Messages in this thread | | | Subject | Re: Threads in linux, contd | Date | Fri, 3 Sep 1999 10:56:45 +0100 (BST) | From | Alan Cox <> |
| |
> As pointed by Mark Hans to me ia32 has 32byte cache lines. Thus if the > above example gets compiled so that variables A and B are further than 32
You also care about the L2 cache line and the amount of data that has to be bounced across CPU's. There are specific algorithms for minimising this kind of thing when solving many classic problem types like finite element work. Even on a PII it is expensive to bounce data between CPUs.
You can reference the same data read-only on multiple CPU's relatively cheaply, but a writer and one or more other reads costs. Multiple writers costs a lot.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |