Re: [RFC][PATCH 2/8] perf, arch: Use early_initcall() for all arch pmu implementations

[Paul Mundt <lethal@linux-sh.org>]

On Thu, Nov 25, 2010 at 06:55:59PM +0100, Peter Zijlstra wrote:
> On Thu, 2010-11-25 at 11:25 +0100, Peter Zijlstra wrote:
> > 
> > Right, so hw perf init happens from (after this patch):
> > 
> >  arch_initcall: powerpc, arm, sh, mips
> >  early_initcall: x86, sparc, alpha
> > 
> > 
> > Now the problem is that the generic watchdog code (kernel/watchdog.c)
> > tries to create hw perf events, and that too runs from early_initcall.
> > 
> > So my question is, how do we go about curing this, because powerpc, arm,
> > sh and mips are too late and the rest depends on link order to work, not
> > really a nice situation.
> > 
> > There's two categories of solutions:
> >  - move the watchdog later, and
> >  - move the hw perf init earlier.
> > 
> > The former is undesired because we want the watchdog as early as
> > possible, the later needs new infrastructure (also, I don't know if the
> > arch implementations can actually run this early).
> > 
> > So do I create a perf_initcall() or is there another solution that
> > avoids things like calling the watchdog code from all arch init code? 
> 
> So the perf_event_init() site is _waaay_ to early to init stuff.. I
> guess I'll move it all to early_initcall() and I'll move the watchdog to
> an explicit call right after it.
> 
> Something like the below,.. now I guess the question to all of you is,
> can your arch pmu code cope with early_initcall() or does it need to be
> some other place?
> 
Any of the initcall levels are ok for SH at least. The only dependency we
have for the perf code is the clock framework, which comes up during
time_init().