<div><div dir="auto">I meant if the command line says “-mvaes -mno-aes” we should make sure to disable vaes </div><br><div class="gmail_quote"><div>On Tue, Dec 26, 2017 at 9:47 AM coby via Phabricator <<a href="mailto:reviews@reviews.llvm.org">reviews@reviews.llvm.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">coby added inline comments.<br>
<br>
<br>
================<br>
Comment at: lib/Basic/Targets/X86.cpp:573<br>
     setMMXLevel(Features, AMD3DNowAthlon, Enabled);<br>
   } else if (Name == "aes") {<br>
     if (Enabled)<br>
----------------<br>
craig.topper wrote:<br>
> Shouldn't -aes imply -vaes?<br>
how come? perhaps i'm missing here something?<br>
why would the first imply the latter?<br>
following this road an atom z8XXX should be capable of supporting vaes, for example (where it lacks avx, for example, <a href="https://www.intel.com/content/www/us/en/processors/atom/atom-z8000-datasheet-vol-1.html" rel="noreferrer" target="_blank">https://www.intel.com/content/www/us/en/processors/atom/atom-z8000-datasheet-vol-1.html</a>)<br>
also, in that sense, pclmul is implying vpclmulqdq<br>
<br>
<br>
Repository:<br>
  rC Clang<br>
<br>
<a href="https://reviews.llvm.org/D41583" rel="noreferrer" target="_blank">https://reviews.llvm.org/D41583</a><br>
<br>
<br>
<br>
</blockquote></div></div><div dir="ltr">-- <br></div><div class="gmail_signature" data-smartmail="gmail_signature">~Craig</div>