From mboxrd@z Thu Jan 1 00:00:00 1970 From: Florian Westphal Subject: [PATCH v2 next 0/2] tcp: make undo_cwnd mandatory for congestion modules Date: Mon, 21 Nov 2016 14:18:36 +0100 Message-ID: <1479734318-30607-1-git-send-email-fw@strlen.de> To: Return-path: Received: from Chamillionaire.breakpoint.cc ([146.0.238.67]:38102 "EHLO Chamillionaire.breakpoint.cc" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754179AbcKUNSY (ORCPT ); Mon, 21 Nov 2016 08:18:24 -0500 Sender: netdev-owner@vger.kernel.org List-ID: highspeed, illinois, scalable, veno and yeah congestion control algorithms don't provide a 'cwnd_undo' function. This makes the stack default to a 'reno undo' which doubles cwnd. However, the ssthresh implementation of these algorithms do not halve the slowstart threshold. This causes similar issue as the one fixed for dctcp in ce6dd23329b1e ("dctcp: avoid bogus doubling of cwnd after loss"). In light of this it seems better to remove the fallback and make undo_cwnd mandatory. First patch fixes those spots where reno undo seems incorrect by providing .cwnd_undo functions, second patch removes the fallback.