This is great, IBM probably should have done it unless there are proprietary, 
undocumented oocodes.Chrles, why won't you consider publishing it on CBTThanksZA

Sent from Yahoo Mail on Android 
 
  On Mon, Jun 17, 2019 at 11:27 AM, Charles Mills<[email protected]> wrote:   +1

I have a table that I maintain to keep it all straight for myself.

Short version: ARCH() + 2 = marketing model number. ARCH(11) corresponds to 
z13. 

ARCH() - 4 = ZS number. ARCH(11) is equivalent to ZS-7.

I will leave the correspondence of ZS to marketing number as an exercise for 
the reader.

I don't know if ARCH() is for "all" of the compilers but I believe it is the 
same across C/C++, COBOL and PL/I.

Charles


-----Original Message-----
From: IBM Mainframe Assembler List [mailto:[email protected]] On 
Behalf Of John McKown
Sent: Monday, June 17, 2019 6:33 AM
To: [email protected]
Subject: idiot thought on OPTABLE vs. ARCH

I have just been looking around and noticed that most of the z/OS compilers
have an ARCH(n) option to specify the level of instructions to generate. As
best as I can tell, the number "n" means the same for all of the compilers.
So, I was thinking it would be nice if HLASM would bow to "peer pressure"
and adopt this parameter as well, in addition to the current OPTABLE. It
might even be nice if this would somehow affect the IBM macros similar to
the SYSSTATE ARCHLVL= . As an aside, I also wish that the ARCHLVL= numbers
were the same as the ARCH(n) numbers. But I recognise that this is
impossible at this late date. And I don't know which came first.  

Reply via email to