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][VTD] Fix apic pin to interrupt remapping table in

To: 'Keir Fraser' <keir.fraser@xxxxxxxxxxxxx>
Subject: RE: [xen-devel][PATCH][VTD] Fix apic pin to interrupt remapping table index
From: "Han, Weidong" <weidong.han@xxxxxxxxx>
Date: Thu, 4 Jun 2009 15:47:43 +0800
Accept-language: en-US
Acceptlanguage: en-US
Cc: 'Isaku Yamahata' <yamahata@xxxxxxxxxxxxx>, 'xen-devel' <xen-devel@xxxxxxxxxxxxxxxxxxx>, 'Jan Beulich' <JBeulich@xxxxxxxxxx>, "Cui, Dexuan" <dexuan.cui@xxxxxxxxx>
Delivery-date: Thu, 04 Jun 2009 00:49:10 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <C64D37FF.CB9A%keir.fraser@xxxxxxxxxxxxx>
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/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <715D42877B251141A38726ABF5CABF2C054590BADE@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> <C64D37FF.CB9A%keir.fraser@xxxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
Thread-index: AcnkLZrJFGVElBkwRdefjzY4Cbw2dwAFaH6lABtrdeAADY5d9QAAGJDQ
Thread-topic: [xen-devel][PATCH][VTD] Fix apic pin to interrupt remapping table index
Keir Fraser wrote:
> Ok, we do already support x2apic though, so I don't know what extra
> Dexuan's patch will do? Is x2apic+intremap currently broken?
> 

No break now. I think current x2apic implementation is incomplete. Per spec, 
interrupt remapping is pre-requirement for x2apic. So should enable interrupt 
remapping before x2apic. I think your modified patch is fine for now.  We can 
change it when it's really necessary.

Regards,
Weidong

>  -- Keir
> 
> On 04/06/2009 02:16, "Han, Weidong" <weidong.han@xxxxxxxxx> wrote:
> 
>> Keir,
>> 
>> Yes, your modified patch saves memory. I wanted to do like it. But
>> Dexuan is working on a x2apic patch, which will move interrupt
>> remapping enabling before IOAPIC setup. So I'm wondering ioapic
>> stuffs (nr_ioapic_registers[], nr_ioapics, etc.) aren't ready when
>> enable interrupt remapping after that moving. Dexuan, can you have a
>> look at it? 
>> 
>> Regards,
>> Weidong
>> 
>> Keir Fraser wrote:
>>> Wasteful of memory, so I checked in a modified version as c/s 19707,
>>> which dynamically sizes the array. Please take a look and check
>>> it's okay. 
>>> 
>>> It probably breaks ia64 build due to undefined nr_ioapics and
>>> nr_ioapic_registers[], but I think yours broke ia64 too so we're
>>> even. :-) 
>>> 
>>> Isaku: can you suggest ia64 equivalents for nr_ioapics and
>>> nr_ioapic_registers[]? We can do some ifdef magic at the top of
>>> intremap.c, including defining a nr_ioapic_registers() macro, if
>>> that helps. 
>>> 
>>>  Thanks,
>>>  Keir
>>> 
>>> On 03/06/2009 10:28, "Han, Weidong" <weidong.han@xxxxxxxxx> wrote:
>>> 
>>>> Originally, it calls xmalloc to set index in
>>>> ioapic_rte_to_remap_entry(). When make with debug=y, it may trigger
>>>> spinlock BUG_ON because allocate memory with interrupt disabled.
>>>> 
>>>> This patch doesn't allocate list_head entry in
>>>> ioapic_rte_to_remap_entry(), instead allocate the array in
>>>> enable_intremap() to avoid allocating memory with interrupt
>>>> disabled. 
>>>> 
>>>> 
>>>> Signed-off-by: Weidong Han <weidong.han@xxxxxxxxx>


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