On Fri, Jun 15, 2001 at 12:40:36AM +0200, Pavel Machek wrote:
> Hi!
>
> > After some more work on the driver, it's bugfree(tm) now, possibly
> > except for a couple harmless race conditions and some stuff in the
> > multicast filters, which I wasn't able to test much. It does about
> > 600-700 KB/sec on UHCI, haven't measured it on OHCI yet.
> >
> > It should work with the following USB devices:
> >
> > Belkin F5U111
> > CATC Netmate
> > CATC Netmate II
> > smartBridges smartNIC
> >
> > The patch is agains 2.4.5-ac12, but should patch cleanly against almost
> > any other recent kernel.
> >
> > Johannes: Although marked EXPERIMENTAL, because I was the only one to
> > test it, it should be pretty stable. Please send it to Alan/Linus for
> > inclusion in the kernel.
>
> +static int catc_ctrl_msg(struct catc *catc, u8 dir, u8 request, u16
> value, u16 index, void *buf, int len)
> +{
> + int retval = usb_control_msg(catc->usbdev,
> + dir ? usb_rcvctrlpipe(catc->usbdev, 0) :
>usb_sndctrlpipe(catc->usbdev, 0),
> + request, 0x40 | dir, value, index, buf, len, HZ);
> ~~~~
> + return retval < 0 ? retval : 0;
> +}
> +
> +static void catc_ctrl_run(struct catc *catc)
> +{
> + struct ctrl_queue *q = catc->ctrl_queue + catc->ctrl_tail;
> + struct usb_device *usbdev = catc->usbdev;
> + struct urb *urb = &catc->ctrl_urb;
> + devrequest *dr = &catc->ctrl_dr;
> + int status;
> +
> + dr->request = q->request;
> + dr->requesttype = 0x40 | q->dir;
> ~~~~--- should probably be USB_TYPE_VENDOR.
You're right, that way it is cleaner. I'll fix this in the next update.
--
Vojtech Pavlik
SuSE Labs
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
http://lists.sourceforge.net/lists/listinfo/linux-usb-devel
