> This idiom: "if (GET_CODE (body) == ASM_INPUT || asm_noperands (body) > > >= 0)" appears in multiple places. There's even one place where the > > idiom above is used in reverse (making the GET_CODE... check > redundant). A few more places to the equivalent by checking > extract_asm_operands != NULL.
I think that the last point is not clear: asm_noperands can return -1 and yet extract_asm_operands has returned non-NULL. And, at least in some cases, I think that the right predicate is extract_asm_operands. In fact, I wonder whether in most cases the right combined predicate would be: GET_CODE (body) == ASM_INPUT || extract_asm_operands (body) != NULL and asm_noperands only used when you really care about the operands. > It made sense to me, at least, to replace those equivalent checks with > a new function: insn_with_asm_operands_p(). The first hunk for config/ia64/ia64.c looks incorrect. -- Eric Botcazou