Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!aplcen!aplpy.jhuapl.edu!cfw From: cfw@aplpy.jhuapl.edu (Chuck Waltrip) Newsgroups: comp.sys.next Subject: Re: parity memory (on a NeXT) costs a cycle Keywords: parity 9 Message-ID: <1990Dec27.193942.8303@aplcen.apl.jhu.edu> Date: 27 Dec 90 19:39:42 GMT References: <1038@tokio.cs.utexas.edu> Sender: news@aplcen.apl.jhu.edu (USENET News System) Organization: Johns Hopkins University Applied Physics Lab Lines: 23 In article <1038@tokio.cs.utexas.edu> garnett@cs.utexas.edu (John William Garnett) writes: >I'm not sure if the cost of using parity memory on a NeXT was >ever resolved. I just submitted a question to the local campus NeXT I'm not sure that there was ever an answer to the question about what NeXT OS does when a parity error is encountered...if anyone knows, I'd sure be interested in the answer (sorry if it was posted and I missed it). >representative and was informed that memory accesses to parity >memory (on a NeXT) cost an extra cycle (over that required by >non-parity memory). If this issue is important to you, you should I don't know the answer either but I believe this cost may not be terribly significant if you're running with a high cache hit rate. >probably confirm (or hopefully invalidate) this news with NeXT before >purchasing parity memory. > >-- >John Garnett > University of Texas at Austin >garnett@cs.utexas.edu Department of Computer Science > Austin, Texas c.f.waltrip Opinions expressed are my own.