This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug tree-optimization/30032] sqrt(CGAMMA*PRES[j]/DENS[j]) much slower than compiting compiler
- From: "burnus at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 30 Nov 2006 21:14:39 -0000
- Subject: [Bug tree-optimization/30032] sqrt(CGAMMA*PRES[j]/DENS[j]) much slower than compiting compiler
- References: <bug-30032-13404@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #4 from burnus at gcc dot gnu dot org 2006-11-30 21:14 -------
> Also you are using VLAs so maybe ICC also converts those VLAs to normal
> arrays.
Some tests with
void eos(const int NODES, const float CGAMMA, float CS[NODES], const float
PRES[NODES], const float DENS[NODES])
icc -O3 -xW -ipo -static
real: 0m1.879s 0m1.878s
gcc -march=opteron -ffast-math -funroll-loops -ftree-vectorize -msse3 -O3
real: 0m6.478s 0m6.791s
gcc -march=opteron -ffast-math -funroll-loops -ftree-vectorize -msse3 -O3
-mfpmath=sse,387
real: 0m6.907s 0m6.501s
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30032