The chips that are suppose to use en29f002a already use 29f040b. Ok, I've added the feature bit field and FEATURE_REGISTERMAP. I've restored all the old write code and modified as little as I can, we should work on the design some more before anyone changes the write functions.
Well, it seems I missed this in the last review. We should always read the first part of the ID at 0x00/0x01 and I thought did_addr/mid_addr specified the followup locations used for largeid1/largeid2.
One chip expect to read the id using 0x00 and 0x02 on the first run... probe_m29f400bt => probe_jedec_aaa_did_02
I have to admit I didn't check the conversion of the chip drivers. They look sane on a quick glance, but I'll have to verify calling conventions (int start vs. chipaddr dst) closely before giving a final Ack.
To be honest, I will kill all progress printing first thing after this is committed. Better no progress printing than the current slow (and sometimes buggy) stuff.
Regards, Carl-Daniel
Sound great! Thanks for the review, help, and everything. Thanks to Michael Karcher for the review and the double check on Don't Cares in the datasheets.
Also added my Copyrights lines to jedec.c and flashchips.c.