Hi Sebastien,
There are good cons and pros arguments for moving to CMake.
Just like many here I don' t have preference for one or another, but
we need to analyze what is better for NuttX evolution and make a good
decision.
The main pros of moving to CMake:
1) It is easier to integrate with new projects and IDEs
2) It will speed up the compilation of the building system
3) It will avoid cumbersome integration to make things work (see
Brennan comments)
The main cons of moving to CMake:
1) The current building system already works "fine" ("If it is not
broken, don't fix", comfort zone)
2) It could spend time and energy to maintain two building systems
during the transition period
3) People will need to learn a new too, although CMake seems easy at
first look, it is hard when you want to do something different.
I think I depicted the three more important point, case someone else
thing about some other important point of each side, please bring it
to the table.
BR,
Alan
On 5/22/23, Sebastien Lorquet <sebast...@lorquet.fr> wrote:
> I very much agree with all of these arguments.
>
> Thats a too large disruption for too little benefits.
>
> I dont want to be forced to use cmake.
>
> Everything we use here to integrate NuttX is based on makefiles.
>
> Why do we have to bring in yet another dependency? No, cmake is not
> installed in our build systems.
>
> Sebastien
>
>
> Le 20/05/2023 à 00:23, Tomek CEDRO a écrit :
>> I am thinking about this. "If it works don't fix it" comes to my mind.
>>
>> Current build system is amazingly simple coherent and fast. Building
>> firmware takes 17 seconds. Why change it?
>>
>> Such change will flip everything upside down. Adds lots of work and
>> even more possible problems.
>>
>> What would be the real benefit?
>>
>> How would it improve that 17 seconds?
>>
>> I think the practical presentation and comparison of results (i.e. ide
>> integration, ci automation) along with numbers (i.e. build time) needs
>> to take place before making any decisions.
>>
>
