Skip to contentSkip to conversation

usb: renesas_usbhs: gadget: fix NULL pointer dereference in ep_disable()

PATCH Superseded Yoshihiro Shimoda 3 participants 5 messages Sep 10, 2014 – Sep 11, 2014

Subsystems (2)
USB SuperH

v1 of 6 Superseded Latest is v6 →

lore
Share & tools
lore

Yoshihiro Shimoda

author
#0 lore raw

PATCH 1/4 usb: renesas_usbhs: gadget: fix NULL pointer dereference in ep_disable()

This patch fixes an issue that the NULL pointer dereference happens when we use g_audio driver. Since the g_audio driver will call usb_ep_disable() in afunc_set_alt() before it calls usb_ep_enable(), the uep->pipe of renesas usbhs driver will be NULL. So, this patch adds a condition to avoid the oops.

Signed-off-by Kazuya Mizuguchi <[email protected]>
Signed-off-by Takeshi Kihara <[email protected]>
Signed-off-by Yoshihiro Shimoda <[email protected]>
---
 drivers/usb/renesas_usbhs/mod_gadget.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/usb/renesas_usbhs/mod_gadget.c +3 -0
index 04e6505..a582222 100644
@@ -602,6 +602,9 @@ static int usbhsg_ep_disable(struct usb_ep *ep)
 	struct usbhsg_uep *uep = usbhsg_ep_to_uep(ep);
 	struct usbhs_pipe *pipe = usbhsg_uep_to_pipe(uep);
 
+	if (!uep || !uep->pipe)
+		return -EINVAL;
+
 	usbhsg_pipe_disable(uep);
 	usbhs_pipe_free(pipe);
Signature · 1 line
1.7.9.5
#1 lore raw
Quoted 9 lines · Yoshihiro Shimoda

This patch fixes an issue that the NULL pointer dereference happens when we use g_audio driver. Since the g_audio driver will call usb_ep_disable() in afunc_set_alt() before it calls usb_ep_enable(), the uep->pipe of renesas usbhs driver will be NULL. So, this patch adds a condition to avoid the oops.

Signed-off-by Kazuya Mizuguchi <[email protected]>
Signed-off-by Takeshi Kihara <[email protected]>
Signed-off-by Yoshihiro Shimoda <[email protected]>

Which commit does this fix ? Do we need a Cc: stable ?

Signature · 1 line
balbi
#2 lore raw
In reply to Yoshihiro Shimoda

Hi Shimoda-san

drivers/usb/renesas_usbhs/mod_gadget.c static int usbhsg_ep_disable(struct usb_ep *ep) · line 609

Quoted 11 lines
drivers/usb/renesas_usbhs/mod_gadget.c +3 -0
@@ -602,6 +602,9 @@ static int usbhsg_ep_disable(struct usb_ep *ep)
 	struct usbhsg_uep *uep = usbhsg_ep_to_uep(ep);
 	struct usbhs_pipe *pipe = usbhsg_uep_to_pipe(uep);
 
+	if (!uep || !uep->pipe)
+		return -EINVAL;
+
 	usbhsg_pipe_disable(uep);
 	usbhs_pipe_free(pipe);

If uep can be NULL,
we need care about usbhsg_uep_to_pipe(uep) too.

and, "uep->pipe" is same as "pipe" ?

Best regards
---
Kuninori Morimoto

author
#3 lore raw
In reply to Yoshihiro Shimoda

Hi,

Quoted 7 lines · Felipe Balbi
Quoted 6 lines · Yoshihiro Shimoda

This patch fixes an issue that the NULL pointer dereference happens when we use g_audio driver. Since the g_audio driver will call usb_ep_disable() in afunc_set_alt() before it calls usb_ep_enable(), the uep->pipe of renesas usbhs driver will be NULL. So, this patch adds a condition to avoid the oops.

Quoted 6 lines
Quoted 3 lines
Signed-off-by Kazuya Mizuguchi <[email protected]>
Signed-off-by Takeshi Kihara <[email protected]>
Signed-off-by Yoshihiro Shimoda <[email protected]>

Which commit does this fix ? Do we need a Cc: stable ?

I think we need a Cc: stable in this patch because the previous code causes the oops.

Best regards,
Yoshihiro Shimoda

author
#4 lore raw
In reply to Yoshihiro Shimoda

Hi Morimoto-san,

Quoted 9 lines · Kuninori Morimoto

Hi Shimoda-san

Quoted 6 lines
drivers/usb/renesas_usbhs/mod_gadget.c +0 -0
@@ -602,6 +602,9 @@ static int usbhsg_ep_disable(struct usb_ep *ep)
 	struct usbhsg_uep *uep = usbhsg_ep_to_uep(ep);
 	struct usbhs_pipe *pipe = usbhsg_uep_to_pipe(uep);
Quoted 8 lines

drivers/usb/renesas_usbhs/mod_gadget.c static int usbhsg_ep_disable(struct usb_ep *ep) · line 609

Quoted 5 lines
drivers/usb/renesas_usbhs/mod_gadget.c +3 -0
+	if (!uep || !uep->pipe)
+		return -EINVAL;
+
 	usbhsg_pipe_disable(uep);
 	usbhs_pipe_free(pipe);

If uep can be NULL,
we need care about usbhsg_uep_to_pipe(uep) too.

Thank you for the point.
I will check the uep can be NULL or not.

Quoted 1 line

and, "uep->pipe" is same as "pipe" ?

Yes. I will use "pipe" instead of "uep->pipe".

Best regards,
Yoshihiro Shimoda

Quoted 4 lines

Best regards
---
Kuninori Morimoto