Messages in this thread | | | From | "Stephen C. Tweedie" <> | Date | Fri, 9 Apr 1999 11:44:03 +0100 (BST) | Subject | Re: [PFC]: hash instrumentation |
| |
Hi,
On Thu, 8 Apr 1999 13:22:25 -0400 (EDT), Chuck Lever <cel@monkey.org> said:
> i'm discovering that a 13 bit hash mitigates the spikey size distribution > in the page hash *better* than the +offset change. although i've been > able to push the system into swap, i still haven't seen any degenerate > hash behavior that's as bad as the buffer cache's hash function.
Pushing it into swap is not sufficient: to trigger the swap cache you need to be actively using data which has previously been swapped out, which requires dirtying the pages, swapping them out and then touching them again for read access. In that special case, when the newly-swapped-back-in page still matches the copy on disk, the swap cache mechanism keeps both the in-memory and on-disk copies linked to each other.
> i'll have more as i test this further.
Don't forget that you can use the "mem=*m" option at boot time to run in a reduced memory configuration. That might make it a bit easier for you to trigger this stuff. :)
--Stephen
- 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/
| |