Skip to contentSkip to conversation

NAND FSL UPM: driver re-write using the hwcontrol callback

PATCH New Wolfgang Grandegger 5 participants 7 messages Jun 2, 2008 – Jun 4, 2008

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 NAND FSL UPM: driver re-write using the hwcontrol callback

NAND FSL UPM: driver re-write using the hwcontrol callback

This is a re-write of the NAND FSL UPM driver using the more universal hwcontrol callback (instead of the cmdfunc callback). Here is a brief list of furher modifications:

- For the time being, the UPM setup writing the UPM array has been removed from the driver and must now be done by the board specific code.

- The bus width definition in "struct fsl_upm_nand" is now in bits to comply with the corresponding Linux driver.

- chip->dev_read is only set if fun->dev_ready != NULL, which is required for boards not connecting the R/B pin.

- A few issue have been fixed with MxMR bit manipulation like in the corresponding Linux driver.

Note: I think the "io_addr" field of "struct fsl_upm" could be removed
as well, because the address is already determined by "nand->IO_ADDR_[RW]", but I'm not 100% sure.

This patch has been tested on a TQM8548 modules with the NAND chip Micron MT29F8G08FABWP.

This patch is based on the following patches posted to this list a few minutes ago:

[PATCH] PPC: add accessor macros to clear and set bits in one shot [PATCH] 83xx/85xx/86xx: add more MxMR local bus definitions

Anton, could you please verify if it works on your MPC8360ERDK board as well. A patch will follow. In principle, the NAND driver of the TQM8272 should work with it as well.

---
 drivers/mtd/nand/fsl_upm.c  |  133 +++++++++++++++++---------------------------
 include/linux/mtd/fsl_upm.h |    1 
 2 files changed, 52 insertions(+), 82 deletions(-)
drivers/mtd/nand/fsl_upm.c +52 -81
Index: u-boot/drivers/mtd/nand/fsl_upm.c
===================================================================
@@ -20,112 +20,89 @@
 #include <linux/mtd/fsl_upm.h>
 #include <nand.h>
 
-#define FSL_UPM_MxMR_OP_NO (0 << 28) /* normal operation */
-#define FSL_UPM_MxMR_OP_WA (1 << 28) /* write array */
-#define FSL_UPM_MxMR_OP_RA (2 << 28) /* read array */
-#define FSL_UPM_MxMR_OP_RP (3 << 28) /* run pattern */
+static int fsl_upm_in_pattern;
 
 static void fsl_upm_start_pattern(struct fsl_upm *upm, u32 pat_offset)
 {
-	out_be32(upm->mxmr, FSL_UPM_MxMR_OP_RP | pat_offset);
+	clrsetbits_be32(upm->mxmr, MxMR_MAD_MSK, MxMR_OP_RUNP | pat_offset);
 }
 
 static void fsl_upm_end_pattern(struct fsl_upm *upm)
 {
-	out_be32(upm->mxmr, FSL_UPM_MxMR_OP_NO);
-	while (in_be32(upm->mxmr) != FSL_UPM_MxMR_OP_NO)
+	clrbits32(upm->mxmr, MxMR_OP_RUNP);
+
+	while (in_be32(upm->mxmr) & MxMR_OP_RUNP)
 		eieio();
 }
 
 static void fsl_upm_run_pattern(struct fsl_upm *upm, int width, u32 cmd)
 {
-	out_be32(upm->mar, cmd << (32 - width * 8));
-	out_8(upm->io_addr, 0x0);
-}
-
-static void fsl_upm_setup(struct fsl_upm *upm)
-{
-	int i;
-
-	/* write upm array */
-	out_be32(upm->mxmr, FSL_UPM_MxMR_OP_WA);
-
-	for (i = 0; i < 64; i++) {
-		out_be32(upm->mdr, upm->array[i]);
+	out_be32(upm->mar, cmd << (32 - width));
+	switch (width) {
+	case 8:
 		out_8(upm->io_addr, 0x0);
+		break;
+	case 16:
+		out_be16(upm->io_addr, 0x0);
+		break;
+	case 32:
+		out_be32(upm->io_addr, 0x0);
+		break;
 	}
-
-	/* normal operation */
-	out_be32(upm->mxmr, FSL_UPM_MxMR_OP_NO);
-	while (in_be32(upm->mxmr) != FSL_UPM_MxMR_OP_NO)
-		eieio();
 }
 
-static void fun_cmdfunc(struct mtd_info *mtd, unsigned command, int column,
-			int page_addr)
+static void nand_hwcontrol (struct mtd_info *mtd, int cmd)
 {
 	struct nand_chip *chip = mtd->priv;
 	struct fsl_upm_nand *fun = chip->priv;
 
-	fsl_upm_start_pattern(&fun->upm, fun->upm_cmd_offset);
-
-	if (command == NAND_CMD_SEQIN) {
-		int readcmd;
-
-		if (column >= mtd->oobblock) {
-			/* OOB area */
-			column -= mtd->oobblock;
-			readcmd = NAND_CMD_READOOB;
-		} else if (column < 256) {
-			/* First 256 bytes --> READ0 */
-			readcmd = NAND_CMD_READ0;
-		} else {
-			column -= 256;
-			readcmd = NAND_CMD_READ1;
-		}
-		fsl_upm_run_pattern(&fun->upm, fun->width, readcmd);
+	switch (cmd) {
+	case NAND_CTL_SETCLE:
+		fsl_upm_start_pattern(&fun->upm, fun->upm_cmd_offset);
+		fsl_upm_in_pattern++;
+		break;
+	case NAND_CTL_SETALE:
+		fsl_upm_start_pattern(&fun->upm, fun->upm_addr_offset);
+		fsl_upm_in_pattern++;
+		break;
+	case NAND_CTL_CLRCLE:
+	case NAND_CTL_CLRALE:
+		fsl_upm_end_pattern(&fun->upm);
+		fsl_upm_in_pattern--;
+		break;
 	}
+#if 1
+	/* Temorary check */
+	if (fsl_upm_in_pattern < 0 || fsl_upm_in_pattern > 1)
+		printf("fsl_upm: Oops, unexpected fsl_upm_in_pattern %d\n",
+		       fsl_upm_in_pattern);
+#endif
+}
 
-	fsl_upm_run_pattern(&fun->upm, fun->width, command);
-
-	fsl_upm_end_pattern(&fun->upm);
-
-	fsl_upm_start_pattern(&fun->upm, fun->upm_addr_offset);
-
-	if (column != -1)
-		fsl_upm_run_pattern(&fun->upm, fun->width, column);
+static void nand_write_byte(struct mtd_info *mtd, u_char byte)
+{
+	struct nand_chip *chip = mtd->priv;
 
-	if (page_addr != -1) {
-		fsl_upm_run_pattern(&fun->upm, fun->width, page_addr);
-		fsl_upm_run_pattern(&fun->upm, fun->width,
-				    (page_addr >> 8) & 0xFF);
-		if (chip->chipsize > (32 << 20)) {
-			fsl_upm_run_pattern(&fun->upm, fun->width,
-					    (page_addr >> 16) & 0x0f);
-		}
-	}
+	if (fsl_upm_in_pattern) {
+		struct fsl_upm_nand *fun = chip->priv;
 
-	fsl_upm_end_pattern(&fun->upm);
+		fsl_upm_run_pattern(&fun->upm, fun->width, byte);
 
-	if (fun->wait_pattern) {
 		/*
 		 * Some boards/chips needs this. At least on MPC8360E-RDK we
 		 * need it. Probably weird chip, because I don't see any need
 		 * for this on MPC8555E + Samsung K9F1G08U0A. Usually here are
 		 * 0-2 unexpected busy states per block read.
 		 */
-		while (!fun->dev_ready())
-			debug("unexpected busy state\n");
+		if (fun->wait_pattern) {
+			while (!fun->dev_ready())
+				debug("unexpected busy state\n");
+		}
+	} else {
+		out_8(chip->IO_ADDR_W, byte);
 	}
 }
 
-static void nand_write_byte(struct mtd_info *mtd, u_char byte)
-{
-	struct nand_chip *chip = mtd->priv;
-
-	out_8(chip->IO_ADDR_W, byte);
-}
-
 static u8 nand_read_byte(struct mtd_info *mtd)
 {
 	struct nand_chip *chip = mtd->priv;
@@ -164,10 +141,6 @@ static int nand_verify_buf(struct mtd_in
 	return 0;
 }
 
-static void nand_hwcontrol(struct mtd_info *mtd, int cmd)
-{
-}
-
 static int nand_dev_ready(struct mtd_info *mtd)
 {
 	struct nand_chip *chip = mtd->priv;
@@ -179,22 +152,20 @@ static int nand_dev_ready(struct mtd_inf
 int fsl_upm_nand_init(struct nand_chip *chip, struct fsl_upm_nand *fun)
 {
 	/* yet only 8 bit accessors implemented */
-	if (fun->width != 1)
+	if (fun->width != 8 && fun->width != 16 && fun->width != 32)
 		return -ENOSYS;
 
-	fsl_upm_setup(&fun->upm);
-
 	chip->priv = fun;
 	chip->chip_delay = fun->chip_delay;
 	chip->eccmode = NAND_ECC_SOFT;
-	chip->cmdfunc = fun_cmdfunc;
 	chip->hwcontrol = nand_hwcontrol;
 	chip->read_byte = nand_read_byte;
 	chip->read_buf = nand_read_buf;
 	chip->write_byte = nand_write_byte;
 	chip->write_buf = nand_write_buf;
 	chip->verify_buf = nand_verify_buf;
-	chip->dev_ready = nand_dev_ready;
+	if (fun->dev_ready)
+		chip->dev_ready = nand_dev_ready;
 
 	return 0;
 }
include/linux/mtd/fsl_upm.h +0 -1
Index: u-boot/include/linux/mtd/fsl_upm.h
===================================================================
@@ -16,7 +16,6 @@
 #include <linux/mtd/nand.h>
 
 struct fsl_upm {
-	const u32 *array;
 	void __iomem *mdr;
 	void __iomem *mxmr;
 	void __iomem *mar;
#1 lore raw

PATCH 1/1 NAND FSL UPM: driver re-write using the hwcontrol callback

Quoted 29 lines · Wolfgang Grandegger

NAND FSL UPM: driver re-write using the hwcontrol callback

This is a re-write of the NAND FSL UPM driver using the more universal hwcontrol callback (instead of the cmdfunc callback). Here is a brief list of furher modifications:

- For the time being, the UPM setup writing the UPM array has been removed from the driver and must now be done by the board specific code.

- The bus width definition in "struct fsl_upm_nand" is now in bits to comply with the corresponding Linux driver.

- chip->dev_read is only set if fun->dev_ready != NULL, which is required for boards not connecting the R/B pin.

- A few issue have been fixed with MxMR bit manipulation like in the corresponding Linux driver.

Note: I think the "io_addr" field of "struct fsl_upm" could be removed
as well, because the address is already determined by "nand->IO_ADDR_[RW]", but I'm not 100% sure.

This patch has been tested on a TQM8548 modules with the NAND chip Micron MT29F8G08FABWP.

This patch is based on the following patches posted to this list a few minutes ago:

Quoted 6 lines

[PATCH] PPC: add accessor macros to clear and set bits in one shot [PATCH] 83xx/85xx/86xx: add more MxMR local bus definitions

Anton, could you please verify if it works on your MPC8360ERDK board as well. A patch will follow. In principle, the NAND driver of the TQM8272 should work with it as well.

Works great here (tested on MPC8360E-RDK). Thanks!

Quoted 1 line
#2 lore raw

PATCH 1/1 NAND FSL UPM: driver re-write using the hwcontrol callback

Quoted 30 lines · Anton Vorontsov
Quoted 29 lines · Wolfgang Grandegger

NAND FSL UPM: driver re-write using the hwcontrol callback

This is a re-write of the NAND FSL UPM driver using the more universal hwcontrol callback (instead of the cmdfunc callback). Here is a brief list of furher modifications:

- For the time being, the UPM setup writing the UPM array has been removed from the driver and must now be done by the board specific code.

- The bus width definition in "struct fsl_upm_nand" is now in bits to comply with the corresponding Linux driver.

- chip->dev_read is only set if fun->dev_ready != NULL, which is required for boards not connecting the R/B pin.

- A few issue have been fixed with MxMR bit manipulation like in the corresponding Linux driver.

Note: I think the "io_addr" field of "struct fsl_upm" could be removed
as well, because the address is already determined by "nand->IO_ADDR_[RW]", but I'm not 100% sure.

This patch has been tested on a TQM8548 modules with the NAND chip Micron MT29F8G08FABWP.

This patch is based on the following patches posted to this list a few minutes ago:

Quoted 8 lines
Quoted 6 lines

[PATCH] PPC: add accessor macros to clear and set bits in one shot [PATCH] 83xx/85xx/86xx: add more MxMR local bus definitions

Anton, could you please verify if it works on your MPC8360ERDK board as well. A patch will follow. In principle, the NAND driver of the TQM8272 should work with it as well.

Works great here (tested on MPC8360E-RDK). Thanks!

Great.

Quoted 3 lines
Quoted 1 line

All 4 patches are maintained by different people. What is the fastest way to get these patches in? Maybe Stefan could pick them up directly?

Wolfgang.

#3 lore raw

PATCH 1/1 NAND FSL UPM: driver re-write using the hwcontrol callback

Quoted 6 lines · Wolfgang Grandegger
Quoted 3 lines
Quoted 1 line

All 4 patches are maintained by different people. What is the fastest way to get these patches in? Maybe Stefan could pick them up directly?

No. I can't. Simply because it's not my responsibility anymore. :)

Scott Wood is now the custodian for NAND FLASH stuff. So he will probably pick it up. Scott?

Best regards,
Stefan

=====================================================================
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80 Email: office at denx.de
=====================================================================

#4 lore raw

PATCH 1/1 NAND FSL UPM: driver re-write using the hwcontrol callback

Quoted 7 lines · Stefan Roese
Quoted 4 lines · Wolfgang Grandegger
Quoted 2 lines
Quoted 1 line

All 4 patches are maintained by different people. What is the fastest way to get these patches in? Maybe Stefan could pick them up directly?

No. I can't. Simply because it's not my responsibility anymore. :)

Ah, right, no I remember. At http://www.denx.de/wiki/UBoot/Custodians
and http://git.denx.de/?p=u-boot.git;a=forks you are still listed as the owner of u-boot-nand-flash.git.

Quoted 1 line

Scott Wood is now the custodian for NAND FLASH stuff. So he will

probably pick

Quoted 1 line

it up. Scott?

That would be nice.

Wolfgang.

#5 lore raw

PATCH 1/1 NAND FSL UPM: driver re-write using the hwcontrol callback

Hi,

Quoted 4 lines
Quoted 3 lines · Stefan Roese
Quoted 2 lines · Wolfgang Grandegger
Quoted 2 lines
Quoted 1 line
Quoted 8 lines
Quoted 4 lines
Quoted 2 lines

All 4 patches are maintained by different people. What is the fastest way to get these patches in? Maybe Stefan could pick them up directly?

No. I can't. Simply because it's not my responsibility anymore. :)

Ah, right, no I remember. At http://www.denx.de/wiki/UBoot/Custodians
and http://git.denx.de/?p=u-boot.git;a=forks you are still listed as the owner of u-boot-nand-flash.git.

At least the first is a wiki :-P

Anyway, changed.

Cheers
Detlev

Signature · 8 lines
We can forgive a man for making a useful thing as long as he does not
admire it.  The only excuse  for making  a useless  thing is that one
admires it intensely.
                                    --- Oscar Wilde
--
DENX Software Engineering GmbH,      MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich,  Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-40 Fax: (+49)-8142-66989-80 Email: dzu at denx.de
#6 lore raw

PATCH 1/1 NAND FSL UPM: driver re-write using the hwcontrol callback

In reply to Wolfgang Grandegger
Quoted 5 lines · Wolfgang Grandegger

NAND FSL UPM: driver re-write using the hwcontrol callback

This is a re-write of the NAND FSL UPM driver using the more universal hwcontrol callback (instead of the cmdfunc callback). Here is a brief list of furher modifications:

[snip]

Quoted 5 lines

This patch is based on the following patches posted to this list a few minutes ago:

[PATCH] PPC: add accessor macros to clear and set bits in one shot [PATCH] 83xx/85xx/86xx: add more MxMR local bus definitions

[snip]

drivers/mtd/nand/fsl_upm.c static int nand_dev_ready(struct mtd_inf · line 155

Quoted 3 lines
drivers/mtd/nand/fsl_upm.c +1 -1
 	/* yet only 8 bit accessors implemented */
-	if (fun->width != 1)
+	if (fun->width != 8 && fun->width != 16 && fun->width != 32)

The above comment looks like it should be removed.

Otherwise, Acked-by: Scott Wood <[email protected]>

Feel free to send via a powerpc tree due to the dependencies.

-Scott