Oops, try gcc -v on the various compilers, not gcc --version... See below, you can see how differently my v4.1.1 versus v4.3.0 are (disregard 32-bit vs. 64-bit). Also, I can compile a binary -O2 -g and the two different compilers and get different results. It's not unheard of. Again, it begs the question of source code! -brian 20_ gcc -v Using built-in specs. Target: i386-redhat-linux Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux Thread model: posix gcc version 4.1.1 20070105 (Red Hat 4.1.1-51) And 44_ gcc -v Using built-in specs. Target: x86_64-redhat-linux Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-bugurl=http://bugzilla.redhat.com/bugzilla --enable-bootstrap --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --enable-plugin --with-java-home=/usr/lib/jvm/java-1.5.0-gcj-1.5.0.0/jre --enable-libgcj-multifile --enable-java-maintainer-mode --with-ecj-jar=/usr/share/java/eclipse-ecj.jar --disable-libjava-multilib --with-cpu=generic --build=x86_64-redhat-linux Thread model: posix gcc version 4.3.0 20080428 (Red Hat 4.3.0-8) (GCC) -----Original Message----- From: Łukasz Lew [mailto:lukasz.lew@xxxxxxxxx] Sent: Sunday, September 21, 2008 2:41 PM To: Brian D. McGrew Cc: gcc-help@xxxxxxxxxxx Subject: Re: Performance problem Thanks for a fast answer, I use -O3 -march=native -fomit-frame-pointer -DDEBUG -ffast-math -frename-registers -ggdb3 -Wall -Wextra -Wswitch-enum but I have the same effect with only -O3 Versions: g++-4.2 (GCC) 4.2.4 (Ubuntu 4.2.4-1ubuntu1) g++-4.3 (Ubuntu 4.3.2-1ubuntu7) 4.3.2 The original library is here: http://www.mimuw.edu.pl/~lew/hg/libego/?shortlog I extracted only the benchmark part: http://www.mimuw.edu.pl/~lew/libego_benchmark.tgz I also found strange one-line difference that degrades performance of g++-4.2 compiled binary to the level of g++-4.3. I don't get it at all .. If you have any insights please let me know, it is very important to me. Lukasz On Sun, Sep 21, 2008 at 23:24, Brian D. McGrew <brian@xxxxxxxxxxxxx> wrote: > Source code would be a good start??? Optimization flags??? Architecture flags??? How was gcc-4.3 vs. gcc-4.2 built??? (g++ --version will show this). > > -brian > > -----Original Message----- > From: gcc-help-owner@xxxxxxxxxxx [mailto:gcc-help-owner@xxxxxxxxxxx] On Behalf Of Lukasz Lew > Sent: Sunday, September 21, 2008 2:20 PM > To: gcc-help@xxxxxxxxxxx > Subject: Performance problem > > Hi > > I have a program that when compiled with g++-4.3 is 10% slower than with g++-4.2 > I tried to look at the asm output, but its way too different. > How should I proceed to get my 10% back? > > Thanks > Lukasz >
