[PATCH] finish iso support

castet.matthieu at free.fr castet.matthieu at free.fr
Thu Jun 30 08:01:43 EDT 2005

Hi Duncan
On Wed, Jun 29, 2005 at 10:06:10PM +0200, Duncan Sands wrote:
> Hi Matthieu, first of all I should say that I'm not against having an
> explicit "iso" flag, in fact I think it's likely better than doing some
> kind of autodetection based on the endpoint.  I just wanted to hear your
> reasons.
Could you explain a bit more how you would like to do the detection ?
You propose to scan all the device endpoint on all interfaces, and see if
there is one matching direction and number ?

Wouldn't be a bit cleanner to had a flag to the endpoint number to know
if it is bulk or iso ?
iso = number | PIPE_ISOCHRONOUS < 30
bulk = number | PIPE_BULK < 30

> In any case, I think it would be convenient if minidrivers could modify
> their struct usbatm_driver in the bind() call.
I think they could do it as only driver_name and bind are used before
bind call. But in case of multiple devices I don't know who should
duplicate the struct : the driver as it know it will fill it with
specific data or usbatm.



More information about the Usbatm mailing list