[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Bug in Mint 1.12?



In Email on Mon, 28 Nov 1994 12:11:52 -0500 (EST), jeffmay@magnus.acs.ohio-stat
>
>Hi.  I haven't been able to get Mint > 1.10h6 running at all on my
>TT (including the binary version 1.12).  I had some free time this
>weekend, and found that one of the calls to rsconf in init_xbios 
>was crashing. (I think it was the third one in the loop, but don't remeber
>offhand.  I can check when I get home if anyone is interested.) 
>Replacing the rsconf procedure in version 1.12h1 with
>the one from 1.10h6 and recompiling gave me a functional kernel.
>I don't really know enough about programming the serial ports to 
>narrow it down further than that.
>

>Anyway, has anyone else had a problem running 1.12?  I'm using a 
>TT030 4/16 (revision h motherboard).  I've tried running it on a 
>completely bare setup (no auto programs, no accessories, etc.) and
>get the same results.  
>

I have no problem running MiNT 1.12H1 nor the older ones. I run the 
hsmoda4 package before mint. I even used 38400 baud on modem2.


The only problem I got with mint is that i doesnt understand my 
graphics card. I found the problem and it is really the redirect.prg
program that comes with NVDI-ET4000. Physbase() is not returning the
correct screen address. It return the address that should have been if
I was using the original screen, but Logbase() and address 0x44E contains
the correct values. I have two "dirty" fixes for this. It is was easy to 
install a new Physbase routine in xbios.c that simply returned the value
of 0x44E.. Allso in mem.c (somewhere, I have not the sources available 
here right now) I changed:

     newbase = s_realloc(...);

to:

     newbase = 0;

X11R5 runs happily on my graphics card in monocrome resolutions 
(640x400 and 640x480, did not work with 1600x1016 :-() with these 
fixes installed.

Oh!: My graphics card is a Spectrum HC card. I use NVDI V2.01.


//Mats Loman -- <lom@tc.multi.se>

--

-----------------------------
Mats Loman <lom@tc.multi.se>
-----------------------------