blob: c1c1ff876a7539d8a6c576c9e8d34a99d814415f [file] [log] [blame]
/*
* drivers/mtd/nand/docg4.c
*
* Copyright (C) 2013 Mike Dunn <mikedunn@newsguy.com>
*
* SPDX-License-Identifier: GPL-2.0+
*
* mtd nand driver for M-Systems DiskOnChip G4
*
* Tested on the Palm Treo 680. The G4 is also present on Toshiba Portege, Asus
* P526, some HTC smartphones (Wizard, Prophet, ...), O2 XDA Zinc, maybe others.
* Should work on these as well. Let me know!
*
* TODO:
*
* Mechanism for management of password-protected areas
*
* Hamming ecc when reading oob only
*
* According to the M-Sys documentation, this device is also available in a
* "dual-die" configuration having a 256MB capacity, but no mechanism for
* detecting this variant is documented. Currently this driver assumes 128MB
* capacity.
*
* Support for multiple cascaded devices ("floors"). Not sure which gadgets
* contain multiple G4s in a cascaded configuration, if any.
*/
#include <common.h>
#include <asm/arch/hardware.h>
#include <asm/io.h>
#include <asm/bitops.h>
#include <asm/errno.h>
#include <malloc.h>
#include <nand.h>
#include <linux/bch.h>
#include <linux/bitrev.h>
#include <linux/mtd/docg4.h>
/*
* The device has a nop register which M-Sys claims is for the purpose of
* inserting precise delays. But beware; at least some operations fail if the
* nop writes are replaced with a generic delay!
*/
static inline void write_nop(void __iomem *docptr)
{
writew(0, docptr + DOC_NOP);
}
static int poll_status(void __iomem *docptr)
{
/*
* Busy-wait for the FLASHREADY bit to be set in the FLASHCONTROL
* register. Operations known to take a long time (e.g., block erase)
* should sleep for a while before calling this.
*/
uint8_t flash_status;
/* hardware quirk requires reading twice initially */
flash_status = readb(docptr + DOC_FLASHCONTROL);
do {
flash_status = readb(docptr + DOC_FLASHCONTROL);
} while (!(flash_status & DOC_CTRL_FLASHREADY));
return 0;
}
static void write_addr(void __iomem *docptr, uint32_t docg4_addr)
{
/* write the four address bytes packed in docg4_addr to the device */
writeb(docg4_addr & 0xff, docptr + DOC_FLASHADDRESS);
docg4_addr >>= 8;
writeb(docg4_addr & 0xff, docptr + DOC_FLASHADDRESS);
docg4_addr >>= 8;
writeb(docg4_addr & 0xff, docptr + DOC_FLASHADDRESS);
docg4_addr >>= 8;
writeb(docg4_addr & 0xff, docptr + DOC_FLASHADDRESS);
}
/*
* This is a module parameter in the linux kernel version of this driver. It is
* hard-coded to 'off' for u-boot. This driver uses oob to mark bad blocks.
* This can be problematic when dealing with data not intended for the mtd/nand
* subsystem. For example, on boards that boot from the docg4 and use the IPL
* to load an spl + u-boot image, the blocks containing the image will be
* reported as "bad" because the oob of the first page of each block contains a
* magic number that the IPL looks for, which causes the badblock scan to
* erroneously add them to the bad block table. To erase such a block, use
* u-boot's 'nand scrub'. scrub is safe for the docg4. The device does have a
* factory bad block table, but it is read-only, and is used in conjunction with
* oob bad block markers that are written by mtd/nand when a block is deemed to
* be bad. To read data from "bad" blocks, use 'read.raw'. Unfortunately,
* read.raw does not use ecc, which would still work fine on such misidentified
* bad blocks. TODO: u-boot nand utilities need the ability to ignore bad
* blocks.
*/
static const int ignore_badblocks; /* remains false */
struct docg4_priv {
int status;
struct {
unsigned int command;
int column;
int page;
} last_command;
uint8_t oob_buf[16];
uint8_t ecc_buf[7];
int oob_page;
struct bch_control *bch;
};
/*
* Oob bytes 0 - 6 are available to the user.
* Byte 7 is hamming ecc for first 7 bytes. Bytes 8 - 14 are hw-generated ecc.
* Byte 15 (the last) is used by the driver as a "page written" flag.
*/
static struct nand_ecclayout docg4_oobinfo = {
.eccbytes = 9,
.eccpos = {7, 8, 9, 10, 11, 12, 13, 14, 15},
.oobavail = 7,
.oobfree = { {0, 7} }
};
static void reset(void __iomem *docptr)
{
/* full device reset */
writew(DOC_ASICMODE_RESET | DOC_ASICMODE_MDWREN, docptr + DOC_ASICMODE);
writew(~(DOC_ASICMODE_RESET | DOC_ASICMODE_MDWREN),
docptr + DOC_ASICMODECONFIRM);
write_nop(docptr);
writew(DOC_ASICMODE_NORMAL | DOC_ASICMODE_MDWREN,
docptr + DOC_ASICMODE);
writew(~(DOC_ASICMODE_NORMAL | DOC_ASICMODE_MDWREN),
docptr + DOC_ASICMODECONFIRM);
writew(DOC_ECCCONF1_ECC_ENABLE, docptr + DOC_ECCCONF1);
poll_status(docptr);
}
static void docg4_select_chip(struct mtd_info *mtd, int chip)
{
/*
* Select among multiple cascaded chips ("floors"). Multiple floors are
* not yet supported, so the only valid non-negative value is 0.
*/
void __iomem *docptr = CONFIG_SYS_NAND_BASE;
if (chip < 0)
return; /* deselected */
if (chip > 0)
printf("multiple floors currently unsupported\n");
writew(0, docptr + DOC_DEVICESELECT);
}
static void read_hw_ecc(void __iomem *docptr, uint8_t *ecc_buf)
{
/* read the 7 hw-generated ecc bytes */
int i;
for (i = 0; i < 7; i++) { /* hw quirk; read twice */
ecc_buf[i] = readb(docptr + DOC_BCH_SYNDROM(i));
ecc_buf[i] = readb(docptr + DOC_BCH_SYNDROM(i));
}
}
static int correct_data(struct mtd_info *mtd, uint8_t *buf, int page)
{
/*
* Called after a page read when hardware reports bitflips.
* Up to four bitflips can be corrected.
*/
struct nand_chip *nand = mtd->priv;
struct docg4_priv *doc = nand->priv;
void __iomem *docptr = CONFIG_SYS_NAND_BASE;
int i, numerrs;
unsigned int errpos[4];
const uint8_t blank_read_hwecc[8] = {
0xcf, 0x72, 0xfc, 0x1b, 0xa9, 0xc7, 0xb9, 0 };
read_hw_ecc(docptr, doc->ecc_buf); /* read 7 hw-generated ecc bytes */
/* check if read error is due to a blank page */
if (!memcmp(doc->ecc_buf, blank_read_hwecc, 7))
return 0; /* yes */
/* skip additional check of "written flag" if ignore_badblocks */
if (!ignore_badblocks) {
/*
* If the hw ecc bytes are not those of a blank page, there's
* still a chance that the page is blank, but was read with
* errors. Check the "written flag" in last oob byte, which
* is set to zero when a page is written. If more than half
* the bits are set, assume a blank page. Unfortunately, the
* bit flips(s) are not reported in stats.
*/
if (doc->oob_buf[15]) {
int bit, numsetbits = 0;
unsigned long written_flag = doc->oob_buf[15];
for (bit = 0; bit < 8; bit++) {
if (written_flag & 0x01)
numsetbits++;
written_flag >>= 1;
}
if (numsetbits > 4) { /* assume blank */
printf("errors in blank page at offset %08x\n",
page * DOCG4_PAGE_SIZE);
return 0;
}
}
}
/*
* The hardware ecc unit produces oob_ecc ^ calc_ecc. The kernel's bch
* algorithm is used to decode this. However the hw operates on page
* data in a bit order that is the reverse of that of the bch alg,
* requiring that the bits be reversed on the result. Thanks to Ivan
* Djelic for his analysis!
*/
for (i = 0; i < 7; i++)
doc->ecc_buf[i] = bitrev8(doc->ecc_buf[i]);
numerrs = decode_bch(doc->bch, NULL, DOCG4_USERDATA_LEN, NULL,
doc->ecc_buf, NULL, errpos);
if (numerrs == -EBADMSG) {
printf("uncorrectable errors at offset %08x\n",
page * DOCG4_PAGE_SIZE);
return -EBADMSG;
}
BUG_ON(numerrs < 0); /* -EINVAL, or anything other than -EBADMSG */
/* undo last step in BCH alg (modulo mirroring not needed) */
for (i = 0; i < numerrs; i++)
errpos[i] = (errpos[i] & ~7)|(7-(errpos[i] & 7));
/* fix the errors */
for (i = 0; i < numerrs; i++) {
/* ignore if error within oob ecc bytes */
if (errpos[i] > DOCG4_USERDATA_LEN * 8)
continue;
/* if error within oob area preceeding ecc bytes... */
if (errpos[i] > DOCG4_PAGE_SIZE * 8)
__change_bit(errpos[i] - DOCG4_PAGE_SIZE * 8,
(unsigned long *)doc->oob_buf);
else /* error in page data */
__change_bit(errpos[i], (unsigned long *)buf);
}
printf("%d error(s) corrected at offset %08x\n",
numerrs, page * DOCG4_PAGE_SIZE);
return numerrs;
}
static int read_progstatus(struct docg4_priv *doc, void __iomem *docptr)
{
/*
* This apparently checks the status of programming. Done after an
* erasure, and after page data is written. On error, the status is
* saved, to be later retrieved by the nand infrastructure code.
*/
/* status is read from the I/O reg */
uint16_t status1 = readw(docptr + DOC_IOSPACE_DATA);
uint16_t status2 = readw(docptr + DOC_IOSPACE_DATA);
uint16_t status3 = readw(docptr + DOCG4_MYSTERY_REG);
MTDDEBUG(MTD_DEBUG_LEVEL3, "docg4: %s: %02x %02x %02x\n",
__func__, status1, status2, status3);
if (status1 != DOCG4_PROGSTATUS_GOOD ||
status2 != DOCG4_PROGSTATUS_GOOD_2 ||
status3 != DOCG4_PROGSTATUS_GOOD_2) {
doc->status = NAND_STATUS_FAIL;
printf("read_progstatus failed: %02x, %02x, %02x\n",
status1, status2, status3);
return -EIO;
}
return 0;
}
static int pageprog(struct mtd_info *mtd)
{
/*
* Final step in writing a page. Writes the contents of its
* internal buffer out to the flash array, or some such.
*/
struct nand_chip *nand = mtd->priv;
struct docg4_priv *doc = nand->priv;
void __iomem *docptr = CONFIG_SYS_NAND_BASE;
int retval = 0;
MTDDEBUG(MTD_DEBUG_LEVEL3, "docg4: %s\n", __func__);
writew(DOCG4_SEQ_PAGEPROG, docptr + DOC_FLASHSEQUENCE);
writew(DOC_CMD_PROG_CYCLE2, docptr + DOC_FLASHCOMMAND);
write_nop(docptr);
write_nop(docptr);
/* Just busy-wait; usleep_range() slows things down noticeably. */
poll_status(docptr);
writew(DOCG4_SEQ_FLUSH, docptr + DOC_FLASHSEQUENCE);
writew(DOCG4_CMD_FLUSH, docptr + DOC_FLASHCOMMAND);
writew(DOC_ECCCONF0_READ_MODE | 4, docptr + DOC_ECCCONF0);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
retval = read_progstatus(doc, docptr);
writew(0, docptr + DOC_DATAEND);
write_nop(docptr);
poll_status(docptr);
write_nop(docptr);
return retval;
}
static void sequence_reset(void __iomem *docptr)
{
/* common starting sequence for all operations */
writew(DOC_CTRL_UNKNOWN | DOC_CTRL_CE, docptr + DOC_FLASHCONTROL);
writew(DOC_SEQ_RESET, docptr + DOC_FLASHSEQUENCE);
writew(DOC_CMD_RESET, docptr + DOC_FLASHCOMMAND);
write_nop(docptr);
write_nop(docptr);
poll_status(docptr);
write_nop(docptr);
}
static void read_page_prologue(void __iomem *docptr, uint32_t docg4_addr)
{
/* first step in reading a page */
sequence_reset(docptr);
writew(DOCG4_SEQ_PAGE_READ, docptr + DOC_FLASHSEQUENCE);
writew(DOCG4_CMD_PAGE_READ, docptr + DOC_FLASHCOMMAND);
write_nop(docptr);
write_addr(docptr, docg4_addr);
write_nop(docptr);
writew(DOCG4_CMD_READ2, docptr + DOC_FLASHCOMMAND);
write_nop(docptr);
write_nop(docptr);
poll_status(docptr);
}
static void write_page_prologue(void __iomem *docptr, uint32_t docg4_addr)
{
/* first step in writing a page */
sequence_reset(docptr);
writew(DOCG4_SEQ_PAGEWRITE, docptr + DOC_FLASHSEQUENCE);
writew(DOCG4_CMD_PAGEWRITE, docptr + DOC_FLASHCOMMAND);
write_nop(docptr);
write_addr(docptr, docg4_addr);
write_nop(docptr);
write_nop(docptr);
poll_status(docptr);
}
static uint32_t mtd_to_docg4_address(int page, int column)
{
/*
* Convert mtd address to format used by the device, 32 bit packed.
*
* Some notes on G4 addressing... The M-Sys documentation on this device
* claims that pages are 2K in length, and indeed, the format of the
* address used by the device reflects that. But within each page are
* four 512 byte "sub-pages", each with its own oob data that is
* read/written immediately after the 512 bytes of page data. This oob
* data contains the ecc bytes for the preceeding 512 bytes.
*
* Rather than tell the mtd nand infrastructure that page size is 2k,
* with four sub-pages each, we engage in a little subterfuge and tell
* the infrastructure code that pages are 512 bytes in size. This is
* done because during the course of reverse-engineering the device, I
* never observed an instance where an entire 2K "page" was read or
* written as a unit. Each "sub-page" is always addressed individually,
* its data read/written, and ecc handled before the next "sub-page" is
* addressed.
*
* This requires us to convert addresses passed by the mtd nand
* infrastructure code to those used by the device.
*
* The address that is written to the device consists of four bytes: the
* first two are the 2k page number, and the second is the index into
* the page. The index is in terms of 16-bit half-words and includes
* the preceeding oob data, so e.g., the index into the second
* "sub-page" is 0x108, and the full device address of the start of mtd
* page 0x201 is 0x00800108.
*/
int g4_page = page / 4; /* device's 2K page */
int g4_index = (page % 4) * 0x108 + column/2; /* offset into page */
return (g4_page << 16) | g4_index; /* pack */
}
static void docg4_command(struct mtd_info *mtd, unsigned command, int column,
int page_addr)
{
/* handle standard nand commands */
struct nand_chip *nand = mtd->priv;
struct docg4_priv *doc = nand->priv;
uint32_t g4_addr = mtd_to_docg4_address(page_addr, column);
MTDDEBUG(MTD_DEBUG_LEVEL3, "%s %x, page_addr=%x, column=%x\n",
__func__, command, page_addr, column);
/*
* Save the command and its arguments. This enables emulation of
* standard flash devices, and also some optimizations.
*/
doc->last_command.command = command;
doc->last_command.column = column;
doc->last_command.page = page_addr;
switch (command) {
case NAND_CMD_RESET:
reset(CONFIG_SYS_NAND_BASE);
break;
case NAND_CMD_READ0:
read_page_prologue(CONFIG_SYS_NAND_BASE, g4_addr);
break;
case NAND_CMD_STATUS:
/* next call to read_byte() will expect a status */
break;
case NAND_CMD_SEQIN:
write_page_prologue(CONFIG_SYS_NAND_BASE, g4_addr);
/* hack for deferred write of oob bytes */
if (doc->oob_page == page_addr)
memcpy(nand->oob_poi, doc->oob_buf, 16);
break;
case NAND_CMD_PAGEPROG:
pageprog(mtd);
break;
/* we don't expect these, based on review of nand_base.c */
case NAND_CMD_READOOB:
case NAND_CMD_READID:
case NAND_CMD_ERASE1:
case NAND_CMD_ERASE2:
printf("docg4_command: unexpected nand command 0x%x\n",
command);
break;
}
}
static void docg4_read_buf(struct mtd_info *mtd, uint8_t *buf, int len)
{
int i;
struct nand_chip *nand = mtd->priv;
uint16_t *p = (uint16_t *)buf;
len >>= 1;
for (i = 0; i < len; i++)
p[i] = readw(nand->IO_ADDR_R);
}
static int docg4_read_oob(struct mtd_info *mtd, struct nand_chip *nand,
int page)
{
struct docg4_priv *doc = nand->priv;
void __iomem *docptr = CONFIG_SYS_NAND_BASE;
uint16_t status;
MTDDEBUG(MTD_DEBUG_LEVEL3, "%s: page %x\n", __func__, page);
/*
* Oob bytes are read as part of a normal page read. If the previous
* nand command was a read of the page whose oob is now being read, just
* copy the oob bytes that we saved in a local buffer and avoid a
* separate oob read.
*/
if (doc->last_command.command == NAND_CMD_READ0 &&
doc->last_command.page == page) {
memcpy(nand->oob_poi, doc->oob_buf, 16);
return 0;
}
/*
* Separate read of oob data only.
*/
docg4_command(mtd, NAND_CMD_READ0, nand->ecc.size, page);
writew(DOC_ECCCONF0_READ_MODE | DOCG4_OOB_SIZE, docptr + DOC_ECCCONF0);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
/* the 1st byte from the I/O reg is a status; the rest is oob data */
status = readw(docptr + DOC_IOSPACE_DATA);
if (status & DOCG4_READ_ERROR) {
printf("docg4_read_oob failed: status = 0x%02x\n", status);
return -EIO;
}
MTDDEBUG(MTD_DEBUG_LEVEL3, "%s: status = 0x%x\n", __func__, status);
docg4_read_buf(mtd, nand->oob_poi, 16);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
writew(0, docptr + DOC_DATAEND);
write_nop(docptr);
return 0;
}
static int docg4_write_oob(struct mtd_info *mtd, struct nand_chip *nand,
int page)
{
/*
* Writing oob-only is not really supported, because MLC nand must write
* oob bytes at the same time as page data. Nonetheless, we save the
* oob buffer contents here, and then write it along with the page data
* if the same page is subsequently written. This allows user space
* utilities that write the oob data prior to the page data to work
* (e.g., nandwrite). The disdvantage is that, if the intention was to
* write oob only, the operation is quietly ignored. Also, oob can get
* corrupted if two concurrent processes are running nandwrite.
*/
/* note that bytes 7..14 are hw generated hamming/ecc and overwritten */
struct docg4_priv *doc = nand->priv;
doc->oob_page = page;
memcpy(doc->oob_buf, nand->oob_poi, 16);
return 0;
}
static int docg4_block_neverbad(struct mtd_info *mtd, loff_t ofs, int getchip)
{
/* only called when module_param ignore_badblocks is set */
return 0;
}
static void docg4_write_buf16(struct mtd_info *mtd, const uint8_t *buf, int len)
{
int i;
struct nand_chip *nand = mtd->priv;
uint16_t *p = (uint16_t *)buf;
len >>= 1;
for (i = 0; i < len; i++)
writew(p[i], nand->IO_ADDR_W);
}
static int write_page(struct mtd_info *mtd, struct nand_chip *nand,
const uint8_t *buf, int use_ecc)
{
void __iomem *docptr = CONFIG_SYS_NAND_BASE;
uint8_t ecc_buf[8];
writew(DOC_ECCCONF0_ECC_ENABLE |
DOC_ECCCONF0_UNKNOWN |
DOCG4_BCH_SIZE,
docptr + DOC_ECCCONF0);
write_nop(docptr);
/* write the page data */
docg4_write_buf16(mtd, buf, DOCG4_PAGE_SIZE);
/* oob bytes 0 through 5 are written to I/O reg */
docg4_write_buf16(mtd, nand->oob_poi, 6);
/* oob byte 6 written to a separate reg */
writew(nand->oob_poi[6], docptr + DOCG4_OOB_6_7);
write_nop(docptr);
write_nop(docptr);
/* write hw-generated ecc bytes to oob */
if (likely(use_ecc)) {
/* oob byte 7 is hamming code */
uint8_t hamming = readb(docptr + DOC_HAMMINGPARITY);
hamming = readb(docptr + DOC_HAMMINGPARITY); /* 2nd read */
writew(hamming, docptr + DOCG4_OOB_6_7);
write_nop(docptr);
/* read the 7 bch bytes from ecc regs */
read_hw_ecc(docptr, ecc_buf);
ecc_buf[7] = 0; /* clear the "page written" flag */
}
/* write user-supplied bytes to oob */
else {
writew(nand->oob_poi[7], docptr + DOCG4_OOB_6_7);
write_nop(docptr);
memcpy(ecc_buf, &nand->oob_poi[8], 8);
}
docg4_write_buf16(mtd, ecc_buf, 8);
write_nop(docptr);
write_nop(docptr);
writew(0, docptr + DOC_DATAEND);
write_nop(docptr);
return 0;
}
static int docg4_write_page_raw(struct mtd_info *mtd, struct nand_chip *nand,
const uint8_t *buf, int oob_required)
{
return write_page(mtd, nand, buf, 0);
}
static int docg4_write_page(struct mtd_info *mtd, struct nand_chip *nand,
const uint8_t *buf, int oob_required)
{
return write_page(mtd, nand, buf, 1);
}
static int read_page(struct mtd_info *mtd, struct nand_chip *nand,
uint8_t *buf, int page, int use_ecc)
{
struct docg4_priv *doc = nand->priv;
void __iomem *docptr = CONFIG_SYS_NAND_BASE;
uint16_t status, edc_err, *buf16;
writew(DOC_ECCCONF0_READ_MODE |
DOC_ECCCONF0_ECC_ENABLE |
DOC_ECCCONF0_UNKNOWN |
DOCG4_BCH_SIZE,
docptr + DOC_ECCCONF0);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
/* the 1st byte from the I/O reg is a status; the rest is page data */
status = readw(docptr + DOC_IOSPACE_DATA);
if (status & DOCG4_READ_ERROR) {
printf("docg4_read_page: bad status: 0x%02x\n", status);
writew(0, docptr + DOC_DATAEND);
return -EIO;
}
docg4_read_buf(mtd, buf, DOCG4_PAGE_SIZE); /* read the page data */
/* first 14 oob bytes read from I/O reg */
docg4_read_buf(mtd, nand->oob_poi, 14);
/* last 2 read from another reg */
buf16 = (uint16_t *)(nand->oob_poi + 14);
*buf16 = readw(docptr + DOCG4_MYSTERY_REG);
/*
* Diskonchips read oob immediately after a page read. Mtd
* infrastructure issues a separate command for reading oob after the
* page is read. So we save the oob bytes in a local buffer and just
* copy it if the next command reads oob from the same page.
*/
memcpy(doc->oob_buf, nand->oob_poi, 16);
write_nop(docptr);
if (likely(use_ecc)) {
/* read the register that tells us if bitflip(s) detected */
edc_err = readw(docptr + DOC_ECCCONF1);
edc_err = readw(docptr + DOC_ECCCONF1);
/* If bitflips are reported, attempt to correct with ecc */
if (edc_err & DOC_ECCCONF1_BCH_SYNDROM_ERR) {
int bits_corrected = correct_data(mtd, buf, page);
if (bits_corrected == -EBADMSG)
mtd->ecc_stats.failed++;
else
mtd->ecc_stats.corrected += bits_corrected;
}
}
writew(0, docptr + DOC_DATAEND);
return 0;
}
static int docg4_read_page_raw(struct mtd_info *mtd, struct nand_chip *nand,
uint8_t *buf, int oob_required, int page)
{
return read_page(mtd, nand, buf, page, 0);
}
static int docg4_read_page(struct mtd_info *mtd, struct nand_chip *nand,
uint8_t *buf, int oob_required, int page)
{
return read_page(mtd, nand, buf, page, 1);
}
static int docg4_erase_block(struct mtd_info *mtd, int page)
{
struct nand_chip *nand = mtd->priv;
struct docg4_priv *doc = nand->priv;
void __iomem *docptr = CONFIG_SYS_NAND_BASE;
uint16_t g4_page;
MTDDEBUG(MTD_DEBUG_LEVEL3, "%s: page %04x\n", __func__, page);
sequence_reset(docptr);
writew(DOCG4_SEQ_BLOCKERASE, docptr + DOC_FLASHSEQUENCE);
writew(DOC_CMD_PROG_BLOCK_ADDR, docptr + DOC_FLASHCOMMAND);
write_nop(docptr);
/* only 2 bytes of address are written to specify erase block */
g4_page = (uint16_t)(page / 4); /* to g4's 2k page addressing */
writeb(g4_page & 0xff, docptr + DOC_FLASHADDRESS);
g4_page >>= 8;
writeb(g4_page & 0xff, docptr + DOC_FLASHADDRESS);
write_nop(docptr);
/* start the erasure */
writew(DOC_CMD_ERASECYCLE2, docptr + DOC_FLASHCOMMAND);
write_nop(docptr);
write_nop(docptr);
poll_status(docptr);
writew(DOCG4_SEQ_FLUSH, docptr + DOC_FLASHSEQUENCE);
writew(DOCG4_CMD_FLUSH, docptr + DOC_FLASHCOMMAND);
writew(DOC_ECCCONF0_READ_MODE | 4, docptr + DOC_ECCCONF0);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
write_nop(docptr);
read_progstatus(doc, docptr);
writew(0, docptr + DOC_DATAEND);
write_nop(docptr);
poll_status(docptr);
write_nop(docptr);
return nand->waitfunc(mtd, nand);
}
static int read_factory_bbt(struct mtd_info *mtd)
{
/*
* The device contains a read-only factory bad block table. Read it and
* update the memory-based bbt accordingly.
*/
struct nand_chip *nand = mtd->priv;
uint32_t g4_addr = mtd_to_docg4_address(DOCG4_FACTORY_BBT_PAGE, 0);
uint8_t *buf;
int i, block, status;
buf = kzalloc(DOCG4_PAGE_SIZE, GFP_KERNEL);
if (buf == NULL)
return -ENOMEM;
read_page_prologue(CONFIG_SYS_NAND_BASE, g4_addr);
status = docg4_read_page(mtd, nand, buf, 0, DOCG4_FACTORY_BBT_PAGE);
if (status)
goto exit;
/*
* If no memory-based bbt was created, exit. This will happen if module
* parameter ignore_badblocks is set. Then why even call this function?
* For an unknown reason, block erase always fails if it's the first
* operation after device power-up. The above read ensures it never is.
* Ugly, I know.
*/
if (nand->bbt == NULL) /* no memory-based bbt */
goto exit;
/*
* Parse factory bbt and update memory-based bbt. Factory bbt format is
* simple: one bit per block, block numbers increase left to right (msb
* to lsb). Bit clear means bad block.
*/
for (i = block = 0; block < DOCG4_NUMBLOCKS; block += 8, i++) {
int bitnum;
uint8_t mask;
for (bitnum = 0, mask = 0x80;
bitnum < 8; bitnum++, mask >>= 1) {
if (!(buf[i] & mask)) {
int badblock = block + bitnum;
nand->bbt[badblock / 4] |=
0x03 << ((badblock % 4) * 2);
mtd->ecc_stats.badblocks++;
printf("factory-marked bad block: %d\n",
badblock);
}
}
}
exit:
kfree(buf);
return status;
}
static int docg4_block_markbad(struct mtd_info *mtd, loff_t ofs)
{
/*
* Mark a block as bad. Bad blocks are marked in the oob area of the
* first page of the block. The default scan_bbt() in the nand
* infrastructure code works fine for building the memory-based bbt
* during initialization, as does the nand infrastructure function that
* checks if a block is bad by reading the bbt. This function replaces
* the nand default because writes to oob-only are not supported.
*/
int ret, i;
uint8_t *buf;
struct nand_chip *nand = mtd->priv;
struct nand_bbt_descr *bbtd = nand->badblock_pattern;
int block = (int)(ofs >> nand->bbt_erase_shift);
int page = (int)(ofs >> nand->page_shift);
uint32_t g4_addr = mtd_to_docg4_address(page, 0);
MTDDEBUG(MTD_DEBUG_LEVEL3, "%s: %08llx\n", __func__, ofs);
if (unlikely(ofs & (DOCG4_BLOCK_SIZE - 1)))
printf("%s: ofs %llx not start of block!\n",
__func__, ofs);
/* allocate blank buffer for page data */
buf = kzalloc(DOCG4_PAGE_SIZE, GFP_KERNEL);
if (buf == NULL)
return -ENOMEM;
/* update bbt in memory */
nand->bbt[block / 4] |= 0x01 << ((block & 0x03) * 2);
/* write bit-wise negation of pattern to oob buffer */
memset(nand->oob_poi, 0xff, mtd->oobsize);
for (i = 0; i < bbtd->len; i++)
nand->oob_poi[bbtd->offs + i] = ~bbtd->pattern[i];
/* write first page of block */
write_page_prologue(CONFIG_SYS_NAND_BASE, g4_addr);
docg4_write_page(mtd, nand, buf, 1);
ret = pageprog(mtd);
if (!ret)
mtd->ecc_stats.badblocks++;
kfree(buf);
return ret;
}
static uint8_t docg4_read_byte(struct mtd_info *mtd)
{
struct nand_chip *nand = mtd->priv;
struct docg4_priv *doc = nand->priv;
MTDDEBUG(MTD_DEBUG_LEVEL3, "%s\n", __func__);
if (doc->last_command.command == NAND_CMD_STATUS) {
int status;
/*
* Previous nand command was status request, so nand
* infrastructure code expects to read the status here. If an
* error occurred in a previous operation, report it.
*/
doc->last_command.command = 0;
if (doc->status) {
status = doc->status;
doc->status = 0;
}
/* why is NAND_STATUS_WP inverse logic?? */
else
status = NAND_STATUS_WP | NAND_STATUS_READY;
return status;
}
printf("unexpectd call to read_byte()\n");
return 0;
}
static int docg4_wait(struct mtd_info *mtd, struct nand_chip *nand)
{
struct docg4_priv *doc = nand->priv;
int status = NAND_STATUS_WP; /* inverse logic?? */
MTDDEBUG(MTD_DEBUG_LEVEL3, "%s...\n", __func__);
/* report any previously unreported error */
if (doc->status) {
status |= doc->status;
doc->status = 0;
return status;
}
status |= poll_status(CONFIG_SYS_NAND_BASE);
return status;
}
int docg4_nand_init(struct mtd_info *mtd, struct nand_chip *nand, int devnum)
{
uint16_t id1, id2;
struct docg4_priv *docg4;
int retval;
docg4 = kzalloc(sizeof(*docg4), GFP_KERNEL);
if (!docg4)
return -1;
mtd->priv = nand;
nand->priv = docg4;
/* These must be initialized here because the docg4 is non-standard
* and doesn't produce an id that the nand code can use to look up
* these values (nand_scan_ident() not called).
*/
mtd->size = DOCG4_CHIP_SIZE;
mtd->name = "Msys_Diskonchip_G4";
mtd->writesize = DOCG4_PAGE_SIZE;
mtd->erasesize = DOCG4_BLOCK_SIZE;
mtd->oobsize = DOCG4_OOB_SIZE;
nand->IO_ADDR_R =
(void __iomem *)CONFIG_SYS_NAND_BASE + DOC_IOSPACE_DATA;
nand->IO_ADDR_W = nand->IO_ADDR_R;
nand->chipsize = DOCG4_CHIP_SIZE;
nand->chip_shift = DOCG4_CHIP_SHIFT;
nand->bbt_erase_shift = DOCG4_ERASE_SHIFT;
nand->phys_erase_shift = DOCG4_ERASE_SHIFT;
nand->chip_delay = 20;
nand->page_shift = DOCG4_PAGE_SHIFT;
nand->pagemask = 0x3ffff;
nand->badblockpos = NAND_LARGE_BADBLOCK_POS;
nand->badblockbits = 8;
nand->ecc.layout = &docg4_oobinfo;
nand->ecc.mode = NAND_ECC_HW_SYNDROME;
nand->ecc.size = DOCG4_PAGE_SIZE;
nand->ecc.prepad = 8;
nand->ecc.bytes = 8;
nand->ecc.strength = DOCG4_T;
nand->options = NAND_BUSWIDTH_16 | NAND_NO_SUBPAGE_WRITE;
nand->controller = &nand->hwcontrol;
/* methods */
nand->cmdfunc = docg4_command;
nand->waitfunc = docg4_wait;
nand->select_chip = docg4_select_chip;
nand->read_byte = docg4_read_byte;
nand->block_markbad = docg4_block_markbad;
nand->read_buf = docg4_read_buf;
nand->write_buf = docg4_write_buf16;
nand->scan_bbt = nand_default_bbt;
nand->erase = docg4_erase_block;
nand->ecc.read_page = docg4_read_page;
nand->ecc.write_page = docg4_write_page;
nand->ecc.read_page_raw = docg4_read_page_raw;
nand->ecc.write_page_raw = docg4_write_page_raw;
nand->ecc.read_oob = docg4_read_oob;
nand->ecc.write_oob = docg4_write_oob;
/*
* The way the nand infrastructure code is written, a memory-based bbt
* is not created if NAND_SKIP_BBTSCAN is set. With no memory bbt,
* nand->block_bad() is used. So when ignoring bad blocks, we skip the
* scan and define a dummy block_bad() which always returns 0.
*/
if (ignore_badblocks) {
nand->options |= NAND_SKIP_BBTSCAN;
nand->block_bad = docg4_block_neverbad;
}
reset(CONFIG_SYS_NAND_BASE);
/* check for presence of g4 chip by reading id registers */
id1 = readw(CONFIG_SYS_NAND_BASE + DOC_CHIPID);
id1 = readw(CONFIG_SYS_NAND_BASE + DOCG4_MYSTERY_REG);
id2 = readw(CONFIG_SYS_NAND_BASE + DOC_CHIPID_INV);
id2 = readw(CONFIG_SYS_NAND_BASE + DOCG4_MYSTERY_REG);
if (id1 != DOCG4_IDREG1_VALUE || id2 != DOCG4_IDREG2_VALUE)
return -1;
/* initialize bch algorithm */
docg4->bch = init_bch(DOCG4_M, DOCG4_T, DOCG4_PRIMITIVE_POLY);
if (docg4->bch == NULL)
return -1;
retval = nand_scan_tail(mtd);
if (retval)
return -1;
/*
* Scan for bad blocks and create bbt here, then add the factory-marked
* bad blocks to the bbt.
*/
nand->scan_bbt(mtd);
nand->options |= NAND_BBT_SCANNED;
retval = read_factory_bbt(mtd);
if (retval)
return -1;
retval = nand_register(devnum);
if (retval)
return -1;
return 0;
}