On Thu, Nov 14, 2024 at 08:38:38PM +0100, Jan Hubicka wrote:
> Concerning the other uses outside of inliner:
> - tree_inlinable_function_p and expand_call_inline check it to warn. We
> could silence the wraning via NO_INLINE_WRANING flag.
I know and initially I've even had that flag set in c-attribs.cc, but didn't
want to propagate that flag too everywhere in the FEs.
Though maybe it should be done anyway because C++ FE sets it on the
methods without inline keyword and should propagate to ctor/dtor clones etc.
> - ipa-icf only wants to be sure that both functions are same. It will
> compare attribute lists, so it will likely bail out on the new
> attribute differences.
> - ipa-split indeed would need to check it.
>
> Thinking of it, extra flag in DECL is probably slightly easier to use,
> so the IPA changes are OK as they are...
> >
> > And DECL_AGGRESSIVE_INLINING_P looks reasonable.
> ... perhaps with this change - I think it represents the intended use
> more clearly.
I'll test the patch with DECL_AGGRESSIVE_INLINING_P then tonight.
Jakub
