[U-Boot] [PATCH 5/5] PXA: PXA3xx NAND Driver
Scott Wood
scottwood at freescale.com
Fri Jul 9 22:04:00 CEST 2010
On Tue, Jul 06, 2010 at 03:12:49AM +0200, Marek Vasut wrote:
> From: Compulab uboot <none at none>
Hmm?
> Signed-off-by: Marek Vasut <marek.vasut at gmail.com>
> ---
> drivers/mtd/nand/Makefile | 1 +
> drivers/mtd/nand/pxa3xx_nand.c | 848 ++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 849 insertions(+), 0 deletions(-)
> create mode 100644 drivers/mtd/nand/pxa3xx_nand.c
>
> diff --git a/drivers/mtd/nand/Makefile b/drivers/mtd/nand/Makefile
> index 28f27da..cd840cd 100644
> --- a/drivers/mtd/nand/Makefile
> +++ b/drivers/mtd/nand/Makefile
> @@ -50,6 +50,7 @@ COBJS-$(CONFIG_NAND_S3C64XX) += s3c64xx.o
> COBJS-$(CONFIG_NAND_SPEAR) += spr_nand.o
> COBJS-$(CONFIG_NAND_OMAP_GPMC) += omap_gpmc.o
> COBJS-$(CONFIG_NAND_PLAT) += nand_plat.o
> +COBJS-$(CONFIG_NAND_PXA3XX) += pxa3xx_nand.o
> endif
>
> COBJS := $(COBJS-y)
> diff --git a/drivers/mtd/nand/pxa3xx_nand.c b/drivers/mtd/nand/pxa3xx_nand.c
> new file mode 100644
> index 0000000..380c918
> --- /dev/null
> +++ b/drivers/mtd/nand/pxa3xx_nand.c
> @@ -0,0 +1,848 @@
> +/*
> + * drivers/mtd/nand/pxa3xx_nand.c
> + *
> + * Copyright ? 2005 Intel Corporation
> + * Copyright ? 2006 Marvell International Ltd.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + */
> +
> +#include <common.h>
> +
> +#if defined(CONFIG_CMD_NAND)
The makefile already limits compilation of this file to when
CONFIG_NAND_PXA3XX is defined.
> +#ifdef CONFIG_NEW_NAND_CODE
Hmm?
> +#define WAIT_EVENT_TIMEOUT (2 * CONFIG_SYS_HZ/10)
> +
> +static int wait_for_event(struct pxa3xx_nand_info *info, uint32_t event)
> +{
> + int timeout = WAIT_EVENT_TIMEOUT;
> + uint32_t ndsr;
> +
> + while (timeout--) {
> + ndsr = NDSR & NDSR_MASK;
> + if (ndsr & event) {
> + NDSR = ndsr;
> + return 0;
> + }
> + udelay(10);
> + }
You're defining WAIT_EVENT_TIMEOUT in terms of CONFIG_SYS_HZ ticks (which
are supposed to be 1ms, but in any case you should only use CONFIG_SYS_HZ
when you're interacting with get_ticks() or similar), but you're
interpreting it as a number of 10us intervals.
> + if (info->col_addr_cycles == 2) {
> + /* large block, 2 cycles for column address
> + * row address starts from 3rd cycle
> + */
> + info->ndcb1 |= page_addr << 16;
> + if (info->row_addr_cycles == 3)
> + info->ndcb2 = (page_addr >> 16) & 0xff;
> + } else
> + /* small block, 1 cycles for column address
> + * row address starts from 2nd cycle
> + */
> + info->ndcb1 = page_addr << 8;
If you put braces on one half of the if, please put it on the other half
(and also when you have a multi-line body even if it's technically only one
statement).
> +/* calculate delta between OSCR values start and now */
> +static unsigned long get_delta(unsigned long start)
> +{
> + unsigned long cur = OSCR;
> +
> + if(cur < start) /* OSCR overflowed */
> + return (cur + (start^0xffffffff));
> + else
> + return (cur - start);
> +}
What's wrong with get_ticks()?
> +
> +/* wait_event with timeout */
> +static unsigned int wait_event(unsigned long event)
> +{
> + unsigned long ndsr, timeout, start = OSCR;
> +
> + if(!event)
> + return 0xff000000;
> +
> + if(event & (NDSR_CS0_CMDD | NDSR_CS0_BBD))
> + timeout = CONFIG_SYS_NAND_PROG_ERASE_TO * OSCR_CLK_FREQ;
> + else
> + timeout = CONFIG_SYS_NAND_OTHER_TO * OSCR_CLK_FREQ;
> +
> + while(1) {
Space after keywords like "if" and "while".
> +static int pxa3xx_nand_do_cmd(struct pxa3xx_nand_info *info, uint32_t event)
> +{
> + unsigned int status;
> +
> + if (write_cmd(info)) {
> + info->retcode = ERR_SENDCMD;
> + goto fail_stop;
> + }
> +
> + info->state = STATE_CMD_HANDLE;
> +
> + status = wait_event(event);
> + if (status & (NDSR_RDDREQ | NDSR_DBERR)) {
> + if (status & NDSR_DBERR)
> + info->retcode = ERR_DBERR;
> +
> + info->state = STATE_PIO_READING;
> + } else if (status & NDSR_WRDREQ) {
> + info->state = STATE_PIO_WRITING;
> + } else if (status & (NDSR_CS0_BBD | NDSR_CS0_CMDD)) {
> + if (status & NDSR_CS0_BBD)
> + info->retcode = ERR_BBERR;
> +
> + info->state = STATE_READY;
> + }
> +/* NDSR = status; */
Why commented out?
> +fail_stop:
> + NDCR &= ~NDCR_ND_RUN;
> + udelay(10);
> + return -ETIMEDOUT;
Why is the udelay needed?
> + default:
> + printk(KERN_ERR "non-supported command.\n");
That's a bit vague -- tell the user this message comes from the NAND driver,
and what the bad command is.
> +/*
> + * not required for Monahans DFC
> + */
> +static void pxa3xx_nand_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl)
> +{
> + return;
> +}
Just don't provide it at all.
> +static int __readid(struct pxa3xx_nand_info *info, uint32_t *id)
> +{
> + const struct pxa3xx_nand_flash *f = info->flash_info;
> + const struct pxa3xx_nand_cmdset *cmdset = f->cmdset;
> + uint8_t id_buff[8];
> + int i;
> + unsigned long *long_buf;
> +
> + if (prepare_other_cmd(info, cmdset->read_id)) {
> + printk(KERN_ERR "failed to prepare command\n");
> + return -EINVAL;
> + }
> +
> + /* Send command */
> + if (write_cmd(info))
> + goto fail_timeout;
> +
> + /* Wait for CMDDM(command done successfully) */
> + if (wait_for_event(info, NDSR_RDDREQ))
> + goto fail_timeout;
> +
> + for (i = 0; i < 8; i += 4) {
> + long_buf = (unsigned long *) &id_buff[i];
> + *long_buf = NDDB;
> + }
> + *id = id_buff[0] | (id_buff[1] << 8);
> + return 0;
> +
> +fail_timeout:
> + NDCR &= ~NDCR_ND_RUN;
> + udelay(10);
> + return -ETIMEDOUT;
> +}
> +
Why is this done differently here than when the generic layer issues a
READID command?
> +static int pxa3xx_nand_config_flash(struct pxa3xx_nand_info *info,
> + const struct pxa3xx_nand_flash *f)
> +{
> + uint32_t ndcr = 0x00000FFF; /* disable all interrupts */
> +
> + if (f->page_size != 2048 && f->page_size != 512)
> + return -EINVAL;
> +
> + if (f->flash_width != 16 && f->flash_width != 8)
> + return -EINVAL;
> +
> + /* calculate flash information */
> + info->oob_size = (f->page_size == 2048) ? 64 : 16;
> + info->read_id_bytes = (f->page_size == 2048) ? 4 : 2;
> +
> + /* calculate addressing information */
> + info->col_addr_cycles = (f->page_size == 2048) ? 2 : 1;
> +
> + if (f->num_blocks * f->page_per_block > 65536)
> + info->row_addr_cycles = 3;
> + else
> + info->row_addr_cycles = 2;
This looks duplicative of pxa3xx_nand_detect_config.
> + /* set info fields needed to __readid */
> + info->flash_info = &default_flash;
> + info->read_id_bytes = (default_flash.page_size == 2048) ? 4 : 2;
> + info->reg_ndcr = ndcr;
> +
> + if (__readid(info, &id))
> + return -ENODEV;
> +
> + /* Lookup the flash id */
> + id = (id >> 8) & 0xff; /* device id is byte 2 */
> +
> + for (i = 0; nand_flash_ids[i].name != NULL; i++) {
> + if (id == nand_flash_ids[i].id) {
> + type = &nand_flash_ids[i];
> + break;
> + }
> + }
Why do you need to do this here? The generic NAND code will look this up.
> +static int pxa3xx_nand_detect_flash(struct pxa3xx_nand_info *info)
> +{
> + uint32_t id = -1;
> + int i;
> +
> + if (pxa3xx_nand_detect_config(info) == 0)
> + return 0;
> +
> + for (i = 0; i < 1; ++i) {
What's this one-iteration loop for?
> + if (pxa3xx_nand_config_flash(info, info->flash_info))
> + continue;
> +
> + if (__readid(info, &id))
> + continue;
> +
> + return 0;
> + }
> +
> + printf("failed to detect configured nand flash; found %04x instead of\n", id);
Instead of what? Did you really "find" anything in this case?
pxa3xx_nand_detect_config already does the __readid, why do you need to do
it again here? You don't do anything with the id if __readid succeeds.
> +
> + return -ENODEV;
> +}
> +
> +static struct nand_ecclayout hw_smallpage_ecclayout = {
> + .eccbytes = 6,
> + .eccpos = {8, 9, 10, 11, 12, 13 },
> + .oobfree = { {2, 6} }
> +};
Normally with small page NAND the bad block marker is at byte 5, at least
with 8-bit chips.
If you really have bad block information somewhere else, you'll want to
provide a non-default badblock_pattern.
> +#else
> +#error "U-Boot legacy NAND support not available for Monahans DFC."
> +#endif
> +#endif
Legacy NAND has been removed from u-boot altogether. Please don't
reintroduce references to it.
-Scott
More information about the U-Boot
mailing list