On Thu, Feb 16, 2012 at 5:12 PM, Bill Nottingham <nott...@redhat.com> wrote: > Vladimir Makarov (vmaka...@redhat.com) said: >> >Since I was a bit (intentionally) curt and dismissive in my other >> >response in this thread, I'll add some anecdata here. I have actually >> >tried building xserver with clang and running the standard set of >> >microbenchmarks. I found one relevant path where the clang build was >> >~15% faster [1]. Something like 60% of the rest were within ±3%. For >> >everything else clang was uniformly worse by usually about 5%. >> > >> The another usual mistake when people compare speed of GCC and LLVM >> is to use -O2 for the both compilers. But the true is that -O1 of >> GCC is -O2 of LLVM with the point of code generation quality. The >> compiler speed of GCC with -O1 is the same as for LLVM with -O2. >> You can find the latest comparison of LLVM and GCC on >> http://vmakarov.fedorapeople.org/spec/ (see 2011 comparison at the >> bottom of the left frame). > > Speaking of potential magic bullets... is there any reason > we don't enable auto-vectorization by default (with -O3, or with the > assorted -f/-m flags?) > > - Is it not stable enough? > - Does it not take effect often enough? > - Is it not done generically enough that we would run into instruction set > problems? (I belive on x86_64 we can assume SSE2. x86 is a mess, obvs.)
Yes SSE2 is part of the x86_64 abi, but last time I asked Jakub said that it generates larger code which might end up slower due to not being as cache efficient. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
