Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Fri, 25 Sep 2020 17:13:21 -0400
From:   Peter Xu <peterx@redhat.com>
To:     Linus Torvalds <torvalds@linux-foundation.org>
Cc:     Jason Gunthorpe <jgg@ziepe.ca>, John Hubbard <jhubbard@nvidia.com>,
        Linux-MM <linux-mm@kvack.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Jan Kara <jack@suse.cz>, Michal Hocko <mhocko@suse.com>,
        Kirill Tkhai <ktkhai@virtuozzo.com>,
        Kirill Shutemov <kirill@shutemov.name>,
        Hugh Dickins <hughd@google.com>,
        Christoph Hellwig <hch@lst.de>,
        Andrea Arcangeli <aarcange@redhat.com>,
        Oleg Nesterov <oleg@redhat.com>,
        Leon Romanovsky <leonro@nvidia.com>,
        Jann Horn <jannh@google.com>
Subject: Re: [PATCH 1/5] mm: Introduce mm_struct.has_pinned
Message-ID: <20200925211321.GC188812@xz-x1>
References: <20200923002735.GN19098@xz-x1>
 <20200923170759.GA9916@ziepe.ca>
 <20200924143517.GD79898@xz-x1>
 <20200924165152.GE9916@ziepe.ca>
 <20200924175531.GH79898@xz-x1>
 <20200924181501.GF9916@ziepe.ca>
 <20200924183418.GJ79898@xz-x1>
 <20200924183953.GG9916@ziepe.ca>
 <20200924213010.GL79898@xz-x1>
 <CAHk-=wgz5SXKA6-uZ_BimOP1C7pHJag0ndz=tnJDAZS_Z+FrGQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <CAHk-=wgz5SXKA6-uZ_BimOP1C7pHJag0ndz=tnJDAZS_Z+FrGQ@mail.gmail.com>
Precedence: bulk

On Fri, Sep 25, 2020 at 12:56:05PM -0700, Linus Torvalds wrote:
> So I think we can simply add a
> 
>         if (page_mapcount(page) != 1)
>                 return false;
> 
> to page_maybe_dma_pinned(), and that very naturally protects against
> the "is the page count perhaps elevated due to a lot of forking?"

How about the MAP_SHARED case where the page is pinned by some process but also
shared (so mapcount can be >1)?

> And honestly, since this is all getting fairly late in the rc, and it
> took longer than I thought, I think we should do the GFP_ATOMIC
> approach for now - not great, but since it only triggers for this case
> that really should never happen anyway, I think it's probably the best
> thing for 5.9, and we can improve on things later.

Sorry for that.  Maybe I should have moved even faster.

Would the ATOMIC version always work?  I mean, I thought it could fail anytime,
so any fork() can start to fail for the tests too.

PS. I do plan to post a GFP_KERNEL version soon today, no matter for this
release or the next one.

-- 
Peter Xu