<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: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=utf-8">
<meta name="Generator" content="Microsoft Word 14 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:Helvetica;
panose-1:2 11 6 4 2 2 2 2 2 4;}
@font-face
{font-family:Helvetica;
panose-1:2 11 6 4 2 2 2 2 2 4;}
@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;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
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.MsoListParagraph, li.MsoListParagraph, div.MsoListParagraph
{mso-style-priority:34;
margin-top:0cm;
margin-right:0cm;
margin-bottom:0cm;
margin-left:36.0pt;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman","serif";}
span.apple-tab-span
{mso-style-name:apple-tab-span;}
span.EmailStyle18
{mso-style-type:personal-reply;
font-family:"Calibri","sans-serif";
color:#1F497D;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:72.0pt 72.0pt 72.0pt 72.0pt;}
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-GB" link="blue" vlink="purple">
<div class="WordSection1">
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Hi Steve,<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">Thanks for the explanations and pointers!<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">> Anything I missed?<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">Maybe. Reading your reply and also N1945 again, we got confused (my colleague Simon Tatham helped me here). You seem to be interpreting FLT_EVAL_METHOD=0 as
saying that fp16 operations are done in fp16 precision, and FLT_EVAL_METHOD=32 as saying that fp16 operations are done in fp32 precision.<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">But by our reading of N1945, FLT_EVAL_METHOD=0 and FLT_EVAL_METHOD=32 mean the same thing, at least on a platform where float is 32 bits wide. Each one says
that for some type T, expressions whose semantic type is T or smaller are evaluated in type T; and for FLT_EVAL_METHOD=0, T = float, whereas for FLT_EVAL_METHOD=32, T = _Float32, i.e. the same physical type in both cases.<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">So surely, to indicate that fp16 operations are done in fp16 precision, we would have to set FLT_EVAL_METHOD to 16, not to 0?<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">Cheers,<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Sjoerd.<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 0cm 0cm 0cm">
<p class="MsoNormal"><b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">From:</span></b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif""> cfe-dev [mailto:cfe-dev-bounces@lists.llvm.org]
<b>On Behalf Of </b>Stephen Canon via cfe-dev<br>
<b>Sent:</b> 12 May 2017 00:57<br>
<b>To:</b> Hal Finkel<br>
<b>Cc:</b> nd; clang developer list<br>
<b>Subject:</b> Re: [cfe-dev] [RFC] implementation of _Float16<o:p></o:p></span></p>
</div>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal">On May 11, 2017, at 7:11 PM, Hal Finkel via cfe-dev <<a href="mailto:cfe-dev@lists.llvm.org">cfe-dev@lists.llvm.org</a>> wrote:<o:p></o:p></p>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:"Helvetica","sans-serif"">That's what's been asserted here as well. The question is: If we're going to want a type that represents half precision without the implied extend/truncate operations, do
we a) Introduce a new type that is "really" a half or b) change half not to imply the extend/truncate and then autoupgrade?</span><o:p></o:p></p>
</div>
</blockquote>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">Just to try to try to be precise, I want to broaden this slightly and try to sketch out all the questions around this. Apologies if the answers to these are obvious or you feel like they’re already settled. I’d like to make sure we define
the scope of the decisions pretty clearly before bike shedding it to death =)<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">(a) For targets that do not have fp16 hardware support, what is FLT_EVAL_METHOD (I’m using the C-language bindings here so that there are semi-formal definitions that people can look up, but this is at least partially a non-language specific
policy decision)?<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>- We could choose FLT_EVAL_METHOD = 0, which requires us to “simulate” _Float16 operations by upconverting to a legal type (float), doing the operation in float, and converting back to _Float16
for every operation (this works for all the arithmetic instructions, except fma, which would require a libcall or other special handling, but we would want fma formation from mul + add to still be licensed when allowed by program semantics).<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>- We could choose FLT_EVAL_METHOD = 32, which allows us to maintain extra precision by eliding the conversions to/from _Float16 around each operation (leaving intermediate results in `float`).<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">The second option obviously yields better performance on many targets, but slightly reduces portability; targets without _Float16 support now get different answers than targets that have _Float16 support for basic arithmetic. The second
option matches (I think?) the intended behavior of the arm __fp16 extension.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">(b) For targets that have fp16 hardware support, we still get to choose FLT_EVAL_METHOD.<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>- Use the fp16 hardware. FLT_EVAL_METHOD = 0.<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>- The other choice is FLT_EVAL_METHOD = 32 (matching the existing behavior of __fp16, but making it much harder for people to take advantage of the shiny new instructions—they would have to
use intrinsics—and severely hampering the autovectorizer’s options).<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">It sounds like everyone is settled on the first choice (and I agree with that), but let’s be clear that this *is* a decision that we’re making.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">(c) Assuming FLT_EVAL_METHOD = 0 for targets with fp16 hardware, do we need to support a type with the __fp16 extension semantics of “implicitly promote everything to float” for the purposes of source compatibility?<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">Sounds like “yes”, at least for some toolchains.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">(d) If yes, does that actually require a separate type at LLVM IR layer?<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">I don’t immediately see that it would, but I am not an expert.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">Anything I missed?<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">– Steve<o:p></o:p></p>
</div>
</div>
</body>
</html>