Re: [PATCH 2/4] perf_counter: provide pagefault software events

[Peter Zijlstra <a.p.zijlstra@chello.nl>]

On Tue, 2009-03-10 at 15:59 +0100, Ingo Molnar wrote:
> * Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
> 
> > Index: linux-2.6/arch/powerpc/mm/fault.c
> > ===================================================================
> > --- linux-2.6.orig/arch/powerpc/mm/fault.c
> > +++ linux-2.6/arch/powerpc/mm/fault.c
> > @@ -29,6 +29,7 @@
> >  #include <linux/module.h>
> >  #include <linux/kprobes.h>
> >  #include <linux/kdebug.h>
> > +#include <linux/perf_counter.h>
> >  
> >  #include <asm/firmware.h>
> >  #include <asm/page.h>
> > @@ -170,6 +171,8 @@ int __kprobes do_page_fault(struct pt_re
> >  		die("Weird page fault", regs, SIGSEGV);
> >  	}
> >  
> > +	perf_swcount_event(PERF_COUNT_PAGE_FAULTS, 1, regs);
> 
> Wow, that's really nice - this way we can display a profile of 
> pagefault events in KernelTop - as if it was a real cycles or 
> other hw event based profile. Have you tried it out?

Yeah, but unsurprising all pagefaults I generated were in userspace, so
kerneltop stayed empty.

Found that perfstat went bokers though, forgot to handle the case where
there's no overflow interrupt.

/me goes poke at the code a bit more.