[coreboot] Problem with M57SLI and newly installed MX25L4005

[Carl-Daniel Hailfinger]

On 06.02.2008 16:40, Chris Lingard wrote:
> Carl-Daniel Hailfinger wrote:
>   
>> On 06.02.2008 15:05, Chris Lingard wrote:
>>     
>>> Carl-Daniel Hailfinger wrote:
>>>
>>>   
>>>       
>>>> Thanks for testing.
>>>> The output should be mostly identical on the first look, but not exactly
>>>> identical. If it is exactly identical, the executed binary didn't
>>>> contain the patch.
>>>> I'm looking for the following output changes:
>>>>
>>>>     
>>>>         
>>>>> probe_spi: id1 0x7f, id2 0x9d7e
>>>>>       
>>>>>           
>>>> should become
>>>>     
>>>>         
>>>>> probe_spi: id1 0x7f9d, id2 0x7e
>>>>>       
>>>>>           
>>>> and there should be three additional lines near the end:
>>>>     
>>>>         
>>>>> Probing for PMC unknown SPI chip, 0 KB
>>>>> WARNING: size: 0 -> 4096 (page size)
>>>>> RDID returned 7f 9d 7d.
>>>>>       
>>>>>           
>>> Sorry.  Think I may have got it right now :-(
>>>   
>>>       
>> Yes, indeed.
>>
>> Thanks for the logs. You hit a bug in the existing code which was
>> uncovered by my patch. Can you change the second line of probe_spi() in
>> spi.c from
>> uint8_t manuf_id;
>> to
>> uint32_t manuf_id;
>> and post full results for the original chip again? This time it should
>> find a generic PMC SPI chip.
>>
>> Regards,
>> Carl-Daniel
>>     
>
> Am amazing prediction :-)
>   

Thanks. I like it when I'm right ;-)

> Switch down
> bash-3.2# ./flashrom    -m gigabyte:m57sli  -V
> Found chipset "NVIDIA MCP55", enabling flash write... OK.
> Found board "GIGABYTE GA-M57SLI-S4": enabling flash write... Serial 
> [...]
> Probing for PMC unknown SPI chip, 0 KB
> WARNING: size: 0 -> 4096 (page size)
> RDID returned 7f 9d 7e.
> probe_spi: id1 0x7f9d, id2 0x7e
> PMC unknown SPI chip found at physical address 0x100000000.
> Flash part is PMC unknown SPI chip (0 KB).
> No operations were specified.
>   

OK, so the generic detection works for the first chip.

> bash-3.2# ./flashrom    -m gigabyte:m57sli  -V
> Found chipset "NVIDIA MCP55", enabling flash write... OK.
> Found board "GIGABYTE GA-M57SLI-S4": enabling flash write... Serial 
> [...]
> Probing for PMC unknown SPI chip, 0 KB
> WARNING: size: 0 -> 4096 (page size)
> RDID returned 7f 9d 7e.
> probe_spi: id1 0x7f9d, id2 0x7e
> PMC unknown SPI chip found at physical address 0x100000000.
> Flash part is PMC unknown SPI chip (0 KB).
> No operations were specified.
>   

Same log as above, maybe double cut-n-paste?

> Switch up
>
>
> bash-3.2# ./flashrom    -m gigabyte:m57sli  -V
> Found chipset "NVIDIA MCP55", enabling flash write... OK.
> Found board "GIGABYTE GA-M57SLI-S4": enabling flash write... Serial 
> [...]
> Probing for PMC unknown SPI chip, 0 KB
> WARNING: size: 0 -> 4096 (page size)
> RDID returned 7f 9d 7d.
> probe_spi: id1 0x7f9d, id2 0x7d
> PMC unknown SPI chip found at physical address 0x100000000.
> Flash part is PMC unknown SPI chip (0 KB).
> No operations were specified.
>   

Second chip detected with the generic routine as well.

Can you try the patch I posted in a separate thread with subject
"[PATCH] flashrom: Support Pm25LV*"?

I only need normal flashrom output (without -V) unless a chip is
detected as "PMC unknown SPI chip".
If it works for you, can you please ack the patch? More details about
acking at http://www.coreboot.org/Development_Guidelines sections
"Sign-off Procedure"
<http://www.coreboot.org/Development_Guidelines#Sign-off_Procedure> and
"Reviews".

Regards,
Carl-Daniel

-- 
http://www.hailfinger.org/