imx27: Unsupported ECC algorithm

[Miquel Raynal]

Hello,

Fabio Estevam <festevam at gmail.com> wrote on Tue, 29 Sep 2020 13:31:03
-0300:

> Hi Miquel,
> 
> On Tue, Sep 29, 2020 at 8:39 AM Fabio Estevam <festevam at gmail.com> wrote:
> 
> > linux-next from 20200928 still reports the same error:
> >
> > https://storage.kernelci.org/next/master/next-20200928/arm/imx_v4_v5_defconfig/gcc-8/lab-pengutronix/baseline-imx27-phytec-phycard-s-rdk.html
> >
> > I checked that the fix pointed out by Sascha is not present in this tree though.
> >
> > I do see it fixed in linux-next from today (20200929), but the report
> > from 20200929 is not yet available.
> >
> > I will let you know after kernelci publishes the boot results from today's next.  
> 
> linux-next 20200929 has an even worse log with a new division by zero:
> 
> https://storage.kernelci.org/next/master/next-20200929/arm/imx_v4_v5_defconfig/gcc-8/lab-pengutronix/baseline-imx27-phytec-phycard-s-rdk.html

I found a board (NTC C.H.I.P, sun5i R8 SoC) with a parallel NAND.

with the same DT properties but a different NAND controller driver I
could not reproduce the issue.

However, I did tried with soft BCH and I found a related error which I
just fixed. I also got a bug report from Intel's robot which I also
fixed. Perhaps what you see is related (still around this ->total entry
not being set at the right moment at probe time).

If tomorrow's report still returns an error, we will need someone
to get physical access to the device to add debugging code and track
down the failure.

Thanks,
Miquèl