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

Re: [Xen-devel] [xen-unstable test] 11574: tolerable FAIL

Tim Deegan writes ("Re: [Xen-devel] [xen-unstable test] 11574: tolerable FAIL"):
> At 16:00 +0000 on 24 Jan (1327420844), Ian Jackson wrote:
> Hmm, Didn't have to pull on that thread too hard to find it's not tied
> to anything.  The access_* arguments to hvm_hap_nested_page_fault()
> aren't plumbed in on AMD:
>     ret = hvm_hap_nested_page_fault(gpa, 0, ~0ul, 0, 0, 0, 0);
> so gating behaviour on them is not going to work so well.  Sorry about
> that - I should definitely have caught this.  (But Andres, did you test
> this, or any of your mm work, on AMD?)

In general it's probably unrealistic to ask every submitter to test
every patch on two different systems...

> The attached patch ought to fix it.  Smoke-tested but I won't have
> good enough access to my test machines to check Windows installs before
> Thursday. 

I'd be quite happy if this patch went into -unstable right away.  We
should be able to tell from the automatic tests whether it is awful
:-) and the current situation is rather poor.

Also when we have got rid of this host-specific failure I can push my
new test machinery branch which is intended to (mostly) prevent
host-specific failures being regarded as heisenbugs.


Xen-devel mailing list



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