OK, LFI/IILF is probably best if length is known as assembly time.  As best
as I can recall, I never had that case.  I did however, do that to set up a
primitive GETMAIN with subpool and length.

sas

On Fri, Mar 6, 2026 at 4:33 PM Steve Smith <[email protected]> wrote:

> OILH Rx,X'xx00' is safe and efficient after the length is loaded in any
> normal way.  I use IILH sometimes when I know the length is <64K-1.  But
> there's no rational reason for that.
>
> LLILH Rx,X'xx00' is safe and efficient if you want to insert the length
> afterwards.  Only reason I can think of is if length is an unsigned
> halfword in memory.
>
> These don't allow for some dynamic specification of the pad character.
> I've never had that case, but if you do, then ICM Rx,B'1000',pad would be
> how I'd do that.
>
> sas
>

Reply via email to