Messages in this thread | | | Date | Thu, 12 Aug 1999 21:18:56 -0400 (EDT) | From | Chuck Lever <> | Subject | Re: clustering page-ins |
| |
On Thu, 12 Aug 1999, Dr. Michael Weller wrote: > I didn't try the already posted patches, but if it is not already the case > I'd suggest to make the behaviour configurable by madvice, you know: > > I was told somehow the origins of this function might be BSD, so it might > be worthwhile to support. I saw it in many unixes but it is usually > documented to have no effect at all.
someone (randy appleton?) is already looking at adding madvise(). i hope the work i'm doing doesn't make his work more difficult to implement.
FreeBSD has madvise() and adjusts its read-around depending on the application's suggested access pattern. it doesn't appear that it has any sophisticated guessing heuristics about whether an application is accessing in sequential or random order.
interestingly, though, it appears that FreeBSD's read() simply does clustered reads rather than any complicated read-ahead. but i could be wrong about that.
- Chuck Lever -- corporate: <chuckl@netscape.com> personal: <chucklever@netscape.net> or <cel@monkey.org>
The Linux Scalability project: http://www.citi.umich.edu/projects/linux-scalability/
- 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/
| |