<div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div class="gmail_default" style="font-family:monospace,monospace;font-size:small">From a quick glance at the code, it looks like both inner loops were unrolled at least once (vary by architecture)<br></div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small"> and have the expected number of multiplications, four and two (if unrolled once), respectively.</div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small"></div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small"><br></div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small"><a href="https://gcc.godbolt.org/z/1ks4Pt">https://gcc.godbolt.org/z/1ks4Pt</a><br></div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small"></div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small"><br></div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small">--</div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small">Momchil Velikov,</div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small">Senior Compiler Engineer</div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small">Arm</div><div class="gmail_default" style="font-family:monospace,monospace;font-size:small"><br></div></div></div></div></div>