Hi everyone,<br>I have replayed the "unix benchmark v3.6 dhrystone 2" test.<br><br>You can find the archive of the test here : <a href="http://www.phoronix-test-suite.com/benchmark-files/byte-benchmark-1.tar.gz">http://www.phoronix-test-suite.com/benchmark-files/byte-benchmark-1.tar.gz</a><br>
To play the test on gcc :<br>make<br>./Run dhry2<br>To play the test on llvm-gcc :<br>Replace in Makefile : CC=gcc by CC=llvm-gcc<br>in Run : CC=gcc by CC=llvm-gcc<br>make<br>./Run dhry2<br><br>Some information on the test platform :<br>
<br>cat /etc/lsb-release<br>DISTRIB_ID=Ubuntu<br>DISTRIB_RELEASE=9.04<br>DISTRIB_CODENAME=jaunty<br>DISTRIB_DESCRIPTION="Ubuntu 9.04"<br><br>uname -a<br>Linux zaraki 2.6.28-15-generic #49-Ubuntu SMP Tue Aug 18 18:40:08 UTC 2009 i686 GNU/Linux<br>
<br>/proc/cpuinfo<br>dual core Intel(R) Core(TM)2 CPU X6800 @ 2.93GHz (cache size 4096)<br><br>llvm-gcc -v<br>Using built-in specs.<br>Target: i686-pc-linux-gnu<br>Configured with: ../llvm-gcc4.2-2.5.source/<br>configure --prefix=/mounts/zion/disks/0/localhome/tbrethou/2.5/prerelease2/llvm-gcc/obj/../install --program-prefix=llvm- --enable-llvm=/localhome/tbrethou/2.5/prerelease2/llvm-2.5/ --disable-bootstrap --enable-languages=c,c++,fortran<br>
Thread model: posix<br>gcc version 4.2.1 (Based on Apple Inc. build 5636) (LLVM build 2.5)<br>(Binary from official x86 2.5 release)<br><br>gcc -v<br>Using built-in specs.<br>Target: i486-linux-gnu<br>Configured with: ../src/configure -v --with-pkgversion='Ubuntu 4.3.3-5ubuntu4' --with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --enable-nls --with-gxx-include-dir=/usr/include/c++/4.3 --program-suffix=-4.3 --enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc --enable-mpfr --enable-targets=all --with-tune=generic --enable-checking=release --build=i486-linux-gnu --host=i486-linux-gnu --target=i486-linux-gnu<br>
Thread model: posix<br>gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4)<br><br>The test displays its results in form of a number with unit seems to be "loops per seconds". So it's logical to say that the bigger is better.<br>
I run the test 10 times for each compiler (for an approximate duration of 20 minutes each compiler).<br><br>Results:<br>LLVM average : 10696687.14<br>GCC average : 9563046.99<br>LLVM is on average 10.58% faster than GCC on this test.<br>
<br>If you need more informations or if you think my test was not correctly made... ping me. :)<br><br>Here a quick report on the test. (shows individual result of each test) :<br><br>
<meta http-equiv="CONTENT-TYPE" content="text/html; charset=utf-8">
<title></title>
<meta name="GENERATOR" content="OpenOffice.org 3.0 (Linux)">
<style>
<!--
BODY,DIV,TABLE,THEAD,TBODY,TFOOT,TR,TH,TD,P { font-family:"Nimbus Sans L"; font-size:x-small }
-->
</style>
<table border="0" cellspacing="0" cols="5" rules="none" frame="void">
<colgroup><col width="142"><col width="86"><col width="86"><col width="86"><col width="86"></colgroup>
<tbody>
<tr>
<td align="left" width="142" height="19"><br></td>
<td align="left" width="86">LLVM</td>
<td align="left" width="86">GCC</td>
<td align="left" width="86">Difference</td>
<td align="left" width="86">Difference %</td>
</tr>
<tr>
<td align="left" height="19"><br></td>
<td align="left"><br></td>
<td align="left"><br></td>
<td align="left"><br></td>
<td align="left"><br></td>
</tr>
<tr>
<td align="left" height="19">Run 1</td>
<td align="right">10720897.7</td>
<td align="right">9875431</td>
<td align="right">845466.7</td>
<td align="right">7.89</td>
</tr>
<tr>
<td align="left" height="19">Run 2</td>
<td align="right">10411474.6</td>
<td align="right">9666138.5</td>
<td align="right">745336.1</td>
<td align="right">7.16</td>
</tr>
<tr>
<td align="left" height="19">Run 3</td>
<td align="right">10580705.2</td>
<td align="right">9265904.6</td>
<td align="right">1314800.6</td>
<td align="right">12.43</td>
</tr>
<tr>
<td align="left" height="19">Run 4</td>
<td align="right">10746408</td>
<td align="right">9539499.9</td>
<td align="right">1206908.1</td>
<td align="right">11.23</td>
</tr>
<tr>
<td align="left" height="19">Run 5</td>
<td align="right">10818845.5</td>
<td align="right">9766210.7</td>
<td align="right">1052634.8</td>
<td align="right">9.73</td>
</tr>
<tr>
<td align="left" height="19">Run 6</td>
<td align="right">10904404.3</td>
<td align="right">10034620.8</td>
<td align="right">869783.5</td>
<td align="right">7.98</td>
</tr>
<tr>
<td align="left" height="19">Run 7</td>
<td align="right">10861482.8</td>
<td align="right">9028718.6</td>
<td align="right">1832764.2</td>
<td align="right">16.87</td>
</tr>
<tr>
<td align="left" height="19">Run 8</td>
<td align="right">10480568.5</td>
<td align="right">9568849.4</td>
<td align="right">911719.1</td>
<td align="right">8.7</td>
</tr>
<tr>
<td align="left" height="19">Run 9</td>
<td align="right">10857968.2</td>
<td align="right">9306331.6</td>
<td align="right">1551636.6</td>
<td align="right">14.29</td>
</tr>
<tr>
<td align="left" height="19">Run 10</td>
<td align="right">10584116.6</td>
<td align="right">9578764.8</td>
<td align="right">1005351.8</td>
<td align="right">9.5</td>
</tr>
<tr>
<td align="left" height="19"><br></td>
<td align="left"><br></td>
<td align="left"><br></td>
<td align="left"><br></td>
<td align="left"><br></td>
</tr>
<tr>
<td align="left" height="19">Average</td>
<td align="right">10696687.14</td>
<td align="right">9563046.99</td>
<td align="right">1133640.15</td>
<td align="right">10.58</td>
</tr>
<tr>
<td align="left" height="19">Standard deviation</td>
<td align="right">172714.92</td>
<td align="right">300445.48</td>
<td align="right">345992.26</td>
<td align="right">3.13</td>
</tr>
</tbody>
</table>
<br>Cheers,<br>Olivier.<br><br><br>On Mon, Sep 14, 2009 at 10:02 PM, Keir Mierle <<a href="mailto:mierle@gmail.com">mierle@gmail.com</a>> wrote:<br>><br>> On Mon, Sep 14, 2009 at 12:23 PM, Renato Golin <<a href="mailto:rengolin@systemcall.org">rengolin@systemcall.org</a>> wrote:<br>
>><br>>> 2009/9/14 Edward O'Callaghan <<a href="mailto:eocallaghan@auroraux.org">eocallaghan@auroraux.org</a>>:<br>>> > screw that site, its useless info run by a linux gnu zealot.<br>>><br>
>> Well, being a Linux GNU zealot doesn't invalidate numbers. What does<br>>> invalidate is that he has no numbers!<br>>><br>>> He just fired a few applications and counted the seconds, this is<br>
>> utterly pointless. No reference to how many times he ran, standard<br>>> deviations, possible effects of OS cache, compilation parameters, etc,<br>>> etc. There simply isn't anything in there. It even lack text in<br>
>> between the charts, a conclusion, and... well... everything.<br>>><br>>> "Ignore that site", would also be more polite... ;)<br>><br>> Nevertheless, the site has a large audience. From a publicity perspective, it may be worthwhile to work with the author to get more representative numbers.<br>
> Keir<br>> <br>>><br>>> cheers,<br>>> --renato<br>>><br>>> Reclaim your digital rights, eliminate DRM, learn more at<br>>> <a href="http://www.defectivebydesign.org/what_is_drm">http://www.defectivebydesign.org/what_is_drm</a><br>
>> _______________________________________________<br>>> LLVM Developers mailing list<br>>> <a href="mailto:LLVMdev@cs.uiuc.edu">LLVMdev@cs.uiuc.edu</a> <a href="http://llvm.cs.uiuc.edu">http://llvm.cs.uiuc.edu</a><br>
>> <a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev</a><br>><br>><br>> _______________________________________________<br>> LLVM Developers mailing list<br>
> <a href="mailto:LLVMdev@cs.uiuc.edu">LLVMdev@cs.uiuc.edu</a> <a href="http://llvm.cs.uiuc.edu">http://llvm.cs.uiuc.edu</a><br>> <a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev</a><br>
><br><br>