--OgqxwSJOaUobr8KG
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Mon, Mar 14, 2005 at 08:25:58AM -0800, Matt Thomas wrote:
> >Also now that I think about it, since the SACK information needs to
> >be in time sequence, we should clear out the sack block list (receiver)
> >each time we send SACK information.
>=20
> That makes no sense.  The SACK processing for each direction is
> independent of the opposite direction.  The fact we are sending
> SACK blocks makes no difference to our side which is receiving
> SACK information.

Erhm, sorry, I guess I wasn't clear enough.  I did not mean that we
need to clear out the scoreboard each time we send a SACK, just that
tp->rcv_sack_num =3D 0; in the SACK send path is a good idea.

--=20
Kentaro A. Kurahone
SIGUSR1 Research and Development

--OgqxwSJOaUobr8KG
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (FreeBSD)
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=UJqo
-----END PGP SIGNATURE-----

--OgqxwSJOaUobr8KG--