Re: next/master bisection: baseline.login on ox820-cloudengines-pogoplug-series-3

[Ard Biesheuvel]

On Fri, 18 Dec 2020 at 20:12, Ard Biesheuvel <ardb@xxxxxxxxxx> wrote:
>
> On Fri, 18 Dec 2020 at 15:01, Ard Biesheuvel <ardb@xxxxxxxxxx> wrote:
> >
> > On Fri, 18 Dec 2020 at 15:00, Russell King - ARM Linux admin
> > <linux@xxxxxxxxxxxxxxx> wrote:
> > >
> > > On Fri, Dec 18, 2020 at 01:48:09PM +0000, Guillaume Tucker wrote:
> > > > Please see the bisection report below about a boot failure on
> > > > ox820-cloudengines-pogoplug-series-3.  There was also a bisection
> > > > yesterday with next-20201216 which landed on the same commit, on
> > > > the same platform and also with oxnas_v6_defconfig.  I'm not
> > > > aware of any other platform on kernelci.org showing the same
> > > > regression.
> > >
> > > Ah, I bet I know what's happening.
> > >
> > > We test for the presence of VFP by issuing an instruction to read
> > > FPSID. If VFP is not present, this will raise an undefined instruction
> > > exception, and we expect to head into the vfp_testing_entry code.
> > >
> > > I bet Pogoplug, being an ARM11 MPCore platform, either raises an
> > > exception here.
> > >
> > > We probably need to also rework the code in vfp_init() as well to
> > > register a temporary hook when reading the FPSID.
> > >
> >
> > Thanks for diagnosing that - I wasn't quite sure what was going on.
> >
> > I will look into this later today.
>
> Working again with my fix applied:

https://kernelci.org/test/plan/id/5fdcebf1a6350e2bd1c94cd3/

I'll drop it into the patch system tomorrow (unless there is a
pressing need to do it earlier)