__attribute__((leafify))

[Michael Matz]

Hi,

On Thu, 5 Jun 2003, Richard Guenther wrote:

> On Tue, 3 Jun 2003, Joseph S. Myers wrote:
>
> > On Tue, 3 Jun 2003, Richard Guenther wrote:
> >
> > > So what is this __attribute__((leafify))? Its a way to tell gcc to make
> > > the marked function a leaf of the cfg, i.e. to inline every call made in
> > > this function ignoring all inlining limits. A patch for mainline was
> > > posted at
> > > http://gcc.gnu.org/ml/gcc-patches/2003-05/msg01662.html
> >
> > Your patch doesn't have testcases.  They should include:
>
> Yup, I know. I'm not familiar with DejaGNU, so didnt bother to fiddle with
> it before at least one person is in agreement with the concept of
> __attribute__((leafify)). Sofar I have received no comments on the
> concept.

I think I said that I agree with the concept (so, start writing testcases
;-) ), and the fact that only the name of the option was discussed
indicates that there is no heavy opposition.  The patch in the mail above
also contains the hunk which removes the artificial "MAX_INLINE_INSNS *
128" limit.  I was opposed to that, and it's anyway unrelated to
(leafify).  There are ChangeLog entries missing, plus maybe a testcase or
two.  The patch itself without the above hunk looks good.  But I can't
approve it.  I had hoped that either a global one or Alexandre (as author
of tree-inline.c) commented, but that doesn't seem to be the case.


Ciao,
Michael.