Re: Fix powerTOP regression with 2.6.39-rc5

[Frederic Weisbecker <fweisbec@gmail.com>]

On Sat, May 07, 2011 at 10:59:36AM -0700, Arjan van de Ven wrote:
> On 5/7/2011 10:20 AM, Steven Rostedt wrote:
> >On Sat, 2011-05-07 at 16:44 +0200, Ingo Molnar wrote:
> >>* Steven Rostedt<rostedt@goodmis.org>  wrote:
> >>
> >>>2) we separate perf from ftrace and keep the "stable" ABI for perf, and let
> >>>ftrace advance into a more efficient tracer.
> >>The thing is, ftrace is still largely separated from perf, and this is why this
> >>regression came in: a random tracing 'cleanup' churn was done to 'tracing'
> >>which broke PowerTop.
> >>
> >>Look at the commit itself:
> >>
> >>   e6e1e2593592: tracing: Remove lock_depth from event entry
> >>
> >>Clearly you didnt even *realize* that there's a whole tooling world behind this
> >>mechanism ...
> >Note, I discussed this change with Frederic and he totally agreed with
> >me on removing it. In fact, we are in discussions about getting rid of
> >pid, preempt-count, and irq flags as well. But according to your logic,
> >that is a no go. I guess Frederic also does not *realize* there's a
> >whole tooling world behind this mechanism too.
> 
> btw if you remove some of these, how is userland supposed to find
> out if an event happened in irq context?

You can use the irq events for that.