Re: Re: Matrox Millenium G450 AGP

Petr Vandrovec <[email protected]>
Newsgroups gmane.linux.fbdev.user
Message-ID <[email protected]>
On Sat, Apr 03, 2004 at 03:41:18PM +0200, Sven Luther wrote:
> On Sat, Apr 03, 2004 at 02:43:02PM +0200, Johan Gullden wrote:
> > On Sat, 3 Apr 2004 13:51:25 +0200
> > Johan Gullden <[email protected]> wrote:
> > 
> > > On Sat, 3 Apr 2004 12:49:42 +0200
> > > Sven Luther <[email protected]> wrote:
> > > 
> > > > On Sat, Apr 03, 2004 at 12:32:54PM +0200, Johan Gullden wrote:
> > > > > (I am not subscribed to the list, so please CC your replies to me.)
> > > > > 
> > > > > Hello,
> > > > > 
> > > > > I am having trouble getting my Matrox G450 to work on my pegasos2,
> > > > > after and while booting the console gets weird. It doubles the output with a space of 4 characters, so it's really hard to use it.
> > > > > 
> > > > > The kernel I am using is 2.6.4 (gentoo/ppc-development-sources), I have also tried the 2.4 kernel and getting the same problems.
> > > > > 
> > > > > my settings from #Graphics support
> > > > > 
> > > > > CONFIG_FB=y
> > > > > CONFIG_FB_MATROX=y
> > > > > CONFIG_FB_MATROX_G450=y
> > > > > CONFIG_FB_MATROX_G100=y
> > > > > CONFIG_FB_MATROX_I2C=y
> > > > > 
> > > > > I have also tried to use diffrent boot parameters, but with no luck. The output is just a screen filled with colorful pixels.
> > > > 
> > > > Johan, could you provide a bit more detail about the screen. You told me
> > > > that you saw like 4 different screens or soemthing such. This kind of
> > > > information, or even a digicam picture of the screen, is essential in
> > > > detecting the problem, and hopefully fixing it.
> > > > 
> > > 
> > > Ok, there seems to be some missunderstanding, I will take a picture of the console. So you all can se waht I mean.
> > > 
> > 
> > The pictures are really bad. But you should see what I mean.
> > Pictures are visible at: http://hem.passagen.se/stimp
> 
> Nope, i don't understand what the problem is from those picutres, they
> seem rather normal to me. I may be wrong though.

They are quite normal when using G100 in 2.4.x with weird memory chips. But
it should not happen with 2.6.x which reads setup from ROM.

So I have couple of questions:

(1) What is pegasos? Is it i386/x86_64/alpha architecture which executes/emulates ROM BIOS,
    or is Matrox initialized only by matroxfb? Can you try 'video=matroxfb:init' or 'video=matroxfb:noinit'
    options (default value is autodetect whether card was initialized or not; using noinit on
    system which did not initialize matrox hardware may lock up PCI bus, so have reset button ready).

(2) If 'video=matroxfb:init' alone will not help, you can try playing with g450_memory_init() function.
    It completely relies on PINS data, so if pins data are incorrect (like on GXT130P IBM distributed),
    driver will not work without manual tweaking default_pins5() with values your hardware needs.

(2a) Do you use card from Matrox, or is it some 3rd party product - either onboard (from f.e. HP) or
     bundled with their own firmware (f.e. IBM)? Onboard devices are only in x86 systems, as in onboard
     solutions there is usually no EEPROM connected to the graphics chip, and so it is not possible to 
     find PINS structure.

(3) You can try 'video=matroxfb:sdram'. But I do not think that g450/g550 portion of driver honors this
    option - as I said, it completely relies on on pins...

(4) Otherwise I have no further idea. Problem is definitely caused by wrong memory type, mctlwtst* or 
    memmisc/memrdbk value.

								Best regards,
									Petr Vandrovec
									[email protected]



-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.