<snip>Could you please concede that VD is valid HLASM syntax as per:Can you 
look at 
https://www.ibm.com/docs/en/zos/2.1.0?topic=instruction-subfield-3-type-extension
</snip>
Given the evidence, I do concede.
I'm quite sure that the initial implementation of HLASM for 8-byte relocatable 
entities did not support VD(xxx), and no one ever bothered informing the troops 
of the new function when it was provided. 
It is true that I never worked on the assembler or binder, but I sure used them 
both a lot and I know what we had to code (and I know what the PL/X compiler 
implemented when the support was introduced).
<snip>when callers AMODE64 is bit 63 of R14. 
</snip>
bit 63 of R14 is *not* caller's AMODE 64. Caller's AMODE is in R14 only for 
AMODE 24 and AMODE 31 targets, and an AMODE 24/31 target with an AMODE 64 
invoker is not given the information that the invoker of the service was AMODE 
64. You can see that in Jim Mulder's examples showing 64->24 and 64->31.
These choices were designed to preserve compatibility for existing programs. 
Anything that would have required an existing program to return via BSM would 
have been incompatible.
Peter Relson






----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to