diff mbox

[v7] mtd: gpmi: Deal with bitflips in erased regions

Message ID 1388784439-23004-2-git-send-email-eliedebrauwer@gmail.com
State Superseded
Headers show

Commit Message

Elie De Brauwer Jan. 3, 2014, 9:27 p.m. UTC 
The BCH block typically used with a GPMI block on an i.MX28/i.MX6 is only
able to correct bitflips on data actually streamed through the block.
When erasing a block the data does not stream through the BCH block
and therefore no ECC data is written to the NAND chip. This causes
gpmi_ecc_read_page to return failure as soon as a single non-1-bit is
found in an erased page. Typically causing problems at higher levels
(ubifs corrupted empty space warnings). This problem was also observed
when using SLC NAND devices.

This patch configures the BCH block to mark a block as 'erased' if
not too much bitflips are found (by setting the erase threshold). A
consequence of this is that whenever an erased page is read, the
number of bitflips will be counted and corrected in software,
allowing the upper layers to take proper actions.

Signed-off-by: Elie De Brauwer <eliedebrauwer@gmail.com>
Acked-by: Peter Korsgaard <peter@korsgaard.com>
Acked-by: Huang Shijie <b32955@freescale.com>
---
 drivers/mtd/nand/gpmi-nand/bch-regs.h  |  1 +
 drivers/mtd/nand/gpmi-nand/gpmi-lib.c  |  9 +++++++
 drivers/mtd/nand/gpmi-nand/gpmi-nand.c | 43 +++++++++++++++++++++++++++++++---
 3 files changed, 50 insertions(+), 3 deletions(-)

Comments

Huang Shijie Jan. 4, 2014, 12:50 p.m. UTC | #1 
On Fri, Jan 03, 2014 at 10:27:19PM +0100, Elie De Brauwer wrote:
> The BCH block typically used with a GPMI block on an i.MX28/i.MX6 is only
> able to correct bitflips on data actually streamed through the block.
> When erasing a block the data does not stream through the BCH block
> and therefore no ECC data is written to the NAND chip. This causes
> gpmi_ecc_read_page to return failure as soon as a single non-1-bit is
> found in an erased page. Typically causing problems at higher levels
> (ubifs corrupted empty space warnings). This problem was also observed
> when using SLC NAND devices.
> 
> This patch configures the BCH block to mark a block as 'erased' if
> not too much bitflips are found (by setting the erase threshold). A
> consequence of this is that whenever an erased page is read, the
> number of bitflips will be counted and corrected in software,
> allowing the upper layers to take proper actions.
> 
> Signed-off-by: Elie De Brauwer <eliedebrauwer@gmail.com>
> Acked-by: Peter Korsgaard <peter@korsgaard.com>
> Acked-by: Huang Shijie <b32955@freescale.com>

I am okay with this patch.

thanks
Huang Shijie
Huang Shijie Jan. 7, 2014, 2:13 a.m. UTC | #2 
On Fri, Jan 03, 2014 at 10:27:19PM +0100, Elie De Brauwer wrote:
> 
> @@ -1003,15 +1028,27 @@ static int gpmi_ecc_read_page(struct mtd_info *mtd, struct nand_chip *chip,
>  	status = auxiliary_virt + nfc_geo->auxiliary_status_offset;
>  
>  	for (i = 0; i < nfc_geo->ecc_chunk_count; i++, status++) {
> -		if ((*status == STATUS_GOOD) || (*status == STATUS_ERASED))
> +		if (*status == STATUS_GOOD)
>  			continue;
>  
>  		if (*status == STATUS_UNCORRECTABLE) {
>  			mtd->ecc_stats.failed++;
>  			continue;
>  		}
> -		mtd->ecc_stats.corrected += *status;
> -		max_bitflips = max_t(unsigned int, max_bitflips, *status);
> +
> +		/* 

Hi Brian:
  Please do not forget to remove the whitespace at the end of the above line
  when you merge this patch.

thanks
Huang Shijie


> +		 * The number of bitflips are either counted in software
> +		 * in case of an erased chunk or otherwise reported by
> +		 * the BCH block.
> +		 */
> +		if (*status == STATUS_ERASED)
> +			flips = erased_sector_bitflips(payload_virt, i,
> +							       nfc_geo);
> +		else
> +			flips = *status;
> +
diff mbox

Patch

diff --git a/drivers/mtd/nand/gpmi-nand/bch-regs.h b/drivers/mtd/nand/gpmi-nand/bch-regs.h
index 588f537..b2104de 100644
--- a/drivers/mtd/nand/gpmi-nand/bch-regs.h
+++ b/drivers/mtd/nand/gpmi-nand/bch-regs.h
@@ -31,6 +31,7 @@ 
 
 #define HW_BCH_STATUS0				0x00000010
 #define HW_BCH_MODE				0x00000020
+#define BM_BCH_MODE_ERASE_THRESHOLD_MASK	0xff
 #define HW_BCH_ENCODEPTR			0x00000030
 #define HW_BCH_DATAPTR				0x00000040
 #define HW_BCH_METAPTR				0x00000050
diff --git a/drivers/mtd/nand/gpmi-nand/gpmi-lib.c b/drivers/mtd/nand/gpmi-nand/gpmi-lib.c
index dd1df60..b6a7aa8 100644
--- a/drivers/mtd/nand/gpmi-nand/gpmi-lib.c
+++ b/drivers/mtd/nand/gpmi-nand/gpmi-lib.c
@@ -256,6 +256,7 @@  int bch_set_geometry(struct gpmi_nand_data *this)
 	unsigned int ecc_strength;
 	unsigned int page_size;
 	unsigned int gf_len;
+	unsigned int erase_threshold;
 	int ret;
 
 	if (common_nfc_set_geometry(this))
@@ -298,6 +299,14 @@  int bch_set_geometry(struct gpmi_nand_data *this)
 			| BF_BCH_FLASH0LAYOUT1_DATAN_SIZE(block_size, this),
 			r->bch_regs + HW_BCH_FLASH0LAYOUT1);
 
+	/* Set the tolerance for bitflips when reading erased blocks. */
+	erase_threshold = gf_len / 2;
+	if (erase_threshold > bch_geo->ecc_strength)
+		erase_threshold = bch_geo->ecc_strength;
+
+	writel(erase_threshold & BM_BCH_MODE_ERASE_THRESHOLD_MASK,
+		r->bch_regs + HW_BCH_MODE);
+
 	/* Set *all* chip selects to use layout 0. */
 	writel(0, r->bch_regs + HW_BCH_LAYOUTSELECT);
 
diff --git a/drivers/mtd/nand/gpmi-nand/gpmi-nand.c b/drivers/mtd/nand/gpmi-nand/gpmi-nand.c
index e2f5820..eac6714 100644
--- a/drivers/mtd/nand/gpmi-nand/gpmi-nand.c
+++ b/drivers/mtd/nand/gpmi-nand/gpmi-nand.c
@@ -958,6 +958,30 @@  static void block_mark_swapping(struct gpmi_nand_data *this,
 	p[1] = (p[1] & mask) | (from_oob >> (8 - bit));
 }
 
+/*
+ * Count the number of 0 bits in a supposed to be
+ * erased region and correct them. Return the number
+ * of bitflips or zero when the region was correct.
+ */
+static unsigned int erased_sector_bitflips(unsigned char *data,
+					unsigned int chunk,
+					struct bch_geometry *geo)
+{
+	unsigned int flip_bits = 0;
+	int i;
+	int base = geo->ecc_chunk_size * chunk;
+
+	/* Count bitflips */
+	for (i = 0; i < geo->ecc_chunk_size; i++)
+		flip_bits += hweight8(~data[base + i]);
+
+	/* Correct bitflips by 0xFF'ing this chunk. */
+	if (flip_bits)
+		memset(&data[base], 0xFF, geo->ecc_chunk_size);
+
+	return flip_bits;
+}
+
 static int gpmi_ecc_read_page(struct mtd_info *mtd, struct nand_chip *chip,
 				uint8_t *buf, int oob_required, int page)
 {
@@ -969,6 +993,7 @@  static int gpmi_ecc_read_page(struct mtd_info *mtd, struct nand_chip *chip,
 	dma_addr_t    auxiliary_phys;
 	unsigned int  i;
 	unsigned char *status;
+	unsigned int  flips;
 	unsigned int  max_bitflips = 0;
 	int           ret;
 
@@ -1003,15 +1028,27 @@  static int gpmi_ecc_read_page(struct mtd_info *mtd, struct nand_chip *chip,
 	status = auxiliary_virt + nfc_geo->auxiliary_status_offset;
 
 	for (i = 0; i < nfc_geo->ecc_chunk_count; i++, status++) {
-		if ((*status == STATUS_GOOD) || (*status == STATUS_ERASED))
+		if (*status == STATUS_GOOD)
 			continue;
 
 		if (*status == STATUS_UNCORRECTABLE) {
 			mtd->ecc_stats.failed++;
 			continue;
 		}
-		mtd->ecc_stats.corrected += *status;
-		max_bitflips = max_t(unsigned int, max_bitflips, *status);
+
+		/* 
+		 * The number of bitflips are either counted in software
+		 * in case of an erased chunk or otherwise reported by
+		 * the BCH block.
+		 */
+		if (*status == STATUS_ERASED)
+			flips = erased_sector_bitflips(payload_virt, i,
+							       nfc_geo);
+		else
+			flips = *status;
+
+		mtd->ecc_stats.corrected += flips;
+		max_bitflips = max_t(unsigned int, max_bitflips, flips);
 	}
 
 	if (oob_required) {