From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 32D55350A18 for ; Tue, 21 Oct 2025 19:28:50 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=170.10.133.124 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1761074933; cv=none; b=XdRMeL84sZkIx39UCD9p0PwW6jRbjJXOpq7r4NgEQISPC6VBlg7HdjleDWsgvefJzrScwFsY3CHifh4FijK7v5kPZCtBgEwrSBWGx4i+0/yAz6RMOUNmbyh7n+aJmv5UgQPOkN8rnDr/02dxs3CJEi+W0oJYHu+yy4RgFdam+0Q= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1761074933; c=relaxed/simple; bh=5BCq9RRv7KamLtUB79/4SrvmAL0LwrRQ7PGI4CZ2MM8=; h=From:Message-ID:Date:MIME-Version:Subject:To:Cc:References: In-Reply-To:Content-Type; b=kQZbOTLVBetsLDggoQaYQCwbZBwsp35eYkmO44UwrCuZk5dLQ/028tFJOQy5d0uKQ1/hLWx1dyJDIpQSFcBLoBH2Q/7V8fxag8syrwD8pLNob3UhFbe3MGEK0fp/me/ZXAVi8BlGa3yP+qdfCSp75N+isyIIqNytk1X9N4YXL8o= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=redhat.com; spf=pass smtp.mailfrom=redhat.com; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b=PgrniAbq; arc=none smtp.client-ip=170.10.133.124 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="PgrniAbq" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1761074930; h=from:from: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=OAcllF9nVIfCniwCIoLGKjAXFJuJdiDQqFfGC/k/Yfk=; b=PgrniAbq6A8Ta+KQ1dpnmaF1fFvTyAjBLjGFPcaTqIBLTWMMgeaQdpZAkqqb2iNY5flQMR T/S+JYIv9pRdQJILeM/Fsig91O9lAB8JddQ9mHbVQ/mKMi8ewWA3hFJY1CC3Rc+RX3bLjc 1YtZodE2nJ8+vYDkX28nwfsfCI+2yzU= Received: from mail-qk1-f197.google.com (mail-qk1-f197.google.com [209.85.222.197]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-322-MtJ77Lk0M_SOR8wzeGgS_A-1; Tue, 21 Oct 2025 15:28:47 -0400 X-MC-Unique: MtJ77Lk0M_SOR8wzeGgS_A-1 X-Mimecast-MFC-AGG-ID: MtJ77Lk0M_SOR8wzeGgS_A_1761074927 Received: by mail-qk1-f197.google.com with SMTP id af79cd13be357-88ea6169a96so403234185a.2 for ; Tue, 21 Oct 2025 12:28:47 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1761074927; x=1761679727; h=content-transfer-encoding:in-reply-to:content-language:references :cc:to:subject:user-agent:mime-version:date:message-id:from :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=OAcllF9nVIfCniwCIoLGKjAXFJuJdiDQqFfGC/k/Yfk=; b=WHlG/G1dMrBN5H0kzKAMlE7+kUdYpIDUrj6GsrZn0eAubgXMpmXFXj1mkY/AWoYGqR y4CSdNeLUOQ3QNYABLl66UTlLGx8sE72PAhWKfuSwPUTYy3S9CpbPWUvdbQgoEk36Ngf d9OrjWQOafY84XYgtJQgzt3sb9XMbsECAlP/jyM9VQSXvI90eeuKR5mT/Ir3xIyJiruX k4bdRt4Icr64Pt7fYYiZewZ24UdKye6y1gOsjSPsHlL/i5H+U0RVtb3T8b22Dsb42pTI Xu9u2E9xF8n9RHN7zz0x6h0aeqKe/n76Kqf1aUldLf1lq9e1Bhxh+gMFCZLEkV8z3oaa z6BQ== X-Forwarded-Encrypted: i=1; AJvYcCUDM8V+Xb/frO53dJfO1y41P+VoqB87NhbZ8jRzGsqrOeL57gakrmsU56NG2w1plNFbk6WsHVuc275Q5w==@vger.kernel.org X-Gm-Message-State: AOJu0YzeQTM9ThilFo0pd+T0s2WFO1orXvOJ23nILR2PUs+bj99Y0TNq bojHWZk0rjmqJbqdstQBSioTFN4Z3NXt5eTcI/ODNJXcOzc2KRopspkv6sx08FIUpld8y3SKQgY foa/VkuvN75tFlWmwNbutNvrEM9T4DtWsJZ/C0nKbaP/d31Hja7w4cANlfTaW/PGO X-Gm-Gg: ASbGncsXkHc+qalninwG2+GSUSNpScdIYbwK7hQYmEWxpupkfoVUynYs7B0dsxoFtAT MVsS01wDhDZ4LFVjjI/vk4N36MObuDIhTADowE8pQqB7oQ58B7h+/735YEB1/jzmmc38FTJKZhp 5idATTNRbSK6KGpISMzshFLYo+IvGXgeHP+pxw7V5N7JLkB8zHRvZNHpsAAh1zREVPXgnFJ9wRy kWYHZUqgOGxv4xf/Z56DTSx/Erexdo2eWKYQ+GleRC95WqpyKglc5PAnnsMDK08KmB9oKe5MbfH MfKRXfhVGLAR86a2rfWWHNE9QX5uKXAHhkqZvrsAF0at11CiabE3CMRLCzDMltswCW3IzJRtak3 fmvaLBvuA/MpOOid09CEiXmWDGimXpBiTLw2TQAKgUOWk4g== X-Received: by 2002:a05:620a:488e:b0:7e8:67b8:476 with SMTP id af79cd13be357-89070114f94mr2346436585a.57.1761074927214; Tue, 21 Oct 2025 12:28:47 -0700 (PDT) X-Google-Smtp-Source: AGHT+IG+u+8bH+lrXiYca/LtZ13ni/QSfiNRKtdF54+j0jtBiSwDvm50yRps0a3A4dxIJsM4iefKKw== X-Received: by 2002:a05:620a:488e:b0:7e8:67b8:476 with SMTP id af79cd13be357-89070114f94mr2346432285a.57.1761074926666; Tue, 21 Oct 2025 12:28:46 -0700 (PDT) Received: from ?IPV6:2601:600:947f:f020:85dc:d2b2:c5ee:e3c4? ([2601:600:947f:f020:85dc:d2b2:c5ee:e3c4]) by smtp.gmail.com with ESMTPSA id af79cd13be357-891cefba8f0sm817018785a.40.2025.10.21.12.28.42 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Tue, 21 Oct 2025 12:28:46 -0700 (PDT) From: Waiman Long X-Google-Original-From: Waiman Long Message-ID: Date: Tue, 21 Oct 2025 15:28:42 -0400 Precedence: bulk X-Mailing-List: linux-block@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH 14/33] sched/isolation: Flush memcg workqueues on cpuset isolated partition change To: Frederic Weisbecker , LKML Cc: =?UTF-8?Q?Michal_Koutn=C3=BD?= , Andrew Morton , Bjorn Helgaas , Catalin Marinas , Danilo Krummrich , "David S . Miller" , Eric Dumazet , Gabriele Monaco , Greg Kroah-Hartman , Ingo Molnar , Jakub Kicinski , Jens Axboe , Johannes Weiner , Lai Jiangshan , Marco Crivellari , Michal Hocko , Muchun Song , Paolo Abeni , Peter Zijlstra , Phil Auld , "Rafael J . Wysocki" , Roman Gushchin , Shakeel Butt , Simon Horman , Tejun Heo , Thomas Gleixner , Vlastimil Babka , Will Deacon , cgroups@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-block@vger.kernel.org, linux-mm@kvack.org, linux-pci@vger.kernel.org, netdev@vger.kernel.org References: <20251013203146.10162-1-frederic@kernel.org> <20251013203146.10162-15-frederic@kernel.org> <364e084a-ef37-42ab-a2ae-5f103f1eb212@redhat.com> Content-Language: en-US In-Reply-To: <364e084a-ef37-42ab-a2ae-5f103f1eb212@redhat.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit On 10/21/25 3:16 PM, Waiman Long wrote: > On 10/13/25 4:31 PM, Frederic Weisbecker wrote: >> The HK_TYPE_DOMAIN housekeeping cpumask is now modifyable at runtime. In >> order to synchronize against memcg workqueue to make sure that no >> asynchronous draining is still pending or executing on a newly made >> isolated CPU, the housekeeping susbsystem must flush the memcg >> workqueues. >> >> However the memcg workqueues can't be flushed easily since they are >> queued to the main per-CPU workqueue pool. >> >> Solve this with creating a memcg specific pool and provide and use the >> appropriate flushing API. >> >> Acked-by: Shakeel Butt >> Signed-off-by: Frederic Weisbecker >> --- >>   include/linux/memcontrol.h |  4 ++++ >>   kernel/sched/isolation.c   |  2 ++ >>   kernel/sched/sched.h       |  1 + >>   mm/memcontrol.c            | 12 +++++++++++- >>   4 files changed, 18 insertions(+), 1 deletion(-) >> >> diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h >> index 873e510d6f8d..001200df63cf 100644 >> --- a/include/linux/memcontrol.h >> +++ b/include/linux/memcontrol.h >> @@ -1074,6 +1074,8 @@ static inline u64 cgroup_id_from_mm(struct >> mm_struct *mm) >>       return id; >>   } >>   +void mem_cgroup_flush_workqueue(void); >> + >>   extern int mem_cgroup_init(void); >>   #else /* CONFIG_MEMCG */ >>   @@ -1481,6 +1483,8 @@ static inline u64 cgroup_id_from_mm(struct >> mm_struct *mm) >>       return 0; >>   } >>   +static inline void mem_cgroup_flush_workqueue(void) { } >> + >>   static inline int mem_cgroup_init(void) { return 0; } >>   #endif /* CONFIG_MEMCG */ >>   diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c >> index 95d69c2102f6..9ec365dea921 100644 >> --- a/kernel/sched/isolation.c >> +++ b/kernel/sched/isolation.c >> @@ -144,6 +144,8 @@ int housekeeping_update(struct cpumask *mask, >> enum hk_type type) >>         synchronize_rcu(); >>   +    mem_cgroup_flush_workqueue(); >> + >>       kfree(old); >>         return 0; >> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h >> index 8fac8aa451c6..8bfc0b4b133f 100644 >> --- a/kernel/sched/sched.h >> +++ b/kernel/sched/sched.h >> @@ -44,6 +44,7 @@ >>   #include >>   #include >>   #include >> +#include >>   #include >>   #include >>   #include >> diff --git a/mm/memcontrol.c b/mm/memcontrol.c >> index 1033e52ab6cf..1aa14e543f35 100644 >> --- a/mm/memcontrol.c >> +++ b/mm/memcontrol.c >> @@ -95,6 +95,8 @@ static bool cgroup_memory_nokmem __ro_after_init; >>   /* BPF memory accounting disabled? */ >>   static bool cgroup_memory_nobpf __ro_after_init; >>   +static struct workqueue_struct *memcg_wq __ro_after_init; >> + >>   static struct kmem_cache *memcg_cachep; >>   static struct kmem_cache *memcg_pn_cachep; >>   @@ -1975,7 +1977,7 @@ static void schedule_drain_work(int cpu, >> struct work_struct *work) >>   { >>       guard(rcu)(); >>       if (!cpu_is_isolated(cpu)) >> -        schedule_work_on(cpu, work); >> +        queue_work_on(cpu, memcg_wq, work); >>   } >>     /* >> @@ -5092,6 +5094,11 @@ void mem_cgroup_sk_uncharge(const struct sock >> *sk, unsigned int nr_pages) >>       refill_stock(memcg, nr_pages); >>   } >>   +void mem_cgroup_flush_workqueue(void) >> +{ >> +    flush_workqueue(memcg_wq); >> +} >> + >>   static int __init cgroup_memory(char *s) >>   { >>       char *token; >> @@ -5134,6 +5141,9 @@ int __init mem_cgroup_init(void) >>       cpuhp_setup_state_nocalls(CPUHP_MM_MEMCQ_DEAD, >> "mm/memctrl:dead", NULL, >>                     memcg_hotplug_cpu_dead); >>   +    memcg_wq = alloc_workqueue("memcg", 0, 0); > > Should we explicitly mark the memcg_wq as WQ_PERCPU even though I > think percpu is the default. The schedule_work_on() schedules work on > the system_percpu_wq. According to commit dadb3ebcf39 ("workqueue: WQ_PERCPU added to alloc_workqueue users"), the default may be changed to WQ_UNBOUND in the future. Cheers, Longman