As of today, after installing yesterday's binary snapshot from fd.o instead of CVS, I'm happy to report the bug is gone.
Actually, I thought I had updated everything that needed to be, but I had forgotten to copy the 2D (ati_drv.o, radeon.o) driver when I updated the dri (r200_dri.so) driver. That became apparent when I got a desktop lock, followed an hour later by a xscreensaver (2D, no GL) lock. So the problem was 2D and not 3D. Using the snapshots' excellent install.sh script is a breeze, btw. Thanks a lot, sorry for the false bug report. On Fri, 22 Oct 2004 16:13:38 +0200 Dieter N�tzel <[EMAIL PROTECTED]> wrote: > > The problem *was* : > > random lockups / freezes when playing UT (original one) > > Do you mean the old (first) one (No 2k3/2k4)? > I (we) havent't tested that for ages. I meant the original one, yes. > > those lockups happen within 30 seconds of starting > > any of these games. > > Using driconf and : > > * disabling entirely TCL (software) > > * setting frame throttling to Usleeps > > * VSync to OFF > > * Number of texture units to 2 > > actually helps alleviate the problem (I get lockups every 10 minutes or > > so). > > > > I then applied Mr Scheidegger 's patch at : > > Rolands...;-) heh, sorry :) > > https://freedesktop.org/bugzilla/show_bug.cgi?id=814 > > eventhough the problem was thought to be corrected in current CVS. > > > > Lo and behold, I know could play for about 10 minutes with both > > "Bypass MESA's pipeline with state-based code generation" and > > "Software interrupts" enabled ; better visuals, and faster. > > That said, UT locked up after 12 minutes (DM-Morpheus). > > I still haven't been able to pinpoint a common denominator for the > > crashes. > > It seems to me that the patch from Mr Scheidegger is still needed, > > because I already have dbg = 0x6; in r200_texstate.c ; but that's not > > enough, I fear :/ > > dbg = 0x06; :-) hum. it was late, I presume :) > Weird. I thought so. > > The whole PC is known to be reliable, it's able to do 4 hours of 3DMark > > 2001SE benchmark-looping in 1280x1024x32 mode, so I don't think > > there is a hardware problem. > > Sorry, but this say 'nothing' (on Linux). I couldn't test the card on Linux as it would lock any 3D app within the hour. Trying 3DMark was the only thing I could think of that would validate "a bit" the 3D hardware. > Try memtest-3.0 for min. 24 hours. ;-) Done last week, 2 days. But that actually tests nothing about the video card. > > relevant Hardware specs : > > PIII/933 coppermine > > asus tusl2-c / i815ep agp4x > > Try with > "AGPMode" "2" tested, even with "1", it still crashed > Maybe without fast write, too. > "AGPFastWrite" "0" tested, same result > /etc/X11/XF86Config > Or where ever it is. > > > 2x256MB PC133 Crucial, tested recently for defects with memtest > > How long? (see above) 2 days :) > > ATI FireGL 8800/128MB > > X log? not needed anymore, but will attach next time > > relevant software specs : > > linux kernel 2.6.9 (vanilla) > > quite up-to-date gentoo system with xorg-6.8.0-r1 > > Standard drm module (Gentoo)? drm from 2.6.9, drm from dri CVS, both experienced the behaviour. > -Dieter Thanks ! ------------------------------------------------------- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
