[PATCH] usb: fotg210: validate endpoint index in ep0 handlers

Liu Chao posted 1 patch 1 week, 4 days ago
drivers/usb/fotg210/fotg210-udc.c | 17 +++++++++++------
1 file changed, 11 insertions(+), 6 deletions(-)
[PATCH] usb: fotg210: validate endpoint index in ep0 handlers
Posted by Liu Chao 1 week, 4 days ago
fotg210_set_feature(), fotg210_clear_feature() and fotg210_get_status()
use wIndex from the USB setup packet to index the fotg210->ep[] array
without verifying that the endpoint number falls below
FOTG210_MAX_NUM_EP (5).  USB_ENDPOINT_NUMBER_MASK is 0x0f, so a
malicious host can issue a setup packet with wIndex 5..15, resulting in
an out-of-bounds array read.  The resulting wild pointer is then
dereferenced in fotg210_set_epnstall() or fotg210_is_epnstall(), which
compute an MMIO register offset from ep->epnum and perform iowrite32
through it.

fotg210_clear_feature() is especially problematic: the out-of-bounds
access occurs at function entry (source-level) regardless of which
USB_RECIP_* case is taken, because the ep pointer is computed before the
switch statement.

Add upper-bound checks on the endpoint number derived from wIndex in
all three functions.  Invalid endpoint numbers in clear_feature now
trigger fotg210_request_error() (STALL) instead of silently falling
through to fotg210_set_cxdone().  Also add the missing le16_to_cpu()
conversion for ctrl->wIndex in fotg210_get_status() to fix a sparse
endianness warning.

Fixes: b84a8dee23fd ("usb: gadget: add Faraday fotg210_udc driver")
Cc: stable@vger.kernel.org
Signed-off-by: Liu Chao <liuc63@xiaopeng.com>
---
 drivers/usb/fotg210/fotg210-udc.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/fotg210/fotg210-udc.c b/drivers/usb/fotg210/fotg210-udc.c
index d9e024873..127a259c8 100644
--- a/drivers/usb/fotg210/fotg210-udc.c
+++ b/drivers/usb/fotg210/fotg210-udc.c
@@ -661,7 +661,7 @@ static void fotg210_set_feature(struct fotg210_udc *fotg210,
 	case USB_RECIP_ENDPOINT: {
 		u8 epnum;
 		epnum = le16_to_cpu(ctrl->wIndex) & USB_ENDPOINT_NUMBER_MASK;
-		if (epnum)
+		if (epnum && epnum < FOTG210_MAX_NUM_EP)
 			fotg210_set_epnstall(fotg210->ep[epnum]);
 		else
 			fotg210_set_cxstall(fotg210);
@@ -677,8 +677,8 @@ static void fotg210_set_feature(struct fotg210_udc *fotg210,
 static void fotg210_clear_feature(struct fotg210_udc *fotg210,
 				struct usb_ctrlrequest *ctrl)
 {
-	struct fotg210_ep *ep =
-		fotg210->ep[ctrl->wIndex & USB_ENDPOINT_NUMBER_MASK];
+	u8 epnum = le16_to_cpu(ctrl->wIndex) & USB_ENDPOINT_NUMBER_MASK;
+	struct fotg210_ep *ep;
 
 	switch (ctrl->bRequestType & USB_RECIP_MASK) {
 	case USB_RECIP_DEVICE:
@@ -688,7 +688,12 @@ static void fotg210_clear_feature(struct fotg210_udc *fotg210,
 		fotg210_set_cxdone(fotg210);
 		break;
 	case USB_RECIP_ENDPOINT:
-		if (ctrl->wIndex & USB_ENDPOINT_NUMBER_MASK) {
+		if (epnum >= FOTG210_MAX_NUM_EP) {
+			fotg210_request_error(fotg210);
+			break;
+		}
+		if (epnum) {
+			ep = fotg210->ep[epnum];
 			if (ep->wedged) {
 				fotg210_set_cxdone(fotg210);
 				break;
@@ -744,8 +749,8 @@ static void fotg210_get_status(struct fotg210_udc *fotg210,
 		fotg210->ep0_data = cpu_to_le16(0);
 		break;
 	case USB_RECIP_ENDPOINT:
-		epnum = ctrl->wIndex & USB_ENDPOINT_NUMBER_MASK;
-		if (epnum)
+		epnum = le16_to_cpu(ctrl->wIndex) & USB_ENDPOINT_NUMBER_MASK;
+		if (epnum && epnum < FOTG210_MAX_NUM_EP)
 			fotg210->ep0_data =
 				cpu_to_le16(fotg210_is_epnstall(fotg210->ep[epnum])
 					    << USB_ENDPOINT_HALT);

base-commit: 2f0c1cf72f4682178506f513bbf015e591b1aa4a
-- 
2.50.1
Re: [PATCH] usb: fotg210: validate endpoint index in ep0 handlers
Posted by Linus Walleij 1 week, 4 days ago
On Sun, Sep 13, 2026 at 2:05 PM Liu Chao <liuc63@xiaopeng.com> wrote:

> fotg210_set_feature(), fotg210_clear_feature() and fotg210_get_status()
> use wIndex from the USB setup packet to index the fotg210->ep[] array
> without verifying that the endpoint number falls below
> FOTG210_MAX_NUM_EP (5).  USB_ENDPOINT_NUMBER_MASK is 0x0f, so a
> malicious host can issue a setup packet with wIndex 5..15, resulting in
> an out-of-bounds array read.  The resulting wild pointer is then
> dereferenced in fotg210_set_epnstall() or fotg210_is_epnstall(), which
> compute an MMIO register offset from ep->epnum and perform iowrite32
> through it.
>
> fotg210_clear_feature() is especially problematic: the out-of-bounds
> access occurs at function entry (source-level) regardless of which
> USB_RECIP_* case is taken, because the ep pointer is computed before the
> switch statement.
>
> Add upper-bound checks on the endpoint number derived from wIndex in
> all three functions.  Invalid endpoint numbers in clear_feature now
> trigger fotg210_request_error() (STALL) instead of silently falling
> through to fotg210_set_cxdone().  Also add the missing le16_to_cpu()
> conversion for ctrl->wIndex in fotg210_get_status() to fix a sparse
> endianness warning.
>
> Fixes: b84a8dee23fd ("usb: gadget: add Faraday fotg210_udc driver")
> Cc: stable@vger.kernel.org
> Signed-off-by: Liu Chao <liuc63@xiaopeng.com>

That's a good catch.
Reviewed-by: Linus Walleij <linusw@kernel.org>

Yours,
Linus Walleij