------- Comment #8 from rguenth at gcc dot gnu dot org 2009-12-11 12:16 ------- I wonder if we should simply ignore -fno-exceptions/-fexceptions passed to lto1 and rely on the automated discovery.
The testcase btw works for me on i?86. The proposed patch would delete handling of OPT_fexceptions from register_user_option_p and add flag_exceptions = 0 to lto_init_options in case you can still reproduce the issue. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41756