From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH net-next] udp: be less conservative with sock rmem accounting Date: Sat, 03 Dec 2016 16:15:02 -0500 (EST) Message-ID: <20161203.161502.2062069811901046717.davem@davemloft.net> References: Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, brouer@redhat.com To: pabeni@redhat.com Return-path: Received: from shards.monkeyblade.net ([184.105.139.130]:37420 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750873AbcLCVPE (ORCPT ); Sat, 3 Dec 2016 16:15:04 -0500 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: From: Paolo Abeni Date: Fri, 2 Dec 2016 17:35:49 +0100 > Before commit 850cbaddb52d ("udp: use it's own memory accounting > schema"), the udp protocol allowed sk_rmem_alloc to grow beyond > the rcvbuf by the whole current packet's truesize. After said commit > we allow sk_rmem_alloc to exceed the rcvbuf only if the receive queue > is empty. As reported by Jesper this cause a performance regression > for some (small) values of rcvbuf. > > This commit is intended to fix the regression restoring the old > handling of the rcvbuf limit. > > Reported-by: Jesper Dangaard Brouer > Fixes: 850cbaddb52d ("udp: use it's own memory accounting schema") > Signed-off-by: Paolo Abeni Applied.