This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Object size checking (stage 2 project)


On Tue, May 10, 2005 at 05:47:13PM -0400, Diego Novillo wrote:
> On Mon, May 02, 2005 at 01:40:01PM -0400, Jakub Jelinek wrote:
> 
> > +/* Simple pass to optimize all __builtin_object_size () builtins.  */
> > +
> > +static void
> > +compute_object_sizes (void)
> > +{
> > +  basic_block bb;
> > +  FOR_EACH_BB (bb)
> > +    {
> > +      block_stmt_iterator i;
> > +      for (i = bsi_start (bb); !bsi_end_p (i); bsi_next (&i))
> > +	{
> >
> I don't like this part.  __builtin_object_size is an almost
> perfect fit for using the SSA propagator.  You are duplicating
> much of the propagation functionality in ssa_propagate.  Any
> particular reason why not to use the propagator?
> 
I have spoken with Jakub about this offline.  While I would like
to see this propagation done using the propagation engine, I
realize that Jakub implemented this before the propagator existed
and it may or it may not be a trivial change.

Given that the stage2 deadline is fast approaching and neither
Jakub nor myself will likely have time to work on this any time
soon, I withdraw my objection.

Jakub has agreed to eventually rewrite the propagation using the
engine.  Since that would be just a cleanup, the behaviour of the
patch would not be affected.  This cleanup can be delayed until
the next stage1.

If the builtin bits are approved, then this patch can go in.


Diego.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]