<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:x="urn:schemas-microsoft-com:office:excel" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"><head><meta http-equiv=Content-Type content="text/html; charset=us-ascii"><meta name=Generator content="Microsoft Word 14 (filtered medium)"><base href="x-msg://321/"><style><!--
/* Font Definitions */
@font-face
{font-family:Helvetica;
panose-1:2 11 6 4 2 2 2 2 2 4;}
@font-face
{font-family:SimSun;
panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
{font-family:SimSun;
panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:Tahoma;
panose-1:2 11 6 4 3 5 4 4 2 4;}
@font-face
{font-family:"\@SimSun";
panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
p.MsoAcetate, li.MsoAcetate, div.MsoAcetate
{mso-style-priority:99;
mso-style-link:"Balloon Text Char";
margin:0in;
margin-bottom:.0001pt;
font-size:8.0pt;
font-family:"Tahoma","sans-serif";}
span.apple-tab-span
{mso-style-name:apple-tab-span;}
span.apple-converted-space
{mso-style-name:apple-converted-space;}
span.EmailStyle19
{mso-style-type:personal-reply;
font-family:"Calibri","sans-serif";
color:#1F497D;}
span.BalloonTextChar
{mso-style-name:"Balloon Text Char";
mso-style-priority:99;
mso-style-link:"Balloon Text";
font-family:"Tahoma","sans-serif";}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]--></head><body lang=EN-US link=blue vlink=purple><div class=WordSection1><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>I agree. They are different from floor, round, etc. <o:p></o:p></span></p><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>For the transformation, we can limit the condition to both the input and output be single FP and guard them with fast-math.<o:p></o:p></span></p><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'><o:p> </o:p></span></p><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'><o:p> </o:p></span></p><div><div style='border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in 0in 0in'><p class=MsoNormal><b><span style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span></b><span style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'> Stephen Canon [mailto:scanon@apple.com] <br><b>Sent:</b> Wednesday, August 15, 2012 1:49 PM<br><b>To:</b> Weiming Zhao<br><b>Cc:</b> llvm-commits@cs.uiuc.edu<br><b>Subject:</b> Re: [llvm-commits] Fix Bug 13574: Add more double float shrinking optimizations<o:p></o:p></span></p></div></div><p class=MsoNormal><o:p> </o:p></p><div><p class=MsoNormal>We want functions with integral-valued results (floor, ceil, round, rint, nearbyint, trunc) to get these transformations universally; because the result is always exactly representable, this transformation does not change the resulting values.<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal>For functions that can return inexact results (cos, sin, pow, exp, log, etc), this should be defeatable. There are platforms with "skanky" math libraries that may have 10+ ulps of error in these single-precision routines; using the double-precision versions instead is a common idiom for developers to get better quality results.<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal>The transform that I'm especially concerned about, is where the result is not casted back to float:<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal><span class=apple-tab-span> </span>func(floatval) --> (double)funcf(floatval)<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal>this should simply not done for the inexact operations (it discards 29 bits of precision!), but is fine for the functions that produce integral results.<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal>Someone who knows the guts here may have a better idea, but it seems to me that we should have two distinct optimizations (one for floor/ceil/round/etc, and one for cos/sin/pow/etc).<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal>- Steve<o:p></o:p></p></div><p class=MsoNormal><o:p> </o:p></p><div><div><p class=MsoNormal>On Aug 15, 2012, at 4:37 PM, Weiming Zhao <<a href="mailto:weimingz@codeaurora.org">weimingz@codeaurora.org</a>> wrote:<o:p></o:p></p></div><p class=MsoNormal><br><br><o:p></o:p></p><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Hi Steve,</span><o:p></o:p></p></div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'> </span><o:p></o:p></p></div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Yes, it transforms (float)func((double) val) to funcf(v), which saves 2 conversions and possibly cheaper math function.</span><o:p></o:p></p></div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'> </span><o:p></o:p></p></div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>GCC O1 does the same transformation. For LLVM, we can guard the transformations by flags like fast-math or a separate flag.</span><o:p></o:p></p></div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Furthermore, such transformation can be disabled by default. So it also solves the performance issue of double/single FP on some platforms.</span><o:p></o:p></p></div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'> </span><o:p></o:p></p></div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Thanks,</span><o:p></o:p></p></div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'>Weiming</span><o:p></o:p></p></div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'> </span><o:p></o:p></p></div><div><div style='border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in 0in 0in'><div><p class=MsoNormal><b><span style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span></b><span class=apple-converted-space><span style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'> </span></span><span style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'>Stephen Canon [mailto:scanon@<a href="http://apple.com"><span style='color:purple'>apple.com</span></a>]<span class=apple-converted-space> </span><br><b>Sent:</b><span class=apple-converted-space> </span>Wednesday, August 15, 2012 11:54 AM<br><b>To:</b><span class=apple-converted-space> </span>Weiming Zhao<br><b>Cc:</b><span class=apple-converted-space> </span><a href="mailto:llvm-commits@cs.uiuc.edu"><span style='color:purple'>llvm-commits@cs.uiuc.edu</span></a><br><b>Subject:</b><span class=apple-converted-space> </span>Re: [llvm-commits] Fix Bug 13574: Add more double float shrinking optimizations</span><o:p></o:p></p></div></div></div><div><p class=MsoNormal> <o:p></o:p></p></div><div><div><p class=MsoNormal>Do I understand correctly that this optimization would transform cos((double)floatval) --> (double)cosf(floatval)? That's not a acceptable transformation, numerically.<o:p></o:p></p></div></div><div><div><p class=MsoNormal> <o:p></o:p></p></div></div><div><div><p class=MsoNormal>One should also consider that there are a number of platforms supported by LLVM (not naming names) on which the float variants of libm functions are significantly *slower* than the double-precision variants.<o:p></o:p></p></div></div><div><div><p class=MsoNormal> <o:p></o:p></p></div></div><div><div><p class=MsoNormal>- Steve<o:p></o:p></p></div></div><div><p class=MsoNormal> <o:p></o:p></p></div><div><div><div><p class=MsoNormal>On Aug 15, 2012, at 2:33 PM, Weiming Zhao <<a href="mailto:weimingz@codeaurora.org"><span style='color:purple'>weimingz@codeaurora.org</span></a>> wrote:<o:p></o:p></p></div></div><div><p class=MsoNormal><br><br><br><o:p></o:p></p></div><div><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'>Hi,</span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'> </span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'>This patch fixes<span class=apple-converted-space> </span><a href="http://llvm.org/bugs/show_bug.cgi?id=13574"><span style='color:purple'>http://llvm.org/bugs/show_bug.cgi?id=13574</span></a></span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'> </span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:9.0pt;font-family:"Arial","sans-serif"'>Current LLVM only supports 5 double->float shrinking optimizations (floor->floorf, ceil->ceilf,round->roundf, rint->rintf() and nearbyint->nearbyintf).</span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:9.0pt;font-family:"Arial","sans-serif"'> </span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:9.0pt;font-family:"Arial","sans-serif"'>This patch adds more math functions, for example, pow->powf, sin->sinf, cos->cosf, etc.</span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:9.0pt;font-family:"Arial","sans-serif"'> </span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:9.0pt;font-family:"Arial","sans-serif"'>In this patch:</span><o:p></o:p></p></div></div><div style='margin-left:.5in'><div><p class=MsoNormal style='text-indent:-.25in'><span style='font-size:9.0pt;font-family:"Arial","sans-serif"'>1.</span><span style='font-size:7.0pt'> <span class=apple-converted-space> </span></span><span style='font-size:9.0pt;font-family:"Arial","sans-serif"'>Since cos(), pow() and exp2() already have their own optimizers: CosPot, PowOpt and Exp2Opt, respectively, I let them inherit from UnaryDoubleFPOpt and in their CallOptimizer(), I let them to call the base class’s CallOptimizer() first.</span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:9.0pt;font-family:"Arial","sans-serif"'>For this change, I have to hoist the definition of UnaryDoubleFPOpt ahead of CosOpt.</span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:9.0pt;font-family:"Arial","sans-serif"'> </span><o:p></o:p></p></div></div><div style='margin-left:.5in'><div><p class=MsoNormal style='text-indent:-.25in'><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'>2.</span><span style='font-size:7.0pt'> <span class=apple-converted-space> </span></span><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'>Since some functions are not available on Windows, I updated TargetLibraryInfo to mark them as unavailable.</span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'> </span><o:p></o:p></p></div></div><div style='margin-left:.5in'><div><p class=MsoNormal style='text-indent:-.25in'><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'>3.</span><span style='font-size:7.0pt'> <span class=apple-converted-space> </span></span><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'>A unit test case is updated to test this patch.</span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'> </span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'>Please help to review them.</span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'> </span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'>Thanks,</span><o:p></o:p></p></div></div><div><div><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif"'>Weiming</span><o:p></o:p></p></div></div><div><p class=MsoNormal><span style='font-size:13.5pt;font-family:"Helvetica","sans-serif"'><0001-Bug-13574-More-double-float-shrinking-optimizations-f.patch>_______________________________________________<br>llvm-commits mailing list<br><a href="mailto:llvm-commits@cs.uiuc.edu"><span style='color:purple'>llvm-commits@cs.uiuc.edu</span></a><br><a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits"><span style='color:purple'>http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits</span></a></span><o:p></o:p></p></div></div></div><div><p class=MsoNormal> <o:p></o:p></p></div></div></div><p class=MsoNormal><o:p> </o:p></p></div></body></html>