[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Xen-ia64-devel] Problems with Xen/ia64 on xen-unstable tip


  • To: "Magenheimer, Dan \(HP Labs Fort Collins\)" <dan.magenheimer@xxxxxx>, <xen-ia64-devel@xxxxxxxxxxxxxxxxxxx>
  • From: "Tian, Kevin" <kevin.tian@xxxxxxxxx>
  • Date: Fri, 8 Jul 2005 18:13:27 +0800
  • Delivery-date: Fri, 08 Jul 2005 10:12:13 +0000
  • List-id: DIscussion of the ia64 port of Xen <xen-ia64-devel.lists.xensource.com>
  • Thread-index: AcWDEr491+R29zYdTaC0laPe9cD7LQAkN28Q
  • Thread-topic: [Xen-ia64-devel] Problems with Xen/ia64 on xen-unstable tip

Hi, Dan,
        I sent out a question to xen-dev-list for possible issue within
sedf scheduler, which makes IDLE domain scheduled more than Dom0.
Previously based on BVT scheduler, IDLE domain will never be scheduled
once dom0 is up to run, unless dom0 explicitly requests to sleep (Should
be no requests on transparent xenolinux by far?). So this may be a clue
for you considering from context switch point which will happen now even
when dom0 is running. Some code may not apply to IDLE domain... 

        Actually we already encountered a similar issue on VTI, with
patch attached where IDLE domain has different context as Dom0. Due to
bad fact that I never tried your xenolinux successfully, so only some
though provided and hope helpful.

Thanks,
Kevin

>-----Original Message-----
>From: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx
>[mailto:xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of
>Magenheimer, Dan (HP Labs Fort Collins)
>Sent: Friday, July 08, 2005 12:42 AM
>To: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
>Subject: [Xen-ia64-devel] Problems with Xen/ia64 on xen-unstable tip
>
>In preparing to catch up with the backlog of patches due
>to the switchover from bk to hg, I grabbed the latest
>tip for xen-unstable.hg.  Unfortunately, it doesn't work.
>After limited debugging, it appears that the sedf
>scheduler (recently changed from bvt) doesn't work in Xen/ia64
>and (after changing this back to see if I get further)
>a hypercall to HYPERVISOR_event_channel_op fails
>(possibly due to an interface change?).
>
>These failures are due to recent Xen core changes as
>a previous tree worked fine.
>
>I will have limited time to track these problems down
>in the next couple of days (and all patches are on hold
>until tip runs); if someone else could volunteer to track
>these down and identify fixes, I would very much
>appreciate it!
>
>Thanks,
>Dan
>
>_______________________________________________
>Xen-ia64-devel mailing list
>Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
>http://lists.xensource.com/xen-ia64-devel

Attachment: patch_domain_switch
Description: patch_domain_switch

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

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.