deleting urb->status

Hi all,

The .status field of the struct urb is routinely used improperly by
drivers to try to see if the urb is still in flight or not (and yes, I
know that I wrote some of those drivers, that's why I'm proposing
this...)  I'd like to just drop that field completely, and replace it by
changing the urb callback function to contain a status variable instead.

As an example, the following fragment of a patch to the pl2303 driver
would be what is needed to change for most USB drivers:


--- 1.111/drivers/usb/serial/pl2303.c	2004-11-22 10:42:02 -08:00
+++ edited/pl2303.c	2005-03-24 00:15:24 -08:00
@@ -942,7 +942,7 @@ static void pl2303_read_bulk_callback (s
 
 
 
-static void pl2303_write_bulk_callback (struct urb *urb, struct pt_regs
*regs)
+static void pl2303_write_bulk_callback (struct urb *urb, int status, struct pt_regs
*regs)
 {
 	struct usb_serial_port *port = (struct usb_serial_port *) urb->context;
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
@@ -950,7 +950,7 @@ static void pl2303_write_bulk_callback (
 
 	dbg("%s - port %d", __FUNCTION__, port->number);
 
-	switch (urb->status) {
+	switch (status) {
 	case 0:
 		/* success */
 		break;
@@ -958,13 +958,13 @@ static void pl2303_write_bulk_callback (
 	case -ENOENT:
 	case -ESHUTDOWN:
 		/* this urb is terminated, clean up */
-		dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
+		dbg("%s - urb shutting down with status: %d", __FUNCTION__, status);
 		priv->write_urb_in_use = 0;
 		return;
 	default:
 		/* error in the urb, so we have to resubmit it */
 		dbg("%s - Overflow in write", __FUNCTION__);
-		dbg("%s - nonzero write bulk status received: %d", __FUNCTION__,
urb->status);
+		dbg("%s - nonzero write bulk status received: %d", __FUNCTION__, status);
 		port->write_urb->transfer_buffer_length = 1;
 		port->write_urb->dev = port->serial->dev;
 		result = usb_submit_urb (port->write_urb, GFP_ATOMIC);



Now yes, I do realise this is going to require changing every urb
callback function, but hey, I'm the one volunteering to do this, not you
:)

Anyway, any big objections to this?  As I change everything, I'll also
fixup the drivers that improperly use this field.  

The HCD drivers will need to bigest changes, as they will have to put
the status value in a hcd specific location, but that's not a big deal,
right?

thanks,

greg k-h


-------------------------------------------------------
This SF.net email is sponsored by Microsoft Mobile & Embedded DevCon 2005
Attend MEDC 2005 May 9-12 in Vegas. Learn more about the latest Windows
Embedded(r) & Windows Mobile(tm) platforms, applications & content.  Register
by 3/29 & save $300 http://ads.osdn.com/?ad_id=6883&alloc_id=15149&op...
_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel