108401 – GPU lockup with [AMD/ATI] RV730 XT [Radeon HD 4670]

Bug 108401 - GPU lockup with [AMD/ATI] RV730 XT [Radeon HD 4670]

Summary: GPU lockup with [AMD/ATI] RV730 XT [Radeon HD 4670]

Status:	NEW

Alias:	None

Product:	Drivers
Classification:	Unclassified
Component:	Video(DRI - non Intel) (show other bugs)
Hardware:	x86-64 Linux

Importance:	P1 normal
Assignee:	drivers_video-dri

URL:
Keywords:

Depends on:
Blocks:

Reported:	2015-11-24 19:23 UTC by joe.r.floss.user
Modified:	2019-10-02 23:27 UTC (History)
CC List:	3 users (show)

See Also:
Kernel Version:	4.2.6
Subsystem:
Regression:	No
Bisected commit-id:

Attachments
X.org log file (44.51 KB, application/x-trash) 2015-11-24 19:23 UTC, joe.r.floss.user	Details
Output of 'journalctl -kb -1' after the first reboot (105.64 KB, text/plain) 2015-11-24 19:26 UTC, joe.r.floss.user	Details
Add an attachment (proposed patch, testcase, etc.)

Description joe.r.floss.user 2015-11-24 19:23:49 UTC

Created attachment 195301 [details]
X.org log file

Hello,

I am having relatively frequent GPU lockups when using FlightGear, which uses 3D a lot. I am on Debian unstable using the following packages:

ii  libc6                                  2.19-22
ii  libudev1                               228-2
ii  xserver-xorg-core [xorg-video-abi-19]  2:1.17.3-2

ii  libdrm-amdgpu1:amd64     2.4.65-3
ii  libdrm-dev:amd64         2.4.65-3
ii  libdrm-intel1:amd64      2.4.65-3
ii  libdrm-nouveau2:amd64    2.4.65-3
ii  libdrm-radeon1:amd64     2.4.65-3
ii  libdrm-radeon1-dbg:amd64 2.4.65-3
ii  libdrm2:amd64            2.4.65-3

My graphics adapter is:

01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] RV730 XT [Radeon HD 4670] (prog-if 00 [VGA controller])
        Subsystem: PC Partner Limited / Sapphire Technology Device e100
        Flags: bus master, fast devsel, latency 0, IRQ 34
        Memory at e0000000 (64-bit, prefetchable) [size=256M]
        Memory at fb9e0000 (64-bit, non-prefetchable) [size=64K]
        I/O ports at be00 [size=256]
        [virtual] Expansion ROM at fb900000 [disabled] [size=128K]
        Capabilities: [50] Power Management version 3
        Capabilities: [58] Express Legacy Endpoint, MSI 00
        Capabilities: [a0] MSI: Enable+ Count=1/1 Maskable- 64bit+
        Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
        Kernel driver in use: radeon

The kernel log corresponding to the last lockup is:

Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: ring 0 stalled for more than 10004msec
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: GPU lockup (current fence id 0x000000000024264d last fence id 0x000000000024267e on ring 0)
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: Saved 1561 dwords of commands on ring 0.
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: GPU softreset: 0x00000009
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_008010_GRBM_STATUS      = 0xE7730028
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_008014_GRBM_STATUS2     = 0x00330302
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_000E50_SRBM_STATUS      = 0x200038C0
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_008678_CP_STALLED_STAT2 = 0x00010802
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_00867C_CP_BUSY_STAT     = 0x00028186
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_008680_CP_STAT          = 0x80038647
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_00D034_DMA_STATUS_REG   = 0x44C83D57
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: R_008020_GRBM_SOFT_RESET=0x00007F6B
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: SRBM_SOFT_RESET=0x00000100
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_008010_GRBM_STATUS      = 0x00003028
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_008014_GRBM_STATUS2     = 0x00000002
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_000E50_SRBM_STATUS      = 0x200000C0
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_008678_CP_STALLED_STAT2 = 0x00000000
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_00867C_CP_BUSY_STAT     = 0x00000000
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_008680_CP_STAT          = 0x00000000
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0:   R_00D034_DMA_STATUS_REG   = 0x44C83D57
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: GPU reset succeeded, trying to resume
Nov 24 17:34:19 zita kernel: [drm] enabling PCIE gen 2 link speeds, disable with radeon.pcie_gen2=0
Nov 24 17:34:19 zita kernel: [drm] PCIE GART of 1024M enabled (table at 0x000000000025E000).
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: WB enabled
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000020000c00 and cpu addr 0xffff88003578dc00
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: fence driver on ring 3 use gpu addr 0x0000000020000c0c and cpu addr 0xffff88003578dc0c
Nov 24 17:34:19 zita kernel: radeon 0000:01:00.0: fence driver on ring 5 use gpu addr 0x000000000005c598 and cpu addr 0xffffc9000181c598
Nov 24 17:34:20 zita kernel: [drm] ring test on 0 succeeded in 1 usecs
Nov 24 17:34:20 zita kernel: [drm] ring test on 3 succeeded in 2 usecs
Nov 24 17:34:20 zita kernel: [drm] ring test on 5 succeeded in 1 usecs
Nov 24 17:34:20 zita kernel: [drm] UVD initialized successfully.
Nov 24 17:34:20 zita kernel: [drm:rv770_dpm_set_power_state [radeon]] *ERROR* rv770_set_sw_state failed
Nov 24 17:34:20 zita kernel: [drm] ib test on ring 0 succeeded in 0 usecs
Nov 24 17:34:20 zita kernel: [drm] ib test on ring 3 succeeded in 0 usecs
Nov 24 17:37:00 zita kernel: INFO: task Xorg:1301 blocked for more than 120 seconds.
Nov 24 17:37:00 zita kernel:       Not tainted 4.2.0-1-amd64 #1
Nov 24 17:37:00 zita kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Nov 24 17:37:00 zita kernel: Xorg            D 0000000000000003     0  1301   1287 0x00000004
Nov 24 17:37:00 zita kernel:  ffff880035836040 0000000000000082 ffff8800b1b6ecc0 ffff8800d9ea71c0
Nov 24 17:37:00 zita kernel:  0000000000000000 ffff880213e80000 ffff880035cc0030 ffff880035cc0018
Nov 24 17:37:00 zita kernel:  ffffffff00000001 ffffffff00000000 fffffffe00000001 ffffffff815500ff
Nov 24 17:37:00 zita kernel: Call Trace:
Nov 24 17:37:00 zita kernel:  [<ffffffff815500ff>] ? schedule+0x2f/0x70
Nov 24 17:37:00 zita kernel:  [<ffffffff8155270d>] ? rwsem_down_write_failed+0x1dd/0x300
Nov 24 17:37:00 zita kernel:  [<ffffffff812d26b3>] ? call_rwsem_down_write_failed+0x13/0x20
Nov 24 17:37:00 zita kernel:  [<ffffffff81551fd9>] ? down_write+0x29/0x40
Nov 24 17:37:00 zita kernel:  [<ffffffffa052db02>] ? radeon_gpu_reset+0x22/0x2f0 [radeon]
Nov 24 17:37:00 zita kernel:  [<ffffffff81404609>] ? fence_wait_timeout+0x39/0x120
Nov 24 17:37:00 zita kernel:  [<ffffffff8140566b>] ? reservation_object_wait_timeout_rcu+0xeb/0x280
Nov 24 17:37:00 zita kernel:  [<ffffffffa055beee>] ? radeon_gem_handle_lockup.part.3+0xe/0x20 [radeon]
Nov 24 17:37:00 zita kernel:  [<ffffffffa04850b2>] ? drm_ioctl+0x172/0x550 [drm]
Nov 24 17:37:00 zita kernel:  [<ffffffff8107bfd3>] ? signal_setup_done+0x63/0xa0
Nov 24 17:37:00 zita kernel:  [<ffffffffa052b04a>] ? radeon_drm_ioctl+0x4a/0x80 [radeon]
Nov 24 17:37:00 zita kernel:  [<ffffffff811cf2c3>] ? do_vfs_ioctl+0x2c3/0x4a0
Nov 24 17:37:00 zita kernel:  [<ffffffff81078f67>] ? recalc_sigpending+0x17/0x50
Nov 24 17:37:00 zita kernel:  [<ffffffff811cf516>] ? SyS_ioctl+0x76/0x90
Nov 24 17:37:00 zita kernel:  [<ffffffff81553a32>] ? system_call_fast_compare_end+0xc/0x6b
Nov 24 17:37:00 zita kernel: INFO: task kworker/u16:2:8971 blocked for more than 120 seconds.
Nov 24 17:37:00 zita kernel:       Not tainted 4.2.0-1-amd64 #1
Nov 24 17:37:00 zita kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Nov 24 17:37:00 zita kernel: kworker/u16:2   D ffffffff8109a1b0     0  8971      2 0x00000000
Nov 24 17:37:00 zita kernel: Workqueue: radeon-crtc radeon_flip_work_func [radeon]
Nov 24 17:37:00 zita kernel:  ffff8800b1b6ecc0 0000000000000046 00000000000004b0 ffff880216218d40
Nov 24 17:37:00 zita kernel:  0000000000000004 ffff8801e62d0000 ffff8801e62cfc58 ffff880035cc0000
Nov 24 17:37:00 zita kernel:  ffff8801e62cfbf0 ffff880035cc1490 ffff880035cc0018 ffffffff815500ff
Nov 24 17:37:00 zita kernel: Call Trace:
Nov 24 17:37:00 zita kernel:  [<ffffffff815500ff>] ? schedule+0x2f/0x70
Nov 24 17:37:00 zita kernel:  [<ffffffff81552bc7>] ? schedule_timeout+0x1f7/0x290
Nov 24 17:37:00 zita kernel:  [<ffffffffa05464fe>] ? radeon_fence_process+0xe/0x30 [radeon]
Nov 24 17:37:00 zita kernel:  [<ffffffffa0546801>] ? radeon_fence_wait_seq_timeout.constprop.8+0x271/0x2c0 [radeon]
Nov 24 17:37:00 zita kernel:  [<ffffffff810a9ae0>] ? wait_woken+0x80/0x80
Nov 24 17:37:00 zita kernel:  [<ffffffffa0546af9>] ? radeon_fence_wait+0x79/0xb0 [radeon]
Nov 24 17:37:00 zita kernel:  [<ffffffffa060eb1b>] ? uvd_v1_0_ib_test+0xfb/0x150 [radeon]
Nov 24 17:37:00 zita kernel:  [<ffffffffa0601d58>] ? radeon_ib_ring_tests+0x58/0xc0 [radeon]
Nov 24 17:37:00 zita kernel:  [<ffffffffa052dcd4>] ? radeon_gpu_reset+0x1f4/0x2f0 [radeon]
Nov 24 17:37:00 zita kernel:  [<ffffffffa055385b>] ? radeon_flip_work_func+0x14b/0x170 [radeon]
Nov 24 17:37:00 zita kernel:  [<ffffffff81084c5a>] ? process_one_work+0x14a/0x3b0
Nov 24 17:37:00 zita kernel:  [<ffffffff810855fd>] ? worker_thread+0x4d/0x450
Nov 24 17:37:00 zita kernel:  [<ffffffff810855b0>] ? rescuer_thread+0x2e0/0x2e0
Nov 24 17:37:00 zita kernel:  [<ffffffff8108aa41>] ? kthread+0xc1/0xe0
Nov 24 17:37:00 zita kernel:  [<ffffffff8108a980>] ? kthread_create_on_node+0x170/0x170
Nov 24 17:37:00 zita kernel:  [<ffffffff81553e5f>] ? ret_from_fork+0x3f/0x70
Nov 24 17:37:00 zita kernel:  [<ffffffff8108a980>] ? kthread_create_on_node+0x170/0x170
Nov 24 17:39:00 zita kernel: INFO: task Xorg:1301 blocked for more than 120 seconds.
Nov 24 17:39:00 zita kernel:       Not tainted 4.2.0-1-amd64 #1
Nov 24 17:39:00 zita kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Nov 24 17:39:00 zita kernel: Xorg            D 0000000000000003     0  1301   1287 0x00000004
Nov 24 17:39:00 zita kernel:  ffff880035836040 0000000000000082 ffff8800b1b6ecc0 ffff8800d9ea71c0
Nov 24 17:39:00 zita kernel:  0000000000000000 ffff880213e80000 ffff880035cc0030 ffff880035cc0018
Nov 24 17:39:00 zita kernel:  ffffffff00000001 ffffffff00000000 fffffffe00000001 ffffffff815500ff
Nov 24 17:39:00 zita kernel: Call Trace:
Nov 24 17:39:00 zita kernel:  [<ffffffff815500ff>] ? schedule+0x2f/0x70
Nov 24 17:39:00 zita kernel:  [<ffffffff8155270d>] ? rwsem_down_write_failed+0x1dd/0x300
Nov 24 17:39:00 zita kernel:  [<ffffffff812d26b3>] ? call_rwsem_down_write_failed+0x13/0x20
Nov 24 17:39:00 zita kernel:  [<ffffffff81551fd9>] ? down_write+0x29/0x40
Nov 24 17:39:00 zita kernel:  [<ffffffffa052db02>] ? radeon_gpu_reset+0x22/0x2f0 [radeon]
Nov 24 17:39:00 zita kernel:  [<ffffffff81404609>] ? fence_wait_timeout+0x39/0x120
Nov 24 17:39:00 zita kernel:  [<ffffffff8140566b>] ? reservation_object_wait_timeout_rcu+0xeb/0x280
Nov 24 17:39:00 zita kernel:  [<ffffffffa055beee>] ? radeon_gem_handle_lockup.part.3+0xe/0x20 [radeon]
Nov 24 17:39:00 zita kernel:  [<ffffffffa04850b2>] ? drm_ioctl+0x172/0x550 [drm]
Nov 24 17:39:00 zita kernel:  [<ffffffff8107bfd3>] ? signal_setup_done+0x63/0xa0
Nov 24 17:39:00 zita kernel:  [<ffffffffa052b04a>] ? radeon_drm_ioctl+0x4a/0x80 [radeon]
Nov 24 17:39:00 zita kernel:  [<ffffffff811cf2c3>] ? do_vfs_ioctl+0x2c3/0x4a0
Nov 24 17:39:00 zita kernel:  [<ffffffff81078f67>] ? recalc_sigpending+0x17/0x50
Nov 24 17:39:00 zita kernel:  [<ffffffff811cf516>] ? SyS_ioctl+0x76/0x90
Nov 24 17:39:00 zita kernel:  [<ffffffff81553a32>] ? system_call_fast_compare_end+0xc/0x6b
Nov 24 17:39:00 zita kernel: INFO: task kworker/u16:2:8971 blocked for more than 120 seconds.
Nov 24 17:39:00 zita kernel:       Not tainted 4.2.0-1-amd64 #1
Nov 24 17:39:00 zita kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Nov 24 17:39:00 zita kernel: kworker/u16:2   D ffffffff8109a1b0     0  8971      2 0x00000000
Nov 24 17:39:00 zita kernel: Workqueue: radeon-crtc radeon_flip_work_func [radeon]
Nov 24 17:39:00 zita kernel:  ffff8800b1b6ecc0 0000000000000046 00000000000004b0 ffff880216218d40
Nov 24 17:39:00 zita kernel:  0000000000000004 ffff8801e62d0000 ffff8801e62cfc58 ffff880035cc0000
Nov 24 17:39:00 zita kernel:  ffff8801e62cfbf0 ffff880035cc1490 ffff880035cc0018 ffffffff815500ff
Nov 24 17:39:00 zita kernel: Call Trace:
Nov 24 17:39:00 zita kernel:  [<ffffffff815500ff>] ? schedule+0x2f/0x70
Nov 24 17:39:00 zita kernel:  [<ffffffff81552bc7>] ? schedule_timeout+0x1f7/0x290
Nov 24 17:39:00 zita kernel:  [<ffffffffa05464fe>] ? radeon_fence_process+0xe/0x30 [radeon]
Nov 24 17:39:00 zita kernel:  [<ffffffffa0546801>] ? radeon_fence_wait_seq_timeout.constprop.8+0x271/0x2c0 [radeon]
Nov 24 17:39:00 zita kernel:  [<ffffffff810a9ae0>] ? wait_woken+0x80/0x80
Nov 24 17:39:00 zita kernel:  [<ffffffffa0546af9>] ? radeon_fence_wait+0x79/0xb0 [radeon]
Nov 24 17:39:00 zita kernel:  [<ffffffffa060eb1b>] ? uvd_v1_0_ib_test+0xfb/0x150 [radeon]
Nov 24 17:39:00 zita kernel:  [<ffffffffa0601d58>] ? radeon_ib_ring_tests+0x58/0xc0 [radeon]
Nov 24 17:39:00 zita kernel:  [<ffffffffa052dcd4>] ? radeon_gpu_reset+0x1f4/0x2f0 [radeon]
Nov 24 17:39:00 zita kernel:  [<ffffffffa055385b>] ? radeon_flip_work_func+0x14b/0x170 [radeon]
Nov 24 17:39:00 zita kernel:  [<ffffffff81084c5a>] ? process_one_work+0x14a/0x3b0
Nov 24 17:39:00 zita kernel:  [<ffffffff810855fd>] ? worker_thread+0x4d/0x450
Nov 24 17:39:00 zita kernel:  [<ffffffff810855b0>] ? rescuer_thread+0x2e0/0x2e0
Nov 24 17:39:00 zita kernel:  [<ffffffff8108aa41>] ? kthread+0xc1/0xe0
Nov 24 17:39:00 zita kernel:  [<ffffffff8108a980>] ? kthread_create_on_node+0x170/0x170
Nov 24 17:39:00 zita kernel:  [<ffffffff81553e5f>] ? ret_from_fork+0x3f/0x70
Nov 24 17:39:00 zita kernel:  [<ffffffff8108a980>] ? kthread_create_on_node+0x170/0x170

I am attaching the Xorg.0.log for the session that had the lockup.

Thanks in advance for considering.

Comment 1 joe.r.floss.user 2015-11-24 19:26:11 UTC

Created attachment 195311 [details]
Output of 'journalctl -kb -1' after the first reboot

Full kernel log

Comment 2 Alex Deucher 2015-11-24 19:27:31 UTC

What version of mesa are you using?  Does updating your version of mesa help?

Comment 3 joe.r.floss.user 2015-11-24 19:38:57 UTC

% COLUMNS=120 dpkg -l|grep 'mesa.*amd64'

→ ...

ii  libegl1-mesa:amd64         11.0.5-1
ii  libegl1-mesa-dev:amd64     11.0.5-1
ii  libgl1-mesa-dev:amd64      11.0.5-1
ii  libgl1-mesa-dri:amd64      11.0.5-1
ii  libgl1-mesa-glx:amd64      11.0.5-1
ii  libglapi-mesa:amd64        11.0.5-1
ii  libgles1-mesa:amd64        11.0.5-1
ii  libgles2-mesa:amd64        11.0.5-1
ii  libgles2-mesa-dev:amd64    11.0.5-1
ii  libglu1-mesa:amd64         9.0.0-2.
ii  libglu1-mesa-dev:amd64     9.0.0-2.
ii  libwayland-egl1-mesa:amd64 11.0.5-1
ii  mesa-common-dev:amd64      11.0.5-1
ii  mesa-utils                 8.2.0-3+b1
ii  mesa-vdpau-drivers:amd64   11.0.5-1

Currently, I haven't tried upgrading it (mesa), but if this is the way to go, I'll try and report. Thank you.

Comment 4 Alex Deucher 2015-11-24 19:40:55 UTC

Or try downgrading.  The problem is most likely in mesa.

Comment 5 joe.r.floss.user 2015-11-24 22:52:59 UTC

For now, I have installed the latest mesa from git master. A short test was OK, but is not enough to conclude. Should I report here or on the mesa bug tracker if the problem appears again?

Thanks.

Comment 6 Alex Deucher 2015-11-25 14:16:06 UTC

Please report Mesa bugs on https://bugs.freedesktop.org

Comment 7 joe.r.floss.user 2015-11-28 11:54:04 UTC

Right, I've had no such problem with recent mesa from Git master (47b3a0d23504de43ce2dc028e84037f7cd89bd89). Besides, should the problem arise again, I would report it where you said, therefore I suppose you may close this report.

Thank you!

Note You need to log in before you can comment on or make changes to this bug.