[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [Patch 2/4] Refining Xsave/Xrestore support


  • To: Jan Beulich <JBeulich@xxxxxxxxxx>
  • From: Haitao Shan <maillists.shan@xxxxxxxxx>
  • Date: Thu, 28 Oct 2010 10:31:34 +0800
  • Cc: xen-devel@xxxxxxxxxxxxxxxxxxx, Weidong Han <weidong.han@xxxxxxxxx>, Keir Fraser <keir.fraser@xxxxxxxxxxxxx>
  • Delivery-date: Wed, 27 Oct 2010 19:32:59 -0700
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=E2VwPlfbYT03sgc6l2WrAN2rcIv2eg0PUUPXrvvz3yShdifhm+P2qNMXnSL+sMA0e9 qloAvbfKbP0DVl5IUNlowTv0/tnxJl6SruK3aGLTp381jyQ3omY5xBIGw1mXyLWlwUgf JfgsGV39LMiX5yK+t4kYNFeJ21dKjNT1qRcK4=
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>

I don't know but I suppose this won't be a costing instruction at
lease no larger than writing to other CRx registers. But agreed that a
comparison to minimize the cost would be great. I can add that.

Thanks for pointing out.

Shan Haitao

2010/10/27 Jan Beulich <JBeulich@xxxxxxxxxx>:
>>@@ -1367,6 +1384,8 @@ static void __context_switch(void)
>>         memcpy(stack_regs,
>>                &n->arch.guest_context.user_regs,
>>                CTXT_SWITCH_STACK_BYTES);
>>+        if ( cpu_has_xsave )
>>+            set_xcr0(n->arch.xcr0);
>
> How slow is a write to xcr0? I.e. is it worth checking whether
> current xcr0 matches the to be written value?
>
> Jan
>
>

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.