UM1653 datasheet(14/45 Pages) STMICROELECTRONICS

14 / 45 page

NAND driver firmware modules

UM1653

14/45

DocID025024 Rev 1

NAND_ADDRESS NAND_GetAddress (uint32_t Address)

The above function translates a logical address into a physical one and stores it in a

structure element of type NAND_ADDRESS.

While writing to the NAND, the wear level algorithm should return the block to be written of

which the erase count is least. To maintain the list of USED, FREE and BAD blocks an array

is maintained: LUT[ ].

The previously fetched address for writing, and the free block obtained with least erase

counts are swapped in LUT and updated, this ensures that the write takes place at the block

with least erase count.

uint16_t NAND_GetFreeBlock (void)

The above function is called to get the free block for swap. The function returns the first free

block it finds in the LUT[ ]. This implies that the LUT[ ] should have the free blocks arranged

in the increasing order of erase count. The LUT[ ] is updated by the following function.

uint16_t NAND_BuildLUT (uint8_t ZoneNbr)

The above function arranges the bad block at the bottom of the array and the used and free

blocks are located in the upper part of the array.

uint16_t NAND_WearLeveling (void)

The above function sorts the free blocks in the ascending order based on the wear level

count. Now, the free block used for writing in the NAND would be the one with the least

erase count.

2.3

ECC

Unlike NOR Flash memory that does not require error correction code, NAND memory

needs to ensure data integrity.

The disadvantage of the NAND configuration is that when a cell is read, the sense amplifier

detects a signal much lower than for the NOR configuration because many transistors are in

series. Therefore access to a cell is not straightforward and must necessarily go through all

the cells in series which reduces precision and makes code error correction required.

There are three error correction codes:



The Hamming code can correct only one bit error.



The Reed Solomon code can correct more errors.



The BCH code can correct many errors and is more efficient than Reed Solomon.

2.3.1

Hamming code for NAND Flash

The Hamming code algorithm used by NAND Flash-based applications calculates two

values of ECC for a data packet. Each bit in the values of ECC parity represents half of the

bits of the data packet.

UM1653 датащи(PDF) 14 Page - STMicroelectronics

UM1653 датащи(HTML) 14 Page - STMicroelectronics

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

датащи скачать

ссылки URL