Re: [PATCH 00/50] Add log level to show_stack()

[Dmitry Safonov <dima@arista.com>]

On 11/6/19 8:35 AM, Petr Mladek wrote:
> On Wed 2019-11-06 03:04:51, Dmitry Safonov wrote:
>> Add log level argument to show_stack().
>> Done in three stages:
>> 1. Introducing show_stack_loglvl() for every architecture
>> 2. Migrating old users with an explicit log level
>> 3. Renaming show_stack_loglvl() into show_stack()
>>
>> Justification:
>> o It's a design mistake to move a business-logic decision
>>   into platform realization detail.
>> o I have currently two patches sets that would benefit from this work:
>>   Removing console_loglevel jumps in sysrq driver [1]
> 
> Just to clarify. The problem in sysrq driver is a bit different.
> It modifies console_loglevel to show even less important message
> on the console.
> 
> IMHO, it should be solved by printing the header line with pr_error().
> It is not ideal. A cleaner solution might be to introduce another
> loglevel that will always get pushed to the console. But I am
> not sure if it is worth this single line.

I believe why it's not done - there is a comment in sysrq code that said
the userspace relies on the loglevel of sysrq messages (and may trigger
alerts from emerg/err log level):
  * Raise the apparent loglevel to maximum so that the sysrq header
  * is shown to provide the user with positive feedback.  We do not
  * simply emit this at KERN_EMERG as that would change message
  * routing in the consumers of /proc/kmsg.

But I don't mind any solution.

>>   Hung task warning before panic [2] - suggested by Tetsuo (but he
>>   probably didn't realise what it would involve).
>> o While doing (1), (2) the backtraces were adjusted to headers
>>   and other messages for each situation - so there won't be a situation
>>   when the backtrace is printed, but the headers are missing because
>>   they have lesser log level (or the reverse).
>> o As the result in (2) plays with console_loglevel for kdb are removed.
> 
>> The least important for upstream, but maybe still worth to note that
>> every company I've worked in so far had an off-list patch to print
>> backtrace with the needed log level (but only for the architecture they
>> cared about).
>> If you have other ideas how you will benefit from show_stack() with
>> a log level - please, reply to this cover letter.
> 
> I agree with all the other justification.
> 
> I would add. The backtrace is really useful for debugging. It should
> be possible to print it even in less critical situations.
> 
> I am afraid that many people use WARN() for this purpose. But WARN()
> is not always appropriate. WARN() misuse huts when panic_on_warn
> option is used.

Thanks, Petr.

-- 
          Dmitry