--- On Thu, 9/30/10, Gene Smith <[email protected]> wrote:

> From: Gene Smith <[email protected]>
> Subject: Re: [Openocd-development] cortex-r4 core
> To: [email protected]
> Date: Thursday, September 30, 2010, 12:42 PM
> Laurent Gauch wrote:
> > 
> > Built-in XDS100v2 !!!!
> > 
> 
> Not sure if this is good "!!!!" or bad  ??

XDS100v2 is an FT2232H (high speed) JTAG adapter,
so in a sense it's good; you won't need to buy
another, and cabling can be simpler.

Do you have complete board schematics including
the XDS100 stuff?

The deal so far has been that OpenOCD doesn't have
an XDS100 (v1 or v2) driver.  There are a few
technical issues there:

- (a) Need to toggle some FT2232 GPIO to latch
some critical state.

Pure XDS100 schematics are circulated with some
legalese saying more or less, as I recall, that
they can't be used to support open source software
(like OpenOCD).  So if your schematics (R4 board)
don't say that, then the toggling can be coded in
a way that OpenOCD can finally ship a basic driver.

NOTE:  you can write an xds100 (v or v2) driver
using TI documentation, just fine.  The issue is
distribution restrictions on such code.  (Much
the same way GPL code has its own different set of
distribution restrictions.

There are also two signals that TI uses, EMU0 and
EMU1 as I recall, which it would be nice to support;
but probably not essential just yet.

One nice thing about a "real" XDS100" is that it
comes with a TI-14 JTAG connector, so it works well
with most TI development bords. (Including Beagle,
the upcoming Panda, and so on.   If the XDS100 is
built into your board, that won't matter; you'll
not need a separate JTAG connector.


_______________________________________________
Openocd-development mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/openocd-development

Reply via email to