Re: port-sparc64/46461: Sun E450 drops to DDB on boot of 6.0_BETA

[christos%zoulas.com@localhost (Christos Zoulas)]

The following reply was made to PR port-sparc64/46461; it has been noted by 
GNATS.

From: christos%zoulas.com@localhost (Christos Zoulas)
To: gnats-bugs%NetBSD.org@localhost, 
port-sparc64-maintainer%netbsd.org@localhost, 
        gnats-admin%netbsd.org@localhost, netbsd-bugs%netbsd.org@localhost, 
he%NetBSD.org@localhost
Cc: 
Subject: Re: port-sparc64/46461: Sun E450 drops to DDB on boot of 6.0_BETA
Date: Sun, 20 May 2012 10:03:53 -0400

 On May 20,  8:25am, he%NetBSD.org@localhost (Havard Eidnes) wrote:
 -- Subject: Re: port-sparc64/46461: Sun E450 drops to DDB on boot of 6.0_BETA
 
 | The following reply was made to PR port-sparc64/46461; it has been noted by 
GNATS.
 | 
 | From: Havard Eidnes <he%NetBSD.org@localhost>
 | To: macallan%netbsd.org@localhost
 | Cc: gnats-bugs%NetBSD.org@localhost
 | Subject: Re: port-sparc64/46461: Sun E450 drops to DDB on boot of 6.0_BETA
 | Date: Sun, 20 May 2012 10:25:05 +0200 (CEST)
 | 
 |  >  Ouch, I think I see where the problem comes from.
 |  >  That path is apparently never taken if the console is explicitly set
 |  >  to another fb or a tty since in that case OF doesn't setup the
 |  >  graphics hardware and drivers abort much earlier, which is why I never
 |  >  saw it. In your case OF sets it up, then decides it doesn't have a
 |  >  keyboard and switches to ttya. I'll fix that, there's probably a bunch>  
 |  >  of other drivers which may get bitten by that :/
 |  >  I'm fairly sure that used to work though, at least sbus and UPA
 |  >  framebuffers are set up even if they're not the console.
 |  >  
 |  >  Does this help:
 |  >  @@ -326,11 +335,10 @@
 |  >                       defattr);
 |  >                   vcons_replay_msgbuf(&sc->sc_console_screen);
 |  >           } else {
 |  >  -               /*
 |  >  -                * since we're not the console we can postpone the rest
 |  >  -                * until someone actually allocates a screen for us
 |  >  -                */
 |  >  -               (*ri->ri_ops.allocattr)(ri, 0, 0, 0, &defattr);
 |  >  +               if (sc->sc_console_screen.scr_ri.ri_rows == 0) {
 |  >  +                       /* do some minimal setup to avoid weirdnesses 
later */
 |  >  +                       vcons_init_screen(&sc->vd, 
&sc->sc_console_screen, 1, &defattr);
 |  >  +               }
 |  >
 |  >  ?
 |  
 |  Yes, this fixes the problem, I now have the 6.0_BETA kernel
 |  running with the pm2fb driver enabled.
 
 Can you disable to pm2fb and see why genfb crashes too?
 
 christos