On 8/28/20 7:25 AM, Alan Modra wrote:
On Fri, Aug 28, 2020 at 01:17:27AM -0500, Segher Boessenkool wrote:
1) Very many unnecessary moves are generated during expand *anyway*, a
few more will not hurt;
2) In practice we always generate a move from a pseudo into r12 here,
never a copy from r12 into r12;
3) Why generate dead code optimising cases that do not happen, making
the compiler bigger and slower, but more importantly, making the
compiler code harder to read?
Uh, OK point 2 and followup 3 is persuasive. I should have thought of
that myself, thanks.
OK. I had already upstreamed the patch on David's ok, but I can go back
in and clear out those two unnecessary if's in _call and _sibcall fns.
Bill