Re: IBM Workpad z50 dmesg collection

[David Brownlee <abs%absd.org@localhost>]

On Tue, 25 May 2021 at 22:38, Alexander Bochmann <ab%lists.gxis.de@localhost> wrote:
>
> ...on Sun, May 23, 2021 at 03:42:58PM +0100, David Brownlee wrote:
>
>  > Re the pcmcia issue, as Jason has fixed the ifattr attach issue,
>  > tomorrow's nyftp HEAD build may give more information, plus if you
>
> As I wrote off list, HEAD as of 2021-05-23 boots when using the
> serial console after the latest changes (yay!), but panics in
> wsdisplay.c on the local screen.

OK, so that's a separate slice of fun to track down, but the fact that
we have working pcmcia on HEAD is very encouraging. Does the system
boot to multi-user (should be able to still test with a 7.0.2
userland), and does it suffer from the random userland segfaults?

>  > have time to test some of the 2018-2020 test kernels to see when it
>  > was introduced I can fill in more builds in whatever gap that
>  > identifies.
>
> 2020-01-01 is the first from that batch that doesn't recognize wd0 as
> a boot device. Before looks fine in that regard. Didn't find the time
> to go back using the older kernels yet. Depending on the state of the
> system, power and batteries have to be removed to boot back into Windows
> CE and restart from there...

So, to confirm my understanding we have three issues

1) wsdisplay panic: new in HEAD (not in -9)
Fixing this is obviously going to be needed going forward (I'll start
a new thread for that to try to reduce confusion, or at least move it
around a little)

2) pcmcia issue: in -9 (fixed in HEAD)
It's excellent news that this is fixed in HEAD. It would be nice to
run some more tests to try to track down when it was fixed and see if
we can get a change pulled up into -9. As we know 2020-01-01 has the
issue we want to test later kernels. Can you test 2021-01-01 , and
I'll start filling in between 2020-01 and 2021-01? We could start
digging through code, but it's always nice to have a "this binary
works and that one does not" check, and this can narrow down the
change window.

3) userland random crashes: in -8 & -9 (unknown if fixed in HEAD?)
This one is the big unknown - if it's fixed in HEAD we're in a very
good place, otherwise we have more fun ahead :)

David