What events cause the GPMC CYCLETIME counter to reset and start counting? I am seeing an occasional corruption of GPMC_AD6 on a burst read. The corruption tends to occur when the bit was set on the previous cycle. Mostly I have seen this on the third burst. By my calculations I should have adequate setup margin, so the only thing I can think of is the SoC has a different view of the CYCLETIME counter than I am expecting.
Thanks,
John