Patchwork [U-Boot] mtd: nand: Allow caller to pass alternative ID table to nand_scan_ident()

login
register
mail settings
Submitter Lei Wen
Date Jan. 3, 2011, 3:06 p.m.
Message ID <1294067164-21747-1-git-send-email-leiwen@marvell.com>
Download mbox | patch
Permalink /patch/77283/
State Superseded
Headers show

Comments

Lei Wen - Jan. 3, 2011, 3:06 p.m.
This patch sync with newest mtd code in Linux on handling
nand_scan_ident.

Signed-off-by: Lei Wen <leiwen@marvell.com>
---
 drivers/mtd/nand/atmel_nand.c |    2 +-
 drivers/mtd/nand/nand_base.c  |   29 +++++++++++++++--------------
 drivers/mtd/nand/nand_ids.c   |    2 +-
 include/linux/mtd/nand.h      |    6 ++++--
 4 files changed, 21 insertions(+), 18 deletions(-)
Mike Frysinger - Jan. 3, 2011, 3:35 p.m.
On Mon, Jan 3, 2011 at 10:06 AM, Lei Wen <leiwen@marvell.com> wrote:
> This patch sync with newest mtd code in Linux on handling
> nand_scan_ident.

is there any point to the sync ?  doesnt seem like anyone is using the
new option.

> --- a/drivers/mtd/nand/nand_ids.c
> +++ b/drivers/mtd/nand/nand_ids.c
> -const struct nand_flash_dev nand_flash_ids[] = {
> +struct nand_flash_dev nand_flash_ids[] = {

this is wrong
-mike
Lei Wen - Jan. 3, 2011, 3:48 p.m.
Hi Mike,

On Mon, Jan 3, 2011 at 11:35 PM, Mike Frysinger <vapier@gentoo.org> wrote:
> On Mon, Jan 3, 2011 at 10:06 AM, Lei Wen <leiwen@marvell.com> wrote:
>> This patch sync with newest mtd code in Linux on handling
>> nand_scan_ident.
>
> is there any point to the sync ?  doesnt seem like anyone is using the
> new option.

This patch sync with David's patch on Linux.
commit 5e81e88a4c140586d9212999cea683bcd66a15c6
Author: David Woodhouse <David.Woodhouse@intel.com>
Date:   Fri Feb 26 18:32:56 2010 +0000

    mtd: nand: Allow caller to pass alternative ID table to nand_scan_ident()

    Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>

Yes, for this time uboot's code hasn't used this kind of code. But
this prepare for
my submition of pxa3xx_nand.c code in the next several weeks, which
need this patch
in advance.

>
>> --- a/drivers/mtd/nand/nand_ids.c
>> +++ b/drivers/mtd/nand/nand_ids.c
>> -const struct nand_flash_dev nand_flash_ids[] = {
>> +struct nand_flash_dev nand_flash_ids[] = {
>
> this is wrong

This would bring warning to the compilation, and it is also what Linux
does now for the
same code...

Best regards,
Lei
Mike Frysinger - Jan. 3, 2011, 4:32 p.m.
On Mon, Jan 3, 2011 at 10:48 AM, Lei Wen wrote:
> On Mon, Jan 3, 2011 at 11:35 PM, Mike Frysinger wrote:
>> On Mon, Jan 3, 2011 at 10:06 AM, Lei Wen wrote:
>>> --- a/drivers/mtd/nand/nand_ids.c
>>> +++ b/drivers/mtd/nand/nand_ids.c
>>> -const struct nand_flash_dev nand_flash_ids[] = {
>>> +struct nand_flash_dev nand_flash_ids[] = {
>>
>> this is wrong
>
> This would bring warning to the compilation, and it is also what Linux
> does now for the same code...

then fix the callers to take a const flash table instead of removing
the const markings on a const table
-mike
Wolfgang Denk - Jan. 3, 2011, 4:57 p.m.
Dear Lei Wen,

In message <AANLkTi=t-Ho8pV=Nd_uN7K+XOwmWVo8L825Q9_weqD3g@mail.gmail.com> you wrote:
> This patch sync with David's patch on Linux.
> commit 5e81e88a4c140586d9212999cea683bcd66a15c6
> Author: David Woodhouse <David.Woodhouse@intel.com>
> Date:   Fri Feb 26 18:32:56 2010 +0000
> 
>     mtd: nand: Allow caller to pass alternative ID table to nand_scan_ident()
> 
>     Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>

If you ever post such patches, then please pay speacial attention to
properly attribute such patches. See
http://www.denx.de/wiki/view/U-Boot/Patches#Attributing_Code_Copyrights_Sign

All patches that do not adhere to this policy will be rejected.

Best regards,

Wolfgang Denk
Scott Wood - Jan. 3, 2011, 6:38 p.m.
On Mon, 3 Jan 2011 10:35:45 -0500
Mike Frysinger <vapier@gentoo.org> wrote:

> On Mon, Jan 3, 2011 at 10:06 AM, Lei Wen <leiwen@marvell.com> wrote:
> > This patch sync with newest mtd code in Linux on handling
> > nand_scan_ident.
> 
> is there any point to the sync ?  doesnt seem like anyone is using the
> new option.

Even if Lei didn't need this for an upcoming driver, I see value
in keeping in sync unless we have a good reason to deviate.  It
makes it easier to share drivers and bring over new features that we
*do* want to use.

> > --- a/drivers/mtd/nand/nand_ids.c
> > +++ b/drivers/mtd/nand/nand_ids.c
> > -const struct nand_flash_dev nand_flash_ids[] = {
> > +struct nand_flash_dev nand_flash_ids[] = {
> 
> this is wrong

I agree.

-Scott

Patch

diff --git a/drivers/mtd/nand/atmel_nand.c b/drivers/mtd/nand/atmel_nand.c
index d5eb54a..ab8bbb3 100644
--- a/drivers/mtd/nand/atmel_nand.c
+++ b/drivers/mtd/nand/atmel_nand.c
@@ -296,7 +296,7 @@  int board_nand_init(struct nand_chip *nand)
 	mtd->priv = nand;
 
 	/* Detect NAND chips */
-	if (nand_scan_ident(mtd, 1)) {
+	if (nand_scan_ident(mtd, 1, NULL)) {
 		printk(KERN_WARNING "NAND Flash not found !\n");
 		return -ENXIO;
 	}
diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
index 5239c1f..39fcc94 100644
--- a/drivers/mtd/nand/nand_base.c
+++ b/drivers/mtd/nand/nand_base.c
@@ -2414,10 +2414,10 @@  static void nand_set_defaults(struct nand_chip *chip, int busw)
  */
 static const struct nand_flash_dev *nand_get_flash_type(struct mtd_info *mtd,
 						  struct nand_chip *chip,
-						  int busw, int *maf_id)
+						  int busw, int *maf_id,
+						  struct nand_flash_dev *type)
 {
-	const struct nand_flash_dev *type = NULL;
-	int i, dev_id, maf_idx;
+	int dev_id, maf_idx;
 	int tmp_id, tmp_manf;
 
 	/* Select the device */
@@ -2456,15 +2456,14 @@  static const struct nand_flash_dev *nand_get_flash_type(struct mtd_info *mtd,
 		return ERR_PTR(-ENODEV);
 	}
 
-	/* Lookup the flash id */
-	for (i = 0; nand_flash_ids[i].name != NULL; i++) {
-		if (dev_id == nand_flash_ids[i].id) {
-			type =  &nand_flash_ids[i];
-			break;
-		}
-	}
+	if (!type)
+		type = nand_flash_ids;
+
+	for (; type->name != NULL; type++)
+		if (dev_id == type->id)
+                        break;
 
-	if (!type) {
+	if (!type->name) {
 		/* supress warning if there is no nand */
 		if (*maf_id != 0x00 && *maf_id != 0xff &&
 		    dev_id  != 0x00 && dev_id  != 0xff)
@@ -2580,13 +2579,15 @@  static const struct nand_flash_dev *nand_get_flash_type(struct mtd_info *mtd,
  * nand_scan_ident - [NAND Interface] Scan for the NAND device
  * @mtd:	     MTD device structure
  * @maxchips:	     Number of chips to scan for
+ * @table:	     Alternative NAND ID table
  *
  * This is the first phase of the normal nand_scan() function. It
  * reads the flash ID and sets up MTD fields accordingly.
  *
  * The mtd->owner field must be set to the module of the caller.
  */
-int nand_scan_ident(struct mtd_info *mtd, int maxchips)
+int nand_scan_ident(struct mtd_info *mtd, int maxchips,
+		    struct nand_flash_dev *table)
 {
 	int i, busw, nand_maf_id;
 	struct nand_chip *chip = mtd->priv;
@@ -2598,7 +2599,7 @@  int nand_scan_ident(struct mtd_info *mtd, int maxchips)
 	nand_set_defaults(chip, busw);
 
 	/* Read the flash type */
-	type = nand_get_flash_type(mtd, chip, busw, &nand_maf_id);
+	type = nand_get_flash_type(mtd, chip, busw, &nand_maf_id, table);
 
 	if (IS_ERR(type)) {
 #ifndef CONFIG_SYS_NAND_QUIET_TEST
@@ -2869,7 +2870,7 @@  int nand_scan(struct mtd_info *mtd, int maxchips)
 {
 	int ret;
 
-	ret = nand_scan_ident(mtd, maxchips);
+	ret = nand_scan_ident(mtd, maxchips, NULL);
 	if (!ret)
 		ret = nand_scan_tail(mtd);
 	return ret;
diff --git a/drivers/mtd/nand/nand_ids.c b/drivers/mtd/nand/nand_ids.c
index 8d7ea76..67da52a 100644
--- a/drivers/mtd/nand/nand_ids.c
+++ b/drivers/mtd/nand/nand_ids.c
@@ -22,7 +22,7 @@ 
 +	256	256 Byte page size
 *	512	512 Byte page size
 */
-const struct nand_flash_dev nand_flash_ids[] = {
+struct nand_flash_dev nand_flash_ids[] = {
 
 #ifdef CONFIG_MTD_NAND_MUSEUM_IDS
 	{"NAND 1MiB 5V 8-bit",		0x6e, 256, 1, 0x1000, 0},
diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h
index 1128f5a..bcbf4dd 100644
--- a/include/linux/mtd/nand.h
+++ b/include/linux/mtd/nand.h
@@ -33,11 +33,13 @@ 
 
 
 struct mtd_info;
+struct nand_flash_dev;
 /* Scan and identify a NAND device */
 extern int nand_scan (struct mtd_info *mtd, int max_chips);
 /* Separate phases of nand_scan(), allowing board driver to intervene
  * and override command or ECC setup according to flash type */
-extern int nand_scan_ident(struct mtd_info *mtd, int max_chips);
+extern int nand_scan_ident(struct mtd_info *mtd, int max_chips,
+			   struct nand_flash_dev *table);
 extern int nand_scan_tail(struct mtd_info *mtd);
 
 /* Free resources held by the NAND device */
@@ -471,7 +473,7 @@  struct nand_manufacturers {
 	char * name;
 };
 
-extern const struct nand_flash_dev nand_flash_ids[];
+extern struct nand_flash_dev nand_flash_ids[];
 extern const struct nand_manufacturers nand_manuf_ids[];
 
 extern int nand_scan_bbt(struct mtd_info *mtd, struct nand_bbt_descr *bd);