Subject: Re: buffer cache memory management revision
To: None <>
From: Jason Thorpe <>
List: tech-kern
Date: 12/01/2003 13:43:11
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=US-ASCII; format=flowed

On Dec 1, 2003, at 12:17 PM, <> wrote:

> I for sure have an upper bound of it anyway because the kernel vm
> area may be very limited in size, so the only question is if the
> buffers must fight with other users of the same map.

Well, but your upper-bound should come from a run-time tunable, rather 
than a boot-time limit.  We should be striving to rid ourselves of as 
many boot-time limits as is feasible.

> Is the uarea the only other frequent user of the kernel_map today?
> In that case it may not be a big problem, but it still is lurking
> around down there.

U-area, bus_dma(9), and pools (on systems that don't have a 
direct-mapped memory segment that can be used).

         -- Jason R. Thorpe <>

content-type: application/pgp-signature; x-mac-type=70674453;
content-description: This is a digitally signed message part
content-disposition: inline; filename=PGP.sig
content-transfer-encoding: 7bit

Version: GnuPG v1.2.3 (Darwin)