On Wed, 2023-08-30 at 11:43 +0100, Richard Purdie via
lists.openembedded.org wrote:
> On Fri, 2023-08-25 at 07:34 +0100, Richard Purdie via
> lists.openembedded.org wrote:
> > > > 
> > > > qemu 8.1.0 brings with it a new set of problems but I've reproduced the
> > > > hang with 8.1.0 so it does not solve that.
> > > > 
> > > > I'm really struggling to understand which change brought in these
> > > > issues for qemuppc.
> > > 
> > > Are these issues visible on mickledore branch? Maybe mickledore with 
> > > kernel 6.1 stable update or
> > > qemu 7.2 update to 8.y.x could be tested too. At least then kernel or 
> > > qemu could be blamed
> > > for the issues.
> > 
> > Not that I know of.
> > 
> > I have now also reproduced the failure with glibc 2.37 instead of 2.38
> > including the fortify sources change and the 6.1.34 kernel so there is
> > something else causing this.
> > 
> > I've wondered if we need to try going back to qemu 7.2. It may also be
> > worth ruling out binutils.
> > 
> > It shouldn't be systemd as the sysvinit images show the issue too.
> 
> I've ruled out the binutils upgrade, the glibc upgrade, systemd, the
> kernel changes and the tar, libarchive and qemu upgrades.
> 
> I've continued to try and narrow things down and we see the issue from
> this commit onwards:
> 
> https://git.yoctoproject.org/poky/commit/?id=12d9280c3de24c1c2b835e80fa1b8be72e9bc63a
> 
> I did get three clean runs with:
> 
> https://git.yoctoproject.org/poky/commit/?id=fb51e196a978d452e6a14a8343832659da97fdc7
> 
> but that still could be false negatives as it is intermittent.
> 
> I'm trying builds of the commits between those two to see if any
> pattern emerges.
> 
> The qemu 8.1.0 upgrade breaks x86 with qemu kernel hangs seemingly with
> nfs root relatively consistently:
> 
> https://autobuilder.yoctoproject.org/typhoon/#/builders/145/builds/424/steps/13/logs/stdio
> https://autobuilder.yoctoproject.org/typhoon/#/builders/148/builds/430/steps/12/logs/stdio
> https://autobuilder.yoctoproject.org/typhoon/#/builders/145/builds/427/steps/12/logs/stdio
> 
> https://autobuilder.yoctoproject.org/typhoon/#/builders/148/builds/433/steps/12/logs/stdio

This can be reproduced with a "bitbake core-image-minimal" then:

runqemu nographic 
/XXX/tmp/deploy/images/qemux86-64/core-image-minimal-qemux86-64.rootfs.tar.bz2  
bootparams=" printk.time=1"

If you add "kvm" to the command, it works.

Interestingly dropping the nographic resulted in a different crash with
rcu stalls and fun timestamps.

> I think this was where I was previously worrying about x86 kernel hangs
> from. Now we know the 8.1.0 upgrade is the cause of this, we can
> hopefully get to the bottom of it more quickly.

Bisection brings us fairly conclusively to:

https://git.yoctoproject.org/poky/commit/?id=ffd73bef9b9bb5c94c050387941eee29719ca697

"yocto-uninative: Update to 4.2 for glibc 2.38"

which means somehow qemu for qemuppc is breaking when used with glibc
2.38.

Why? No idea.

Cheers,

Richard
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#186887): 
https://lists.openembedded.org/g/openembedded-core/message/186887
Mute This Topic: https://lists.openembedded.org/mt/100733646/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to