[PATCH v6] mtd: rawnand: use bit-wise majority to recover the contents of ONFI parameter

[PATCH v6] mtd: rawnand: use bit-wise majority to recover the contents of ONFI parameter

[Jane Wan]

Per ONFI specification (Rev. 4.0), if all parameter pages have invalid
CRC values, the bit-wise majority may be used to recover the contents of
the parameter pages from the parameter page copies present.

Signed-off-by: Jane Wan <Jane.Wan@nokia.com>
---
v6: support the cases that srcbufs are not contiguous
v5: make the bit-wise majority functon generic
v4: move the bit-wise majority code in a separate function
v3: fix warning message detected by kbuild test robot
v2: rebase the changes on top of v4.17-rc1

 drivers/mtd/nand/raw/nand_base.c |   52 ++++++++++++++++++++++++++++++++++----
 1 file changed, 47 insertions(+), 5 deletions(-)

diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c
index 72f3a89..acf905c 100644
--- a/drivers/mtd/nand/raw/nand_base.c
+++ b/drivers/mtd/nand/raw/nand_base.c
@@ -5087,6 +5087,35 @@ static int nand_flash_detect_ext_param_page(struct nand_chip *chip,
 }
 
 /*
+ * Recover data with bit-wise majority
+ */
+static void nand_bit_wise_majority(const void **srcbufs,
+				   unsigned int nsrcbufs,
+				   void *dstbuf,
+				   unsigned int bufsize)
+{
+	int i, j, k;
+
+	for (i = 0; i < bufsize; i++) {
+		u8 cnt, val;
+
+		val = 0;
+		for (j = 0; j < 8; j++) {
+			cnt = 0;
+			for (k = 0; k < nsrcbufs; k++) {
+				const u8 *srcbuf = srcbufs[k];
+
+				if (srcbuf[i] & BIT(j))
+					cnt++;
+			}
+			if (cnt > nsrcbufs / 2)
+				val |= BIT(j);
+		}
+		((u8 *)dstbuf)[i] = val;
+	}
+}
+
+/*
  * Check if the NAND chip is ONFI compliant, returns 1 if it is, 0 otherwise.
  */
 static int nand_flash_detect_onfi(struct nand_chip *chip)
@@ -5102,7 +5131,7 @@ static int nand_flash_detect_onfi(struct nand_chip *chip)
 		return 0;
 
 	/* ONFI chip: allocate a buffer to hold its parameter page */
-	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	p = kzalloc((sizeof(*p) * 3), GFP_KERNEL);
 	if (!p)
 		return -ENOMEM;
 
@@ -5113,21 +5142,34 @@ static int nand_flash_detect_onfi(struct nand_chip *chip)
 	}
 
 	for (i = 0; i < 3; i++) {
-		ret = nand_read_data_op(chip, p, sizeof(*p), true);
+		ret = nand_read_data_op(chip, &p[i], sizeof(*p), true);
 		if (ret) {
 			ret = 0;
 			goto free_onfi_param_page;
 		}
 
-		if (onfi_crc16(ONFI_CRC_BASE, (uint8_t *)p, 254) ==
+		if (onfi_crc16(ONFI_CRC_BASE, (u8 *)&p[i], 254) ==
 				le16_to_cpu(p->crc)) {
+			if (i)
+				memcpy(p, &p[i], sizeof(*p));
 			break;
 		}
 	}
 
 	if (i == 3) {
-		pr_err("Could not find valid ONFI parameter page; aborting\n");
-		goto free_onfi_param_page;
+		const void *srcbufs[3] = {p, p + 1, p + 2};
+
+		pr_err("Could not find valid ONFI parameter page\n");
+		pr_info("Recover ONFI params with bit-wise majority\n");
+
+		nand_bit_wise_majority(srcbufs, ARRAY_SIZE(srcbufs), p,
+				       sizeof(*p));
+
+		if (onfi_crc16(ONFI_CRC_BASE, (u8 *)p, 254) !=
+				le16_to_cpu(p->crc)) {
+			pr_err("ONFI parameter recovery failed, aborting\n");
+			goto free_onfi_param_page;
+		}
 	}
 
 	/* Check version */
-- 
1.7.9.5

Re: [PATCH v6] mtd: rawnand: use bit-wise majority to recover the contents of ONFI parameter

[Boris Brezillon]

On Thu, 10 May 2018 14:28:37 -0700
Jane Wan <Jane.Wan@nokia.com> wrote:

> Per ONFI specification (Rev. 4.0), if all parameter pages have invalid
> CRC values, the bit-wise majority may be used to recover the contents of
> the parameter pages from the parameter page copies present.
> 
> Signed-off-by: Jane Wan <Jane.Wan@nokia.com>
> ---
> v6: support the cases that srcbufs are not contiguous
> v5: make the bit-wise majority functon generic
> v4: move the bit-wise majority code in a separate function
> v3: fix warning message detected by kbuild test robot
> v2: rebase the changes on top of v4.17-rc1
> 
>  drivers/mtd/nand/raw/nand_base.c |   52 ++++++++++++++++++++++++++++++++++----
>  1 file changed, 47 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c
> index 72f3a89..acf905c 100644
> --- a/drivers/mtd/nand/raw/nand_base.c
> +++ b/drivers/mtd/nand/raw/nand_base.c
> @@ -5087,6 +5087,35 @@ static int nand_flash_detect_ext_param_page(struct nand_chip *chip,
>  }
>  
>  /*
> + * Recover data with bit-wise majority
> + */
> +static void nand_bit_wise_majority(const void **srcbufs,
> +				   unsigned int nsrcbufs,
> +				   void *dstbuf,
> +				   unsigned int bufsize)
> +{
> +	int i, j, k;
> +
> +	for (i = 0; i < bufsize; i++) {
> +		u8 cnt, val;
> +
> +		val = 0;
> +		for (j = 0; j < 8; j++) {
> +			cnt = 0;
> +			for (k = 0; k < nsrcbufs; k++) {
> +				const u8 *srcbuf = srcbufs[k];
> +
> +				if (srcbuf[i] & BIT(j))
> +					cnt++;
> +			}
> +			if (cnt > nsrcbufs / 2)
> +				val |= BIT(j);
> +		}
> +		((u8 *)dstbuf)[i] = val;
> +	}
> +}
> +
> +/*
>   * Check if the NAND chip is ONFI compliant, returns 1 if it is, 0 otherwise.
>   */
>  static int nand_flash_detect_onfi(struct nand_chip *chip)
> @@ -5102,7 +5131,7 @@ static int nand_flash_detect_onfi(struct nand_chip *chip)
>  		return 0;
>  
>  	/* ONFI chip: allocate a buffer to hold its parameter page */
> -	p = kzalloc(sizeof(*p), GFP_KERNEL);
> +	p = kzalloc((sizeof(*p) * 3), GFP_KERNEL);
>  	if (!p)
>  		return -ENOMEM;
>  
> @@ -5113,21 +5142,34 @@ static int nand_flash_detect_onfi(struct nand_chip *chip)
>  	}
>  
>  	for (i = 0; i < 3; i++) {
> -		ret = nand_read_data_op(chip, p, sizeof(*p), true);
> +		ret = nand_read_data_op(chip, &p[i], sizeof(*p), true);
>  		if (ret) {
>  			ret = 0;
>  			goto free_onfi_param_page;
>  		}
>  
> -		if (onfi_crc16(ONFI_CRC_BASE, (uint8_t *)p, 254) ==
> +		if (onfi_crc16(ONFI_CRC_BASE, (u8 *)&p[i], 254) ==
>  				le16_to_cpu(p->crc)) {
> +			if (i)
> +				memcpy(p, &p[i], sizeof(*p));
>  			break;
>  		}
>  	}
>  
>  	if (i == 3) {
> -		pr_err("Could not find valid ONFI parameter page; aborting\n");
> -		goto free_onfi_param_page;
> +		const void *srcbufs[3] = {p, p + 1, p + 2};
> +
> +		pr_err("Could not find valid ONFI parameter page\n");

Maybe pr_warn() here

> +		pr_info("Recover ONFI params with bit-wise majority\n");

and maybe you can pack the 2 messages:

		pr_warn("Could not find a valid ONFI parameter page, trying bit-wise majority to recover it");

> +
> +		nand_bit_wise_majority(srcbufs, ARRAY_SIZE(srcbufs), p,
> +				       sizeof(*p));
> +
> +		if (onfi_crc16(ONFI_CRC_BASE, (u8 *)p, 254) !=
> +				le16_to_cpu(p->crc)) {
> +			pr_err("ONFI parameter recovery failed, aborting\n");
> +			goto free_onfi_param_page;
> +		}
>  	}
>  
>  	/* Check version */

RE: [PATCH v6] mtd: rawnand: use bit-wise majority to recover the contents of ONFI parameter

[Wan, Jane (Nokia - US/Sunnyvale)]

Hi Boris,

I've sent v7 of the patch.

Thanks.
Jane

> -----Original Message-----
> From: Boris Brezillon [mailto:boris.brezillon@bootlin.com]
> Sent: Saturday, May 12, 2018 1:21 AM
> To: Wan, Jane (Nokia - US/Sunnyvale) <jane.wan@nokia.com>
> Cc: miquel.raynal@bootlin.com; dwmw2@infradead.org;
> computersforpeace@gmail.com; richard@nod.at; marek.vasut@gmail.com;
> yamada.masahiro@socionext.com; prabhakar.kushwaha@nxp.com;
> shawnguo@kernel.org; jagdish.gediya@nxp.com;
> shreeya.patel23498@gmail.com; linux-mtd@lists.infradead.org; linux-
> kernel@vger.kernel.org; Bos, Ties (Nokia - US/Sunnyvale) <ties.bos@nokia.com>
> Subject: Re: [PATCH v6] mtd: rawnand: use bit-wise majority to recover the
> contents of ONFI parameter
> 
> On Thu, 10 May 2018 14:28:37 -0700
> Jane Wan <Jane.Wan@nokia.com> wrote:
> 
> > Per ONFI specification (Rev. 4.0), if all parameter pages have invalid
> > CRC values, the bit-wise majority may be used to recover the contents
> > of the parameter pages from the parameter page copies present.
> >
> > Signed-off-by: Jane Wan <Jane.Wan@nokia.com>
> > ---
> > v6: support the cases that srcbufs are not contiguous
> > v5: make the bit-wise majority functon generic
> > v4: move the bit-wise majority code in a separate function
> > v3: fix warning message detected by kbuild test robot
> > v2: rebase the changes on top of v4.17-rc1
> >
> >  drivers/mtd/nand/raw/nand_base.c |   52
> ++++++++++++++++++++++++++++++++++----
> >  1 file changed, 47 insertions(+), 5 deletions(-)
> >
> > diff --git a/drivers/mtd/nand/raw/nand_base.c
> > b/drivers/mtd/nand/raw/nand_base.c
> > index 72f3a89..acf905c 100644
> > --- a/drivers/mtd/nand/raw/nand_base.c
> > +++ b/drivers/mtd/nand/raw/nand_base.c
> > @@ -5087,6 +5087,35 @@ static int
> > nand_flash_detect_ext_param_page(struct nand_chip *chip,  }
> >
> >  /*
> > + * Recover data with bit-wise majority  */ static void
> > +nand_bit_wise_majority(const void **srcbufs,
> > +				   unsigned int nsrcbufs,
> > +				   void *dstbuf,
> > +				   unsigned int bufsize)
> > +{
> > +	int i, j, k;
> > +
> > +	for (i = 0; i < bufsize; i++) {
> > +		u8 cnt, val;
> > +
> > +		val = 0;
> > +		for (j = 0; j < 8; j++) {
> > +			cnt = 0;
> > +			for (k = 0; k < nsrcbufs; k++) {
> > +				const u8 *srcbuf = srcbufs[k];
> > +
> > +				if (srcbuf[i] & BIT(j))
> > +					cnt++;
> > +			}
> > +			if (cnt > nsrcbufs / 2)
> > +				val |= BIT(j);
> > +		}
> > +		((u8 *)dstbuf)[i] = val;
> > +	}
> > +}
> > +
> > +/*
> >   * Check if the NAND chip is ONFI compliant, returns 1 if it is, 0 otherwise.
> >   */
> >  static int nand_flash_detect_onfi(struct nand_chip *chip) @@ -5102,7
> > +5131,7 @@ static int nand_flash_detect_onfi(struct nand_chip *chip)
> >  		return 0;
> >
> >  	/* ONFI chip: allocate a buffer to hold its parameter page */
> > -	p = kzalloc(sizeof(*p), GFP_KERNEL);
> > +	p = kzalloc((sizeof(*p) * 3), GFP_KERNEL);
> >  	if (!p)
> >  		return -ENOMEM;
> >
> > @@ -5113,21 +5142,34 @@ static int nand_flash_detect_onfi(struct
> nand_chip *chip)
> >  	}
> >
> >  	for (i = 0; i < 3; i++) {
> > -		ret = nand_read_data_op(chip, p, sizeof(*p), true);
> > +		ret = nand_read_data_op(chip, &p[i], sizeof(*p), true);
> >  		if (ret) {
> >  			ret = 0;
> >  			goto free_onfi_param_page;
> >  		}
> >
> > -		if (onfi_crc16(ONFI_CRC_BASE, (uint8_t *)p, 254) ==
> > +		if (onfi_crc16(ONFI_CRC_BASE, (u8 *)&p[i], 254) ==
> >  				le16_to_cpu(p->crc)) {
> > +			if (i)
> > +				memcpy(p, &p[i], sizeof(*p));
> >  			break;
> >  		}
> >  	}
> >
> >  	if (i == 3) {
> > -		pr_err("Could not find valid ONFI parameter page; aborting\n");
> > -		goto free_onfi_param_page;
> > +		const void *srcbufs[3] = {p, p + 1, p + 2};
> > +
> > +		pr_err("Could not find valid ONFI parameter page\n");
> 
> Maybe pr_warn() here
> 
> > +		pr_info("Recover ONFI params with bit-wise majority\n");
> 
> and maybe you can pack the 2 messages:
> 
> 		pr_warn("Could not find a valid ONFI parameter page, trying
> bit-wise majority to recover it");

[Jane] Changed as suggested.

> 
> > +
> > +		nand_bit_wise_majority(srcbufs, ARRAY_SIZE(srcbufs), p,
> > +				       sizeof(*p));
> > +
> > +		if (onfi_crc16(ONFI_CRC_BASE, (u8 *)p, 254) !=
> > +				le16_to_cpu(p->crc)) {
> > +			pr_err("ONFI parameter recovery failed, aborting\n");
> > +			goto free_onfi_param_page;
> > +		}
> >  	}
> >
> >  	/* Check version */