Re: grfcl breakage on current?

[Radoslaw Kujawa <radoslaw.kujawa%c0ff33.net@localhost>]

Hi.

On Nov 6, 2012, at 3:32 PM, Chuck Silvers <chuq%chuq.com@localhost> wrote:

> On Tue, Nov 06, 2012 at 01:09:34AM +0100, Radoslaw Kujawa wrote:
>> Hi.
>> 
>> Can anyone confirm/deny breakage of grfcl driver with -current ? I have a 
>> PicassoIV card, but now when trying to boot current kernel on A4000 it only 
>> gives "CIRRUS" on top of the screen and that's it. NetBSD 6.0 kernel is 
>> working correctly. I suspect that this driver got broken recently. 
> 
> it looks like this is due to my recent device/softc changes.
> the amiga video drivers abuse the autoconf framework in horrible ways

Yes, I know, they're awful. I'm thinking about rewriting these drivers, but 
that's a lot of work and I'd need to have all these cards to test (I only have 
PIcassoIV, A2410 and CV64/3D). CV64/3D and CV64 drivers are even more messy 
now, since we've (me and Frank) added wsdisplay support to them. I'd really 
like to phase out grf/ite in favour of wsdisplay for NetBSD 7.

> and I had some trouble with updating them.  I could only test the grfcc driver

Hm, so I wonder where's the difference. Perhaps grfcc works because it is 
attached in a bit different way (since it lives on mainbus, and other grf 
drivers live on zbus).

> but I hoped that I got the others right too... alas, I didn't.
> I think the attached patch will repair things, could you give it a try?

I've tested it, unfortunately it didn't help. But that's weird, looking at cvs 
logs and diffs, I expected that the patch you've sent would fix this problem. 
Maybe I'm missing something else.

I also tested the build from 20121015 and it works correctly, so it confirms 
that latest changes are indeed at fault. 

-- 
Best regards,
Radoslaw Kujawa