Summary: | kwin crashes if I choose "i965 gallium" in ems-7.10.2 | ||
---|---|---|---|
Product: | [Plasma] kwin | Reporter: | Toralf Förster <toralf.foerster> |
Component: | general | Assignee: | KWin default assignee <kwin-bugs-null> |
Status: | RESOLVED DUPLICATE | ||
Severity: | crash | ||
Priority: | NOR | ||
Version: | unspecified | ||
Target Milestone: | --- | ||
Platform: | Compiled Sources | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: |
Description
Toralf Förster
2011-04-16 10:25:24 UTC
driver bug *** This bug has been marked as a duplicate of bug 252817 *** Well, rather the xorg-server itself. I switched from xorg-server 1.95 to 1.10.1 and now the issue seems to be fixed. (In reply to comment #2) > Well, rather the xorg-server itself. > > I switched from xorg-server 1.95 to 1.10.1 and now the issue seems to be fixed. nope - not fixed :-( it's a gallium issue, see the dupe (have no intel chip to test, but believe their findings) (In reply to comment #4) > it's a gallium issue, see the dupe (have no intel chip to test, but believe > their findings) Hhm, I'm wondering then why I get this error b/c I set Gentoo to use classic (intel GM45 graphic) : foerste@n22 ~ $ eselect mesa list i915 (Intel 915, 945) [1] classic * [2] gallium i965 (Intel 965, G/Q3x, G/Q4x) [1] classic * [2] gallium r300 (Radeon R300-R500) r600 (Radeon R600-R700, Evergreen, Northern Islands) sw (Software renderer) [1] classic * [2] gallium Just for the record this error I found in Xorg.log with current environment (xorg 1.10.1, mesa 7.10.2, intel 2.15.0, KDE 4.6.2) : Backtrace: [ 3251.280] 0: /usr/bin/X (xorg_backtrace+0x3c) [0x80ea36c] [ 3251.280] 1: /usr/bin/X (0x8048000+0x60e71) [0x80a8e71] [ 3251.280] 2: (vdso) (__kernel_rt_sigreturn+0x0) [0xb788140c] [ 3251.280] 3: /usr/lib/dri/i965_dri.so (0xb6e25000+0x199b6) [0xb6e3e9b6] [ 3251.280] 4: /usr/lib/dri/i965_dri.so (0xb6e25000+0x1c800) [0xb6e41800] [ 3251.280] 5: /usr/lib/dri/i965_dri.so (0xb6e25000+0x1f70c8) [0xb701c0c8] [ 3251.280] 6: /usr/lib/xorg/modules/extensions/libglx.so (0xb7426000+0x838a) [0xb742e38a] [ 3251.280] 7: /usr/lib/xorg/modules/extensions/libglx.so (0xb7426000+0x322a2) [0xb74582a2] [ 3251.280] 8: /usr/lib/xorg/modules/extensions/libglx.so (0xb7426000+0x37b73) [0xb745db73] [ 3251.280] 9: /usr/bin/X (0x8048000+0x2ad27) [0x8072d27] [ 3251.280] 10: /usr/bin/X (0x8048000+0x1d88a) [0x806588a] [ 3251.280] 11: /lib/libc.so.6 (__libc_start_main+0xe6) [0xb759dbb6] [ 3251.280] 12: /usr/bin/X (0x8048000+0x1d441) [0x8065441] [ 3251.280] Segmentation fault at address 0x30 [ 3251.280] Fatal server error: [ 3251.280] Caught signal 11 (Segmentation fault). Server aborting Errrr... the summary and first post state that you tried the gallium driver & got a crash in kwin while the classic driver works fine. This is however a segfault in X11 (clients should not be able to segfault X11, therefore it's a pure driver/X11 bug), ie. a completely different issue (though GLX related - basically do never attempt to run two OpenGL clients on the intel driver, so at least suspend compositing before running any other OpenGL application or use the XRender backend) (In reply to comment #6) (though > GLX related - basically do never attempt to run two OpenGL clients on the intel > driver, so at least suspend compositing before running any other OpenGL > application or use the XRender backend) Right - Thanks - I changed to XRender and the crashes went away :-) Should I file a bug against freedesktop and/or intel driver ? (And BTW unmark this bug as a duplicate ?) This bug as presented is a dupe of #252817 - if you've other crashes with different backtraces you should file them independently, but i foresee a dupe there as well (and at least upstream, the intel driver has a "history" since they the kms move :-( I assume the original (dupe) bug is known to freedesktop/intel but you could file or search the bug there to track its progress. wrong dupe... :S (though very likely same upstream source) *** This bug has been marked as a duplicate of bug 255094 *** |