[PATCH][RESEND] RE: [Xen-ia64-devel] [PATCH] Add support to physdev_ops

["Tian, Kevin" <kevin.tian@xxxxxxxxx>]

Thanks,
Kevin

>-----Original Message-----
>From: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx
>[mailto:xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of
>Tian, Kevin
>Sent: 2006年4月21日 11:12
>To: Alex Williamson
>Cc: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
>Subject: RE: [Xen-ia64-devel] [PATCH] Add support to physdev_ops
>
>>From: Alex Williamson [mailto:alex.williamson@xxxxxx]
>>Sent: 2006年4月21日 10:51
>>
>>On Thu, 2006-04-20 at 17:01 +0800, Tian, Kevin wrote:
>>
>>> 2. This patch has no obvious performance influence (Why
>>> I say 'obvious' is because latest p2m tree seems not very
>>> stable even without this patch. The output drifts with -10-+10%
>>> range... It's also possible due to my test environment. )
>>
>>Hi Kevin,
>>
>>   I ran some performance tests from xen-ia64-unstable on this patch
>>today and also didn't see any obvious performance difference.  My
>>results for wget for network and bonnie++ for disk were very consistent
>>between runs though.  It's nice to see that iosapic virtualization
>>appears to have such minimal overhead.  Thanks,
>>
>>      Alex
>>
>
>Hi, Alex,
>       Thanks for test with the patch. It's good to see your data always
>consistent, and that proves my environment really a bit messed.
>
>       Currently the minimal overhead is because dom0 has seldom
>access to RTE entries at run-time. Later when EOI is also virtualized,
>the overhead may increase. Just guess, and we can see exact
>influence later when final part is ready.
>
>       I'll resend the patch with Isaku's comment.
>
>Thanks,
>Kevin
>
>_______________________________________________
>Xen-ia64-devel mailing list
>Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
>http://lists.xensource.com/xen-ia64-devel

physdev_op_support.patch
Description: physdev_op_support.patch

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