[Pkg-xen-devel] Bug#1020787: linux-image-5.19.0-2-amd64: After updating to 5.19 kernel the VMs are started without XSAVE CPU flags

[Diederik de Haas]

Control: reassign -1 src:xen
Control: retitle -1 Xen: After updating to 5.19 kernel the VMs are started without XSAVE CPU flags
Control: tag -1 -moreinfo patch upstream fixed-upstream

On dinsdag 27 september 2022 14:05:59 CEST Marek Marczykowski-Górecki wrote:
> On Tue, Sep 27, 2022 at 01:59:42PM +0200, Diederik de Haas wrote:
> > On Tue, Sep 27, 2022 at 01:39:46AM +0200, Diederik de Haas wrote:
> > > In kernel 5.19 the following commits were added under
> > > ``arch/x86/kernel/fpu/``:
> > > 
> > > b91c0922bf1ed15b67a6faa404bc64e3ed532ec2 x86/fpu: Cleanup variable
> > > shadowing
> > > 8ad7e8f696951f192c6629a0cbda9ac94c773159 x86/fpu/xsave: Support
> > > XSAVEC in the kernel
> > > f5c0b4f30416c670408a77be94703d04d22b57df x86/prctl:
> > > Remove pointless task argument
> > 
> > 8ad7e8f696951f192c6629a0cbda9ac94c773159 seems to be the 'offending' commit
> > 
> > http://xenbits.xen.org/gitweb/?p=xen.git;a=commit;
> > h=c3bd0b83ea5b7c0da6542687436042eeea1e7909 is where it's committed
> > in Xen's master branch. I haven't seen a backport to 4.16 (yet?)

This would be the patch, but it needs to be backported.

> > Looking at the patch thread you linked, I get the impression that the
> > change in the Linux kernel exposed the problem on the Xen side, but the
> > Linux kernel change didn't introduce a bug (Andrew's patch seems to make
> > it more informative, not correct a bug).
> > 
> > But I'm not sure that assesment is correct. What's your take on this?
> 
> Yes, that seems correct.

Thanks, reassigning accordingly.

On dinsdag 27 september 2022 14:20:29 CEST Marek Marczykowski-Górecki wrote:
> On Tue, Sep 27, 2022 at 12:08:58PM +0000, Andy Smith wrote:
> > On Tue, Sep 27, 2022 at 01:59:42PM +0200, Diederik de Haas wrote:
> > > http://xenbits.xen.org/gitweb/?p=xen.git;a=commit;
> > > h=c3bd0b83ea5b7c0da6542687436042eeea1e7909 is where it's committed 
> > > in Xen's master branch. I haven't seen a backport to 4.16 (yet?)
> > 
> > Does this mean that all versions of hypervisor need this patch
> > backported in order to support Linux kernel 5.19+ as dom0 or domU?
> 
> Yes, I think so. I've seen the issue on Xen 4.14 too.

Agreed.

AFAIC, the question that remains is about its impact.
On my Xen server I did not see the mentioned problem, so it looks like it is 
hardware dependent? (I did not do an extensive test though)

And for systems that are affected is how bad the consequences are. What I've
seen in this bug report is an error message in the log wrt gnutls. I don't know
whether this makes every application using gnutls not working at all anymore.
I also don't know whether other applications and how many are affected and to
what extend.
Marek and Patrick, can you shed some light on this?

FTR: it 'still' is a bug that should be fixed; the question is about urgency.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 228 bytes
Desc: This is a digitally signed message part.
URL: <http://alioth-lists.debian.net/pipermail/pkg-xen-devel/attachments/20220927/2db45791/attachment.sig>