<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;">On Sep 11, 2013, at 3:29 PM, Hal Finkel <<a href="mailto:hfinkel@anl.gov">hfinkel@anl.gov</a>> wrote:<br><div><br class="Apple-interchange-newline"><blockquote type="cite"><div style="font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: 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;">----- Original Message -----<br><blockquote type="cite">----- Original Message -----<br><blockquote type="cite">Hal —<br><br>Am I understanding this correctly:<br><br><span class="Apple-tab-span" style="white-space: pre;"> </span>vector4float x;<br><span class="Apple-tab-span" style="white-space: pre;"> </span>vector4int16 y = (vector4int16)x; // this is a conversion<br><span class="Apple-tab-span" style="white-space: pre;"> </span>vector4int32 z = (vector4int32)x; // this is a bitcast<br><br>This seems confusing in the extreme.<br></blockquote></blockquote><br>On the other hand, I don't really like the implied int-to-float bitcast semantics. Maybe it would be better to just add a warning for this case. What do you think?</div></blockquote><br></div><div>bitcast semantics for vector casts are an essential tool for SIMD programming. Essentially all warnings generated would be false positives.</div><div><br></div><div>– Steve</div></body></html>