Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20190215024830.GA26477@jordon-HP-15-Notebook-PC>
 <20190728180611.GA20589@mail-itl> <CAFqt6zaMDnpB-RuapQAyYAub1t7oSdHH_pTD=f5k-s327ZvqMA@mail.gmail.com>
In-Reply-To: <CAFqt6zaMDnpB-RuapQAyYAub1t7oSdHH_pTD=f5k-s327ZvqMA@mail.gmail.com>
From:   Souptick Joarder <jrdr.linux@gmail.com>
Date:   Mon, 29 Jul 2019 14:02:54 +0530
Message-ID: <CAFqt6zY+07JBxAVfMqb+X78mXwFOj2VBh0nbR2tGnQOP9RrNkQ@mail.gmail.com>
Subject: Re: [Xen-devel] [PATCH v4 8/9] xen/gntdev.c: Convert to use vm_map_pages()
To:     =?UTF-8?Q?Marek_Marczykowski=2DG=C3=B3recki?= 
        <marmarek@invisiblethingslab.com>
Cc:     Andrew Morton <akpm@linux-foundation.org>,
        Matthew Wilcox <willy@infradead.org>,
        Michal Hocko <mhocko@suse.com>,
        Boris Ostrovsky <boris.ostrovsky@oracle.com>,
        Juergen Gross <jgross@suse.com>,
        Russell King - ARM Linux <linux@armlinux.org.uk>,
        robin.murphy@arm.com, xen-devel@lists.xenproject.org,
        linux-kernel@vger.kernel.org, Linux-MM <linux-mm@kvack.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Jul 29, 2019 at 1:35 PM Souptick Joarder <jrdr.linux@gmail.com> wro=
te:
>
> On Sun, Jul 28, 2019 at 11:36 PM Marek Marczykowski-G=C3=B3recki
> <marmarek@invisiblethingslab.com> wrote:
> >
> > On Fri, Feb 15, 2019 at 08:18:31AM +0530, Souptick Joarder wrote:
> > > Convert to use vm_map_pages() to map range of kernel
> > > memory to user vma.
> > >
> > > map->count is passed to vm_map_pages() and internal API
> > > verify map->count against count ( count =3D vma_pages(vma))
> > > for page array boundary overrun condition.
> >
> > This commit breaks gntdev driver. If vma->vm_pgoff > 0, vm_map_pages
> > will:
> >  - use map->pages starting at vma->vm_pgoff instead of 0
>
> The actual code ignores vma->vm_pgoff > 0 scenario and mapped
> the entire map->pages[i]. Why the entire map->pages[i] needs to be mapped
> if vma->vm_pgoff > 0 (in original code) ?
>
> are you referring to set vma->vm_pgoff =3D 0 irrespective of value passed
> from user space ? If yes, using vm_map_pages_zero() is an alternate
> option.
>
>
> >  - verify map->count against vma_pages()+vma->vm_pgoff instead of just
> >    vma_pages().
>
> In original code ->
>
> diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c
> index 559d4b7f807d..469dfbd6cf90 100644
> --- a/drivers/xen/gntdev.c
> +++ b/drivers/xen/gntdev.c
> @@ -1084,7 +1084,7 @@ static int gntdev_mmap(struct file *flip, struct
> vm_area_struct *vma)
> int index =3D vma->vm_pgoff;
> int count =3D vma_pages(vma);
>
> Count is user passed value.
>
> struct gntdev_grant_map *map;
> - int i, err =3D -EINVAL;
> + int err =3D -EINVAL;
> if ((vma->vm_flags & VM_WRITE) && !(vma->vm_flags & VM_SHARED))
> return -EINVAL;
> @@ -1145,12 +1145,9 @@ static int gntdev_mmap(struct file *flip,
> struct vm_area_struct *vma)
> goto out_put_map;
> if (!use_ptemod) {
> - for (i =3D 0; i < count; i++) {
> - err =3D vm_insert_page(vma, vma->vm_start + i*PAGE_SIZE,
> - map->pages[i]);
>
> and when count > i , we end up with trying to map memory outside
> boundary of map->pages[i], which was not correct.

typo.
s/count > i / count > map->count
>
> - if (err)
> - goto out_put_map;
> - }
> + err =3D vm_map_pages(vma, map->pages, map->count);
> + if (err)
> + goto out_put_map;
>
> With this commit, inside __vm_map_pages(), we have addressed this scenari=
o.
>
> +static int __vm_map_pages(struct vm_area_struct *vma, struct page **page=
s,
> + unsigned long num, unsigned long offset)
> +{
> + unsigned long count =3D vma_pages(vma);
> + unsigned long uaddr =3D vma->vm_start;
> + int ret, i;
> +
> + /* Fail if the user requested offset is beyond the end of the object */
> + if (offset > num)
> + return -ENXIO;
> +
> + /* Fail if the user requested size exceeds available object size */
> + if (count > num - offset)
> + return -ENXIO;
>
> By checking count > num -offset. (considering vma->vm_pgoff !=3D 0 as wel=
l).
> So we will never cross the boundary of map->pages[i].
>
>
> >
> > In practice, this breaks using a single gntdev FD for mapping multiple
> > grants.
>
> How ?
>
> >
> > It looks like vm_map_pages() is not a good fit for this code and IMO it
> > should be reverted.
>
> Did you hit any issue around this code in real time ?
>
>
> >
> > > Signed-off-by: Souptick Joarder <jrdr.linux@gmail.com>
> > > Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
> > > ---
> > >  drivers/xen/gntdev.c | 11 ++++-------
> > >  1 file changed, 4 insertions(+), 7 deletions(-)
> > >
> > > diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c
> > > index 5efc5ee..5d64262 100644
> > > --- a/drivers/xen/gntdev.c
> > > +++ b/drivers/xen/gntdev.c
> > > @@ -1084,7 +1084,7 @@ static int gntdev_mmap(struct file *flip, struc=
t vm_area_struct *vma)
> > >       int index =3D vma->vm_pgoff;
> > >       int count =3D vma_pages(vma);
> > >       struct gntdev_grant_map *map;
> > > -     int i, err =3D -EINVAL;
> > > +     int err =3D -EINVAL;
> > >
> > >       if ((vma->vm_flags & VM_WRITE) && !(vma->vm_flags & VM_SHARED))
> > >               return -EINVAL;
> > > @@ -1145,12 +1145,9 @@ static int gntdev_mmap(struct file *flip, stru=
ct vm_area_struct *vma)
> > >               goto out_put_map;
> > >
> > >       if (!use_ptemod) {
> > > -             for (i =3D 0; i < count; i++) {
> > > -                     err =3D vm_insert_page(vma, vma->vm_start + i*P=
AGE_SIZE,
> > > -                             map->pages[i]);
> > > -                     if (err)
> > > -                             goto out_put_map;
> > > -             }
> > > +             err =3D vm_map_pages(vma, map->pages, map->count);
> > > +             if (err)
> > > +                     goto out_put_map;
> > >       } else {
> > >  #ifdef CONFIG_X86
> > >               /*
> >
> > --
> > Best Regards,
> > Marek Marczykowski-G=C3=B3recki
> > Invisible Things Lab
> > A: Because it messes up the order in which people normally read text.
> > Q: Why is top-posting such a bad thing?