Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965159AbeALSMk (ORCPT + 1 other); Fri, 12 Jan 2018 13:12:40 -0500 Received: from mail-io0-f193.google.com ([209.85.223.193]:36010 "EHLO mail-io0-f193.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S964971AbeALSMh (ORCPT ); Fri, 12 Jan 2018 13:12:37 -0500 X-Google-Smtp-Source: ACJfBos1yaq+ub5ObnI5kxDxnex8VAfDsERkVoO3GIB8227oyvp7XS1Iq8Oig7sBdcOY8nRP+kQiaAL7S1nqnSG7Xeo= MIME-Version: 1.0 In-Reply-To: <20180112143448.GA1950@lerouge> References: <1515735354-19279-1-git-send-email-frederic@kernel.org> <1515735354-19279-2-git-send-email-frederic@kernel.org> <20180112143448.GA1950@lerouge> From: Linus Torvalds Date: Fri, 12 Jan 2018 10:12:32 -0800 X-Google-Sender-Auth: YOymOXtJpwFIm3r0k8suw2WbQdY Message-ID: Subject: Re: [RFC PATCH 1/2] softirq: Account time and iteration stats per vector To: Frederic Weisbecker Cc: Eric Dumazet , LKML , Levin Alexander , Peter Zijlstra , Hannes Frederic Sowa , "Paul E . McKenney" , Wanpeng Li , Dmitry Safonov , Thomas Gleixner , Radu Rendec , Ingo Molnar , Stanislaw Gruszka , Paolo Abeni , Rik van Riel , Andrew Morton , David Miller Content-Type: text/plain; charset="UTF-8" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Return-Path: On Fri, Jan 12, 2018 at 6:34 AM, Frederic Weisbecker wrote: > > That's right. But I thought it was bit large for the stack: > > struct { > u64 time; > u64 count; > } [NR_SOFTIRQS] Note that you definitely don't want "u64" here. Both of these values had better be very limited. The "count" is on the order of 10 - it fits in 4 _bits_ without any overflow. And 'time' is on the order of 2ms, so even if it's in nanoseconds, we already know that we want to limit it to a single ms or so (yes, yes, right now our limit is 2ms, but I think that's long). So even that doesn't need 64-bit. Finally, I think you can join them. If we do a "time or count" limit, let's just make the "count" act as some arbitrary fixed time, so that we limit things that way. Say, if we want to limit it to 2ms, consider one count to be 0.2ms. So instead of keeping track of count at all, just say "make each softirq call count as at least 200,000ns even if the scheduler clock says it's less". End result: we'd loop at most ten times. So now you only need one value, and you know it can't be bigger than 2 million, so it can be a 32-bit one. Boom. Done. Also, don't you want these to be percpu, and keep accumulating them until you decide to either age them away (just clear it in timer interrupt?) or if the value gets so big that you want o fall back to the thread instead (and then the thread can clear it every iteration, so you don't need to track whether the thread is active or not). I don't know. I'm traveling today, so I didn't actually have time to really look at the patches, I'm just reacting to Eric's reaction. Linus