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 AF581C3DA61 for ; Mon, 29 Jul 2024 06:04:31 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 45A426B009E; Mon, 29 Jul 2024 02:04:31 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 40A0D6B00A0; Mon, 29 Jul 2024 02:04:31 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 284106B00A1; Mon, 29 Jul 2024 02:04:31 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0011.hostedemail.com [216.40.44.11]) by kanga.kvack.org (Postfix) with ESMTP id 01F816B009E for ; Mon, 29 Jul 2024 02:04:30 -0400 (EDT) Received: from smtpin17.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay02.hostedemail.com (Postfix) with ESMTP id B683A12011F for ; Mon, 29 Jul 2024 06:04:30 +0000 (UTC) X-FDA: 82391750700.17.36A49AB Received: from mgamail.intel.com (mgamail.intel.com [198.175.65.16]) by imf03.hostedemail.com (Postfix) with ESMTP id 319F620028 for ; Mon, 29 Jul 2024 06:04:28 +0000 (UTC) Authentication-Results: imf03.hostedemail.com; dkim=pass header.d=intel.com header.s=Intel header.b=a8mKe77W; dmarc=pass (policy=none) header.from=intel.com; spf=pass (imf03.hostedemail.com: domain of ying.huang@intel.com designates 198.175.65.16 as permitted sender) smtp.mailfrom=ying.huang@intel.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1722233065; a=rsa-sha256; cv=none; b=FSHDCZNNbbbLjFz3244L1NqYJdQ1zfESzAFRLSGCofHAMH+jpqDbXwX9V9uTh6nZe+VPUc v0yTgpJrxCkfW7lqy44mAdqCcZE09pFmTDXITaSYfN0mD2jUtI5rD39u1s2gZH+J2m2vJ7 bFXq0uY3V4vuvTUp7sEPiEFh7YssAx0= ARC-Authentication-Results: i=1; imf03.hostedemail.com; dkim=pass header.d=intel.com header.s=Intel header.b=a8mKe77W; dmarc=pass (policy=none) header.from=intel.com; spf=pass (imf03.hostedemail.com: domain of ying.huang@intel.com designates 198.175.65.16 as permitted sender) smtp.mailfrom=ying.huang@intel.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1722233065; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references:dkim-signature; bh=gJnHol0epqRXGwBygOQtzpq80iAO4dJj+1WWsZmWSbo=; b=yQCWx5lyKXlfsanS4nibjhthP0fna6KvncpWBlI2U1LhOmKqu9ZM+SZIequJWQg6zKlhFv p/swLK7vQQUmopkibRt5mQg2CiUx5LDTpaI//Bc8BltcRJUjEXLlDdK+Tl4FQphzHH2H/s c1mAb4mvKBqAvJvej4Tjx7XJbilKHPg= DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1722233068; x=1753769068; h=from:to:cc:subject:in-reply-to:references:date: message-id:mime-version:content-transfer-encoding; bh=GLaZYZ2wRM3RRSd0t3BrgE5K4lT+uQ8sms8irJGF6sE=; b=a8mKe77WTMtnDWWgYKLfVigYTpoGBrXLSf52kEPuwfjwcxLrpGJfQPaw V+IYzbyvf/e3YMUgGGOKe6BDXTyEEhWIrf/OUn8GFhtK3NRuj4lnuE7Kz LMgNEkRLv40t1RSb6DPl5enixykKD3Z+97pU1VrgziCKthCo1ASLttJM1 ZGWzfv6vajZF3aHsSD3beILbS6A3Mm1eWPwurFM8Bs2tsXk0T0UKx4Wtz wGAksMq+a/i/cNSNEXqCB8A0woQxWco1Rinwoi+xUdqdqvQcd6vY9ZB8e np2fpp5ki+/ArQGXlyf/1VpQzFFEpMHDLatiDSb5UHEh/Ys7wXbRhAsJp w==; X-CSE-ConnectionGUID: tL71X5/hS1qHP78QXobEWA== X-CSE-MsgGUID: +AeWLstaR2q+BbrsBZm6vA== X-IronPort-AV: E=McAfee;i="6700,10204,11147"; a="20105587" X-IronPort-AV: E=Sophos;i="6.09,245,1716274800"; d="scan'208";a="20105587" Received: from fmviesa010.fm.intel.com ([10.60.135.150]) by orvoesa108.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 28 Jul 2024 23:04:26 -0700 X-CSE-ConnectionGUID: NlCD3uh6Tvm0xaRXeu/Yag== X-CSE-MsgGUID: lwKzjcMXSd+nbSngruZ/Ig== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.09,245,1716274800"; d="scan'208";a="53955694" Received: from yhuang6-desk2.sh.intel.com (HELO yhuang6-desk2.ccr.corp.intel.com) ([10.238.208.55]) by fmviesa010-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 28 Jul 2024 23:04:24 -0700 From: "Huang, Ying" To: Yafang Shao Cc: akpm@linux-foundation.org, mgorman@techsingularity.net, linux-mm@kvack.org, Matthew Wilcox , David Rientjes Subject: Re: [PATCH v2 3/3] mm/page_alloc: Introduce a new sysctl knob vm.pcp_batch_scale_max In-Reply-To: (Yafang Shao's message of "Mon, 29 Jul 2024 14:00:04 +0800") References: <20240729023532.1555-1-laoar.shao@gmail.com> <20240729023532.1555-4-laoar.shao@gmail.com> <878qxkyjfr.fsf@yhuang6-desk2.ccr.corp.intel.com> <874j88ye65.fsf@yhuang6-desk2.ccr.corp.intel.com> <87zfq0wxub.fsf@yhuang6-desk2.ccr.corp.intel.com> Date: Mon, 29 Jul 2024 14:00:51 +0800 Message-ID: <87v80owxd8.fsf@yhuang6-desk2.ccr.corp.intel.com> User-Agent: Gnus/5.13 (Gnus v5.13) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-Rspam-User: X-Rspamd-Queue-Id: 319F620028 X-Rspamd-Server: rspam01 X-Stat-Signature: f1zpijzspargsgn9ajqpk3x9tzhjf6ey X-HE-Tag: 1722233068-159204 X-HE-Meta: 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 vFY1+MMB kJcRKrQFoa1PUVJSMEfHAnt+waq4kRLxVRxPihnNYf6PMg8WvjjQ6h3Orw9TpR0URt0nV9J8o1xOOzmM6RBODGWrx3Aczy4rpVqtHuvNorYlgbx5VADe5W5wbEB5ON/8cO7y4toZOvYcS3Y1+sJ751R/FseZ0puF8peQhUZl3qTipeG3Px/uDqLd1jY7EeeT82KRQpKDYykYM0H8gHNPKWbTJj7mH43D8LZ7O15IsCakEEjm0+02wITUaFzyE4BM7Lm4tu8GbAXUp8w626sw0qGgHN4i+0pqJBvqNpf86UboHSTwT4Iy3OnQpczdNvZujQysBD002xHl881VueYq13JYtFzq4uuFa2T17KBdzSrJfXUvH0ylMYBviEg== 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: List-Subscribe: List-Unsubscribe: Yafang Shao writes: > On Mon, Jul 29, 2024 at 1:54=E2=80=AFPM Huang, Ying wrote: >> >> Yafang Shao writes: >> >> > On Mon, Jul 29, 2024 at 1:16=E2=80=AFPM Huang, Ying wrote: >> >> >> >> Yafang Shao writes: >> >> >> >> > On Mon, Jul 29, 2024 at 11:22=E2=80=AFAM Huang, Ying wrote: >> >> >> >> >> >> Hi, Yafang, >> >> >> >> >> >> Yafang Shao writes: >> >> >> >> >> >> > During my recent work to resolve latency spikes caused by zone->= lock >> >> >> > contention[0], I found that CONFIG_PCP_BATCH_SCALE_MAX is diffic= ult to use >> >> >> > in practice. >> >> >> >> >> >> As we discussed before [1], I still feel confusing about the descr= iption >> >> >> about zone->lock contention. How about change the description to >> >> >> something like, >> >> > >> >> > Sure, I will change it. >> >> > >> >> >> >> >> >> Larger page allocation/freeing batch number may cause longer run t= ime of >> >> >> code holding zone->lock. If zone->lock is heavily contended at th= e same >> >> >> time, latency spikes may occur even for casual page allocation/fre= eing. >> >> >> Although reducing the batch number cannot make zone->lock contended >> >> >> lighter, it can reduce the latency spikes effectively. >> >> >> >> >> >> [1] https://lore.kernel.org/linux-mm/87ttgv8hlz.fsf@yhuang6-desk2.= ccr.corp.intel.com/ >> >> >> >> >> >> > To demonstrate this, I wrote a Python script: >> >> >> > >> >> >> > import mmap >> >> >> > >> >> >> > size =3D 6 * 1024**3 >> >> >> > >> >> >> > while True: >> >> >> > mm =3D mmap.mmap(-1, size) >> >> >> > mm[:] =3D b'\xff' * size >> >> >> > mm.close() >> >> >> > >> >> >> > Run this script 10 times in parallel and measure the allocation = latency by >> >> >> > measuring the duration of rmqueue_bulk() with the BCC tools >> >> >> > funclatency[1]: >> >> >> > >> >> >> > funclatency -T -i 600 rmqueue_bulk >> >> >> > >> >> >> > Here are the results for both AMD and Intel CPUs. >> >> >> > >> >> >> > AMD EPYC 7W83 64-Core Processor, single NUMA node, KVM virtual s= erver >> >> >> > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D >> >> >> > >> >> >> > - Default value of 5 >> >> >> > >> >> >> > nsecs : count distribution >> >> >> > 0 -> 1 : 0 | = | >> >> >> > 2 -> 3 : 0 | = | >> >> >> > 4 -> 7 : 0 | = | >> >> >> > 8 -> 15 : 0 | = | >> >> >> > 16 -> 31 : 0 | = | >> >> >> > 32 -> 63 : 0 | = | >> >> >> > 64 -> 127 : 0 | = | >> >> >> > 128 -> 255 : 0 | = | >> >> >> > 256 -> 511 : 0 | = | >> >> >> > 512 -> 1023 : 12 | = | >> >> >> > 1024 -> 2047 : 9116 | = | >> >> >> > 2048 -> 4095 : 2004 | = | >> >> >> > 4096 -> 8191 : 2497 | = | >> >> >> > 8192 -> 16383 : 2127 | = | >> >> >> > 16384 -> 32767 : 2483 | = | >> >> >> > 32768 -> 65535 : 10102 | = | >> >> >> > 65536 -> 131071 : 212730 |******************* = | >> >> >> > 131072 -> 262143 : 314692 |***************************= ** | >> >> >> > 262144 -> 524287 : 430058 |***************************= *************| >> >> >> > 524288 -> 1048575 : 224032 |******************** = | >> >> >> > 1048576 -> 2097151 : 73567 |****** = | >> >> >> > 2097152 -> 4194303 : 17079 |* = | >> >> >> > 4194304 -> 8388607 : 3900 | = | >> >> >> > 8388608 -> 16777215 : 750 | = | >> >> >> > 16777216 -> 33554431 : 88 | = | >> >> >> > 33554432 -> 67108863 : 2 | = | >> >> >> > >> >> >> > avg =3D 449775 nsecs, total: 587066511229 nsecs, count: 1305242 >> >> >> > >> >> >> > The avg alloc latency can be 449us, and the max latency can be h= igher >> >> >> > than 30ms. >> >> >> > >> >> >> > - Value set to 0 >> >> >> > >> >> >> > nsecs : count distribution >> >> >> > 0 -> 1 : 0 | = | >> >> >> > 2 -> 3 : 0 | = | >> >> >> > 4 -> 7 : 0 | = | >> >> >> > 8 -> 15 : 0 | = | >> >> >> > 16 -> 31 : 0 | = | >> >> >> > 32 -> 63 : 0 | = | >> >> >> > 64 -> 127 : 0 | = | >> >> >> > 128 -> 255 : 0 | = | >> >> >> > 256 -> 511 : 0 | = | >> >> >> > 512 -> 1023 : 92 | = | >> >> >> > 1024 -> 2047 : 8594 | = | >> >> >> > 2048 -> 4095 : 2042818 |****** = | >> >> >> > 4096 -> 8191 : 8737624 |************************** = | >> >> >> > 8192 -> 16383 : 13147872 |***************************= *************| >> >> >> > 16384 -> 32767 : 8799951 |************************** = | >> >> >> > 32768 -> 65535 : 2879715 |******** = | >> >> >> > 65536 -> 131071 : 659600 |** = | >> >> >> > 131072 -> 262143 : 204004 | = | >> >> >> > 262144 -> 524287 : 78246 | = | >> >> >> > 524288 -> 1048575 : 30800 | = | >> >> >> > 1048576 -> 2097151 : 12251 | = | >> >> >> > 2097152 -> 4194303 : 2950 | = | >> >> >> > 4194304 -> 8388607 : 78 | = | >> >> >> > >> >> >> > avg =3D 19359 nsecs, total: 708638369918 nsecs, count: 36604636 >> >> >> > >> >> >> > The avg was reduced significantly to 19us, and the max latency i= s reduced >> >> >> > to less than 8ms. >> >> >> > >> >> >> > - Conclusion >> >> >> > >> >> >> > On this AMD CPU, reducing vm.pcp_batch_scale_max significantly h= elps reduce >> >> >> > latency. Latency-sensitive applications will benefit from this t= uning. >> >> >> > >> >> >> > However, I don't have access to other types of AMD CPUs, so I wa= s unable to >> >> >> > test it on different AMD models. >> >> >> > >> >> >> > Intel(R) Xeon(R) Platinum 8260 CPU @ 2.40GHz, two NUMA nodes >> >> >> > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D >> >> >> > >> >> >> > - Default value of 5 >> >> >> > >> >> >> > nsecs : count distribution >> >> >> > 0 -> 1 : 0 | = | >> >> >> > 2 -> 3 : 0 | = | >> >> >> > 4 -> 7 : 0 | = | >> >> >> > 8 -> 15 : 0 | = | >> >> >> > 16 -> 31 : 0 | = | >> >> >> > 32 -> 63 : 0 | = | >> >> >> > 64 -> 127 : 0 | = | >> >> >> > 128 -> 255 : 0 | = | >> >> >> > 256 -> 511 : 0 | = | >> >> >> > 512 -> 1023 : 2419 | = | >> >> >> > 1024 -> 2047 : 34499 |* = | >> >> >> > 2048 -> 4095 : 4272 | = | >> >> >> > 4096 -> 8191 : 9035 | = | >> >> >> > 8192 -> 16383 : 4374 | = | >> >> >> > 16384 -> 32767 : 2963 | = | >> >> >> > 32768 -> 65535 : 6407 | = | >> >> >> > 65536 -> 131071 : 884806 |***************************= *************| >> >> >> > 131072 -> 262143 : 145931 |****** = | >> >> >> > 262144 -> 524287 : 13406 | = | >> >> >> > 524288 -> 1048575 : 1874 | = | >> >> >> > 1048576 -> 2097151 : 249 | = | >> >> >> > 2097152 -> 4194303 : 28 | = | >> >> >> > >> >> >> > avg =3D 96173 nsecs, total: 106778157925 nsecs, count: 1110263 >> >> >> > >> >> >> > - Conclusion >> >> >> > >> >> >> > This Intel CPU works fine with the default setting. >> >> >> > >> >> >> > Intel(R) Xeon(R) Platinum 8260 CPU @ 2.40GHz, single NUMA node >> >> >> > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D >> >> >> > >> >> >> > Using the cpuset cgroup, we can restrict the test script to run = on NUMA >> >> >> > node 0 only. >> >> >> > >> >> >> > - Default value of 5 >> >> >> > >> >> >> > nsecs : count distribution >> >> >> > 0 -> 1 : 0 | = | >> >> >> > 2 -> 3 : 0 | = | >> >> >> > 4 -> 7 : 0 | = | >> >> >> > 8 -> 15 : 0 | = | >> >> >> > 16 -> 31 : 0 | = | >> >> >> > 32 -> 63 : 0 | = | >> >> >> > 64 -> 127 : 0 | = | >> >> >> > 128 -> 255 : 0 | = | >> >> >> > 256 -> 511 : 46 | = | >> >> >> > 512 -> 1023 : 695 | = | >> >> >> > 1024 -> 2047 : 19950 |* = | >> >> >> > 2048 -> 4095 : 1788 | = | >> >> >> > 4096 -> 8191 : 3392 | = | >> >> >> > 8192 -> 16383 : 2569 | = | >> >> >> > 16384 -> 32767 : 2619 | = | >> >> >> > 32768 -> 65535 : 3809 | = | >> >> >> > 65536 -> 131071 : 616182 |***************************= *************| >> >> >> > 131072 -> 262143 : 295587 |******************* = | >> >> >> > 262144 -> 524287 : 75357 |**** = | >> >> >> > 524288 -> 1048575 : 15471 |* = | >> >> >> > 1048576 -> 2097151 : 2939 | = | >> >> >> > 2097152 -> 4194303 : 243 | = | >> >> >> > 4194304 -> 8388607 : 3 | = | >> >> >> > >> >> >> > avg =3D 144410 nsecs, total: 150281196195 nsecs, count: 1040651 >> >> >> > >> >> >> > The zone->lock contention becomes severe when there is only a si= ngle NUMA >> >> >> > node. The average latency is approximately 144us, with the maxim= um >> >> >> > latency exceeding 4ms. >> >> >> > >> >> >> > - Value set to 0 >> >> >> > >> >> >> > nsecs : count distribution >> >> >> > 0 -> 1 : 0 | = | >> >> >> > 2 -> 3 : 0 | = | >> >> >> > 4 -> 7 : 0 | = | >> >> >> > 8 -> 15 : 0 | = | >> >> >> > 16 -> 31 : 0 | = | >> >> >> > 32 -> 63 : 0 | = | >> >> >> > 64 -> 127 : 0 | = | >> >> >> > 128 -> 255 : 0 | = | >> >> >> > 256 -> 511 : 24 | = | >> >> >> > 512 -> 1023 : 2686 | = | >> >> >> > 1024 -> 2047 : 10246 | = | >> >> >> > 2048 -> 4095 : 4061529 |********* = | >> >> >> > 4096 -> 8191 : 16894971 |***************************= *************| >> >> >> > 8192 -> 16383 : 6279310 |************** = | >> >> >> > 16384 -> 32767 : 1658240 |*** = | >> >> >> > 32768 -> 65535 : 445760 |* = | >> >> >> > 65536 -> 131071 : 110817 | = | >> >> >> > 131072 -> 262143 : 20279 | = | >> >> >> > 262144 -> 524287 : 4176 | = | >> >> >> > 524288 -> 1048575 : 436 | = | >> >> >> > 1048576 -> 2097151 : 8 | = | >> >> >> > 2097152 -> 4194303 : 2 | = | >> >> >> > >> >> >> > avg =3D 8401 nsecs, total: 247739809022 nsecs, count: 29488508 >> >> >> > >> >> >> > After setting it to 0, the avg latency is reduced to around 8us,= and the >> >> >> > max latency is less than 4ms. >> >> >> > >> >> >> > - Conclusion >> >> >> > >> >> >> > On this Intel CPU, this tuning doesn't help much. Latency-sensit= ive >> >> >> > applications work well with the default setting. >> >> >> > >> >> >> > It is worth noting that all the above data were tested using the= upstream >> >> >> > kernel. >> >> >> > >> >> >> > Why introduce a systl knob? >> >> >> > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D >> >> >> > >> >> >> > From the above data, it's clear that different CPU types have va= rying >> >> >> > allocation latencies concerning zone->lock contention. Typically= , people >> >> >> > don't release individual kernel packages for each type of x86_64= CPU. >> >> >> > >> >> >> > Furthermore, for latency-insensitive applications, we can keep t= he default >> >> >> > setting for better throughput. In our production environment, we= set this >> >> >> > value to 0 for applications running on Kubernetes servers while = keeping it >> >> >> > at the default value of 5 for other applications like big data. = It's not >> >> >> > common to release individual kernel packages for each applicatio= n. >> >> >> >> >> >> Thanks for detailed performance data! >> >> >> >> >> >> Is there any downside observed to set CONFIG_PCP_BATCH_SCALE_MAX t= o 0 in >> >> >> your environment? If not, I suggest to use 0 as default for >> >> >> CONFIG_PCP_BATCH_SCALE_MAX. Because we have clear evidence that >> >> >> CONFIG_PCP_BATCH_SCALE_MAX hurts latency for some workloads. After >> >> >> that, if someone found some other workloads need larger >> >> >> CONFIG_PCP_BATCH_SCALE_MAX, we can make it tunable dynamically. >> >> >> >> >> > >> >> > The decision doesn=E2=80=99t rest with us, the kernel team at our c= ompany. >> >> > It=E2=80=99s made by the system administrators who manage a large n= umber of >> >> > servers. The latency spikes only occur on the Kubernetes (k8s) >> >> > servers, not in other environments like big data servers. We have >> >> > informed other system administrators, such as those managing the big >> >> > data servers, about the latency spike issues, but they are unwilling >> >> > to make the change. >> >> > >> >> > No one wants to make changes unless there is evidence showing that = the >> >> > old settings will negatively impact them. However, as you know, >> >> > latency is not a critical concern for big data; throughput is more >> >> > important. If we keep the current settings, we will have to release >> >> > different kernel packages for different environments, which is a >> >> > significant burden for us. >> >> >> >> Totally understand your requirements. And, I think that this is bett= er >> >> to be resolved in your downstream kernel. If there are clear evidenc= es >> >> to prove small batch number hurts throughput for some workloads, we c= an >> >> make the change in the upstream kernel. >> >> >> > >> > Please don't make this more complicated. We are at an impasse. >> > >> > The key issue here is that the upstream kernel has a default value of >> > 5, not 0. If you can change it to 0, we can persuade our users to >> > follow the upstream changes. They currently set it to 5, not because >> > you, the author, chose this value, but because it is the default in >> > Linus's tree. Since it's in Linus's tree, kernel developers worldwide >> > support it. It's not just your decision as the author, but the entire >> > community supports this default. >> > >> > If, in the future, we find that the value of 0 is not suitable, you'll >> > tell us, "It is an issue in your downstream kernel, not in the >> > upstream kernel, so we won't accept it." PANIC. >> >> I don't think so. I suggest you to change the default value to 0. If >> someone reported that his workloads need some other value, then we have >> evidence that different workloads need different value. At that time, >> we can suggest to add an user tunable knob. >> > > The problem is that others are unaware we've set it to 0, and I can't > constantly monitor the linux-mm mailing list. Additionally, it's > possible that you can't always keep an eye on it either. IIUC, they will use the default value. Then, if there is any performance regression, they can report it. > I believe we should hear Andrew's suggestion. Andrew, what is your opinio= n? -- Best Regards, Huang, Ying