0

I am debugging an problem wherein sometimes, reboot command just does not boot. Very similar to https://e2e.ti.com/support/processors-group/processors/f/processors-forum/1086480/am3352-linux-reboot-command-hangs-for-10-minutes-while-booting-down-then-succeeds but not the same.

Can "tracefs" be used for it ? I ask this because in all examples I have seen the trace-buffer is stored in /sys/kernel/tracing/trace, and then later this file is dumped.

Since the issue is in the reboot sequence, I will not have the option to dump this.

Is there a way that I can have the trace-buffer to be directly printed to console ?

I looked at tp_printk kernel cmdline option. However when I enable the "function_graph" I still see the output in the trace file but not console.

Could you let me know the parameters that need to be set to have the tracing subsystem print the trace to console.

Thanks

ksrikanth
  • 71
  • 9
  • set the loglevel and console on the kernel command line https://elixir.bootlin.com/linux/v5.19-rc8/source/Documentation/admin-guide/kernel-parameters.txt#L698 – stark Sep 13 '22 at 12:26
  • 1
    @stark tracefs does not print to dmesg, so that doesn't help OP – Marco Bonelli Sep 14 '22 at 13:45
  • Yes, there is a specific command line for that (I don't remember by heart), just read documentation, it's all written there. – 0andriy Sep 21 '22 at 12:33

0 Answers0