[PATCH v5 0/2] Add support for read retry

[PATCH v5 0/2] Add support for read retry

[Cheng Ming Lin]

From: Cheng Ming Lin <chengminglin@mxic.com.tw>

When the host ECC fails to correct the data error of NAND device,
there's a special read for data recovery method which can be setup
by the host for the next read. There are several retry levels that
can be attempted until the lost data is recovered or definitely
assumed lost.

For more detailed information, please refer to the link below:
Link: https://www.macronix.com/Lists/Datasheet/Attachments/9034/MX35LF1G24AD,%203V,%201Gb,%20v1.4.pdf

v5:
* rebase on nand/next

v4:
* If set_read_retry fails, it should back to read_retry level 0

v3:
* If set_read_retry fails, it should return an error

v2:
* Remove fixups
* Remove the function of init_read_retry

Cheng Ming Lin (2):
  mtd: spi-nand: Add read retry support
  mtd: spi-nand: macronix: Add support for read retry

 drivers/mtd/nand/spi/core.c     | 35 ++++++++++++++-
 drivers/mtd/nand/spi/macronix.c | 79 ++++++++++++++++++++++++++-------
 include/linux/mtd/spinand.h     | 15 +++++++
 3 files changed, 112 insertions(+), 17 deletions(-)

-- 
2.25.1

[PATCH v5 1/2] mtd: spi-nand: Add read retry support

[Cheng Ming Lin]

From: Cheng Ming Lin <chengminglin@mxic.com.tw>

When the host ECC fails to correct the data error of NAND device,
there's a special read for data recovery method which can be setup
by the host for the next read. There are several retry levels that
can be attempted until the lost data is recovered or definitely
assumed lost.

Signed-off-by: Cheng Ming Lin <chengminglin@mxic.com.tw>
---
 drivers/mtd/nand/spi/core.c | 35 +++++++++++++++++++++++++++++++++--
 include/linux/mtd/spinand.h | 15 +++++++++++++++
 2 files changed, 48 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c
index dc5b11fa7a15..d16e42cf8fae 100644
--- a/drivers/mtd/nand/spi/core.c
+++ b/drivers/mtd/nand/spi/core.c
@@ -700,11 +700,15 @@ static int spinand_mtd_regular_page_read(struct mtd_info *mtd, loff_t from,
 {
 	struct spinand_device *spinand = mtd_to_spinand(mtd);
 	struct nand_device *nand = mtd_to_nanddev(mtd);
+	struct mtd_ecc_stats old_stats;
 	struct nand_io_iter iter;
 	bool disable_ecc = false;
 	bool ecc_failed = false;
+	unsigned int retry_mode = 0;
 	int ret;
 
+	old_stats = mtd->ecc_stats;
+
 	if (ops->mode == MTD_OPS_RAW || !mtd->ooblayout)
 		disable_ecc = true;
 
@@ -716,18 +720,43 @@ static int spinand_mtd_regular_page_read(struct mtd_info *mtd, loff_t from,
 		if (ret)
 			break;
 
+read_retry:
 		ret = spinand_read_page(spinand, &iter.req);
 		if (ret < 0 && ret != -EBADMSG)
 			break;
 
-		if (ret == -EBADMSG)
+		if (ret == -EBADMSG && spinand->set_read_retry) {
+			if (spinand->read_retries && (++retry_mode <= spinand->read_retries)) {
+				ret = spinand->set_read_retry(spinand, retry_mode);
+				if (ret < 0) {
+					spinand->set_read_retry(spinand, 0);
+					return ret;
+				}
+
+				/* Reset ecc_stats; retry */
+				mtd->ecc_stats = old_stats;
+				goto read_retry;
+			} else {
+				/* No more retry modes; real failure */
+				ecc_failed = true;
+			}
+		} else if (ret == -EBADMSG) {
 			ecc_failed = true;
-		else
+		} else {
 			*max_bitflips = max_t(unsigned int, *max_bitflips, ret);
+		}
 
 		ret = 0;
 		ops->retlen += iter.req.datalen;
 		ops->oobretlen += iter.req.ooblen;
+
+		/* Reset to retry mode 0 */
+		if (retry_mode) {
+			retry_mode = 0;
+			ret = spinand->set_read_retry(spinand, retry_mode);
+			if (ret < 0)
+				return ret;
+		}
 	}
 
 	if (ecc_failed && !ret)
@@ -1320,6 +1349,8 @@ int spinand_match_and_init(struct spinand_device *spinand,
 		spinand->set_cont_read = table[i].set_cont_read;
 		spinand->fact_otp = &table[i].fact_otp;
 		spinand->user_otp = &table[i].user_otp;
+		spinand->read_retries = table[i].read_retries;
+		spinand->set_read_retry = table[i].set_read_retry;
 
 		op = spinand_select_op_variant(spinand,
 					       info->op_variants.read_cache);
diff --git a/include/linux/mtd/spinand.h b/include/linux/mtd/spinand.h
index 73424405232a..5837a09ab9d8 100644
--- a/include/linux/mtd/spinand.h
+++ b/include/linux/mtd/spinand.h
@@ -452,6 +452,8 @@ struct spinand_user_otp {
  * @set_cont_read: enable/disable continuous cached reads
  * @fact_otp: SPI NAND factory OTP info.
  * @user_otp: SPI NAND user OTP info.
+ * @read_retries: the number of read retry modes supported
+ * @set_read_retry: enable/disable read retry for data recovery
  *
  * Each SPI NAND manufacturer driver should have a spinand_info table
  * describing all the chips supported by the driver.
@@ -474,6 +476,9 @@ struct spinand_info {
 			     bool enable);
 	struct spinand_fact_otp fact_otp;
 	struct spinand_user_otp user_otp;
+	unsigned int read_retries;
+	int (*set_read_retry)(struct spinand_device *spinand,
+			     unsigned int read_retry);
 };
 
 #define SPINAND_ID(__method, ...)					\
@@ -520,6 +525,10 @@ struct spinand_info {
 		.ops = __ops,						\
 	}
 
+#define SPINAND_READ_RETRY(__read_retries, __set_read_retry)		\
+	.read_retries = __read_retries,				\
+	.set_read_retry = __set_read_retry,
+
 #define SPINAND_INFO(__model, __id, __memorg, __eccreq, __op_variants,	\
 		     __flags, ...)					\
 	{								\
@@ -572,6 +581,8 @@ struct spinand_dirmap {
  * @priv: manufacturer private data
  * @fact_otp: SPI NAND factory OTP info.
  * @user_otp: SPI NAND user OTP info.
+ * @read_retries: the number of read retry modes supported
+ * @set_read_retry: Enable/disable the read retry feature
  */
 struct spinand_device {
 	struct nand_device base;
@@ -607,6 +618,10 @@ struct spinand_device {
 
 	const struct spinand_fact_otp *fact_otp;
 	const struct spinand_user_otp *user_otp;
+
+	unsigned int read_retries;
+	int (*set_read_retry)(struct spinand_device *spinand,
+			     unsigned int retry_mode);
 };
 
 /**
-- 
2.25.1

[PATCH v5 2/2] mtd: spi-nand: macronix: Add support for read retry

[Cheng Ming Lin]

From: Cheng Ming Lin <chengminglin@mxic.com.tw>

Add read retry support.

The Special Read for Data Recovery operation is enabled by
Set Feature function.

There are 5 modes for the user to recover the lost data.

Signed-off-by: Cheng Ming Lin <chengminglin@mxic.com.tw>
---
 drivers/mtd/nand/spi/macronix.c | 79 ++++++++++++++++++++++++++-------
 1 file changed, 64 insertions(+), 15 deletions(-)

diff --git a/drivers/mtd/nand/spi/macronix.c b/drivers/mtd/nand/spi/macronix.c
index 3dc4d63d6832..1ef08ad850a2 100644
--- a/drivers/mtd/nand/spi/macronix.c
+++ b/drivers/mtd/nand/spi/macronix.c
@@ -14,6 +14,8 @@
 #define MACRONIX_ECCSR_BF_LAST_PAGE(eccsr) FIELD_GET(GENMASK(3, 0), eccsr)
 #define MACRONIX_ECCSR_BF_ACCUMULATED_PAGES(eccsr) FIELD_GET(GENMASK(7, 4), eccsr)
 #define MACRONIX_CFG_CONT_READ         BIT(2)
+#define MACRONIX_FEATURE_ADDR_READ_RETRY 0x70
+#define MACRONIX_NUM_READ_RETRY_MODES 5
 
 #define STATUS_ECC_HAS_BITFLIPS_THRESHOLD (3 << 4)
 
@@ -136,6 +138,23 @@ static int macronix_set_cont_read(struct spinand_device *spinand, bool enable)
 	return 0;
 }
 
+/**
+ * macronix_set_read_retry - Set the retry mode
+ * @spinand: SPI NAND device
+ * @retry_mode: Specify which retry mode to set
+ *
+ * Return: 0 on success, a negative error code otherwise.
+ */
+static int macronix_set_read_retry(struct spinand_device *spinand,
+					     unsigned int retry_mode)
+{
+	struct spi_mem_op op = SPINAND_SET_FEATURE_OP(MACRONIX_FEATURE_ADDR_READ_RETRY,
+						      spinand->scratchbuf);
+
+	*spinand->scratchbuf = retry_mode;
+	return spi_mem_exec_op(spinand->spimem, &op);
+}
+
 static const struct spinand_info macronix_spinand_table[] = {
 	SPINAND_INFO("MX35LF1GE4AB",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x12),
@@ -168,7 +187,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 		     SPINAND_HAS_QE_BIT,
 		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout,
 				     macronix_ecc_get_status),
-		     SPINAND_CONT_READ(macronix_set_cont_read)),
+		     SPINAND_CONT_READ(macronix_set_cont_read),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35LF4GE4AD",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x37, 0x03),
 		     NAND_MEMORG(1, 4096, 128, 64, 2048, 40, 1, 1, 1),
@@ -179,7 +200,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 		     SPINAND_HAS_QE_BIT,
 		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout,
 				     macronix_ecc_get_status),
-		     SPINAND_CONT_READ(macronix_set_cont_read)),
+		     SPINAND_CONT_READ(macronix_set_cont_read),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35LF1G24AD",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x14, 0x03),
 		     NAND_MEMORG(1, 2048, 128, 64, 1024, 20, 1, 1, 1),
@@ -188,7 +211,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 					      &write_cache_variants,
 					      &update_cache_variants),
 		     SPINAND_HAS_QE_BIT,
-		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout, NULL)),
+		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout, NULL),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35LF2G24AD",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x24, 0x03),
 		     NAND_MEMORG(1, 2048, 128, 64, 2048, 40, 2, 1, 1),
@@ -198,7 +223,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 					      &update_cache_variants),
 		     SPINAND_HAS_QE_BIT |
 		     SPINAND_HAS_PROG_PLANE_SELECT_BIT,
-		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout, NULL)),
+		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout, NULL),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35LF2G24AD-Z4I8",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x64, 0x03),
 		     NAND_MEMORG(1, 2048, 128, 64, 2048, 40, 1, 1, 1),
@@ -207,7 +234,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 					      &write_cache_variants,
 					      &update_cache_variants),
 		     SPINAND_HAS_QE_BIT,
-		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout, NULL)),
+		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout, NULL),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35LF4G24AD",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x35, 0x03),
 		     NAND_MEMORG(1, 4096, 256, 64, 2048, 40, 2, 1, 1),
@@ -217,7 +246,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 					      &update_cache_variants),
 		     SPINAND_HAS_QE_BIT |
 		     SPINAND_HAS_PROG_PLANE_SELECT_BIT,
-		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout, NULL)),
+		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout, NULL),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35LF4G24AD-Z4I8",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x75, 0x03),
 		     NAND_MEMORG(1, 4096, 256, 64, 2048, 40, 1, 1, 1),
@@ -226,7 +257,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 					      &write_cache_variants,
 					      &update_cache_variants),
 		     SPINAND_HAS_QE_BIT,
-		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout, NULL)),
+		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout, NULL),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX31LF1GE4BC",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x1e),
 		     NAND_MEMORG(1, 2048, 64, 64, 1024, 20, 1, 1, 1),
@@ -270,7 +303,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 		     SPINAND_HAS_QE_BIT |
 		     SPINAND_HAS_PROG_PLANE_SELECT_BIT,
 		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout,
-				     macronix_ecc_get_status)),
+				     macronix_ecc_get_status),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35UF4G24AD-Z4I8",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xf5, 0x03),
 		     NAND_MEMORG(1, 4096, 256, 64, 2048, 40, 1, 1, 1),
@@ -280,7 +315,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 					      &update_cache_variants),
 		     SPINAND_HAS_QE_BIT,
 		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout,
-				     macronix_ecc_get_status)),
+				     macronix_ecc_get_status),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35UF4GE4AD",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xb7, 0x03),
 		     NAND_MEMORG(1, 4096, 256, 64, 2048, 40, 1, 1, 1),
@@ -291,7 +328,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 		     SPINAND_HAS_QE_BIT,
 		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout,
 				     macronix_ecc_get_status),
-		     SPINAND_CONT_READ(macronix_set_cont_read)),
+		     SPINAND_CONT_READ(macronix_set_cont_read),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35UF2G14AC",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xa0),
 		     NAND_MEMORG(1, 2048, 64, 64, 2048, 40, 2, 1, 1),
@@ -314,7 +353,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 		     SPINAND_HAS_QE_BIT |
 		     SPINAND_HAS_PROG_PLANE_SELECT_BIT,
 		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout,
-				     macronix_ecc_get_status)),
+				     macronix_ecc_get_status),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35UF2G24AD-Z4I8",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xe4, 0x03),
 		     NAND_MEMORG(1, 2048, 128, 64, 2048, 40, 1, 1, 1),
@@ -324,7 +365,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 					      &update_cache_variants),
 		     SPINAND_HAS_QE_BIT,
 		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout,
-				     macronix_ecc_get_status)),
+				     macronix_ecc_get_status),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35UF2GE4AD",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xa6, 0x03),
 		     NAND_MEMORG(1, 2048, 128, 64, 2048, 40, 1, 1, 1),
@@ -335,7 +378,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 		     SPINAND_HAS_QE_BIT,
 		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout,
 				     macronix_ecc_get_status),
-		     SPINAND_CONT_READ(macronix_set_cont_read)),
+		     SPINAND_CONT_READ(macronix_set_cont_read),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35UF2GE4AC",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xa2, 0x01),
 		     NAND_MEMORG(1, 2048, 64, 64, 2048, 40, 1, 1, 1),
@@ -366,7 +411,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 					      &update_cache_variants),
 		     SPINAND_HAS_QE_BIT,
 		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout,
-				     macronix_ecc_get_status)),
+				     macronix_ecc_get_status),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35UF1GE4AD",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x96, 0x03),
 		     NAND_MEMORG(1, 2048, 128, 64, 1024, 20, 1, 1, 1),
@@ -377,7 +424,9 @@ static const struct spinand_info macronix_spinand_table[] = {
 		     SPINAND_HAS_QE_BIT,
 		     SPINAND_ECCINFO(&mx35lfxge4ab_ooblayout,
 				     macronix_ecc_get_status),
-		     SPINAND_CONT_READ(macronix_set_cont_read)),
+		     SPINAND_CONT_READ(macronix_set_cont_read),
+		     SPINAND_READ_RETRY(MACRONIX_NUM_READ_RETRY_MODES,
+					macronix_set_read_retry)),
 	SPINAND_INFO("MX35UF1GE4AC",
 		     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x92, 0x01),
 		     NAND_MEMORG(1, 2048, 64, 64, 1024, 20, 1, 1, 1),
-- 
2.25.1

Re: [PATCH v5 0/2] Add support for read retry

[Miquel Raynal]

On Mon, 24 Feb 2025 15:03:47 +0800, Cheng Ming Lin wrote:
> From: Cheng Ming Lin <chengminglin@mxic.com.tw>
> 
> When the host ECC fails to correct the data error of NAND device,
> there's a special read for data recovery method which can be setup
> by the host for the next read. There are several retry levels that
> can be attempted until the lost data is recovered or definitely
> assumed lost.
> 
> [...]

Applied to nand/next, thanks!

[1/2] mtd: spi-nand: Add read retry support
      commit: c2c64e15172a1def9ccd104e1e7e3d41ac5d9853
[2/2] mtd: spi-nand: macronix: Add support for read retry
      commit: da3558d940c685048245c58ea536ca6f80025c87

Patche(s) should be available on mtd/linux.git and will be
part of the next PR (provided that no robot complains by then).

Kind regards,
Miquèl