Trying to get 1Gb Samsung nand flash working

Trying to get 1Gb Samsung nand flash working

[extabe]

The flash works well in U-boot enviroment with jffs2 but I can´t get it running i Linux.
The chip i detected but then it reads badblocks on the entire chip.

"Scanning device for bad blocks
nand_read_oob: from = 0x00000000, len = 2
Bad eraseblock 0 at 0x00000000"
..and so on.
 I´ve used the same driver running a 64Mb nandflash with good result.

Anyone else who used this chip?


/Andre

Re: Trying to get 1Gb Samsung nand flash working

[Thomas Gleixner]

On Fri, 2005-03-04 at 18:49 +0000, extabe@bredband.net wrote:
> The flash works well in U-boot enviroment with jffs2 but I can´t get it running i Linux.
> The chip i detected but then it reads badblocks on the entire chip.
> 
> "Scanning device for bad blocks
> nand_read_oob: from = 0x00000000, len = 2
> Bad eraseblock 0 at 0x00000000"
> ..and so on.
>  I´ve used the same driver running a 64Mb nandflash with good result.
>
> Anyone else who used this chip?

I don't think its a chip problem. 

What's the exact chip type ?
Which kernel version ? 
Are you using MTD CVS code ?

Please tell me the CVS $Id string from drivers/mtd/nand/nand_bbt.c

tglx

RE: Trying to get 1Gb Samsung nand flash working

[Kyungmin Park]

> The flash works well in U-boot enviroment with jffs2 but I 
> can´t get it running i Linux.
> The chip i detected but then it reads badblocks on the entire chip.
> 
> "Scanning device for bad blocks
> nand_read_oob: from = 0x00000000, len = 2 Bad eraseblock 0 at 
> 0x00000000"
> ..and so on.
>  I´ve used the same driver running a 64Mb nandflash with good result.
> 
> Anyone else who used this chip?

omap_nor_cs3: Found 1 x16 devices at 0x0 in 16-bit bank
omap_nor_cs3: Found 1 x16 devices at 0x1000000 in 16-bit bank
 Intel/Sharp Extended Query Table at 0x0031
Using buffer write method
cfi_cmdset_0001: Erase suspend on write enabled
Creating 4 MTD partitions on "omap_nor_cs3":
0x00000000-0x00020000 : "bootloader"
0x00020000-0x00040000 : "params"
0x00040000-0x00240000 : "kernel"
0x00240000-0x02000000 : "filesystem"
MUX: initialized L3_1610_FLASH_CS2B_OE
MUX: initialized M8_1610_FLASH_CS2B_WE
No NAND device found!!!
No NAND device found!!!
NAND device: Manufacturer ID: 0xec, Chip ID: 0x79 (<NULL> NAND 128MiB 3,3V
8-bi)
NAND bus width 16 instead 8 bit
NAND device: Manufacturer ID: 0xec, Chip ID: 0x79 (Samsung NAND 128MiB 3,3V
8-b)
Scanning device for bad blocks
Creating 5 MTD partitions on "NAND 128MiB 3,3V 8-bit":
0x00000000-0x00010000 : "Booting Image"
0x00010000-0x00050000 : "U-Boot"
0x00050000-0x00080000 : "U-Boot Environment"
0x00080000-0x00280000 : "Kernel"
0x00280000-0x08000000 : "File System"

There's no problem :)

% Test environment

kernel: Linux version 2.6.11-omap1

Kyungmin Park

RE: Trying to get 1Gb Samsung nand flash working

[Thomas Gleixner]

On Sat, 2005-03-05 at 12:02 +0900, Kyungmin Park wrote:
> NAND device: Manufacturer ID: 0xec, Chip ID: 0x79 (<NULL> NAND 128MiB 3,3V
> 8-bi)
> NAND bus width 16 instead 8 bit
> NAND device: Manufacturer ID: 0xec, Chip ID: 0x79 (Samsung NAND 128MiB 3,3V
> 8-b)
> Scanning device for bad blocks
> Creating 5 MTD partitions on "NAND 128MiB 3,3V 8-bit":
> 0x00000000-0x00010000 : "Booting Image"
> 0x00010000-0x00050000 : "U-Boot"
> 0x00050000-0x00080000 : "U-Boot Environment"
> 0x00080000-0x00280000 : "Kernel"
> 0x00280000-0x08000000 : "File System"

Can you please tell me the CVS $Id string from
drivers/mtd/nand/nand_bbt.c ?

tglx

RE: Trying to get 1Gb Samsung nand flash working

[Kyungmin Park]

> 
> Can you please tell me the CVS $Id string from 
> drivers/mtd/nand/nand_bbt.c ?
> 

I just use default mtd in linux 2.6.11 kernel inside

* $Id: nand_bbt.c,v 1.28 2004/11/13 10:19:09 gleixner Exp $

Kyungmin Park

Re: Trying to get 1Gb Samsung nand flash working

[Charles Manning]

1Gb is not enough info.
The devices with 512-byte pages should work much the same (perhaps a 
difference in the addressing).
The devices with 2048-byte pages will work quite a bit differentrly.

Also, check that the pins are the same on the packages. I have seen some 
devices with extra pins that had to be held in a different state to work.


On Saturday 05 March 2005 07:49, extabe@bredband.net wrote:
> The flash works well in U-boot enviroment with jffs2 but I can´t get it
> running i Linux. The chip i detected but then it reads badblocks on the
> entire chip.
>
> "Scanning device for bad blocks
> nand_read_oob: from = 0x00000000, len = 2
> Bad eraseblock 0 at 0x00000000"
> ..and so on.
>  I´ve used the same driver running a 64Mb nandflash with good result.
>
> Anyone else who used this chip?
>
>
> /Andre
>
>
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: Trying to get 1Gb Samsung nand flash working

[Thomas Gleixner]

Andre, 

please place your answers below the citation and configure your mail
client to break the lines at 80 chars. See also
http://www.infradead.org/~dwmw2/email.html

Thanks

On Sat, 2005-03-05 at 12:56 +0000, extabe@bredband.net wrote:
> The chip is a Samsung 1Gx8 Bit NAND Flash (K9W8G08U1M). It is 
> detected as a 512Mb chip (K9k4G08U0M) since it consist of two 
> K9k4G08U0M stacked.

Should work.

> I use 2.4.25 kernel patched with "mtd-snapshot-20050223"
> 
> The CVS $Id string is:
> $Id: nand_bbt.c,v1.31 2005/02/16 17:09:36 dedekind Exp $

There was a change in the bad block table code recently. I'm not sure if
it is relevant to your problem - it should not. I send you a patch
seperately to backout the changes. Please give it a try.

> As I mentioned erlier I´ve used a Samsung 64Mb chip (K9F1208U0M) 
> with the same driver and it´s working good.

This chip is a small page chip (512Byte + 16Byte oob) and therefor a
different thing.

Can I have a look on your board driver please ?

tglx