[PATCH 1/1] Fix ECC Correction bug for SMC ordering for NDFC driver.

Fix ECC Correction bug where the byte offset location were double
fliped causing correction routine to toggle the wrong byte location
in the ECC segment. The ndfc_calculate_ecc routine change the order
of getting the ECC code.
        /* The NDFC uses Smart Media (SMC) bytes order */
        ecc_code[0] = p[2];
        ecc_code[1] = p[1];
        ecc_code[2] = p[3];
But in the Correction algorithm when calculating the byte offset
location, the b1 is used as the upper part of the address. Which
again reverse the order making the final byte offset address 
location incorrect.
	byte_addr = (addressbits[b1] << 4) + addressbits[b0];
The order is change to read it in straight and let the correction
function to revert it to SMC order.

Signed-off-by: Feng Kan <redacted>
Acked-by: Victor Gallardo <redacted>
Acked-by: Prodyut Hazarika <redacted>
---
 drivers/mtd/nand/ndfc.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/ndfc.c b/drivers/mtd/nand/ndfc.c
index 5906c40..d9d3e6e 100644
--- a/drivers/mtd/nand/ndfc.c
+++ b/drivers/mtd/nand/ndfc.c

@@ -101,8 +101,8 @@ static int ndfc_calculate_ecc(struct mtd_info *mtd,
 	wmb();
 	ecc = in_be32(ndfc->ndfcbase + NDFC_ECC);
 	/* The NDFC uses Smart Media (SMC) bytes order */
-	ecc_code[0] = p[2];
-	ecc_code[1] = p[1];
+	ecc_code[0] = p[1];
+	ecc_code[1] = p[2];
 	ecc_code[2] = p[3];
 
 	return 0;

-- 
1.5.5

On Thu, 20 Aug 2009 17:19:17 -0700
Feng Kan [off-list ref] wrote:

Acked-by: Sean MacLennan <redacted>

Cheers,
   Sean

On Friday 21 August 2009 02:19:17 Feng Kan wrote:

Fix ECC Correction bug where the byte offset location were double
fliped causing correction routine to toggle the wrong byte location
in the ECC segment. The ndfc_calculate_ecc routine change the order
of getting the ECC code.
        /* The NDFC uses Smart Media (SMC) bytes order */
        ecc_code[0] = p[2];
        ecc_code[1] = p[1];
        ecc_code[2] = p[3];
But in the Correction algorithm when calculating the byte offset
location, the b1 is used as the upper part of the address. Which
again reverse the order making the final byte offset address
location incorrect.
	byte_addr = (addressbits[b1] << 4) + addressbits[b0];
The order is change to read it in straight and let the correction
function to revert it to SMC order.

Signed-off-by: Feng Kan <redacted>
Acked-by: Victor Gallardo <redacted>
Acked-by: Prodyut Hazarika <redacted>

Acked-by: Stefan Roese <sr@denx.de>

Would be great if we could get this fix into 2.6.31.

Cheers,
Stefan

On Thu, 20 Aug 2009 17:19:17 -0700
Feng Kan [off-list ref] wrote:

Fix ECC Correction bug where the byte offset location were double
fliped causing correction routine to toggle the wrong byte location
in the ECC segment. The ndfc_calculate_ecc routine change the order
of getting the ECC code.

It looks like another fix for this bug is to leave the current code
alone and turn off CONFIG_MTD_NAND_ECC_SMC.

This could be a better fix if this is the way u-boot currently works.
Has anybody verified if the current u-boot has the ECC problem?

Cheers,
   Sean

Yes, I have considered that. However, it would make the #define rather confusing
for the rest.

Cheers,
Feng


-----Original Message-----
From: Sean MacLennan [mailto:smaclennan@pikatech.com]
Sent: Fri 8/21/2009 11:55 AM
To: Feng Kan
Cc: linuxppc-dev@ozlabs.org; linux-mtd@lists.infradead.org; Feng Kan
Subject: Re: [PATCH 1/1] Fix ECC Correction bug for SMC ordering for NDFC driver.
 
On Thu, 20 Aug 2009 17:19:17 -0700
Feng Kan [off-list ref] wrote:

Fix ECC Correction bug where the byte offset location were double
fliped causing correction routine to toggle the wrong byte location
in the ECC segment. The ndfc_calculate_ecc routine change the order
of getting the ECC code.

It looks like another fix for this bug is to leave the current code
alone and turn off CONFIG_MTD_NAND_ECC_SMC.

This could be a better fix if this is the way u-boot currently works.
Has anybody verified if the current u-boot has the ECC problem?

Cheers,
   Sean

Hi Sean=0A=0A> =0A> It looks like another fix for this bug is to leave the =
current code=0A> alone and turn off CONFIG_MTD_NAND_ECC_SMC.=0A=0AThis=A0wo=
uld fix the problem by hiding the issue. Fengs patch is=A0the correct way t=
o go.=0A=0ABest Regards,=0A=0AVictor Gallardo