Subject: Re: KVA usage of old-style buffers
To: Thor Lancelot Simon <email@example.com>
From: Chuck Silvers <firstname.lastname@example.org>
Date: 10/02/2001 21:37:28
currently the UBC/UVM code uses MAXBSIZE, but that could be changed
easily enough. the somewhat bigger issue is just that the default
MAXBSIZE has been 64k ~forever, so those few people that use
filesystems with a 64k blocksize no doubt expect that those filesystems
will work with a default kernel. if we reduce MAXBSIZE to 8k,
then any filesystem with a blocksize larger than 8k won't work
until the kernel is rebuilt with a larger MAXBSIZE.
it'd be nice if we could have each buffer use only the KVA
that it actually needs, essentially getting rid of MAXBSIZE,
but that would be tricky to make work as time-efficiently as
our current buffer cache. the current code has extremely fast
KVA management, but at the cost of using more KVA.
On Tue, Oct 02, 2001 at 01:50:47PM -0400, Thor Lancelot Simon wrote:
> On Mon, Oct 01, 2001 at 07:52:58PM -0700, Chuck Silvers wrote:
> > putting an upper bound on the amount of KVA used for buffers
> > is a fine thing for now, for all ports, especially ones that
> > can have large amounts of RAM or that have less KVA to begin with.
> > eventually it would be good to enhance the buffer cache to be
> > more efficient in its use of KVA, but that isn't critical yet.
> Does UBC use MAXBSIZE or MAXPHYS to limit the size of transfers it
> hands to drivers? If it uses MAXPHYS, we could (correctly) reduce
> MAXBSIZE and thus address this problem.
> Thor Lancelot Simon email@example.com
> And now he couldn't remember when this passion had flown, leaving him so
> foolish and bewildered and astray: can any man?
> William Styron