Hi Marco,
On Thu, Jun 4, 2026 at 10:57 AM Marco Felsch <m.felsch(a)pengutronix.de> wrote:
>
> Hi Jens,
>
> On 26-05-25, Jens Wiklander wrote:
> > Hi,
> >
> > Tomorrow, Tuesday, it's time for another OP-TEE contributors' monthly
> > meeting. For time and connection details, see the calendar at
> > https://www.trustedfirmware.org/meetings/
>
> Sorry I saw this e-mail to late since I was in Nice last week to attend
> to the Embedded Recipes.
>
> > Regarding Kconfig support:https://github.com/OP-TEE/optee_os/pull/7714
> > It appears that the transition will take quite some time. We'll have a
> > window when some variables have transitioned to be managed by Kconfig,
> > but others remain managed by the makefiles. Do we care that variables
> > passed on the command line may work differently, depending on whether
> > they are managed by makefiles or Kconfig? Can something be done about
> > it?
>
> Could you forward me the outcome of this discussion? I hope to continue
> the work next week, since we _really_ want Kconfig support.
We agreed that the transition may take a couple of years and that we
should try to minimize confusion in the meantime. Until we've
completed the transition, it would be preferable if, from a user
(someone who builds OP-TEE) point of view, it doesn't matter whether a
particular config variable is managed by Kconfig or by Makefiles. At
least when building using the build git. The mindset should be to
avoid causing any problems for the users. Documentation is helpful,
but changing well-established behavior will cause problems.
If the above isn't possible, then my view is that we must plan which
config variables we transition. Then, after each release, we select a
new group of config variables to be merged in a single PR. Or
something like that, so there's some logic behind how different config
variables behave. This will make the transition longer and more
difficult, but that's the price of making incompatible changes.
Cheers,
Jens