Hadn't tried it that way. Thanks for the head's up! The reason I'm trying to avoid ICMY is the memory fetch although it is probably not a big hit.
In most cases, this is used to "clear" memory to a specific value so the length is 0. There are a few cases where data is being moved but with a very few exceptions, the source length and destination length are the same with the pad byte set to NULL. *Mark Hammack* [email protected] On Thu, Mar 5, 2026 at 2:42 PM Jonathan Scott < [email protected]> wrote: > It's not clear whether you're assuming 31-bit or 64-bit mode. If it's > only 31-bit, then IILF (also known as LFI if you have a recent maintenance > level of HLASM) would probably be best as a hardware solution, and you can > write the immediate value as follows if you like: > > LFI R15,C' '*X'1000000' > > Jonathan Scott > > -----Original Message----- > From: IBM Mainframe Assembler List <[email protected]> On > Behalf Of Mark Hammack > Sent: 05 March 2026 20:32 > To: [email protected] > Subject: Loading the pad byte for MVCL > > So, I was wondering if anyone had a "better" way. > > I've done: > > IC R15,C' ' > SLL R15,24 > > and > > LGFI R15,X'40000000' > > and > > LY R15,PADCHAR > ... > PADCHAR DC C' ',X'000000' > > What I'd like is to do something like the LGFI (or LLILH) but using the > more "self documenting" PADCHAR format. > > I'm probably just too picky. I could probably write a SETPAD macro and > use SLL, just haven't tried that yet. > > > *Mark Hammack* > [email protected] >
