hard coded value for idcode field in struct cpu_table

[Sławomir Cygan]

2011/3/6 Madhavi Manchala <madhavi.linux at gmail.com>:
> 2011/3/5 Sławomir Cygan <slawomir.cygan at gmail.com>:
>> On Thu, Mar 3, 2011 at 2:49 PM, Madhavi Manchala
>> <madhavi.linux at gmail.com> wrote:
>>> Dear All,
>>>
>>> I do not understand the value of idcode field in the struct cpu_table
>>> i.e. the value is hard coded. I did not find that value in the user
>>> manuals also. The code is available at
>>> linux/arch/arm/plat-s3c24xx/cpu.c file. Currently, I am implementing
>>> code for Samsung S3C2510A processor which is not available in the
>>> current sources. Where can I find the idcode value for Samsung
>>> S3C2510A processor?
>>>
>>> Any ideas will be appreciated.
>>>
>>> Thanks and  Regards,
>>> Madhavi M.
>>>
>> Hi,
>> This field is defined for example in this datasheed:
>>
>> http://www.alldatasheet.com/datasheet-pdf/pdf/84872/SAMSUNG/S3C2410A.html
>>
>> PDF page 282:
>>
>> GENERAL STATUS REGISTER (GSTATUSn)
>> Register Address R/W Description Reset Value
>> GSTATUS0 0x560000AC R External pin status Undefined
>>
>>>>>>>   GSTATUS1 0x560000B0 R Chip ID 0x32410000 <<<
>>
>> GSTATUS2 0x560000B4 R/W Reset status 0x1
>> GSTATUS3 0x560000B8 R/W Infrom register 0x0
>> GSTATUS4 0x560000BC R/W Infrom register 0x0
>>
>> This is "GSTATUS1" register initial value
>>
>>
>> I don't think that S3c25 are so similar to S3c25, so they provide
>> exactly the same method of chip identification.
>
> Thank you very much Slawomir Cygan for the information. However, I did
> not find any CHIP ID in the S3C2510A documentation, as you said. There
> are IDs which are PCI vendor ID and device ID. I guess, this is
> different from that chip ID as in case S3C2410A.
>
> Thanks and Regards,
> Madhavi M.
>
>>
>> --
>> Pozdrawiam,
>> Sławomir Cygan
>>
>

At first glance the datasheets are quite different, the assumption
that S3C24xx are similar to S3C25xx may not be so good.

And the second thing: You were asking about cpu identification code
snippet, which can be omitted at all at least for the beginning (we
now know which cpu model we boot, aren't we?).

-- 
Pozdrawiam,
Sławomir Cygan