This patch adds a comment to function hdm_configure_channel() to clarify
its execution paths.
Signed-off-by: Christian Gromm <christian.gr...@microchip.com>
Reported-by: Dan Carpenter <dan.carpen...@oracle.com>
---
v2: added Reported-by tag
drivers/staging/most/hdm-usb/hdm_usb.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/drivers/staging/most/hdm-usb/hdm_usb.c
b/drivers/staging/most/hdm-usb/hdm_usb.c
index 3433646..d6db0bd 100644
--- a/drivers/staging/most/hdm-usb/hdm_usb.c
+++ b/drivers/staging/most/hdm-usb/hdm_usb.c
@@ -634,6 +634,15 @@ static int hdm_enqueue(struct most_interface *iface, int
channel,
* @iface: interface
* @channel: channel ID
* @conf: structure that holds the configuration information
+ *
+ * The attached network interface controller (NIC) supports a padding mode
+ * to avoid short packets on USB, hence increasing the performance due to a
+ * lower interrupt load. This mode is default for synchronous data and can
+ * be switched on for isochronous data. In case padding is active the
+ * driver needs to know the frame size of the payload in order to calculate
+ * the number of bytes it needs to pad when transmitting or to cut off when
+ * receiving data.
+ *
*/
static int hdm_configure_channel(struct most_interface *iface, int channel,
struct most_channel_config *conf)
@@ -667,6 +676,11 @@ static int hdm_configure_channel(struct most_interface
*iface, int channel,
!(conf->data_type == MOST_CH_ISOC &&
conf->packets_per_xact != 0xFF)) {
mdev->padding_active[channel] = false;
+ /*
+ * Since the NIC's padding mode is not going to be
+ * used, we can skip the frame size calculations and
+ * move directly on to exit.
+ */
goto exit;
}
--
1.9.1
_______________________________________________
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
