[PATCH v4 0/2] mtd: hisilicon: add a new driver for NAND controller of hisilicon hip04 Soc

[Brian Norris]

On Thu, Dec 11, 2014 at 08:02:27PM +0800, Zhou Wang wrote:
> I have tested this NAND controller driver on the MTD test modules
> (drivers/mtd/tests/*). All tests passed except mtd_nandbiterrs.ko.
> 
> Here is the test log for mtd_nandbiterrs:
> /home # insmod mtd_nandbiterrs.ko dev=2 page_offset=1 seed=110 mode=0
> [  100.484995]
> [  100.490082] ==================================================
> [  100.509657] mtd_nandbiterrs: MTD device: 2
> [  100.523413] mtd_nandbiterrs: MTD device size 8388608,
> eraseblock=131072, page=2048, oob=64
> [  100.551134] mtd_nandbiterrs: Device uses 2 subpages of 1024 bytes
> [  100.571585] mtd_nandbiterrs: Using page=1, offset=2048, eraseblock=0
> [  104.431136] mtd_nandbiterrs: incremental biterrors test
> [  104.448872] mtd_nandbiterrs: write_page
> [  104.463193] mtd_nandbiterrs: rewrite page
> [  104.477620] mtd_nandbiterrs: read_page
> [  104.490898] mtd_nandbiterrs: verify_page
> [  104.504338] mtd_nandbiterrs: Successfully corrected 0 bit errors
> per subpage
> [  104.527985] mtd_nandbiterrs: Inserted biterror @ 0/5
> [  104.544673] mtd_nandbiterrs: Inserted biterror @ 1024/2
> [  104.562197] mtd_nandbiterrs: rewrite page
> [  104.576766] mtd_nandbiterrs: read_page
> [  104.590052] mtd_nandbiterrs: verify_page
> [  104.603252] mtd_nandbiterrs: Error: page offset 0, expected 23, got 03
> [  104.625203] mtd_nandbiterrs: Error: page offset 1024, expected 06, got 02
> [  104.648056] mtd_nandbiterrs: ECC failure, read data is incorrect
> despite read success
> insmod: can't insert 'mtd_nandbiterrs.ko': Input/output error
> 
> The reason for above failure is that:
> In ECC mode, when rewriting page data to NAND flash, the NAND
> controller will also produce ECC code and write them to NAND flash
> as well. So when we read data from NAND flash, there is no need to
> correct the error bit. We read what we write to the flash.

BTW, your explanation doesn't seem quite right. The problem is that
even though mtd_read() didn't report errors, the data doesn't match
what's written. It's not that there was "no need to correct the error
bit".

I'd recommend digging a little more to figure out what's wrong here. You
might need to instrument the nandbiterrs test. This is possibly
highlighting a driver bug [1].

Brian

[1] Besised simply that you didn't implement write_page_raw(). The
default nand_write_page_raw() implementation looks just like your
non-raw version.