[PATCH v5 00/26] memory: omap-gpmc: mtd: nand: Support GPMC NAND on non-OMAP platforms

[nick]

On 2016-02-22 03:05 PM, Roger Quadros wrote:
> On 22/02/16 18:42, Tony Lindgren wrote:
>> * Roger Quadros <rogerq at ti.com> [160222 02:16]:
>>> On 20/02/16 00:04, Tony Lindgren wrote:
>>>> * Roger Quadros <rogerq at ti.com> [160219 13:27]:
>>>>> Hi,
>>>>>
>>>>> @Tony
>>>>> Patches 15 and 24 are new and will need your review.
>>>>> I've modified patch 22 to include the new am335x boards introduced since v4.4.
>>>>>
>>>>> Patches are based on top of omap-for-v4.6/dt so that the DT changes apply cleanly.
>>>>
>>>> Looks OK to me. Can we merge the dts changes separately? Otherwise
>>>> we will easily end up with tons of conflicts..
>>>
>>> I agree. But we just need to keep in mind that NAND functionality will be
>>> broken till all the patches in this series are merged. We don't maintain
>>> backward compatibility with the old DT implementation.
>>
>> Please let's not do that! That breaks booting and git bisect.
>> It's better to have a minimal branch where each patch boots fine.
> 
> Understood. I'll send out a patch set with only the minimal DT changes
> that doesn't break anything. This should be a preparatory step for
> the DT clean up.
> 
Would you mind and try testing this on a few boards as even miminal DT
changes can break things and thus I would like to see you test it if
possible on the boards affected by the DT change.
>>
>> Also, I think you should at least print a warning for the old
>> binding. Otherwise people with out of tree boards will have
>> hard time updating their patches to send to mainline tree.
> 
Second that I agree with Tony on this it's just the reality.
Nick
> OK. I'll add this warning mechanism to the omap-gpmc driver.
> 
> cheers,
> -roger
> 
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
>