On 21/12/2016 20:21, Florian Fainelli wrote:
> Hallo,
>
> On 12/20/2016 12:53 PM, John Crispin wrote:
>> Hi,
>>
>> i have been thinking of how to split up the kernel patches folder in a
>> better way. for a start we have all the backports inside the same folder
>> as our normal patches, unfortunately we have some backports and patches
>> that have in the meantime been sent upstream not numbered as 0xx-*.patch.
>>
>> i was thinking that we could add a backports-4.4 which would hold all
>> the backported patches.
>>
>> we could then split up the remaining patches into those that still need
>> to be sent upstream and those that we wont be able to send upstream. the
>> order in which we apply them would then be
>>
>> * backports
>> * pending patches
>> * lede patches
>
> Seems like a good directory structure proposal. How about having the
> patches also contain some additional metadata like:
>
> Upstream-status: pending|submitted|accepted|not relevant
> Kernel version: 4.8 (if accepted)
yes, i was thinking of exactly that. also add reasons why it failed,
what the todos are etc. I sent the pci quirk patch upstream and it was
NAK'ed today as it is a disable and not enable patch. in case this leads
to a big rework of a patch, that info should be added aswell.
>
> Kind of what Yocto/OE does with their patches so people know if they can
> help with the submission process of the patches?
never looked at what yocto does, will do so the next few days.
John
_______________________________________________
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev
