Re: [PATCH 4/6] perf kmem: Analyze page allocator events also

[Namhyung Kim <namhyung@kernel.org>]

Hi Jiri,

On Thu, Mar 12, 2015 at 12:01:27PM +0100, Jiri Olsa wrote:
> On Thu, Mar 12, 2015 at 04:32:49PM +0900, Namhyung Kim wrote:
> 
> SNIP
> 
> > +static int parse_slab_opt(const struct option *opt __maybe_unused,
> > +			  const char *arg __maybe_unused,
> > +			  int unset __maybe_unused)
> > +{
> > +	kmem_slab = (kmem_page + 1);
> > +	return 0;
> > +}
> > +
> > +static int parse_page_opt(const struct option *opt __maybe_unused,
> > +			  const char *arg __maybe_unused,
> > +			  int unset __maybe_unused)
> > +{
> > +	kmem_page = (kmem_slab + 1);
> > +	return 0;
> > +}
> 
> hum, just curious.. why not just assign 1, I'm missing the magic ;-)

Well, basically I just followed the existing convention. :)

Anyway it controls the effect of -s/--sort option.  The single -s
option will be used to set sort keys for both of --alloc and --caller
stat.  So it needs to known currently selected mode at option parsing
time.  I extended it to page/slab selection also.

For example,

  # perf kmem --page --alloc -s bytes --slab --caller -s hit

The first -s option will set sort keys for page/alloc stat and the
second -s option will set sort keys for slab/caller stat.  So the -s
option should be preceded by at least one of the mode selection
options (i.e. --page/slab/alloc/caller).  I'll add it to the man page
also.

Thanks,
Namhyung