Little C-benchmarks of gcc-XXX: Compilation timings of make for each GCC's version.

[jc-nospam at jr-pizarro dot jazztel dot es]

I did continue since yesterday with the same options/flags & architecture ...

    http://gcc.gnu.org/ml/gcc/2004-06/msg00045.html

 and i'd added the -O3 option (even if it was not very important).

Timings of make from gzip-1.2.4: (they are the real's times)
--------------------------------
gcc2  -Os : 0m1.184s
gcc33 -Os : 0m1.897s
gcc34 -Os : 0m1.877s # +58.5% slower than gcc2
gcc35 -Os : 0m2.454s # +30.7% slower than gcc34 (+107.2% slower than gcc2)
---------- ---------
gcc2  -O2 : 0m1.373s
gcc33 -O2 : 0m2.240s
gcc34 -O2 : 0m2.208s # +60.8% slower than gcc2
gcc35 -O2 : 0m2.836s # +28.4% slower than gcc34 (+106.5% slower than gcc2)
---------- ---------
gcc2  -O3 : 0m1.420s
gcc33 -O3 : 0m2.436s
gcc34 -O3 : 0m2.572s # +81.1% slower than gcc2
gcc35 -O3 : 0m3.270s # +27.1% slower than gcc34 (+130.2% slower than gcc2)
---------- ---------

Timings of make from bzip2-1.0.2: (they are the real's times)
---------------------------------
gcc2  -Os : 0m3.282s
gcc33 -Os : 0m4.754s
gcc34 -Os : 0m5.018s # +52.8% slower than gcc2
gcc35 -Os : 0m5.290s #  +5.4% slower than gcc34 (+61.1% slower than gcc2)
---------- ---------
gcc2  -O2 : 0m3.343s
gcc33 -O2 : 0m5.036s
gcc34 -O2 : 0m5.744s # +71.8% slower than gcc2
gcc35 -O2 : 0m8.043s # +40.0% slower than gcc34 (+140.5% slower than gcc2)
---------- ---------
gcc2  -O3 : 0m3.662s
gcc33 -O3 : 0m5.741s
gcc34 -O3 : 0m6.871s # +87.6% slower than gcc2
gcc35 -O3 : 0m9.626s # +40.0% slower than gcc34 (+162.8% slower than gcc2)
---------- ---------

Later, i'd tested gcc3X with the option -ftime-report, but i didn't understand
 very well the meanings of theirs results, so i did forgot it because it was
 not very important for my benchmarks's results.

                      Pizarro (c) 2004-June-02 (licensed it as GNU's FDL).