On Wednesday 11 October 2006 07:22, Alan Stern wrote: > On Mon, 9 Oct 2006, Inaky Perez-Gonzalez wrote:
> > > > pseudo allocates buffer 'maxp' bytes in size using usb_maxpacket() > > [?? which gets the max packet size from the interrupt > > endpoint]. That's maybe not as proper. According to the spec > > (see below) the amount of data to read for the hub status is > > (PORTS + 1 + 7) / 8 bytes. > > That is the amount of data the hub is _supposed_ to send. Some hubs have > been known to send more. Using the maxpacket size helps to avoid > EOVERFLOW errors. Huh, so that *is* the reason -- then it'd be a good idea to document it in the code, so the next guy who goes around it doesn't try to do the same thing I did. And on the way there, check that the maxpacketsize is bigger or equal to (PORTS + 1 + 7) / 8 and flag a warning otherwise. > > Haven't been able to track how a fake endpoint descriptor is > > generated for the root hub's interrupt endpoint, but somehow we > > ended up with wrong data in there and thus we always had two > > bytes of buffer space instead of the three that we needed for 16 > > ports (for example) or more... > > It's a static array defined near the beginning of drivers/usb/core/hcd.c. How about I change the maxpacketsize definition to be (USB_MAXCHILDREN + 1 + 7) / 8? That way we won't have to worry about it any more (and I doubt it'll ever go over 255 ... so we don't have to worry about endianess either. > > hub_activate() > > usb_hcd_submit_urb() > > rh_urb_enqueue() > > urb->pipe is neither int nor ctl, so it errors out > > rh_queue_status() > > Returns -EINVAL because the buffer length is smaller > > than the minimum needed to report all the hub port > > bits as in accordance with USB2.0[11.12.3]. There has > > to be trunc((PORTS + 1 + 7) / 8) bytes of space at > > least. > > > > So this patch simply changes that size calculation to be based on the > > number of ports declared by the hub on its hub descriptor, so it > > always floats based on that. > > It would be better to change the maxpacket value returned in the > endpoint descriptor. I am guessing you refer to changing the value in the static array, right? ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
