Hi,

I was looking at the numbers here: http://phoronix.com/forums/showthread.php?32696-CompilerDeathMatch-64bit-Final-results and there are quite stunning results when you make GCC more aggressive. Lets for example take the apache static page serving.

With just the -o2 compile flag you get (in that test) ~2478.45 requests per second. (which seems to be the archlinux default)
However, if you make it more aggresive (very aggressive) then the number nearly doubles to ~4117.43 requests per second!

The flags are: -O3 -ffast-math -march=native -mtune=native -fexcess-precision=fast -ftree-parallelize-loops=5 -floop-parallelize-all -floop-interchange -floop-strip-mine -floop-block -flto

And for completeness:
- o2 benchmarks: http://www.global.phoronix-test-suite.com/index.php?k=profile&u=staalmannen-6171-9881-28731
- custom flags benchmarks: http://global.phoronix-test-suite.com/index.php?k=profile&u=staalmannen-28797-31696-27314

That is from the same user on the same hardware and the same software from the before mentioned phoronix thread.

So, i think adding those flags (or tweaking it even more for more performance) is certainly worth looking at since it can nearly double performance!

Note: i did not test this myself on archlinux to see if performance really doubles with those flags.

Regards,
Mark