Although the system seems to mostly run OK (using ntpd and the reference clock 20 driver (https://www.eecis.udel.edu/~mills/ntp/h ... ver20.html), I have seen the C2 get in to a hard-loop that just spews:
Code: Select all
------------ [ cut here ] ------------ WARNING: CPU: 0 PID: 0 at kernel/timer.c:1038 del_timer_sync+0x44/0x68() ... Call trace: ... bust_spinlocks+0x28/0x4c die+0xb8/0x1a8 arm64_notify_die+0x48/0x8c do_undefinstr+0x108/0x258 Exception stack ... update_wall_time+0x168/0x690 tick_do_update_jiffies64+0xe4/x0140 tick_sched_do_timer+0x4c/0x58 __run_hrtimer+0x94/0x32c hrtimer_interrupt+0x114/0x2a8 meson_timer_interrupt+0x34/0xa4 handle_irq_event_percpu+0x78/0x290 handle_irq_event+0x54/0x80 handle_fasteoi_irq+0xc4/0x1a4 generic_handle_irq+0x3c/0x54 handle_IRQ+0x74/0xe8 gic_handle_irq+0x48/0x8c Exception stack ... el1_irq+0x64/0xd0 cpu_startup_entry+0x1b0/0x274 rest_init+0x8c/0x94 start_kernel_0x38c/0x3a0
Any idea what might be causing this? I have the CPU governor set to "performance", the clock speed underclocked to 1000 MHz, and the CPU temps are good (46°C / 114°F).
Unfortunately, this isn't easily reproducible. But the whole point of getting the C2 was to have a very reliable local source of time on the network, so if the C2 is going to randomly lock up, that's not going to work out very well!
Any hints would be appreciated.