--- "Robert S. Kerr" <[EMAIL PROTECTED]> wrote: > Alex, > > I'm sending this to you directly because I have been unable to get > this > message sent to the dri-developer's list. For whatever reason, the > dri-list doesn't appear to like my e-mail address. I knew you would > have some interest in this issue, but I think it merits posting to > the > list. Perhaps I'll be able to figure out how to get it up there > soon. > > Rob Kerr > > <begin message> > Based on Alex Deucher's recent request to test the Savage driver he > and > Felix have been working on, I recently pulled the savage-2-0-0-branch > > and compiled the driver for testing. > > My configuration is Fedora Core 1, AMD Athlon 2200, 512 MB, > ProsavageDDR > in a Shuttle SK41G. > > My results have been good so far. I was able to get the driver up > and > running on a monitor quickly. X seems stable and I haven't had any > crashes so far.
Have you tried 3D? how has that worked out? what about modes? 1280x1024? 800x600? etc. any problems with Xv? > > What isn't working, however is TV out. Any time I start X with a TV > connected, I get an error saying that displays have been found, but > no > valid modes for the displays are available. I did some looking > around > and I know what the problem is, I'm just not sure of the correct fix > for > it. > > In savage_vbe.c, there is a method 'SavageQueryVBEInfo' that queries > for > supported displays and sets up some TV Out information. Noteably, > the > TvSizeX & TvSizeY are set to 0. For all chips other than S3_TWISTER, > > they stay at TvSizeX =0, & TvSizeY = 0. For S3_TWISTER chips, they > are > set to a size based on the external TV encoder. In my case they stay > at > 0 because I have an S3_PROSAVAGEDDR chip. > > Later on, when the driver is trying to find valid modes in > 'SavageValidMode' (savage_driver.c), since TvSizeX & TvSizeY are 0, > all > modes are bad. > > My fix would be to add a check for my chip in the > 'SavageQueryVBEInfo' > method, (in fact I did this last night and was able to get TV out > working at least a little bit), but I don't know if that is really > the > right answer. Seems like that check is in there because when the > driver > was written, S3_TWISTER was the only chip that supported TV out. I'm > That's probably the right answer. the DDR chips are basically prosavage/twister chips that support DDR memory. I think the only difference between prosavage and twister was chipset support. Prosavage was for Intel and twister was for AMD. the cores are pretty much the same. > certain that there must be others and so I wanted to know if anyone > knows which of the savage chips support TV out so we can set up the > right 'if' test. I don't have a prosavage-based board to test with, but I suspect all prosavage/twister/ddr chips should work the same for tv-out. > > So, does anyone know which savage chips support TV out? Just about all of them do to a certain degree. > Do they all use the external encoders like the S3_TWISTER does? The laptop chips (mx/ix/supersavage) included a built in tv-encoder. they also have two crtcs and can do dualhead. the prosavage/twister chips only have one crtc and no built in tv-encoder, hence to have to use an external one to support tv-out. > Is adding additional chips into the 'if' test the right answer here? > I think so. I'd say add twister/prosavage/prosavageddr. Attach a patch if you like. I'll probably mess with the mode/Xv code at some point this week (depending on time). I'll take a look at you patch then if you provide one, otherwise I'll just take a look at the code in question. felix has a ddr and perhaps he can test. Alex > Thoughts? > > Rob Kerr. > > BTW, I've been following the work of Alex and Felix for these last > few > weeks excitedly. Kudos to them for getting the Savage driver to the > state that it is in such short order. > <end message> __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free web site building tool. Try it! http://webhosting.yahoo.com/ps/sb/ ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn -- _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
