http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47209
Zdenek Sojka <zsojka at seznam dot cz> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|ICE: SIGSEGV in |ICE: SIGSEGV in |should_emit_struct_debug |should_emit_struct_debug |(dwarf2out.c:627) with |(dwarf2out.c:627) with |-femit-struct-debug-baseonl |-f{no-,}emit-struct-debug-{ |y -g |baseonly,reduced} -g --- Comment #3 from Zdenek Sojka <zsojka at seznam dot cz> 2011-01-08 20:45:38 UTC --- It fails the same way with -f{no-,}emit-struct-debug-reduced and -f{no-,}emit-struct-debug-baseonly , with very similiar crash location and the same backtrace: $ gcc -femit-struct-debug-reduced -g pr47209.C ==15753== Invalid read of size 2 ==15753== at 0x78AC90: should_emit_struct_debug (dwarf2out.c:624) pr47209.C:6:8: internal compiler error: Segmentation fault Please submit a full bug report, with preprocessed source if appropriate. See <http://gcc.gnu.org/bugs.html> for instructions. $ gcc -fno-emit-struct-debug-reduced -g pr47209.C ==15579== Invalid read of size 2 ==15579== at 0x78AC90: should_emit_struct_debug (dwarf2out.c:624) pr47209.C:6:8: internal compiler error: Segmentation fault Please submit a full bug report, with preprocessed source if appropriate. See <http://gcc.gnu.org/bugs.html> for instructions. $ gcc -femit-struct-debug-baseonly -g pr47209.C ==5087== Invalid read of size 2 ==5087== at 0x78AC0E: should_emit_struct_debug (dwarf2out.c:627) pr47209.C:6:8: internal compiler error: Segmentation fault Please submit a full bug report, with preprocessed source if appropriate. See <http://gcc.gnu.org/bugs.html> for instructions. $ gcc -fno-emit-struct-debug-baseonly -g pr47209.C ==13603== Invalid read of size 2 ==13603== at 0x78AC0E: should_emit_struct_debug (dwarf2out.c:627) pr47209.C:6:8: internal compiler error: Segmentation fault Please submit a full bug report, with preprocessed source if appropriate. See <http://gcc.gnu.org/bugs.html> for instructions. (In reply to comment #1) > I don't know if the code is valid or invalid (I have no clue what is the > meaning of "base::foo;"). According to people ##C++ at irc.freenode.org, it is equivalent to "using base::foo;", in some pre-standard notation (if I understood that correctly).