El Mar 13, 2009, a las 19:08 , Duane Ellis escribió:
duane>7) Why the "10 second" - vrs 1 second timeout in arm7_9_common.c?jeff> Because I kept wanting to dump the entire RAM at 1KHz! :-) jeff> Perhaps the right thing to do here is to time out after onejeff> second of nothing received instead of one second for the jeff> entire op.Hmm, sounds like you are asking for too much at a time and need to "chunk the request".But - more importantly - I think it is the 1khz rate - yuck.I see the MC1322x chip has an "RTCK signal" - that means the chip is actually an arm7tdmi-s (S means synth core, not hard core) You should look at the FAQ #1 - in the manual, and enable it if you can. I think the JLINK supports this, and if you do, the thing will run as fast as the chip can handle it.
When I first started, RTCK was completely unworkable. Then just a few days ago (after all of my changes) I tried it and it works great! I'm attaching the config file that I use for this part.
OPTIONS as I see it: (a) Based on the jtag clock rate (freq) adjust the timeout period.(b) chunk the transfer requests into smaller packets when running at a slower speed.Which do you think is easy to implement? You know the issues that led you to this problem, not me.
I think the answer is c) don't ask too much! :-)For now let's set it back to what it was. There really is no reason to download all of RAM. I was just doing it to verify against the results I was getting from the Segger tools.
-Duane.
MC13224.cfg
Description: Binary data
_______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
