Stefan Roese wrote:
I think I can run tests on a small set (~5-10 different AMD-commandset and 2 intel-commandset, all 16 bit) flashs, but that still leaves the (small) possibility that there are flash roms that behave different.
Right. But it will be an improvement to the current implementation, where a random command is written as RESET command. And also an improvement to the fixed AMD/Intel RESET command. So I vote for trying this solution. I'll test on a few of mine platforms too.
Michael, could please you provide a patch?
Yes, but it may take some days - I probably need to run the tests on a board that has socketed flash, using a special test case that really puts the flash in the middle of a command cycle and tests reset behaviour.
cu Michael