On Tue, Feb 08, 2005 at 11:48:11AM -0800, David Brownell wrote:
> On Tuesday 08 February 2005 11:05 am, Andi Kleen wrote:
> > My plan was to just drive a normal host2host cable through 
> > debug port. It'll need some hacks, but should not be too bad.
> 
> I'd have expected some of them to _be_ "debug devices" in fact,
> but I'm not sure how common that is.  My extremely limited sample
> shows no success there ... they didn't have debug descriptors.

Just debug port should work even without debug descriptors,
shouldn't it?  (with some manual scanning of addresses)

> > On Tue, Feb 08, 2005 at 10:11:25AM -0800, David Brownell wrote:
> > > > It's annoying 
> > > > to have to fight against such dumb code in the kernel then.
> > > 
> > > Do you actually have a debug port driver for Linux?  And some
> > > hardware that it can talk to?
> > 
> > Not yet. But I have(had) plans to write one. Unfortunately 
> > it dropped back a bit on the todo list, so not too much has happened
> > yet:/
> 
> I have the untested guts of one ... attached, purely FYI.  Adds
> a couple KBytes of object code, and doesn't care if IRQs or DMA
> are working (obviously).

Thanks.

> 
> 
> > > I started looking at that issue a while back:
> > > 
> > >   - What model to use for it?  OK, "console" for output is
> > >     at least a (unidirectional) start; how about remote GDB?
> > 
> > Not sure I understand the q. (model?) My plan was to start with output 
> > only, and then later implement receiving too.
> 
> The question is how to _use_ the port.  What data to write to
> it (console data?  binary event records?), or read from it.
> It's a bidirectional channel, so more featureful models are
> possible, like GDB stubs for debug-as-target.

Just ordinary ASCII console data. Other protocols like kgdb can 
run over that fine. In fact the kgdb remote protocol is ASCII
based.

> No; but that gets back to the model question.  My assumption
> was that something simple should work:  if the port is in use
> by the debug channel when EHCI starts, then ignore it from
> then on ... or at least, until the debug port driver drops it.

Yes, obviously when a debug device is plugged in the port
shouldn't be used by the normal drivers (ignoring hubs here,
I think not supporting them is not a issue for debugging) 

-Andi


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to