From mboxrd@z Thu Jan  1 00:00:00 1970
From: Gleb Natapov <gleb@redhat.com>
Subject: Re: [PATCH kvm-unit-tests v2] vmexit: time the number of cycles for
 simple PIO
Date: Tue, 18 Dec 2012 12:56:01 +0200
Message-ID: <20121218105601.GH11016@redhat.com>
References: <1355402437-21935-1-git-send-email-pbonzini@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: kvm@vger.kernel.org, mtosatti@redhat.com
To: Paolo Bonzini <pbonzini@redhat.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:64399 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1754695Ab2LRK4E (ORCPT <rfc822;kvm@vger.kernel.org>);
	Tue, 18 Dec 2012 05:56:04 -0500
Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23])
	by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id qBIAu3bN022015
	(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK)
	for <kvm@vger.kernel.org>; Tue, 18 Dec 2012 05:56:03 -0500
Content-Disposition: inline
In-Reply-To: <1355402437-21935-1-git-send-email-pbonzini@redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Thu, Dec 13, 2012 at 01:40:37PM +0100, Paolo Bonzini wrote:
> This patch adds three scenarios to the vmexit test.  Two are very simple
> PIO cases that are handled in the kernel (reading from and writing
> to ELCR).  The other is an unmapped PIO that is handled in userspace.
> 
> The difference between the two reading scenarios is roughly the cost of a
> userspace exit; the existing inl_from_pmtimer test is not precise enough,
> because the device model has a pretty high cost.
> 
> The difference between the kernel read and write is the cost of emulation,
> because inl_from_kernel goes through the whole emulation stuff while outl
> does not (it is used for virtio, while the speed of inl matters less).
> 
> Example:
> 
>     vmcall 3898
>     inl_from_pmtimer 24615
>     inl_from_qemu 20574
>     inl_from_kernel 7237
>     outl_to_kernel 4451
> 
> So the cost of exiting to userspace is 13000 cycles on this machine,
> and the cost of emulation is 3300 cycles.
> 
> Suggested-by: Avi Kivity <avi.kivity@gmail.com>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Thanks, applied. On my machine out is faster than vmcall.

vmcall 2681
outl_to_kernel 2358

> ---
>  x86/vmexit.c | 30 ++++++++++++++++++++++++++++++
>  1 file changed, 30 insertions(+)
> 
> diff --git a/x86/vmexit.c b/x86/vmexit.c
> index ad8ab55..98f0ead 100644
> --- a/x86/vmexit.c
> +++ b/x86/vmexit.c
> @@ -4,6 +4,18 @@
>  #include "processor.h"
>  #include "atomic.h"
>  
> +static void outb(unsigned short port, int val)
> +{
> +    asm volatile("outb %b0, %w1" : "=a"(val) : "Nd"(port));
> +}
> +
> +static unsigned int inb(unsigned short port)
> +{
> +    unsigned int val;
> +    asm volatile("xorl %0, %0; inb %w1, %b0" : "=a"(val) : "Nd"(port));
> +    return val;
> +}
> +
>  static unsigned int inl(unsigned short port)
>  {
>      unsigned int val;
> @@ -82,6 +94,21 @@ static void inl_pmtimer(void)
>      inl(0xb008);
>  }
>  
> +static void inl_nop_qemu(void)
> +{
> +    inl(0x1234);
> +}
> +
> +static void inl_nop_kernel(void)
> +{
> +    inb(0x4d0);
> +}
> +
> +static void outl_elcr_kernel(void)
> +{
> +    outb(0x4d0, 0);
> +}
> +
>  static void ple_round_robin(void)
>  {
>  	struct counter {
> @@ -116,6 +143,9 @@ static struct test {
>  	{ mov_to_cr8, "mov_to_cr8" , .parallel = 1, },
>  #endif
>  	{ inl_pmtimer, "inl_from_pmtimer", .parallel = 1, },
> +	{ inl_nop_qemu, "inl_from_qemu", .parallel = 1 },
> +	{ inl_nop_kernel, "inl_from_kernel", .parallel = 1 },
> +	{ outl_elcr_kernel, "outl_to_kernel", .parallel = 1 },
>  	{ ipi, "ipi", is_smp, .parallel = 0, },
>  	{ ipi_halt, "ipi+halt", is_smp, .parallel = 0, },
>  	{ ple_round_robin, "ple-round-robin", .parallel = 1 },
> -- 
> 1.8.0.2

--
			Gleb.