|
I actually agree. We're going to encounter more relaxed ordering in multiprocessors. The question is, what do the hardware designers consider conservative? Forcing an interlock at both the beginning and finish of a locked part seems to be fairly conservative to me, but I clearly am not imaginative enough. The Pro manuals go into excruciating element in describing the caches and what keeps them coherent however don’t seem to care to say something detailed about execution or learn ordering. The reality is that we haven't any means of figuring out whether we’re conservative enough. 0 end result, and that the Pentium Pro simply had larger pipelines and write queues that exposed the behavior more typically. The Intel architect additionally wrote: Loosely talking, this implies the ordering of events originating from anyone processor in the system, as observed by other processors, is always the identical. Nevertheless, totally different observers are allowed to disagree on the interleaving of events from two or extra processors.
Look at my web site :: Memory Wave Protocol |
|