On 29 Jul 2022 10:15, Alexander Kanavin <[email protected]> wrote:

The standard SDK is not going anywhere. Obviously we're not removing
something that has a lot of existing users.

However I think it's still worth for you to play with the 'direct SDK'
(the patches just landed in master) [1]. The built-in extensibility
and ability to trivially update everything is beneficial. But you do
need a bulletproof sstate infra if the developers run things on
underpowered laptops.


This is the point I'm most worried about, keeping in mind that most of the users need to use 3rd party layers which aren't sstate stable at all, there might be a chance that this new approach doesn't really work well, even though I really like the idea of reusing the caches for this purpose.



[1] https://git.yoctoproject.org/poky/commit/?id=9b3fcb0d91648ae3b53ec8ffcb31fb6eac9209dd

Alex

On Fri, 29 Jul 2022 at 08:37, Matthias Klein <[email protected]> wrote:
>
> I have not been working with the eSDK so far.
>
> We are heavy users of the standard SDK, and use it daily in the team for application development.
> We use it with QtCreator and CMake including remote debugging. During setup, we obtain a single time information from the environment to create the kit within QtCreator.
> Later, when using the kit, the environment is no longer used. This has the advantage that you can use multiple SDKs for different devices and versions at the same time in QtCreator.
>
> Please don't eliminate the standard SDK with its separate installer and independence to BitBake etc.
> From the user perspective of an application developer it is perfect.
>
> Best regards,
> Matthias
>
>
>
>





-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#1605): 
https://lists.openembedded.org/g/openembedded-architecture/message/1605
Mute This Topic: https://lists.openembedded.org/mt/90990557/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-architecture/unsub 
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to