Patchwork [U-Boot,NAND] Fixes 16bit NAND support with the NDFC

login
register
mail settings
Submitter Alex
Date May 19, 2011, 8:23 p.m.
Message ID <4DD57C56.9050904@dawning.com>
Download mbox | patch
Permalink /patch/96454/
State Superseded
Headers show

Comments

Alex - May 19, 2011, 8:23 p.m.
From: Alex Waterman <awaterman@dawning.com>
Date: Thu, 19 May 2011 15:08:36 -0400
Subject: [PATCH] [NAND] Fixes 16bit NAND support with the NDFC

This patch adds support for 16 bit NAND devices attached to the
NDFC on ppc4xx processors. Two config entries were added:

  CONFIG_SYS_NDFC_16        - Setting this tells the NDFC that a
  			      16 bit device is attached.
  CONFIG_SYS_NDFC_EBC0_CFG  - This is for the External Bus
  			      Controller configuration register.

Also, a new ndfc_read_byte() function was added which does not
first convert the data to little endian.

The NAND SPL was also modified to do 16bit bad block testing
when a 16 bit chip is being used.

Signed-off-by: Alex Waterman <awaterman@dawning.com>
Cc: Scott Wood <scottwood@freescale.com>
Cc: Stefan Roese <sr@denx.de>
---
 README                  |    8 ++++++++
 drivers/mtd/nand/ndfc.c |   34 +++++++++++++++++++++++++++++-----
 nand_spl/nand_boot.c    |   11 ++++++++---
 3 files changed, 45 insertions(+), 8 deletions(-)

Here is the new patch with the README updated to reflect the config
options added.
Alex - May 26, 2011, 1:40 p.m.
I know there has been a lot of activity related to other things in the U-Boot source, but has anyone had a chance to review this patch?

Regards,
Alex
Scott Wood - May 26, 2011, 4:11 p.m.
On Thu, 26 May 2011 09:40:46 -0400
Alex Waterman <awaterman@dawning.com> wrote:

> 
> I know there has been a lot of activity related to other things in the U-Boot source, but has anyone had a chance to review this patch?

Looks mostly OK to me --  I was going to consider it for next, rather than
master, as despite "fix" in the name it's really adding new hardware support.

You may want to use an #ifdef for bus width in nand_boot.c rather than
spending bytes to check it dynamically.

Likewise in the non-SPL driver code, you already have
CONFIG_SYS_NDFC_16BIT, so why check dynamically in ndfc_read_byte()?

-Scott
Alex - May 26, 2011, 6:33 p.m.
Scott,

> Looks mostly OK to me --  I was going to consider it for next, rather than
> master, as despite "fix" in the name it's really adding new hardware support.

Ahh, yeah, that makes sense. I will change "Fixes" to "Adds" for next 
submission.

> You may want to use an #ifdef for bus width in nand_boot.c rather than
> spending bytes to check it dynamically.

OK. I suppose its unlikely that a NAND chip will swap from 8bit to 16bit (or 
vice versa) from one compile to another :).

> Likewise in the non-SPL driver code, you already have
> CONFIG_SYS_NDFC_16BIT, so why check dynamically in ndfc_read_byte()?

My thought process was something along the lines of one compiled binary working
on potentially multiple similar boards... Though in thinking about it, that 
seems like a needless requirement. I will change to #ifdefs for smaller code 
size. That's probably more useful overall.

Regards,
Alex

Patch

diff --git a/README b/README
index 6f3748d..3ede798 100644
--- a/README
+++ b/README
@@ -2912,6 +2912,14 @@  Low Level (hardware related) configuration options:
 - CONFIG_SYS_SRIOn_MEM_SIZE:
 		Size of SRIO port 'n' memory region
 
+- CONFIG_SYS_NDFC_16
+		Defined to tell the NDFC that the NAND chip is using a
+		16 bit bus.
+
+- CONFIG_SYS_NDFC_EBC0_CFG
+		Sets the EBC0_CFG register for the NDFC. If not defined
+		a default value will be used.
+
 - CONFIG_SPD_EEPROM
 		Get DDR timing information from an I2C EEPROM. Common
 		with pluggable memory modules such as SODIMMs
diff --git a/drivers/mtd/nand/ndfc.c b/drivers/mtd/nand/ndfc.c
index 0729e0c..b533474 100644
--- a/drivers/mtd/nand/ndfc.c
+++ b/drivers/mtd/nand/ndfc.c
@@ -37,6 +37,13 @@ 
 #include <asm/io.h>
 #include <asm/ppc4xx.h>
 
+#ifndef CONFIG_SYS_NAND_BCR
+#define CONFIG_SYS_NAND_BCR 0x80002222
+#endif
+#ifndef CONFIG_SYS_NDFC_EBC0_CFG
+#define CONFIG_SYS_NDFC_EBC0_CFG 0xb8400000
+#endif
+
 /*
  * We need to store the info, which chip-select (CS) is used for the
  * chip number. For example on Sequoia NAND chip #0 uses
@@ -140,11 +147,23 @@  static int ndfc_verify_buf(struct mtd_info *mtdinfo, const uint8_t *buf, int len
 
 	return 0;
 }
-#endif /* #ifndef CONFIG_NAND_SPL */
 
-#ifndef CONFIG_SYS_NAND_BCR
-#define CONFIG_SYS_NAND_BCR 0x80002222
-#endif
+/*
+ * Read a byte from the NDFC.
+ */
+static uint8_t ndfc_read_byte(struct mtd_info *mtd)
+{
+
+	struct nand_chip *chip = mtd->priv;
+
+	if (chip->options & NAND_BUSWIDTH_16)
+		return (uint8_t) readw(chip->IO_ADDR_R);
+	else
+		return readb(chip->IO_ADDR_R);
+
+}
+
+#endif /* #ifndef CONFIG_NAND_SPL */
 
 void board_nand_select_device(struct nand_chip *nand, int chip)
 {
@@ -198,16 +217,21 @@  int board_nand_init(struct nand_chip *nand)
 	nand->ecc.bytes = 3;
 	nand->select_chip = ndfc_select_chip;
 
+#ifdef CONFIG_SYS_NDFC_16BIT
+	nand->options |= NAND_BUSWIDTH_16;
+#endif
+
 #ifndef CONFIG_NAND_SPL
 	nand->write_buf  = ndfc_write_buf;
 	nand->verify_buf = ndfc_verify_buf;
+	nand->read_byte = ndfc_read_byte;
 
 	chip++;
 #else
 	/*
 	 * Setup EBC (CS0 only right now)
 	 */
-	mtebc(EBC0_CFG, 0xb8400000);
+	mtebc(EBC0_CFG, CONFIG_SYS_NDFC_EBC0_CFG);
 
 	mtebc(PB0CR, CONFIG_SYS_EBC_PB0CR);
 	mtebc(PB0AP, CONFIG_SYS_EBC_PB0AP);
diff --git a/nand_spl/nand_boot.c b/nand_spl/nand_boot.c
index 9545a9a..4683c7c 100644
--- a/nand_spl/nand_boot.c
+++ b/nand_spl/nand_boot.c
@@ -122,10 +122,15 @@  static int nand_is_bad_block(struct mtd_info *mtd, int block)
 	nand_command(mtd, block, 0, CONFIG_SYS_NAND_BAD_BLOCK_POS, NAND_CMD_READOOB);
 
 	/*
-	 * Read one byte
+	 * Read one byte (or two if it's a 16 bit chip).
 	 */
-	if (readb(this->IO_ADDR_R) != 0xff)
-		return 1;
+	if (this->options & NAND_BUSWIDTH_16) {
+		if (readw(this->IO_ADDR_R) != 0xffff)
+			return 1;
+	} else {
+		if (readb(this->IO_ADDR_R) != 0xff)
+			return 1;
+	}
 
 	return 0;
 }