Re: [linux-5.4 bisection] complete test-amd64-amd64-dom0pvh-xl-intel

[Roger Pau Monné <roger.pau@xxxxxxxxxx>]

On Wed, Jan 20, 2021 at 10:04:13AM +0100, Jan Beulich wrote:
> On 19.01.2021 19:43, osstest service owner wrote:
> > branch xen-unstable
> > xenbranch xen-unstable
> > job test-amd64-amd64-dom0pvh-xl-intel
> > testid xen-boot
> > 
> > Tree: linux 
> > git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git
> > Tree: linuxfirmware git://xenbits.xen.org/osstest/linux-firmware.git
> > Tree: ovmf git://xenbits.xen.org/osstest/ovmf.git
> > Tree: qemu git://xenbits.xen.org/qemu-xen-traditional.git
> > Tree: qemuu git://xenbits.xen.org/qemu-xen.git
> > Tree: seabios git://xenbits.xen.org/osstest/seabios.git
> > Tree: xen git://xenbits.xen.org/xen.git
> > 
> > *** Found and reproduced problem changeset ***
> > 
> >   Bug is in tree:  xen git://xenbits.xen.org/xen.git
> >   Bug introduced:  0f7bcaf19f7c67ee8c07d1674b2bbb70f16c2d11
> >   Bug not present: 04b090366ca59e8a75837c822df261a8d0bd1a30
> >   Last fail repro: http://logs.test-lab.xenproject.org/osstest/logs/158526/
> > 
> > 
> >   commit 0f7bcaf19f7c67ee8c07d1674b2bbb70f16c2d11
> >   Author: Jan Beulich <jbeulich@xxxxxxxx>
> >   Date:   Tue Jan 5 13:18:26 2021 +0100
> >   
> >       x86/vPCI: check address in vpci_msi_update()
> >       
> >       If the upper address bits don't match the interrupt delivery address
> >       space window, entirely different behavior would need to be 
> > implemented.
> >       Refuse such requests for the time being.
> >       
> >       Replace adjacent hard tabs while introducing MSI_ADDR_BASE_MASK.
> >       
> >       Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
> >       Reviewed-by: Roger Pau Monné <roger.pau@xxxxxxxxxx>
> 
> Hmm, I'm puzzled. I see
> 
> (XEN) irq.c:2812: dom0: -1:-1 already mapped to -640
> (XEN) vmsi.c:762:d0v0 0000:00:19.0: failed to map PIRQ: -16
> 
> and
> 
> (XEN) irq.c:2812: dom0: -1:-1 already mapped to -640
> (XEN) vmsi.c:762:d0v0 0000:03:00.0: failed to map PIRQ: -16
> (XEN) d0v0 0000:03:00.0: unable to enable entry 0: -16
> (XEN) irq.c:2812: dom0: -1:-1 already mapped to -640
> (XEN) vmsi.c:762:d0v0 0000:03:00.0: failed to map PIRQ: -16
> (XEN) d0v0 0000:03:00.0: unable to enable entry 1: -16
> (XEN) irq.c:2812: dom0: -1:-1 already mapped to -640
> (XEN) vmsi.c:762:d0v0 0000:03:00.0: failed to map PIRQ: -16
> (XEN) d0v0 0000:03:00.0: unable to enable entry 2: -16
> (XEN) irq.c:2812: dom0: -1:-1 already mapped to -640
> (XEN) vmsi.c:762:d0v0 0000:03:00.0: failed to map PIRQ: -16
> (XEN) d0v0 0000:03:00.0: unable to enable entry 3: -16
> (XEN) irq.c:2812: dom0: -1:-1 already mapped to -640
> (XEN) vmsi.c:762:d0v0 0000:03:00.0: failed to map PIRQ: -16
> (XEN) d0v0 0000:03:00.0: unable to enable entry 4: -16
> 
> but I specifically do not see any instance of the debugging
> message that the supposedly offending commit added. Yet without
> any instance of that message I don't see how the change could
> make any difference to the behavior.

I don't seem to be able to reproduce with my test box, so I'm setting
up a repro on the test lab to debug. Will update when I get more
information.

Roger.