------- Comment #13 from dominiq at lps dot ens dot fr 2010-06-29 22:40 ------- Created an attachment (id=21042) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21042&action=view) preprocessed tree.i
The patch in http://gcc.gnu.org/ml/gcc-patches/2010-06/msg03011.html fixes the reported ICE but causes a new one with tree.c: [macbook] build_w/gcc% /opt/gcc/build_w/prev-gcc/xgcc -B/opt/gcc/build_w/prev-gcc/ -c -O2 -DIN_GCC -DHAVE_CONFIG_H -I. -I. -I../../work/gcc -I../../work/gcc/. -I../../work/include -I../../work/libcpp/include -I/opt/sw64/include ../../work/gcc/tree.c ../../work/gcc/tree.c: In function 'build_real': ../../work/gcc/tree.c:1441:1: internal compiler error: Segmentation fault The backtrace for tree.i is Reason: KERN_INVALID_ADDRESS at address: 0x000000000000000c 0x0000000100ad23c0 in useless_type_conversion_p () (gdb) bt #0 0x0000000100ad23c0 in useless_type_conversion_p () #1 0x0000000100a0035a in may_propagate_copy () #2 0x0000000100a8863d in replace_phi_args_in () #3 0x0000000100a889cc in substitute_and_fold () #4 0x0000000100b3c2e1 in vrp_finalize () #5 0x0000000100b3c440 in execute_vrp () #6 0x00000001007e24fa in execute_one_pass () #7 0x00000001007e28eb in execute_pass_list () #8 0x00000001007e2909 in execute_pass_list () #9 0x000000010099668b in tree_rest_of_compilation () #10 0x0000000100c2b2f6 in cgraph_expand_function () #11 0x0000000100c2b58d in cgraph_expand_all_functions () #12 0x0000000100c2bc5f in cgraph_optimize () #13 0x0000000100c292e9 in cgraph_finalize_compilation_unit () #14 0x000000010002ef94 in c_write_global_declarations () #15 0x00000001008fbeb9 in compile_file () #16 0x00000001008fe96b in do_compile () #17 0x00000001008fea35 in toplev_main () #18 0x000000010010d450 in main () -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44699
