Re: [Dri-devel] Strange slowdown with DRI drivers after reboot onRadeon VE
On Fre, 2002-12-20 at 11:00, Nick Kurshev wrote: On 19 Dec 2002 03:03:09 +0100 you wrote: On Son, 2002-12-15 at 09:25, Nick Kurshev wrote: It seems that DRI-CVS doesn't configure some registers or something similar. Means that before reboot - DRI works with hardware which was configured by Xfree86 drivers but after reboot it is not able to configure video card properly. I've attached 3 dumps of Radeon's registers. These dump were generated with using of slightly modified by me 'gfxdump' utility from old-gatos CVS (at linuxvideo.org). Could please someone to investigate this problem? I investigated the three dumps with meld and none of the registers that differ seem performance critical to me. except 'rbbm_cmdfifo_stat'! ;) As the name suggests, that's a read only status register. Someone asked a similar question on the Xpert list lately, and it turned out to be a BIOS feature to regulate system performance, could it be something like that? First - my BIOS may configure of AGP speed and AGP aperture size only! Second - if it's problem of the BIOS then why it doesn't affect XFree86? Good point, but then that could just be coincidence, one never knows with things like that. Have you compared the AGP setup yet? One thing that has changed is that AGP Fast Writes are disabled by default. -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast --- This SF.NET email is sponsored by: The Best Geek Holiday Gifts! Time is running out! Thinkgeek.com has the coolest gifts for your favorite geek. Let your fingers do the typing. Visit Now. T H I N K G E E K . C O Mhttp://www.thinkgeek.com/sf/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] R200 notes issues
There are sporadic rendering bugs in FlightGear, however. Every ~40 frames or so, I'll see a large triangle or two flash on the screen. Like these ones ? http://document.ihg.uni-duisburg.de/bitmap/Radeon/Mesa-4.0.4_3.png Did you know that 'Solace' - recently mentioned on this list - also shows this sort of artifacts ? http://document.ihg.uni-duisburg.de/bitmap/Radeon/Solace-4.x.png Maybe this program is not that complex and could server as a test case for DRI/Mesa-4.x, Martin. -- Unix _IS_ user friendly - it's just selective about who its friends are ! -- --- This SF.NET email is sponsored by: The Best Geek Holiday Gifts! Time is running out! Thinkgeek.com has the coolest gifts for your favorite geek. Let your fingers do the typing. Visit Now. T H I N K G E E K . C O Mhttp://www.thinkgeek.com/sf/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] Radeon PCI patch
C. Brewer wrote: I would like to beg for the inclusion of #define PCIGART_ENABLED in radeon_dri.c, radeon_cp.c and the applied ring buffer tweak to the XFree86 CVS. The pci changes shouldnt affect any other system and do allow rather stable 3D accelerated environment, with a properly configured XF86Config-4 file. I have been running these as such for a while,and I proposed to get them in through regular XFree86, and was told they were forwarded to you guys. That ws at the end of 4.1.0. I am submitting directly to you now in hopes of faster response. The attached ring buffer tweak is not entirely my own, the bulk of it being released publically on the Internet by [EMAIL PROTECTED], whom I have been unable to contact for some time. All other minor tweaking is my own,and the whole patch should be GPL. Thank you for your consideration. Chuck, I just want to point out the XFree86 only accepts BSD style licensing, not GPL. The code from the DRI project get's merged into XFree86 (we're a subproject, if you will), so we need to follow that licensing as well. See http://www.xfree86.org/legal/licence.html for more details. Regards, Jens -- /\ Jens Owen/ \/\ _ [EMAIL PROTECTED] /\ \ \ Steamboat Springs, Colorado --- This SF.NET email is sponsored by: The Best Geek Holiday Gifts! Time is running out! Thinkgeek.com has the coolest gifts for your favorite geek. Let your fingers do the typing. Visit Now. T H I N K G E E K . C O Mhttp://www.thinkgeek.com/sf/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] S3TC
Am Freitag, 20. Dezember 2002 14:38 schrieb Alexander Stohr: [-] And thats why all the world on the stock marked emphasizes on their patents protfolio, its money and power that those patenst sometimes do represent. Of course there is much vapourware, but such key technologies like S3TC that anybody wants to use are valueable to its owner. Sorry, Alexander. ...but such key technologies like S3TC... What? The algorithm is a very trivial one. -Dieter --- This SF.NET email is sponsored by: The Best Geek Holiday Gifts! Time is running out! Thinkgeek.com has the coolest gifts for your favorite geek. Let your fingers do the typing. Visit Now. T H I N K G E E K . C O Mhttp://www.thinkgeek.com/sf/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] S3TC
On Fri, 20 Dec 2002, Dieter [iso-8859-1] Nützel wrote: Am Freitag, 20. Dezember 2002 14:38 schrieb Alexander Stohr: [-] And thats why all the world on the stock marked emphasizes on their patents protfolio, its money and power that those patenst sometimes do represent. Of course there is much vapourware, but such key technologies like S3TC that anybody wants to use are valueable to its owner. Sorry, Alexander. ...but such key technologies like S3TC... What? The algorithm is a very trivial one. That doesn't make it any less of a key technology. As Alexander pointed out, it's valuable to it's owner and it is, in fact, being used more and more in games and implemented in hardware. Key technology doesn't necessarily mean complex. Adam --- This SF.NET email is sponsored by: The Best Geek Holiday Gifts! Time is running out! Thinkgeek.com has the coolest gifts for your favorite geek. Let your fingers do the typing. Visit Now. T H I N K G E E K . C O Mhttp://www.thinkgeek.com/sf/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] S3TC
On Fri, Dec 20, 2002 at 03:29:35PM +0100, Dieter Nützel wrote: Am Freitag, 20. Dezember 2002 14:38 schrieb Alexander Stohr: [-] And thats why all the world on the stock marked emphasizes on their patents protfolio, its money and power that those patenst sometimes do represent. Of course there is much vapourware, but such key technologies like S3TC that anybody wants to use are valueable to its owner. Sorry, Alexander. ...but such key technologies like S3TC... s/S3TC/public-key encryption/ What? The algorithm is a very trivial one. Doesn't matter. -- http://trikuare.cx --- This SF.NET email is sponsored by: The Best Geek Holiday Gifts! Time is running out! Thinkgeek.com has the coolest gifts for your favorite geek. Let your fingers do the typing. Visit Now. T H I N K G E E K . C O Mhttp://www.thinkgeek.com/sf/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] R200 notes issues
Am Freitag, 20. Dezember 2002 19:48 schrieb magenta: On Fri, Dec 20, 2002 at 12:47:34PM +0100, Martin Spott wrote: There are sporadic rendering bugs in FlightGear, however. Every ~40 frames or so, I'll see a large triangle or two flash on the screen. Like these ones ? http://document.ihg.uni-duisburg.de/bitmap/Radeon/Mesa-4.0.4_3.png Did you know that 'Solace' - recently mentioned on this list - also shows this sort of artifacts ? And I can guarantee it's not a bug in Solace. ;) http://document.ihg.uni-duisburg.de/bitmap/Radeon/Solace-4.x.png Maybe this program is not that complex and could server as a test case for DRI/Mesa-4.x, Wow, that's a really cool failure mode... What's your GL_draw_method (in ~/.Solace/registry) set to? Try changing it to 1 or 0... also, I notice a problem with the outlines on the various objects which have them (the group of torii and the bubble-shaded thing on the right). The easy workaround (in Solace anyway) is to change GL_line_method to 0, which switches it to immediate mode (it still records a displaylist though). BTW, for an explanation of the various settings (so that people really could use Solace as a driver stress test), type d_set from the controlling terminal (and you can also use this command to change the settings on-the-fly). I would try it on my dual Athlon system but I haven't have a GCC 3.2 system, yet. So the libg++ reference fail...;-( Any chance to get a more compatible version? Thanks, Dieter --- This SF.NET email is sponsored by: The Best Geek Holiday Gifts! Time is running out! Thinkgeek.com has the coolest gifts for your favorite geek. Let your fingers do the typing. Visit Now. T H I N K G E E K . C O Mhttp://www.thinkgeek.com/sf/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
