From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2992694AbcB0XeH (ORCPT ); Sat, 27 Feb 2016 18:34:07 -0500 Received: from mail-pf0-f171.google.com ([209.85.192.171]:36036 "EHLO mail-pf0-f171.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2992581AbcB0XeF (ORCPT ); Sat, 27 Feb 2016 18:34:05 -0500 Subject: Re: Softirq priority inversion from "softirq: reduce latencies" To: David Miller References: <56D1E8B6.6090003@hurleysoftware.com> <1456604037.648.29.camel@edumazet-ThinkPad-T530> <56D20733.1000409@hurleysoftware.com> <20160227.180403.2101360385050644823.davem@davemloft.net> Cc: eric.dumazet@gmail.com, edumazet@google.com, netdev@vger.kernel.org, linux-kernel@vger.kernel.org, gregkh@linuxfoundation.org, dmaengine@vger.kernel.org, john.ogness@linutronix.de, bigeasy@linutronix.de, akpm@linux-foundation.org From: Peter Hurley Message-ID: <56D23266.2080306@hurleysoftware.com> Date: Sat, 27 Feb 2016 15:33:58 -0800 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.5.1 MIME-Version: 1.0 In-Reply-To: <20160227.180403.2101360385050644823.davem@davemloft.net> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 02/27/2016 03:04 PM, David Miller wrote: > From: Peter Hurley > Date: Sat, 27 Feb 2016 12:29:39 -0800 > >> Not really. softirq raised from interrupt context will always execute >> on this cpu and not in ksoftirqd, unless load forces softirq loop abort. > > That guarantee never was specified. ?? Neither is running network socket servers at normal priority as if they're higher priority than softirq. > Or are you saying that by design, on a system under load, your UART > will not function properly? > > Surely you don't mean that. No, that's not what I mean. What I mean is that bypassing the entire SOFTIRQ priority so that sshd can process one network packet makes a mockery of the point of softirq. This hack to workaround NET_RX looping over-and-over-and-over affects every subsystem, not just one uart. HI, TIMER, BLOCK; all of these are skipped: that's straight-up, a bug. Regards, Peter Hurley