[Rpm-maint] [rpm-software-management/rpm] Make %optflags empty on noarch target (PR #2615)

[Panu Matilainen]

To clarify, I'm not particularly hell-bent on pushing this through, my primary motivation here is to have some discussion on and around the topic.

For all I know, the existing behavior of letting host %optflags through to noarch packages could even be intentional. This gets kinda philosophical as to what a noarch package is: should only the output of the build process be considered wrt arch-independence, or does the build process and the tooling used there matter?

As in, if a package compiles a tool  that is then used to produce arch-independent data which is then packaged, is it a noarch package? If we consider anything but the output itself, it gets *extremely* blurry.

No matter which way one looks at it, it'll be wrong for somebody unless we separate all macros for build (the system on which the package is built) and host (the system where the built package will be used), and packagers adopt the convention of treating noarch builds as an actual cross-build kind of thing...

-- 
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/2615#issuecomment-1678587881
You are receiving this because you are subscribed to this thread.

Message ID: <rpm-software-management/rpm/pull/2615/c1678587881 at github.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rpm.org/pipermail/rpm-maint/attachments/20230815/2df4347a/attachment-0001.html>