search for: 175_vpr

...0.9869 0.0810 > MultiSource/Benchmarks/Fhourstones-3_1/fhourstones3_1.exec 3.88% 2.0729 2.1534 > 0.0010 > External/SPEC/CINT2000/256_bzip2/256_bzip2.exec 2.68% 10.3792 10.6571 0.0138 > External/SPEC/CINT2000/164_gzip/164_gzip.exec 2.64% 13.4866 13.8424 0.0737 > External/SPEC/CINT2000/175_vpr/175_vpr.exec 2.61% 4.5220 4.6399 0.0007 > External/SPEC/CINT2000/197_parser/197_parser.compile 2.21% 3.1692 3.2392 0.0075 > MultiSource/Benchmarks/sim/sim.exec 1.97% 6.1997 6.3219 0.0004 > MultiSource/Applications/sqlite3/sqlite3.exec 1.30% 4.6635 4.7241 0.0028 > MultiSource/Benchmarks/...

...t; MultiSource/Benchmarks/Fhourstones-3_1/fhourstones3_1.exec 3.88% 2.0729 2.1534 >> 0.0010 >> External/SPEC/CINT2000/256_bzip2/256_bzip2.exec 2.68% 10.3792 10.6571 0.0138 >> External/SPEC/CINT2000/164_gzip/164_gzip.exec 2.64% 13.4866 13.8424 0.0737 >> External/SPEC/CINT2000/175_vpr/175_vpr.exec 2.61% 4.5220 4.6399 0.0007 >> External/SPEC/CINT2000/197_parser/197_parser.compile 2.21% 3.1692 3.2392 0.0075 >> MultiSource/Benchmarks/sim/sim.exec 1.97% 6.1997 6.3219 0.0004 >> MultiSource/Applications/sqlite3/sqlite3.exec 1.30% 4.6635 4.7241 0.0028 >> MultiSo...

...53 6.7457 0.0206 > External/SPEC/CFP2006/433_milc/433_milc.compile 6.29% 3.6706 3.9015 0.0293 > External/SPEC/CINT2006/462_libquantum/462_libquantum.compile 6.26% 0.9149 > 0.9722 0.0070 > MultiSource/Applications/spiff/spiff.compile 6.25% 0.8954 0.9514 0.0059 > External/SPEC/CINT2000/175_vpr/175_vpr.compile 6.23% 3.0709 3.2621 0.0103 > External/skidmarks10/skidmarks.compile 6.20% 1.9730 2.0953 0.0087 > MultiSource/Applications/SIBsim4/SIBsim4.compile 6.19% 1.2870 1.3667 0.0026 > External/SPEC/CINT2000/186_crafty/186_crafty.compile 6.19% 4.4605 4.7364 0.0167 > External/SPEC/...

...;> External/SPEC/CFP2006/433_milc/433_milc.compile 6.29% 3.6706 3.9015 0.0293 >> External/SPEC/CINT2006/462_libquantum/462_libquantum.compile 6.26% 0.9149 >> 0.9722 0.0070 >> MultiSource/Applications/spiff/spiff.compile 6.25% 0.8954 0.9514 0.0059 >> External/SPEC/CINT2000/175_vpr/175_vpr.compile 6.23% 3.0709 3.2621 0.0103 >> External/skidmarks10/skidmarks.compile 6.20% 1.9730 2.0953 0.0087 >> MultiSource/Applications/SIBsim4/SIBsim4.compile 6.19% 1.2870 1.3667 0.0026 >> External/SPEC/CINT2000/186_crafty/186_crafty.compile 6.19% 4.4605 4.7364 0.0167 >&gt...

..., and got the following results, > > (It is to measure run-time reduction, and negative is better performance) > > spec.cpu2000.ref.183_equake 33.77% > spec.cpu2000.ref.179_art 13.44% > spec.cpu2000.ref.256_bzip2 7.80% > spec.cpu2000.ref.186_crafty 3.69% > spec.cpu2000.ref.175_vpr 2.96% > spec.cpu2000.ref.176_gcc 1.77% > spec.cpu2000.ref.252_eon 1.77% > spec.cpu2000.ref.254_gap 1.19% > spec.cpu2000.ref.197_parser 1.15% > spec.cpu2000.ref.253_perlbmk 1.11% > spec.cpu2000.ref.300_twolf -1.04% > > So we can see almost all got worse performance. > &g...

Hello everyone, One of Google's summer interns, George Burgess IV, created an implementation of the CFL pointer-aliasing analysis algorithm, and this has now been added to LLVM trunk. Now we should determine whether it is worthwhile adding this to the default optimization pipeline. For ease of testing, I've added the command line option -use-cfl-aa which will cause the CFL analysis to be

...>> (It is to measure run-time reduction, and negative is better >> performance) >> >> spec.cpu2000.ref.183_equake 33.77% >> spec.cpu2000.ref.179_art 13.44% >> spec.cpu2000.ref.256_bzip2 7.80% >> spec.cpu2000.ref.186_crafty 3.69% >> spec.cpu2000.ref.175_vpr 2.96% >> spec.cpu2000.ref.176_gcc 1.77% >> spec.cpu2000.ref.252_eon 1.77% >> spec.cpu2000.ref.254_gap 1.19% >> spec.cpu2000.ref.197_parser 1.15% >> spec.cpu2000.ref.253_perlbmk 1.11% >> spec.cpu2000.ref.300_twolf -1.04% >> >> So we can see almost all...

...the following > results, > > (It is to measure run-time reduction, and negative is better > performance) > > spec.cpu2000.ref.183_equake 33.77% > spec.cpu2000.ref.179_art 13.44% > spec.cpu2000.ref.256_bzip2 7.80% > spec.cpu2000.ref.186_crafty 3.69% > spec.cpu2000.ref.175_vpr 2.96% > spec.cpu2000.ref.176_gcc 1.77% > spec.cpu2000.ref.252_eon 1.77% > spec.cpu2000.ref.254_gap 1.19% > spec.cpu2000.ref.197_parser 1.15% > spec.cpu2000.ref.253_perlbmk 1.11% > spec.cpu2000.ref.300_twolf -1.04% > > So we can see almost all got worse performance. > &g...

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".

...mpile=on> > 2.28% 0.2892 0.2958 0.0008 > SingleSource/Benchmarks/Shootout/matrix.compile > <http://llvm.org/perf/db_default/simple/nts/380/graph?pset.0=on&test.SingleSource%2FBenchmarks%2FShootout%2Fmatrix.compile=on> > 2.28% 0.0790 0.0808 0.0004 > External/SPEC/CINT2000/175_vpr/175_vpr.compile > <http://llvm.org/perf/db_default/simple/nts/380/graph?pset.0=on&test.External%2FSPEC%2FCINT2000%2F175_vpr%2F175_vpr.compile=on> > 2.26% 3.0062 3.0740 0.0051 > SingleSource/UnitTests/2005-05-11-Popcount-ffs-fls.compile > <http://llvm.org/perf/db_default/sim...