Aw: Re: [RFC v3 00/11] Add BananaPi R3

Matthias Brugger matthias.bgg at gmail.com
Sun Nov 13 14:28:35 PST 2022



On 11/11/2022 13:37, Frank Wunderlich wrote:
> Hi
> 
>> Gesendet: Freitag, 11. November 2022 um 10:17 Uhr
>> Von: "Matthias Brugger" <matthias.bgg at gmail.com>
>> An: "Frank Wunderlich" <frank-w at public-files.de>, "Rob Herring" <robh+dt at kernel.org>
>> Cc: linux-mediatek at lists.infradead.org, "Krzysztof Kozlowski" <krzysztof.kozlowski+dt at linaro.org>, linux-arm-kernel at lists.infradead.org, devicetree at vger.kernel.org, linux-kernel at vger.kernel.org, "Frank Wunderlich" <linux at fw-web.de>
>> Betreff: Re: Aw: [RFC v3 00/11] Add BananaPi R3
>>
>>
>>
>> On 10/11/2022 18:03, Frank Wunderlich wrote:
>>> Hi Rob,
>>>
>>> can you please take Patches 1+3 of v1 into your tree?
>>> due to reordering in my tree i missed them in this version
>>>
>>> https://patchwork.kernel.org/project/linux-mediatek/list/?series=685798
>>>
>>
>> They should go through the subsystem maintainers trees. I'd advise to resend,
>> dropping the RFC prefix and add the corresponding maintainers.
> 
> ok, wait for your push and i create v4 including the missing binding-patches and the reorder of wed-node together with my fixes.
> 
>> Regards,
>> Matthias
>>
>>> pcie-bindings are here:
>>>
>>> https://patchwork.kernel.org/project/linux-mediatek/list/?series=690172
>>>
>>> i found 1 error in bpi-r3 dts (properties from sfp1 not plural) which i will fix in next version and will drop
>>> the compile comment in dt overlays.
>>>
>>> also i've found this commit:
>>>
>>> https://git.kernel.org/pub/scm/linux/kernel/git/robh/linux.git/commit/?h=dt/next&id=363547d2191cbc32ca954ba75d72908712398ff2
>>>
>>> so maybe i need to rename my overlay dts files to dtso?
>>>
>>> maybe the others can be applied (at least the mt7986 nodes)?
> 
> does this solve your question about dt-overlays?
> 
> just to note for all...it is all 1 board having sd+emmc on 1 mmc-controller and nand+nor on one spi. it depends on hardware-switches which devices are accessable (1mmc+1 spi). Adding dts for all possible combinations will add more dts (sd+nor, sd+nand, emmc+nor, emmc+nand) than adding the 2 overlays or adds redundant config.
> 

I wonder if in the end that wouldn't be part of U-Boot to pass the correct DTB 
depending on the HW switches. Just one idea I had is, to create on DTS with 
sd+nor+nand+emmc if that's possible. And then leave the permutations to the boot 
firmware (U-Boot).

Regards,
Matthias

> or should i copy the emmc/sd dts and adding once nor or nand? so we have the dtsi and dts for combinations above (=5 total)?
> 
>>> regards Frank
>>



More information about the Linux-mediatek mailing list