Gk20a_fifo te_open_trusted and truty_restore_keyslots

JP4.3
Deepstream4

Hi,

I can’t find any mentions of these faults

what could be the source?

Many thanks

[17410.086454] nvgpu: 17000000.gv11b gk20a_fifo_handle_pbdma_intr_0:2713 [ERR] semaphore acquire timeout!
[17410.086671] nvgpu: 17000000.gv11b nvgpu_set_error_notifier_locked:137 [ERR] error notifier set to 24 for ch 509
[37080.153001] nvgpu: 17000000.gv11b gk20a_fifo_handle_pbdma_intr_0:2713 [ERR] semaphore acquire timeout!
[37080.153215] nvgpu: 17000000.gv11b nvgpu_set_error_notifier_locked:137 [ERR] error notifier set to 24 for ch 509
[38845.348799] nvgpu: 17000000.gv11b __nvgpu_timeout_expired_msg_cpu:94 [ERR] Timeout detected @ pmu_wait_message_cond+0x88/0x118 [nvgpu]
[38937.569391] nvgpu: 17000000.gv11b gk20a_fifo_handle_pbdma_intr_0:2713 [ERR] semaphore acquire timeout!
[38937.569615] nvgpu: 17000000.gv11b nvgpu_set_error_notifier_locked:137 [ERR] error notifier set to 24 for ch 509
[48558.754561] nvgpu: 17000000.gv11b gk20a_fifo_handle_pbdma_intr_0:2713 [ERR] semaphore acquire timeout!
[48558.754783] nvgpu: 17000000.gv11b nvgpu_set_error_notifier_locked:137 [ERR] error notifier set to 24 for ch 509
[50245.585659] te_open_trusted_session:ERROR(-110) in receiving response from service
[50245.585832] trusty_restore_keyslots:ERROR(-110): Failed to open trusted session
[50249.605104] nvgpu: 17000000.gv11b gk20a_fifo_handle_pbdma_intr_0:2713 [ERR] semaphore acquire timeout!
[50249.605310] nvgpu: 17000000.gv11b nvgpu_set_error_notifier_locked:137 [ERR] error notifier set to 24 for ch 509
[50250.659035] te_open_trusted_session:ERROR(-110) in receiving response from service
[50250.659189] trusty_restore_keyslots:ERROR(-110): Failed to open trusted session
[50253.666764] nvgpu: 17000000.gv11b gk20a_fifo_handle_pbdma_intr_0:2713 [ERR] semaphore acquire timeout!
[50253.666963] nvgpu: 17000000.gv11b nvgpu_set_error_notifier_locked:137 [ERR] error notifier set to 24 for ch 509
[50255.779054] te_open_trusted_session:ERROR(-110) in receiving response from service
[50255.779264] trusty_restore_keyslots:ERROR(-110): Failed to open trusted session
[50260.899084] te_open_trusted_session:ERROR(-110) in receiving response from service
[50260.899255] trusty_restore_keyslots:ERROR(-110): Failed to open trusted session
[50266.019064] te_open_trusted_session:ERROR(-110) in receiving response from service
[50266.019270] trusty_restore_keyslots:ERROR(-110): Failed to open trusted session
[50271.139453] te_open_trusted_session:ERROR(-110) in receiving response from service
[50271.139619] trusty_restore_keyslots:ERROR(-110): Failed to open trusted session
[50276.259058] te_open_trusted_session:ERROR(-110) in receiving response from service
[50276.259228] trusty_restore_keyslots:ERROR(-110): Failed to open trusted session
[50281.383069] te_open_trusted_session:ERROR(-110) in receiving response from service
[50281.383247] trusty_restore_keyslots:ERROR(-110): Failed to open trusted session
[50286.499055] te_open_trusted_session:ERROR(-110) in receiving response from service
[50286.499227] trusty_restore_keyslots:ERROR(-110): Failed to open trusted session
[50291.619137] te_open_trusted_session:ERROR(-110) in receiving response from service
[50291.619328] trusty_restore_keyslots:ERROR(-110): Failed to open trusted session

Not sure for your case, but if you have a task that can hog the GPU for some time and can afford a frozen GUI, you may try:

sudo echo N > /sys/kernel/debug/17000000.gv11b/timeouts_enabled

hello cdevd,

may I know what’s the actual use-case, please share the detail steps to reproduce the failure.
thanks

Hi JerryChang

Just running Deepstream,
2 x axis p7304 encoders connected via netgear network switch
HDMI, keyboard and mouse go through a KVM

no other peripherals connected

here’s a dmesg from the unit in case that shows something

Thanks

dmesg.txt (81.2 KB)

There is no update from you for a period, assuming this is not an issue any more.
Hence we are closing this topic. If need further support, please open a new one.
Thanks

hello cdevd,

I don’t see gk20a failures from the attach logs.
could you please share which nvpmodel you’re running, i.e. $ nvpmodel -q
also, how long did you reproduce the failure.
thanks