New: r300 lockup

Email lists > Direct Rendering Open Source Project development

Thread: New: r300 lockup

Search this list only Search all lists
New: r300 lockup
United States	2008-03-25 15:11:51
htt p://bugs.freedesktop.org/show_bug.cgi?id=15203 Summary: r300 lockup Product: DRI Version: XOrg CVS Platform: Other OS/Version: All Status: NEW Severity: normal Priority: medium Component: DRM modules AssignedTo: dri-devellists.sourceforge.net ReportedBy: markus.amsleroribi.org Created an attachment (id=15455) --> (h ttp://bugs.freedesktop.org/attachment.cgi?id=15455) Test case The attached test locks up my RV370 [Radeon X300SE] within 1 second or less. It locks up in 3 possible ways: - machine up, xorg 100%, hw cursor works - machine up, xorg 100%, no hw cursor - system freeze If I export RADEON_DEBUG=all the lockup is gone. Attached is also a drm log, where the system didn't freeze. It only locks up if the tris variable in the test case is over 300. I run everything from git (xorg, drm, mesa, xf86-video-ati) Kernel: stock debian unstable 2.6.24-1-686 Compile test with gcc -std=c99 texsub.c -lglut -o texsub I tried to fix it myself without any luck. -- Configure bugmail: h ttp://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------ ------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ -- _______________________________________________ Dri-devel mailing list Dri-devellists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel

r300 lockup
United States	2008-03-26 06:51:09
htt p://bugs.freedesktop.org/show_bug.cgi?id=15203 --- Comment #2 from Oliver McFadden <z3ro.geekgmail.com> 2008-03-26 04:51:08 PST --- I just tested this on my development box (r300-vertprog-branch, ATI Technologies Inc RV410 [Radeon X700 Pro (PCIE)) and I can reproduce the lockup. The first time executed fine, but the second time hard locked. I looked at your debug log and it looks like it dies somewhere during processing of an indirect buffer, then just sits in a loop waiting for the CP to idle. Probably that indirect buffer does something it shouldn't... Mar 25 15:16:53 debian kernel: [drm:drm_unlocked_ioctl] pid=3090, cmd=0xc010644d, nr=0x4d, dev 0xe200, auth=1 Mar 25 15:16:53 debian kernel: [drm:radeon_cp_indirect] idx=11 s=0 e=112 d=0 Mar 25 15:16:53 debian kernel: [drm:radeon_cp_dispatch_indirect] buf=11 s=0x0 e=0x70 Mar 25 15:16:53 debian kernel: [drm:drm_unlocked_ioctl] pid=3090, cmd=0x6444, nr=0x44, dev 0xe200, auth=1 Mar 25 15:16:53 debian kernel: [drm:radeon_cp_idle] Mar 25 15:16:53 debian kernel: [drm:radeon_do_cp_idle] Mar 25 15:16:53 debian kernel: [drm:drm_unlocked_ioctl] ret = -16 Mar 25 15:16:53 debian kernel: [drm:drm_unlocked_ioctl] pid=3090, cmd=0x6444, nr=0x44, dev 0xe200, auth=1 ... Defining RADEON_FIFO_DEBUG to 1 at the top of radeon_cp.c will give some more information, though not a lot; still useful though. -- Configure bugmail: h ttp://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------ ------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216 239;13503038;w?http://sf.net/marketplace -- _______________________________________________ Dri-devel mailing list Dri-devellists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel

r300 lockup
United States	2008-03-25 15:17:06
htt p://bugs.freedesktop.org/show_bug.cgi?id=15203 --- Comment #1 from Markus Amsler <markus.amsleroribi.org> 2008-03-25 13:17:05 PST --- Created an attachment (id=15456) --> (h ttp://bugs.freedesktop.org/attachment.cgi?id=15456) drm log where system didn't freeze. -- Configure bugmail: h ttp://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------ ------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ -- _______________________________________________ Dri-devel mailing list Dri-devellists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel

r300 lockup
United States	2008-03-26 08:17:14
htt p://bugs.freedesktop.org/show_bug.cgi?id=15203 --- Comment #3 from Markus Amsler <markus.amsleroribi.org> 2008-03-26 06:17:14 PST --- Created an attachment (id=15472) --> (h ttp://bugs.freedesktop.org/attachment.cgi?id=15472) drm log with RADEON_FIFO_DEBUG -- Configure bugmail: h ttp://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------ ------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216 239;13503038;w?http://sf.net/marketplace -- _______________________________________________ Dri-devel mailing list Dri-devellists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel

r300 lockup
United States	2008-03-26 19:37:03
htt p://bugs.freedesktop.org/show_bug.cgi?id=15203 --- Comment #4 from Markus Amsler <markus.amsleroribi.org> 2008-03-26 17:37:02 PST --- Created an attachment (id=15493) --> (h ttp://bugs.freedesktop.org/attachment.cgi?id=15493) mesa patch This patch makes the lockup gone. I have absolutely no clue why. I just knew RADEON_DEBUG=sync fixes the lockup. So I peppered the source with radeonWaitForIdle until I found this hack -- Configure bugmail: h ttp://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------ ------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216 239;13503038;w?http://sf.net/marketplace -- _______________________________________________ Dri-devel mailing list Dri-devellists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel

r300 lockup
United States	2008-03-27 09:39:37
htt p://bugs.freedesktop.org/show_bug.cgi?id=15203 --- Comment #5 from Jerome Glisse <glissefreedesktop.org> 2008-03-27 07:39:36 PST --- (In reply to comment #4) > Created an attachment (id=15493) --> (h ttp://bugs.freedesktop.org/attachment.cgi?id=15493) [details] > mesa patch > > This patch makes the lockup gone. I have absolutely no clue why. I just knew > RADEON_DEBUG=sync fixes the lockup. So I peppered the source with > radeonWaitForIdle until I found this hack > This means that this lockup is due to sync problem. My wild guess is that we write a register that shouldn't be written before a flush (like changing vertex program while there is still fragment being rendered with old vertex program). I think we could apply the patch you point to as temporary fix. -- Configure bugmail: h ttp://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------ ------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216 239;13503038;w?http://sf.net/marketplace -- _______________________________________________ Dri-devel mailing list Dri-devellists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel

r300 lockup
United States	2008-03-27 12:52:37
htt p://bugs.freedesktop.org/show_bug.cgi?id=15203 --- Comment #6 from Markus Amsler <markus.amsleroribi.org> 2008-03-27 10:52:37 PST --- Its has a 30% performance impact on Wow, but only 1% on glxgears. It also doesn't fix the Wow lockup I was after in the first place. It's also possible that it doesn't fix the bug, but only changes the timing, so the bug gets not triggered. I'm gonna investigate this further. -- Configure bugmail: h ttp://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------ ------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216 239;13503038;w?http://sf.net/marketplace -- _______________________________________________ Dri-devel mailing list Dri-devellists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel

r300 lockup
United States	2008-03-27 13:19:26
htt p://bugs.freedesktop.org/show_bug.cgi?id=15203 --- Comment #7 from Michel Dänzer <micheltungstengraphics.com> 2008-03-27 11:19:26 PST --- Synchronously waiting for idle on each flush is a pretty large hammer... does just emitting wait for idle commands to the CP instead help as well? -- Configure bugmail: h ttp://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------ ------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216 239;13503038;w?http://sf.net/marketplace -- _______________________________________________ Dri-devel mailing list Dri-devellists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel

r300 lockup
United States	2008-03-28 00:23:04
htt p://bugs.freedesktop.org/show_bug.cgi?id=15203 --- Comment #8 from Oliver McFadden <z3ro.geekgmail.com> 2008-03-27 22:23:03 PST --- I would rather NOT see this patch committed to Mesa, as it really just hides the issue: we're doing something wrong, which should be fixed. This must be a sync problem because more aggressive flushing can solve the problem, though it's not the correct solution. A discussion on IRC between MrCooper and glisse mentioned the NQWAIT_UNTIL (0xe50) register. I've seen the blob write this register in Revenge dumps, but it's never used by our drivers. I wouldn't mind getting some comments from AMD describing the precisely correct way to flush and synchronize the ASIC. We're obviously missing something. I guess tcore might provide this information once it's released... -- Configure bugmail: h ttp://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------ ------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216 239;13503038;w?http://sf.net/marketplace -- _______________________________________________ Dri-devel mailing list Dri-devellists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel

r300 lockup
United States	2008-03-28 15:56:20
htt p://bugs.freedesktop.org/show_bug.cgi?id=15203 --- Comment #9 from Markus Amsler <markus.amsleroribi.org> 2008-03-28 13:56:19 PST --- Emitting wait doesn't help. -- Configure bugmail: h ttp://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------ ------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216 239;13503038;w?http://sf.net/marketplace -- _______________________________________________ Dri-devel mailing list Dri-devellists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel

[1-10] [11-15]

about | contact Other archives ( Real Estate discussion Medical topics )