[U-Boot-Users] [PATCH] NAND update 2

[U-Boot-Users] [PATCH] NAND update 2

[William Juul]

Make nand dump and nand dump.oob work again.

This is a small  update to the patch series I sent earlier
and must be applied on top of those patches.

Signed-off-by: William Juul <william.juul@tandberg.com>
Signed-off-by: Stig Olsen <stig.olsen@tandberg.com>
----

Have anyone tried/looked at the patch series?

On our git server http://git.tandberg.com/tandberg/u-boot.git, I have
updated the branch mtd-update to be based on v1.3.1 and I have also
made a mtd-update-1.3.1 branch. These two branches now contain the patch
included in this email.

I have also put an updated patch on our git http server:
http://git.tandberg.com/tandberg/patches/mtd-update.patch
This patch is a diff between my mtd-update-1.3.1 branch and v1.3.1 tag.

Best regards
William



diff --git a/common/cmd_nand.c b/common/cmd_nand.c
index bb1accd..a3ea1db 100644
--- a/common/cmd_nand.c
+++ b/common/cmd_nand.c
@@ -40,37 +40,44 @@ int find_dev_and_part(const char *id, struct mtd_device **dev,
 
 extern nand_info_t nand_info[];       /* info for NAND chips */
 
-static int nand_dump_oob(nand_info_t *nand, ulong off)
-{
-	return 0;
-}
-
-static int nand_dump(nand_info_t *nand, ulong off)
+static int nand_dump(nand_info_t *nand, ulong off, int only_oob)
 {
 	int i;
-	u_char *buf, *p;
+	u_char *datbuf, *oobbuf, *p;
 
-	buf = malloc(nand->writesize + nand->oobsize);
-	if (!buf) {
+	datbuf = malloc(nand->writesize + nand->oobsize);
+	oobbuf = malloc(nand->oobsize);
+	if (!datbuf || !oobbuf) {
 		puts("No memory for page buffer\n");
 		return 1;
 	}
 	off &= ~(nand->writesize - 1);
 	size_t dummy;
 	loff_t addr = (loff_t) off;
-	i = nand->read(nand, addr, nand->writesize, &dummy, buf);
+	struct mtd_oob_ops ops;
+	memset(&ops, 0, sizeof(ops));
+	ops.datbuf = datbuf;
+	ops.oobbuf = oobbuf; /* must exist, but oob data will be appended to ops.datbuf */
+	ops.len = nand->writesize;
+	ops.ooblen = nand->oobsize;
+	ops.mode = MTD_OOB_RAW;
+	i = nand->read_oob(nand, addr, &ops);
 	if (i < 0) {
 		printf("Error (%d) reading page %08x\n", i, off);
-		free(buf);
+		free(datbuf);
+		free(oobbuf);
 		return 1;
 	}
 	printf("Page %08x dump:\n", off);
-	i = nand->writesize >> 4; p = buf;
+	i = nand->writesize >> 4;
+	p = datbuf;
 	while (i--) {
-		printf( "\t%02x %02x %02x %02x %02x %02x %02x %02x"
-			"  %02x %02x %02x %02x %02x %02x %02x %02x\n",
-			p[0], p[1], p[2], p[3], p[4], p[5], p[6], p[7],
-			p[8], p[9], p[10], p[11], p[12], p[13], p[14], p[15]);
+		if (!only_oob) {
+			printf( "\t%02x %02x %02x %02x %02x %02x %02x %02x"
+					"  %02x %02x %02x %02x %02x %02x %02x %02x\n",
+					p[0], p[1], p[2], p[3], p[4], p[5], p[6], p[7],
+					p[8], p[9], p[10], p[11], p[12], p[13], p[14], p[15]);
+		}
 		p += 16;
 	}
 	puts("OOB:\n");
@@ -80,7 +87,8 @@ static int nand_dump(nand_info_t *nand, ulong off)
 			p[0], p[1], p[2], p[3], p[4], p[5], p[6], p[7]);
 		p += 8;
 	}
-	free(buf);
+	free(datbuf);
+	free(oobbuf);
 
 	return 0;
 }
@@ -304,9 +312,9 @@ int do_nand(cmd_tbl_t * cmdtp, int flag, int argc, char *argv[])
 		off = (int)simple_strtoul(argv[2], NULL, 16);
 
 		if (s != NULL && strcmp(s, ".oob") == 0)
-			ret = nand_dump_oob(nand, off);
+			ret = nand_dump(nand, off, 1);
 		else
-			ret = nand_dump(nand, off);
+			ret = nand_dump(nand, off, 0);
 
 		return ret == 0 ? 1 : 0;
 

[U-Boot-Users] [PATCH] NAND update 2

[Stefan Roese]

Hi William,

On Friday 21 December 2007, William Juul wrote:
> Make nand dump and nand dump.oob work again.
>
> This is a small  update to the patch series I sent earlier
> and must be applied on top of those patches.
>
> Signed-off-by: William Juul <william.juul@tandberg.com>
> Signed-off-by: Stig Olsen <stig.olsen@tandberg.com>
> ----
>
> Have anyone tried/looked at the patch series?
>
> On our git server http://git.tandberg.com/tandberg/u-boot.git, I have
> updated the branch mtd-update to be based on v1.3.1 and I have also
> made a mtd-update-1.3.1 branch. These two branches now contain the patch
> included in this email.

I'm trying to integrate your patches right now. Looks quite good so far but 
I'm facing some problems on a ppc4xx board with a large page NAND chip. Small 
page devices are fine.

Did you ever test this stuff on large page devices? Which ones?

BTW: What is the platform you "normally" test on (ARM, MIPS, PPC...)?

Thanks.

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
=====================================================================

[U-Boot-Users] [PATCH] NAND update 2

[William Juul]

On Jan 4, 2008 3:21 PM, Stefan Roese <sr@denx.de> wrote:
> Hi William,
>
> On Friday 21 December 2007, William Juul wrote:
> > Make nand dump and nand dump.oob work again.
> >
> > This is a small  update to the patch series I sent earlier
> > and must be applied on top of those patches.
> >
> > Signed-off-by: William Juul <william.juul@tandberg.com>
> > Signed-off-by: Stig Olsen <stig.olsen@tandberg.com>
> > ----
> >
> > Have anyone tried/looked at the patch series?
> >
> > On our git server http://git.tandberg.com/tandberg/u-boot.git, I have
> > updated the branch mtd-update to be based on v1.3.1 and I have also
> > made a mtd-update-1.3.1 branch. These two branches now contain the patch
> > included in this email.
>
> I'm trying to integrate your patches right now. Looks quite good so far but
> I'm facing some problems on a ppc4xx board with a large page NAND chip. Small
> page devices are fine.
>
> Did you ever test this stuff on large page devices? Which ones?
We did run this on large page devices with 4KB page size (total size
of 1/2 GB). Unfortunately the devices are/were under NDA as they
are/were not officially released yet.
Stig can you verify if any/both of the flashes can be named here?

>
> BTW: What is the platform you "normally" test on (ARM, MIPS, PPC...)?
We had two custom boards, one ARM based (926) and one PPC83xx.

I will not have access to the build environment and boards the next
couple of months, but Stig Olsen is more than capable of answering
questions (if he has the time :-)

Best regards
William

>
> Thanks.
>
> 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
> =====================================================================
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2005.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> U-Boot-Users mailing list
> U-Boot-Users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/u-boot-users
>

[U-Boot-Users] [PATCH] NAND update 2

[Stefan Roese]

On Friday 04 January 2008, William Juul wrote:
> > I'm trying to integrate your patches right now. Looks quite good so far
> > but I'm facing some problems on a ppc4xx board with a large page NAND
> > chip. Small page devices are fine.
> >
> > Did you ever test this stuff on large page devices? Which ones?
>
> We did run this on large page devices with 4KB page size (total size
> of 1/2 GB). Unfortunately the devices are/were under NDA as they
> are/were not officially released yet.
> Stig can you verify if any/both of the flashes can be named here?

Not needed anymore. I found my problem. A timeout was calculated incorrectly. 
I need to change nand_base.c at one place to make it work on my platform with 
my NAND chips. I'll post a till beginning of next week, so that you (or Stig) 
can test it on your platform too.

> > BTW: What is the platform you "normally" test on (ARM, MIPS, PPC...)?
>
> We had two custom boards, one ARM based (926) and one PPC83xx.
>
> I will not have access to the build environment and boards the next
> couple of months, but Stig Olsen is more than capable of answering
> questions (if he has the time :-)

Thanks.

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
=====================================================================