Re: qemu-arm: zram: mkfs.ext4 : Unable to handle kernel NULL pointer dereference at virtual address 00000140

[Minchan Kim]

On Tue, Jun 07, 2022 at 06:51:27AM +0530, Naresh Kamboju wrote:
> On Mon, 6 Jun 2022 at 23:06, Minchan Kim <minchan@xxxxxxxxxx> wrote:
> >
> > On Thu, Jun 02, 2022 at 02:19:34PM +0530, Naresh Kamboju wrote:
> > > The following kernel crash reported while running selftests: zram: zram.sh
> > > test case on qemu-arm with Linux mainline v5.18.0 kernel version.
> > >
> > > Reported-by: Linux Kernel Functional Testing <lkft@xxxxxxxxxx>
> > >
> > > Started noticing from kernel crash from
> > > BAD: git_sha: cdeffe87f790dfd1baa193020411ce9a538446d7
> > > GOOD:   git_sha: babf0bb978e3c9fce6c4eba6b744c8754fd43d8e
> > >
> > > # git log  --oneline
> > > babf0bb978e3c9fce6c4eba6b744c8754fd43d8e..cdeffe87f790dfd1baa193020411ce9a538446d7
> > >   -- drivers/block/
> > > 98931dd95fd4 Merge tag 'mm-stable-2022-05-25' of
> > > git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
> > > 6140ae41effe zram: fix Kconfig dependency warning
> > > b3fbd58fcbb1 mm: Kconfig: simplify zswap configuration
> > > e7be8d1dd983 zram: remove double compression logic
> > > a2ad63daa88b VFS: add FMODE_CAN_ODIRECT file flag
> > > 30226b69f876 zram: add a huge_idle writeback mode
> >
> > Hi Nalesh,
> >
> > Thanks for the reporting the bug!
> >
> > I locally ran the test in my x86 KVM machine but couldn't reproduce it.
> > How is it easy to reproduce?  Could you help to bisect it?
> 
> It works on qemu_x86-64
> It is always reproducible in qemu arm32 environment.
> 
> steps to reproduce:
> # Please download and use this rootfs and kernel zImage
> 
> rootfs: https://oebuilds.tuxbuild.com/29zhlbEc3EWq2wod9Uy964Bp27q/images/am57xx-evm/rpb-console-image-lkft-am57xx-evm-20220601222434.rootfs.ext4.gz
> kernel: https://builds.tuxbuild.com/29zhqJJizU2Y7Ka7ArhryUOrNDC/zImage
> 
> Boot command,
>  /usr/bin/qemu-system-aarch64 -cpu host,aarch64=off -machine
> virt-2.10,accel=kvm -nographic -net
> nic,model=virtio,maaacaddr=BA:DD:AD:CC:09:04 -net tap -m 2048 -monitor
> none -kernel kernel/zImage --append "console=ttyAMA0 root=/dev/vda rw"
> -hda rootfs/rpb-console-image-lkft-am57xx-evm-20220601222434.rootfs.ext4
> -m 4096 -smp 2
> 
> # cd /opt/kselftests/default-in-kernel/zram
> # ./zram.sh
> 
> Allow me sometime I will try to bisect this problem.

Thanks for sharing the info. 

I managed to work your rootfs with my local arm build
based on the problematic git tip. 
However, I couldn't suceed to reproduce it.

I needed to build zsmalloc/zram built-in instead of modules
Is it related? Hmm,

Yeah, It would be very helpful if you could help to bisect it.
Thanks!

=== &< ====
create '1' zram device(s)
zram modules already loaded, kernel supports zram-control interface
[  936.188541] zram: Added device: zram1
all zram devices (/dev/zram1~1 successfully created
set max_comp_streams to zram device(s)
The device attribute max_comp_streams was deprecated in 4.7
set disk size to zram device(s)
/sys/block/zram1/disksize = '1048576'
zram set disksizes: OK
set memory limit to zram device(s)
[  936.214438] zram1: detected capacity change from 0 to 2048
/sys/block/zram1/mem_limit = '1M'
zram set memory limit: OK
make swap with zram device(s)
[  936.608854] Adding 1020k swap on /dev/zram1.  Priority:-2 extents:1 across:1020k SS
done with /dev/zram1
zram making zram mkswap and swapon: OK
zram swapoff: OK
zram cleanup
[  937.063416] zram1: detected capacity change from 2048 to 0
[  937.189223] zram: Removed device: zram1
zram02 : [PASS]
root@am57xx-evm:/opt/kselftests/default-in-kernel/zram# uname -r
5.18.0-11934-g54eb8462f21f