usb: ehci: do not invalidate a NULL buffer
authorDirk Behme <dirk.behme@gmail.com>
Fri, 17 Nov 2017 14:28:36 +0000 (15:28 +0100)
committerMarek Vasut <marek.vasut+renesas@gmail.com>
Sun, 26 Nov 2017 01:22:36 +0000 (02:22 +0100)
Its a valid use case to call ehci_submit_async() with a NULL buffer
with length 0. E.g. from usb_set_configuration().

As invalidate_dcache_range() isn't able to judge if the address
NULL is valid or not (depending on the SoC hardware configuration it
might be valid) do the check in ehci_submit_async() as here we know
that we don't have to invalidate such a buffer.

Signed-off-by: Dirk Behme <dirk.behme@de.bosch.com>
drivers/usb/host/ehci-hcd.c

index 80cc87c..2582bf3 100644 (file)
@@ -592,8 +592,9 @@ ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
         * dangerous operation, it's responsibility of the calling
         * code to make sure enough space is reserved.
         */
-       invalidate_dcache_range((unsigned long)buffer,
-               ALIGN((unsigned long)buffer + length, ARCH_DMA_MINALIGN));
+       if (buffer != NULL && length > 0)
+               invalidate_dcache_range((unsigned long)buffer,
+                       ALIGN((unsigned long)buffer + length, ARCH_DMA_MINALIGN));
 
        /* Check that the TD processing happened */
        if (QT_TOKEN_GET_STATUS(token) & QT_TOKEN_STATUS_ACTIVE)