Re: [PATCH v3] net/9p: Fix buffer overflow in USB transport layer

[Kees Cook]

On June 22, 2025 6:39:56 AM PDT, Dominique Martinet via B4 Relay <devnull+asmadeus.codewreck.org@xxxxxxxxxx> wrote:
> [...]
>Add validation in usb9pfs_rx_complete() to ensure req->actual does not
>exceed the buffer capacity before copying data.
> [...]
>+	if (req_size > p9_rx_req->rc.capacity) {
>+		dev_err(&cdev->gadget->dev,
>+			"%s received data size %u exceeds buffer capacity %zu\n",
>+			ep->name, req_size, p9_rx_req->rc.capacity);
>+		req_size = 0;
>+		status = REQ_STATUS_ERROR;
>+	}
> 
>-	p9_rx_req->rc.size = req->actual;
>+	memcpy(p9_rx_req->rc.sdata, req->buf, req_size);

Is rc.sdata always rc.capacity sized? If so, this world be a good first adopter of the __counted_by annotation for pointer struct members, available in Clang trunk and soon in GCC:
https://gcc.gnu.org/pipermail/gcc-patches/2025-May/683696.html

-Kees

 

-- 
Kees Cook