Marc Leeman wrote:
I wrote a version of your function some time ago, and included it in the set of changes for the 83xx tree that was made available to Wolfgang a few weeks ago, but he hasn't pulled them yet. You can find those changes here:
Thanks,
I switched to your function (since this is the one that will be merged in u-boot anyway) and it still hangs while programming UPM A (writing) (so exactly the same behaviour as before).
When we switched to UPM B (no other change but CFG_BR1_PRELIM and the upmconfig call), we started seeing the expected signals that we programmed in our RAM for HPI.
Although my function is supposed to handle all UPMs, I've only tested it with UPM A, because that's what my 8349 ITX uses for CF programming. It's odd that UPM A doesn't work for you but UPM B does.
I don't know a whole lot about the UPM, but doesn't the board layout determine which UPM you can use? So maybe, for whatever device you're programming, you have to use UPM B?