| Message ID | 20101229110434.GB724@alvy.suse.cz |
|---|---|
| State | New |
| Headers | show |
On Wed, Dec 29, 2010 at 12:04 PM, Martin Jambor <mjambor@suse.cz> wrote: > Hi, > > the section type passed to lto_free_section_data in > input_cgraph_opt_section seems like it was copied from ipa-prop.c but > the section type was not changed afterwards as it should have been. > (If it is actually correct, it probably deserves a comment what is > going on.) At least there is no other trace in > LTO_section_jump_functions in lto-cgraph.c. > > I'm currently bootstrapping and testing the following, OK if it > passes? IMHO obvious (the parameter isn't used). Richard. > Thanks, > > Martin > > > 2010-12-29 Martin Jambor <mjambor@suse.cz> > > * lto-cgraph.c (input_cgraph_opt_section): Use the correct section type. > > Index: icln/gcc/lto-cgraph.c > =================================================================== > --- icln.orig/gcc/lto-cgraph.c 2010-12-29 11:32:14.000000000 +0100 > +++ icln/gcc/lto-cgraph.c 2010-12-29 11:32:18.000000000 +0100 > @@ -1797,7 +1797,7 @@ input_cgraph_opt_section (struct lto_fil > input_node_opt_summary (VEC_index (cgraph_node_ptr, nodes, ref), > &ib_main, data_in); > } > - lto_free_section_data (file_data, LTO_section_jump_functions, NULL, data, > + lto_free_section_data (file_data, LTO_section_cgraph_opt_sum, NULL, data, > len); > lto_data_in_delete (data_in); > } >
Index: icln/gcc/lto-cgraph.c =================================================================== --- icln.orig/gcc/lto-cgraph.c 2010-12-29 11:32:14.000000000 +0100 +++ icln/gcc/lto-cgraph.c 2010-12-29 11:32:18.000000000 +0100 @@ -1797,7 +1797,7 @@ input_cgraph_opt_section (struct lto_fil input_node_opt_summary (VEC_index (cgraph_node_ptr, nodes, ref), &ib_main, data_in); } - lto_free_section_data (file_data, LTO_section_jump_functions, NULL, data, + lto_free_section_data (file_data, LTO_section_cgraph_opt_sum, NULL, data, len); lto_data_in_delete (data_in); }