Subject: Re: mapping shared memory at a fixed address
To: Rahul Kulkarni <crypto_rahul@yahoo.com>
From: Bill Studenmund <wrstuden@netbsd.org>
List: tech-kern
Date: 02/08/2005 15:03:59
--cNdxnHkX5QqsyA0e
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Feb 08, 2005 at 02:28:47PM -0800, Rahul Kulkarni wrote:
> On a potentially large system with a huge code base,
> this cannot work, due to a lot of duplication in the
> text and if if you fork a lot of children then the
> text  sizes will be a concern..=20

If you have something you want usable in a huge code base, I think you're
much more likely to run into issues with the address range not being
available. You may be lucky, and something around 2 GB or so in the VA=20
will be fine. Or something around 1 GB maybe. Depends on how big your=20
programs and libs are and how big your heap usage is. Also, I think we now=
=20
have a top-down heap, so the middle is open.

Also, I assume you're running on an ILP32 system (x86, so on). If you're=20
on an LP64, you will probably have no problem. :-)

Look at X and other things that already do shared-memory transport.

Take care,

Bill

--cNdxnHkX5QqsyA0e
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)

iD8DBQFCCUVfWz+3JHUci9cRAoaaAJ9JEfoDMR7ikHmf/vyZcLdDqr+qNACgjqNw
ILs6wBdWxB4fC/nN5dQKXQM=
=soK1
-----END PGP SIGNATURE-----

--cNdxnHkX5QqsyA0e--