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-ppc-devel

Re: [XenPPC] Problem in running on 970 simulator

To: hs8848.jang@xxxxxxxxxxx
Subject: Re: [XenPPC] Problem in running on 970 simulator
From: Hollis Blanchard <hollisb@xxxxxxxxxx>
Date: Fri, 10 Nov 2006 13:54:41 -0600
Cc: xen-ppc-devel@xxxxxxxxxxxxxxxxxxx
Delivery-date: Fri, 10 Nov 2006 11:55:03 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
In-reply-to: <4644424.264301162965230403.JavaMail.weblogic@ep_ml02>
List-help: <mailto:xen-ppc-devel-request@lists.xensource.com?subject=help>
List-id: Xen PPC development <xen-ppc-devel.lists.xensource.com>
List-post: <mailto:xen-ppc-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-ppc-devel>, <mailto:xen-ppc-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-ppc-devel>, <mailto:xen-ppc-devel-request@lists.xensource.com?subject=unsubscribe>
Organization: IBM Linux Technology Center
References: <4644424.264301162965230403.JavaMail.weblogic@ep_ml02>
Sender: xen-ppc-devel-bounces@xxxxxxxxxxxxxxxxxxx
On Wed, 2006-11-08 at 05:53 +0000, 장현승 wrote:
> I am using 970 Simulator for testing PPC Xen but I found that PPC Xen did not 
> work on 970 simulator from rev 11963.
> 
> It hangs after domain 0 is initialized and before shell prompt comes.
> 
>  
> 
> By some tracing, I found that there is no hypervisor decrementer exception at 
> all.
> 
> So it causes the domain 0 not to get scheduled again once it sets timer and 
> be blocked by hypercall.
> 
>  
> 
> Comparing the revisions, I found that it’s because of enabling of external 
> machine check interrupts, which was added from rev 11963.
> 
> I checked out the latest xen tree and commented out that line like the 
> following:
> 
>  
> 
> diff -r b30cb72ed5e2 xen/arch/powerpc/powerpc64/ppc970.c
> 
> --- a/xen/arch/powerpc/powerpc64/ppc970.c      Fri Nov 03 16:53:17 2006 -0500
> 
> +++ b/xen/arch/powerpc/powerpc64/ppc970.c      Wed Nov 08 09:37:07 2006 +0900
> 
> @@ -200,7 +200,7 @@ void cpu_initialize(int cpuid)
> 
>      hid0.bits.en_therm = 0; /* ! Enable ext thermal ints */
> 
>      /* only debug Xen should activate ATTN */
> 
>      hid0.bits.en_attn = 1;  /* Enable attn instruction */
> 
> -    hid0.bits.en_mck = 1;   /* Enable external machine check interrupts */
> 
> +//    hid0.bits.en_mck = 1;   /* Enable external machine check interrupts */
> 
>  
> 
>  #ifdef SERIALIZE
> 
>      hid0.bits.one_ppc = 1;
> 
>  
> 
> Then it worked on 970 simulator.
> 
>  
> 
> I am curious about why it does not work on simulator with en_mck = 1
> 
> Is there anyone who can explain this? Is it kind of a bug in simulator?

Hi, sorry for the delay in responding.

It does look like a simulator bug. What version of systemsim are you
using? Mark, this code works for you, right? What version are you using?

Jang, could you report this on the systemsim forums
(http://www.alphaworks.ibm.com/tech/systemsim970/forum) and see if they
have any comments?

In the meantime, we can use on_mambo() to conditionalize your patch.

-- 
Hollis Blanchard
IBM Linux Technology Center


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

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