From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?UTF-8?Q?Christian_K=c3=b6nig?= Subject: Re: [PATCH] mm, oom: distinguish blockable mode for mmu notifiers Date: Fri, 24 Aug 2018 15:28:33 +0200 Message-ID: <86bd94d5-0ce8-c67f-07a5-ca9ebf399cdd@gmail.com> References: <20180824113248.GH29735@dhcp22.suse.cz> <20180824115226.GK29735@dhcp22.suse.cz> <20180824120339.GL29735@dhcp22.suse.cz> <20180824123341.GN29735@dhcp22.suse.cz> <20180824130132.GP29735@dhcp22.suse.cz> <23d071d2-82e4-9b78-1000-be44db5f6523@gmail.com> <20180824132442.GQ29735@dhcp22.suse.cz> Reply-To: christian.koenig@amd.com Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit Return-path: In-Reply-To: <20180824132442.GQ29735@dhcp22.suse.cz> Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org To: Michal Hocko , christian.koenig@amd.com Cc: kvm@vger.kernel.org, =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= , Tetsuo Handa , Sudeep Dutt , dri-devel@lists.freedesktop.org, linux-mm@kvack.org, Andrea Arcangeli , Dimitri Sivanich , Jason Gunthorpe , linux-rdma@vger.kernel.org, amd-gfx@lists.freedesktop.org, David Airlie , Doug Ledford , David Rientjes , xen-devel@lists.xenproject.org, intel-gfx@lists.freedesktop.org, Leon Romanovsky , =?UTF-8?B?SsOpcsO0bWUgR2xpc3Nl?= , Rodrigo Vivi , Boris Ostrovsky , Juergen Gross List-Id: dri-devel@lists.freedesktop.org Am 24.08.2018 um 15:24 schrieb Michal Hocko: > On Fri 24-08-18 15:10:08, Christian König wrote: >> Am 24.08.2018 um 15:01 schrieb Michal Hocko: >>> On Fri 24-08-18 14:52:26, Christian König wrote: >>>> Am 24.08.2018 um 14:33 schrieb Michal Hocko: >>> [...] >>>>> Thiking about it some more, I can imagine that a notifier callback which >>>>> performs an allocation might trigger a memory reclaim and that in turn >>>>> might trigger a notifier to be invoked and recurse. But notifier >>>>> shouldn't really allocate memory. They are called from deep MM code >>>>> paths and this would be extremely deadlock prone. Maybe Jerome can come >>>>> up some more realistic scenario. If not then I would propose to simplify >>>>> the locking here. We have lockdep to catch self deadlocks and it is >>>>> always better to handle a specific issue rather than having a code >>>>> without a clear indication how it can recurse. >>>> Well I agree that we should probably fix that, but I have some concerns to >>>> remove the existing workaround. >>>> >>>> See we added that to get rid of a real problem in a customer environment and >>>> I don't want to that to show up again. >>> It would really help to know more about that case and fix it properly >>> rather than workaround it like this. Anyway, let me think how to handle >>> the non-blocking notifier invocation then. I was not able to come up >>> with anything remotely sane yet. >> With avoiding allocating memory in the write lock path I don't see an issue >> any more with that. >> >> All what the write lock path does now is adding items to a linked lists, >> arrays etc.... > Can we change it to non-sleepable lock then? No, the write side doesn't sleep any more, but the read side does. See amdgpu_mn_invalidate_node() and that is where you actually need to handle the non-blocking flag correctly. Christian.