[m-dev.] for review: The first bit of deep profiling

[Thomas Conway]

On Mon, Sep 13, 1999 at 01:11:36PM EST, Fergus Henderson wrote:
> On 13-Sep-1999, Thomas Conway <conway at cs.mu.OZ.AU> wrote:
> > 
> > Index: compiler/handle_options.m
> > ===================================================================
> > RCS file: /home/staff/zs/imp/mercury/compiler/handle_options.m,v
> > retrieving revision 1.77
> > diff -u -r1.77 handle_options.m
> > --- handle_options.m	1999/08/13 01:42:59	1.77
> > +++ handle_options.m	1999/08/27 00:20:17
> > @@ -356,6 +356,15 @@
> >  		[]
> >  	),
> >  
> > +	% Deep profiling requires `procid' stack layouts
> > +	globals__io_lookup_bool_option(profile_deep, DeepProfiling),
> > +	(
> > +		{ DeepProfiling = yes },
> > +		globals__io_set_option(procid_stack_layout, bool(yes))
> > +	;
> > +		{ DeepProfiling = no }
> > +	),
> 
> I think you could write that more simply as
> 
> 	% Deep profiling requires `procid' stack layouts
> 	option_implies(profile_deep, procid_stack_layout, bool(yes)),
> 

Fixed.

> > --- options.m	1999/08/13 01:43:07	1.267
> > +++ options.m	1999/08/27 00:20:27
> > @@ -1099,6 +1105,9 @@
> >  	map__set(OptionTable0, profile_time, bool(no), OptionTable1),
> >  	map__set(OptionTable1, profile_calls, bool(yes), OptionTable2),
> >          map__set(OptionTable2, profile_memory, bool(yes), OptionTable).
> > +special_handler(deep_profiling, none, OptionTable0, ok(OptionTable)) :-
> > +	map__set(OptionTable0, profile_time, bool(no), OptionTable1),
> > +	map__set(OptionTable1, profile_deep, bool(yes), OptionTable).
> 
> Shouldn't the `bool(no)' here be `bool(yes)'?
> 
> Also, for consistency with the way the other --foo-profiling options are
> handled, and with the way that the --deep-profiling option is handled in
> scripts/parse_grade_options.sh-subr, you should set `profile_calls'
> and `profile_memory' to `bool(no)' here.  Likewise the special_handlers for
> memory_profiling, call_profiling, and time_profiling should set
> `profile_deep' to `bool(no)'.

Fixed.

> 
> (I notice that the new options are not yet documented, but I presume
> that will come as part of a later change?)
> 

Yes. Once there is enough for it to actually do something....

-- 
 Thomas Conway )O+     Every sword has two edges.
     Mercurian            <conway at cs.mu.oz.au>
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------