free_pardevice() is called by parport_unregister_device() and already frees
pp->pdev->name, don't try to do it again.
This bug causes kernel crashes.
I found and verified this with KASAN and some added pr_emerg()s:
[ 60.316568] pp_release: pp->pdev->name == ffff88039cb264c0
[ 60.316692] free_pardevice: freeing par_dev->name at ffff88039cb264c0
[ 60.316706] pp_release: kfree(ffff88039cb264c0)
[ 60.316714] ==========================================================
[ 60.316722] BUG: Double free or freeing an invalid pointer
[ 60.316731] Unexpected shadow byte: 0xFB
[ 60.316801] Object at ffff88039cb264c0, in cache kmalloc-32 size: 32
[ 60.316813] Allocated:
[ 60.316824] PID = 1695
[ 60.316869] Freed:
[ 60.316880] PID = 1695
[ 60.316935] ==========================================================
Signed-off-by: Jann Horn <[email protected]>
---
drivers/char/ppdev.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/char/ppdev.c b/drivers/char/ppdev.c
index d23368874710..6af1ce04b3da 100644
--- a/drivers/char/ppdev.c
+++ b/drivers/char/ppdev.c
@@ -748,10 +748,7 @@ static int pp_release(struct inode *inode, struct file *file)
}
if (pp->pdev) {
- const char *name = pp->pdev->name;
-
parport_unregister_device(pp->pdev);
- kfree(name);
pp->pdev = NULL;
pr_debug(CHRDEV "%x: unregistered pardevice\n", minor);
}
--
2.1.4
On Sun, Oct 30, 2016 at 11:19:24PM +0100, Jann Horn wrote:
> free_pardevice() is called by parport_unregister_device() and already frees
> pp->pdev->name, don't try to do it again.
>
> This bug causes kernel crashes.
>
> I found and verified this with KASAN and some added pr_emerg()s:
>
> [ 60.316568] pp_release: pp->pdev->name == ffff88039cb264c0
> [ 60.316692] free_pardevice: freeing par_dev->name at ffff88039cb264c0
> [ 60.316706] pp_release: kfree(ffff88039cb264c0)
> [ 60.316714] ==========================================================
> [ 60.316722] BUG: Double free or freeing an invalid pointer
> [ 60.316731] Unexpected shadow byte: 0xFB
> [ 60.316801] Object at ffff88039cb264c0, in cache kmalloc-32 size: 32
> [ 60.316813] Allocated:
> [ 60.316824] PID = 1695
> [ 60.316869] Freed:
> [ 60.316880] PID = 1695
> [ 60.316935] ==========================================================
>
> Signed-off-by: Jann Horn <[email protected]>
> ---
> drivers/char/ppdev.c | 3 ---
> 1 file changed, 3 deletions(-)
Sudip, I need an ack here before I can take this...
On Sun, Oct 30, 2016 at 11:19:24PM +0100, Jann Horn wrote:
> free_pardevice() is called by parport_unregister_device() and already frees
> pp->pdev->name, don't try to do it again.
>
> This bug causes kernel crashes.
>
> I found and verified this with KASAN and some added pr_emerg()s:
>
> [ 60.316568] pp_release: pp->pdev->name == ffff88039cb264c0
> [ 60.316692] free_pardevice: freeing par_dev->name at ffff88039cb264c0
> [ 60.316706] pp_release: kfree(ffff88039cb264c0)
> [ 60.316714] ==========================================================
> [ 60.316722] BUG: Double free or freeing an invalid pointer
> [ 60.316731] Unexpected shadow byte: 0xFB
> [ 60.316801] Object at ffff88039cb264c0, in cache kmalloc-32 size: 32
> [ 60.316813] Allocated:
> [ 60.316824] PID = 1695
> [ 60.316869] Freed:
> [ 60.316880] PID = 1695
> [ 60.316935] ==========================================================
>
> Signed-off-by: Jann Horn <[email protected]>
Acked-by: Sudip Mukherjee <[email protected]>
On Sunday, October 30, 2016 11:19:24 PM CET Jann Horn wrote:
> diff --git a/drivers/char/ppdev.c b/drivers/char/ppdev.c
> index d23368874710..6af1ce04b3da 100644
> --- a/drivers/char/ppdev.c
> +++ b/drivers/char/ppdev.c
> @@ -748,10 +748,7 @@ static int pp_release(struct inode *inode, struct file *file)
> }
>
> if (pp->pdev) {
> - const char *name = pp->pdev->name;
> -
> parport_unregister_device(pp->pdev);
> - kfree(name);
> pp->pdev = NULL;
> pr_debug(CHRDEV "%x: unregistered pardevice\n", minor);
> }
>
I took a closer look at this and found that we also leak the name
that is passed in register_device() in the same file:
name = kasprintf(GFP_KERNEL, CHRDEV "%x", minor);
...
pdev = parport_register_dev_model(port, name, &ppdev_cb, minor);
parport_register_dev_model() copies the name using kstrdup() and
we should really free it after parport_register_dev_model().
It's not a huge problem, just leaking a few bytes of memory, but
the extra kfree() probably came from this confusion.
Arnd
On Thu, Nov 10, 2016 at 02:18:12PM +0100, Arnd Bergmann wrote:
> On Sunday, October 30, 2016 11:19:24 PM CET Jann Horn wrote:
> > diff --git a/drivers/char/ppdev.c b/drivers/char/ppdev.c
> > index d23368874710..6af1ce04b3da 100644
> > --- a/drivers/char/ppdev.c
> > +++ b/drivers/char/ppdev.c
> > @@ -748,10 +748,7 @@ static int pp_release(struct inode *inode, struct file *file)
> > }
> >
> > if (pp->pdev) {
> > - const char *name = pp->pdev->name;
> > -
> > parport_unregister_device(pp->pdev);
> > - kfree(name);
> > pp->pdev = NULL;
> > pr_debug(CHRDEV "%x: unregistered pardevice\n", minor);
> > }
> >
>
> I took a closer look at this and found that we also leak the name
> that is passed in register_device() in the same file:
>
> name = kasprintf(GFP_KERNEL, CHRDEV "%x", minor);
> ...
> pdev = parport_register_dev_model(port, name, &ppdev_cb, minor);
>
> parport_register_dev_model() copies the name using kstrdup() and
> we should really free it after parport_register_dev_model().
yes. I missed that while converting the driver to use device model.
Thanks. I will send a patch to fix this tonight.
>
> It's not a huge problem, just leaking a few bytes of memory, but
> the extra kfree() probably came from this confusion.
No, it was the old code which was not in the device-model. The old code
uses parport_register_device() which just uses the same string that is
passed to it and so it was freed while releasing the device.
Regards
Sudip