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

Re: [Xen-devel] [xen-4.4-testing test] 25979: regressions - FAIL

On 5/2/2014 11:20 AM, Aravind Gopalakrishnan wrote:
On 5/2/2014 4:38 AM, Jan Beulich wrote:

See my other email.  I think this is host-specific.  lace-bug is an
AMD machine.

I think perhaps this is a performance problem with SVM.
I don't think this is a new thing. I've heard speculation that this may
have been the case since the page-sharing p2m locking changes way back
in 4.2.

I've lost track of who at AMD I should be CCing with this mail -- Jan, I
guess you know?
SVM maintainers - can one of you please take looking into this as
an action item?

Yes, I'll look into this.
I've got a

cpu family    : 16
model        : 9
stepping    : 1

Will need winxpsp3 image though.

Will keep you guys posted.

Sorry for lack of activity as I have not been able dedicate much time on this issue last week.

I did try booting a winXpSp3 guest, and it came up fine for me. I am working towards setting this up as a continuous test now as the logs indicated it was stalling on 4th boot..

Anyway, these links don't seem to work anymore-
1. http://www.chiark.greenend.org.uk/~xensrcts/logs/25979/test-amd64-amd64-xl-qemut-winxpsp3/info.html 2. http://www.chiark.greenend.org.uk/~xensrcts/logs/25979/test-amd64-amd64-xl-qemut-winxpsp3/

basically all logs for test '25979' don't exist.(It could help me verify my set up..)

Something went wrong? Or I am looking at the wrong place?


Xen-devel mailing list



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