WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-ia64-devel

RE: [Xen-ia64-devel] Question about vmx_ivt.S

To: "Tristan Gingold" <tgingold@xxxxxxx>
Subject: RE: [Xen-ia64-devel] Question about vmx_ivt.S
From: "Xu, Anthony" <anthony.xu@xxxxxxxxx>
Date: Wed, 10 Oct 2007 09:09:10 +0800
Cc: Xen-ia64-devel <xen-ia64-devel@xxxxxxxxxxxxxxxxxxx>
Delivery-date: Tue, 09 Oct 2007 18:10:07 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
In-reply-to: <20071010010831.GA2542@saphi>
List-help: <mailto:xen-ia64-devel-request@lists.xensource.com?subject=help>
List-id: Discussion of the ia64 port of Xen <xen-ia64-devel.lists.xensource.com>
List-post: <mailto:xen-ia64-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-ia64-devel>, <mailto:xen-ia64-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-ia64-devel>, <mailto:xen-ia64-devel-request@lists.xensource.com?subject=unsubscribe>
References: <20070930152627.GA2622@saphi> <51CFAB8CB6883745AE7B93B3E084EBE2011E1516@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> <1191920896.470b4500bde50@xxxxxxxxxxx> <51CFAB8CB6883745AE7B93B3E084EBE2011E1672@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> <20071010010831.GA2542@saphi>
Sender: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx
Thread-index: AcgK2U4lZjRktIrYTXK/RHsiJLPEGwAALNew
Thread-topic: [Xen-ia64-devel] Question about vmx_ivt.S
Hi Tristan,

You are right, mf can be deleted.

Thanks
Anthony


>From: Tristan Gingold [mailto:tgingold@xxxxxxx]
>Sent: 2007年10月10日 9:09
>To: Xu, Anthony
>Cc: tgingold@xxxxxxx; Xen-ia64-devel
>Subject: Re: [Xen-ia64-devel] Question about vmx_ivt.S
>
>On Tue, Oct 09, 2007 at 04:23:20PM +0800, Xu, Anthony wrote:
>>
>> VHPT speculative load happens in the same time when tlb_miss handler is
>executing.
>[...]
>>
>> Mf is to make sure that before modifying vhpt entry, vhpt entry must be
>disabled, otherwise VHPT walker hardware may see enabled half modified vhpt
>entry(definitely wrong entry), and load it into TLB cache.
>
>I do understand why the VHPT entry must be disabled before being modified
>and
>enabled after.  I do not understand why the MF is required.
>>
>>
>> Notice
>> For example, Write2 is after write1,
>> Write2 may be visible before write1.
>
>'Visible' is defined only wrt other processor.  But only the local
>processor
>will use the VHPT.  Therefore, no need to make the write visible.
>
>In your example, if the memory is read between write1 and write2, the value
>will *always* be the value of write1. (if no other processor modifies the
>memory at the address).
>
>Tristan.

_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel

<Prev in Thread] Current Thread [Next in Thread>