From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Angelo Dell'Aera" Subject: Re: [PATCH] (1/3) tcp - choose congestion algorithm at initialization Date: Mon, 4 Oct 2004 14:42:32 +0200 Sender: netdev-bounce@oss.sgi.com Message-ID: <20041004144232.2bfa5a9e.buffer@olografix.org> References: <20040927111834.48c7baab@zqx3.pdx.osdl.net> <20040927121610.68f942a4.davem@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: shemminger@osdl.org, netdev@oss.sgi.com Return-path: To: "David S. Miller" In-Reply-To: <20040927121610.68f942a4.davem@redhat.com> Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 27 Sep 2004 12:16:10 -0700 "David S. Miller" wrote: >> The choice of congestion algorithm needs to be made when connection >> is setup to avoid problems when the sysctl values change later and the >> necessary data hasn't been collected. >Looks great, applied. I completely agree with you. Really a cool work. Just a proposal now. If the choice is made when the connection is established we do not need a struct for every supported congestion control algorithm in the struct tcp_opt any more. We could use for example a generic void * and then refer it properly after Stephen's controls such as tcp_is_vegas(), tcp_is_bic(), etc. This should slim the struct tcp_opt because if a connection is using Westwood its tcpbic and vegas structs will never be used. Moreover, I think it would make the code more readable. Regards. - -- Angelo Dell'Aera 'buffer' Antifork Research, Inc. http://buffer.antifork.org Metro Olografix PGP information in e-mail header -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.6 (GNU/Linux) iD8DBQFBYUU4pONIzxnBXKIRArT5AJ0Ztak7xFH9DIEd24kRxCvMMxHrBACghKmE e0CHh6gMpJaW02Y5HVA6u78= =av9u -----END PGP SIGNATURE-----