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 X-Spam-Level: X-Spam-Status: No, score=-9.6 required=3.0 tests=DKIM_SIGNED,DKIM_VALID, DKIM_VALID_AU,FREEMAIL_FORGED_FROMDOMAIN,FREEMAIL_FROM, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY, SPF_HELO_NONE,SPF_PASS,USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 8A40BC38A29 for ; Sat, 18 Apr 2020 15:13:44 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 437692076A for ; Sat, 18 Apr 2020 15:13:44 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="Y32CSBdM" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 437692076A Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=gmail.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id E999E8E0008; Sat, 18 Apr 2020 11:13:43 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id E4A288E0003; Sat, 18 Apr 2020 11:13:43 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id D5F208E0008; Sat, 18 Apr 2020 11:13:43 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0243.hostedemail.com [216.40.44.243]) by kanga.kvack.org (Postfix) with ESMTP id BDD2C8E0003 for ; Sat, 18 Apr 2020 11:13:43 -0400 (EDT) Received: from smtpin20.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay02.hostedemail.com (Postfix) with ESMTP id 826BF63FA for ; Sat, 18 Apr 2020 15:13:43 +0000 (UTC) X-FDA: 76721320326.20.car83_77ca5596ebe07 X-HE-Tag: car83_77ca5596ebe07 X-Filterd-Recvd-Size: 10152 Received: from mail-pg1-f196.google.com (mail-pg1-f196.google.com [209.85.215.196]) by imf46.hostedemail.com (Postfix) with ESMTP for ; Sat, 18 Apr 2020 15:13:43 +0000 (UTC) Received: by mail-pg1-f196.google.com with SMTP id o15so2119268pgi.1 for ; Sat, 18 Apr 2020 08:13:42 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=rDs3m5c7QgR5dvRqd0B9A5lPWUbsdNS/xsLnI70SQss=; b=Y32CSBdMVLj5BVb72xkC3mQ2pMz5jPw1D+8bd/Pt5C6XhVIVKFPz02h+1Pv84tES+K WxOnKUxPslnAKuHqWu/0AetvzzZG+r8Hh9zsyNl8GOZd/4YNT1A08HMJKSYYqoSnGDn5 nymN0TsYz2fUTGNwjsh8GTxQqxN6WSN4+q0f4Gj3d1J/zPvmY4bgn2avHQwpSvtgYqbP ieaRfdrlfUvpqM88kWIwQPIloWr64lBX2nbSZEfJ7RPQPo1ZO6Th8HwW/zpvMMufFVZF a9ApWmLsuSQjDdU+JUBl5XaTkbZkcXIP/838fmLiyimw+APrrm1yNKbR7vUIYPnlZBLn I+cA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=rDs3m5c7QgR5dvRqd0B9A5lPWUbsdNS/xsLnI70SQss=; b=S6AIn6i9vGdlpKNo4yKX7hq7ft8ljq+zm2WFi+PVm4YK9WbetlCS7aKvenZFtfLzJc JiaEvc+jU+v9qQHrRM+YibCIPEdWpJ2R59hdc2K7neyV3AiCVqNPZYNJbwCII5EKMkuS WLZhkjsKxCOIqf5nYBtnBYc+NdpbFY74+4b0Np5Apk3ukPXQ7lyJn2k4vbk38xIaFQCj mXAnkK6BAf9JgnKey9Espi3Q6LoEQ7A1nclk5F6ZzXUAel6GmYtW7JRl6tr0JeMNf1KF CA5gTi9B5DtB+YdYJnjqlrODNv9iosgi6SwW7PwalmXCdrWj8tasRefrU6WeD8UTUbeL Bniw== X-Gm-Message-State: AGi0PuYNTaK9g9/ic/krZEW/ogZrNomW1e01jcovTdUd3AVOcjajrTkV Tof06oX0/CtbetLfiCeNZ5A= X-Google-Smtp-Source: APiQypIBndF88G89GiIx3faHP7ouPXOTAPd5mwmfOuwz9fUnPz140ZeUXGMUlq3WxEpnPBKd9Gnm/w== X-Received: by 2002:a62:62c3:: with SMTP id w186mr8264161pfb.238.1587222822222; Sat, 18 Apr 2020 08:13:42 -0700 (PDT) Received: from localhost.localdomain ([203.100.54.194]) by smtp.gmail.com with ESMTPSA id j32sm9613028pgb.55.2020.04.18.08.13.39 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 18 Apr 2020 08:13:41 -0700 (PDT) From: Yafang Shao To: hannes@cmpxchg.org, mhocko@kernel.org, vdavydov.dev@gmail.com, akpm@linux-foundation.org Cc: linux-mm@kvack.org, Yafang Shao Subject: [PATCH 3/3] memcg oom: bail out from the charge path if no victim found Date: Sat, 18 Apr 2020 11:13:11 -0400 Message-Id: <20200418151311.7397-4-laoar.shao@gmail.com> X-Mailer: git-send-email 2.18.1 In-Reply-To: <20200418151311.7397-1-laoar.shao@gmail.com> References: <20200418151311.7397-1-laoar.shao@gmail.com> 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: Without considering the manually triggered OOM, if no victim found in system OOM, the system will be deadlocked on memory, however if no victim found in memcg OOM, it can charge successfully and runs well. This behavior in memcg oom is not proper because that can prevent the memcg from being limited. Take an easy example. $ cd /sys/fs/cgroup/foo/ $ echo $$ > cgroup.procs $ echo 200M > memory.max $ cat memory.max 209715200 $ echo -1000 > /proc/$$/oom_score_adj Then, let's run a memhog task in memcg foo, which will allocate 1G memory and keeps running. $ /home/yafang/test/memhog & Then memory.current will be greater than memory.max. Run bellow command in another shell. $ cat /sys/fs/cgroup/foo/memory.current 1097228288 The tasks which have already allocated memory and won't allocate new memory still runs well. This behavior makes nonsense. This patch is to improve it. If no victim found in memcg oom, we should force the current task to wait until there's available pages. That is similar with the behavior in memcg1 when oom_kill_disable is set. In memcg2, the memcg oom can also be triggered manually - by reducing memory.max. We should distinguish this manually triggered memcg oom with other reguler memcg oom, so a magic key "oom_control.order == -2" is set in this situation. The tasks waiting in memcg oom will be waked up when we enlarge the memory.max. As these tasks is killable, we can also kill them directly. In memcg1, it cooperates well with memory.oom_control(oom_kill_disable). The tasks waiting in memcg1 oom can be waked up by enlarging memory.limit_in_bytes or disabling oom_kill_disable. As oom_kill_disable can be set manually, we should distinguish it with other reguler memcg oom as well. A member named memcg_oom_wait is introduced in struct task_struct to handle it. Signed-off-by: Yafang Shao --- include/linux/memcontrol.h | 8 ++++++-- include/linux/sched.h | 1 + mm/memcontrol.c | 25 +++++++++++++++++++++---- mm/oom_kill.c | 22 ++++++++++++++++++++++ 4 files changed, 50 insertions(+), 6 deletions(-) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 767bac135787..5df2c08e2720 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -569,7 +569,8 @@ static inline bool task_in_memcg_oom(struct task_struct *p) static inline enum oom_status task_in_memcg_oom_set(struct task_struct *p, struct mem_cgroup *memcg, gfp_t mask, - int order) + int order, + bool force) { if (!current->in_user_fault) return OOM_SKIPPED; @@ -578,6 +579,8 @@ static inline enum oom_status task_in_memcg_oom_set(struct task_struct *p, p->memcg_in_oom = memcg; p->memcg_oom_gfp_mask = mask; p->memcg_oom_order = order; + if (force) + p->memcg_oom_wait = true; return OOM_ASYNC; } @@ -1051,7 +1054,8 @@ static inline bool task_in_memcg_oom(struct task_struct *p) static inline enum oom_status task_in_memcg_oom_set(struct task_struct *p, struct mem_cgroup *memcg, gfp_t mask, - int order) + int order, + bool force) { return OOM_SUCCESS; } diff --git a/include/linux/sched.h b/include/linux/sched.h index 4418f5cb8324..cc1c7de7c248 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -774,6 +774,7 @@ struct task_struct { #endif #ifdef CONFIG_MEMCG unsigned in_user_fault:1; + unsigned memcg_oom_wait:1; #endif #ifdef CONFIG_COMPAT_BRK unsigned brk_randomized:1; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index d6cb1b786045..a637e13a4964 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1801,7 +1801,8 @@ static enum oom_status mem_cgroup_oom(struct mem_cgroup *memcg, gfp_t mask, int * victim and then we have to bail out from the charge path. */ if (memcg->oom_kill_disable) - return task_in_memcg_oom_set(current, memcg, mask, order); + return task_in_memcg_oom_set(current, memcg, mask, + order, false); mem_cgroup_mark_under_oom(memcg); @@ -1863,7 +1864,8 @@ bool mem_cgroup_oom_synchronize(bool handle) if (locked) mem_cgroup_oom_notify(memcg); - if (locked && !memcg->oom_kill_disable) { + if (locked && !memcg->oom_kill_disable && + !current->memcg_oom_wait) { mem_cgroup_unmark_under_oom(memcg); finish_wait(&memcg_oom_waitq, &owait.wait); mem_cgroup_out_of_memory(memcg, current->memcg_oom_gfp_mask, @@ -1883,7 +1885,10 @@ bool mem_cgroup_oom_synchronize(bool handle) */ memcg_oom_recover(memcg); } + cleanup: + if (current->memcg_oom_wait) + current->memcg_oom_wait = false; current->memcg_in_oom = NULL; css_put(&memcg->css); return true; @@ -6056,6 +6061,7 @@ static ssize_t memory_max_write(struct kernfs_open_file *of, struct mem_cgroup *memcg = mem_cgroup_from_css(of_css(of)); unsigned int nr_reclaims = MEM_CGROUP_RECLAIM_RETRIES; bool drained = false; + bool enlarge = false; unsigned long max; int err; @@ -6069,7 +6075,12 @@ static ssize_t memory_max_write(struct kernfs_open_file *of, for (;;) { unsigned long nr_pages = page_counter_read(&memcg->memory); - if (nr_pages <= max) + if (nr_pages < max) { + enlarge = true; + break; + } + + if (nr_pages == max) break; if (signal_pending(current)) @@ -6081,6 +6092,9 @@ static ssize_t memory_max_write(struct kernfs_open_file *of, continue; } + if (memcg->under_oom) + break; + if (nr_reclaims) { if (!try_to_free_mem_cgroup_pages(memcg, nr_pages - max, GFP_KERNEL, true)) @@ -6089,11 +6103,14 @@ static ssize_t memory_max_write(struct kernfs_open_file *of, } memcg_memory_event(memcg, MEMCG_OOM); - if (mem_cgroup_out_of_memory(memcg, GFP_KERNEL, 0) != + if (mem_cgroup_out_of_memory(memcg, GFP_KERNEL, -2) != OOM_SUCCESS) break; } + if (enlarge) + memcg_oom_recover(memcg); + memcg_wb_domain_size_changed(memcg); return nbytes; } diff --git a/mm/oom_kill.c b/mm/oom_kill.c index d5a941bea2d7..df564495c8b2 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -158,6 +158,15 @@ static inline bool is_sysrq_oom(struct oom_control *oc) return oc->order == -1; } +/* + * order == -2 means the oom kill is triggered by reducing memcg max, + * otherwise only for display puerposes. + */ +static bool is_max_write_oom(struct oom_control *oc) +{ + return oc->order == -2; +} + /* return true if the task is not adequate as candidate victim task. */ static bool oom_unkillable_task(struct task_struct *p) { @@ -1108,6 +1117,19 @@ enum oom_status out_of_memory(struct oom_control *oc) */ if (!is_sysrq_oom(oc) && !is_memcg_oom(oc)) panic("System is deadlocked on memory\n"); + + /* Bail out from the charge path if we can't find a victim. */ + if (is_memcg_oom(oc)) { + if (is_max_write_oom(oc)) + return OOM_SKIPPED; + + return task_in_memcg_oom_set(current, + oc->memcg, + oc->gfp_mask, + oc->order, + true); + } + } if (oc->chosen && oc->chosen != (void *)-1UL) oom_kill_process(oc, !is_memcg_oom(oc) ? "Out of memory" : -- 2.18.2