HLASM correctly refers to POPs since it is not describing each instruction.  
POPs uses 'treated as' for describing different operands.  The term 'treated 
as' is used frequently  with different instructions and appears to accurately 
describe the operand.  How is it ambiguous?  The limits of a 16-bit signed 
binary integer are well known aren't they?  Not sure why you have examples of 
unsigned.

Sorry for getting defensive but I don't see a problem here,
...chris.
-----Original Message-----
From: IBM Mainframe Assembler List [mailto:[email protected]] On 
Behalf Of Paul Gilmartin
Sent: February-23-17 11:05 AM
To: [email protected]
Subject: Re: HLASM anomaly

On 2017-02-23, at 11:16, Paul Gilmartin wrote:
> 
> What are the limits on the operand of AHI?  What of:
> 
> A        EQU -32768
>         DC  Y(A)
>         AHI R1,A    OK, I believe.
> 
> B        EQU  32767
>         DC  Y(B)
>         AHI R1,B    OK, I believe.
> 
> c        EQU  65535
>         DC  Y(C)
>         AHI R1,C    OK ???
> 
> Is the operand of AHI treated modulo 2^16, or ?
>  
So, I RTFM and learned little:
        * HLASM
        * HLASM Language Reference
        * Machine instruction statements
        * Operand entries
        * Immediate data
HLASM Language Reference
SC26-4940-06
    ...
    Immediate data must be specified as absolute expressions whose range
    of values depends on the machine instruction for which the data is
   required. The immediate data is assembled into its binary representation.

"depends on the machine instruction".  So I need to read the PoOps, which says, 
for AHI:
    ... The second operand is two bytes in length and is treated
    as a 16-bit signed binary integer.  ...

Does "treated as" mean it's treated modulo size of field?  Or ...?
That seems to be a HLASM question, not a zSeries question, so I guess HLASM 
deserves an RCF; it shouldn't pass the responsibility (to PoOps?), which offers 
no clarification.  Are the limits of a halfword operand:
(-32768,32767)
(0,65535)
(-32768,65535)
Other (specify)?

Lore is not documentation.

-- gil

Reply via email to