On Thu, 22 Jan 2026, Jeffrey Law wrote: > How is this materially different than any indirect function call where the > caller and callee don't have matching modes? Presumably those are handled via > some kind of stub or trampoline mechanism. I don't offhand see why this > couldn't be handled in the same manner (and why it isn't already). So I think > a bit more digging is necessary here.
Cf. <https://inbox.sourceware.org/gcc-patches/[email protected]/>. Maciej
