add mulq, 4
add lenq, 32
jl .loop
- REP_RET
+%if notcpuflag(sse2)
+ ;; cvtpi2ps switches to MMX even if the source is a memory location
+ ;; possible an error in documentation since every tested CPU disagrees with
+ ;; that. Use emms anyway since the vast majority of machines will use the
+ ;; SSE2 variant
+ emms
+%endif
+ RET
%endmacro
INIT_XMM sse