On Mon, May 02, 2022 at 01:31:00PM +0000, Liam Howlett wrote:
> * Heiko Carstens <hca@linux.ibm.com> [220502 06:18]:
> > On Sun, May 01, 2022 at 05:24:12PM -0700, Andrew Morton wrote:
> > > (cc S390 maintainers)
> > > (cc stable & Greg)
> > > 
> > > > I have been porting my patches over and hit a bit of a snag. It looked
> > > > like my patches were not booting on the s390 - but not all the time. So
> > > > I reverted back to mm-stable (059342d1dd4e) and found that also failed
> > > > to boot sometimes on my qemu setup.  When it fails it's ~4-5sec into
> > > > booting.  The last thing I see is:
> > > > 
> > > > "[    4.668916] Spectre V2 mitigation: execute trampolines"
> > > > 
> > > > I've bisected back to commit e553f62f10d9 (mm, page_alloc: fix
> > > > build_zonerefs_node())
> > > > 
> > > > With the this commit, I am unable to boot one out of three times.  When
> > > > using the previous commit I was not able to get it to hang after trying
> > > > 10+ times.  This is a qemu s390 install with KASAN on and I see no error
> > > > messages.  I think it's likely it is this patch, but no guaranteed.
> > > > 
> > > 
> > > Great, thanks.  So mapletree is absolved.
> > > 
> > > Unfortunately 059342d1dd4e was cc:stable.  Greg, you might want to pull
> > > the plug on that one if it isn't too late.
> > > 
> > > I'll await input from the S390 team, but from my reading the issues
> > > which that patch addresses aren't terribly serious, so perhaps the
> > > thing to do is to revert 059342d1dd4e (with a cc:stable) while
> > > 059342d1dd4e gets a redo?
> > 
> > I cannot confirm that Linus' tree currently has problems on s390 with
> > commit e553f62f10d9 ("mm, page_alloc: fix build_zonerefs_node()").
> > 
> > This commit was applied by Linus on 15th of April, and our CI didn't
> > report any problems since then. At least nothing that would point to
> > this commit. Also I just gave e553f62f10d9 a try (defconfig + KASAN):
> > no problems to report.
> > 
> > Same with 059342d1dd4e (mm-stable): it just works for me.
> > 
> > Liam, could you share your kernel config?
> 
> Sure thing.  See attached.

So, I can reproduce the hanging system now. However this looks like a
qemu problem on s390, since I can reproduce this only with Qemu+TCG.
Qemu with kvm works without any problems (same if I use z/VM as
hypervisor).

Janosch, Claudio, can you have a look at this please?

The commit in question is available via this git tree:
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git

(Note: I didn't try any other commit or branch, except the one
reported above)

Kernel config (again) attached.

qemu command line which I used to recreate the problem:

qemu-system-s390x -nographic -nodefaults -chardev stdio,id=c1 -device sclpconsole,chardev=c1 -m 4G -kernel bzImage

(adding -enable-kvm makes the hangs go away).

Thanks,
Heiko