<div dir="ltr">This came up randomly because of a typo, but rather than respond for them I've changed the thread name to something descriptive and copies the authors of the paper.<div><br></div><div>The small bit that I know:</div>
<div><ul style><li style>It is a new backend that takes Core from GHC and does the rest itself (including having a non-GHC runtime).</li><li style>There are some patches to GHC (somewhere?) to support a new kind of initialization-only vector.</li>
<li style>The IFLC backend uses a strict intermediate representation and performs auto-vectorization on loops. In particular, the paper evaluates the technique on loops from (slightly modified) REPA and vector library code.</li>
<li style>It handles a pretty sizable subset of the language, but is missing some things (e.g. asynchronous exceptions).</li><li>It is not released yet</li></ul></div><div><div class="gmail_extra"><div class="gmail_quote">
On Tue, Jul 2, 2013 at 7:54 PM, Carter Schonwald <span dir="ltr"><<a href="mailto:carter.schonwald@gmail.com" target="_blank">carter.schonwald@gmail.com</a>></span> wrote:<br></div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
What branch is that work in? I've not seen mention of it in the commit logs. I'd really love to see how they're Doing it. <div><br></div><div>Likewise, it's worth remarking that a carefullywritten simd primop that explicitly uses the instruction intrinsics can easily be 1.5-2x faster than auto vectorized simd code. I've some examples I tested for matrix multiplication where the performance is robustly In that range. </div>
<div class=""><div class="h5">
<div><br></div><div><br>On Tuesday, July 2, 2013, Ryan Newton wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<div dir="ltr">Gosh, sorry, yes it's not obvious from the list of accepted papers:<div>
<br></div><div><dt style="line-height:18px;font-family:Arial,Helvetica,sans-serif">"Automatic SIMD Vectorization for Haskell" Leaf Petersen, Dominic Orchard and Neal Glew</dt>
<dt style="line-height:18px;font-family:Arial,Helvetica,sans-serif"><br></dt><dt style="line-height:18px;font-family:Arial,Helvetica,sans-serif">One of the authors has a link for it but it appears there's no preprint up yet:</dt>
</div><div> <a href="http://www.leafpetersen.com/leaf/publications.htm" target="_blank">http://www.leafpetersen.com/leaf/publications.htm</a><br><div><br></div><div><br></div><div><br></div><div><br></div></div></div>
<div class="gmail_extra">
<br><br><div class="gmail_quote">On Tue, Jul 2, 2013 at 5:36 AM, Nicolas Trangez <span dir="ltr"><<a>nicolas@incubaid.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<div>On Mon, 2013-07-01 at 12:05 -0400, Ryan Newton wrote:<br>
> Err, GCC replacement. But, ironically, GHC [backend] replacement as well,<br>
> as of the recent ICFP paper.<br>
<br>
</div>Got a link or reference?<br>
<span><font color="#888888"><br>
Nicolas<br>
</font></span><div><div><br>
<br>
_______________________________________________<br>
ghc-devs mailing list<br>
<a>ghc-devs@haskell.org</a><br>
<a href="http://www.haskell.org/mailman/listinfo/ghc-devs" target="_blank">http://www.haskell.org/mailman/listinfo/ghc-devs</a><br>
</div></div></blockquote></div><br></div>
</blockquote></div>
</div></div></blockquote></div><br></div></div></div>