Patch "usb: dwc3: gadget: delay unmap of bounced requests" has been added to the 4.4-stable tree

[<gregkh@linuxfoundation.org>]

This is a note to let you know that I've just added the patch titled

    usb: dwc3: gadget: delay unmap of bounced requests

to the 4.4-stable tree which can be found at:
    http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
     usb-dwc3-gadget-delay-unmap-of-bounced-requests.patch
and it can be found in the queue-4.4 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.


>From de288e36fe33f7e06fa272bc8e2f85aa386d99aa Mon Sep 17 00:00:00 2001
From: Janusz Dziedzic <januszx.dziedzic@intel.com>
Date: Mon, 13 Mar 2017 14:11:32 +0200
Subject: usb: dwc3: gadget: delay unmap of bounced requests

From: Janusz Dziedzic <januszx.dziedzic@intel.com>

commit de288e36fe33f7e06fa272bc8e2f85aa386d99aa upstream.

In the case of bounced ep0 requests, we must delay DMA operation until
after ->complete() otherwise we might overwrite contents of req->buf.

This caused problems with RNDIS gadget.

Signed-off-by: Janusz Dziedzic <januszx.dziedzic@intel.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>


---
 drivers/usb/dwc3/gadget.c |   21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

--- a/drivers/usb/dwc3/gadget.c
+++ b/drivers/usb/dwc3/gadget.c
@@ -235,6 +235,7 @@ void dwc3_gadget_giveback(struct dwc3_ep
 		int status)
 {
 	struct dwc3			*dwc = dep->dwc;
+	unsigned int			unmap_after_complete = false;
 	int				i;
 
 	if (req->queued) {
@@ -259,11 +260,19 @@ void dwc3_gadget_giveback(struct dwc3_ep
 	if (req->request.status == -EINPROGRESS)
 		req->request.status = status;
 
-	if (dwc->ep0_bounced && dep->number <= 1)
+	/*
+	 * NOTICE we don't want to unmap before calling ->complete() if we're
+	 * dealing with a bounced ep0 request. If we unmap it here, we would end
+	 * up overwritting the contents of req->buf and this could confuse the
+	 * gadget driver.
+	 */
+	if (dwc->ep0_bounced && dep->number <= 1) {
 		dwc->ep0_bounced = false;
-
-	usb_gadget_unmap_request(&dwc->gadget, &req->request,
-			req->direction);
+		unmap_after_complete = true;
+	} else {
+		usb_gadget_unmap_request(&dwc->gadget,
+				&req->request, req->direction);
+	}
 
 	dev_dbg(dwc->dev, "request %p from %s completed %d/%d ===> %d\n",
 			req, dep->name, req->request.actual,
@@ -273,6 +282,10 @@ void dwc3_gadget_giveback(struct dwc3_ep
 	spin_unlock(&dwc->lock);
 	usb_gadget_giveback_request(&dep->endpoint, &req->request);
 	spin_lock(&dwc->lock);
+
+	if (unmap_after_complete)
+		usb_gadget_unmap_request(&dwc->gadget,
+				&req->request, req->direction);
 }
 
 int dwc3_send_gadget_generic_command(struct dwc3 *dwc, unsigned cmd, u32 param)


Patches currently in stable-queue which might be from januszx.dziedzic@intel.com are

queue-4.4/usb-dwc3-gadget-delay-unmap-of-bounced-requests.patch