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-devel

Re: [Xen-devel] [PATCH] linux/i386: allow CONFIG_HIGHPTE on i386

To: Jan Beulich <jbeulich@xxxxxxxxxx>, Keir Fraser <keir@xxxxxxxxxxxxx>
Subject: Re: [Xen-devel] [PATCH] linux/i386: allow CONFIG_HIGHPTE on i386
From: Keir Fraser <keir@xxxxxxxxxxxxx>
Date: Wed, 10 Jan 2007 11:28:43 +0000
Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
Delivery-date: Wed, 10 Jan 2007 03:28:46 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
In-reply-to: <45A4D8A8.76E4.0078.0@xxxxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
Thread-index: Acc0qnvVuq/0TaCdEduNDgAX8io7RQ==
Thread-topic: [Xen-devel] [PATCH] linux/i386: allow CONFIG_HIGHPTE on i386
User-agent: Microsoft-Entourage/11.2.5.060620
On 10/1/07 11:14, "Jan Beulich" <jbeulich@xxxxxxxxxx> wrote:

>> Is it actually safe to circumvent tlb_remove_page()? It does rather more
>> than just free_page()! That's why I used a ForeignPage hook when writing the
>> i386 code. It's certainly *not* safe to assume that whoever wrote that part
>> of the x86/64 port understood all the ramifications of what they were doing.
> 
> Not really, as I understand it - it either frees the page (when the TLB is in
> fast
> mode) or inserts the page into the gather list, bumping the contained pages
> count and freeing all of them and flushing the hardware TLBs if exceeding the
> threshold - the latter step is otherwise done from tlb_finish_mmu().

Presumably this gathering happens because it may not be safe to free the
pages without having first flushed the TLBs? Freeing the page with no TLB
flush management at all would be a bad idea in that case.

And I said that sync'ing with i386 is a good general rule. I trust the code
a lot more. :-)

 -- Keir


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