mtd: spinand: winbond: Add support for continuous reads on W35NxxJW

W35N{01,02,04}JW support being read continuously under certain
circumstances. A bit must be set in their configuration register, and
a specific read from cache operation, a bit shorter than usual because
it no longer requires the address cycles, must be used for the occasion.

Setting the "enable" bit is already supported by the core, aside from
the subtlety of making sure the HFREQ bit is also set in octal DTR mode
above 90MHz. However, handling two different read from cache templates
involves creating a list of read from cache variants adapted the
continuous reads, ie. without address cycles.

Unfortunately, these operations, despite being very close to their
original read from cache cousins, are often unsupported by smart SPI
controller drivers because reading from cache historically allowed
changing the offset at which the host would start by providing a 2-byte
column address. In order to prevent issues with this, it has been
decided to implement these variants with a single "ignored" address byte
(respectively two in the octal DTR case), further reducing the amount of
dummy cycles needed before the first bit of data.

Enabling continuous reads has a side effect: the ECC status register now
may also return the value b11, which means that more than 1
uncorrectable error happened during the read. This non standard
behaviour requires to re-implement, almost identically the "get ECC"
helper from the core, with just an extra case for this value (it is
prefixed "w25w35nxxjw" because all these chips have the same behaviour).

Speed gain is substantial, see below. The flash_speed -C benchmark has
been run on a TI AM62A7 LP SK with CPU power management disabled,
mounted with a W35N01JW chip.

1S-8S-8S:

     1 page read speed is 15058 KiB/s
     2 page read speed is 15058 KiB/s
     3 page read speed is 16800 KiB/s
     4 page read speed is 17066 KiB/s
     5 page read speed is 18461 KiB/s
     6 page read speed is 18461 KiB/s
     7 page read speed is 19384 KiB/s
     8 page read speed is 19692 KiB/s
     9 page read speed is 19384 KiB/s
     10 page read speed is 20000 KiB/s
     11 page read speed is 20000 KiB/s
     12 page read speed is 20000 KiB/s
     13 page read speed is 20800 KiB/s
     14 page read speed is 20363 KiB/s
     15 page read speed is 20000 KiB/s
     16 page read speed is 19692 KiB/s
     32 page read speed is 19692 KiB/s
     64 page read speed is 19692 KiB/s

8D-8D-8D:

     1 page read speed is 23272 KiB/s
     2 page read speed is 23272 KiB/s
     3 page read speed is 28000 KiB/s
     4 page read speed is 32000 KiB/s
     5 page read speed is 34285 KiB/s
     6 page read speed is 34285 KiB/s
     7 page read speed is 36000 KiB/s
     8 page read speed is 36571 KiB/s
     9 page read speed is 36000 KiB/s
     10 page read speed is 34285 KiB/s
     11 page read speed is 36666 KiB/s
     12 page read speed is 40000 KiB/s
     13 page read speed is 41600 KiB/s
     14 page read speed is 37333 KiB/s
     15 page read speed is 40000 KiB/s
     16 page read speed is 36571 KiB/s
     32 page read speed is 42666 KiB/s
     64 page read speed is 42666 KiB/s

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
Not all configurations have been tested/validated yet.

diff --git a/drivers/mtd/nand/spi/winbond.c b/drivers/mtd/nand/spi/winbond.c
index f538cf2228f8ce8fe477c665fb3f072eb5c789d5..82383c8839f0d91670fb582b7277105d887206f2 100644 (file)
--- a/drivers/mtd/nand/spi/winbond.c
+++ b/drivers/mtd/nand/spi/winbond.c
@@ -15,9 +15,11 @@
 
 #define SPINAND_MFR_WINBOND            0xEF
 
+#define WINBOND_CFG_HFREQ              BIT(0)
 #define WINBOND_CFG_BUF_READ           BIT(3)
 
 #define W25N04KV_STATUS_ECC_5_8_BITFLIPS       (3 << 4)
+#define W25W35NXXJW_STATUS_ECC_MULT_UNCOR      (3 << 4)
 
 #define W25N0XJW_SR4                   0xD0
 #define W25N0XJW_SR4_HS                        BIT(2)
@@ -29,6 +31,49 @@
 #define W35N01JW_VCR_IO_MODE_OCTAL_DDR         0xC7
 #define W35N01JW_VCR_DUMMY_CLOCK_REG   0x01
 
+/*
+ * Winbond chips ignore the address bytes during continuous reads, and
+ * because the dummy cycles are enough they indicate dropping the
+ * address cycles from the continuous read from cache variants. This is
+ * very poorly supported by SPI controller drivers which are "wired" to
+ * always at least provide the column. Keep using address cycles, but
+ * reduce the number of dummy cycles accordingly.
+ */
+#define WINBOND_CONT_READ_FROM_CACHE_FAST_1S_1S_1S_OP(ndummy, buf, len, freq) \
+       SPI_MEM_OP(SPI_MEM_OP_CMD(0x0b, 1),                             \
+                  SPI_MEM_OP_ADDR(1, 0, 1),                            \
+                  SPI_MEM_OP_DUMMY(ndummy - 1, 1),                     \
+                  SPI_MEM_OP_DATA_IN(len, buf, 1),                     \
+                  SPI_MEM_OP_MAX_FREQ(freq))
+
+#define WINBOND_CONT_READ_FROM_CACHE_1S_1S_8S_OP(ndummy, buf, len, freq) \
+       SPI_MEM_OP(SPI_MEM_OP_CMD(0x8b, 1),                             \
+                  SPI_MEM_OP_ADDR(1, 0, 1),                            \
+                  SPI_MEM_OP_DUMMY(ndummy - 1, 1),                     \
+                  SPI_MEM_OP_DATA_IN(len, buf, 8),                     \
+                  SPI_MEM_OP_MAX_FREQ(freq))
+
+#define WINBOND_CONT_READ_FROM_CACHE_1S_1D_8D_OP(ndummy, buf, len, freq) \
+       SPI_MEM_OP(SPI_MEM_OP_CMD(0x9d, 1),                             \
+                  SPI_MEM_DTR_OP_ADDR(1, 0, 1),                        \
+                  SPI_MEM_DTR_OP_DUMMY(ndummy - 1, 1),                 \
+                  SPI_MEM_DTR_OP_DATA_IN(len, buf, 8),                 \
+                  SPI_MEM_OP_MAX_FREQ(freq))
+
+#define WINBOND_CONT_READ_FROM_CACHE_1S_8S_8S_OP(ndummy, buf, len, freq) \
+       SPI_MEM_OP(SPI_MEM_OP_CMD(0xcb, 1),                             \
+                  SPI_MEM_OP_ADDR(1, 0, 8),                            \
+                  SPI_MEM_OP_DUMMY(ndummy - 1, 8),                     \
+                  SPI_MEM_OP_DATA_IN(len, buf, 8),                     \
+                  SPI_MEM_OP_MAX_FREQ(freq))
+
+#define WINBOND_CONT_READ_FROM_CACHE_8D_8D_8D_OP(ndummy, buf, len, freq) \
+       SPI_MEM_OP(SPI_MEM_DTR_OP_RPT_CMD(0x9d, 8),                     \
+                  SPI_MEM_DTR_OP_ADDR(2, 0, 8),                        \
+                  SPI_MEM_DTR_OP_DUMMY(ndummy - 2, 8),                 \
+                  SPI_MEM_DTR_OP_DATA_IN(len, buf, 8),                 \
+                  SPI_MEM_OP_MAX_FREQ(freq))
+
 /*
  * "X2" in the core is equivalent to "dual output" in the datasheets,
  * "X4" in the core is equivalent to "quad output" in the datasheets.
@@ -49,6 +94,19 @@ static SPINAND_OP_VARIANTS(read_cache_octal_variants,
                SPINAND_PAGE_READ_FROM_CACHE_FAST_1S_1S_1S_OP(0, 1, NULL, 0, 0),
                SPINAND_PAGE_READ_FROM_CACHE_1S_1S_1S_OP(0, 1, NULL, 0, 0));
 
+static SPINAND_OP_VARIANTS(cont_read_cache_octal_variants,
+               WINBOND_CONT_READ_FROM_CACHE_8D_8D_8D_OP(24, NULL, 0, 120 * HZ_PER_MHZ),
+               WINBOND_CONT_READ_FROM_CACHE_8D_8D_8D_OP(16, NULL, 0, 86 * HZ_PER_MHZ),
+               WINBOND_CONT_READ_FROM_CACHE_1S_1D_8D_OP(3, NULL, 0, 120 * HZ_PER_MHZ),
+               WINBOND_CONT_READ_FROM_CACHE_1S_1D_8D_OP(2, NULL, 0, 105 * HZ_PER_MHZ),
+               WINBOND_CONT_READ_FROM_CACHE_1S_8S_8S_OP(20, NULL, 0, 0),
+               WINBOND_CONT_READ_FROM_CACHE_1S_8S_8S_OP(16, NULL, 0, 162 * HZ_PER_MHZ),
+               WINBOND_CONT_READ_FROM_CACHE_1S_8S_8S_OP(12, NULL, 0, 124 * HZ_PER_MHZ),
+               WINBOND_CONT_READ_FROM_CACHE_1S_8S_8S_OP(8, NULL, 0, 86 * HZ_PER_MHZ),
+               WINBOND_CONT_READ_FROM_CACHE_1S_1S_8S_OP(2, NULL, 0, 0),
+               WINBOND_CONT_READ_FROM_CACHE_1S_1S_8S_OP(1, NULL, 0, 133 * HZ_PER_MHZ),
+               WINBOND_CONT_READ_FROM_CACHE_FAST_1S_1S_1S_OP(1, NULL, 0, 0));
+
 static SPINAND_OP_VARIANTS(write_cache_octal_variants,
                SPINAND_PROG_LOAD_8D_8D_8D_OP(true, 0, NULL, 0),
                SPINAND_PROG_LOAD_1S_8S_8S_OP(true, 0, NULL, 0),
@@ -326,6 +384,26 @@ static int w25n02kv_ecc_get_status(struct spinand_device *spinand,
        return -EINVAL;
 }
 
+static int w25w35nxxjw_ecc_get_status(struct spinand_device *spinand, u8 status)
+{
+       switch (status & STATUS_ECC_MASK) {
+       case STATUS_ECC_NO_BITFLIPS:
+               return 0;
+
+       case STATUS_ECC_HAS_BITFLIPS:
+               return 1;
+
+       case STATUS_ECC_UNCOR_ERROR:
+       case W25W35NXXJW_STATUS_ECC_MULT_UNCOR:
+               return -EBADMSG;
+
+       default:
+               break;
+       }
+
+       return -EINVAL;
+}
+
 static int w25n0xjw_hs_cfg(struct spinand_device *spinand,
                           enum spinand_bus_interface iface)
 {
@@ -451,6 +529,18 @@ static int w35n0xjw_vcr_cfg(struct spinand_device *spinand,
        return 0;
 }
 
+static int w35n0xjw_set_cont_read(struct spinand_device *spinand, bool enable)
+{
+       const struct spi_mem_op *cont_op = spinand->op_templates->cont_read_cache;
+       u8 mask = enable ? 0 : WINBOND_CFG_BUF_READ;
+
+       if (cont_op && enable && spinand_op_is_odtr(cont_op) &&
+           cont_op->max_freq >= 90 * HZ_PER_MHZ)
+               mask |= WINBOND_CFG_HFREQ;
+
+       return spinand_upd_cfg(spinand, WINBOND_CFG_BUF_READ | WINBOND_CFG_HFREQ, mask);
+}
+
 static const struct spinand_info winbond_spinand_table[] = {
        /* 512M-bit densities */
        SPINAND_INFO("W25N512GW", /* 1.8V */
@@ -504,13 +594,15 @@ static const struct spinand_info winbond_spinand_table[] = {
                     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xdc, 0x21),
                     NAND_MEMORG(1, 4096, 128, 64, 512, 10, 1, 1, 1),
                     NAND_ECCREQ(1, 512),
-                    SPINAND_INFO_OP_VARIANTS(&read_cache_octal_variants,
-                                             &write_cache_octal_variants,
-                                             &update_cache_octal_variants),
+                    SPINAND_INFO_OP_VARIANTS_WITH_CONT(&read_cache_octal_variants,
+                                                       &write_cache_octal_variants,
+                                                       &update_cache_octal_variants,
+                                                       &cont_read_cache_octal_variants),
                     0,
                     SPINAND_INFO_VENDOR_OPS(&winbond_w35_ops),
-                    SPINAND_ECCINFO(&w35n01jw_ooblayout, NULL),
-                    SPINAND_CONFIGURE_CHIP(w35n0xjw_vcr_cfg)),
+                    SPINAND_ECCINFO(&w35n01jw_ooblayout, w25w35nxxjw_ecc_get_status),
+                    SPINAND_CONFIGURE_CHIP(w35n0xjw_vcr_cfg),
+                    SPINAND_CONT_READ(w35n0xjw_set_cont_read)),
        /* 2G-bit densities */
        SPINAND_INFO("W25M02GV", /* 2x1G-bit 3.3V */
                     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xab, 0x21),
@@ -555,13 +647,15 @@ static const struct spinand_info winbond_spinand_table[] = {
                     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xdf, 0x22),
                     NAND_MEMORG(1, 4096, 128, 64, 512, 10, 1, 2, 1),
                     NAND_ECCREQ(1, 512),
-                    SPINAND_INFO_OP_VARIANTS(&read_cache_octal_variants,
-                                             &write_cache_octal_variants,
-                                             &update_cache_octal_variants),
+                    SPINAND_INFO_OP_VARIANTS_WITH_CONT(&read_cache_octal_variants,
+                                                       &write_cache_octal_variants,
+                                                       &update_cache_octal_variants,
+                                                       &cont_read_cache_octal_variants),
                     SPINAND_ODTR_PACKED_PAGE_READ,
                     SPINAND_INFO_VENDOR_OPS(&winbond_w35_ops),
-                    SPINAND_ECCINFO(&w35n01jw_ooblayout, NULL),
-                    SPINAND_CONFIGURE_CHIP(w35n0xjw_vcr_cfg)),
+                    SPINAND_ECCINFO(&w35n01jw_ooblayout, w25w35nxxjw_ecc_get_status),
+                    SPINAND_CONFIGURE_CHIP(w35n0xjw_vcr_cfg),
+                    SPINAND_CONT_READ(w35n0xjw_set_cont_read)),
        /* 4G-bit densities */
        SPINAND_INFO("W25N04KV", /* 3.3V */
                     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xaa, 0x23),
@@ -585,13 +679,15 @@ static const struct spinand_info winbond_spinand_table[] = {
                     SPINAND_ID(SPINAND_READID_METHOD_OPCODE_DUMMY, 0xdf, 0x23),
                     NAND_MEMORG(1, 4096, 128, 64, 512, 10, 1, 4, 1),
                     NAND_ECCREQ(1, 512),
-                    SPINAND_INFO_OP_VARIANTS(&read_cache_octal_variants,
-                                             &write_cache_octal_variants,
-                                             &update_cache_octal_variants),
+                    SPINAND_INFO_OP_VARIANTS_WITH_CONT(&read_cache_octal_variants,
+                                                       &write_cache_octal_variants,
+                                                       &update_cache_octal_variants,
+                                                       &cont_read_cache_octal_variants),
                     SPINAND_ODTR_PACKED_PAGE_READ,
                     SPINAND_INFO_VENDOR_OPS(&winbond_w35_ops),
-                    SPINAND_ECCINFO(&w35n01jw_ooblayout, NULL),
-                    SPINAND_CONFIGURE_CHIP(w35n0xjw_vcr_cfg)),
+                    SPINAND_ECCINFO(&w35n01jw_ooblayout, w25w35nxxjw_ecc_get_status),
+                    SPINAND_CONFIGURE_CHIP(w35n0xjw_vcr_cfg),
+                    SPINAND_CONT_READ(w35n0xjw_set_cont_read)),
 };
 
 static int winbond_spinand_init(struct spinand_device *spinand)