http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56242
--- Comment #10 from Jeffrey A. Law <law at redhat dot com> 2013-02-20 19:47:53 UTC --- My first thought WRT the actual patch itself is why not put the logic to fixup the insn chain for SEQUENCEs at the end of reorder_insns_nobb? That would avoid similar problems for any other code that used this function after reorg.c had started. It also seems to me there are 4 cases that need to be fixed. FROM is a SEQUENCE Need to fix PREV_INSN of the first insn in the SEQUENCE TO is a SEQUENCE Need to fix NEXT_INSN of the last insn in the SEQUENCE PREV_INSN (FROM) is a sequence Need to fix NEXT_INSN of the last INSN in the SEQUENCE NEXT_INSN (TO) is a sequence Need to fix PREV_INSN of the first INSN in the SEQUENCE Am I missing something here?
