Call trace on Jetson Nano

[chhg82]

Hello everyone
On my Jetson Nano/kirkstone-l4t-r32.7.x is a lot of call trace in dmesg when starting up, probably from the latest kernel version 4.9.337. But during runtime, I didn't notice anything negative about the GPU/CPU performance. On the Jetson NX machine nothing is logged. Is anyone aware of something like this?

[    9.207623] CPU: 2 PID: 3268 Comm: nvpmodel Tainted: G        W       4.9.337-l4t-r32.7.4+ga1cf17bc0aa0 #1
[    9.207625] Hardware name: NVIDIA Jetson Nano Developer Kit (DT)
[    9.207627] task:         (ptrval) task.stack:         (ptrval)
[    9.207630] PC is at kmem_cache_create+0x1c4/0x250
[    9.207632] LR is at kmem_cache_create+0xcc/0x250
[    9.207634] pc : [<ffffff800820e024>] lr : [<ffffff800820df2c>] pstate: 60400045
[    9.207635] sp : ffffffc0f4a63840
[    9.207639] x29: ffffffc0f4a63840 x28: 0000000000000001 
[    9.207642] x27: 0000000100000000 x26: 0000000000000080 
[    9.207645] x25: 0000000000000000 x24: 0000000000000000 
[    9.207648] x23: 0000000000000080 x22: ffffffc0f90dff10 
[    9.207650] x21: ffffff800a287dd0 x20: ffffff800a287df8 
[    9.207653] x19: ffffff800a287d90 x18: ffffffffffffffff 
[    9.207656] x17: 0000000000000000 x16: 0000000000000000 
[    9.207659] x15: ffffffc0f90dff90 x14: ffffffffffffffff 
[    9.207662] x13: ffffffc0f90dff33 x12: 0000000000000000 
[    9.207665] x11: 0101010101010101 x10: ffffffc0f4a638b0 
[    9.207668] x9 : 0000000000000080 x8 : ffffffc0f90dff26 
[    9.207670] x7 : 000000007f200000 x6 : ffffffc0f4a638b0 
[    9.207673] x5 : ffffff800a287df8 x4 : 0000000000000040 
[    9.207676] x3 : 0000000000000010 x2 : 0000000000000020 
[    9.207678] x1 : 0000000000000020 x0 : ffffffc0f90dff1e 

[    9.207681] ---[ end trace 90374a96e6eb674d ]---
[    9.207682] Call trace:
[    9.207686] [<        (ptrval)>] kmem_cache_create+0x1c4/0x250
[    9.207977] [<        (ptrval)>] nvgpu_kmem_cache_create+0x88/0x100 [nvgpu]
[    9.208269] [<        (ptrval)>] nvgpu_buddy_allocator_init+0x140/0x4a0 [nvgpu]
[    9.208545] [<        (ptrval)>] __nvgpu_vm_init+0x3a8/0x6d0 [nvgpu]
[    9.208826] [<        (ptrval)>] nvgpu_vm_init+0x90/0xe0 [nvgpu]
[    9.209103] [<        (ptrval)>] nvgpu_init_mm_setup_sw+0x360/0x4b0 [nvgpu]
[    9.209374] [<        (ptrval)>] nvgpu_init_mm_support+0x80/0xa0 [nvgpu]
[    9.209646] [<        (ptrval)>] gk20a_finalize_poweron+0x4b0/0x930 [nvgpu]
[    9.209919] [<        (ptrval)>] gk20a_pm_finalize_poweron+0xe8/0x400 [nvgpu]
[    9.210197] [<        (ptrval)>] gk20a_pm_runtime_resume+0x40/0x74 [nvgpu]
[    9.210205] [<        (ptrval)>] pm_generic_runtime_resume+0x40/0x60
[    9.210209] [<        (ptrval)>] __rpm_callback+0x78/0xa0
[    9.210212] [<        (ptrval)>] rpm_callback+0x38/0xa0
[    9.210215] [<        (ptrval)>] rpm_resume+0x4b8/0x710
[    9.210218] [<        (ptrval)>] pm_runtime_forbid+0x50/0x64
[    9.210221] [<        (ptrval)>] control_store+0xac/0x120
[    9.210225] [<        (ptrval)>] dev_attr_store+0x48/0x60
[    9.210230] [<        (ptrval)>] sysfs_kf_write+0x60/0x70
[    9.210233] [<        (ptrval)>] kernfs_fop_write+0xc4/0x1dc
[    9.210236] [<        (ptrval)>] __vfs_write+0x4c/0x130
[    9.210238] [<        (ptrval)>] vfs_write+0xb0/0x1b4
[    9.210240] [<        (ptrval)>] SyS_write+0x60/0xd0
[    9.210245] [<        (ptrval)>] __sys_trace_return+0x0/0x4
[    9.242226] ------------[ cut here ]------------
[    9.242236] WARNING: CPU: 1 PID: 1934 at mm/slab_common.c:98 kmem_cache_create+0x1c4/0x250
[    9.242241] Modules linked in: nvgpu

[    9.242247] CPU: 1 PID: 1934 Comm: irq/78-gk20a_st Tainted: G        W       4.9.337-l4t-r32.7.4+ga1cf17bc0aa0 #1
[    9.242248] Hardware name: NVIDIA Jetson Nano Developer Kit (DT)
[    9.242250] task:         (ptrval) task.stack:         (ptrval)
[    9.242253] PC is at kmem_cache_create+0x1c4/0x250
[    9.242256] LR is at kmem_cache_create+0xcc/0x250
[    9.242258] pc : [<ffffff800820e024>] lr : [<ffffff800820df2c>] pstate: 60400045
[    9.242260] sp : ffffffc0f8b93a80
[    9.242264] x29: ffffffc0f8b93a80 x28: 0000000000000000 
[    9.242267] x27: 0000000000000000 x26: 0000000000000040 
[    9.242270] x25: 0000000000000000 x24: 0000000000000000 
[    9.242273] x23: 0000000000000040 x22: ffffffc0f88a7a90 
[    9.242276] x21: ffffff800a287dd0 x20: ffffff800a287df8 
[    9.242279] x19: ffffff800a287d90 x18: ffffffffffffffff 
[    9.242282] x17: 0000000000000000 x16: 0000000000000000 
[    9.242285] x15: ffffffc0f88a7b10 x14: ffffffffffffffff 
[    9.242287] x13: ffffffc0f88a7ab2 x12: ffffff800900aac0 
[    9.242290] x11: 0000000000000001 x10: ffffffc0f8b93af0 
[    9.242293] x9 : 0000000000000040 x8 : ffffffc0f88a7aa6 
[    9.242296] x7 : 000000007f200000 x6 : ffffffc0f8b93af0 
[    9.242299] x5 : ffffff800a287df8 x4 : 0000000000000040 
[    9.242302] x3 : 0000000000000010 x2 : 0000000000000020 
[    9.242304] x1 : 0000000000000020 x0 : ffffffc0f88a7a9e 

[    9.242307] ---[ end trace 90374a96e6eb674e ]---
[    9.242308] Call trace:
[    9.242312] [<        (ptrval)>] kmem_cache_create+0x1c4/0x250
[    9.242612] [<        (ptrval)>] nvgpu_kmem_cache_create+0x88/0x100 [nvgpu]
[    9.242902] [<        (ptrval)>] nvgpu_bitmap_allocator_init+0x23c/0x2d0 [nvgpu]
[    9.243183] [<        (ptrval)>] nvgpu_pmu_process_init_msg+0x278/0x300 [nvgpu]
[    9.243472] [<        (ptrval)>] nvgpu_pmu_process_message+0x1f0/0x250 [nvgpu]
[    9.243752] [<        (ptrval)>] gk20a_pmu_isr+0x220/0x28c [nvgpu]
[    9.244020] [<        (ptrval)>] gm20b_mc_isr_stall+0xd4/0x170 [nvgpu]
[    9.244286] [<        (ptrval)>] nvgpu_intr_thread_stall+0x54/0x1e0 [nvgpu]
[    9.244559] [<        (ptrval)>] gk20a_intr_thread_stall+0x24/0x30 [nvgpu]
[    9.244565] [<        (ptrval)>] irq_thread_fn+0x34/0x84
[    9.244568] [<        (ptrval)>] irq_thread+0x174/0x1e0
[    9.244571] [<        (ptrval)>] kthread+0xf0/0xf4
[    9.244575] [<        (ptrval)>] ret_from_fork+0x10/0x20

Many Thanks!

[madisongh]

Looks like #1364... I merged some fixes with #1370 that should help.

[chhg82]

I'm already up to date with kirkstone-l4t-r32.7.x but the boot problem still occurs. I'll take a closer look at it again tomorrow.

[madisongh]

Adding haveged or some other entropy generator in your build, to ensure that the kernel's entropy pool gets to a suitable size quickly enough, may be needed. We include that in the demo distro here, but if you've created your own distro, you'll need to add it in yourself.

[chhg82]

Perfect thank you!