Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit c9ffc787 authored by Harish Jenny K N's avatar Harish Jenny K N Committed by Felipe Balbi
Browse files

usb: gadget: NCM: Protect dev->port_usb using dev->lock 

This commit incorporates findings from
https://lkml.org/lkml/2016/4/25/594



The function has been modified to make sure we hold
the dev lock when accessing the net device pointer.

Acked-by: default avatarJim Baxter <jim_baxter@mentor.com>
Signed-off-by: default avatarHarish Jenny K N <harish_kandiga@mentor.com>
Signed-off-by: default avatarFelipe Balbi <felipe.balbi@linux.intel.com>
parent 79775f44

drivers/usb/gadget/function/u_ether.c

+5 −2
Original line number Original line Diff line number Diff line
@@ -553,14 +553,16 @@ static netdev_tx_t eth_start_xmit(struct sk_buff *skb, 
		spin_lock_irqsave(&dev->lock, flags);

		spin_lock_irqsave(&dev->lock, flags);

		if (dev->port_usb)

		if (dev->port_usb)

			skb = dev->wrap(dev->port_usb, skb);

			skb = dev->wrap(dev->port_usb, skb);

		spin_unlock_irqrestore(&dev->lock, flags);

		if (!skb) {

		if (!skb) {

			/* Multi frame CDC protocols may store the frame for

			/* Multi frame CDC protocols may store the frame for

			 * later which is not a dropped frame.

			 * later which is not a dropped frame.

			 */

			 */

			if (dev->port_usb &&

			if (dev->port_usb &&

					dev->port_usb->supports_multi_frame)

					dev->port_usb->supports_multi_frame) {

				spin_unlock_irqrestore(&dev->lock, flags);

				goto multiframe;

				goto multiframe;

			}

			spin_unlock_irqrestore(&dev->lock, flags);

			goto drop;

			goto drop;

		}

		}

	}

	}
@@ -578,6 +580,7 @@ static netdev_tx_t eth_start_xmit(struct sk_buff *skb, 
		req->zero = 0;

		req->zero = 0;

	else

	else

		req->zero = 1;

		req->zero = 1;

	spin_unlock_irqrestore(&dev->lock, flags);





	/* use zlp framing on tx for strict CDC-Ether conformance,

	/* use zlp framing on tx for strict CDC-Ether conformance,

	 * though any robust network rx path ignores extra padding.

	 * though any robust network rx path ignores extra padding.