Quote:

'd be interested to know the precise details of how you ensure that, but I guess it'd take too long!

See my PM... actually, you were right: There is a minimum, and that's because 12.5 % is 1/8 of 100 %, i.e. if you have 8 or more buffers per slab, then you're sure to limit the internal fragmentation...

_________________
Thomas, the kernel guy

"I don't have a frigging clue. I'm norwegian" -- Ole-Egil

All opinions expressed are my own and do not necessarily represent those of Hyperion Entertainment

OK, having looked at it in a lot more detail, I've realised that I was actually right in my original suspicion about how to calculate a family of suitable slab sizes (which keep fragmentation below 12.5%). Glad that I'm not completely thick :)

Last edited by ChrisH on 22-Dec-2005 at 06:06 PM.
Last edited by ChrisH on 22-Dec-2005 at 05:45 PM.
Last edited by ChrisH on 22-Dec-2005 at 05:21 PM.
Last edited by ChrisH on 22-Dec-2005 at 04:35 PM.
Last edited by ChrisH on 22-Dec-2005 at 04:16 PM.
Last edited by ChrisH on 22-Dec-2005 at 04:14 PM.
Last edited by ChrisH on 22-Dec-2005 at 04:13 PM.

_________________
Author of the PortablE programming language.
It is pitch black. You are likely to be eaten by a grue...