--7mxbaLlpDEyR1+x6
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Nov 15, 2005 at 12:08:17AM +0100, haad wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>=20
> Hi list I have some questions I don't know if this is correct list for
> my questions.
> I look at new freebsd 6 features and found that they implement
> something  like
>=20
> 1)
> ibm_acpi. Acpi functions for Ibm notebooks.
> can we port it to our ACPI code or not ??

No idea.

> 3)
> We still use Giant Lock ??

Yes.

> 2)
> "Fine-grained locking to allow much of the VFS stack to run without the
> Giant lock has been added."
> Sometime ago i have seen some benchmarks with
> Linux & freebsd and Netbsd had very poor performance with SMP machine.
> Our SMP performance was worst than on single processor machine.

Depends on what you're doing. If your application spends most (almost=20
all?) of its time in userland, then our SMP will be useful.=20

If however you are spending a lot of time in the kernel, then yes, our SMP=
=20
isn't very helpful.

Take care,

Bill

--7mxbaLlpDEyR1+x6
Content-Type: application/pgp-signature
Content-Disposition: inline

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

iD8DBQFDeS8DWz+3JHUci9cRAsHIAJ90qGYGXQrsszbX2SmSutuuDXnziwCfW1XT
Q9zahEc4BfxhAsKC+eC30+U=
=9XwX
-----END PGP SIGNATURE-----

--7mxbaLlpDEyR1+x6--