From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id 91F9ECCA473 for ; Fri, 24 Jun 2022 06:01:18 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 2D48E6B02AE; Fri, 24 Jun 2022 02:01:18 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 284D28E01E7; Fri, 24 Jun 2022 02:01:18 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 14D478E01E3; Fri, 24 Jun 2022 02:01:18 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0010.hostedemail.com [216.40.44.10]) by kanga.kvack.org (Postfix) with ESMTP id 080A46B02AE for ; Fri, 24 Jun 2022 02:01:18 -0400 (EDT) Received: from smtpin02.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay06.hostedemail.com (Postfix) with ESMTP id D9EB53587E for ; Fri, 24 Jun 2022 06:01:17 +0000 (UTC) X-FDA: 79612081794.02.5E2CFAF Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by imf12.hostedemail.com (Postfix) with ESMTP id DAD59400B3 for ; Fri, 24 Jun 2022 06:01:04 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1656050464; x=1687586464; h=date:from:to:cc:subject:message-id:references: mime-version:in-reply-to; bh=6o4KFuPBreV56Xtx9KWgdlQDzfvBbkhilorMy6YcUL8=; b=NwTFWOV8Zg7RTGvApe/7khHcYsiV7ulviJtxDk8P2j/L/r5P6YhAqDUp UiSXcgtzunKgSEvgeUybW5qXOj9wyOFOoktk60szPTEyHiBnLI9vPFU4i IlqPJ3sHtN0t9aD1pPyQgYpwV5x8MMe+SHql8IMSt21PVnh2+KxDp/rQp o/nd3F6RVaLo0jlLpEZ4ZGDEHfz6ScYMAlL+NX9g/n3lu+8Xv56N9gVHb 3cNvauanvkoJMEsdwri5dhciZqiubDa68N/Iup8BO0CDxbE2OwEnsBzF8 Z0QlNw1xF56Kqw2qT6UtUfXUblmStEJtvba/yWtGIKnb+YOTlJNb7xOun g==; X-IronPort-AV: E=McAfee;i="6400,9594,10387"; a="367249582" X-IronPort-AV: E=Sophos;i="5.92,218,1650956400"; d="scan'208";a="367249582" Received: from fmsmga007.fm.intel.com ([10.253.24.52]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 23 Jun 2022 23:00:58 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.92,218,1650956400"; d="scan'208";a="593076463" Received: from shbuild999.sh.intel.com (HELO localhost) ([10.239.146.138]) by fmsmga007.fm.intel.com with ESMTP; 23 Jun 2022 23:00:54 -0700 Date: Fri, 24 Jun 2022 14:00:53 +0800 From: Feng Tang To: Eric Dumazet Cc: Jakub Kicinski , Xin Long , Marcelo Ricardo Leitner , kernel test robot , Shakeel Butt , Soheil Hassas Yeganeh , LKML , Linux Memory Management List , network dev , linux-s390@vger.kernel.org, MPTCP Upstream , "linux-sctp @ vger . kernel . org" , lkp@lists.01.org, kbuild test robot , Huang Ying , zhengjun.xing@linux.intel.com, fengwei.yin@intel.com, Ying Xu Subject: Re: [net] 4890b686f4: netperf.Throughput_Mbps -69.4% regression Message-ID: <20220624060053.GD79500@shbuild999.sh.intel.com> References: <20220619150456.GB34471@xsang-OptiPlex-9020> <20220622172857.37db0d29@kernel.org> <20220623185730.25b88096@kernel.org> <20220624051351.GA72171@shbuild999.sh.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: X-Rspam-User: Authentication-Results: imf12.hostedemail.com; dkim=temperror ("DNS error when getting key") header.d=intel.com header.s=Intel header.b=NwTFWOV8; spf=temperror (imf12.hostedemail.com: error in processing during lookup of feng.tang@intel.com: DNS error) smtp.mailfrom=feng.tang@intel.com; dmarc=temperror reason="query timed out" header.from=intel.com (policy=temperror) X-Rspamd-Server: rspam05 X-Rspamd-Queue-Id: DAD59400B3 X-Stat-Signature: r9wzhzapsq9to4e9p194f3xrs6jyar9s X-HE-Tag: 1656050464-10462 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: On Fri, Jun 24, 2022 at 07:45:00AM +0200, Eric Dumazet wrote: > On Fri, Jun 24, 2022 at 7:14 AM Feng Tang wrote: > > > > Hi Eric, > > > > On Fri, Jun 24, 2022 at 06:13:51AM +0200, Eric Dumazet wrote: > > > On Fri, Jun 24, 2022 at 3:57 AM Jakub Kicinski wrote: > > > > > > > > On Thu, 23 Jun 2022 18:50:07 -0400 Xin Long wrote: > > > > > From the perf data, we can see __sk_mem_reduce_allocated() is the one > > > > > using CPU the most more than before, and mem_cgroup APIs are also > > > > > called in this function. It means the mem cgroup must be enabled in > > > > > the test env, which may explain why I couldn't reproduce it. > > > > > > > > > > The Commit 4890b686f4 ("net: keep sk->sk_forward_alloc as small as > > > > > possible") uses sk_mem_reclaim(checking reclaimable >= PAGE_SIZE) to > > > > > reclaim the memory, which is *more frequent* to call > > > > > __sk_mem_reduce_allocated() than before (checking reclaimable >= > > > > > SK_RECLAIM_THRESHOLD). It might be cheap when > > > > > mem_cgroup_sockets_enabled is false, but I'm not sure if it's still > > > > > cheap when mem_cgroup_sockets_enabled is true. > > > > > > > > > > I think SCTP netperf could trigger this, as the CPU is the bottleneck > > > > > for SCTP netperf testing, which is more sensitive to the extra > > > > > function calls than TCP. > > > > > > > > > > Can we re-run this testing without mem cgroup enabled? > > > > > > > > FWIW I defer to Eric, thanks a lot for double checking the report > > > > and digging in! > > > > > > I did tests with TCP + memcg and noticed a very small additional cost > > > in memcg functions, > > > because of suboptimal layout: > > > > > > Extract of an internal Google bug, update from June 9th: > > > > > > -------------------------------- > > > I have noticed a minor false sharing to fetch (struct > > > mem_cgroup)->css.parent, at offset 0xc0, > > > because it shares the cache line containing struct mem_cgroup.memory, > > > at offset 0xd0 > > > > > > Ideally, memcg->socket_pressure and memcg->parent should sit in a read > > > mostly cache line. > > > ----------------------- > > > > > > But nothing that could explain a "-69.4% regression" > > > > We can double check that. > > > > > memcg has a very similar strategy of per-cpu reserves, with > > > MEMCG_CHARGE_BATCH being 32 pages per cpu. > > > > We have proposed patch to increase the batch numer for stats > > update, which was not accepted as it hurts the accuracy and > > the data is used by many tools. > > > > > It is not clear why SCTP with 10K writes would overflow this reserve constantly. > > > > > > Presumably memcg experts will have to rework structure alignments to > > > make sure they can cope better > > > with more charge/uncharge operations, because we are not going back to > > > gigantic per-socket reserves, > > > this simply does not scale. > > > > Yes, the memcg statitics and charge/unchage update is very sensitive > > with the data alignemnt layout, and can easily trigger peformance > > changes, as we've seen quite some similar cases in the past several > > years. > > > > One pattern we've seen is, even if a memcg stats updating or charge > > function only takes about 2%~3% of the CPU cycles in perf-profile data, > > once it got affected, the peformance change could be amplified to up to > > 60% or more. > > > > Reorganizing "struct mem_cgroup" to put "struct page_counter memory" > in a separate cache line would be beneficial. That may help. And I also want to say the benchmarks(especially micro one) are very sensitive to the layout of mem_cgroup. As the 'page_counter' is 112 bytes in size, I recently made a patch to make it cacheline aligned (take 2 cachelines), which improved some hackbench/netperf test cases, but caused huge (49%) drop for some vm-scalability tests. > Many low hanging fruits, assuming nobody will use __randomize_layout on it ;) > > Also some fields are written even if their value is not changed. > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > index abec50f31fe64100f4be5b029c7161b3a6077a74..53d9c1e581e78303ef73942e2b34338567987b74 > 100644 > --- a/mm/memcontrol.c > +++ b/mm/memcontrol.c > @@ -7037,10 +7037,12 @@ bool mem_cgroup_charge_skmem(struct mem_cgroup > *memcg, unsigned int nr_pages, > struct page_counter *fail; > > if (page_counter_try_charge(&memcg->tcpmem, nr_pages, &fail)) { > - memcg->tcpmem_pressure = 0; > + if (READ_ONCE(memcg->tcpmem_pressure)) > + WRITE_ONCE(memcg->tcpmem_pressure, 0); > return true; > } > - memcg->tcpmem_pressure = 1; > + if (!READ_ONCE(memcg->tcpmem_pressure)) > + WRITE_ONCE(memcg->tcpmem_pressure, 1); > if (gfp_mask & __GFP_NOFAIL) { > page_counter_charge(&memcg->tcpmem, nr_pages); > return true; I will also try this patch, which may take some time. Thanks, Feng