Now, I for one would be -VERY- interested in seeing same system comparsion
between the flags: -Os -O2 and -O3 (on the same -march= ) for a
single system.
You can get a gist of this (not necessarily benchmarks, but stability results) in the CFLags central thread. It took me the better part of 3 or 4 days to read all of the posts before confirming that -O2 was the best for my situation. There were WAY to many situations where -03 created instabililties in a small set of applications (I don't want any instabililties) and the -Os seemed to be the best (its -O2 with size opts as well) but in a handful of situations it caused REALLY obscure crashes with some pretty major apps (like X or Gnome or something I don't quite recall).
Anyway I ended up with march=p3 (on a p4 system) and -O2 -pipe (left out the other one that everyone uses, something about pipeline because I saw a handful of problems that caused with some niche applications that I might have used) and I'm pretty happy with everything. Granted these 3 machines are all servers, but over all great success.
As regards to -ffast-math. It makes a hell of a difference in math related things, but all applications that the developers think are safe and should have it, are already so (mplayer,ffmpeg,xine... check compile logs ;)
This is a very interesting fact that I'm glad to hear! Now I don't need to feel bad about not abusing my make flags.
Its very interesting that you don't remember wether using prelink or
not, nor wether you use the Gentoo kernel's specific grsecurity patch or
not (Which I think will have some, albeit minimal, impact)
another thing to add that a lot of people seem to be confused about is "Gentoo's default optimization"
This is -O2 -pipe
Yes. it is, go look in /etc/make.globals The -O3 are recommendations and examples, genflags should have this even
better theese days.
genflags suggests -O3 on the systems I mention above, even though -O3 isn't promised to be stable... any concerns? Or are the packages that -O3 would typically effect, written to clear the make flags (gcc, binutils, etc) before compiling?
Some more points:
Comparing compilations is perhaps defacto praxis, but generally a useless comparsion as gcc overrules all optimizations (check the sources of gcc or ebuilds yourself) in most of all places. its almost impossible to optimize gcc, and if you want fast compiles, i'd suggest gcc-2.95.3 derivations. (still installable)
-fomit-frame-pointer is generally giving some boost, but nothing I run on my own systems (I debug a lot ;)
Yep that was the one I was thinking about.
//Spider
-- [EMAIL PROTECTED] mailing list
