Re: [Xen-devel] [xen-4.6-testing test] 123408: regressions - FAIL

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 01.06.18 at 06:43, <osstest-admin@xxxxxxxxxxxxxx> wrote:
> flight 123408 xen-4.6-testing real [real]
> http://logs.test-lab.xenproject.org/osstest/logs/123408/ 
> 
> Regressions :-(
> 
> Tests which did not succeed and are blocking,
> including tests which could not be run:
>  test-amd64-amd64-xl-xsm       7 xen-boot                 fail REGR. vs. 
> 122997

Now this is unexpected:

(XEN) Xen-e820 RAM map:
(XEN) 
(XEN) ****************************************
(XEN) Panic on CPU 0:
(XEN) Not enough memory to relocate the dom0 kernel image.
(XEN) ****************************************

Not very helpful of course that we don't see any portion of the memory
map. However, looking at the code I think this suggests that the table
has zero entries.

All the failures in this flight are on one of the albana-s, so there's surely
some connection (but I'm not meaning to say at this point that it's a host
problem - we might well be corrupting memory somewhere). Going back
in serial log I see that 4.7 and 4.8 have the same issue, while 4.11 is fine.
But it's booting in EFI mode, which explains the absence of any E820
entries:

(XEN) EFI RAM map:

Assuming that the EFI booting relies on Daniel's patches, I don't think
branches older than 4.9 should be tested on these hosts, unless booting
without grub (i.e. launching xen.efi either from the EFI shell or from the
EFI boot manager).

Jan



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel