[Dri-devel] DRM+ATI-Radeon-Mobility
The problem identified by DRM+ATI-Radeon-Mobility appeared already with kernel 2.4.23-pre8 Freezing takes place when function, gl_init_all_x86_transform_asm called from /usr/X11R6/lib/modules/dri/radeon_dri.so is to be executed (see eg., gdb tunnel ) Regards Wlodek --- 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
[Dri-devel] DRM+ATI-Radeon-Mobility
On Mon, Feb 09, 2004 Christian Schoenebeck wrote: Hi! I have the same problem as you mentioned on DRI-Devel ML (freeze of X with Radeon Mobility whenever DRI access occurs). Has there been any progress in this field meanwhile? Did anybody solve the problem? Best regards Christian Schoenebeck Unfortunately, the problem persists. I 'veseen few advices concerning upgrading it: making upgrade to Xfree86-4.3.0 or 4.4.0 or compiling DRI tree from CVS The DRI from CVS requires Mesa source and seems to cause problems (see message from Alan Hourihane of Feb 4th), I cannot consider upgrading my XFree86-4.1.0-50 (from RedHat) as this would be like changing system in case of my setup. I am persuing the dri-devel list and I have impression this may be something not only in DRM but also in other place in the kernel tree starting from about 2.4.20 (many users report similar effects). Now, I boot to 2.4.18 after switch on, start an DRI using application that writes sth. to static cache memory (i.e. probably so) and then I do warm restart to 2.4.23 and DRI works with the software provided with the kernel. This is half-solution of course (more details is in my previous messages in dri-devel list of January ). Regards Wlodek --i19FC5sW030023.1076339525/atol.icm.edu.pl-- --- 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
Re: [Dri-devel] DRM+ATI-Radeon-Mobility
On Thu, 2004-01-15 at 03:10, Wlodzimierz Kozlowski wrote: I observe the following problem concerning, most probably, radeon.o module compiled from the kernel linux-2.4.23 source (generally later than 2.4.18): An effort of launching a graphic application using DRI results in total freezing of the keyboard, mouse, screen so that only switch off can be done (glxinfo shows that DRI , AGP and all the rest are enabled). Can you post the output of LIBGL_DEBUG=verbose glxinfo as well as any relevant parts of the X server log and kernel output? However , I have observed that launching the graphic application using DRI with XFree86-4.1.0 under kernel 2.4.18 (Linux kernel 2.4.18 from http://www.kernel.org) and subsequent warm restart to 2.4.23 removes this unrequired effect as long as the machine is switched on (subsequently, a number of WARM restarts can be done and this will not restore the freezing effect). I have found this has most probably nothing with APM or using the EXT3 and occurs with XFree86 subversions 4.1.0, 4.2.0 and apparently does not depend on the XFree86 subversion number. Have you tried 4.3 or newer though? Assignment of RAM intervals appears to be the only difference between log files [...] What RAM intervals? (others report the problem preserves also with Linux kernel 2.6.0, e.g. https://sourceforge.net/mailarchive/forum.php?thread_id=3713780forum_id=6511 That URL doesn't work for me, but maybe that's just sf.net... All in all, it sounds to me like newer DRMs fail to initialise something they should, at least with older DRI userland. -- Earthling Michel Dnzer | Debian (powerpc), X and DRI developer Software libre enthusiast| http://svcs.affero.net/rm.php?r=daenzer --- 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
[Dri-devel] DRM+ATI-Radeon-Mobility
Re: Problem with kernel modules drm with XFree86-4.1.0, linux-2.4.23 for ATI Radeon Mobility Setup: Laptop Clevo 4600 (vel Eurocom 4600) with Pentium III Mobility 1.2. GHz CPU: L1 I cache: 16K, L1 D cache: 16K CPU: L2 cache: 512K ATI Radeon Mobility (Integrated) AGP 4x M6-P (indicated also as R100/7000, or LY) 32 MB VideoRAM: 64bit DDR SGRAM (i830MP) 512 MB DRAM kernel Linux-2.4.23 (from, www.kernel.org) XFree86-4.1.0-50 (from RedHat) Mesa-3.4.2-10 (from RedHat for RH-7.2 and use with XFree86-4.1.0) distribution Linux RedHat-7.2 (with erratic supplements) I observe the following problem concerning, most probably, radeon.o module compiled from the kernel linux-2.4.23 source (generally later than 2.4.18): An effort of launching a graphic application using DRI results in total freezing of the keyboard, mouse, screen so that only switch off can be done (glxinfo shows that DRI , AGP and all the rest are enabled). Kernel and all modules have been compiled using the same version of the compiler, gcc2.96-112.7.2 (from RedHat-7.2 Errata) However , I have observed that launching the graphic application using DRI with XFree86-4.1.0 under kernel 2.4.18 (Linux kernel 2.4.18 from http://www.kernel.org) and subsequent warm restart to 2.4.23 removes this unrequired effect as long as the machine is switched on (subsequently, a number of WARM restarts can be done and this will not restore the freezing effect). I have found this has most probably nothing with APM or using the EXT3 and occurs with XFree86 subversions 4.1.0, 4.2.0 and apparently does not depend on the XFree86 subversion number. Assignment of RAM intervals appears to be the only difference between log files with X starting under linux-2.4.23 booted directly after switching the machine on and the log files with X under the 2.4.23 warm restarted from 2.4.18 under which a graphic application using DRI has been launched. Note, starting X under 2.4.18 without launching such application will not remove the effect of freezing after the warm restart to 2.4.23. (examples of the application: gears, tunnel, tunnel2 ) It seems that cache RAM may keep some information copied there while starting aplication using DRI under X with linux-2.4.18 and this enables start of such applications after warm restart to 2.4.23 (does it have any sense ??). I am sending this message in hope it will appear helpful in solving the known problem of using DRI with ATI RADEON for kernels later than 2.4.18 (others report the problem preserves also with Linux kernel 2.6.0, e.g. https://sourceforge.net/mailarchive/forum.php?thread_id=3713780forum_id=6511 From: Rafal Gala [EMAIL PROTECTED] Mobility Radeon + DRI 2004-01-08 05:39 ). Kind regards Wlodek --- This SF.net email is sponsored by: Perforce Software. Perforce is the Fast Software Configuration Management System offering advanced branching capabilities and atomic changes on 50+ platforms. Free Eval! http://www.perforce.com/perforce/loadprog.html -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
