My problem is that i can't reset the DSP after i download new code via JTAG port. Essentially I need to turn the power off and on in order for DSP to start executing the code. My reset signal stays low for about 200ms and i think that should be more than enough to reset DSP. Should i do something else after the code is loaded to DSP, is there any portion of memory that needs to be flushed before it starts executing? I would appreciate any information regarding this. |
|
reset problem with 56807
Started by ●November 22, 2002
Reply by ●November 22, 20022002-11-22
Well, this might seem a little obvious but are you clicking the run
button? We have seen some strange instances where the DSP hangs after
downloading to flash and the only way to run the code is by a hardware reset or
cycling the power but once in a while we get reports that sometimes a push
button reset does nothing to kick start the processor. We have not found a
solution to this problem other than re-flashing the program, but I have noticed
that logical errors in the code seem to amplify this problem. A not so
trust worthy indication that you are headed for trouble is if your program
is not jumping into main() after downloading. Sometimes the compiler just
blows up and we end up having to stop the down loader in the middle of the
process, reset the board or kill the power and flash again. Luckily this
doesn't happen very often to warrant serious attention but the
aforementioned fix seems to resolve the problem. Lastly, you might
want to check that your power source is as clean as
possible.
Sorry this is not much help but I thought I would share
this info just so you don't start ripping your hair out and go on a hunger
strike.
P.S. Its not your fault, your not alone.
-miramont
|
Reply by ●November 25, 20022002-11-25
I think your problem is that JTAG/OnCE is still in debug mode after the code is downloaded into flash. You must reset DSP and its on-chip JTAG/OnCE at same time. The solution can be found in Motorola FAQ database. The following Q&A was copied from Motorola web site. I strongly recommend everyone who uses Motorola products to use this database. http://e-www.motorola.com/cgi-bin/faq.cgi Question: The 56F80x/56F82x DSP on my production board sometimes fails to boot properly on power-up when I'm not debugging. What could be wrong? Answer: The problem stems from the TSRT\ pin of the JTAG port. (On power-on the JTAG port comes up in an unknown and unpredictable state.) When you are debugging TRST\ is controlled by Codewarrior. When not debugging it should be tied low. One way to ground it is via a zero ohm resistor. (Just remove the resistor for debug.) Or to save cost just ground TRST\ pin through a trace. (Then just cut the trace for debugging, otherwise leave it alone.) On Motorola EVMs just jumper pins 2 and 14 of the JTAG connector Charlie W --- ljube20022000 <> wrote: > My problem is that i can't reset the DSP after i > download new code > via JTAG port. Essentially I need to turn the power > off and on in > order for DSP to start executing the code. My reset > signal stays low > for about 200ms and i think that should be more than > enough to reset > DSP. Should i do something else after the code is > loaded to DSP, is > there any portion of memory that needs to be flushed > before it starts > executing? I would appreciate any information > regarding this. > __________________________________________________ |
|
Reply by ●November 25, 20022002-11-25
Thanks, that was a problem. TRST\ signal stayed high after programming my board causing not complete reset of the DSP. --- In motoroladsp@y..., Charlie W <charliewtx@y...> wrote: > I think your problem is that JTAG/OnCE is still in > debug mode after the code is downloaded into flash. > You must reset DSP and its on-chip JTAG/OnCE at same > time. The solution can be found in Motorola FAQ > database. The following Q&A was copied from Motorola > web site. I strongly recommend everyone who uses > Motorola products to use this database. > > http://e-www.motorola.com/cgi-bin/faq.cgi > > Question: > The 56F80x/56F82x DSP on my production board sometimes > fails to boot properly on power-up when I'm not > debugging. What could be wrong? > Answer: > The problem stems from the TSRT\ pin of the JTAG port. > (On power-on the JTAG port comes up in an unknown and > unpredictable state.) When you are debugging TRST\ is > controlled by Codewarrior. When not debugging it > should be tied low. One way to ground it is via a zero > ohm resistor. (Just remove the resistor for debug.) Or > to save cost just ground TRST\ pin through a trace. > (Then just cut the trace for debugging, otherwise > leave it alone.) On Motorola EVMs just jumper pins 2 > and 14 of the JTAG connector > > Charlie W > > --- ljube20022000 <ljube172@h...> wrote: > > My problem is that i can't reset the DSP after i > > download new code > > via JTAG port. Essentially I need to turn the power > > off and on in > > order for DSP to start executing the code. My reset > > signal stays low > > for about 200ms and i think that should be more than > > enough to reset > > DSP. Should i do something else after the code is > > loaded to DSP, is > > there any portion of memory that needs to be flushed > > before it starts > > executing? I would appreciate any information > > regarding this. > > > > > > __________________________________________________ > |