This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: pretty-ipa merge 2: master clones removal
> > Typo. Also the tree-ssa-structalias.c change broke gcc.dg/torture/ipa-pta-1.c,
> > I think the following will fix it, but I wonder what to do about
> > clones - just process
> > them?
> >
> > --- trunk/gcc/tree-ssa-structalias.c 2009/03/28 10:48:04 145174
> > +++ trunk/gcc/tree-ssa-structalias.c 2009/03/28 11:00:07 145175
> > @@ -5669,7 +5669,7 @@
> >
> > for (node = cgraph_nodes; node; node = node->next)
> > {
> > - if (!node->analyzed || cgraph_is_master_clone (node))
> > + if (!node->analyzed)
> > {
> > unsigned int varid;
> >
> > also remove the !node->analyzed check and process all nodes.
>
> Uh yes, it should work. I will test the change and commit if suceeds.
Bootstrapped/regtested x86_64-linux and comitted.
* tree-ssa-structalias.c (ipa_pta_execute): Fix bogus node->analyzed
test introduced by my previous patch.
Index: tree-ssa-structalias.c
===================================================================
*** tree-ssa-structalias.c (revision 145191)
--- tree-ssa-structalias.c (working copy)
*************** ipa_pta_execute (void)
*** 5669,5686 ****
for (node = cgraph_nodes; node; node = node->next)
{
! if (!node->analyzed)
! {
! unsigned int varid;
! varid = create_function_info_for (node->decl,
! cgraph_node_name (node));
! if (node->local.externally_visible)
! {
! varinfo_t fi = get_varinfo (varid);
! for (; fi; fi = fi->next)
! make_constraint_from (fi, anything_id);
! }
}
}
for (node = cgraph_nodes; node; node = node->next)
--- 5669,5683 ----
for (node = cgraph_nodes; node; node = node->next)
{
! unsigned int varid;
! varid = create_function_info_for (node->decl,
! cgraph_node_name (node));
! if (node->local.externally_visible)
! {
! varinfo_t fi = get_varinfo (varid);
! for (; fi; fi = fi->next)
! make_constraint_from (fi, anything_id);
}
}
for (node = cgraph_nodes; node; node = node->next)