Subject: Re: buffer cache memory management revision
To: Paul Kranenburg <firstname.lastname@example.org>
From: Matt Thomas <email@example.com>
Date: 12/23/2003 13:14:12
At 11:47 AM 12/23/2003, Paul Kranenburg wrote:
> > i've been running tests with this patch for a few hours on my
> > U5/270 without incident. works for me!
>I'm still trying to decide what the best compromise is in order to
>avoid excessive memory fragmentation (a danger with any pool based
>allocation scheme) and still allow a fair degree of freedom for
>configuring the buffer cache sizes. I'm thinking of using MAXBSIZE
>pool page sizes for all buffer pools on machines with lots of
>physical memory, while being more economical on low memory machines.
pool(9) doesn't support having items that are bigger than a page.
(since on alpha, mips, powerpc pool are mapped 1:1 with physical
addresses so virutally contiguous pages probably won't be physically
Matt Thomas email: firstname.lastname@example.org
3am Software Foundry www: http://3am-software.com/bio/matt/
Cupertino, CA disclaimer: I avow all knowledge of this message.