[coreboot] USB Debug support

[Gregg C Levine]

Hello! (Duplicated message.)
Okay. Thank you folks for breaking down the complexity of this issue. I
didn't want to go into the complexity of that logic area that the parts are
wearing as I've been given to understand from reading the message traffic on
the Linux-USB regarding gadgets that this is indeed a moving target. 

Now as to obtaining that Net20DC gizmo, well I'm not that closer to doing
that. I've not chosen my target, nor decided how far I want to dig into
things. Ideally there's a list of things I want to work on, or will be RSN,
and naturally one of them does include your notes regarding the
configuration of the bridges. As for the reading (and understanding) of the
documentation from the site, and the USB group, that's first on the list.

I am resending this just to the list because it still hasn't arrived that
way. Peter if you've got a complaint, or even a comment please do so in the
usual fashion.
--
Gregg C Levine hansolofalcon at worldnet.att.net
"The Force will be with you always." Obi-Wan Kenobi
  


> -----Original Message-----
> From: coreboot-bounces at coreboot.org [mailto:coreboot-bounces at coreboot.org]
On
> Behalf Of Peter Stuge
> Sent: Wednesday, July 02, 2008 8:28 AM
> To: Coreboot
> Subject: Re: [coreboot] USB Debug support
> 
> Hi,
> 
> On Wed, Jul 02, 2008 at 07:45:25AM -0400, Gregg C Levine wrote:
> > A USB cable configured to connect two systems together, such as a
> > laptop to receive files from a desktop is simply two USB-A cables
> > wired to an appropriate logic array. This would (Or "should be".
> > Either way) probably be the latest generation in USB Gadget
> > hardware.
> 
> Yes. But the logic "array" is quite complex, and implements several
> layers of protocol. Please look at chapters 5, 8 and 9 in the USB
> specification for a description of the protocol.
> 
> 
> > Ideally a Linux system can use this to talk to an appropriately
> > configured system, provided the target was built using the debug
> > features turned on for the USB stack layer.
> 
> Not enough. These are the hardware requirements:
> 
> * The target USB controller must be EHCI
> * The target USB controller must implement the optional Debug Port
> * The target board must have a USB connector for the physical port
>   that supports Debug Port. It is always only one particular port.
> * The "cable" must implement the Debug Device functional
>   specification. See links on http://coreboot.org/EHCI_Debug_Port
> 
> 
> > While I have here hardware for trying out this idea, namely my
> > Linux development station, I have not as yet chosen a target.
> 
> You will also need the debug device. I suggest ordering the NET20DC
> product from semiconductorstore.com.
> 
> 
> //Peter
> 
> --
> coreboot mailing list
> coreboot at coreboot.org
> http://www.coreboot.org/mailman/listinfo/coreboot