This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug c++/50614] [4.7 Regression] [C++0x] ICE: tree check: expected field_decl, have identifier_node in component_ref_field_offset, at expr.c:6697 with -fcompare-debug and a non-static initializer
- From: "jakub at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Wed, 05 Oct 2011 12:10:15 +0000
- Subject: [Bug c++/50614] [4.7 Regression] [C++0x] ICE: tree check: expected field_decl, have identifier_node in component_ref_field_offset, at expr.c:6697 with -fcompare-debug and a non-static initializer
- Auto-submitted: auto-generated
- References: <bug-50614-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50614
--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-10-05 12:10:15 UTC ---
-fdump-final-insns calls dump_enumerated_decls and that ICEs on the C++
specific trees (ARROW_EXPR etc.) left over in DECL_INITIAL of the FIELD_DECL by
the NSDMI support.
I wonder if the DECL_INITIAL of the FIELD_DECLs shouldn't be cleared during
genericization.