Re: [linux-next:master 1937/6910] drivers/mtd/nand/raw/qcom_nandc.c:2941 qcom_op_cmd_mapping() error: uninitialized symbol 'ret'.

[Sricharan Ramabadhran <quic_srichara@quicinc.com>]

Hi Miquel,

<..>

>>>>>>
>>>>>> Look at your qcom_op_cmd_mapping, you don't even have a path for 
>>>>>> reads.
>>>>>> I bet something along:
>>>>>>     CMD_READ0:
>>>>>>         ret = XXX_OPCODE_READ;
>>>>>>         break;
>>>>>> will make it work.
>>>>>>
>>>>>> Please fix the driver and test with nandbiterrs -i. If this test 
>>>>>> works,
>>>>>> it is encouraging. Otherwise it is still broken.
>>>>>
>>>>>     ok understand. Will fix this up.
>>>>
>>>> This is urgent now. The driver in -next is broken, shall I revert
>>>> everything or will you test the above fix quickly?
>>>
>>>     While we are able to reproduce the issue, add path for
>>>     CMD_READ0/READ_START for IPQ platforms. All mtd tests including
>>>     nandbiterrs works fine.  But we are not able to get hold of a SDX
>>>     device where we could reproduce the issue to debug. We are working
>>>     offline with Manivannan for the SDX remote debug. Will keep you
>>>     updated here and if we are not able to solve this on SDX in next
>>>     couple of days, only way looks like revert and re-apply it for 6.7.
>>
>> If the READ path works on IPQ platforms it's already a huge step
>> forward, please send the fix on top of Mannivan series.
>>
>    ok, will send. Please confirm if its all fine.
>    There are more abstraction/common mode in the driver to be pulled
>    out, but that will do later.
> 
      Just now Alam posted this series for this.
       "[PATCH 0/3]mtd: rawnand: qcom: Fixes for exec_op".

Regards,
  Sricharan


>> Now, what is the issue with SDX?
> 
>     Looks like SDX device that Mani is having is a 4K nand device.
>     For some reason, after exec ops nand boot is broken on that board.
>     We applied the fixes we did on top of Manivannan series.
>     With that enumeration works, but still mount fails. But 4K nand
>     device boot on IPQ works fine. Problem is we do not have a SDX board
>     where we could debug that. We are trying to get a remote setup
>     enabled for this.
> 
> Regards,
>   Sricharan