src/kernel/linux/v4.14/Documentation/mtd/nand/pxa3xx-nand.txt

About this document
===================

Some notes about Marvell's NAND controller available in PXA and Armada 370/XP
SoC (aka NFCv1 and NFCv2), with an emphasis on the latter.

NFCv2 controller background
===========================

The controller has a 2176 bytes FIFO buffer. Therefore, in order to support
larger pages, I/O operations on 4 KiB and 8 KiB pages is done with a set of
chunked transfers.

For instance, if we choose a 2048 data chunk and set "BCH" ECC (see below)
we'll have this layout in the pages:

  ------------------------------------------------------------------------------
  | 2048B data | 32B spare | 30B ECC || 2048B data | 32B spare | 30B ECC | ... |
  ------------------------------------------------------------------------------

The driver reads the data and spare portions independently and builds an internal
buffer with this layout (in the 4 KiB page case):

  ------------------------------------------
  |     4096B data     |     64B spare     |
  ------------------------------------------

Also, for the READOOB command the driver disables the ECC and reads a 'spare + ECC'
OOB, one per chunk read.

  -------------------------------------------------------------------
  |     4096B data     |  32B spare | 30B ECC | 32B spare | 30B ECC |
  -------------------------------------------------------------------

So, in order to achieve reading (for instance), we issue several READ0 commands
(with some additional controller-specific magic) and read two chunks of 2080B
(2048 data + 32 spare) each.
The driver accommodates this data to expose the NAND core a contiguous buffer
(4096 data + spare) or (4096 + spare + ECC + spare + ECC).

ECC
===

The controller has built-in hardware ECC capabilities. In addition it is
configurable between two modes: 1) Hamming, 2) BCH.

Note that the actual BCH mode: BCH-4 or BCH-8 will depend on the way
the controller is configured to transfer the data.

In the BCH mode the ECC code will be calculated for each transferred chunk
and expected to be located (when reading/programming) right after the spare
bytes as the figure above shows.

So, repeating the above scheme, a 2048B data chunk will be followed by 32B
spare, and then the ECC controller will read/write the ECC code (30B in
this case):

  ------------------------------------
  | 2048B data | 32B spare | 30B ECC |
  ------------------------------------

If the ECC mode is 'BCH' then the ECC is *always* 30 bytes long.
If the ECC mode is 'Hamming' the ECC is 6 bytes long, for each 512B block.
So in Hamming mode, a 2048B page will have a 24B ECC.

Despite all of the above, the controller requires the driver to only read or
write in multiples of 8-bytes, because the data buffer is 64-bits.

OOB
===

Because of the above scheme, and because the "spare" OOB is really located in
the middle of a page, spare OOB cannot be read or write independently of the
data area. In other words, in order to read the OOB (aka READOOB), the entire
page (aka READ0) has to be read.

In the same sense, in order to write to the spare OOB the driver has to write
an *entire* page.

Factory bad blocks handling
===========================

Given the ECC BCH requires to layout the device's pages in a split
data/OOB/data/OOB way, the controller has a view of the flash page that's
different from the specified (aka the manufacturer's) view. In other words,

Factory view:

  -----------------------------------------------
  |                    Data           |x  OOB   |
  -----------------------------------------------

Driver's view:

  -----------------------------------------------
  |      Data      | OOB |      Data   x  | OOB |
  -----------------------------------------------

It can be seen from the above, that the factory bad block marker must be
searched within the 'data' region, and not in the usual OOB region.

In addition, this means under regular usage the driver will write such
position (since it belongs to the data region) and every used block is
likely to be marked as bad.

For this reason, marking the block as bad in the OOB is explicitly
disabled by using the NAND_BBT_NO_OOB_BBM option in the driver. The rationale
for this is that there's no point in marking a block as bad, because good
blocks are also 'marked as bad' (in the OOB BBM sense) under normal usage.

Instead, the driver relies on the bad block table alone, and should only perform
the bad block scan on the very first time (when the device hasn't been used).