@@ -207,6 +207,14 @@ config MTD_CFI_STAA
sets which a CFI-compliant chip may claim to implement. This code
provides support for one of those command sets.
+config MTD_PSD4256G
+ tristate "ST PSD4256G compatible flash chip support"
+ help
+ Driver for flashes compatible with ST PSD4256G. This is a non-CFI
+ and non-JEDEC compliant flash and you really shouldn't buy this.
+
+ If unsure, say N here.
+
config MTD_CFI_UTIL
tristate
@@ -13,3 +13,4 @@ obj-$(CONFIG_MTD_JEDECPROBE) += jedec_probe.o
obj-$(CONFIG_MTD_RAM) += map_ram.o
obj-$(CONFIG_MTD_ROM) += map_rom.o
obj-$(CONFIG_MTD_ABSENT) += map_absent.o
+obj-$(CONFIG_MTD_PSD4256G) += stm_flash.o
new file mode 100644
@@ -0,0 +1,747 @@
+/*
+ * Driver for ST PSD4256G flash
+ *
+ * Copyright 2004-2009 Analog Devices Inc.
+ * Licensed under the GPL-2 or later.
+ */
+
+#define pr_fmt(fmt) "stm_flash: " fmt
+
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mtd/map.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/flashchip.h>
+#include <linux/sched.h>
+#include <linux/types.h>
+
+/* Addresses */
+#define ADDR_MANUFACTURER 0x0000
+#define ADDR_DEVICE_ID 0x0002
+#define ADDR_UNLOCK_1 0x0AAA
+#define ADDR_UNLOCK_2 0x0554
+
+/* Commands */
+#define CMD_UNLOCK_DATA_1 0x00aa
+#define CMD_UNLOCK_DATA_2 0x0055
+#define CMD_MANUFACTURER_UNLOCK_DATA 0x0090
+#define CMD_PROGRAM_UNLOCK_DATA 0x00A0
+#define CMD_RESET_DATA 0x00F0
+#define CMD_SECTOR_ERASE_UNLOCK_DATA 0x0080
+#define CMD_SECTOR_ERASE_UNLOCK_DATA_2 0x0030
+
+#define D6_MASK 0x40
+
+struct stm_flash_private {
+ unsigned long chipshift;
+ struct flchip chip;
+};
+
+static void send_unlock(struct map_info *map, unsigned long base)
+{
+ map_word test;
+ test.x[0] = 0x00aa;
+ map_write(map, test, base + ADDR_UNLOCK_1);
+ test.x[0] = 0x0055;
+ map_write(map, test, base + ADDR_UNLOCK_2);
+}
+
+static void send_cmd(struct map_info *map, unsigned long base,
+ unsigned long cmd)
+{
+ map_word test;
+ test.x[0] = cmd;
+ send_unlock(map, base);
+ map_write(map, test, base + ADDR_UNLOCK_1);
+}
+
+static void send_cmd_to_addr(struct map_info *map, unsigned long base,
+ unsigned long cmd, unsigned long addr)
+{
+ map_word test;
+ test.x[0] = cmd;
+ send_unlock(map, base);
+ map_write(map, test, addr);
+}
+
+static void stm_flash_sync(struct mtd_info *mtd)
+{
+ struct map_info *map = mtd->priv;
+ struct stm_flash_private *private = map->fldrv_priv;
+ struct flchip *chip;
+ DECLARE_WAITQUEUE(wait, current);
+
+ chip = &private->chip;
+
+ retry:
+ spin_lock_bh(chip->mutex);
+
+ switch (chip->state) {
+ case FL_READY:
+ case FL_STATUS:
+ case FL_CFI_QUERY:
+ case FL_JEDEC_QUERY:
+ chip->oldstate = chip->state;
+ chip->state = FL_SYNCING;
+ case FL_SYNCING:
+ spin_unlock_bh(chip->mutex);
+ break;
+ default:
+ /* Not an idle state */
+ add_wait_queue(&chip->wq, &wait);
+
+ spin_unlock_bh(chip->mutex);
+ schedule();
+ remove_wait_queue(&chip->wq, &wait);
+
+ goto retry;
+ }
+
+ chip = &private->chip;
+
+ spin_lock_bh(chip->mutex);
+
+ if (chip->state == FL_SYNCING) {
+ chip->state = chip->oldstate;
+ wake_up(&chip->wq);
+ }
+ spin_unlock_bh(chip->mutex);
+}
+
+static int read_one_chip(struct map_info *map, struct flchip *chip,
+ loff_t addr, size_t len, unsigned char *buf)
+{
+ DECLARE_WAITQUEUE(wait, current);
+ unsigned long timeo = jiffies + HZ;
+
+ retry:
+ spin_lock_bh(chip->mutex);
+
+ if (chip->state != FL_READY) {
+ pr_info("waiting for chip to read, state = %d\n",
+ chip->state);
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ add_wait_queue(&chip->wq, &wait);
+
+ spin_unlock_bh(chip->mutex);
+
+ schedule();
+ remove_wait_queue(&chip->wq, &wait);
+
+ if (signal_pending(current))
+ return -EINTR;
+
+ timeo = jiffies + HZ;
+
+ goto retry;
+ }
+
+ addr += chip->start;
+
+ chip->state = FL_READY;
+
+ map_copy_from(map, buf, addr, len);
+
+ wake_up(&chip->wq);
+ spin_unlock_bh(chip->mutex);
+
+ return 0;
+}
+
+static int stm_flash_read(struct mtd_info *mtd, loff_t from, size_t len,
+ size_t *retlen, unsigned char *buf)
+{
+ struct map_info *map = mtd->priv;
+ struct stm_flash_private *private = map->fldrv_priv;
+ unsigned long offset;
+ int chipnum, ret = 0;
+
+ if ((from + len) > mtd->size) {
+ pr_warning("read request past end of device (0x%lx)\n",
+ (unsigned long)from + len);
+ return -EINVAL;
+ }
+
+ /* Offset within the first chip that the first read should start. */
+ chipnum = (from >> private->chipshift);
+ offset = from - (chipnum << private->chipshift);
+
+ *retlen = 0;
+
+ while (len) {
+ unsigned long this_len;
+
+ if ((len + offset - 1) >> private->chipshift)
+ this_len = (1 << private->chipshift) - offset;
+ else
+ this_len = len;
+
+ ret = read_one_chip(map, &private->chip, offset,
+ this_len, buf);
+
+ if (ret)
+ break;
+
+ *retlen += this_len;
+ len -= this_len;
+ buf += this_len;
+
+ offset = 0;
+ chipnum++;
+ }
+
+ return ret;
+}
+
+static int flash_is_busy(struct map_info *map, unsigned long addr)
+{
+ unsigned short toggled;
+ map_word read11, read21;
+
+ read11 = map_read(map, addr);
+ read21 = map_read(map, addr);
+
+ toggled = (unsigned short)read11.x[0] ^ (unsigned short)read21.x[0];
+
+ toggled &= (((unsigned short)1) << 6);
+ return toggled;
+}
+
+static int write_one_word(struct map_info *map, struct flchip *chip,
+ unsigned long addr, unsigned long datum)
+{
+ unsigned long timeo = jiffies + HZ;
+ DECLARE_WAITQUEUE(wait, current);
+ int ret = 0;
+ int times_left;
+ map_word test;
+
+ retry:
+ spin_lock_bh(chip->mutex);
+
+ if (chip->state != FL_READY) {
+ pr_info("%s: waiting for chip to write, state = %d\n",
+ map->name, chip->state);
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ add_wait_queue(&chip->wq, &wait);
+
+ spin_unlock_bh(chip->mutex);
+
+ schedule();
+ remove_wait_queue(&chip->wq, &wait);
+ pr_info("%s: woke up to write\n", map->name);
+ if (signal_pending(current))
+ return -EINTR;
+
+ timeo = jiffies + HZ;
+
+ goto retry;
+ }
+
+ chip->state = FL_WRITING;
+
+ addr += chip->start;
+
+ send_cmd(map, chip->start, CMD_PROGRAM_UNLOCK_DATA);
+
+ test.x[0] = datum;
+ map_write(map, test, addr);
+
+ times_left = 50000;
+ while (times_left-- && flash_is_busy(map, addr)) {
+ if (need_resched()) {
+ spin_unlock_bh(chip->mutex);
+ schedule();
+ spin_lock_bh(chip->mutex);
+ }
+ }
+
+ if (!times_left) {
+ pr_warning("write to 0x%lx timed out!\n", addr);
+ ret = -EIO;
+ } else {
+ unsigned long verify;
+ map_word test;
+
+ test = map_read(map, addr);
+ verify = test.x[0];
+ if (verify != datum) {
+ pr_warning("write to 0x%lx failed "
+ "datum = %lx, verify = %lx\n",
+ addr, datum, verify);
+ ret = -EIO;
+ }
+ }
+ chip->state = FL_READY;
+ wake_up(&chip->wq);
+ spin_unlock_bh(chip->mutex);
+
+ return ret;
+}
+
+static int stm_flash_write(struct mtd_info *mtd, loff_t to, size_t len,
+ size_t *retlen, const unsigned char *buf)
+{
+ struct map_info *map = mtd->priv;
+ struct stm_flash_private *private = map->fldrv_priv;
+ int ret = 0;
+ int chipnum;
+ unsigned long offset;
+
+ *retlen = 0;
+ if (!len)
+ return 0;
+
+ chipnum = to >> private->chipshift;
+ offset = to - (chipnum << private->chipshift);
+
+ /* If it's not bus-aligned, do the first byte write. */
+ if (offset & (map->bankwidth - 1)) {
+ unsigned long bus_offset = offset & ~(map->bankwidth - 1);
+ int i = offset - bus_offset;
+ int n = 0;
+ unsigned char tmp_buf[4];
+ unsigned long datum;
+
+ map_copy_from(map, tmp_buf,
+ bus_offset + private->chip.start,
+ map->bankwidth);
+ while (len && i < map->bankwidth)
+ tmp_buf[i++] = buf[n++], len--;
+
+ if (map->bankwidth == 2)
+ datum = *(__u16 *)tmp_buf;
+ else if (map->bankwidth == 4)
+ datum = *(__u32 *)tmp_buf;
+ else
+ return -EINVAL;
+
+ ret = write_one_word(map, &private->chip, bus_offset,
+ datum);
+
+ if (ret)
+ return ret;
+
+ offset += n;
+ buf += n;
+ (*retlen) += n;
+
+ if (offset >> private->chipshift)
+ return 0;
+ }
+
+ /* We are now aligned, write as much as possible. */
+ while (len >= map->bankwidth) {
+ unsigned long datum;
+
+ if (map->bankwidth == 1)
+ datum = *(unsigned char *)buf;
+ else if (map->bankwidth == 2)
+ datum = *(unsigned short *)buf;
+ else if (map->bankwidth == 4)
+ datum = *(unsigned long *)buf;
+ else
+ return -EINVAL;
+
+ ret = write_one_word(map, &private->chip, offset,
+ datum);
+
+ if (ret)
+ return ret;
+
+ offset += map->bankwidth;
+ buf += map->bankwidth;
+ (*retlen) += map->bankwidth;
+ len -= map->bankwidth;
+
+ if (offset >> private->chipshift)
+ return 0;
+ }
+
+ if (len & (map->bankwidth - 1)) {
+ int i = 0, n = 0;
+ unsigned char tmp_buf[2];
+ unsigned long datum;
+
+ map_copy_from(map, tmp_buf,
+ offset + private->chip.start,
+ map->bankwidth);
+
+ while (len--)
+ tmp_buf[i++] = buf[n++];
+
+ if (map->bankwidth == 2)
+ datum = *(unsigned short *)tmp_buf;
+ else if (map->bankwidth == 4)
+ datum = *(unsigned long *)tmp_buf;
+ else
+ return -EINVAL;
+
+ ret = write_one_word(map, &private->chip, offset,
+ datum);
+
+ if (ret)
+ return ret;
+
+ (*retlen) += n;
+ }
+
+ return 0;
+}
+
+static int erase_one_block(struct map_info *map, struct flchip *chip,
+ unsigned long addr, unsigned long size)
+{
+ unsigned long timeo = jiffies + HZ;
+ DECLARE_WAITQUEUE(wait, current);
+
+ retry:
+ spin_lock_bh(chip->mutex);
+
+ if (chip->state != FL_READY) {
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ add_wait_queue(&chip->wq, &wait);
+
+ spin_unlock_bh(chip->mutex);
+
+ schedule();
+ remove_wait_queue(&chip->wq, &wait);
+
+ if (signal_pending(current))
+ return -EINTR;
+
+ timeo = jiffies + HZ;
+
+ goto retry;
+ }
+
+ chip->state = FL_ERASING;
+
+ addr += chip->start;
+
+ send_cmd(map, chip->start, CMD_SECTOR_ERASE_UNLOCK_DATA);
+ send_cmd_to_addr(map, chip->start, CMD_SECTOR_ERASE_UNLOCK_DATA_2, addr);
+
+ timeo = jiffies + (HZ * 20);
+
+ spin_unlock_bh(chip->mutex);
+ schedule_timeout(HZ);
+ spin_lock_bh(chip->mutex);
+
+ while (flash_is_busy(map, chip->start)) {
+ if (chip->state != FL_ERASING) {
+ /* Someone's suspended the erase. Sleep. */
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ add_wait_queue(&chip->wq, &wait);
+
+ spin_unlock_bh(chip->mutex);
+ pr_info("erase suspended, sleeping\n");
+ schedule();
+ remove_wait_queue(&chip->wq, &wait);
+
+ if (signal_pending(current))
+ return -EINTR;
+
+ timeo = jiffies + (HZ*2);
+ spin_lock_bh(chip->mutex);
+ continue;
+ }
+
+ /* OK Still waiting */
+ if (time_after(jiffies, timeo)) {
+ chip->state = FL_READY;
+ spin_unlock_bh(chip->mutex);
+ pr_warning("waiting for erase to complete timed out\n");
+ return -EIO;
+ }
+
+ /* Latency issues. Drop the lock, wait a while, and retry. */
+ spin_unlock_bh(chip->mutex);
+
+ if (need_resched())
+ schedule();
+ else
+ udelay(1);
+
+ spin_lock_bh(chip->mutex);
+ }
+
+ {
+ /* Verify every single word */
+ int address;
+ int error = 0;
+ int verify;
+ map_word test;
+
+ for (address = addr; address < (addr + size); address += 2) {
+ test = map_read(map, address);
+ verify = test.x[0];
+ if (verify != 0xFFFF) {
+ error = 1;
+ break;
+ }
+ }
+
+ if (error) {
+ chip->state = FL_READY;
+ spin_unlock_bh(chip->mutex);
+ pr_warning("verify error at 0x%x, size %ld\n",
+ address, size);
+ return -EIO;
+ }
+ }
+
+ chip->state = FL_READY;
+ wake_up(&chip->wq);
+ spin_unlock_bh(chip->mutex);
+
+ return 0;
+}
+
+static int stm_flash_erase(struct mtd_info *mtd, struct erase_info *instr)
+{
+ struct map_info *map = mtd->priv;
+ struct stm_flash_private *private = map->fldrv_priv;
+ unsigned long addr, len, shift;
+ int chipnum;
+ int ret = 0;
+ int i, first;
+ struct mtd_erase_region_info *regions = mtd->eraseregions;
+
+ if (instr->addr > mtd->size)
+ return -EINVAL;
+
+ if ((instr->len + instr->addr) > mtd->size)
+ return -EINVAL;
+
+ /*
+ * Check that both start and end of the requested erase are aligned
+ * with the erasesize at the appropriate addresses.
+ */
+ i = 0;
+
+ /*
+ * Skip all erase regions which are ended before the start of the
+ * requested erase. Actually, to save on the calculations, we skip
+ * to the first erase region which starts after the start of the
+ * requested erase, and then go back one.
+ */
+ while ((i < mtd->numeraseregions) &&
+ (instr->addr >= regions[i].offset))
+ ++i;
+ --i;
+
+ /*
+ * OK. Now i is pointing at the erase region in which this erase
+ * request starts. Check the start of the requested erase range
+ * is aligned with the erase size which is in effect here.
+ */
+ if (instr->addr & (regions[i].erasesize - 1))
+ return -EINVAL;
+
+ /*
+ * Remember the erase region we start on.
+ */
+ first = i;
+
+ /*
+ * Next, theck that the end of the requested erase is aligned with
+ * the erase region at that address.
+ */
+ while ((i < mtd->numeraseregions) &&
+ ((instr->addr + instr->len) >= regions[i].offset))
+ ++i;
+ --i;
+
+ if ((instr->addr + instr->len) & (regions[i].erasesize - 1))
+ return -EINVAL;
+
+ chipnum = instr->addr >> private->chipshift;
+ addr = instr->addr - (chipnum << private->chipshift);
+ len = instr->len;
+
+ i = first;
+ shift = (1 << private->chipshift) - 1;
+ while (len) {
+ ret = erase_one_block(map, &private->chip, addr,
+ regions[i].erasesize);
+
+ if (ret)
+ return ret;
+
+ addr += regions[i].erasesize;
+ len -= regions[i].erasesize;
+
+ if ((addr & shift) ==
+ ((regions[i].offset +
+ (regions[i].erasesize * regions[i].numblocks))
+ & shift))
+ ++i;
+
+ if (addr & ~shift)
+ break;
+ }
+
+ instr->state = MTD_ERASE_DONE;
+ if (instr->callback)
+ instr->callback(instr);
+
+ return 0;
+}
+
+static void stm_flash_destroy(struct mtd_info *mtd)
+{
+ struct map_info *map = mtd->priv;
+ struct stm_flash_private *private = map->fldrv_priv;
+ kfree(private);
+}
+
+struct stm_flash_info {
+ u8 mfr_id, dev_id;
+ char *name;
+ unsigned long size;
+ int numeraseregions;
+ struct mtd_erase_region_info regions[1];
+};
+
+const struct stm_flash_info stm_data[] = {
+ {
+ .mfr_id = 0x20,
+ .dev_id = 0xe9,
+ .name = "STM PSD4256G6V",
+ .size = 0x100000,
+ .numeraseregions = 1,
+ .regions = {
+ {
+ .offset = 0x000000,
+ .erasesize = 0x10000,
+ .numblocks = 16
+ },
+ },
+ },
+};
+
+static int stm_probe_new_chip(struct mtd_info *mtd)
+{
+ const struct stm_flash_info *std;
+ struct map_info *map = mtd->priv;
+ int i;
+ u8 mfr_id, dev_id;
+ map_word mfr_word, dev_word;
+
+ /* Enter autoselect mode */
+ send_cmd(map, 0, CMD_RESET_DATA);
+ send_cmd(map, 0, CMD_MANUFACTURER_UNLOCK_DATA);
+
+ mfr_word = map_read(map, ADDR_MANUFACTURER);
+ dev_word = map_read(map, ADDR_DEVICE_ID);
+ mfr_id = mfr_word.x[0] & 0xFF;
+ dev_id = dev_word.x[0] & 0xFF;
+
+ /* Exit autoselect mode */
+ send_cmd(map, 0, CMD_RESET_DATA);
+
+ for (i = 0; i < ARRAY_SIZE(stm_data); ++i) {
+ std = &stm_data[i];
+ if (mfr_id == std->mfr_id && dev_id == std->dev_id)
+ break;
+ }
+ if (i == ARRAY_SIZE(stm_data)) {
+ pr_warning("unknown mfr/dev id: 0x%02x 0x%02x\n", mfr_id, dev_id);
+ return 1;
+ }
+
+ mtd->size += std->size;
+ mtd->numeraseregions += std->numeraseregions;
+
+ mtd->eraseregions = kmalloc(sizeof(*mtd->eraseregions) *
+ mtd->numeraseregions, GFP_KERNEL);
+ if (!mtd->eraseregions) {
+ pr_warning("failed to allocate memory for MTD erase region\n");
+ return 1;
+ }
+
+ for (i = 0; i < std->numeraseregions; ++i) {
+ mtd->eraseregions[i].offset = std->regions[i].offset;
+ mtd->eraseregions[i].erasesize = std->regions[i].erasesize;
+ mtd->eraseregions[i].numblocks = std->regions[i].numblocks;
+ if (mtd->erasesize < mtd->eraseregions[i].erasesize)
+ mtd->erasesize = mtd->eraseregions[i].erasesize;
+ }
+
+ pr_info("found %s device at %p in %d-bit bank\n",
+ std->name, map->virt, map->bankwidth * 8);
+
+ return 0;
+}
+
+static struct mtd_info *stm_flash_probe(struct map_info *map);
+
+static struct mtd_chip_driver stm_flash_chipdrv = {
+ .probe = stm_flash_probe,
+ .destroy = stm_flash_destroy,
+ .name = "stm_flash",
+ .module = THIS_MODULE
+};
+
+static struct mtd_info *stm_flash_probe(struct map_info *map)
+{
+ struct mtd_info *mtd;
+ struct flchip *chip;
+ struct stm_flash_private *private;
+ unsigned long size;
+
+ mtd = kzalloc(sizeof(*mtd) + sizeof(*private), GFP_KERNEL);
+ if (!mtd) {
+ pr_warning("kmalloc failed for info structure\n");
+ return NULL;
+ }
+ mtd->priv = map;
+ map->fldrv_priv = private = (void *)&mtd[1];
+ map->fldrv = &stm_flash_chipdrv;
+
+ if (stm_probe_new_chip(mtd)) {
+ kfree(mtd);
+ return NULL;
+ }
+
+ chip = &private->chip;
+ init_waitqueue_head(&chip->wq);
+ spin_lock_init(&chip->_spinlock);
+ chip->start = 0;
+ chip->state = FL_READY;
+ chip->mutex = &chip->_spinlock;
+ for (size = mtd->size; size > 1; size >>= 1)
+ ++private->chipshift;
+
+ mtd->type = MTD_NORFLASH;
+ mtd->flags = MTD_CAP_NORFLASH;
+ mtd->name = map->name;
+ mtd->erase = stm_flash_erase;
+ mtd->read = stm_flash_read;
+ mtd->write = stm_flash_write;
+ mtd->sync = stm_flash_sync;
+ mtd->writesize = 1;
+
+ return mtd;
+}
+
+static int __init stm_flash_init(void)
+{
+ register_mtd_chip_driver(&stm_flash_chipdrv);
+ return 0;
+}
+module_init(stm_flash_init);
+
+static void __exit stm_flash_exit(void)
+{
+ unregister_mtd_chip_driver(&stm_flash_chipdrv);
+}
+module_exit(stm_flash_exit);
+
+MODULE_DESCRIPTION("MTD chip driver for ST PSD4256G flashes");
+MODULE_LICENSE("GPL");
The ST PSD4256G Flash In-System Programming (ISP) part is a multifunction device. Since all of the different pieces are memory mapped, this is a driver for the NOR flash component. All the other pieces can use generic memory based drivers. Unfortunately, this is not a CFI compliant part in any way, so we need a dedicated chip driver for it. Signed-off-by: Mike Frysinger <vapier@gentoo.org> --- drivers/mtd/chips/Kconfig | 8 + drivers/mtd/chips/Makefile | 1 + drivers/mtd/chips/stm_flash.c | 747 +++++++++++++++++++++++++++++++++++++++++ 3 files changed, 756 insertions(+), 0 deletions(-) create mode 100644 drivers/mtd/chips/stm_flash.c