R.Doss wrote:
>
>>> now on The 
>>>     
>>>>> List with numerous bullets beneath it.
>>>>>   
>>>>>         
>>>> Yes, this is an FPGA with a serial port on it?    The is 
>>>>       
>>> your universal 
>>>     
>>>> JTAG TAP emulator, which gives feedback about the path than 
>>>>       
>>> an actual 
>>>     
>>>> TAP is traversing.
>>>>
>>>> Without a device to connect to, one that gives responses 
>>>>       
>>> back, how do 
>>>     
>>>> you verify success?
>>>>       
> A JTAG Tap is interesting. It can be put in chain and collect datas 
> inside.
>
> The JTAG emulator have to commnunicate over the JTAG byself. No serial 
> port. A RAM behind collects and readout over the JTAG. The goal of 
> JTAG is to communicate to all IC also to the emulator. Before starting 
> such emulator, the SVF or XSVF insid openocd have to be ready. The 
> emulator can reprogrammed for new TAP interf


I would not rule out the serial port which can be used to "control and 
monitor the *emulation*".  I think reprogramming the FPGA is one path, 
but another perhaps more convenient path would be to have a universal 
FPGA program inside the emulator, and then the test analysis side of the 
OpenOCD would look at the serial port and read and write to it to 
control they way the emulation works.   In my idea the FGPA or CPLD 
simply implements the tap state management, and passes through 
everything else to the serial port.  The device is really an 
easedropping tool but the brains behind it would be back on the PC, 
perhaps a separate process which only looks at the serial port, and is 
probably written in Python if I were to do it.

This keeps the cost out of the device, and puts the emulation software 
on a platform where there is abundant environmental support.

Having said all this, I wonder if the MPSSE protocol would let somebody 
simply use another standard cable to be the emulator?  Then we just tie 
them together and pass the crap back up to a python program. 

My motto on open source projects is this:  "he who does the work, gets 
to decide."

So I could not be persuaded to do this in C, but then that is moot, 
because I could probably not be persuaded to do it at all.

But maybe some of these ideas will stimulate somebody else into action 
or onto refined ideas,

Dick


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

Reply via email to