Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
Date:   Tue, 15 Feb 2022 20:36:05 +0100
From:   Michal Hocko <mhocko@suse.com>
To:     Suren Baghdasaryan <surenb@google.com>
Cc:     Yang Shi <shy828301@gmail.com>,
        syzbot <syzbot+2ccf63a4bd07cf39cab0@syzkaller.appspotmail.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Christian Brauner <brauner@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Linux MM <linux-mm@kvack.org>,
        syzkaller-bugs <syzkaller-bugs@googlegroups.com>
Subject: Re: [syzbot] KASAN: use-after-free Read in __oom_reap_task_mm
Message-ID: <YgwApdvLhxFU26j5@dhcp22.suse.cz>
References: <00000000000072ef2c05d7f81950@google.com>
 <CAHbLzkpr=xHC7C2+wkOGUB24924T4L1Lx_v+m1Ke8Ku3ym7wYw@mail.gmail.com>
 <Ygt5cAS78BFxVwkg@dhcp22.suse.cz>
 <CAJuCfpGf8HLh=tfrnK3VFd2QO9bgzuM9HyDY8PYYxJkbdbmYRA@mail.gmail.com>
 <CAHbLzkpKGq8XaLx3G1=oJ-TsZiqC+A+CaVK5=Mjzo1Jo0qA3DQ@mail.gmail.com>
 <CAJuCfpE833u+Z909yaLgGEJV7O10fVw-_eG1nH=w4b3+PbxrLg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAJuCfpE833u+Z909yaLgGEJV7O10fVw-_eG1nH=w4b3+PbxrLg@mail.gmail.com>
Precedence: bulk

On Tue 15-02-22 10:10:53, Suren Baghdasaryan wrote:
> On Tue, Feb 15, 2022 at 9:53 AM Yang Shi <shy828301@gmail.com> wrote:
[...]
> > Isn't the below race possible?
> >
> >           CPU A                                     CPU B
> > exiting:
> >   mmap_write_lock
> >   remove_vma()
> >   mmap_write_unlock
> >                                                 process_mrelease:
> >                                                   mmap_read_lock
> >                                                   __oom_reap_task_mm
> >                                                   mmap_read_unlock
> >
> 
> Sure, that sequence (would not call it a race) is possible but in this
> case __oom_reap_task_mm will find no vmas in the mm because exit_mmap
> freed and removed all of them.

I didn't really have chance to have a closer look but I do not see
exit_mmap doing mm->mmap = NULL so the pointer can be a freed vma unless
I am missing something. I thought we've had it in your patches? Has this
got lost somewhere in the process?
-- 
Michal Hocko
SUSE Labs