On Fri, 28 Jul 2006 12:34:33 -0700, Edward Jaffe <[EMAIL PROTECTED]> wrote:
>... >> Your interpretation is correct, Ed. A CPU fetching a doubleword in a >> concurrent manner (as LM is defined to do when fetching from a >> doubleword boundary) is protected once the fetch starts from any >> updates by other CPUs. > >Thanks for confirming this, Walt! It's comforting to know I don't have a >lot of "broken" software due to a misreading of the rules of the >architecture! ;-) >... This thread is one of the real gems that occassionally come up on IBM-Main. I don't do programming that needs this kind of information so I have not paid any attention to this kind of topic for years. But I do have to do diagnosis where this could be important - where contents of registers don't match storage from where they were loaded. I would never have thought of the concurrency issue; I would just been baffled. And even if I never run into this, the topic was interesting. Thanks to all those that contributed. And thanks for coming to a definitive answer. Pat O'Keefe ---------------------------------------------------------------------- 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
