This is the mail archive of the
gcc-help@gcc.gnu.org
mailing list for the GCC project.
Re: Difference between two optimization settings
- From: Eljay Love-Jensen <eljay at adobe dot com>
- To: Thomas Heinz <thomasheinz at gmx dot net>, GCC-help <gcc-help at gcc dot gnu dot org>
- Date: Sat, 28 Jun 2008 08:16:13 -0500
- Subject: Re: Difference between two optimization settings
Hi Thomas,
> What is the difference between
>
> -O0
>
> and
>
> -O1 -fno-defer-pop -fno-merge-constants -fno-thread-jumps -fno-loop-optimize
> -fno-if-conversion -fno-if-conversion2 -fno-delayed-branch
> -fno-guess-branch-probability -fno-cprop-registers -fno-omit-frame-pointer
-O0
Performs no optimizations. It disables (bypasses) all the optimization
machinery. Should make compile time faster, and easier to debug (-g), but
may hide certain bad (language non-compliant) bugs at runtime (which may
give a false impression that the code is correct), and won't generate a few
warnings from -Wall -Wextra which depend on optimization analysis.
-O1 -fno-yadayada...
Enables many optimizations, and the mentioned -fno-yadayada... flags will
disable that handful of specific optimizations. (Most of the optimizations
have no independently twiddle-able flags.)
I like this trick to see which twiddle-able flags were enabled:
gcc -O1 -S -fverbose-asm -x c <(echo '') -o O1.s
cat O1.s
(Your command-line may differ, depending on your platform.)
HTH,
--Eljay