On 2018/02/26 07:50, Israel Brewster wrote:
>     On Feb 24, 2018, at 3:06 AM, Stuart Henderson <s...@spacehopper.org> 
> wrote:
> 
>     On 2018-02-24, Israel Brewster <isr...@ravnalaska.net> wrote:
> 
>         I have an HP Compaq Pro 6300 machine on which I am trying to run
>         OpenBSD. The installer boots and runs fine, but after rebooting into 
> the
>         newly installed OS, I start getting the boot sequence (the white text 
> on
>         blue background stuff - don't know what that is officially called), 
> but
>         after a second or so the screen goes blank and that's all she wrote.
> 
>         My first thought was that it was just a display issue, and that I
>         should be able to ssh in and tweak stuff, but as it turns out, the
>         machine never shows up on the network, either, so apparently it never
>         gets far enough in the boot process to enable the network (networking
>         *does* work while I am running through the installer, so I don't think
>         it's just a missing network driver there).
> 
> 
>     It sounds like it's crashing after the video mode is changed. The
>     machine probably has inteldrm so at the boot loader prompt, try
>     "boot -c", then at UKC "disable inteldrm" and "quit".
> 
> 
> Bingo! that did the trick - got a good clean boot after running that command. 
> The only issue
> appears to be that I'm going to have to do that every time I boot. Given 
> that, how do I make it
> stick? Or, now that we know that is the issue, is there some other, more 
> permanent "fix" I can
> try? Do I need that inteldrm for any reason?

You'll want it if the machine will be running X.

It's possible to modify an on-disk kernel with config(8)'s -e flag, but
that has other problems (not least, syspatch won't be able to update the
kernel).

>     That may let it boot, if not then you're at least more likely to see
>     a hidden error message of some sort.
> 
>     If this is 6.2, try -current instead. If it's OpenBSD/i386, try amd64
>     instead.
> 
> 
> Since disabling inteldrm seemed to bypass the issue, if only temporarily, 
> would these still be
> worth trying, or were they just additional suggestions if the inteldrm thing 
> didn't work?

Definitely worth trying, it would be better to have a fix than a workaround,
and without trying -current you won't know if it's already been fixed. With
the information you've given so far we have very little idea about what you're
running or what hardware.

Please send a bug report with the files generated from sendbug (run as root).
It's often easiest to do "sendbug -P > /tmp/template.txt" then copy that to
another machine, edit to add a description etc, and send the whole thing
to b...@openbsd.org.

Reply via email to