diff mbox series

ieee802154: ca8210: Fix a potential UAF in ca8210_probe

Message ID 20230925072423.24772-1-dinghao.liu@zju.edu.cn (mailing list archive)
State Superseded
Headers show
Series ieee802154: ca8210: Fix a potential UAF in ca8210_probe | expand

Commit Message

Dinghao Liu Sept. 25, 2023, 7:24 a.m. UTC
If of_clk_add_provider() fails in ca8210_register_ext_clock(),
it calls clk_unregister() to release priv->clk and returns an
error. However, the caller ca8210_probe() then calls ca8210_remove(),
where priv->clk is freed again in ca8210_unregister_ext_clock(). In
this case, a use-after-free may happen in the second time we call
clk_unregister().

Fix this by nulling priv->clk after the first clk_unregister(). Also
refine the pointer checking in ca8210_unregister_ext_clock().

Fixes: ded845a781a5 ("ieee802154: Add CA8210 IEEE 802.15.4 device driver")
Signed-off-by: Dinghao Liu <dinghao.liu@zju.edu.cn>
---
 drivers/net/ieee802154/ca8210.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Comments

Miquel Raynal Sept. 25, 2023, 8:29 a.m. UTC | #1
Hi Dinghao,

dinghao.liu@zju.edu.cn wrote on Mon, 25 Sep 2023 15:24:22 +0800:

> If of_clk_add_provider() fails in ca8210_register_ext_clock(),
> it calls clk_unregister() to release priv->clk and returns an
> error. However, the caller ca8210_probe() then calls ca8210_remove(),
> where priv->clk is freed again in ca8210_unregister_ext_clock(). In
> this case, a use-after-free may happen in the second time we call
> clk_unregister().
> 
> Fix this by nulling priv->clk after the first clk_unregister(). Also
> refine the pointer checking in ca8210_unregister_ext_clock().
> 
> Fixes: ded845a781a5 ("ieee802154: Add CA8210 IEEE 802.15.4 device driver")
> Signed-off-by: Dinghao Liu <dinghao.liu@zju.edu.cn>
> ---
>  drivers/net/ieee802154/ca8210.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/net/ieee802154/ca8210.c b/drivers/net/ieee802154/ca8210.c
> index aebb19f1b3a4..1d545879c000 100644
> --- a/drivers/net/ieee802154/ca8210.c
> +++ b/drivers/net/ieee802154/ca8210.c
> @@ -2760,6 +2760,7 @@ static int ca8210_register_ext_clock(struct spi_device *spi)
>  	ret = of_clk_add_provider(np, of_clk_src_simple_get, priv->clk);
>  	if (ret) {
>  		clk_unregister(priv->clk);
> +		priv->clk = NULL;

This function is a bit convoluted. You could just return the result of
of_clk_add_provider() (keep the printk's if you want, they don't seem
very useful) and let ca8210_unregister_ext_clock() do the cleanup.

>  		dev_crit(
>  			&spi->dev,
>  			"Failed to register external clock as clock provider\n"
> @@ -2780,7 +2781,7 @@ static void ca8210_unregister_ext_clock(struct spi_device *spi)
>  {
>  	struct ca8210_priv *priv = spi_get_drvdata(spi);
>  
> -	if (!priv->clk)
> +	if (IS_ERR_OR_NULL(priv->clk))

Does not look useful as you are enforcing priv->clock to be valid or
NULL, it cannot be an error code.

Thanks,
Miquèl
Dinghao Liu Sept. 25, 2023, 8:48 a.m. UTC | #2
Hi Miquèl,

> > index aebb19f1b3a4..1d545879c000 100644
> > --- a/drivers/net/ieee802154/ca8210.c
> > +++ b/drivers/net/ieee802154/ca8210.c
> > @@ -2760,6 +2760,7 @@ static int ca8210_register_ext_clock(struct spi_device *spi)
> >  	ret = of_clk_add_provider(np, of_clk_src_simple_get, priv->clk);
> >  	if (ret) {
> >  		clk_unregister(priv->clk);
> > +		priv->clk = NULL;
> 
> This function is a bit convoluted. You could just return the result of
> of_clk_add_provider() (keep the printk's if you want, they don't seem
> very useful) and let ca8210_unregister_ext_clock() do the cleanup.

Thanks for your advice! I will resend a new patch as suggested.

> 
> >  		dev_crit(
> >  			&spi->dev,
> >  			"Failed to register external clock as clock provider\n"
> > @@ -2780,7 +2781,7 @@ static void ca8210_unregister_ext_clock(struct spi_device *spi)
> >  {
> >  	struct ca8210_priv *priv = spi_get_drvdata(spi);
> >  
> > -	if (!priv->clk)
> > +	if (IS_ERR_OR_NULL(priv->clk))
> 
> Does not look useful as you are enforcing priv->clock to be valid or
> NULL, it cannot be an error code.

I find that ca8210_register_ext_clock() uses IS_ERR to check priv->clk
after calling clk_register_fixed_rate(). So I think priv->clk could be
a non-null pointer even on failure. And a null pointer check may miss
this case in ca8210_unregister_ext_clock(). 

Regards,
Dinghao
diff mbox series

Patch

diff --git a/drivers/net/ieee802154/ca8210.c b/drivers/net/ieee802154/ca8210.c
index aebb19f1b3a4..1d545879c000 100644
--- a/drivers/net/ieee802154/ca8210.c
+++ b/drivers/net/ieee802154/ca8210.c
@@ -2760,6 +2760,7 @@  static int ca8210_register_ext_clock(struct spi_device *spi)
 	ret = of_clk_add_provider(np, of_clk_src_simple_get, priv->clk);
 	if (ret) {
 		clk_unregister(priv->clk);
+		priv->clk = NULL;
 		dev_crit(
 			&spi->dev,
 			"Failed to register external clock as clock provider\n"
@@ -2780,7 +2781,7 @@  static void ca8210_unregister_ext_clock(struct spi_device *spi)
 {
 	struct ca8210_priv *priv = spi_get_drvdata(spi);
 
-	if (!priv->clk)
+	if (IS_ERR_OR_NULL(priv->clk))
 		return
 
 	of_clk_del_provider(spi->dev.of_node);