From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934141AbdC3Vyb (ORCPT ); Thu, 30 Mar 2017 17:54:31 -0400 Received: from mail-io0-f176.google.com ([209.85.223.176]:34843 "EHLO mail-io0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755124AbdC3Vya (ORCPT ); Thu, 30 Mar 2017 17:54:30 -0400 MIME-Version: 1.0 In-Reply-To: References: <20170317231636.142311-1-timmurray@google.com> <20170330155123.GA3929@cmpxchg.org> From: Tim Murray Date: Thu, 30 Mar 2017 14:54:28 -0700 Message-ID: Subject: Re: [RFC 0/1] add support for reclaiming priorities per mem cgroup To: Johannes Weiner Cc: Michal Hocko , Vladimir Davydov , LKML , cgroups@vger.kernel.org, Linux-MM , Suren Baghdasaryan , Patrik Torstensson , Android Kernel Team Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Mar 30, 2017 at 12:40 PM, Tim Murray wrote: > The current critical vmpressure event > hasn't been that successful in avoiding oom-killer (on 3.18, at > least)--I've been able to get oom-killer to trigger without a > vmpressure event. Looked at this some more, and this is almost certainly because vmpressure relies on workqueues. Scheduling delay from CFS workqueues would explain vmpressure latency that results in oom-killer running long before the critical vmpressure notification is received in userspace, even if userspace is running as FIFO. We regularly see 10ms+ latency on workqueues, even when an Android device isn't heavily loaded. From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tim Murray Subject: Re: [RFC 0/1] add support for reclaiming priorities per mem cgroup Date: Thu, 30 Mar 2017 14:54:28 -0700 Message-ID: References: <20170317231636.142311-1-timmurray@google.com> <20170330155123.GA3929@cmpxchg.org> Mime-Version: 1.0 Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20161025; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=r3Yj4HRF2OSIooVWy78Z4jt9YmZTFQ2lyumYJKPj8Dw=; b=v7w5QhdN8Z6qFo1eBMEJ8O+lMajtb+WKPkkbh/4cSunFvk/7+S7PLAsjF43kb7iitB +x8l9SsL08vv/jwiRfc6SBlKpDBwDnHHqKD9o+A+XC0wowl2xM7H+JVhh8dKN0Cpq7rv jBn4y+zmEmFfl8+9Pa20uOeJ6K24IbqMO1heiYJ2+BD/uow8AySBRuEFLeFx1PF6uWBS mId/iIIRC9XVw9T0GplOigqD90hm+AUbbjeI4U9/PbTu/CxBoPNQttECzG1qStivQIt6 ms3lIluSUZfSpRVDUVf80rWnHRJq5G1/PRVVtap3Z4a7ndt2M/NkjQAwQtZt2O+trVKv 4vRA== In-Reply-To: Sender: owner-linux-mm@kvack.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Johannes Weiner Cc: Michal Hocko , Vladimir Davydov , LKML , cgroups@vger.kernel.org, Linux-MM , Suren Baghdasaryan , Patrik Torstensson , Android Kernel Team On Thu, Mar 30, 2017 at 12:40 PM, Tim Murray wrote: > The current critical vmpressure event > hasn't been that successful in avoiding oom-killer (on 3.18, at > least)--I've been able to get oom-killer to trigger without a > vmpressure event. Looked at this some more, and this is almost certainly because vmpressure relies on workqueues. Scheduling delay from CFS workqueues would explain vmpressure latency that results in oom-killer running long before the critical vmpressure notification is received in userspace, even if userspace is running as FIFO. We regularly see 10ms+ latency on workqueues, even when an Android device isn't heavily loaded. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org