Messages in this thread | | | Date | Tue, 7 Jul 1998 13:50:02 -0400 (EDT) | From | "Benjamin C.R. LaHaise" <> | Subject | Re: cp file /dev/zero <-> cache [was Re: increasing page size] |
| |
On Tue, 7 Jul 1998, Stephen C. Tweedie wrote:
[everyone has this thread fresh in their minds, right?] ... > Oh dear. More arbitrary rules in the VM --- just what we need! And > if we have more than one file trying to consume all of cache, what > then? Each one gets 50%, bang, the rule falls over. And the 25% > memory rule --- we expect that to work on 8MB machines as well as > 512MB?
Right. I'd rather see a multi-level lru like policy (ie on each cache hit it gets moved up one level in the cache, with the lru'd pages from a given level being bumped to the head of the next lower level) used for the page cache (it's kinda like the old dcache), whereby a data that is accessed only once ends up being a prime target for replacement. Of course, this can be done with the current aging scheme, it's just that the algorithms used are O(nasty) rather than O(1).
-ben
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu
| |