[OpenWrt-Devel] [PATCH v2] ar71xx: add support for RB SXTsq 2nD

[Thibaut]

Hi,

> Le 15 mars 2020 à 13:52, <mail at adrianschmutzler.de> <mail at adrianschmutzler.de> a écrit :
> 
> Hi,
> 
>> -----Original Message-----
>> From: Thibaut [mailto:hacks at slashdirt.org]
>> Sent: Sonntag, 15. März 2020 13:39
>> To: mail at adrianschmutzler.de
>> Cc: OpenWrt Development List <openwrt-devel at lists.openwrt.org>
>> Subject: Re: [OpenWrt-Devel] [PATCH v2] ar71xx: add support for RB SXTsq
>> 2nD
>> 
>> Hi,
>> 
>>> Le 15 mars 2020 à 13:05, <mail at adrianschmutzler.de>
>> <mail at adrianschmutzler.de> a écrit :
>>> 
>>> Hi,
>>> 
>>>> -----Original Message-----
>>>> From: openwrt-devel [mailto:openwrt-devel-
>> bounces at lists.openwrt.org]
>>>> On Behalf Of Thibaut VARÈNE
>>>> Sent: Sonntag, 15. März 2020 11:35
>>>> To: openwrt-devel at lists.openwrt.org
>>>> Cc: Thibaut VARÈNE <hacks at slashdirt.org>
>>>> Subject: [OpenWrt-Devel] [PATCH v2] ar71xx: add support for RB SXTsq
>> 2nD
>>> 
>>> the ar71xx target will only be supported in already released 19.07 branch
>> and be removed afterwards.
>>> 
>>> Consequently, we do not accept any device support for this target
>> anymore.
>>> 
>>> Please work with the ath79 target instead, which is meant to replace ar71xx
>> and is also already included in 19.07.x (though Mikrotik devices have only
>> been added in master).
>> 
>> I had a feeling this would be the answer.
>> 
>> At least the code is there for those who need it, and it details the necessary
>> bits to get this hardware to work.
>> 
>> Regarding moving to ath79 there are a number of issues which I’m trying to
>> wrap my head around.
>> 
>> Specifically, right now we have a (imo very convenient) « one image fits them
>> all » approach for ar71xx-based mikrotik devices. This relies on load-time
>> detection of the hardware which is based part on the kernel command string
>> passed by the routerboot bootloader, and part on actual hardware detection
>> at kernel load time (MIPS machine name).
>> 
>> I don’t know yet if it is possible to transcribe this in a DTS world: routerboot
>> doesn’t support DTS, so my impression is that in order to maintain this
>> feature an intermediary loader is needed, which is going to be non-trivial.
>> 
>> Second, I don’t know if the machine name detection can also be
>> implemented in DTS: I was under the impression that all DTSes define a fixed
>> string, there doesn’t seem to be a mechanism to lift that string from the mtd
>> device as is currently done on ar71xx. Is this correct?
> 
> The way devices were detected in ar71xx also caused a lot of problem, as it was very confusing even to seasoned developers. Now, we just build a separate image for each device, so we do not need this mess of scripts for detection anymore.

I believe this is a waste of resources and a very suboptimal approach. I’m not sure I’m interested in spending time on this :P

Some devices share the exact same hardware and differ only in (marketing) name, as evidenced by:
https://git.openwrt.org/?p=openwrt/openwrt.git;a=commitdiff;h=5ac974f2145c770431a6eb7e006dd086b70224b1 <https://git.openwrt.org/?p=openwrt/openwrt.git;a=commitdiff;h=5ac974f2145c770431a6eb7e006dd086b70224b1>

(this device uses the 911L platform)

> Just have a look at how the few ath79 devices are implemented, but note that they will be moved to a mikrotik subtarget soon as indicated by Roger already.

I’ve offered in this thread a couple patches to align the ath79 implementation on the existing ramips one wrt mtd partitioning and naming.

Best,
Thibaut
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.infradead.org/pipermail/openwrt-devel/attachments/20200315/db1699a9/attachment.htm>
-------------- next part --------------
_______________________________________________
openwrt-devel mailing list
openwrt-devel at lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel