On Sun, Jun 15, 2003 at 07:47:35PM +0100, José Fonseca wrote: > Something fishy is going on here. My debug statements never appear in > your log and I've just retested on my test box with the latest CVS and > everything works as expected with the radeon driver: [...] > Martin, are you're sure you're using the right (i.e. the latest) > versions of the DRM and 2D drivers? I can't see any other explanation.
I am pretty shure they are. They all stem from the same DRI build (they all carry the appropriate date): -r--r--r-- 1 root root 1342562 Jun 13 15:38 /usr/X11R6/lib/modules/drivers/ati_drv.o -r--r--r-- 1 root root 3821349 Jun 13 15:38 /usr/X11R6/lib/modules/drivers/atimisc_drv.o -r--r--r-- 1 root root 13884724 Jun 13 15:36 /usr/X11R6/lib/modules/dri/radeon_dri.so -r--r--r-- 1 root root 1453282 Jun 13 15:38 /usr/X11R6/lib/modules/drivers/radeon_drv.o lrwxrwxrwx 1 root root 7 Jun 13 15:38 /usr/X11R6/bin/X -> XFree86 -rws--x--x 1 root root 18661010 Jun 13 15:38 /usr/X11R6/bin/XFree86 The procedure I use to build the stuff has not differed for quite a few months, absolutely nothing has changed here. I'm getting quite functional binaries when I build the 'radeon' kernel module from sources _before_ the large APG related patch of June 4th and leave _everything_ else untouched. Just to prevent misunderstanding: I have a distribution provided /usr/X11R6/ tree. I take a copy of the current CVS and do a 'make world'. Afterwards I do a 'make install' as root which copies the CVS build _over_ my previous X11R6 tree. Now I generate a patch from the _current_ CVS trees kernel driver sources (.../os-support/[shared,linux]/drm/kernel/*.[h,c] against the sources in /usr/src/linux/drivers/char/drm/*.[h,c] I apply this patch to my kernel tree from which I rebuild and install the kernel modules. This gives me software OpenGL since 4th June. Now I revert the .../os-support/ directory of my CVS tree to a date before June 4th, I generate an appropriate patch against the kernel tree and rebuild the kernel modules. I stop the X server, unload the 'radeon' module by leaving the 'agpgart' module and load the new 'radeon' module from sources before June 4th. Now I have hardware-accelerated OpenGL after restarting the X server. It does not make _any_ difference if I reload _only_ the 'radeon' module, _all_ modules or if I reboot the whole machine. So at least from my point of view it solely depends on the 'radeon' module. I'll redo the debug output on monday - just to make shure I did not make any mistake. I'm doing this with an OEM Radeon9100. Unfortunately I gave away my Radeon7500 so I don't have any different test case, Martin. -- Unix _IS_ user friendly - it's just selective about who its friends are ! -------------------------------------------------------------------------- ------------------------------------------------------- This SF.NET email is sponsored by: eBay Great deals on office technology -- on eBay now! Click here: http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5 _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
