Re: arm64/acpi: NULL dereference reports from UBSAN at boot

[Hanjun Guo]

On 2020/5/22 16:07, Hanjun Guo wrote:
> Hi Will,
>
> On 2020/5/21 18:09, Will Deacon wrote:
> > Hi folks,
> >
> > I just tried booting the arm64 for-kernelci branch under QEMU (version
> > 4.2.50 (v4.2.0-779-g4354edb6dcc7)) with UBSAN enabled, and I see a couple
> > of NULL pointer dereferences reported at boot. I think they're both GIC
> > related (log below). I don't see a panic with UBSAN disabled, so 
> > something's
> > fishy here.
> >
> > Please can you take a look when you get a chance? I haven't had time 
> > to see
> > if this is a regression or not, but I don't think it's particularly 
> > serious
> > as I have all sorts of horrible stuff enabled in my .config, since I'm
> > trying to chase down another bug:
> >
> > https://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.git/plain/arch/arm64/configs/fuzzing.config?h=fuzzing/arm64-kernelci-20200519&id=c149cf6a51aa4f72d53fc681c6661094e93ef660 
> >
> >
> > (on top of defconfig)
> >
> > CONFIG_FAIL_PAGE_ALLOC may be to blame.
>
> I enabled UBSAN and CONFIG_FAIL_PAGE_ALLOC on top of defconfig,
> testing against the for-kernelci branch on the D06 board, I
> can see some UBSAN warnings from megaraid_sas driver [0], but not
> from any other subsystem including ACPI, I will try all your
> configs above to see if I can get more warnings.

Enabled all the configs except "CONFIG_MODULES=n" and
"CONFIG_SHADOW_CALL_STACK=y" (can't do that via make menuconfig,
do it manually?) in the link, but still got the same warnings and
no other warnings as before, the system is in good function, please
let me know I can do anything more to help the debug.

Thanks
Hanjun