Re: [PATCH] Fix for USB ugen panics

From: Jay Cornwall (jay_at_evilrealms.net)
Date: 01/09/04

  • Next message: Doug Barton: "Re: devfs support of wildcarding" 
    Date: Fri, 09 Jan 2004 02:17:09 +0000
To: Martin <nakal@web.de>

    Martin wrote:

    > or from a sourceforge-project:
    > http://sourceforge.net/project/showfiles.php?group_id=10775&release_id=43893

    > In section 7.5.3 "Bandwidth Management" it says something about
    > "Set Configuration" request and the zero-setting. Maybe I
    > misunderstood the paragraph.

    "...by providing a zero-bandwidth interface (alternate setting zero) and 16
    alternate settings interfaces wth 8Mbps down to 0.5Mbps bandwidth in
    descending 0.5Mbps steps."

    These can be accessed through USB_SET_ALTINTERFACE with a value from 0 to 16.

    "The default alternate setting zero (with zero bandwidth) selected by a Set
    Configuration request..."

    I think this is the badly phrased sentence in question. What the author is
    trying to say is that alternate interface 0 will be selected by default upon
    USB_SET_CONFIGURATION - which seems a bit daft without explaining which
    configurations are available. Try a USB_SET_CONFIGURATION of 1, and see if the
    16 interfaces are available then.

    > Perhaps you can tell me a second thing. In section 7.5.2 they say
    > there are 0-length packets used for frame-sync. I have never seen
    > a 0-length packet coming from the interface, but some packets are
    > (irregulary) shorter. Is the FreeBSD ugen-driver supporting 0-length
    > packets?

    Now that, I'm not sure of. :)

    But if their device does that, which is USB 1.1 compliant, it's probably in
    the spec and so probably supported by ugen. Probably, there's only one real
    way to find out. ;) 

    -- 
Cheers,
Jay
http://www.evilrealms.net/ - Systems Administrator & Developer
http://www.imperial.ac.uk/ - 3rd year CS student
_______________________________________________
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"
  • Next message: Doug Barton: "Re: devfs support of wildcarding"

    Relevant Pages

    • Re: [PATCH] DRTL kernel 2.6.32-rc3 : SCHED_EDF, DI RT-Mutex, Deadline Based Interrupt Handl
      ... The main difference is the bandwidth reservation thing. ... Moreover, as Peter pointed out, coming out with an interface which is ... also SMP has to be there! ... separate scheduling class would be better suited for this. ...
      (Linux-Kernel)
    • Re: Implementing Multi-Interface USB Device Driver
      ... when you need more bandwidth on the isoch endpoints. ... that correspond to that changing interface alternate setting. ... >>> UCHAR ucInterfaceNumber, ...
      (microsoft.public.development.device.drivers)
    • aggregate bandwidth of 2 server each with 2 nics
      ... I'm trying to increase the bandwidth between 2 hosts in the same ... Both server are equipped with 2 GbE NICs and located ... of one interface with netio, netpipe or any other network benchmark. ...
      (comp.os.linux.networking)
    • Re: QoS and available Bandwidth
      ... doesn't know what bandwidth it has available either. ...  class class-voip ... QoS ONLY works outbound on an interface. ... Traffic shaping cannot be used on a dialer. ...
      (comp.dcom.sys.cisco)
    • Re: ATM QoS on 877
      ... I've setup the ATM0 interface and bridged it to the BVI1 interface. ... Hardware is MPC ATMSAR (with Alcatel ADSL Module) ... Insufficient bandwidth on interface or invalid PCR. ...
      (comp.dcom.sys.cisco)