This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Question about GO_IF_LEGITIMIZE_ADDRESS



>> >No, since reload doesn't know what Q or S really mean, so it can't
>> >fix them up if nothing matches.  That's why 'm', 'o', and 'p' are
>> >so special.
>> 
>> I was asking about the predicate (btst_mem_operand), not the constraint.
>
>The point remains: what should reload do if it doesn't match?

Perhaps its about time that a way existed to tell reload what to do.

Since we have 'm', 'o', 'p' that are special to reload, there should
be another one that indicates that reload should just pull the address
of the operand into a register to make it register indirect.  This
reload method would work for most(if not all) of the chips that gcc
currently supports. 

Any suggestions for which character to use for this new
constraint(perhaps '@')?

-- 
Peter Barada                                   Peter.Barada@motorola.com
Wizard                                         781-852-2768 (direct)
WaveMark Solutions(wholly owned by Motorola)   781-270-0193 (fax)


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]