[Macchiato] Trying to install mainline debian 9

[Steve McIntyre]

Hi Duncan,

On Mon, Oct 29, 2018 at 01:12:28AM +0000, Duncan Guthrie wrote:
>
>I'm having significant trouble booting anything. Each thing I've tried
>appears to have similar issues and I'm wanting to rule out issues with
>my UEFI build before anything.
>
>My UEFI is written to internal SPI flash, as my board has significant
>issues booting from directly from SD card (which I've contacted SolidRun
>about) - for an identical, sample error message, see the following
>forum: https://macchiatobin.net/forums/topic/sbsa-support/#post-470

ACK. My first board had exactly those problems, making it almost
impossible to do anything useful. We had *one* SD card that would boot
reliably, but otherwise almost total failure. I resorted to booting
off the network with that machine for a while. Serial boot might be an
option as a one-off, but you wouldn't want to depend on that in your
normal process!

My second board is (thankfully!) much better.

>This should not pose an issue updating the UEFI, as I can use an SPI
>programmer and the pin header. But, I am reluctant to do so.
>
>I've tried:
>
>* debian-buster-DI-alpha3-arm64-netinst.iso.
>
>This works best of everything, albeit Grub installation just hangs, and
>there's no way to get logs as I can't change to a TTY from my serial
>adaptor.

d-i on serial runs screen underneath everything. IIRC you can hit C-a
<number> (maybe C-a a <number> to switch to other sessions).

>This image appears to use the 4.16.x kernel. Stretch installer doesn't
>even work at all, seemingly.

I managed my initial installation using Stretch, but with a USB
network adapter. But that was with quite an old UEFI build.

>* Debian mini.iso from 28/19/2018.
>
>This just fails entirely. This appears to be a kernel panic, verbose log
>here: https://paste.debian.net/hidden/4cd96c8c/. Similarly the latest
>Ubuntu etc also fails, so it's not Debian-specific in this case, as far
>as I can tell. I also had to set console=uart0 for anything of note to
>appear (earlycon prints up to the console change to tty0, and the panic
>happens after this). I confess I don't understand the kernel panic
>message, but I hope it is useful to anyone reading.

"Internal error: synchronous external abort: 96000210 [#1] SMP", with
the stack trace pointing to the middle of _memcpy. Ugh.

-- 
Steve McIntyre, Cambridge, UK.                                steve at einval.com
"I can't ever sleep on planes ... call it irrational if you like, but I'm
 afraid I'll miss my stop" -- Vivek Das Mohapatra