On 16/06/2015 9:23 PM, Charles Mills wrote:
Obviously a nit-pick and I knew that when I posted it. Hence the LOL. But a 
shop might write 100 million SMF records a day. These instructions will get 
exercised for most (IEFU83) of them. Even one cycle per tends to add up. I have 
had to learn the new way of thinking: instructions don't count; storage 
references count. An unneeded reference to storage is potentially very wasteful 
of cycles.

True. But it's a sample so hack away. The trouble with assembler is it doesn't have an optimizer. The programmer is the optimizer.

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

Reply via email to