But why an option and burden the programmer with the responsibility, if the
HLASM can do it all automatically by itself? It is only required to flag the
literal with a new information, if it is used in a relative addressing context,
so that, additional to its length and type, it has to be halfword aligned. And
then, when sorting the literal pool (which is done anyway), this new flag has to
be respected.
Kind regards
Bernd
B.D schrieb:
> On Mon, 23 Aug 2010 10:48:08 +0200 Bernd Oppolzer wrote:
>
> :>I believe that literals could be sorted by alignment requirement
> :>instead of length, this way minimizing the needed padding bytes.
> :>That is, multiples of 8 first, then multiples of 4, then multiples of
> :>2, then odd lenghts, that need to be aligned on even addresses
> :>because of relative addressing, then "true" odds, that dont have
> :>this special requirement.
>
> To me it would be a lot simpler to have an assembler option which specifies
> the minimum alignment for each literal. It would default as is (byte), but
> could be halfword/word/doubleword/quadword.
>
> Much easier than sorting them and interleaving the odd length non-LARL
> referenced byte aligned items between the odd length LARL referenced byte
> aligned items.
>
>
|