shll $1, %ecx # count * 2
.p2align 4
-.loop1:
+.Loop1:
movq (%eax), %mm0
pfmul (%edx), %mm0
pfadd %mm0, %mm4
addl $8, %edx
addl $8, %eax
decl %ecx
- jne .loop1
+ jne .Loop1
# at this point mm4 contains partial sums
.Lfe1:
.size sse_float_dotprod,.Lfe1-sse_float_dotprod
.ident "Hand coded x86 3DNow! assembly"
+
+#if defined(__linux__) && defined(__ELF__)
+.section .note.GNU-stack,"",%progbits
+#endif