Bug#1003574: marked as done (segfault in libc-2.33.so during i386 boot ofde QEMU VM)
Your message dated Wed, 12 Jan 2022 14:08:35 +0100
with message-id <563e9bc6-b084-0a40-4572-56d40298e8a3@debian.org>
and subject line Re: Bug#1003574: segfault in libc-2.33.so during i386 boot ofde QEMU VM
has caused the Debian Bug report #1003574,
regarding segfault in libc-2.33.so during i386 boot ofde QEMU VM
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)
--
1003574: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1003574
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message ---
Package: libc6
Version: 2.33-2
Severity: normal
When booting an i386 VM built for autopkgtests, I see the following
segfault during boot:
> [ 1.374128] Freeing unused kernel image (initmem) memory: 940K
> [ 1.384002] Write protecting kernel text and read-only data: 11292k
> [ 1.384526] Run /init as init process
> Loading, please wait...
> Starting version 250.2-1
> [ 1.406157] udevadm[106]: segfault at bc0000 ip b7d9f638 sp bf989cb8 error 6 in libc-2.33.so[b7c6e000
> [ 1.407017] Code: 1c 8b 01 ca ff e3 29 d9 8d b4 26 00 00 00 00 8d 76 00 0f 18 8a c0 03 00 00 0f 18 8a
> Segmentation fault
Boot continues briefly after that, but then drops to an emergency shell.
I've tried the other popular architectures, but I only saw this on i386.
To reproduce, this requires qemu-system-x86 and autopkgtest >= 5.17.
# Build image
$ sudo autopkgtest-build-qemu \
--mirror http://deb.debian.org/debian
--arch i386 \
unstable i386.img
# Boot image. -enable-kvm assumes that this is being tested on amd64
# Optionally use -nographic for terminal output instead of GUI
$ qemu-system-i386 \
-machine q35 \
-enable-kvm \
-device virtio-serial \
-nic user,model=virtio \
-m 1024 -smp 1 \
i386.img
Filing as severity "normal" as it can't be ruled out that this is a QEMU
issue, though I would be surprised. Unfortunately, I no longer have i386
hardware on which I could test this.
--- End Message ---
--- Begin Message ---
Hi Aurelien,
thank you for the quick reply.
On 2022-01-12 11:45, Aurelien Jarno wrote:
>> # Boot image. -enable-kvm assumes that this is being tested on amd64
>> # Optionally use -nographic for terminal output instead of GUI
>> $ qemu-system-i386 \
>> -machine q35 \
>> -enable-kvm \
>
> You might also want to try without -enable-kvm
Indeed, this fixed the issue.
So sorry for the noise. I was 120% sure that I had tried that.
>> -device virtio-serial \
>> -nic user,model=virtio \
>> -m 1024 -smp 1 \
>> i386.img
>
> Unfortunately I have not been able to reproduce this issue here, the
> image boots perfectly. This is using an up to date sid system. The
> version of QEMU might be an important factor, and maybe your CPU.
Just to document this: This was on a Ryzen 3900X, on bullseye, with qemu
from bullseye-backports (which is just one minor release behind sid).
I'll try reproduce this on sid as soon as I have one up running again,
but this bug report can definitely be closed.
Thank you for your help!
Best,
Christian
--- End Message ---
Reply to: