GTA04 ... JTAG

Werner Almesberger werner at
Wed Apr 30 12:22:28 CEST 2008

Joerg Reisenweber wrote:
> In fact both are programmable to support a variety of different modes, IIRC.

Yup. What I meant is the configuration we're using. We could of course
change modes on the fly, but that may get a wee bit confusing ;-)

> Seen this, a plain bug, nothing else (or maybe the selfdestruct feature ;-)

Good :)

> Both ports are identical. Only in UART mode that is the usual mode for
> channel B on debugboard_v3. No?

They're subtly different. A quick look reveals the following differences:

- MPSSE is only available on channel A
- they have different roles in MCU Host Bus Emulation Mode (not that
  we'd care)
- if both are in Fast Opto-Isolated Serial Mode, only channel B does
  I/O (again, not relevant for us)

> Hey what's wrong with 'muxing' JTAG?

Design risk. More moving parts -> higher risk.

> I suggested a jumper block to have 6400, or 43, or 430 + 6400, depending
> on jumper setting. You could also use it to directly connect to any one
> of the 2 chains.

Hmm, for GTA04v0 that may actually be a nice choice, given that we don't
care about space anyway. Jumpers are a sufficiently basic technology that
I feel I can trust it ;-) Much safer than a JTAG mux.

> Hmm, I don't see this: Use JTAG to disable, to prepare to use JTAG?

The idea would be to catch the CPU/MPU with JTAG before it gets a
chance to do bad things to I2C.

- Werner

More information about the openmoko-kernel mailing list