On Thu, 2 Jul 2015 17:08:39 -0700, John Ehrman <ehr...@us.ibm.com> wrote:
>Local labels in HLASM would require identifying the scope of the labels: >do labels generated in open code or in a macro correspond in any way to >labels generated at the same, higher, or lower levels of macro expansions? I think to be practical they would have to be scoped such that local labels in open code were only affected by other open code labels (probably including the label on a macro invocation), and local labels within a macro were affected only by labels generated by that macro (not by inner or outer macros). Otherwise there would be no way of being sure when a change to some macro might invalidate a local label. I'm trying to remember exactly how they were used in the S/360 assembler the guys back at Caltech wrote in the '60s. I don't think they had this problem because their "macro" processing via was so very different. I think I gave you a copy of the reference manual for that assembler many years ago; it might say more about that. -- Walt
