Hello 计龙 杨,
I confirmed and according to the reference manual, the BCH encode process creates t * 13/14 parity symbols for each data block that are written on the flash device after the corresponding data block. Since t is the number of correctable bits (ecc strength) and you can only select Galois Field of length 13 or 14, this means that the length of the appended data is E * G and not just E. Therefore, the used formula is correct, the diagram is not.
Thanks for pointing this out!
Regards,