Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <000000000000258e5e05fae79fc1@google.com> <20230507135844.1231056-1-lrh2000@pku.edu.cn>
 <CA+CK2bBe2YKYM3rUTCnZ0RF=NFUR9VqO-QYn3ygPsFJWLY1MUA@mail.gmail.com>
 <ZFlrbDft1QfMyIDc@casper.infradead.org> <CA+CK2bDVjovwB9v-Zv4Fn7EUfp5FV2XK36iJKYKY7pYNOFfOGA@mail.gmail.com>
 <ZFlvJEfs1ufh1UUD@casper.infradead.org> <CA+CK2bDC-FVv1tZg9MDn-N735Ak3OAtdZPf+LEYM-JHsO90YcQ@mail.gmail.com>
 <fa1dac7a-406e-30ea-6aba-ded2e0e871fa@redhat.com> <CA+CK2bAHbHHwLUoGJkz8n6mrM5dy7oMojeNksdVOMYn+qFYngA@mail.gmail.com>
 <CA+CK2bD=bv1vPGZaNcxDy-uUFj2ZAKkdPmAtJaweXhgTe91oEw@mail.gmail.com> <366ab078-1101-421c-691d-34f5efe006b5@redhat.com>
In-Reply-To: <366ab078-1101-421c-691d-34f5efe006b5@redhat.com>
From:   Pasha Tatashin <pasha.tatashin@soleen.com>
Date:   Mon, 8 May 2023 17:07:26 -0700
Message-ID: <CA+CK2bBwYGd2_CVQ4Nr87CGEjp7BqVbBeoP++yHB7Cr9_peBOQ@mail.gmail.com>
Subject: Re: usbdev_mmap causes type confusion in page_table_check
To:     David Hildenbrand <david@redhat.com>
Cc:     Matthew Wilcox <willy@infradead.org>,
        Ruihan Li <lrh2000@pku.edu.cn>,
        syzbot+fcf1a817ceb50935ce99@syzkaller.appspotmail.com,
        akpm@linux-foundation.org, linux-kernel@vger.kernel.org,
        linux-mm@kvack.org, gregkh@linuxfoundation.org,
        linux-usb@vger.kernel.org, syzkaller-bugs@googlegroups.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Mon, May 8, 2023 at 4:37=E2=80=AFPM David Hildenbrand <david@redhat.com>=
 wrote:
>
> On 09.05.23 01:21, Pasha Tatashin wrote:
> >> For normal Kernel-MM operations, vm_normal_page() should be used to
> >> get "struct page" based on vma+addr+pte combination, but
> >> page_table_check does not use vma for its operation in order to
> >> strengthen the verification of no invalid page sharing. But, even
>
> I'm not sure if that's the right approach for this case here, though.
>
> >> vm_normal_page() can cause access to the "struct page" for VM_PFNMAP
> >> if pfn_valid(pfn) is true. So, vm_normal_page() can return a struct
> >> page for a user mapped slab page.
> >
> > Only for !ARCH_HAS_PTE_SPECIAL case, otherwise NULL is returned.
>
> That would violate VM_PFNMAP semantics, though. I remember that there
> was a trick to it.
>
> Assuming we map /dev/mem, what stops a page we mapped and determined to
> be !anon to be freed and reused, such that we suddenly have an anon page
> mappped?
>
> In that case, we really don't want to look at the "struct page" ever, no?

Good point. page_table_check just does not work well /dev/mem. I am
thinking of adding BUG_ON(PageSlab(page); and also "depends on
!DEVMEM" for the PAGE_TABLE_CHECK config option.

Pasha