[PATCH] USB: dbgp gadget: fix use after free in dbgp_unbind()

From: Alexey Khoroshilov
Date: Sun Aug 10 2014 - 12:35:46 EST


After dbgp_bind()-dbgp_unbind() cycle happens, static variable dbgp
contains pointers to already deallocated memory (dbgp.serial and dbgp.req).
If the next dbgp_bind() fails, for example in usb_ep_alloc_request(),
dbgp_bind() calls dbgp_unbind() on failure path,
and dbgp_unbind() frees dbgp.serial that still stores a pointer
to already deallocated memory.

The patch sets pointers to NULL in dbgp_unbind().

Found by Linux Driver Verification project (linuxtesting.org).

Signed-off-by: Alexey Khoroshilov <khoroshilov@xxxxxxxxx>
---
drivers/usb/gadget/legacy/dbgp.c | 2 ++
1 file changed, 2 insertions(+)

diff --git a/drivers/usb/gadget/legacy/dbgp.c b/drivers/usb/gadget/legacy/dbgp.c
index 986fc511a2ed..225e385a6160 100644
--- a/drivers/usb/gadget/legacy/dbgp.c
+++ b/drivers/usb/gadget/legacy/dbgp.c
@@ -222,10 +222,12 @@ static void dbgp_unbind(struct usb_gadget *gadget)
{
#ifdef CONFIG_USB_G_DBGP_SERIAL
kfree(dbgp.serial);
+ dbgp.serial = NULL;
#endif
if (dbgp.req) {
kfree(dbgp.req->buf);
usb_ep_free_request(gadget->ep0, dbgp.req);
+ dbgp.req = NULL;
}

gadget->ep0->driver_data = NULL;
--
1.9.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/