While it could be possible to output_set_got such that we can
individually annotate the instructions, it's simpler to simply
admit that all processors currently being manufactured do want
deep branch prediction.  At which point all of the complication
simply goes away.

        * config/i386/i386.h (X86_TUNE_DEEP_BRANCH_PREDICTION): Remove.
        (TARGET_DEEP_BRANCH_PREDICTION): Remove.
        * config/i386/i386.c: Don't include dwarf2out.h.
        (initial_ix86_tune_features): Remove X86_TUNE_DEEP_BRANCH_PREDICTION.
        (output_set_got): Don't test TARGET_DEEP_BRANCH_PREDICTION, delete
        all code dead thereafter.  Don't do dwarf2out_flush_queued_reg_saves.
        (ix86_expand_prologue): Set REG_CFA_FLUSH_QUEUE on set_got insn.
        (machopic_output_stub): Don't test TARGET_DEEP_BRANCH_PREDICTION.

From-SVN: r175730