On Thu, 10 Aug 2006 15:25:00 +0000 "Jeffrey D. Smith" <[EMAIL PROTECTED]> wrote:
:>-----Original Message----- :>From: "Binyamin Dissen" <[EMAIL PROTECTED]> :>Sent: 8/10/2006 6:00 AM :>To: "IBM-MAIN@BAMA.UA.EDU" <IBM-MAIN@BAMA.UA.EDU> :>Subject: Re: Doing a LM of two words on a double word bounday - is it serialized? :>On Fri, 28 Jul 2006 15:11:22 -0400 Walt Farrell <[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. :>:>Note, for LM, that this does depend on the doubleword alignment of the :>:>storage operand. Consider the following example, where the operand is :>:>initially word-aligned: :>:> 986A 4234 LM 6,10,X'234'(4) :>:>In this case, register 6 would be loaded in a word-concurrent manner, :>:>and registers 7-8 and 9-10 would be loaded in a doubleword-concurrent :>:>manner. :>A follow-up, if I may. :> DWORD DC 0D'0' :> DC F'1,2' :>Cpu0: R14 = F'1,2' (64 bits) :> R15 = F'6,7' :> CSG R14,R15,DWORD :>Cpu1: :> R14 = F'99,2' :> R15 = F'99,4' :> CS R14,R15,DWORD+4 :>If the CSG and the CS are issued at the same time, are the results defined? :>Can both get a condition code of zero? :>Binyamin Dissen <[EMAIL PROTECTED]> :>================================================= :>Greetings, :>Unpredictably one of the CPU will get CC=0 and the other will get CC=1. Thus the same result as if both were CS or CSG, which would be a consistent result. My concern was that different size serializations might take place which could lead to an inconsistent result. :>It's never a good idea to use different sized compare-swap accesses, :>because the point of compare-swap is to maintain a consistent view :>of storage. I don't need that in my case. The LIFO linked-list removers need better serialization than the adders. PLO is an option but the section of code does not have enough free registers. -- Binyamin Dissen <[EMAIL PROTECTED]> http://www.dissensoftware.com Director, Dissen Software, Bar & Grill - Israel Should you use the mailblocks package and expect a response from me, you should preauthorize the dissensoftware.com domain. I very rarely bother responding to challenge/response systems, especially those from irresponsible companies. ---------------------------------------------------------------------- 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
