usb: f_fs: fix ffs_epfile_io returning success on req alloc failure
authorMichal Nazarewicz <mina86@mina86.com>
Mon, 4 Jan 2016 20:28:34 +0000 (21:28 +0100)
committerFelipe Balbi <balbi@kernel.org>
Fri, 4 Mar 2016 13:14:31 +0000 (15:14 +0200)
In the AIO path, if allocating of a request failse, the function simply
goes to the error_lock path whose end result is returning value of ret.
However, at this point ret’s value is zero (assigned as return value from
ffs_mutex_lock).

Fix by adding ‘ret = -ENOMEM’ statement.

Signed-off-by: Michal Nazarewicz <mina86@mina86.com>
Signed-off-by: Felipe Balbi <balbi@kernel.org>
drivers/usb/gadget/function/f_fs.c

index d1a4a86456201a4c38353e393471cc0b9ce0a9b6..13842205b84973482373be79ae4fdc9fc1e7b26d 100644 (file)
@@ -793,8 +793,10 @@ static ssize_t ffs_epfile_io(struct file *file, struct ffs_io_data *io_data)
 
                if (io_data->aio) {
                        req = usb_ep_alloc_request(ep->ep, GFP_KERNEL);
-                       if (unlikely(!req))
+                       if (unlikely(!req)) {
+                               ret = -ENOMEM;
                                goto error_lock;
+                       }
 
                        req->buf      = data;
                        req->length   = data_len;