<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">This is a problem with all floating point folding, not just with these operations. What Matt is proposing is consistent with how we fold other libm intrinsics.<div class=""><br class=""></div><div class="">—Owen<br class=""><div class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Aug 18, 2014, at 1:22 AM, Mueller-Roemer, Johannes Sebastian <<a href="mailto:Johannes.Sebastian.Mueller-Roemer@igd.fraunhofer.de" class="">Johannes.Sebastian.Mueller-Roemer@igd.fraunhofer.de</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div class="WordSection1" style="page: WordSection1; font-family: Helvetica; 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;"><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span lang="EN-US" style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Wouldn’t it be better to use the target’s implementation (if there is one) instead of generically using one option for constant folding? Otherwise target behavior and constant folded behavior would differ, which should be avoided if possible IMO.<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span lang="EN-US" style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> </span></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">--<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Johannes S. Mueller-Roemer, MSc<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Wiss. Mitarbeiter - Interactive Engineering Technologies (IET)<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> </span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Fraunhofer-Institut für Graphische Datenverarbeitung IGD<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Fraunhoferstr. 5 | 64283 Darmstadt | Germany<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Tel +49 6151 155-606 | Fax +49 6151 155-139<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><a href="mailto:johannes.mueller-roemer@igd.fraunhofer.de" style="color: purple; text-decoration: underline;" class="">johannes.mueller-roemer@igd.fraunhofer.de</a> | <span class="Apple-converted-space"> </span><a href="http://www.igd.fraunhofer.de/" style="color: purple; text-decoration: underline;" class="">www.igd.fraunhofer.de</a><o:p class=""></o:p></span></div></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> </span></div><div class=""><div style="border-style: solid none none; border-top-color: rgb(181, 196, 223); border-top-width: 1pt; padding: 3pt 0in 0in;" class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><b class=""><span lang="EN-US" style="font-size: 10pt; font-family: Tahoma, sans-serif;" class="">From:</span></b><span lang="EN-US" style="font-size: 10pt; font-family: Tahoma, sans-serif;" class=""><span class="Apple-converted-space"> </span><a href="mailto:llvmdev-bounces@cs.uiuc.edu" class="">llvmdev-bounces@cs.uiuc.edu</a> [<a href="mailto:llvmdev-bounces@cs.uiuc.edu" class="">mailto:llvmdev-bounces@cs.uiuc.edu</a>]<span class="Apple-converted-space"> </span><b class="">On Behalf Of<span class="Apple-converted-space"> </span></b>Stephen Canon<br class=""><b class="">Sent:</b><span class="Apple-converted-space"> </span>Thursday, August 14, 2014 18:03<br class=""><b class="">To:</b><span class="Apple-converted-space"> </span>Matt Arsenault<br class=""><b class="">Cc:</b><span class="Apple-converted-space"> </span>llvm-commits; LLVM Developers Mailing List<br class=""><b class="">Subject:</b><span class="Apple-converted-space"> </span>Re: [LLVMdev] [PATCH][RFC]: Add fmin/fmax intrinsics<o:p class=""></o:p></span></div></div></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">… actually, now that I’m able double-check this, I’m quite surprised to find that we didn’t define fmax(+0,–0) in IEEE–754, which says [paraphrased]:<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span class="apple-tab-span"> <span class="Apple-converted-space"> </span></span><b class="">minNum</b>(x,y) is x if x < y, y if y < x, and the number if one is a number and the other is NaN. Otherwise, it is either x or y (this means results might differ among implementations).<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">So I think your proposed semantics are perfectly reasonable.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">– Steve<o:p class=""></o:p></div></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div class=""><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On Aug 14, 2014, at 10:55 AM, Steve Canon <<a href="mailto:scanon@apple.com" style="color: purple; text-decoration: underline;" class="">scanon@apple.com</a>> wrote:<o:p class=""></o:p></div></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">I have no position on whether or not these should be added, but if they are they should match the IEEE 754 semantics, which fully specify all of these details.<br class=""><br class="">(Signaling NaNs could still be left unspecified as they're optional in IEEE-754).<br class=""><br class="">- Steve<br class=""><br class="">Sent from my iPhone<br class=""><br class=""><br class=""><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On Aug 13, 2014, at 7:38 PM, Matt Arsenault <<a href="mailto:arsenm2@gmail.com" style="color: purple; text-decoration: underline;" class="">arsenm2@gmail.com</a>> wrote:<br class=""><br class="">Hi,<br class=""><br class="">I’d like to re-propose adding intrinsics for fmin / fmax. These can be used to implement the equivalent libm functions as defined in C99 and OpenCL, which R600 and AArch64 at least have instructions with the same semantics. This is not equivalent to a simple fcmp + select due to its handling of NaNs.<span class="Apple-converted-space"> </span><br class=""><br class="">This has been proposed before, but never delivered (<a href="http://lists.cs.uiuc.edu/pipermail/llvmdev/2012-December/057128.html" style="color: purple; text-decoration: underline;" class="">http://lists.cs.uiuc.edu/pipermail/llvmdev/2012-December/057128.html</a>)<br class=""><br class="">To summarize:<br class="">1. If either operand is a NaN, returns the other operand<br class="">2. If both operands are NaN, returns NaN<br class="">3. If the operands are equal, returns a value that will compare equal to both arguments<br class="">4. In the normal case, returns the smaller / larger operand<br class="">5. Ignore what to do for signaling NaNs, since that’s what the rest of LLVM does currently anyway<br class=""><br class="">- Handling of fmin/fmax (+/- 0.0, +/- 0.0)<br class="">Point 3 is worded as such because this doesn’t seem particularly well specified by any standard I’ve looked at. The most explicit mention of this I’ve found is a footnote in C99 that “Ideally, fmax would be sensitive to the sign of zero, for example fmax(-0.0, 0.0) would return +0; however, implementation in software might be impractical.” It doesn’t really state what the expected behavior is. glibc and OS X’s libc disagree on the (+0, -0) and (-0, +0) cases. To resolve this, the semantics of the intrinsic will be that either will be OK as long as the result compares equal.<br class=""><br class="">For the purposes of constant folding, I’ve tried to follow the literal wording which was most explicit for the expected result from OpenCL (<a href="http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/fmin.html" style="color: purple; text-decoration: underline;" class="">http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/fmin.html</a>) and taking the comparison +/-0.0 < +/-0.0 will fail.<br class=""><br class="">This means the constant folded results will be:<br class=""> fmin(0.0, 0.0) = 0.0<br class=""> fmin(0.0, -0.0) = 0.0<br class=""> fmin(-0.0, 0.0) = -0.0<br class=""> fmin(-0.0, -0.0) = -0.0<br class=""><br class="">Other options would be to always use +0.0, or to be sensitive to the sign and claim -0.0 is less than 0.0.<br class=""><br class=""><0001-Add-fmin-fmax-intrinsics.patch><br class=""><0002-Add-basic-fmin-fmax-instcombines.patch><br class=""><0003-Fold-fmin-fmax-with-infinities.patch><br class=""><0004-Move-fmin-fmax-constant-folding-logic-into-APFloat.patch><o:p class=""></o:p></div></div></blockquote></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><span style="font-family: Helvetica; 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; float: none; display: inline !important;" class="">_______________________________________________</span><br style="font-family: Helvetica; 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;" class=""><span style="font-family: Helvetica; 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; float: none; display: inline !important;" class="">LLVM Developers mailing list</span><br style="font-family: Helvetica; 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;" class=""><a href="mailto:LLVMdev@cs.uiuc.edu" style="color: purple; text-decoration: underline; font-family: Helvetica; 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;" class="">LLVMdev@cs.uiuc.edu</a><span style="font-family: Helvetica; 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; float: none; display: inline !important;" class=""><span class="Apple-converted-space"> </span> </span><a href="http://llvm.cs.uiuc.edu/" style="color: purple; text-decoration: underline; font-family: Helvetica; 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;" class="">http://llvm.cs.uiuc.edu</a><br style="font-family: Helvetica; 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;" class=""><a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev" style="color: purple; text-decoration: underline; font-family: Helvetica; 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;" class="">http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev</a></div></blockquote></div><br class=""></div></div></body></html>