Hi,
Has anyone tested/used Johan's usb_set_maxpacket() patch?
It looks good to me, but I haven't tested it yet.
Could I get some feedback on it before forwarding it
to Linus?
Thanks,
Randy
___________________________________________________
|Randy Dunlap Intel Corp., DAL Sr. SW Engr.|
|randy.dunlap.at.intel.com 503-696-2055|
|NOTE: Any views presented here are mine alone |
|and may not represent the views of my employer. |
|_________________________________________________|
> -----Original Message-----
> From: Johan Verrept [mailto:[EMAIL PROTECTED]]
> Sent: Tuesday, March 28, 2000 10:51 AM
> To: Georg Acher
> Cc: [EMAIL PROTECTED]
> Subject: Re: [linux-usb] errors... stalled endpoint ?
>
>
> Georg Acher wrote:
> >
> > On Tue, Mar 28, 2000 at 06:24:56PM +0000, Johan Verrept wrote:
> > <...>
> > > But that doesn't matter beacuse i have found out what the
> problem is.
> > > I hooked up a CATC and behold!
> > > The alternate interface set by me, specifies an out
> packet size of 32
> > > bytes, and an in packet size of 64. (the alternative
> interfaces only
> > > differ in the packets sizes...).
> > > The linux machine however is sending larger packets (64 bytes).
> > > So there is a bug there somewhere...
> >
> > Look at the maxsze-variable in uhci_submit_bulk_urb, this
> is the pipesize used
> > for assembling packets. If it is wrong there, it's not HCD's fault.
>
> It wrong there.
> The usb_set_maxpacket() runs down all the alternative settings to
> determine what is the maximum pacekt size. I don't think this is
> correct.
> It should run down all the interfaces and all the endpoints,
> but only on
> the active alternate setting. At least, if you want to use this to
> determine which packet size you are allowed to use in transmitting
> packets!
>
> It's actually a bug in usb.c then :-)
>
> patch attached. ($Id: usb.c,v 1.58 2000/03/13 16:44:45 acher Exp $)
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
