<div dir="ltr">(Sorry for the delay, slipped off of my email radar)<br><br>Sure, happy to take a look at them.<div><br></div><div>-eric</div></div><br><div class="gmail_quote">On Mon, Feb 9, 2015 at 3:36 PM reed kotler <<a href="mailto:rkotler@mips.com">rkotler@mips.com</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
  
    
  
  <div bgcolor="#FFFFFF" text="#000000">
    <div>You want me to submit the patches for
      these changes?<br>
      <br>
      It would be nice to clean up this module.<br>
      <br>
      There are probably dejagnu tests worth porting to llvm test suite
      for this too.<br>
      <br>
      My manager gave me approval to work on it.</div></div><div bgcolor="#FFFFFF" text="#000000"><div><br>
      <br>
      On 02/08/2015 11:02 AM, Eric Christopher wrote:<br>
    </div></div><div bgcolor="#FFFFFF" text="#000000">
    <blockquote type="cite">
      
      <div dir="ltr"><br>
        <br>
        <div class="gmail_quote">On Sun Feb 08 2015 at 9:57:05 AM Reid
          Kleckner <<a href="mailto:rnk@google.com" target="_blank">rnk@google.com</a>> wrote:<br>
          <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
            <div dir="ltr">The question isn't about the Intel _mm_*
              intrinsics, it's about the C math builtins we provide
              under __builtin_pow, etc. I think most of them are mapped
              to their libc calls for convenience. Some have been added
              on an ad-hoc basis as needed.
              <div><br>
              </div>
            </div>
          </blockquote>
          <div><br>
          </div>
          <div>Right. Same general principle :)</div>
          <div> </div>
          <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
            <div dir="ltr">
              <div>I believe that GCC lowers most calls to
                __builtin_sqrt and such to instructions before library
                calls, so we can go ahead and map most of those down to
                LLVM intrinsics without worrying about things like
                errno. It's worth checking as you go, though, rather
                than blindly mapping everything to LLVM intrinsics.</div>
            </div>
            <div class="gmail_extra"><br>
            </div>
          </blockquote>
          <div><br>
          </div>
          <div>Agreed.</div>
          <div><br>
          </div>
          <div>-eric</div>
          <div> </div>
          <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
            <div class="gmail_extra">
              <div class="gmail_quote">On Sat, Feb 7, 2015 at 9:29 PM,
                Eric Christopher <span dir="ltr"><<a href="mailto:echristo@gmail.com" target="_blank">echristo@gmail.com</a>></span>
                wrote:<br>
                <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
                  <div dir="ltr">The idea is that the builtins are
                    wrapped by portable intrinsics (e.g. _mm_* on x86)
                    and therefore those should be used as an interface
                    in programming. If there's no suitable intrinsic
                    then we'll expose the builtin.<span><font color="#888888"><br>
                        <div><br>
                        </div>
                        <div>-eric</div>
                      </font></span></div>
                  <div>
                    <div><br>
                      <div class="gmail_quote">On Fri Feb 06 2015 at
                        6:34:35 PM reed kotler <<a href="mailto:rkotler@mips.com" target="_blank">rkotler@mips.com</a>>
                        wrote:<br>
                        <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Why is that many of
                          the gcc builtins are not mapped to the llvm ir<br>
                          builtins, in cases<br>
                          where there is a corresponding llvm ir
                          builtin/intrinsic.<br>
                          <br>
                          For example, sin, cos, log, log2, exp,
                          exp2.....??<br>
                          <br>
                          Of the normal math functions is seems that
                          just pow and fabs are there.<br>
                          <br>
                          Reed<br>
                          _______________________________________________<br>
                          cfe-dev mailing list<br>
                          <a href="mailto:cfe-dev@cs.uiuc.edu" target="_blank">cfe-dev@cs.uiuc.edu</a><br>
                          <a href="http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev" target="_blank">http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev</a><br>
                        </blockquote>
                      </div>
                    </div>
                  </div>
                  <br>
                  _______________________________________________<br>
                  cfe-dev mailing list<br>
                  <a href="mailto:cfe-dev@cs.uiuc.edu" target="_blank">cfe-dev@cs.uiuc.edu</a><br>
                  <a href="http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev" target="_blank">http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev</a><br>
                  <br>
                </blockquote>
              </div>
              <br>
            </div>
          </blockquote>
        </div>
      </div>
    </blockquote>
    <br>
  </div></blockquote></div>