shannon@sun.uucp (Bill Shannon) (09/11/85)
(Don't any of you people know how to change the title of an article to reflect its contents? I missed a lot of good information because I was ignoring all those "PDP11's vs. the micros" titles because the original subject was stupid.) > In a multiprocessor system designed for transparent operation and > which has the ability to allocate processors to tasks dynamicly, it > is possible that a task be re-started on a different processor than > the one that faulted. If there is any difference in the micro-state > between cpu revision levels, it could happen that the restart would > fail due to incompatible micro-state. This is also a problem on the beloved VAX. It's been a long time, so forgive me if the details are wrong, but... I believe when a VAX instruction is interrupted and FPD (first part done) is set, the registers that the instruction uses to do its operation may contain some internal state that can not necessarily be reloaded on a different type or rev level VAX. This caused problems for (a since aborted attempt at) VMS process migration in a cluster environment. Bill Shannon