[PATCH v5 3/6] nand: spi: Add bad block support
Peter Pan
peterpandong at micron.com
Mon Apr 10 00:51:50 PDT 2017
Add isbad and markbad support for SPI NAND. And do not
erase bad blocks in spi_nand_erase. BBT is also enabled
in this patch.
Signed-off-by: Peter Pan <peterpandong at micron.com>
---
drivers/mtd/nand/spi/core.c | 329 +++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 322 insertions(+), 7 deletions(-)
diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c
index eb7af9d..52113fc 100644
--- a/drivers/mtd/nand/spi/core.c
+++ b/drivers/mtd/nand/spi/core.c
@@ -19,6 +19,10 @@
#include <linux/jiffies.h>
#include <linux/mtd/spinand.h>
#include <linux/slab.h>
+#include <linux/of.h>
+
+static int spinand_erase_skip_bbt(struct mtd_info *mtd,
+ struct erase_info *einfo);
/*
* spinand_exec_op - execute SPI NAND operation by controller ->exec_op() hook
@@ -878,11 +882,154 @@ static int spinand_write_oob(struct mtd_info *mtd, loff_t to,
}
/*
- * spinand_erase - [MTD Interface] erase block(s)
+ * spinand_block_bad - check if block at offset is bad by bad block marker
+ * @mtd: MTD device structure
+ * @offs: offset from device start
+ */
+static int spinand_block_bad(struct mtd_info *mtd, loff_t offs)
+{
+ struct nand_device *nand = mtd_to_nand(mtd);
+ struct mtd_oob_ops ops = {0};
+ u32 block_addr;
+ u8 bad[2] = {0, 0};
+ u8 ret = 0;
+ unsigned int max_bitflips;
+
+ block_addr = nand_offs_to_eraseblock(nand, offs);
+ ops.mode = MTD_OPS_PLACE_OOB;
+ ops.ooblen = 2;
+ ops.oobbuf = bad;
+ spinand_read_pages(mtd, nand_eraseblock_to_offs(nand, block_addr),
+ &ops, &max_bitflips);
+ if (bad[0] != 0xFF || bad[1] != 0xFF)
+ ret = 1;
+
+ return ret;
+}
+
+/*
+ * spinand_block_checkbad - check if a block is marked bad
+ * @mtd: MTD device structure
+ * @offs: offset from device start
+ * @allowbbt: 1, if allowe to access the bbt area
+ * Description:
+ * Check, if the block is bad. Either by reading the bad block table or
+ * reading bad block marker.
+ */
+static int spinand_block_checkbad(struct mtd_info *mtd, loff_t offs,
+ int allowbbt)
+{
+ struct nand_device *nand = mtd_to_nand(mtd);
+ int ret;
+
+ if (nand_bbt_is_initialized(nand))
+ ret = nand_isbad_bbt(nand, offs, allowbbt);
+ else
+ ret = spinand_block_bad(mtd, offs);
+
+ return ret;
+}
+
+/*
+ * spinand_block_isbad - [MTD Interface] check if block at offset is bad
+ * @mtd: MTD device structure
+ * @offs: offset from device start
+ */
+static int spinand_block_isbad(struct mtd_info *mtd, loff_t offs)
+{
+ struct spinand_device *chip = mtd_to_spinand(mtd);
+ int ret;
+
+ mutex_lock(&chip->lock);
+ ret = spinand_block_checkbad(mtd, offs, 0);
+ mutex_unlock(&chip->lock);
+
+ return ret;
+}
+
+/*
+ * spinand_block_markbad_lowlevel - mark a block bad
+ * @mtd: MTD device structure
+ * @offs: offset from device start
+ *
+ * This function performs the generic bad block marking steps (i.e., bad
+ * block table(s) and/or marker(s)).
+ *
+ * We try operations in the following order:
+ * (1) erase the affected block, to allow OOB marker to be written cleanly
+ * (2) write bad block marker to OOB area of affected block (unless flag
+ * NAND_BBT_NO_OOB_BBM is present)
+ * (3) update the BBT
+ */
+static int spinand_block_markbad_lowlevel(struct mtd_info *mtd, loff_t offs)
+{
+ struct nand_device *nand = mtd_to_nand(mtd);
+ struct mtd_oob_ops ops = {0};
+ struct erase_info einfo = {0};
+ u32 block_addr;
+ u8 buf[2] = {0, 0};
+ int res, ret = 0;
+
+ if (!nand_bbt_is_initialized(nand) ||
+ !(nand->bbt.options & NAND_BBT_NO_OOB_BBM)) {
+ /*erase bad block before mark bad block*/
+ einfo.mtd = mtd;
+ einfo.addr = offs;
+ einfo.len = nand_eraseblock_size(nand);
+ spinand_erase_skip_bbt(mtd, &einfo);
+
+ block_addr = nand_offs_to_eraseblock(nand, offs);
+ ops.mode = MTD_OPS_PLACE_OOB;
+ ops.ooblen = 2;
+ ops.oobbuf = buf;
+ ret = spinand_do_write_ops(mtd,
+ nand_eraseblock_to_offs(nand,
+ block_addr),
+ &ops);
+ }
+
+ /* Mark block bad in BBT */
+ if (nand_bbt_is_initialized(nand)) {
+ res = nand_markbad_bbt(nand, offs);
+ if (!ret)
+ ret = res;
+ }
+
+ if (!ret)
+ mtd->ecc_stats.badblocks++;
+
+ return ret;
+}
+
+/*
+ * spinand_block_markbad - [MTD Interface] mark block at the given offset
+ * as bad
+ * @mtd: MTD device structure
+ * @offs: offset relative to mtd start
+ */
+static int spinand_block_markbad(struct mtd_info *mtd, loff_t offs)
+{
+ int ret;
+
+ ret = spinand_block_isbad(mtd, offs);
+ if (ret) {
+ /* If it was bad already, return success and do nothing */
+ if (ret > 0)
+ return 0;
+ return ret;
+ }
+
+ return spinand_block_markbad_lowlevel(mtd, offs);
+}
+
+/*
+ * spinand_erase - erase block(s)
* @mtd: MTD device structure
* @einfo: erase instruction
+ * @allowbbt: allow to access bbt
*/
-static int spinand_erase(struct mtd_info *mtd, struct erase_info *einfo)
+static int spinand_erase(struct mtd_info *mtd, struct erase_info *einfo,
+ int allowbbt)
{
struct spinand_device *chip = mtd_to_spinand(mtd);
struct nand_device *nand = mtd_to_nand(mtd);
@@ -901,6 +1048,14 @@ static int spinand_erase(struct mtd_info *mtd, struct erase_info *einfo)
einfo->state = MTD_ERASING;
while (len) {
+ /* Check if we have a bad block, we do not erase bad blocks! */
+ if (spinand_block_checkbad(mtd, offs, allowbbt)) {
+ dev_warn(chip->dev,
+ "attempt to erase a bad block at 0x%012llx\n",
+ offs);
+ einfo->state = MTD_ERASE_FAILED;
+ goto erase_exit;
+ }
spinand_write_enable(chip);
spinand_erase_block(chip, nand_offs_to_page(nand, offs));
ret = spinand_wait(chip, &status);
@@ -937,6 +1092,33 @@ static int spinand_erase(struct mtd_info *mtd, struct erase_info *einfo)
}
/*
+ * spinand_erase_skip_bbt - [MTD Interface] erase block(s) except BBT
+ * @mtd: MTD device structure
+ * @einfo: erase instruction
+ */
+static int spinand_erase_skip_bbt(struct mtd_info *mtd,
+ struct erase_info *einfo)
+{
+ return spinand_erase(mtd, einfo, 0);
+}
+
+/*
+ * spinand_block_isreserved - [MTD Interface] check if a block is
+ * marked reserved.
+ * @mtd: MTD device structure
+ * @offs: offset from device start
+ */
+static int spinand_block_isreserved(struct mtd_info *mtd, loff_t offs)
+{
+ struct nand_device *nand = mtd_to_nand(mtd);
+
+ if (!nand_bbt_is_initialized(nand))
+ return 0;
+ /* Return info from the table */
+ return nand_isreserved_bbt(nand, offs);
+}
+
+/*
* spinand_set_rd_wr_op - choose the best read write command
* @chip: SPI NAND device structure
* Description:
@@ -970,6 +1152,106 @@ static void spinand_set_rd_wr_op(struct spinand_device *chip)
chip->write_cache_op = SPINAND_CMD_PROG_LOAD;
}
+/*
+ * spinand_erase_bbt - erase block(s) including BBT
+ * @nand: nand device structure
+ * @einfo: erase instruction
+ */
+static int spinand_erase_bbt(struct nand_device *nand,
+ struct erase_info *einfo)
+{
+ return spinand_erase(nand_to_mtd(nand), einfo, 1);
+}
+
+/*
+ * spinand_erase_bbt - write bad block marker to certain block
+ * @nand: nand device structure
+ * @block: block to mark bad
+ */
+static int spinand_markbad(struct nand_device *nand, int block)
+{
+ struct mtd_oob_ops ops = {0};
+ u8 buf[2] = {0, 0};
+
+ ops.mode = MTD_OPS_PLACE_OOB;
+ ops.ooboffs = 0;
+ ops.ooblen = 2;
+ ops.oobbuf = buf;
+
+ return spinand_do_write_ops(nand_to_mtd(nand),
+ nand_eraseblock_to_offs(nand, block),
+ &ops);
+}
+
+static const struct nand_ops spinand_ops = {
+ .erase = spinand_erase_bbt,
+ .markbad = spinand_markbad,
+};
+
+/*
+ * Define some generic bad/good block scan pattern which are used
+ * while scanning a device for factory marked good/bad blocks.
+ */
+static u8 scan_ff_pattern[] = { 0xff, 0xff };
+
+#define BADBLOCK_SCAN_MASK (~NAND_BBT_NO_OOB)
+
+/*
+ * spinand_create_badblock_pattern - creates a BBT descriptor structure
+ * @chip: SPI NAND device structure
+ *
+ * This function allocates and initializes a nand_bbt_descr for BBM detection.
+ * The new descriptor is stored in nand->bbt.bbp. Thus, nand->bbt.bbp should
+ * be NULL when passed to this function.
+ */
+static int spinand_create_badblock_pattern(struct spinand_device *chip)
+{
+ struct nand_device *nand = &chip->base;
+ struct nand_bbt_descr *bd;
+
+ if (nand->bbt.bbp) {
+ dev_err(chip->dev,
+ "Bad block pattern already allocated; not replacing\n");
+ return -EINVAL;
+ }
+ bd = devm_kzalloc(chip->dev, sizeof(*bd), GFP_KERNEL);
+ if (!bd)
+ return -ENOMEM;
+ bd->options = nand->bbt.options & BADBLOCK_SCAN_MASK;
+ bd->offs = 0;
+ bd->len = 2;
+ bd->pattern = scan_ff_pattern;
+ bd->options |= NAND_BBT_DYNAMICSTRUCT;
+ nand->bbt.bbp = bd;
+
+ return 0;
+}
+
+/*
+ * spinand_scan_bbt - scan BBT in SPI NAND device
+ * @chip: SPI NAND device structure
+ */
+static int spinand_scan_bbt(struct spinand_device *chip)
+{
+ struct nand_device *nand = &chip->base;
+ int ret;
+
+ /*
+ * It's better to put BBT marker in-band, since some oob area
+ * is not ecc protected by internal(on-die) ECC
+ */
+ if (nand->bbt.options & NAND_BBT_USE_FLASH)
+ nand->bbt.options |= NAND_BBT_NO_OOB;
+ nand->bbt.td = NULL;
+ nand->bbt.md = NULL;
+
+ ret = spinand_create_badblock_pattern(chip);
+ if (ret)
+ return ret;
+
+ return nand_scan_bbt(nand);
+}
+
static const struct spinand_manufacturer *spinand_manufacturers[] = {};
/*
@@ -1024,14 +1306,30 @@ static void spinand_manufacturer_cleanup(struct spinand_device *chip)
}
/*
+ * TODO: move of_get_nand_on_flash_bbt() to generic NAND core
+ */
+static bool of_get_nand_on_flash_bbt(struct device_node *np)
+{
+ return of_property_read_bool(np, "nand-on-flash-bbt");
+}
+
+/*
* spinand_dt_init - Initialize SPI NAND by device tree node
* @chip: SPI NAND device structure
*
- * TODO: put ecc_mode, ecc_strength, ecc_step, bbt, etc in here
- * and move it in generic NAND core.
+ * TODO: put ecc_mode, ecc_strength, ecc_step, etc in here and move
+ * it in generic NAND core.
*/
static void spinand_dt_init(struct spinand_device *chip)
{
+ struct nand_device *nand = &chip->base;
+ struct device_node *dn = nand_get_of_node(nand);
+
+ if (!dn)
+ return;
+
+ if (of_get_nand_on_flash_bbt(dn))
+ nand->bbt.options |= NAND_BBT_USE_FLASH;
}
/*
@@ -1085,13 +1383,14 @@ static int spinand_init(struct spinand_device *chip)
GFP_KERNEL);
if (!chip->buf) {
ret = -ENOMEM;
- goto err;
+ goto err1;
}
chip->oobbuf = chip->buf + nand_page_size(nand);
spinand_manufacturer_init(chip);
+ nand->ops = &spinand_ops;
mtd->name = chip->name;
mtd->size = nand_size(nand);
mtd->erasesize = nand_eraseblock_size(nand);
@@ -1109,11 +1408,14 @@ static int spinand_init(struct spinand_device *chip)
if (ret < 0)
ret = 0;
mtd->oobavail = ret;
- mtd->_erase = spinand_erase;
+ mtd->_erase = spinand_erase_skip_bbt;
mtd->_read = spinand_read;
mtd->_write = spinand_write;
mtd->_read_oob = spinand_read_oob;
mtd->_write_oob = spinand_write_oob;
+ mtd->_block_isbad = spinand_block_isbad;
+ mtd->_block_markbad = spinand_block_markbad;
+ mtd->_block_isreserved = spinand_block_isreserved;
if (!mtd->bitflip_threshold)
mtd->bitflip_threshold = DIV_ROUND_UP(mtd->ecc_strength * 3,
@@ -1121,9 +1423,18 @@ static int spinand_init(struct spinand_device *chip)
/* After power up, all blocks are locked, so unlock it here. */
spinand_lock_block(chip, BL_ALL_UNLOCKED);
+ /* Build bad block table */
+ ret = spinand_scan_bbt(chip);
+ if (ret) {
+ dev_err(chip->dev, "Scan Bad Block Table failed.\n");
+ goto err2;
+ }
+
return nand_register(nand);
-err:
+err2:
+ devm_kfree(chip->dev, chip->buf);
+err1:
return ret;
}
@@ -1191,10 +1502,14 @@ int spinand_register(struct spinand_device *chip)
int spinand_unregister(struct spinand_device *chip)
{
struct nand_device *nand = &chip->base;
+ struct nand_bbt_descr *bd = nand->bbt.bbp;
nand_unregister(nand);
spinand_manufacturer_cleanup(chip);
devm_kfree(chip->dev, chip->buf);
+ kfree(nand->bbt.bbt);
+ if (bd->options & NAND_BBT_DYNAMICSTRUCT)
+ devm_kfree(chip->dev, bd);
return 0;
}
--
1.9.1
More information about the linux-mtd
mailing list