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]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 91055CAC5B0 for ; Fri, 3 Oct 2025 10:10:43 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id D7F308E000A; Fri, 3 Oct 2025 06:10:42 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id D56FB8E0001; Fri, 3 Oct 2025 06:10:42 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id C6D348E000A; Fri, 3 Oct 2025 06:10:42 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0014.hostedemail.com [216.40.44.14]) by kanga.kvack.org (Postfix) with ESMTP id B32CC8E0001 for ; Fri, 3 Oct 2025 06:10:42 -0400 (EDT) Received: from smtpin28.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay08.hostedemail.com (Postfix) with ESMTP id 6449314073F for ; Fri, 3 Oct 2025 10:10:42 +0000 (UTC) X-FDA: 83956383924.28.9111852 Received: from frasgout.his.huawei.com (frasgout.his.huawei.com [185.176.79.56]) by imf28.hostedemail.com (Postfix) with ESMTP id 6A8A5C000E for ; Fri, 3 Oct 2025 10:10:40 +0000 (UTC) Authentication-Results: imf28.hostedemail.com; dkim=none; spf=pass (imf28.hostedemail.com: domain of jonathan.cameron@huawei.com designates 185.176.79.56 as permitted sender) smtp.mailfrom=jonathan.cameron@huawei.com; dmarc=pass (policy=quarantine) header.from=huawei.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1759486240; 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; bh=z5LE5la4vH515j8B8M4gP9iPiY+KxqkA0CjAyviV8oI=; b=fwcf7uKy9J3d5E/Lpu5K7y5u48WtPtFZODfLqZANf/18sMYUVeS/wHriS32CskkdI+iU2N 7kkoJNoummi/s06n0UsHm5PByI/d3/TmGsgxRK+fDb+Rs11WeZ9epjxNVLCkrZfNCJOYLG g56EeaHe0bbZMB/dD4x5W4TaHCerT8A= ARC-Authentication-Results: i=1; imf28.hostedemail.com; dkim=none; spf=pass (imf28.hostedemail.com: domain of jonathan.cameron@huawei.com designates 185.176.79.56 as permitted sender) smtp.mailfrom=jonathan.cameron@huawei.com; dmarc=pass (policy=quarantine) header.from=huawei.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1759486240; a=rsa-sha256; cv=none; b=2NHdfDeMSWGmayvmTkSHjwvTWALMGTG/qV3deTZXVkFdnzbv2s/JPeiQVh8477kf1iFSTP 3+52uUCXJAQoHzov4guET8Zq+SS8rHQIVIah/KUjtgbVTSPeS39JlOhAiXwYu6vBFF13fO cyoW78pjbFEtWJ4RDEzKiqsmEVl74kA= Received: from mail.maildlp.com (unknown [172.18.186.216]) by frasgout.his.huawei.com (SkyGuard) with ESMTP id 4cdPYK6YGHz67bVD; Fri, 3 Oct 2025 18:10:13 +0800 (CST) Received: from dubpeml100005.china.huawei.com (unknown [7.214.146.113]) by mail.maildlp.com (Postfix) with ESMTPS id 256911400D9; Fri, 3 Oct 2025 18:10:37 +0800 (CST) Received: from localhost (10.203.177.15) by dubpeml100005.china.huawei.com (7.214.146.113) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.11; Fri, 3 Oct 2025 11:10:34 +0100 Date: Fri, 3 Oct 2025 11:10:32 +0100 From: Jonathan Cameron To: Raghavendra K T CC: , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , Subject: Re: [RFC PATCH V3 11/17] mm/kscand: Implement migration failure feedback Message-ID: <20251003111032.00004688@huawei.com> In-Reply-To: <20250814153307.1553061-12-raghavendra.kt@amd.com> References: <20250814153307.1553061-1-raghavendra.kt@amd.com> <20250814153307.1553061-12-raghavendra.kt@amd.com> X-Mailer: Claws Mail 4.3.0 (GTK 3.24.42; x86_64-w64-mingw32) MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Originating-IP: [10.203.177.15] X-ClientProxiedBy: lhrpeml500012.china.huawei.com (7.191.174.4) To dubpeml100005.china.huawei.com (7.214.146.113) X-Rspamd-Queue-Id: 6A8A5C000E X-Rspamd-Server: rspam05 X-Stat-Signature: i6jtpmpm544aqwn31h1kazc7kz874iry X-Rspam-User: X-HE-Tag: 1759486240-500787 X-HE-Meta: U2FsdGVkX19jl61VkLVV05XYNkgbnvbMYR2u0A/A8teU1GxNolgogKFTMwGHqwZdjYyqn0JlztvL6QjoWUuqolMnbtUsFXSzbGGa6cUqOyUbxb+rPmYxc3v2FH/YjpF9BU46Jca07LC/tQZBoRmtdXXLj9KhTbjfsMYCD9ux8YmXc9JTp6LuG1QVdT/dpXn+s5UdGnWfidc3wPTupsJcBrgMCWG0T5kWaK45FnRyM+qmb7GOJ0KuX1rnNZjMiaM4rhaqH+20C+GyZUPJazDDwmEg4ie/iTLfu2jCY/equFVVj0NwCw69Sqm5v8DmqydbgsNH4IvRw3IEKN6AoADH/7L3OKRjg2O2PJ051mGIhWI93UUICs3SeOVUfOEz87OObKPmCzWUv9poP51OuFpZli3GSyyxaU3InFB96GPpMgZu7Qz+7tiE00NuTxWVTfxJeqPiQdZ6QVYz0DDguBIaO5zyBEpt9i/GCPhEakCQJo33vP6wGctoA4hVweGbjdBTQOyO/s5INofh6gvUWL9l7YHsYD2e3WFRTsF6dSIMMauf239cs5ThXMOCWxCKU+yM7aufXLVbCUTAd4tUHIZbowiOZIk2BSeh+4YwOqMq6uR1xHG42Lmoh91CxaG/P6/aVXJmqe0JMw9j3KVeAlCMLX5RDiNgiTTq3zkHRHyWh3tmJ5lQvzqlFEq9hgqs3qRgGF6zk5JZLb1MtJgodCuXQhjIsStib75v2nLdyyu+ENgC05ukyLFZyOm2c0XzTQt5SwXR2u1KDeBBWDf2WA87IFDn4tIkWzSDjQ/2oIYV50cQXL5GK5q5YLcvKB7rV532FRDqmCCwP743JoHSCZ+a+TB+oEV9iOwlFGBKxR+szlSvToUgrYs7y93P8BkqIGllwRlBCBkfJwt6OpCOUAn7r0jHFHVAHJl490372fIxNcxnbXVeEnQ7pT5rZpnF/UxIn6rW8w/mSST70utfSV9 PqKRvbfv 3MDEmjjc+tv7OYxCYDc2j9RoB34rhIxcQ+cUhLvZ08lVlp8gGeT3EjT5AOVCpjsrKrOorsrFyzyxfYTuc8HIz3nQdrY7x+e689EZ38wnofyrS7sEfizP+UM+8+eXJnaZkp2XcZFwUpathIWxRO4/iyAPKGRh9tL7207JMQphDRZWGzCn8ED9mDEXApOKYRBJMrazWRq9XuJ0yj6l6wNXkEKNrCBsUAc16OZoOdnh7okBFKVLYTwdeLd3IKJzdmmJfedBR+NRz8NvqDihhTNKucQF0Yb55z5B7rPh4EUNkofIEg0Cy9OtqFhhaBW7uuCn+GNQi5QP6s0jXqE0nH4MiDOMioAFXx+GAQngr 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: On Thu, 14 Aug 2025 15:33:01 +0000 Raghavendra K T wrote: > Before this, scanning kthread continues to scan even after > migration fails. To control migration, scanning is slowed down > based on the failure/success ratio obtained from migration > thread. > > Decaying failure ratio is maintained for 1024 migration window. > The ratio further contributes to approximately 10% scaling of > scan_period. Perhaps it's worth adding a cover letter section describing all the heuristics briefly so we have a central place to understand what needs tuning against workloads before this merges? J > > Signed-off-by: Raghavendra K T > --- > mm/kscand.c | 55 +++++++++++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 55 insertions(+) > > diff --git a/mm/kscand.c b/mm/kscand.c > index bf975e82357d..41321d373be7 100644 > --- a/mm/kscand.c > +++ b/mm/kscand.c > @@ -146,6 +146,8 @@ struct kmigrated_mm_slot { > spinlock_t migrate_lock; > /* Head of per mm migration list */ > struct list_head migrate_head; > + /* Indicates weighted success, failure */ > + int msuccess, mfailed, fratio; > }; > > /* System wide list of mms that maintain migration list */ > @@ -812,13 +814,45 @@ static void kscand_collect_mm_slot(struct kscand_mm_slot *mm_slot) > } > } > > +static int kmigrated_get_mstat_fratio(struct mm_struct *mm) > +{ > + int fratio = 0; > + struct kmigrated_mm_slot *mm_slot = NULL; > + struct mm_slot *slot; > + > + guard(spinlock)(&kscand_migrate_lock); > + > + slot = mm_slot_lookup(kmigrated_slots_hash, mm); > + mm_slot = mm_slot_entry(slot, struct kmigrated_mm_slot, mm_slot); > + > + if (mm_slot) > + fratio = mm_slot->fratio; Extra space after = > + > + return fratio; > +} > + > +static void update_mstat_ratio(struct kmigrated_mm_slot *mm_slot, > + int msuccess, int mfailed) > +{ > + mm_slot->msuccess = (mm_slot->msuccess >> 2) + msuccess; > + mm_slot->mfailed = (mm_slot->mfailed >> 2) + mfailed; > + mm_slot->fratio = mm_slot->mfailed * 100; > + mm_slot->fratio /= (mm_slot->msuccess + mm_slot->mfailed); extra space after = > +} > + > +#define MSTAT_UPDATE_FREQ 1024 > + > static void kmigrated_migrate_mm(struct kmigrated_mm_slot *mm_slot) > { > + int mfailed = 0; > + int msuccess = 0; > + int mstat_counter; > int ret = 0, dest = -1; > struct mm_slot *slot; > struct mm_struct *mm; > struct kscand_migrate_info *info, *tmp; > > + mstat_counter = MSTAT_UPDATE_FREQ; Might as well set at declaration above. > spin_lock(&mm_slot->migrate_lock); > > slot = &mm_slot->mm_slot; > @@ -842,11 +876,23 @@ static void kmigrated_migrate_mm(struct kmigrated_mm_slot *mm_slot) > } > > ret = kmigrated_promote_folio(info, mm, dest); > + mstat_counter--; > + > + /* TBD: encode migrated count here, currently assume folio_nr_pages */ > + if (!ret) > + msuccess++; > + else > + mfailed++; > > kfree(info); > > cond_resched(); > spin_lock(&mm_slot->migrate_lock); > + if (!mstat_counter) { > + update_mstat_ratio(mm_slot, msuccess, mfailed); > + msuccess = mfailed = 0; extra space before = > + mstat_counter = MSTAT_UPDATE_FREQ; > + } > } > }