From: Michal Hocko Subject: Re: mm: mkfs.ext4 invoked oom-killer on i386 - pagecache_get_page Date: Thu, 18 Jun 2020 14:41:29 +0200 Message-ID: <20200618124129.GC15447@dhcp22.suse.cz> References: <20200521163450.GV6462@dhcp22.suse.cz> <20200617135758.GA548179@chrisdown.name> <20200617141155.GQ9499@dhcp22.suse.cz> <20200617160624.GS9499@dhcp22.suse.cz> <20200617210935.GA578452@chrisdown.name> <20200618123743.GA694719@chrisdown.name> Mime-Version: 1.0 Return-path: Content-Disposition: inline In-Reply-To: <20200618123743.GA694719@chrisdown.name> Sender: linux-block-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Chris Down Cc: Yafang Shao , Naresh Kamboju , Anders Roxell , "Linux F2FS DEV, Mailing List" , linux-ext4 , linux-block , Andrew Morton , open list , Linux-Next Mailing List , linux-mm , Arnd Bergmann , Andreas Dilger , Jaegeuk Kim , Theodore Ts'o , Chao Yu , Hugh Dickins , Andrea Arcangeli , Matthew Wilcox , Chao Yu , lkf On Thu 18-06-20 13:37:43, Chris Down wrote: > Yafang Shao writes: > > On Thu, Jun 18, 2020 at 5:09 AM Chris Down wrote: > > > > > > Naresh Kamboju writes: > > > >After this patch applied the reported issue got fixed. > > > > > > Great! Thank you Naresh and Michal for helping to get to the bottom of this :-) > > > > > > I'll send out a new version tomorrow with the fixes applied and both of you > > > credited in the changelog for the detection and fix. > > > > As we have already found that the usage around memory.{emin, elow} has > > many limitations, I think memory.{emin, elow} should be used for > > memcg-tree internally only, that means they can only be used to > > calculate the protection of a memcg in a specified memcg-tree but > > should not be exposed to other MM parts. > > I agree that the current semantics are mentally taxing and we should > generally avoid exposing the implementation details outside of memcg where > possible. Do you have a suggested rework? :-) I would really prefer to do that work on top of the fixes we (used to) have in mmotm (with the fixup). -- Michal Hocko SUSE Labs