latest radeonhd crashes Xorg - Asus A8Jp, Mobility Radeon X1700
Hi! I pulled the latest master and built it: now it crashes the the Xorg on start, with the following backtrace: ========================================= ... (II) RADEONHD(0): RandR: Adding output VGA_CRT1/DAC_A (II) RADEONHD(0): RandR: Adding output PANEL_LCD1/LVDS (II) RADEONHD(0): RandR: Adding output SVIDEO_TV1/DAC_B (II) RADEONHD(0): RandR: Adding output DVI-I_DFP1/TMDS_A (II) RADEONHD(0): Output VGA_CRT1/DAC_A has no monitor section (II) RADEONHD(0): Output PANEL_LCD1/LVDS has no monitor section (II) RADEONHD(0): Output SVIDEO_TV1/DAC_B has no monitor section (II) RADEONHD(0): Output DVI-I_DFP1/TMDS_A has no monitor section (II) RADEONHD(0): Output VGA_CRT1/DAC_A disconnected (II) RADEONHD(0): EDID for output VGA_CRT1/DAC_A (II) RADEONHD(0): Output PANEL_LCD1/LVDS connected Backtrace: 0: /usr/bin/X(xf86SigHandler+0x81) [0x80c9581] 1: [0xffffe420] 2: /usr/bin/X(xf86DoEDID_DDC2+0xd6) [0x8104426] 3: /usr/local/lib/xorg/modules/drivers//radeonhd_drv.so [0xb7c126ec] 4: /usr/bin/X(xf86ProbeOutputModes+0x2c7) [0x80f9087] 5: /usr/bin/X(xf86InitialConfiguration+0x83) [0x80f9763] 6: /usr/local/lib/xorg/modules/drivers//radeonhd_drv.so(RHDRandrPreInit+0x3ba) [0xb7c1245a] 7: /usr/local/lib/xorg/modules/drivers//radeonhd_drv.so [0xb7c0545e] 8: /usr/bin/X(InitOutput+0x9a4) [0x80a8e54] 9: /usr/bin/X(main+0x27b) [0x8076ceb] 10: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7dd5050] 11: /usr/bin/X(FontFileCompleteXLFD+0x1e1) [0x8076241] ========================================= I attached the full Xorg log to the email. Setting the option "noRandR" fixes the crash. Please let me know if I can provide more information. Best regards, Kirill
Hi, Almost same situation on a macBook Pro with X1600... Please let me know if there's anything I can do. (--) RADEONHD(0): Detected a 18bit dual link panel. (--) RADEONHD(0): Attaching Output LVDS to Connector Panel (--) RADEONHD(0): Attaching Output DAC B to Connector DVI-I (--) RADEONHD(0): Attaching Output TMDS A to Connector DVI-I (II) RADEONHD(0): RandR: Adding output Panel/LVDS (II) RADEONHD(0): RandR: Adding output DVI-I/DAC_B (II) RADEONHD(0): RandR: Adding output DVI-I/TMDS_A (II) RADEONHD(0): Output Panel/LVDS using monitor section macBookLVDS (II) RADEONHD(0): Output DVI-I/DAC_B has no monitor section (II) RADEONHD(0): Output DVI-I/TMDS_A has no monitor section Backtrace: 0: /usr/bin/X(xf86SigHandler+0x6a) [0x482daa] 1: /lib/libc.so.6 [0x2b1b959e06a0] 2: /usr/bin/X(xf86DoEDID_DDC2+0xd0) [0x4b8480] 3: /opt/lib/xorg/modules/drivers//radeonhd_drv.so [0x2b1b97567833] 4: /usr/bin/X(xf86ProbeOutputModes+0x23d) [0x4ac80d] 5: /usr/bin/X(xf86InitialConfiguration+0x14c) [0x4acf7c] 6: /opt/lib/xorg/modules/drivers//radeonhd_drv.so(RHDRandrPreInit+0x33f) [0x2b1b975675bf] 7: /opt/lib/xorg/modules/drivers//radeonhd_drv.so [0x2b1b9755c2f4] 8: /usr/bin/X(InitOutput+0xa20) [0x467b60] 9: /usr/bin/X(main+0x29f) [0x43683f] 10: /lib/libc.so.6(__libc_start_main+0xf4) [0x2b1b959ccb44] 11: /usr/bin/X(FontFileCompleteXLFD+0x259) [0x435d59] Fatal server error: Caught signal 11. Server aborting On Wed, 2007-11-21 at 20:50 +0200, Kirill Belokurov wrote:
Hi!
I pulled the latest master and built it: now it crashes the the Xorg on start, with the following backtrace:
========================================= ...
(II) RADEONHD(0): RandR: Adding output VGA_CRT1/DAC_A (II) RADEONHD(0): RandR: Adding output PANEL_LCD1/LVDS (II) RADEONHD(0): RandR: Adding output SVIDEO_TV1/DAC_B (II) RADEONHD(0): RandR: Adding output DVI-I_DFP1/TMDS_A (II) RADEONHD(0): Output VGA_CRT1/DAC_A has no monitor section (II) RADEONHD(0): Output PANEL_LCD1/LVDS has no monitor section (II) RADEONHD(0): Output SVIDEO_TV1/DAC_B has no monitor section (II) RADEONHD(0): Output DVI-I_DFP1/TMDS_A has no monitor section (II) RADEONHD(0): Output VGA_CRT1/DAC_A disconnected (II) RADEONHD(0): EDID for output VGA_CRT1/DAC_A (II) RADEONHD(0): Output PANEL_LCD1/LVDS connected
Backtrace: 0: /usr/bin/X(xf86SigHandler+0x81) [0x80c9581] 1: [0xffffe420] 2: /usr/bin/X(xf86DoEDID_DDC2+0xd6) [0x8104426] 3: /usr/local/lib/xorg/modules/drivers//radeonhd_drv.so [0xb7c126ec] 4: /usr/bin/X(xf86ProbeOutputModes+0x2c7) [0x80f9087] 5: /usr/bin/X(xf86InitialConfiguration+0x83) [0x80f9763] 6: /usr/local/lib/xorg/modules/drivers//radeonhd_drv.so(RHDRandrPreInit+0x3ba) [0xb7c1245a] 7: /usr/local/lib/xorg/modules/drivers//radeonhd_drv.so [0xb7c0545e] 8: /usr/bin/X(InitOutput+0x9a4) [0x80a8e54] 9: /usr/bin/X(main+0x27b) [0x8076ceb] 10: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7dd5050] 11: /usr/bin/X(FontFileCompleteXLFD+0x1e1) [0x8076241]
=========================================
I attached the full Xorg log to the email. Setting the option "noRandR" fixes the crash.
Please let me know if I can provide more information.
Best regards, Kirill
On Nov 21, 07 20:50:51 +0200, Kirill Belokurov wrote:
I pulled the latest master and built it: now it crashes the the Xorg on start, with the following backtrace:
I had another guy with the same issue, but he couldn't help us, and I
cannot reproduce here.
So if you can build the driver with debug support (-O0 -g3) and run X in
gdb during startup (in a remote session, because you won't be able to
see the console :P) in order to get a reasonable stacktrace, that would
be awesome.
I might need more then, but only with a reasonable stacktrace I can
start debugging.
Thanks
Matthias
--
Matthias Hopf
On Wednesday 21 November 2007, Matthias Hopf wrote:
On Nov 21, 07 20:50:51 +0200, Kirill Belokurov wrote:
I pulled the latest master and built it: now it crashes the the Xorg on start, with the following backtrace:
I had another guy with the same issue, but he couldn't help us, and I cannot reproduce here.
So if you can build the driver with debug support (-O0 -g3) and run X in gdb during startup (in a remote session, because you won't be able to see the console :P) in order to get a reasonable stacktrace, that would be awesome.
OK, I rebuilt the radeonhd and the stacktrace is attached: output of bt both brief and full. Let me know if you need more info. Btw, does the fact that actual crash happens inside of xfree86/ddc/xf86DDC.c mean that this is not radeonhd issue and it should be forwarded to Xorg Bugzilla? Regards, Kirill
On Wed, Nov 21, 2007 at 11:08:40PM +0200, Kirill Belokurov wrote:
OK, I rebuilt the radeonhd and the stacktrace is attached: output of bt both brief and full. Let me know if you need more info.
Btw, does the fact that actual crash happens inside of xfree86/ddc/xf86DDC.c mean that this is not radeonhd issue and it should be forwarded to Xorg Bugzilla?
Regards, Kirill
#0 DDCRead_DDC2 (scrnIndex=0, pBus=0x0, start=0, len=128) at ../../../../hw/xfree86/ddc/xf86DDC.c:344 #1 0x08104426 in xf86DoEDID_DDC2 (scrnIndex=0, pBus=0x0) at ../../../../hw/xfree86/ddc/xf86DDC.c:299
*points out the bleeding obvious -- for X devs at least* Urgh, the I2CBus pointer passed is null. Matthias will enjoy digging that one out in the morning :) Luc Verhaegen. SUSE/Novell X Driver Developer and pointer-outer-of-the-bleeding-obvious. -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org
On Nov 21, 07 23:08:40 +0200, Kirill Belokurov wrote:
OK, I rebuilt the radeonhd and the stacktrace is attached: output of bt both brief and full. Let me know if you need more info.
Great, I'll look into that today.
Btw, does the fact that actual crash happens inside of xfree86/ddc/xf86DDC.c mean that this is not radeonhd issue and it should be forwarded to Xorg Bugzilla?
Not yet. I still think this is something in our code. And even if it
isn't, as long as no other driver triggers that one it won't be fixed
upstream otherwise.
CU
Matthias
--
Matthias Hopf
participants (4)
-
Jonas Aamodt Moræus
-
Kirill Belokurov
-
Luc Verhaegen
-
Matthias Hopf