Le jeudi 01 novembre 2007 à 21:22 -0400, Andy Fong a écrit : > With the open source announcement of ATI, sometimes we ask is OGP > still needed. I wonder what part of the spec are released : the 2D engine only or as well the 3D engine? > This thread just remind me why OGP is still worth the effort. You're right and this kind of problem should be widely communicated, in the newsletter first. But I mean that consumers should be aware of what happens behind the curtain: having a driver doesn't make all the story. /end > > They could open the full documentation for you to write a driver but > they hide a lot of > stuff in the bios which I don't foresee they will open up anytime soon > and instead they > might need to move more and more stuff (those need to be kept secret) > into the bios. > > From my experience (a couple years ago), all ATI's chips newer than > the Radeon 9000 > has some kind of mechanism to read the bios and lock up your machine > (or have all > kind of weird behavior) if the bios is tempered even the bios code is > not excuted at > all. Why? Because they sell cheaper versions of the card by disabling > pipelines in the bios. > The chip won't function properly without certain data in the bios. > They certainly don't want > anyone to buy the cheaper cards and unlock the pipelines with a hacked > VGA bios. Don't > know if this has changed since then. > > This is nothing new though. Remember back in the 80's, Intel just > disable the co-processor > on the 486DX and sell it as 486SX. AMD had similar deal I assume. > > With OGP, there will be no such thing. Every bit of the hardware will > be utilized to it's full > potential. You have the chief engineers here to answer your questions > so you can optimize > the hack out of your driver. You definitely don't need the bios for it > to work correctly (weather > as a second head or on other platforms). The bios will be fully > opened, so if you want a custom > splash screen, feel free to hack it. The list go on ..... > > Andy > > > > On 10/31/07, Sebastien Bourdeauducq <[EMAIL PROTECTED]> > wrote: > Hi, > > I've heard that some non-x86 workstations also faced this kind > of problem, and > had to implement an x86 emulator to be able to run the > firmware some PCI > cards shipped with. > > If you have enough time and memory on your board, maybe you > could do something > with QEmu. > > Or if you have access to the precise specs of the VGA card, > write a dedicated > driver... but this breaks compatibility. > > Sebastien > > > On Wednesday 31 October 2007 16:18:48 Rupesh Kumar wrote: > > Hi All, > > > > I got ATI Radeon Card which is working on PC as Secondary > VGA display > > adapter. > > When i try this card on my ARM system it is failing to > detect the videoram > > and failing with the following error. > > radeonfb (0000:00:02.0): Found 0k of SDRAM 64 bits wide > videoram > > adeonfb (0000:00:02.0): cannot map FB > > radeonfb: probe of 0000:00:02.0 failed with error -5 > > > > After some time on google i have found that this error can > be solved by > > running vbetool (which runs the video BIOS). > > This is completely killing my idea of running it on ARM > Board. > > > > My goal is to make a VGA Card work on ARM System which > doesnt support > > legacy VGA Address range. > > Can anyone please clarify me whether this is feasible or > not. > > Anybody have ever tried this kind of setup. > > > > Please help me. > > > > Regards > > Rupesh > > > > On 10/26/07, Rupesh Kumar <[EMAIL PROTECTED]> wrote: > > > Thanks for your prompt reply Paul > > > > > > > My goal is to make this card work on a ARM System. My > System doesn't > > > > > > > > > support VGA Legacy address decoding. So i am expecting > that if a card > > > > > > > > works > > > > > > > > > as secondary VGA Adapter then it can be portable to > ARM System. Am i > > > > > correct ?? > > > > > > > > Probably, but there may be other issues. > > > > > > What kind of issues ? Is it VGA BIOS,or driver issues. > > > > > > I'm surprised the your Cirrus Logic card is failing, > they're one of the > > > > > > > better > > > > documented card series. Maybe the code has just > bitrotted. > > > > > > I checked this card in Windows . It works if it is taken > as primary > > > display. > > > if it is taken as secondary then the driver initialization > is failing. > > > > > > May be this card is old and may not work as secondary > VGA. Just kidding > > > !!!!! (Showing VGA BIOS 2005). > > > > > > I have checked linux cirrusfb driver and it is directly > accessing the > > > Legacy VGA Range. (drivers/video/cirrusfb.c) > > > This implies to me that this card can never be supported > as Secondary VGA > > > with this linux driver.(I dont know how X Server act on > it) > > > > > > Please correct me. > > > > > > Regards > > > Rupesh > > > _______________________________________________ > Open-graphics mailing list > [email protected] > http://lists.duskglow.com/mailman/listinfo/open-graphics > List service provided by Duskglow Consulting, LLC > (www.duskglow.com) > > _______________________________________________ > Open-graphics mailing list > [email protected] > http://lists.duskglow.com/mailman/listinfo/open-graphics > List service provided by Duskglow Consulting, LLC (www.duskglow.com)
_______________________________________________ Open-graphics mailing list [email protected] http://lists.duskglow.com/mailman/listinfo/open-graphics List service provided by Duskglow Consulting, LLC (www.duskglow.com)
