<div dir="ltr"><div>I'm using the bindists for some versions, and the ~simonmar/fp/ installs (which I've assumed are bindist) for the ones that already existed there (ie 7.0.1 to 7.4.1, excepting 7.0.4).</div><div>
</div><div><font face="courier new,monospace">$ md5sum ghc*bz2</font></div><div><font face="courier new,monospace">0fc26b4c1d10ff9dc8b0cbe9453d48d3 ghc-7.0.4-x86_64-unknown-linux.tar.bz2<br>de67ecfe619b0126d8a8b93d26f34555 ghc-7.4.2-x86_64-unknown-linux.tar.bz2<br>
23da3285c5f8fe6716e2795d149c6b96 ghc-7.6.1-x86_64-unknown-linux.tar.bz2<br>230270a985c522af939d9c71aa76343f ghc-7.6.2-x86_64-unknown-linux.tar.bz2</font></div><div> </div></div><div class="gmail_extra"><br><br><div class="gmail_quote">
On Tue, Feb 12, 2013 at 11:09 AM, Simon Marlow <span dir="ltr"><<a href="mailto:marlowsd@gmail.com" target="_blank">marlowsd@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">On 12/02/13 03:17, Johan Tibell wrote:<br>
</div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid"><div class="im">
Hi Nicolas!<br>
<br>
I tried to reproduce the difference between 7.0.4 and 7.6.2 on the<br>
exp3_8, wheel-sieve1, and primes and couldn't get the same percent<br>
difference as you. We need to reconcile these differences somehow. Lets<br>
start with more exact machine specs. I have a:<br>
<br></div><div class="im">
------------------------------<u></u>------------------------------<u></u>--------------------<br>
Program Size Allocs Runtime Elapsed TotalMem<br>
------------------------------<u></u>------------------------------<u></u>--------------------<br>
bernouilli +3.3% +0.2% 0.12 0.13 +0.0%<br>
exp3_8 +1.1% +53.7% 0.14 0.14 +300.0%<br>
gen_regexps +18.7% +3.9% 0.00 0.00 +0.0%<br>
integrate -0.1% +39.0% 0.21 0.23 +0.0%<br>
kahan +1.7% +98.6% +9.9% +7.3% +0.0%<br>
paraffins +1.3% -1.2% 0.06 0.08 +0.0%<br>
primes +1.4% +64.7% 0.04 0.05 +50.0%<br>
queens +0.8% -0.5% 0.02 0.02 +0.0%<br>
rfib +1.7% +42.8% 0.02 0.02 +0.0%<br>
tak +0.9% +12.0% 0.01 0.01 +0.0%<br>
wheel-sieve1 +0.8% +66.6% -4.6% -5.8% -12.5%<br>
wheel-sieve2 +0.9% +0.0% 0.12 0.13 +0.0%<br>
x2n1 +10.3% +87.3% 0.00 0.01 +200.0%<br>
------------------------------<u></u>------------------------------<u></u>--------------------<br>
Min -0.1% -1.2% -4.6% -5.8% -12.5%<br>
Max +18.7% +98.6% +9.9% +7.3% +300.0%<br>
Geometric Mean +3.2% +31.7% +2.4% +0.5% +23.6%<br>
</div></blockquote>
<br>
Some of these benchmarks essentially do no allocation in their inner loops (x2nl, rfib, tak), so differences there just indicate changes in the IO library or elsewhere, and aren't significant.<br>
<br>
Is your 7.6.2 from our binary distributions, or did you build it yourself?<br>
<br>
Cheers,<br>
Simon<br>
<br>
</blockquote></div><br></div>