Re: JTAG Problem

Chris Martin <[email protected]> Fri, 26 Nov 2004 23:01:28 +0100
Newsgroups gmane.comp.handhelds.simpad.linux
Message-ID <[email protected]>
Hi Ralf,

Thanx!! That was it. I had rmmod'ed lp, but not parport_pc and parport. 
After doing that even jtag version 0.5.1 worked. And the verify bug 
mentioned in the HOWTO isn't present using version 0.5.1.

Now I'm back to business ;-)

Chris

Ralf Miunske wrote:

>Hi Chris!
>
>As you can see, your connection works, but not very well. First of all: It would be amazingly euphemistic to call you cable crappy! It's absolutely  minimalistic and doen't hold a single rule for electrical design: One cannot convert voltages by adding some resistors; you can only prevent your hardware from blowing up. This could be caused by other tasks accessing the port without locking it or something similar. First, you should stop all tasks, that do not need to run. I've never looked inside the jtag-code, but this way you can make sure, that nothing uses your port and no cpu time will be wasted, so that even time critical apps can run cleanly. Additionally you could remove all parport-modules from the kernel (lp, parport, parport_pc...) and tell the jtag-utilites to access you port directly without any kernel drivers using the io-bus itself (see "cable" manpage). And still, it might not be working yet: The last reason for your problem could be a burn-up protected parallel port. Most modern printer ports have resistors directly behind the connector to protect every single line from breaking when connected to a different voltage than its very own. Theese about 1k sized resistors add to the jtag-resistors and get highly above the suggested values. The other way around is even worse: one can try to control a 5Vline using 3V, but this'll get very slow, if it even works at all (depending on the hardware inside your pc). You could try to use some old and thus unprotected hardware to flash your device or try to explicitly convert the port voltages by using a MAX212 or something similar.
>(http://www.maxim-ic.com/quick_view2.cfm?qv_pk=1051&ln=)
>Before using this IC you should check the speed of your jtag-iface and compare it to the maximum rating named in the data sheet.
>(http://pdfserv.maxim-ic.com/en/ds/MAX212.pdf)
>I never cared about the speed of jtag-connections, but in case of a too high frequency you'll need to find a different converter or build one yourself using some appropriate (e.g.) transistors and a few additional components.
>
>
>Good luck!
>  Ralf
>PS: Try using a different machine first :-)
>
>
>  
>



-------------------------------------------------------
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://productguide.itmanagersjournal.com/