Using the JTAG analyzer – customer performed writes to some memory addresses and when they read them back the first couple of times, the data was correct. But as they continued to read, the data was coming from the wrong address (I think they described it as the reads were coming back as though they were rotating through consecutive addresses) until after a while, the reads would become stable and they would read the same data each time.
Do you have any ideas what would cause this issue? Or perhaps you know of something we could do with the JTAG interface to narrow down the problem?
Also would you be able to make a recommendation on the largest SLC NAND flash that they can use with the AM35xx given the ECC limitations?