Trace on HD6320 after resume on v3.7-rc6

Trace on HD6320 after resume on v3.7-rc6

[Julian Wollrath]

Hello,

while having problems with the resume of the snd-hda-intel driver, I
found the following in my dmesg output:
[  245.003310] irq 16: nobody cared (try booting with the "irqpoll" option)
[  245.003318] Pid: 691, comm: kworker/u:3 Tainted: G         C   3.7.0-rc6-wl+ #3
[  245.003321] Call Trace:
[  245.003342]  <IRQ>  [<ffffffff810a3c86>] ? __report_bad_irq+0x36/0xd0
[  245.003350]  [<ffffffff810a3f8d>] ? note_interrupt+0x1cd/0x220
[  245.003359]  [<ffffffff810a19ae>] ? handle_irq_event_percpu+0x8e/0x1c0
[  245.003367]  [<ffffffff81040415>] ? __do_softirq+0x105/0x1e0
[  245.003374]  [<ffffffff810a1b16>] ? handle_irq_event+0x36/0x60
[  245.003383]  [<ffffffff810a49ac>] ? handle_fasteoi_irq+0x4c/0xe0
[  245.003390]  [<ffffffff810045c5>] ? handle_irq+0x15/0x20
[  245.003396]  [<ffffffff810042b2>] ? do_IRQ+0x52/0xd0
[  245.003404]  [<ffffffff81405b2a>] ? common_interrupt+0x6a/0x6a
[  245.003430]  <EOI>  [<ffffffffa0349380>] ? ttm_bo_move_memcpy+0x3a0/0x4e0 [ttm]
[  245.003443]  [<ffffffffa03492e4>] ? ttm_bo_move_memcpy+0x304/0x4e0 [ttm]
[  245.003457]  [<ffffffffa0346c24>] ? ttm_bo_handle_move_mem+0x234/0x3e0 [ttm]
[  245.003469]  [<ffffffffa0347910>] ? ttm_bo_mem_space+0x170/0x350 [ttm]
[  245.003503]  [<ffffffffa047f200>] ? atom_execute_table_locked+0x70/0x2d0 [radeon]
[  245.003516]  [<ffffffffa0347c3f>] ? ttm_bo_move_buffer+0x14f/0x160 [ttm]
[  245.003528]  [<ffffffffa0347ce5>] ? ttm_bo_validate+0x95/0x120 [ttm]
[  245.003559]  [<ffffffffa0484bbc>] ? radeon_bo_pin_restricted+0x10c/0x1b0 [radeon]
[  245.003590]  [<ffffffffa0485a43>] ? radeon_gart_table_vram_pin+0x33/0xb0 [radeon]
[  245.003619]  [<ffffffffa0481835>] ? atom_execute_table+0x55/0x70 [radeon]
[  245.003649]  [<ffffffffa04c8c86>] ? evergreen_startup+0x386/0x1610 [radeon]
[  245.003677]  [<ffffffffa0481c05>] ? atom_asic_init+0x145/0x1a0 [radeon]
[  245.003706]  [<ffffffffa04ca1d2>] ? evergreen_resume+0x32/0x80 [radeon]
[  245.003730]  [<ffffffffa046e118>] ? radeon_resume_kms+0x68/0x190 [radeon]
[  245.003737]  [<ffffffff81254190>] ? pci_pm_restore+0xe0/0xe0
[  245.003745]  [<ffffffff812e4314>] ? dpm_run_callback.isra.5+0x14/0x40
[  245.003752]  [<ffffffff812e4844>] ? device_resume+0xb4/0x140
[  245.003759]  [<ffffffff812e48e4>] ? async_resume+0x14/0x40
[  245.003767]  [<ffffffff81060666>] ? async_run_entry_fn+0x86/0x1b0
[  245.003775]  [<ffffffff81052f06>] ? process_one_work+0x126/0x490
[  245.003782]  [<ffffffff810605e0>] ? async_unregister_domain+0x70/0x70
[  245.003788]  [<ffffffff81054c3d>] ? worker_thread+0x15d/0x450
[  245.003795]  [<ffffffff81054ae0>] ? flush_delayed_work+0x40/0x40
[  245.003801]  [<ffffffff810599a3>] ? kthread+0xb3/0xc0
[  245.003807]  [<ffffffff810598f0>] ? kthread_create_on_node+0x110/0x110
[  245.003814]  [<ffffffff814061ac>] ? ret_from_fork+0x7c/0xb0
[  245.003820]  [<ffffffff810598f0>] ? kthread_create_on_node+0x110/0x110
[  245.003823] handlers:
[  245.003838] [<ffffffffa0267050>] azx_interrupt [snd_hda_intel]
[  245.003841] Disabling IRQ #16

I was not able to reproduce it but maybe the trace is helpful
nevertheless. I got it during the resume after a suspend to ram.

If you have further questions, please feel free to ask.


With best regards,
Julian Wollrath

Re: Trace on HD6320 after resume on v3.7-rc6

[Michel Dänzer]

On Mit, 2012-11-21 at 14:24 +0100, Julian Wollrath wrote: 
> 
> while having problems with the resume of the snd-hda-intel driver, I
> found the following in my dmesg output:
> [  245.003310] irq 16: nobody cared (try booting with the "irqpoll" option)
[...] 
> [  245.003823] handlers:
> [  245.003838] [<ffffffffa0267050>] azx_interrupt [snd_hda_intel]
> [  245.003841] Disabling IRQ #16

Does /proc/interrupts show IRQ 16 being shared between snd_hda_intel and
radeon? If not, this looks like a snd_hda_intel (or another driver
sharing the IRQ) issue.


-- 
Earthling Michel Dänzer           |                   http://www.amd.com
Libre software enthusiast         |          Debian, X and DRI developer

Re: Trace on HD6320 after resume on v3.7-rc6

[Julian Wollrath]

> > [  245.003823] handlers:
> > [  245.003838] [<ffffffffa0267050>] azx_interrupt [snd_hda_intel]
> > [  245.003841] Disabling IRQ #16
> 
> Does /proc/interrupts show IRQ 16 being shared between snd_hda_intel
> and radeon? If not, this looks like a snd_hda_intel (or another driver
> sharing the IRQ) issue.
/proc/interrupts shows no sharing between snd_hda_intel and radeon. I
just noticed, that radeon was stated in the trace, therefore I send it
to this mailing list.

Thank you for the response and sorry for the noise.


With best regards,
Julian Wollrath