<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
After reading <a moz-do-not-send="true"
href="https://webkit.org/blog/5852/introducing-the-b3-jit-compiler/"
class="">https://webkit.org/blog/5852/introducing-the-b3-jit-compiler/</a>.,
I jotted down a couple of thoughts of my own here:
<a class="moz-txt-link-freetext" href="http://www.philipreames.com/Blog/2016/02/15/quick-thoughts-on-webkits-b3/">http://www.philipreames.com/Blog/2016/02/15/quick-thoughts-on-webkits-b3/</a><br>
<br>
Philip<br>
<br>
<div class="moz-cite-prefix">On 02/15/2016 03:12 PM, Andrew Trick
via llvm-dev wrote:<br>
</div>
<blockquote
cite="mid:7D3EEF0B-8563-4E26-8CEE-B38C3B6CB2EA@apple.com"
type="cite">
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<br class="">
<div>
<blockquote type="cite" class="">
<div class="">On Feb 9, 2016, at 9:55 AM, Rafael Espíndola via
llvm-dev <<a moz-do-not-send="true"
href="mailto:llvm-dev@lists.llvm.org" class="">llvm-dev@lists.llvm.org</a>>
wrote:</div>
<div class="">
<p dir="ltr" class="">
><br class="">
> JavaScriptCore's [FTL JIT](<a moz-do-not-send="true"
href="https://trac.webkit.org/wiki/FTLJIT" class="">https://trac.webkit.org/wiki/FTLJIT</a>)
is moving away<br class="">
> from using LLVM as its backend, towards [B3 (Bare
Bones<br class="">
> Backend)](<a moz-do-not-send="true"
href="https://webkit.org/docs/b3/" class="">https://webkit.org/docs/b3/</a>).
This includes its own [SSA<br class="">
> IR](<a moz-do-not-send="true"
href="https://webkit.org/docs/b3/intermediate-representation.html"
class="">https://webkit.org/docs/b3/intermediate-</a><a
moz-do-not-send="true"
href="https://webkit.org/docs/b3/intermediate-representation.html"
class="">representation.html</a>),<br class="">
> optimisations, and instruction selection backend.</p>
<p dir="ltr" class="">In the end, what was the main
motivation for creating a new IR?</p>
</div>
</blockquote>
</div>
<div class="">I can't speak to the motivation of the WebKit team.
Those are outlined in <a moz-do-not-send="true"
href="https://webkit.org/blog/5852/introducing-the-b3-jit-compiler/"
class="">https://webkit.org/blog/5852/introducing-the-b3-jit-compiler/</a>.</div>
<div class="">I'll give you my personal perspective on using LLVM
for JITs, which may be interesting to the LLVM community.</div>
<div class=""><br class="">
</div>
<div class="">Most of the payoff for high level languages comes
from the language-specific optimizer. It was simpler for
JavaScriptCore to perform loop optimization at that level, so it
doesn't even make use of LLVM's most powerful optimizations,
particularly SCEV based optimization. There is a relatively
small, finite amount of low-level optimization that is going to
be important for JavaScript benchmarks (most of InstCombine is
not relevant).</div>
<div class=""><br class="">
</div>
<div class="">SelectionDAG ISEL's compile time makes it a very
poor choice for a JIT. We never put the effort into making x86
FastISEL competitive for WebKit's needs. The focus now is on
Global ISEL, but that won't be ready for a while.</div>
<div class=""><br class="">
</div>
<div class="">Even when LLVM's compile time problems are largely
solved, and I believe they can be, there will always be systemic
compile time and memory overhead from design decisions that
achieve generality, flexibility, and layering. These are
software engineering tradeoffs.</div>
<div class=""><br class="">
</div>
<div class="">It is possible to design an extremely lightweight
SSA IR that works well in a carefully controlled, fixed
optimization pipeline. You then benefit from basic SSA
optimizations, which are not hard to write. You end up working
with an IR of arrays, where identifiers are indicies into the
array. It's a different way of writing passes, but very
efficient. It's probably worth it for WebKit, but not LLVM.</div>
<div class=""><br class="">
</div>
<div class="">LLVM's patchpoints and stackmaps features are
critical for managed runtimes. However, directly supporting
these features in a custom IR is simply more convenient. It
takes more time to make design changes to LLVM IR vs. a custom
IR. For example, LLVM does not yet support TBAA on calls, which
would be very useful for optimizating around patchpoints and
runtime calls.</div>
<div class=""><br class="">
</div>
<div class="">Prior to FTL, JavaScriptCore had no dependence on
the LLVM project. Maintaining a dependence on an external
project naturally has integration overhead.</div>
<div class=""><br class="">
</div>
<div class="">So, while LLVM is not the perfect JIT IR, it is very
useful for JIT developers who want a quick solution for
low-level optimization and retargetable codegen. WebKit FTL was
a great example of using it to bootstrap a higher tier JIT.</div>
<div class=""><br class="">
</div>
<div class="">To that end, I think it is important for LLVM to
have a well-supported -Ojit pipeline (compile fast) with the
right set of passes for higher-level languages (e.g. Tail
Duplication).</div>
<div class=""><br class="">
</div>
<div class="">-Andy</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
LLVM Developers mailing list
<a class="moz-txt-link-abbreviated" href="mailto:llvm-dev@lists.llvm.org">llvm-dev@lists.llvm.org</a>
<a class="moz-txt-link-freetext" href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev">http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a>
</pre>
</blockquote>
<br>
</body>
</html>