RE: VTI hitting the blocked state (wasRE:[PATCH][RESEND]RE:[Xen-ia64-devel] [PATCH 0/6] Addfullevtchnmechanism forxen/ia64)

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

As a step to reduce suspicious points, I tried similar experiment on Rev 
10138 ([IA64] update default build configs) before the evtchn patch sets 
and then I observed same 'block' issue for VTI domain. Investigation is 
on going.

Thanks,
Kevin

>-----Original Message-----
>From: Tian, Kevin
>Sent: 2006年5月25日 11:50
>To: Tian, Kevin; Alex Williamson
>Cc: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
>Subject: RE: VTI hitting the blocked state
>(wasRE:[PATCH][RESEND]RE:[Xen-ia64-devel] [PATCH 0/6]
>Addfullevtchnmechanism forxen/ia64)
>
>Hi, Alex,
>       I've reproduced your observation at 8th wget for a 670MB image,
>and
>now is looking into it.
>
>Thanks,
>Kevin
>
>>From: Tian, Kevin
>>Sent: 2006年5月25日 10:36
>>>From: Alex Williamson [mailto:alex.williamson@xxxxxx]
>>>Sent: 2006年5月25日 10:26
>>>
>>>On Thu, 2006-05-25 at 10:04 +0800, Tian, Kevin wrote:
>>>>
>>>> VTI network has not been tested aggressively and thus your
>>>observation
>>>> is new and we'll take a look.
>>>>
>>>> I'll try again by following your steps. As a confirmation required from
>>>you:
>>>>    - Simple dom0+domVTI can trigger the block issue. DomU can
>>>> accelerate
>>>
>>>Correct
>>>
>>>>    - dom0 and domVTI are running on different processors.
>>>
>>>Correct, I disabled threads and had dom0 running on cpu0 and
>domVTI
>>>running on cpu3 (w/ affinity of any cpu. ie it was the default
>>>placement)
>>>>    - can that happen steadily? If not, how about the frequency.
>>>
>>>Yes, happened every time I exercised domVTI.
>>>
>>>>    - which test case do you propose to try? 'wget' or KB.
>Previously I
>>>> always try with wget.
>>>
>>>wget hit it much faster, within a few tries pulling a 512MB file.  The
>>>kernel build ran for maybe 30 minutes (very roughly).  Thanks,
>>>
>>>     Alex
>>>
>>
>>OK, previous my test is based on a 128MB file, but that shouldn't matter
>>since both copied to /dev/null. Anyway, I'll create a bigger file to see
>>whether to reproduce.
>>
>>Thanks,
>>Kevin
>>
>>_______________________________________________
>>Xen-ia64-devel mailing list
>>Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
>>http://lists.xensource.com/xen-ia64-devel

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