On Thu, 24 Apr 2014 12:32:32 +0800, Timothy Sipples wrote: >I remember a school of thought that "thunking" might not have been a good >idea because it encouraged developers to do what they do best: nothing. :-)
I never heard of "thunking" before. It sounds to me like a derogatory term for calling routines that run in different addressing modes. Is that what is intended? >That is, developers would use thunking as a means to achieve the bare >minimum for a too-timid leap to the next addressing width when what they >really should have done (for their customers' and end users' benefit) is >brought their whole product/program over to the new width -- or at least >continued their efforts in subsequent releases to complete the move to the >next width. I don't know if I'd agree with that school of thought, but I >remember it, and there's a certain logic to it. I hope that you are not suggesting that z/OS should not have been released until everything ran AMODE 64. We might still be waiting. >I wonder if thunking is an area a third party and/or the open source >community would find interesting -- a standard set of callable thunking >services. And I wonder if it'd be technically possible as such, and how >difficult. I don't see what you mean. It is not difficult for any program to call any other, regardless of addressing mode. -- Tom Marchant ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
