dma-buf: Move dma_buf_release() from fops to dentry_ops

  Charan Teja reported a 'use-after-free' in dmabuffs_dname [1], which
happens if the dma_buf_release() is called while the userspace is
accessing the dma_buf pseudo fs's dmabuffs_dname() in another process,
and dma_buf_release() releases the dmabuf object when the last reference
to the struct file goes away.

I discussed with Arnd Bergmann, and he suggested that rather than tying
the dma_buf_release() to the file_operations' release(), we can tie it to
the dentry_operations' d_release(), which will be called when the last ref
to the dentry is removed.

The path exercised by __fput() calls f_op->release() first, and then calls
dput, which eventually calls d_op->d_release().

In the 'normal' case, when no userspace access is happening via dma_buf
pseudo fs, there should be exactly one fd, file, dentry and inode, so
closing the fd will kill of everything right away.

In the presented case, the dentry's d_release() will be called only when
the dentry's last ref is released.

Therefore, lets move dma_buf_release() from fops->release() to
d_ops->d_release().

Many thanks to Arnd for his FS insights :)

[1]: https://lore.kernel.org/patchwork/patch/1238278/

Fixes: bb2bb9030425 ("dma-buf: add DMA_BUF_SET_NAME ioctls")
Reported-by: syzbot+3643a18836bce555bff6@syzkaller.appspotmail.com
Cc: <stable@vger.kernel.org> [5.3+]
Cc: Arnd Bergmann <arnd@arndb.de>
Reported-by: Charan Teja Reddy <charante@codeaurora.org>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
---
 drivers/dma-buf/dma-buf.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)
  

On Wed, Jun 10, 2020 at 10:33 AM Sumit Semwal <sumit.semwal@linaro.org> wrote:
>
> Charan Teja reported a 'use-after-free' in dmabuffs_dname [1], which
> happens if the dma_buf_release() is called while the userspace is
> accessing the dma_buf pseudo fs's dmabuffs_dname() in another process,
> and dma_buf_release() releases the dmabuf object when the last reference
> to the struct file goes away.
>
> I discussed with Arnd Bergmann, and he suggested that rather than tying
> the dma_buf_release() to the file_operations' release(), we can tie it to
> the dentry_operations' d_release(), which will be called when the last ref
> to the dentry is removed.
>
> The path exercised by __fput() calls f_op->release() first, and then calls
> dput, which eventually calls d_op->d_release().
>
> In the 'normal' case, when no userspace access is happening via dma_buf
> pseudo fs, there should be exactly one fd, file, dentry and inode, so
> closing the fd will kill of everything right away.
>
> In the presented case, the dentry's d_release() will be called only when
> the dentry's last ref is released.
>
> Therefore, lets move dma_buf_release() from fops->release() to
> d_ops->d_release().
>
> Many thanks to Arnd for his FS insights :)
>
> [1]: https://lore.kernel.org/patchwork/patch/1238278/
>
> Fixes: bb2bb9030425 ("dma-buf: add DMA_BUF_SET_NAME ioctls")
> Reported-by: syzbot+3643a18836bce555bff6@syzkaller.appspotmail.com
> Cc: <stable@vger.kernel.org> [5.3+]
> Cc: Arnd Bergmann <arnd@arndb.de>
> Reported-by: Charan Teja Reddy <charante@codeaurora.org>
> Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>

The patch looks correct to me.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>

Obviously this should still be verified against the original report if possible.

>  drivers/dma-buf/dma-buf.c | 13 +++++++------
>  1 file changed, 7 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c
> index 01ce125f8e8d..92ba4b6ef3e7 100644
> --- a/drivers/dma-buf/dma-buf.c
> +++ b/drivers/dma-buf/dma-buf.c
> @@ -54,8 +54,11 @@ static char *dmabuffs_dname(struct dentry *dentry, char *buffer, int buflen)
>                              dentry->d_name.name, ret > 0 ? name : "");
>  }
>
> +static void dma_buf_release(struct dentry *dentry);
> +
>  static const struct dentry_operations dma_buf_dentry_ops = {
>         .d_dname = dmabuffs_dname,
> +       .d_release = dma_buf_release,
>  };

I'd suggest rearranging the file to avoid the forward declaration, even
if it makes it a little harder to review the change, the resulting code
will remain organized more logically.

>  static struct vfsmount *dma_buf_mnt;
> @@ -77,14 +80,14 @@ static struct file_system_type dma_buf_fs_type = {
>         .kill_sb = kill_anon_super,
>  };
>
> -static int dma_buf_release(struct inode *inode, struct file *file)
> +static void dma_buf_release(struct dentry *dentry)
>  {
>         struct dma_buf *dmabuf;
>
> -       if (!is_dma_buf_file(file))
> -               return -EINVAL;
> +       if (dentry->d_op != &dma_buf_dentry_ops)
> +               return;

I think the check here is redundant and it's clearer without it.

          Arnd
  

On Wed, 10 Jun 2020 at 14:57, Arnd Bergmann <arnd@arndb.de> wrote:
>
> On Wed, Jun 10, 2020 at 10:33 AM Sumit Semwal <sumit.semwal@linaro.org> wrote:
> >
> > Charan Teja reported a 'use-after-free' in dmabuffs_dname [1], which
> > happens if the dma_buf_release() is called while the userspace is
> > accessing the dma_buf pseudo fs's dmabuffs_dname() in another process,
> > and dma_buf_release() releases the dmabuf object when the last reference
> > to the struct file goes away.
> >
> > I discussed with Arnd Bergmann, and he suggested that rather than tying
> > the dma_buf_release() to the file_operations' release(), we can tie it to
> > the dentry_operations' d_release(), which will be called when the last ref
> > to the dentry is removed.
> >
> > The path exercised by __fput() calls f_op->release() first, and then calls
> > dput, which eventually calls d_op->d_release().
> >
> > In the 'normal' case, when no userspace access is happening via dma_buf
> > pseudo fs, there should be exactly one fd, file, dentry and inode, so
> > closing the fd will kill of everything right away.
> >
> > In the presented case, the dentry's d_release() will be called only when
> > the dentry's last ref is released.
> >
> > Therefore, lets move dma_buf_release() from fops->release() to
> > d_ops->d_release().
> >
> > Many thanks to Arnd for his FS insights :)
> >
> > [1]: https://lore.kernel.org/patchwork/patch/1238278/
> >
> > Fixes: bb2bb9030425 ("dma-buf: add DMA_BUF_SET_NAME ioctls")
> > Reported-by: syzbot+3643a18836bce555bff6@syzkaller.appspotmail.com
> > Cc: <stable@vger.kernel.org> [5.3+]
> > Cc: Arnd Bergmann <arnd@arndb.de>
> > Reported-by: Charan Teja Reddy <charante@codeaurora.org>
> > Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
>
> The patch looks correct to me.
>
> Reviewed-by: Arnd Bergmann <arnd@arndb.de>
>
> Obviously this should still be verified against the original report if possible.
Thanks, Arnd!
>
> >  drivers/dma-buf/dma-buf.c | 13 +++++++------
> >  1 file changed, 7 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c
> > index 01ce125f8e8d..92ba4b6ef3e7 100644
> > --- a/drivers/dma-buf/dma-buf.c
> > +++ b/drivers/dma-buf/dma-buf.c
> > @@ -54,8 +54,11 @@ static char *dmabuffs_dname(struct dentry *dentry, char *buffer, int buflen)
> >                              dentry->d_name.name, ret > 0 ? name : "");
> >  }
> >
> > +static void dma_buf_release(struct dentry *dentry);
> > +
> >  static const struct dentry_operations dma_buf_dentry_ops = {
> >         .d_dname = dmabuffs_dname,
> > +       .d_release = dma_buf_release,
> >  };
>
> I'd suggest rearranging the file to avoid the forward declaration, even
> if it makes it a little harder to review the change, the resulting code
> will remain organized more logically.
Got it, will update it in v2.
>
> >  static struct vfsmount *dma_buf_mnt;
> > @@ -77,14 +80,14 @@ static struct file_system_type dma_buf_fs_type = {
> >         .kill_sb = kill_anon_super,
> >  };
> >
> > -static int dma_buf_release(struct inode *inode, struct file *file)
> > +static void dma_buf_release(struct dentry *dentry)
> >  {
> >         struct dma_buf *dmabuf;
> >
> > -       if (!is_dma_buf_file(file))
> > -               return -EINVAL;
> > +       if (dentry->d_op != &dma_buf_dentry_ops)
> > +               return;
>
> I think the check here is redundant and it's clearer without it.
Ok, will remove.
>
>           Arnd

Best,
Sumit.