<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body style="overflow-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;">Paul,<div><br></div><div>Thanks for the correction!</div><div><br></div><div>Note that ISO/IEC 60559 9.2.1 says</div><div><br></div><blockquote style="margin: 0 0 0 40px; border: none; padding: 0px;"><div><div>For some formats under some rounding attributes the rounded magnitude range of <b>atan</b> (<b>atan2</b>) might exceed the unrounded magnitude of π/2 (π). A programmer must then take care to properly handle any anomalous manifold jump that might occur under the inverse operation.</div></div></blockquote><div><br></div><div>prioritizing correct rounding over range bounds. I’m not aware of any place that ISO/IEC 60559 makes an exception to correct rounding for its floating-point operation to preserve a property of the corresponding mathematical operation. </div><div><br></div><div>- Jim</div><div><br id="lineBreakAtBeginningOfMessage"><div><br><blockquote type="cite"><div>On Jul 18, 2025, at 11:36 PM, Paul Zimmermann <Paul.Zimmermann@inria.fr> wrote:</div><br class="Apple-interchange-newline"><div><div><blockquote type="cite">Correctly rounded functions, e.g. sqrt(x) per Annex F, won’t return out-of-range values. C reserves cr_ prefixed names for correctly rounded math functions.<br></blockquote><br>however correct rounding is not always compatible with range constraints,<br>see the example in section 6.4 from https://dl.acm.org/doi/pdf/10.1145/3747840.<br><br>Maybe the CFP group might decide what to do in that case.<br><br>Paul<br></div></div></blockquote></div><br></div></body></html>