We unnecessarily restricted a fold of a splat to a constant vector.
We overlooked this fold in D102404 and earlier patches, but the fixed vector variant is shown in:
https://llvm.org/PR50817
Alive2 agrees on that:
https://alive2.llvm.org/ce/z/HpijPC
I don't think Alive2 understands scalable vectors, but the same logic applies IIUC.