Hey,<div><br></div><div>I was looking at the reverse-complement benchmark on the Language Shootout, and among other things, I noticed that the Haskell implementation was using (filter (/= &#39;\n&#39;)) on ByteStrings, and also using lists as queues.  </div>

<div><br></div><div>I had a few improvements which using -fasm seem to yield about a 19% improvement in speed, and a 35% reduction in allocation, on my computer.  (If both programs are compiled with -fllvm -- I&#39;m not sure whether or not that&#39;s fair game on the Shootout -- my implementation is 35% faster, and does 10% less allocation.)  I&#39;ve checked my code on the Shootout&#39;s test input, as well.</div>

<div><br></div><div>Mostly, the improvement comes from a tightly specialized version of (filter (/= &#39;\n&#39;)), although eliminating an intermediate list entirely (and one used in a queuelike fashion) didn&#39;t seem to hurt.  I managed to cut the program to a point where the program size is about the same as before.</div>

<div><br></div><div>The code is at <a href="http://hpaste.org/fastcgi/hpaste.fcgi/view?id=25865">http://hpaste.org/fastcgi/hpaste.fcgi/view?id=25865</a>; the previous implementation is at <a href="http://shootout.alioth.debian.org/u32/program.php?test=revcomp&amp;lang=ghc&amp;id=2">http://shootout.alioth.debian.org/u32/program.php?test=revcomp&amp;lang=ghc&amp;id=2</a>.</div>

<div><br></div><div>Let the arguing begin?</div><meta http-equiv="content-type" content="text/html; charset=utf-8"><div><br clear="all">Louis Wasserman<br><a href="mailto:wasserman.louis@gmail.com">wasserman.louis@gmail.com</a><br>

<a href="http://profiles.google.com/wasserman.louis">http://profiles.google.com/wasserman.louis</a><br>
</div>