Hi
On Fri, Feb 07, 2025 at 08:40:57AM +0100, Radoslav Bodó wrote:
> Hello,
>
> > On Sun, Jan 26, 2025 at 10:45:33AM +0100, Radoslav Bodó wrote:
> > > > Can you confirm if the issue happens as well if you do not boot into
> > > > Xen and dom0? Can you for the first broken kernel provide both full
> > > > boot logs?
> > > >
> > > > Does the issue happens with a kernel from backports on ideally 6.12.10
> > > > from unstable as well?
>
> I've been able to do experimentation to answer above questions with somewhat
> expected results:
>
> * any linux kernel on it's own works fine
> * last kernel working with xen4.17 is 6.1.0-26
Thanks for doing that and provding the results.
I suspect this, #1088159 and #1093371 have the same root cause.
Candidates for the breaking behaviour could be (just adding this now
for documentation purposes, so that we might not loose track):
6434af166441 ("xen: use correct end address of kernel for conflict checking")
b1e6e80a1b42 ("xen/swiotlb: add alignment check for dma buffers")
1bd4e5a74da9 ("xen/swiotlb: fix allocated size")
(all of those did land in 6.1.113).
Regards,
Salvatore
