mtd: spinand: gigadevice: Add continuous read support
Subsystems (2)
PATCH 1/1 mtd: spinand: gigadevice: Add continuous read support
GD5F1GM9 supports a high-speed Continuous Read (CR) mode which allows the entire memory array to be read out by a single read command without re-issuing the 13h “Page Read to Cache” command for every page.
Datasheet summary
Feature register B0[3] (NR bit)
1 = Normal Read (default)
0 = Continuous Read (CR)
ECC status register 7Ch
Normal read: bits [3:0] → ECC status of last page Continuous read: bits [7:4] → accumulated max bit-flips over the whole sequential read operation.
Datasheet:
https://github.com/WT-886/DATASHEET/blob/main/GD5F1GM9-v1.0.pdf
--- drivers/mtd/nand/spi/gigadevice.c | 107 +++++++++++++++++++++++++++++- 1 file changed, 105 insertions(+), 2 deletions(-)
drivers/mtd/nand/spi/gigadevice.c +105 -2
@@ -4,6 +4,7 @@* Chuanhong Guo <[email protected]> */ +#include <linux/bitfield.h> #include <linux/device.h> #include <linux/kernel.h> #include <linux/mtd/spinand.h>
@@ -23,6 +24,18 @@#define GD5FXGQ4UXFXXG_STATUS_ECC_1_3_BITFLIPS (1 << 4)#define GD5FXGQ4UXFXXG_STATUS_ECC_UNCOR_ERROR (7 << 4) +/* Feature bit definitions */ +#define GD_FEATURE_NR BIT(3)/* Normal Read(1=normal,0=continuous) */ +#define GD_FEATURE_CRDC BIT(2)/* Continuous Read Dummy */ + +/* ECC status extraction helpers */ +#define GD_ECCSR_LAST_PAGE (eccsr) FIELD_GET(GENMASK(3, 0), eccsr) +#define GD_ECCSR_ACCUMULATED (eccsr) FIELD_GET(GENMASK(7, 4), eccsr) + +struct gigadevice_priv { +bool continuous_read; +}; + static SPINAND_OP_VARIANTS(read_cache_variants, SPINAND_PAGE_READ_FROM_CACHE_1S_4S_4S_OP(0, 1, NULL, 0, 0), SPINAND_PAGE_READ_FROM_CACHE_1S_1S_4S_OP(0, 1, NULL, 0, 0),
@@ -63,6 +76,74 @@ static SPINAND_OP_VARIANTS(update_cache_variants,SPINAND_PROG_LOAD_1S_1S_4S_OP (false,0 ,NULL ,0 ), SPINAND_PROG_LOAD_1S_1S_1S_OP(false, 0, NULL, 0)); +static int gd5fxgm9_get_eccsr (struct spinand_device *spinand,u8 *eccsr) +{ +struct gigadevice_priv *priv = spinand->priv; +struct spi_mem_op op =SPI_MEM_OP (SPI_MEM_OP_CMD (0x7c ,1 ), +SPI_MEM_OP_NO_ADDR , +SPI_MEM_OP_DUMMY (1 ,1 ), +SPI_MEM_OP_DATA_IN (1 , eccsr,1 )); +int ret; + + ret =spi_mem_exec_op (spinand->spimem, &op); +if (ret) +return ret; + +if (priv->continuous_read) + *eccsr =GD_ECCSR_ACCUMULATED (*eccsr); +else + *eccsr =GD_ECCSR_LAST_PAGE (*eccsr); + +return 0 ; +} + +static int gd5fxgm9_ecc_get_status (struct spinand_device *spinand,u8 status) +{ +struct nand_device *nand =spinand_to_nand (spinand); +u8 eccsr; +int ret; + +switch (status &STATUS_ECC_MASK ) { +case STATUS_ECC_NO_BITFLIPS : +return 0 ; + +case GD5FXGQ4XA_STATUS_ECC_1_7_BITFLIPS : + ret =gd5fxgm9_get_eccsr (spinand, spinand->scratchbuf); +if (ret) +return nanddev_get_ecc_conf (nand)->strength; + + eccsr = *spinand->scratchbuf; +if (WARN_ON (!eccsr || eccsr >nanddev_get_ecc_conf (nand)->strength)) +return nanddev_get_ecc_conf (nand)->strength; + +return eccsr; + +case GD5FXGQ4XA_STATUS_ECC_8_BITFLIPS : +return 8 ; + +case STATUS_ECC_UNCOR_ERROR : +return -EBADMSG ; + +default : +return -EINVAL ; + } +} + +static int gd5fxgm9_set_continuous_read (struct spinand_device *spinand,bool enable) +{ +struct gigadevice_priv *priv = spinand->priv; +int ret; + + ret =spinand_upd_cfg (spinand,GD_FEATURE_NR , + enable ?0 :GD_FEATURE_NR ); +if (ret) +return ret; + + priv->continuous_read = enable; + +return 0 ; +} +static int gd5fxgq4xa_ooblayout_ecc (struct mtd_info *mtd,int section,struct mtd_oob_region *region) {
@@ -542,7 +623,8 @@ static const struct spinand_info gigadevice_spinand_table[] = { &update_cache_variants), SPINAND_HAS_QE_BIT, SPINAND_ECCINFO(&gd5fxgqx_variant2_ooblayout, - gd5fxgq4uexxg_ecc_get_status)), + gd5fxgm9_ecc_get_status), + SPINAND_CONT_READ(gd5fxgm9_set_continuous_read)), SPINAND_INFO("GD5F1GM9RExxG", SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0x81, 0x01), NAND_MEMORG(1, 2048, 128, 64, 1024, 20, 1, 1, 1),
@@ -552,10 +634,31 @@ static const struct spinand_info gigadevice_spinand_table[] = { &update_cache_variants),SPINAND_HAS_QE_BIT ,SPINAND_ECCINFO (&gd5fxgqx_variant2_ooblayout, - gd5fxgq4uexxg_ecc_get_status)), + gd5fxgm9_ecc_get_status), + SPINAND_CONT_READ(gd5fxgm9_set_continuous_read)), }; +static int gd5fxgm9_spinand_init (struct spinand_device *spinand) +{ +struct gigadevice_priv *priv; + + priv =kzalloc (sizeof (*priv),GFP_KERNEL ); +if (!priv) +return -ENOMEM ; + + spinand->priv = priv; + +return 0 ; +} + +static void gd5fxgm9_spinand_cleanup (struct spinand_device *spinand) +{ +kfree (spinand->priv); +} +static const struct spinand_manufacturer_ops gigadevice_spinand_manuf_ops = { + .init = gd5fxgm9_spinand_init, + .cleanup = gd5fxgm9_spinand_cleanup, }; const struct spinand_manufacturer gigadevice_spinand_manufacturer = {
Signature · 1 line
2.43.0
Miquel Raynal
maintainerHello Teng,
Quoted 14 lines · Teng Wu
GD5F1GM9 supports a high-speed Continuous Read (CR) mode which allows the entire memory array to be read out by a single read command without re-issuing the 13h “Page Read to Cache” command for every page.
Datasheet summary
Feature register B0[3] (NR bit)
1 = Normal Read (default)
0 = Continuous Read (CR)ECC status register 7Ch
Normal read: bits [3:0] → ECC status of last page Continuous read: bits [7:4] → accumulated max bit-flips over the whole sequential read operation.
Quoted 5 lines
Datasheet:
https://github.com/WT-886/DATASHEET/blob/main/GD5F1GM9-v1.0.pdfSigned-off-by Teng Wu <[email protected]>---
I am happy to see more defines benefiting from this feature! Would you mind sharing the result of a speed test?flash_speed -c10 -d /dev/mtdx
Thanks!
Miquèl
Quoted 10 lines · Teng Wu
GD5F1GM9 supports a high-speed Continuous Read (CR) mode which allows the entire memory array to be read out by a single read command without re-issuing the 13h “Page Read to Cache” command for every page.
Datasheet summary
Feature register B0[3] (NR bit)
1 = Normal Read (default)
0 = Continuous Read (CR)[...]
Applied to nand/next, thanks!
[1/1] mtd: spinand: gigadevice: Add continuous read support
commit: 49b9254217048068ae1011619db60a6de5edf7e2
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