross@xxxxxxxx>, Jeremy Fitzhardinge <jeremy@xxxxxxxx>, x86@xxxxxxxxxx, kvm@xxxxxxxxxxxxxxx, Radim Krčmář <rkrcmar@xxxxxxxxxx>, Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx>, Pan Xinhui <xinhui.pan@xxxxxxxxxxxxxxxxxx>, Paolo Bonzini <pbonzini@xxxxxxxxxx>, Rusty Russell <rusty@xxxxxxxxxxxxxxx>, linux-kernel@xxxxxxxxxxxxxxx, virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx, Chris Wright <chrisw@xxxxxxxxxxxx>, Ingo Molnar <mingo@xxxxxxxxxx>, hpa@xxxxxxxxx, xen-devel@xxxxxxxxxxxxxxxxxxxx, Alok Kataria <akataria@xxxxxxxxxx>, Thomas Gleixner <tglx@xxxxxxxxxxxxx>
- Delivery-date: Mon, 13 Feb 2017 22:31:52 +0000
- List-id: Xen developer discussion <xen-devel.lists.xen.org>
On Mon, Feb 13, 2017 at 05:24:36PM -0500, Waiman Long wrote:
> >> movsql %edi, %rax;
> >> movq __per_cpu_offset(,%rax,8), %rax;
> >> cmpb $0, %[offset](%rax);
> >> setne %al;
> I have thought of that too. However, the goal is to eliminate memory
> read/write from/to stack. Eliminating a register sign-extend instruction
> won't help much in term of performance.
Problem here is that all instructions have dependencies, so if you can
get rid of the sign extend mov you kill a bunch of stall cycles (I would
expect).
But yes, peanuts vs the stack load/stores.
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel
|
