Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933186AbcK1Ote (ORCPT ); Mon, 28 Nov 2016 09:49:34 -0500 Received: from mail-wm0-f68.google.com ([74.125.82.68]:34672 "EHLO mail-wm0-f68.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933064AbcK1OtY (ORCPT ); Mon, 28 Nov 2016 09:49:24 -0500 Date: Mon, 28 Nov 2016 15:49:17 +0100 From: Michal Hocko To: Hui Zhu Cc: minchan@kernel.org, ngupta@vflare.org, sergey.senozhatsky.work@gmail.com, dan.j.williams@intel.com, jthumshirn@suse.de, akpm@linux-foundation.org, re.emese@gmail.com, andriy.shevchenko@linux.intel.com, vishal.l.verma@intel.com, hannes@cmpxchg.org, mgorman@techsingularity.net, vbabka@suse.cz, vdavydov.dev@gmail.com, kirill.shutemov@linux.intel.com, ying.huang@intel.com, yang.shi@linaro.org, dave.hansen@linux.intel.com, willy@linux.intel.com, vkuznets@redhat.com, vitalywool@gmail.com, jmarchan@redhat.com, lstoakes@gmail.com, geliangtang@163.com, viro@zeniv.linux.org.uk, hughd@google.com, riel@redhat.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org, teawater@gmail.com Subject: Re: [RFC 0/2] Add interface let ZRAM close swap cache Message-ID: <20161128144917.GQ14788@dhcp22.suse.cz> References: <1480062313-7361-1-git-send-email-zhuhui@xiaomi.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1480062313-7361-1-git-send-email-zhuhui@xiaomi.com> User-Agent: Mutt/1.6.0 (2016-04-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 640 Lines: 17 On Fri 25-11-16 16:25:11, Hui Zhu wrote: > SWAP will keep before swap cache before swap space get full. It will > make swap space cannot be freed. It is harmful to the system that use > ZRAM because its space use memory too. I have hard time to follow what you are trying to say here. Could you be more specific about what is the actual problem? > This two patches will add a sysfs switch to ZRAM that open or close swap > cache without check the swap space. I find this a crude hack to be honest. Please make sure to describe why the swap cache stays in the way and why this is not problem in other setups. -- Michal Hocko SUSE Labs