search for: 255_vortex

...0.102 0.113 +11.1% +46mB SingleSource/Benchmarks/Shootout-C++/hash 0.455 0.507 +11.4% +47mB External/Povray/povray 2.015 2.246 +11.5% +47mB External/SPEC/CINT2000/255_vortex/255_vortex 1.814 2.044 +12.7% +52mB SingleSource/Benchmarks/Shootout-C++/heapsort 1.871 2.132 +13.9% +57mB SingleSource/Benchmarks/Shootout-C++/ary3 1.087 1.264 +16.3% +65mB MultiSource/Ben...

...1.1847 1.2550 0.0041 > External/Nurbs/nurbs.compile 5.90% 1.8470 1.9559 0.0087 > MultiSource/Applications/lemon/lemon.compile 5.90% 1.0516 1.1136 0.0027 > MultiSource/Benchmarks/MiBench/consumer-typeset/consumer-typeset.compile 5.88% > 14.2650 15.1031 0.0341 > External/SPEC/CINT2000/255_vortex/255_vortex.compile 5.87% 9.3345 9.8822 0.0331 > MultiSource/Benchmarks/MallocBench/gs/gs.compile 5.84% 4.3921 4.6488 0.0275 > External/SPEC/CINT95/132_ijpeg/132_ijpeg.compile 5.79% 4.8337 5.1137 0.0279 > External/SPEC/CINT2006/400_perlbench/400_perlbench.compile 5.77% 26.7023 > 28.2429...

On Tue, Apr 17, 2012 at 12:23 AM, Jakob Stoklund Olesen <stoklund at 2pi.dk>wrote: > I am not sure how best to fix this. If possible, InstCombine's > canonicalization shouldn't hide arithmetic progressions behind bit masks. The entire concept of cleverly converting arithmetic to bit masks seems like the perfect domain for DAGCombine instead of InstCombine: 1) We know the

...1 >> External/Nurbs/nurbs.compile 5.90% 1.8470 1.9559 0.0087 >> MultiSource/Applications/lemon/lemon.compile 5.90% 1.0516 1.1136 0.0027 >> MultiSource/Benchmarks/MiBench/consumer-typeset/consumer-typeset.compile 5.88% >> 14.2650 15.1031 0.0341 >> External/SPEC/CINT2000/255_vortex/255_vortex.compile 5.87% 9.3345 9.8822 0.0331 >> MultiSource/Benchmarks/MallocBench/gs/gs.compile 5.84% 4.3921 4.6488 0.0275 >> External/SPEC/CINT95/132_ijpeg/132_ijpeg.compile 5.79% 4.8337 5.1137 0.0279 >> External/SPEC/CINT2006/400_perlbench/400_perlbench.compile 5.77% 26.7023 &...

Look at this silly function: $ cat small.c unsigned f(unsigned a, unsigned *p) { unsigned x = a/4; p[0] = x; p[1] = x+x; return p[1] - 2*p[0]; } GCC turns this into straightforward code and figures out the 0 return value: shrl $2, %edi movl %edi, (%rsi) addl %edi, %edi movl %edi, 4(%rsi) movl $0, %eax ret LLVM optimizes the code: $ clang -O -S -o- small.c -emit-llvm define i32

Hi all, I've started looking at the GlobalMerge pass, enabled by default on ARM and AArch64. I think we should reconsider that, at least for AArch64. As is, the pass just merges all globals together, in groups of 4KB (AArch64, 128B on ARM). At the time it was enabled, the general thinking was "it's almost free, it doesn't affect performance much, we might as well use it".

...013 0.0169 > MultiSource/Benchmarks/Prolangs-C/unix-smail/unix-smail.compile > <http://llvm.org/perf/db_default/simple/nts/380/graph?pset.0=on&test.MultiSource%2FBenchmarks%2FProlangs-C%2Funix-smail%2Funix-smail.compile=on> > 1.57% 0.6414 0.6515 0.0012 > External/SPEC/CINT2000/255_vortex/255_vortex.compile > <http://llvm.org/perf/db_default/simple/nts/380/graph?pset.0=on&test.External%2FSPEC%2FCINT2000%2F255_vortex%2F255_vortex.compile=on> > 1.57% 9.2792 9.4247 0.0097 > MultiSource/Benchmarks/Prolangs-C/loader/loader.compile > <http://llvm.org/perf/db_defau...