On 6/22/05, Richard Fish <[EMAIL PROTECTED]> wrote: > I re-arranged your message a bit to make the answers flow a bit better. > > Mark Knecht wrote: > > > Hey Richard, > > So this Pundit-R is driving the S-Video input on a TV set. > > > Oops, I forgot you were doing that...sorry. > > > Can you recommend where I'd look to find the right settings for > > driving a TV input with the radeon driver and 9100 IGP chipset? > > > Well, you already got farther than I did with the radeon driver and > S-Video output....I never got anything to display! Unfortuantely, I am > 6000 miles away from anything with an S-Video input (in France), and > won't be home for another month, so I cannot do any testing on my end to > see...
That's fine. Enjoy France. Nice time of year I suppose. > > As for a specific guide for getting a TV display to work with the radeon > driver, no I haven't found anything. But, you might want to try setting > the CRT2HSync and CRT2VRefresh settings with the radeon driver. I am > assuming that the S-Video output is being operated as the secondary > monitor, and the defaults for these are "undefined". See "man radeon" > for the gory details. I'll look at this. Actually I have BIOS on these machines set to use SVideo as the primary output, and it all works, in general, with the fglrx driver. > > > Linux agpgart interface v0.100 (c) Dave Jones > > [drm] Initialized drm 1.0.0 20040925 > > ACPI: PCI interrupt 0000:01:05.0[A] -> GSI 16 (level, low) -> IRQ 16 > > [drm] Initialized radeon 1.14.0 20050125 on minor 0: PCI device > > 1002:5834 (ATI Technologies Inc) > > <SNIP> > > [drm:radeon_cp_init] *ERROR* radeon_cp_init called without lock held > > [drm:drm_unlock] *ERROR* Process 6162 using kernel context 0 I haven't been back to check this again yet. I've taken two Pundit-R's and moved them to the new ~x86 ati-driver package. > > > Ouch, that doesn't look good. But maybe this can help: > > http://dri.freedesktop.org/wiki/DriTroubleshooting > > It seems you need to load the chipset-specific AGP driver in addition to > agpgart. > Yes, I found this about an hour or two ago. With that loaded I no longer get the agpgart messages and I'm getting better glxgears performance now so that's cool. However I get about 2000 FPS from my laptop with it's Radeon 9200: 0000:01:05.0 VGA compatible controller: ATI Technologies Inc M9+ 5C61 [Radeon Mobility 9200 (AGP)] (rev 01) but only about 200FPS from the Pundit-R's 9100: 0000:01:05.0 VGA compatible controller: ATI Technologies Inc Radeon 9100 IGP Both machines use the same basic chipset. The laptop: flash ~ $ /sbin/lspci 0000:00:00.0 Host bridge: ATI Technologies Inc Radeon 9100 IGP Host Bridge (rev 02) 0000:00:01.0 PCI bridge: ATI Technologies Inc Radeon 9100 IGP AGP Bridge The Pundit-R: [EMAIL PROTECTED] ~ $ /sbin/lspci 0000:00:00.0 Host bridge: ATI Technologies Inc Radeon 9100 IGP Host Bridge (rev 02) 0000:00:01.0 PCI bridge: ATI Technologies Inc Radeon 9100 IGP AGP Bridge The laptop uses 2.6.11-gentoo-r6 while the Pundit-R uses an older 2.6.11-gentoo-r3. I guess the next thing is to update the kernel. I have a very hard time beleiving the performance difference should be 10x!! Oh, one other thing I found. When I tell the Pundit-R to do this: opengl-update xorg-x11 then glxgears runs. However if I use opengl-update ati then the machine locks up hard wen running glxgears. This happens with both the 8.12.10 driver as well as the newer ~x86 driver. I haven't tried this on my laptop. Take care, Mark -- gentoo-user@gentoo.org mailing list
