> -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On > Behalf Of Edward Jaffe > Sent: Friday, August 07, 2009 6:51 PM > To: [email protected] > Subject: Re: z10 and overlapping/destructive moves > > Bill Fairchild wrote: > > I understand why you would not want to make the change global. But it > > might be that the other COBOL moves are executed much less frequently, > > in which case a global change might be acceptable. > > This will most-likely cause the program to run more slowly. Remember, a > cache line fetch from real memory is now 600 cycles. (And, that doesn't > even count the TLB miss you'll almost certainly have.) There's no reason > to pay such severe penalties to load in the cache line(s) containing the > 120-byte blank literal unless you're going to use it for many similar > operations one right after the other...or in a loop.
I had not thought of the cache-line problem. This particular MOVE is in a subroutine that is invoked multiple times per input record (> 10), which is likely why it is showing up as a spike in Strobe for a file of several million records. Well, I think it's worth a try anyway. I do expect I'll find you are correct, but there's nothing like a test to prove the point. Thanks for the tip. I'll report back on the results. Peter This message and any attachments are intended only for the use of the addressee and may contain information that is privileged and confidential. If the reader of the message is not the intended recipient or an authorized representative of the intended recipient, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail and delete the message and any attachments from your system. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
