Re: [Xen-devel] Regression: Windows 7 unable to boot with latest xen-unstable staging

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 03.11.14 at 16:41, <fabio.fantoni@xxxxxxx> wrote:
> I tested now latest xen-unstable staging branch of git (commit 
> 5283b310e14884341f51be35253cdd59c4cb034c) and there is a regression that 
> make unable to boot windows 7 pro 64 bit that on build of some days ago 
> was working, I not found warning/errors in qemu log and xl dmesg.
> I tested also Fedora 20 hvm domU that instead still boot correctly.
> I also tried disabling some features (viridian, hdaudiocard, usb 
> redirection, vdagent, qxl) but still not boot, remain at initial black 
> screen that says "starting windows", in xl list vcpu still remain 1 even 
> if 2 setted and time continue small increase.
> Probably the regression is caused by one of these commits for now not 
> checked with automatic test system:
> x86/HVM: only kill guest when unknown VM exit occurred...
> VMX: values written to MSR_IA32_SYSENTER_E[IS]P should...
> process softirqs while dumping domains
> 
> If you need more informations/tests tell me and I'll post them.

This doesn't match up with the most recent osstest flight (31315),
which didn't show any booting problems with Win7 guests. Hence
narrowing to one of the three commits as well as telling us any
specifics your guest (or host) configuration may differ from
osstest's (e.g. pass-through being used) would be rather helpful.

Also, just to make sure - you're not using the upstream SeaBIOS
tree, and hence aren't suffering from the regression recently
introduced there (tentative fix was already posted)?

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel