Re: [PATCH v2 1/2] bitbake.conf: DISTRO_FEATURES as overrides

[Richard Purdie <richard.purdie@linuxfoundation.org>]

On Tue, 2017-06-13 at 09:14 +0200, Patrick Ohly wrote:
> Let me also point out that the approach below completely sidesteps
> potential vardeps problems, because the dependency of OVERRIDES on
> DISTRO_FEATURES_OVERRIDES and DISTRO_FEATURES is hidden. My
> expectation is that OVERRIDES are part of the base hash and that thus
> tracking of how it gets computed isn't necessary.

I'm not sure what the best approach for the ordering issue is. The
system is setup to dynamically adjust for OVERRIDES as they change.
Perhaps setting it at the end of parsing is better, I'm not sure.

Regardless, you're not quite right above when you say that OVERRIDES
are part of the basehash, they're not in most cases.

We actually need OVERRIDES to change without changing the basehash, for
example when you switch MACHINE, you don't want allarch packages to
rebuild (or tune shared packages with the same tune but a different
MACHINE).

For this reason, OVERRIDES are very much not part of the hash, we rely
on computed values.

Your patch has broken that in several places and I'm trying to untangle
an unfortunate series of changes locally to resolve this behaviour
correctly.

Cheers,

Richard