Onenand support in msm7227

Onenand support in msm7227

[Mateusz Krawczuk]

 I want ask for any help with developing onenand driver.

 I tried generic onenand driver, but without success. Kernel module
was loaded and did nothing or kernel was crashing, and other strange
things.

 I also tried use msm-nand but when I use msm-nand from code-aurora
kernel crash/panic and get rebooted. When I use msm-nand from htc
kernel source I'm getting invalid num_resources or this:

 [ 22.129249] msm_nand_probe: phys addr 0xa0a00000
 [ 22.129278] msm_nand_probe: dmac 0x7
 [ 22.129339] msm_nand_probe: allocated dma buffer at ffc4c000,
dma_addr 27294000

 with my modification for onenand .

Onenand is connected ebi2.

 Best Regards
 Mateusz Krawczuk

Re: Onenand support in msm7227

[David Brown]

On Fri, Dec 23, 2011 at 02:04:10PM +0100, Mateusz Krawczuk wrote:

>  I want ask for any help with developing onenand driver.
> 
>  I tried generic onenand driver, but without success. Kernel module
> was loaded and did nothing or kernel was crashing, and other strange
> things.

I doubt the generic onenand driver will ever work on MSM.  The MSM
doesn't allow read/write access to the onenand device, it must go
through the NAND controller.

>  I also tried use msm-nand but when I use msm-nand from code-aurora
> kernel crash/panic and get rebooted. When I use msm-nand from htc
> kernel source I'm getting invalid num_resources or this:
> 
>  [ 22.129249] msm_nand_probe: phys addr 0xa0a00000
>  [ 22.129278] msm_nand_probe: dmac 0x7
>  [ 22.129339] msm_nand_probe: allocated dma buffer at ffc4c000,
> dma_addr 27294000
> 
>  with my modification for onenand .

You shouldn't need any modification to the code-aurora msm-nand driver
to support onenand.  Have you tried it unmodified?

The MSM NAND controller's support for onenand is very different from
most.  The NAND controller tries to make the onenand device look like
a NAND device.

The upstream kernel doesn't have any support yet for the NAND/onenand
controller.

BTW, if you are an MSM customer, I highly recommend using the support
channels, since you will get much better support that way.

David

-- 
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

Re: Onenand support in msm7227

[Mateusz Krawczuk]

I'm trying to make working with chip kat007012c, but I got errors:
<6>[    1.436921] flash_onenand_probe: manufacturer_id = 0xec
<6>[    1.450291] flash_onenand_probe: device_id = 0x50
<6>[    1.463681] flash_onenand_probe: version_id = 0x241
<6>[    1.477086] flash_onenand_probe: data_buf_size = 0x800
<6>[    1.490524] flash_onenand_probe: boot_buf_size = 0x200
<6>[    1.503981] flash_onenand_probe: num_of_buffers = 0x101
<6>[    1.517448] flash_onenand_probe: technology = 0x0

So I added values to drivers/mtd/devices/msm_nand.c
I detected but size of  detected blocks wasn`t corrected so I changed
mtd->erasesize = mtd->writesize << 7;
It helped but, nanddump drops from 25mb partition/block over 300mb of
data and whole file contains only FF.
<6>[    0.706874] flash_onenand_probe: manufacturer_id = 0xec
<6>[    0.706889] flash_onenand_probe: device_id = 0x50
<6>[    0.706904] flash_onenand_probe: version_id = 0x13e
<6>[    0.706921] flash_onenand_probe: data_buf_size = 0x800
<6>[    0.706936] flash_onenand_probe: boot_buf_size = 0x200
<6>[    0.706953] flash_onenand_probe: num_of_buffers = 0x101
<6>[    0.706968] flash_onenand_probe: technology = 0x0
<6>[    0.706999] Found a supported onenand device
<5>[    0.707024] Creating 14 MTD partitions on "msm_nand":
<5>[    0.707056] 0x000000000000-0x000000180000 : "mibib"
<5>[    0.710093] 0x000000180000-0x000000200000 : "qcsbl"
<5>[    0.711733] 0x000000200000-0x0000002c0000 : "oemsbl"
<5>[    0.713751] 0x0000002c0000-0x000001bc0000 : "amss"
<5>[    0.746481] 0x000001bc0000-0x000003400000 : "efs2"
<5>[    0.763296] 0x000003400000-0x000003900000 : "nv_backup"
<5>[    0.770368] 0x000003900000-0x000005200000 : "fota"
<5>[    0.803076] 0x000005200000-0x000005400000 : "arm11boot"
<5>[    0.806706] 0x000005400000-0x000005e00000 : "boot"
<5>[    0.820499] 0x000005e00000-0x000006800000 : "recovery"
<5>[    0.834269] 0x000006800000-0x0000068c0000 : "parameter"
<5>[    0.836289] 0x0000068c0000-0x000011a40000 : "system"
<5>[    0.950216] 0x000011a40000-0x00001dc00000 : "userdata"
<5>[    1.073928] 0x00001dc00000-0x00001f500000 : "cache"

I have Samsung: KAT007012C - 4Gb Nand Flash and 3Gb DRAM (2G 1G die).
Do you have any idea how to fix this?

Re: Onenand support in msm7227

[David Brown]

On Thu, Aug 16, 2012 at 08:18:07PM +0200, Mateusz Krawczuk wrote:
> Subject: Re: Onenand support in msm7227

Which kernel are you using?  The mainline kernel only seems to have a
boardfile for 7x27, but it isn't ever compiled.

Thanks,
David

-- 
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

Re: Onenand support in msm7227

[Mateusz Krawczuk]

I use latest version  2.6.35 for msm from Codeaurora forum. I also
tried with 3,0, but I have some problems with regulators.

2012/8/17 David Brown <davidb@codeaurora.org>:
> On Thu, Aug 16, 2012 at 08:18:07PM +0200, Mateusz Krawczuk wrote:
>> Subject: Re: Onenand support in msm7227
>
> Which kernel are you using?  The mainline kernel only seems to have a
> boardfile for 7x27, but it isn't ever compiled.
>
> Thanks,
> David
>
> --
> Sent by an employee of the Qualcomm Innovation Center, Inc.
> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

Re: Onenand support in msm7227

[Pankaj Jangra]

Hi,

On Fri, Aug 17, 2012 at 10:59 AM, Mateusz Krawczuk
<willingmagic@gmail.com> wrote:
> I use latest version  2.6.35 for msm from Codeaurora forum. I also
> tried with 3,0, but I have some problems with regulators.

AFAIK On 3.0 @ codeaurora, there is not much support for 7227. So you might want
to use the 2.6.35 only.

>
> 2012/8/17 David Brown <davidb@codeaurora.org>:
>> On Thu, Aug 16, 2012 at 08:18:07PM +0200, Mateusz Krawczuk wrote:
>>> Subject: Re: Onenand support in msm7227
>>
>> Which kernel are you using?  The mainline kernel only seems to have a
>> boardfile for 7x27, but it isn't ever compiled.
>>
>> Thanks,
>> David
>>
>> --
>> Sent by an employee of the Qualcomm Innovation Center, Inc.
>> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Fwd: Onenand support in msm7227

[Mateusz Krawczuk]

From: Mateusz Krawczuk <willingmagic@gmail.com>
Date: 2012/8/17
Subject: Re: Onenand support in msm7227
To: Pankaj Jangra <jangra.pankaj9@gmail.com>


I use 2.6.35 M76XXTSNCJNLYA6140, but I have no idea how to fix it.

Re: Onenand support in msm7227

[David Brown]

On Fri, Aug 17, 2012 at 07:29:37AM +0200, Mateusz Krawczuk wrote:

> I use latest version  2.6.35 for msm from Codeaurora forum. I also
> tried with 3,0, but I have some problems with regulators.

This mailing list mostly focuses on the code that is in Linus' tree,
so, it might not be the best support venue.  There aren't currently
any plans to support the msm7x27 chips in the mainline kernel.  As
Pankaj pointed out, the Codeaurora 3.0 kernel also doesn't have
support for the msm7x27.

If you are an MSM customer, you should have a support channel that
could give you better results.

Otherwise, I think some of the msm storage people do read this list,
and might be able to dig through their memories.

David

-- 
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.