Here’s the stack trace of the Kernel panic message
[ 105.984169] CPU: 1 UID: 0 PID: 2340 Comm: ovx_pipeline_wi Tainted: G W
[ 105.993284] Tainted: [W]=WARN, [O]=OOT_MODULE
[ 105.997625] Hardware name: Texas Instruments J721E SK (DT)
[ 106.003093] pstate: 80000005 (Nzcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=–)
[ 106.010038] pc : find_vma+0x64/0x70
[ 106.013536] lr : hailo_vdma_buffer_map+0xe0/0x5cc [hailo_pci]
[ 106.019277] sp : ffff800087fe3c20
[ 106.022578] x29: ffff800087fe3c30 x28: ffff00082da71080 x27: 0000000000000000
[ 106.029699] x26: 0000000000000000 x25: 0000000000000000 x24: ffff000831f21980
[ 106.036818] x23: 0000000000009600 x22: 0000000000000002 x21: 0000000000000000
[ 106.043937] x20: ffff0008284550c8 x19: 0000ffffb2556000 x18: 0000000000000000
[ 106.051055] x17: 0000000000000000 x16: 0000000000000000 x15: 0000ffff87ffc148
[ 106.058175] x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000
[ 106.065294] x11: 0000000000000000 x10: 0000ffff87ffc148 x9 : 0000000000000000
[ 106.072412] x8 : ffff000831f21a00 x7 : 0000000000000000 x6 : 000000000000003f
[ 106.079532] x5 : 0000000000000040 x4 : 0000000000000000 x3 : 0000000000000000
[ 106.086650] x2 : ffff00082da71080 x1 : 0000000000000000 x0 : ffff000831e80000
[ 106.093770] Call trace:
[ 106.096204] find_vma+0x64/0x70
[ 106.099337] hailo_vdma_buffer_map+0xe0/0x5cc [hailo_pci]
[ 106.104725] hailo_vdma_buffer_map_ioctl+0xd4/0x338 [hailo_pci]
[ 106.110631] hailo_vdma_ioctl+0xc4/0x248 [hailo_pci]
[ 106.115585] hailo_pcie_fops_unlockedioctl+0x1d8/0x7a8 [hailo_pci]
[ 106.121751] __arm64_sys_ioctl+0xac/0xf0
[ 106.125663] invoke_syscall+0x48/0x10c
[ 106.129401] el0_svc_common.constprop.0+0xc0/0xe0
[ 106.134090] do_el0_svc+0x1c/0x28
[ 106.137395] el0_svc+0x28/0x98
[ 106.140441] el0t_64_sync_handler+0x120/0x12c
[ 106.144785] el0t_64_sync+0x190/0x194
[ 106.148436] —[ end trace 0000000000000000 ]—
[ 106.153157] ------------[ cut here ]------------
I’ve seen the same issue reported on the The stack trace points to an issue with VDMA buffer mapping in the hailo_pci driver 4.20.1. - #5 by ksj
and since the thread is 10 days old and we are using the different version of the Kernel, I created the new thread to report the problem.
As I replied on the thread mentioned above, downgrading is not suitable option for us. We could try way lower version of the kernel with the different SoC, but since the Yocto for the BSP is Hardknott, which has been severed from the official support list long time ago, I doubt it would work with the up to date hailo-pci driver and the libhailort library.
Have a great day!