Bug 91893 - R100: GPU lockup: EQ overflowing
Summary: R100: GPU lockup: EQ overflowing
Status: NEW
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/R100 (show other bugs)
Version: 11.0
Hardware: PowerPC Linux (All)
: medium major
Assignee: Default DRI bug account
QA Contact: Default DRI bug account
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-09-06 10:33 UTC by macro
Modified: 2015-09-28 09:39 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
Full dmesg (47.95 KB, text/plain)
2015-09-06 10:43 UTC, macro
Details
Full Xorg.0.log (37.09 KB, text/plain)
2015-09-06 10:43 UTC, macro
Details
radeontool registers dump before lockup (19.66 KB, text/plain)
2015-09-28 09:25 UTC, macro
Details
radeontool registers dump after lockup (19.75 KB, text/plain)
2015-09-28 09:25 UTC, macro
Details

Note You need to log in before you can comment on or make changes to this bug.
Description macro 2015-09-06 10:33:33 UTC
The following GPU lockup occurs consistently soon after starting Dillo (an fltk-based lightweight web browser) compiled without Cairo support.

I am happy to provide any additional information or diagnose further as directed.

"ATI Radeon 7500 QW (AGP/PCI)" (ChipID = 0x5157)

Linux cube 4.1.6-gentoo #2 Sat Aug 29 23:38:46 AEST 2015 ppc 7400, altivec supported PowerMac5,1 GNU/Linux

mesa-11.0.0_rc1

dmesg
...
[ 2561.090198] radeon 0000:00:10.0: ring 0 stalled for more than 10000msec
[ 2561.090226] radeon 0000:00:10.0: GPU lockup (current fence id 0x0000000000000125 last fence id 0x0000000000000126 on ring 0)
[ 2561.095953] radeon 0000:00:10.0: Saved 27 dwords of commands on ring 0.
[ 2561.095969] radeon 0000:00:10.0: GPU reset succeeded, trying to resume
[ 2561.096012] [drm] PCI GART of 32M enabled (table at 0x000000002EE08000).
[ 2561.096026] radeon 0000:00:10.0: WB disabled
[ 2561.096044] radeon 0000:00:10.0: fence driver on ring 0 use gpu addr 0x0000000096000000 and cpu addr 0xeed95000
[ 2561.096140] [drm] radeon: ring at 0x0000000096001000
[ 2561.243896] [drm:r100_ring_test] *ERROR* radeon: ring test failed (scratch(0x15E8)=0xCAFEDEAD)
[ 2561.243908] [drm:r100_cp_init] *ERROR* radeon: cp isn't working (-22).
[ 2561.243920] radeon 0000:00:10.0: failed initializing CP (-22).
...

Xorg.0.log
...
(EE) [mi] EQ overflowing.  Additional events will be discarded until existing events are processed.
(EE)
(EE) Backtrace:
(EE) 0: /usr/bin/X (xorg_backtrace+0x58) [0x101ccefc]
(EE) 1: /usr/bin/X (mieqEnqueue+0x120) [0x101abe78]
(EE) 2: /usr/bin/X (0x10000000+0x48414) [0x10048414]
(EE) 3: /usr/bin/X (QueuePointerEvents+0x68) [0x1004b56c]
(EE) 4: /usr/bin/X (xf86PostMotionEventM+0x340) [0x1008ecd4]
(EE) 5: /usr/lib/xorg/modules/input/evdev_drv.so (0xe9a7000+0x50e4) [0xe9ac0e4]
(EE) 6: /usr/lib/xorg/modules/input/evdev_drv.so (0xe9a7000+0x5a24) [0xe9aca24]
(EE) 7: /usr/lib/xorg/modules/input/evdev_drv.so (0xe9a7000+0x5d80) [0xe9acd80]
(EE) 8: /usr/bin/X (0x10000000+0x7adec) [0x1007adec]
(EE) 9: /usr/bin/X (0x10000000+0xad47c) [0x100ad47c]
(EE) 10: linux-vdso32.so.1 (__kernel_sigtramp32+0x0) [0x100364]
(EE) 11: /lib/libc.so.6 (ioctl+0xe0) [0xfa3ce00]
(EE) 12: ?? [0x0]
(EE) 13: /usr/lib/libdrm.so.2 (drmIoctl+0x50) [0xfd436ac]
(EE) 14: /usr/lib/libdrm.so.2 (drmCommandWrite+0x38) [0xfd46e3c]
(EE) 15: /usr/lib/libdrm_radeon.so.1 (0xf01c000+0x1650) [0xf01d650]
(EE) 16: /usr/lib/libdrm_radeon.so.1 (0xf01c000+0x1788) [0xf01d788]
(EE) 17: /usr/lib/libdrm_radeon.so.1 (radeon_bo_map+0x20) [0xf01f22c]
(EE) 18: /usr/lib/xorg/modules/drivers/radeon_drv.so (0xf04d000+0x30ab0) [0xf07dab0]
(EE) 19: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x6168) [0xefb0168]
(EE) 20: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x63a0) [0xefb03a0]
(EE) 21: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x90c8) [0xefb30c8]
(EE) 22: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x4bf4) [0xefaebf4]
(EE) 23: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x12ea8) [0xefbcea8]
(EE) 24: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x9e88) [0xefb3e88]
(EE) 25: /usr/bin/X (0x10000000+0x146dfc) [0x10146dfc]
(EE) 26: /usr/bin/X (0x10000000+0x2bacc) [0x1002bacc]
(EE) 27: /usr/bin/X (0x10000000+0x30254) [0x10030254]
(EE) 28: /usr/bin/X (0x10000000+0x349c0) [0x100349c0]
(EE) 29: /usr/bin/X (0x10000000+0x1c2b8) [0x1001c2b8]
(EE) 30: /lib/libc.so.6 (0xf956000+0x21938) [0xf977938]
(EE) 31: /lib/libc.so.6 (0xf956000+0x21ad8) [0xf977ad8]
(EE)
(EE) [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
(EE) [mi] mieq is *NOT* the cause.  It is a victim.
(EE) [mi] EQ overflow continuing.  100 events have been dropped.
(EE)
(EE) Backtrace:
(EE) 0: /usr/bin/X (xorg_backtrace+0x58) [0x101ccefc]
(EE) 1: /usr/bin/X (mieqEnqueue+0x190) [0x101abee8]
(EE) 2: /usr/bin/X (0x10000000+0x48414) [0x10048414]
(EE) 3: /usr/bin/X (QueuePointerEvents+0x68) [0x1004b56c]
(EE) 4: /usr/bin/X (xf86PostMotionEventM+0x340) [0x1008ecd4]
(EE) 5: /usr/lib/xorg/modules/input/evdev_drv.so (0xe9a7000+0x50e4) [0xe9ac0e4]
(EE) 6: /usr/lib/xorg/modules/input/evdev_drv.so (0xe9a7000+0x5a24) [0xe9aca24]
(EE) 7: /usr/lib/xorg/modules/input/evdev_drv.so (0xe9a7000+0x5d80) [0xe9acd80]
(EE) 8: /usr/bin/X (0x10000000+0x7adec) [0x1007adec]
(EE) 9: /usr/bin/X (0x10000000+0xad47c) [0x100ad47c]
(EE) 10: linux-vdso32.so.1 (__kernel_sigtramp32+0x0) [0x100364]
(EE) 11: /lib/libc.so.6 (ioctl+0xe0) [0xfa3ce00]
(EE) 12: ?? [0x0]
(EE) 13: /usr/lib/libdrm.so.2 (drmIoctl+0x50) [0xfd436ac]
(EE) 14: /usr/lib/libdrm.so.2 (drmCommandWrite+0x38) [0xfd46e3c]
(EE) 15: /usr/lib/libdrm_radeon.so.1 (0xf01c000+0x1650) [0xf01d650]
(EE) 16: /usr/lib/libdrm_radeon.so.1 (0xf01c000+0x1788) [0xf01d788]
(EE) 17: /usr/lib/libdrm_radeon.so.1 (radeon_bo_map+0x20) [0xf01f22c]
(EE) 18: /usr/lib/xorg/modules/drivers/radeon_drv.so (0xf04d000+0x30ab0) [0xf07dab0]
(EE) 19: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x6168) [0xefb0168]
(EE) 20: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x63a0) [0xefb03a0]
(EE) 21: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x90c8) [0xefb30c8]
(EE) 22: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x4bf4) [0xefaebf4]
(EE) 23: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x12ea8) [0xefbcea8]
(EE) 24: /usr/lib/xorg/modules/libexa.so (0xefaa000+0x9e88) [0xefb3e88]
(EE) 25: /usr/bin/X (0x10000000+0x146dfc) [0x10146dfc]
(EE) 26: /usr/bin/X (0x10000000+0x2bacc) [0x1002bacc]
(EE) 27: /usr/bin/X (0x10000000+0x30254) [0x10030254]
(EE) 28: /usr/bin/X (0x10000000+0x349c0) [0x100349c0]
(EE) 29: /usr/bin/X (0x10000000+0x1c2b8) [0x1001c2b8]
(EE) 30: /lib/libc.so.6 (0xf956000+0x21938) [0xf977938]
(EE) 31: /lib/libc.so.6 (0xf956000+0x21ad8) [0xf977ad8]
Comment 1 macro 2015-09-06 10:43:10 UTC
Created attachment 118097 [details]
Full dmesg
Comment 2 macro 2015-09-06 10:43:27 UTC
Created attachment 118098 [details]
Full Xorg.0.log
Comment 3 macro 2015-09-28 09:25:06 UTC
Created attachment 118476 [details]
radeontool registers dump before lockup
Comment 4 macro 2015-09-28 09:25:34 UTC
Created attachment 118477 [details]
radeontool registers dump after lockup
Comment 5 macro 2015-09-28 09:39:36 UTC
Kernel arguments:

radeon.no_wb=1
radeon.dynclks=0
radeon.agpmode=-1
radeon.gartsize=32
radeon.msi=0
radeon.dpm=0
radeon.aspm=0
radeon.runpm=0
radeon.hard_reset=1
radeon.use_pflipirq=0
radeon.bapm=0

One or more (not sure exactly which) are required in order for X to start without hanging the machine.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.