[Bug middle-end/58723] [4.9 Regression] ICE in lto_output_edge, at lto-cgraph.c:300 for OpenMP's simd reduction

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58723

--- Comment #16 from Richard Biener <rguenth at gcc dot gnu.org> ---
Author: rguenth
Date: Wed Nov 27 15:18:23 2013
New Revision: 205447

URL: http://gcc.gnu.org/viewcvs?rev=205447&root=gcc&view=rev
Log:
2013-11-27  Richard Biener  <rguent...@suse.de>

    PR middle-end/58723
    * cgraphbuild.c (build_cgraph_edges): Do not build edges
    for internal calls.
    (rebuild_cgraph_edges): Likewise.
    * ipa-inline-analysis.c (estimate_function_body_sizes):
    Skip internal calls.
    * tree-inline.c (estimate_num_insns): Estimate size of internal
    calls as 0.
    (gimple_expand_calls_inline): Do not try inline-expanding
    internal calls.
    * lto-streamer-in.c (input_cfg): Stream loop safelen,
    force_vect and simduid.
    (input_struct_function_base): Stream has_force_vect_loops
    and has_simduid_loops.
    (input_function): Adjust.
    * lto-streamer-out.c (output_cfg): Stream loop safelen,
    force_vect and simduid.
    (output_struct_function_base): Stream has_force_vect_loops
    and has_simduid_loops.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/cgraphbuild.c
    trunk/gcc/ipa-inline-analysis.c
    trunk/gcc/lto-streamer-in.c
    trunk/gcc/lto-streamer-out.c
    trunk/gcc/tree-inline.c