From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: mail.toke.dk; dkim=pass header.d=gentwo.org header.i=@gentwo.org header.a=rsa-sha256 header.s=default header.b=TBXabVTW; arc=none (Message is not ARC signed); dmarc=pass (Used From Domain Record) header.from=gentwo.org policy.dmarc=reject Received: from gentwo.org (gentwo.org [62.72.0.81]) by mail.toke.dk (Postfix) with ESMTPS id 0E12A1934F68 for ; Thu, 29 Oct 2026 22:57:19 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=gentwo.org; s=default; t=1793310319; bh=4hyCxRY5aiuixBCePuRZOoujjI8mLWrM1H0xLX/JSQ0=; h=Date:From:To:cc:Subject:In-Reply-To:References:From; b=TBXabVTW905VLTexygm/a8lMr8AN6K7tvVhOXJszOE+9uVuH8Rrk/wsnSDfauZoeA qyH/hgiIhFdd+3M1ujlkJ1QdBaeyOOd8j03RAPlqeY4pZUUSNnNZ6XcNft1/hL/Bpp N0pz6fx90E1Bkmj53eOsY/sk3dAwfKvDZ2zMjVAY= Received: by gentwo.org (Postfix, from userid 1007) id E042B406E1; Thu, 29 Oct 2026 14:45:19 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by gentwo.org (Postfix) with ESMTP id DC61240112; Thu, 29 Oct 2026 14:45:19 -0700 (PDT) Date: Thu, 29 Oct 2026 14:45:19 -0700 (PDT) From: Shubhang To: Paolo Abeni cc: =?ISO-8859-15?Q?Toke_H=F8iland-J=F8rgensen?= , Jamal Hadi Salim , Jiri Pirko , "David S. Miller" , Eric Dumazet , Jakub Kicinski , Simon Horman , =?ISO-8859-15?Q?Jonas_K=F6ppeler?= , Shubhang Kaushik , "Christoph Lameter (Ampere)" , =?ISO-8859-15?Q?Toke_H=F8iland-J=F8rgensen?= , cake@lists.bufferbloat.net, netdev@vger.kernel.org, linux-kernel@vger.kernel.org In-Reply-To: Message-ID: References: <20261026-net_sched_sch_cake-v2-1-cb1d4e6ed990@gentwo.org> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII; format=flowed X-MailFrom: sh@gentwo.org X-Mailman-Rule-Hits: nonmember-moderation X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; loop; banned-address; emergency; member-moderation Message-ID-Hash: GOWEUWRLOQBVFLBJGNNTX5BNQ3JDWKTA X-Message-ID-Hash: GOWEUWRLOQBVFLBJGNNTX5BNQ3JDWKTA X-Mailman-Approved-At: Thu, 29 Oct 2026 23:02:39 +0100 X-Mailman-Version: 3.3.10 Precedence: list Subject: [Cake] Re: [PATCH v2] net/sched: sch_cake: skip redundant cake_mq rate updates List-Id: Cake - FQ_codel the next generation Archived-At: List-Archive: List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: Hi Paolo, On Thu, 29 Oct 2026, Paolo Abeni wrote: > On 10/27/26 05:32, Shubhang Kaushik (Ampere) wrote: >> cake_mq periodically counts active child qdiscs in cake_dequeue() and >> derives an effective per-child shaper rate. cake_dequeue() currently >> reconfigures every tin after each scan, even when that effective rate >> is unchanged. >> >> The dequeue-time call chain is: >> >> cake_dequeue() >> -> cake_configure_rates(..., true) >> -> cake_config_*() >> -> cake_set_rate() >> >> With rate_adjust=true, cake_set_rate() only refreshes rate-derived >> state. Tin 0 records the unscaled rate argument in every supported >> CAKE tin mode, so its installed rate identifies whether this call chain >> would change any state. >> >> Skip the reconfiguration when the effective rate is already installed, >> while continuing to update the activity-sampling timestamp and >> active-queue statistic. >> >> Fixes: 1bddd758bac2 ("net/sched: sch_cake: share shaper state across >> sub-instances of cake_mq") >> Signed-off-by: Shubhang Kaushik (Ampere) > IMHO this is borderline material for 'net'. Was any regression visible on > top of 1bddd758bac2, e.g. in term of CPU usage? > > /P > Thanks for the feedback. I repeated the test with whole-host CPU accounting on this 80-core system and could not measure a reliable aggregate CPU-use difference; throughput was unchanged. The benefit is visible in focused cake_configure_rates() trace, but I agree this is more appropriate for net-next. Regards, Shubhang Kaushik