Messages in this thread |  | | Date | Tue, 10 Oct 2000 02:46:52 -0700 | From | "David S. Miller" <> | Subject | Re: Calling current() from interrupt context |
| |
Date: Tue, 10 Oct 2000 10:54:11 +0100 (BST) From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Having the stacks aligned also isnt good for the caches. If you are bored some time instrument the cache lines that wake_up() touches on a wake. Its very common to see most of them being on the same line because each kernel thread waiting on the same event (eg poll in apache) is at the same kernel and user stack depth.
Some of us actually have instrumented it :-) I added a coloring mechanism to the task/kstack allocator on sparc64. It made no noticable difference whatsoever to any real life measurement I attempted.
In my experiments I tried both L1 cache and L2 cache coloring.
Later, David S. Miller davem@redhat.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
|  |