Re: [PATCH 2/2] tracing: identify which executable object the userspace address belongs to

[Ingo Molnar <mingo@elte.hu>]

* Török Edwin <edwintorok@gmail.com> wrote:

> Thanks. I can move on to the lock latency tracing ;)

that's a bit more contentious ...

> I'll send out a draft of tracepoints that I would need to trace lock 
> latency. I'll try to put them in same place as lockstat (but not 
> necesarely depending on lockstat being enabled).

> Or I could add the tracepoints inside lockstat (now that it has 
> contend with points feature), and use the information already 
> gathered by lockstat, but augment it with finer grained counts per 
> kernel/user stacktrace. (again there would be an ftrace plugin that 
> would register with the tracepoints, and show the per stacktrace 
> statistic in /sys/kernel/debug/tracing/trace).

yes. The less intrusive your patch is, the more you utilize and 
generalize existing facilities, the better. You could split the 
Kconfig of LOCKSTAT into two bits: LOCKSTAT (core) and LOCKSTAT_PROC, 
where the proc bits are enabled separately.

Your tracing approach could then reuse much of core LOCKSTAT (without 
even touching the code) and just plain "select LOCKSTAT" - without 
creating /proc/lockdep_stats.

Peter, what do you think?

	Ingo