Re: [PATCH 0/3] x86/dumpstack: Print registers with the same log level as the backtrace
From: Dmitry Safonov
Date: Thu Jun 25 2020 - 12:41:32 EST
On 6/24/20 7:50 AM, Petr Mladek wrote:
> On Tue 2020-06-23 17:29:55, Dmitry Safonov wrote:
>> show_trace_log_lvl() provides x86 platform-specific way to unwind
>> backtrace with a given log level. Unfortunately, registers dump(s) are
>> not printed with the same log level - instead, KERN_DEFAULT is always
>> used.
>>
>> Arista's switches uses quite common setup with rsyslog, where only
>> urgent messages goes to console (console_log_level=KERN_ERR), everything
>> else goes into /var/log/ as the console baud-rate often is indecently
>> slow (9600 bps).
>>
>> Backtrace dumps without registers printed have proven to be as useful as
>> morning standups. Furthermore, in order to introduce KERN_UNSUPPRESSED
>> (which I believe is still the most elegant way to fix raciness of sysrq[1])
>> the log level should be passed down the stack to register dumping
>> functions. Besides, I have a potential use-case for printing traces
>> with KERN_DEBUG level [2] (where registers dump shouldn't appear with
>> higher log level than the backtrace).
>>
>> Dmitry Safonov (3):
>> x86/dumpstack: Add log_lvl to show_iret_regs()
>> x86/dumpstack: Add log_lvl to __show_regs()
>> x86/dumpstack: Show registers dump with trace's log level
>
> The change makes sense. It is natural next step after adding log_lvl
> parameter for printing stack traces. For the entire patchset:
>
> Acked-by: Petr Mladek <pmladek@xxxxxxxx>
I'll address the nit by Jann and resend v2 with your Ack, thanks!
> Are there any plans to add this also for other architectures, please?
Yes, I'll look into that.
Thanks,
Dmitry