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]

[PATCH][LTO] Fix most of the Fortran LTO failures

From: Richard Guenther <richard dot guenther at gmail dot com>
To: GCC Patches <gcc-patches at gcc dot gnu dot org>, Diego Novillo <dnovillo at google dot com>, fortran at gcc dot gnu dot org
Date: Fri, 10 Jul 2009 17:02:17 +0200
Subject: [PATCH][LTO] Fix most of the Fortran LTO failures

The fortran frontend ends up assigning locations before
BUILTIN_LOCATION to function declarations.
This causes the LTO reader to skip reading their bodies causing in
link-time failures.  When we
can safely use DECL_BUILT_IN instead of DECL_IS_BUILTIN we should do
so (I didn't found the
place where the Fortran frontend adjusts its initial location - maybe
the FE people can followup).

Built and gfortran-tested (with the fortran enabling patch and Pauls
latest whole-file patch).  Ok
"as obvious" for the branch?

Thanks,
Richard.

2009-07-10  Richard Guenther  <rguenther@suse.de>

        * lto.c (lto_materialize_function): Use DECL_BUILT_IN.
        (materialize_cgraph): Likewise.

Attachment: fix-lto-fortran-1
Description: Binary data

Follow-Ups:
- Re: [PATCH][LTO] Fix most of the Fortran LTO failures
  - From: Tom Tromey
- Re: [PATCH][LTO] Fix most of the Fortran LTO failures
  - From: Diego Novillo

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