Subject: Re: cg4 problems revisited...
To: None <>
From: Brad Spencer <>
List: port-sun3
Date: 07/12/1996 18:30:33
>> Date: Thu, 11 Jul 1996 18:31:14 -0400 (EDT)
>> From: Brad Spencer <>
>> Basically, I redid the mmap function in the driver.  I sorta
>> stumbled on the change by mistake.
>> Here is the config line I use:
>> cgfour0 at obio0 addr 0x0b300000 level 4        # P4 4/100
>> I have not had a lot of time lately to fiddle with my 4/110, but the
>> last time I thunked on it, the following worked [around the 1.2
>> version of cgfour.c].  The information was derrived from staring at
>> the cgfour driver in OpenBSD and by some guesses [I don't know if
>> there ever was a cgfour on the Sun 4 that wasn't on a P4 bus.  That is
>> what the 0x300000 is all about in the return()].
>So, you think the mmap function is incorrect?
>If that is true, why would it work on the 3/60?

Well, for the 4/110, the sparc cgfour mmap function didn't work.  It
returned an invalid page to X.

I didn't really mean to imply that the mmap function I posted would be
directly useful, just posted for reference.  It does make the cgfour
work for me, however.

>The current one looks to me like it is roughly equivalent to
>the one sparc/cg4 given the different hardware layout of each.

I have not checked the latest version to see if it works, but as of
the 1.2 era, it didn't work at all.  Prior to that, it exhibited
problems which sounded exactly like the problem which was originally

>The hardware layout information came from the SunOS header file:
>	/usr/include/sundev/cg4reg.h
>which appears to suggest there are two layouts for the cg4
>("type A" and "type B").  My 3/60 has "type B".  Does yours?
>If we have to support both types, how do we determine which
>one we are looking at?
>Here are the relevant definitions from SunOS:
>/* A type (AMD DAC) definitions */
>#define	CG4A			0
>/* frame buffer and color map addresses */
>#define CG4A_ADDR_OVERLAY	0xFF000000
>#define CG4A_ADDR_ENABLE	0xFE400000
>#define CG4A_ADDR_COLOR		0xFE800000
>#define CG4A_ADDR_CMAP		0x000E0000
>/* B type (Brooktree DAC) definitions */
>/* frame buffer and color map addresses */
>#define CG4B_ADDR_OVERLAY	0xFF400000
>#define CG4B_ADDR_ENABLE	0xFF600000
>#define CG4B_ADDR_COLOR		0xFF800000
>#define CG4B_ADDR_CMAP		0xFF200000

I would appear that my 4/110 [I don't have a 3/xxx with anything
except a bw2 in it, I sorta assumed that the cgfour on the sparc would
be simular to the cgfour on the Sun 3] has a type B, at least the
offsets would seem to indicate this.

Brad Spencer -