Subject: Re: [patch 1/2] mm, mmu_notifier: annotate mmu notifiers with
 blockable invalidate callbacks
To: David Rientjes <rientjes@google.com>,
        Andrew Morton <akpm@linux-foundation.org>
Cc: Michal Hocko <mhocko@suse.com>, Andrea Arcangeli <aarcange@redhat.com>,
        Benjamin Herrenschmidt <benh@kernel.crashing.org>,
        Paul Mackerras <paulus@samba.org>, Oded Gabbay <oded.gabbay@gmail.com>,
        Alex Deucher <alexander.deucher@amd.com>,
        =?UTF-8?Q?Christian_K=c3=b6nig?= <christian.koenig@amd.com>,
        David Airlie <airlied@linux.ie>, Joerg Roedel <joro@8bytes.org>,
        Doug Ledford <dledford@redhat.com>,
        Jani Nikula <jani.nikula@linux.intel.com>,
        Mike Marciniszyn <mike.marciniszyn@intel.com>,
        Sean Hefty <sean.hefty@intel.com>, Dimitri Sivanich <sivanich@sgi.com>,
        Boris Ostrovsky <boris.ostrovsky@oracle.com>,
        =?UTF-8?B?SsOpcsO0bWUgR2xpc3Nl?= <jglisse@redhat.com>,
        =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org
References: <alpine.DEB.2.10.1712111409090.196232@chino.kir.corp.google.com>
From: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <40828fec-a375-fb90-f4f1-fc647651c2f7@redhat.com>
Date: Mon, 11 Dec 2017 23:23:54 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.5.0
MIME-Version: 1.0
In-Reply-To: <alpine.DEB.2.10.1712111409090.196232@chino.kir.corp.google.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1071
Lines: 24

On 11/12/2017 23:11, David Rientjes wrote:
> Commit 4d4bbd8526a8 ("mm, oom_reaper: skip mm structs with mmu notifiers")
> prevented the oom reaper from unmapping private anonymous memory with the
> oom reaper when the oom victim mm had mmu notifiers registered.
> 
> The rationale is that doing mmu_notifier_invalidate_range_{start,end}()
> around the unmap_page_range(), which is needed, can block and the oom
> killer will stall forever waiting for the victim to exit, which may not
> be possible without reaping.
> 
> That concern is real, but only true for mmu notifiers that have blockable
> invalidate_range_{start,end}() callbacks.  This patch adds a "flags" field
> for mmu notifiers that can set a bit to indicate that these callbacks do
> block.

Why not put the flag in the ops, since the same ops should always be
either blockable or unblockable?

Paolo

> The implementation is steered toward an expensive slowpath, such as after
> the oom reaper has grabbed mm->mmap_sem of a still alive oom victim.
> 
> Signed-off-by: David Rientjes <rientjes@google.com>