Bug#374986: tested patch, still crashes
Tried with radeon-force-some-clocks.diff, twice. One of the times I got much further (I could actualy play the game, rather than only reach the menus), but it also crashed after a very short time (1/2 minutes). The second time it just crashed in the menus as usual. The X server log is also different. No loop this time. I grabbed a copy of Xorg.log before running the game (xmoto), so that I could diff it against the complete log after rebooting the system. The following lines were printed after xmoto started: +(**) RADEON(0): RADEONDisplayPowerManagementSet(0,0x0) +(**) RADEON(0): EngineRestore (32/32) +(**) RADEON(0): RADEONSwitchMode() !n(**) RADEON(0): RADEONModeInit() +800x60056.30 800 832 896 1048 600 601 604 631 (24,32) +H +V +800x60056.30 800 832 896 1048 600 601 604 631 (24,32) +H +V +(**) RADEON(0): Pitch = 8388736 bytes (virtualX = 1024, displayWidth = 1024) +(**) RADEON(0): dc=5630, of=22520, fd=100, pd=4 +(**) RADEON(0): RADEONInit returns 0x6fe0b0 +(**) RADEON(0): RADEONRestoreMode() +(**) RADEON(0): RADEONRestoreMode(0x6fe0b0) +(**) RADEON(0): RADEONRestoreMemMapRegisters() : +(**) RADEON(0): MC_FB_LOCATION : 0xd7ffd000 +(**) RADEON(0): MC_AGP_LOCATION : 0xffc0 +(**) RADEON(0): Updating display base addresses... +(**) RADEON(0): Memory map updated. +(**) RADEON(0): Programming CRTC1, offset: 0x +(**) RADEON(0): Wrote: 0x000c 0x00020064 0x (0xa700) +(**) RADEON(0): Wrote: rd=12, fd=100, pd=2 +(**) RADEON(0): GRPH_BUFFER_CNTL from 30004c4c to 20157c7c +(**) RADEON(0): EngineRestore (32/32) This time I also saved stdout/stderr output from xmoto itself: Unknown device ID 5B60, please report. Assuming plain R300. libGL warning: 3D driver claims to not support visual 0x23 libGL warning: 3D driver claims to not support visual 0x24 libGL warning: 3D driver claims to not support visual 0x25 libGL warning: 3D driver claims to not support visual 0x26 libGL warning: 3D driver claims to not support visual 0x27 libGL warning: 3D driver claims to not support visual 0x28 libGL warning: 3D driver claims to not support visual 0x29 libGL warning: 3D driver claims to not support visual 0x2a libGL warning: 3D driver claims to not support visual 0x2b libGL warning: 3D driver claims to not support visual 0x2c libGL warning: 3D driver claims to not support visual 0x2d libGL warning: 3D driver claims to not support visual 0x2e libGL warning: 3D driver claims to not support visual 0x2f libGL warning: 3D driver claims to not support visual 0x30 libGL warning: 3D driver claims to not support visual 0x31 libGL warning: 3D driver claims to not support visual 0x32 *WARN_ONCE* File r300_state.c function r300Enable line 456 TODO - double side stencil ! *** No ctx->FragmentProgram._Current!! radeonWaitIrq: drmRadeonIrqWait: -16 Sidenote: My card is a Radeon X300 SE. Where should I report that ID 5b60 for addition ? I couldn't find this string in the ati driver. -- Robert Millan My spam trap is <[EMAIL PROTECTED]>. Note: this address is only intended for spam harvesters. Writing to it will get you added to my black list. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#374986: tested patch, still crashes
On Tue, 2006-07-04 at 20:12 +0200, Robert Millan wrote: > Tried with radeon-force-some-clocks.diff, twice. One of the times I got much > further (I could actualy play the game, rather than only reach the menus), but > it also crashed after a very short time (1/2 minutes). The second time it > just > crashed in the menus as usual. You're probably hitting a different issue then. Please try Mesa CVS or at least libg11-mesa-dri from experimental. -- Earthling Michel Dänzer | http://tungstengraphics.com Libre software enthusiast | Debian, X and DRI developer
Bug#374986: tested patch, still crashes
On Mon, Jul 17, 2006 at 02:06:21PM +0200, Michel Dänzer wrote: > On Tue, 2006-07-04 at 20:12 +0200, Robert Millan wrote: > > Tried with radeon-force-some-clocks.diff, twice. One of the times I got > > much > > further (I could actualy play the game, rather than only reach the menus), > > but > > it also crashed after a very short time (1/2 minutes). The second time it > > just > > crashed in the menus as usual. > > You're probably hitting a different issue then. Please try Mesa CVS or > at least libg11-mesa-dri from experimental. With libgl1-mesa-dri from experimental (6.5.0.cvs.20060524-1), I get it to work for quite longer (at least, this is my subjective impression), but it still crashes at some point. I couldn't capture any logs as they weren't synced to disk. With Mesa CVS, Xorg.log claims that direct rendering is enabled, but glxinfo says otherwise. xmoto seems to agree with that (no crash, and gameplay horribly slow). My Xorg.log for this attempt is attached. -- Robert Millan My spam trap is [EMAIL PROTECTED] Note: this address is only intended for spam harvesters. Writing to it will get you added to my black list. Xorg.0.log.gz Description: Binary data
Bug#374986: tested patch, still crashes
I forgot to mention that, in order to get Mesa CVS building, I disabled r200 driver that required a version of libdrm not yet present in debian. I suppose this shouldn't be relevant (my card is r300). -- Robert Millan My spam trap is [EMAIL PROTECTED] Note: this address is only intended for spam harvesters. Writing to it will get you added to my black list. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#374986: tested patch, still crashes
On Mon, 2006-07-17 at 19:51 +0200, Robert Millan wrote: > > With Mesa CVS, Xorg.log claims that direct rendering is enabled, but glxinfo > says otherwise. LIBGL_DEBUG=verbose glxinfo should give a hint why that is. -- Earthling Michel Dänzer | http://tungstengraphics.com Libre software enthusiast | Debian, X and DRI developer
Bug#374986: tested patch, still crashes
On Mon, Jul 17, 2006 at 09:25:21PM +0200, Michel Dänzer wrote: > On Mon, 2006-07-17 at 19:51 +0200, Robert Millan wrote: > > > > With Mesa CVS, Xorg.log claims that direct rendering is enabled, but glxinfo > > says otherwise. > > LIBGL_DEBUG=verbose glxinfo > > should give a hint why that is. Ok, it was just broken build of mesa. I've rebuilt it properly, this time with libdrm 2.0.2. Now I hit a completely different bug: after a bit of gameplay, system just hangs (no network activity, SysRq key doesn't work, etc). Nothing relevant was printed in the logs (either in xmoto stdout/stderr or in Xorg.log). I'm afraid I can't provide any useful debug information at this point. I suppose atm r300 support is too inmature for me to play with. -- Robert Millan My spam trap is [EMAIL PROTECTED] Note: this address is only intended for spam harvesters. Writing to it will get you added to my black list. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#374986: tested patch, still crashes
Hi Eduard, On Thu, Jul 20, 2006 at 10:12:34PM +0200, Eduard Bloch wrote: > > > > You're probably hitting a different issue then. Please try Mesa CVS or > > at least libg11-mesa-dri from experimental. > > There is no libgl1-mesa-dri in experimental, AFAICS. There is, but: - Michel made a typo (libg11-mesa-dri). Did you copy-paste it? - It's not available for amd64 (although it compiled fine here). > And about Mesa > CVS... I am not sure how to install it in a way that X modules buy for > sure, there seem to be a funky coupling between X and Mesa. Just fetch the mesa source from experimental and run "cvs update". You'll need to update libdrm to 2.0.2 first (not in debian yet). -- Robert Millan My spam trap is [EMAIL PROTECTED] Note: this address is only intended for spam harvesters. Writing to it will get you added to my black list. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#374986: tested patch, still crashes
#include * Michel Dänzer [Mon, Jul 17 2006, 02:06:21PM]: > On Tue, 2006-07-04 at 20:12 +0200, Robert Millan wrote: > > Tried with radeon-force-some-clocks.diff, twice. One of the times I got > > much > > further (I could actualy play the game, rather than only reach the menus), > > but > > it also crashed after a very short time (1/2 minutes). The second time it > > just > > crashed in the menus as usual. > > You're probably hitting a different issue then. Please try Mesa CVS or > at least libg11-mesa-dri from experimental. There is no libgl1-mesa-dri in experimental, AFAICS. And about Mesa CVS... I am not sure how to install it in a way that X modules buy for sure, there seem to be a funky coupling between X and Mesa. Eduard.
Bug#374986: tested patch, still crashes
#include * Robert Millan [Thu, Jul 20 2006, 10:33:17PM]: > > Hi Eduard, > > On Thu, Jul 20, 2006 at 10:12:34PM +0200, Eduard Bloch wrote: > > > > > > You're probably hitting a different issue then. Please try Mesa CVS or > > > at least libg11-mesa-dri from experimental. > > > > There is no libgl1-mesa-dri in experimental, AFAICS. > > There is, but: > > - Michel made a typo (libg11-mesa-dri). Did you copy-paste it? First, no. And my screen font is fine enough to detect the typo immediately. > - It's not available for amd64 (although it compiled fine here). Second... yes, it compiled fine but does not solve the problem. It still crashes. > > And about Mesa > > CVS... I am not sure how to install it in a way that X modules buy for > > sure, there seem to be a funky coupling between X and Mesa. > > Just fetch the mesa source from experimental and run "cvs update". You'll > need > to update libdrm to 2.0.2 first (not in debian yet). updated libdrm, but mesa does not build: In file included from savagerender.c:115: ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h: In function ‘savage_render_poly_el ts’: ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h:989: error: ‘Light’ undeclared (fir st use in this function) ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h:989: error: (Each undeclared identi fier is reported only once ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h:989: error: for each function it ap pears in.) In file included from savagerender.c:138: ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h: In function ‘savage_flat_render_po ly_elts_s3d’: ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h:989: error: ‘Light’ undeclared (fir st use in this function) make[5]: *** [savagerender.o] Fehler 1 No time to hint the bug now, sorry. Eduard.
Bug#374986: tested patch, still crashes
On Fri, 2006-07-21 at 15:05 +0200, Eduard Bloch wrote: > > > In file included from savagerender.c:138: > ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h: In function > ‘savage_flat_render_po > ly_elts_s3d’: > ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h:989: error: ‘Light’ undeclared > (fir > st use in this function) > make[5]: *** [savagerender.o] Fehler 1 > > No time to hint the bug now, sorry. Assuming you don't need the savage driver, try building with DRI_DIRS=r300. -- Earthling Michel Dänzer | http://tungstengraphics.com Libre software enthusiast | Debian, X and DRI developer
Bug#374986: tested patch, still crashes
#include * Michel Dänzer [Fri, Jul 21 2006, 04:36:46PM]: > On Fri, 2006-07-21 at 15:05 +0200, Eduard Bloch wrote: > > > > > In file included from savagerender.c:138: > > ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h: In function > > ‘savage_flat_render_po > > ly_elts_s3d’: > > ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h:989: error: ‘Light’ undeclared > > (fir > > st use in this function) > > make[5]: *** [savagerender.o] Fehler 1 > > > > No time to hint the bug now, sorry. > > Assuming you don't need the savage driver, try building with > DRI_DIRS=r300. Ok, tried that. Compiled, packages created, installed, uncommented dri module in xorg, restarted X -> got deadloop as usual :-( Eduard.
Bug#374986: tested patch, still crashes
On Fri, 2006-07-21 at 17:04 +0200, Eduard Bloch wrote: > #include > * Michel Dänzer [Fri, Jul 21 2006, 04:36:46PM]: > > On Fri, 2006-07-21 at 15:05 +0200, Eduard Bloch wrote: > > > > > > > In file included from savagerender.c:138: > > > ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h: In function > > > ‘savage_flat_render_po > > > ly_elts_s3d’: > > > ../../../../../src/mesa/tnl_dd/t_dd_dmatmp.h:989: error: ‘Light’ > > > undeclared (fir > > > st use in this function) > > > make[5]: *** [savagerender.o] Fehler 1 > > > > > > No time to hint the bug now, sorry. > > > > Assuming you don't need the savage driver, try building with > > DRI_DIRS=r300. > > Ok, tried that. Compiled, packages created, installed, uncommented dri > module in xorg, restarted X -> got deadloop as usual :-( I'm afraid there's been a misunderstanding; Mesa is only involved when starting a 3D application, so it's not relevant to your original problem. As I pointed out before, Robert is seeing a different problem. One thing you have in common though is that you should report your problems upstream. :) Eduard, if you still feel adventurous, you could try current xf86-video-ati git as well. Maybe there have been other changes that helped r300 DRI stability. -- Earthling Michel Dänzer | http://tungstengraphics.com Libre software enthusiast | Debian, X and DRI developer