<html><head><meta http-equiv="Content-Type" content="text/html charset=windows-1252"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;"><div>Hi Tyler, </div><div><br></div><div>Try adding -ffast-math. We can only vectorize reduction variables if it is safe to reorder floating point operations. </div><div><br></div><div>Thanks,</div><div>Nadav</div><div><br></div><div><div>On Apr 3, 2013, at 10:29 AM, "Nowicki, Tyler" <<a href="mailto:tyler.nowicki@intel.com">tyler.nowicki@intel.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div lang="EN-US" link="blue" vlink="purple" style="letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><div class="WordSection1" style="page: WordSection1;"><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;">Hi,<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><o:p> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;">I have a question about LoopVectorize. I wrote a simple test case, a dot product loop and found that packed instructions are generated when input arrays are integer, but not when they are float or double.<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><o:p> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;">If I modify the float example in<span class="Apple-converted-space"> </span><a href="http://llvm.org/docs/Vectorizers.html" style="color: purple; text-decoration: underline;">http://llvm.org/docs/Vectorizers.html</a><span class="Apple-converted-space"> </span>by adding restrict to the input arrays packed instructions are generated. Although it should not be required I tried doing the same to the input arrays of my dot product example but it still doesn’t generate packed float or double instructions.<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><o:p> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;">Is the loop vectorizer supposed to generate packed float and double instructions? Is this a bug, or am I doing something wrong?<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><o:p> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;">Tyler<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><o:p> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;">float dotproduct(float *A, float *B, int n) {<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> float sum = 0;<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> for(int i = 0; i < n; ++i) {<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> sum += A[i] * B[i];<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> }<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> return sum;<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;">}<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><o:p> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;">clang dotproduct.cpp -O3 -fvectorize -march=atom -S -o -<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><o:p> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><loop body><o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;">.LBB1_1:<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> movss (%rdi), %xmm1<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> addq $4, %rdi<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> mulss (%rsi), %xmm1<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> addq $4, %rsi<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> decl %edx<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> addss %xmm1, %xmm0<o:p></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"> jne .LBB1_1</div></div></div></blockquote></div><br></body></html>