search for: optnz

Attached are the Polyhedron 2005 benchmark results for current llvm/dragonegg svn on x86_64-apple-darwin11 built against Xcode 4.3.2 and FSF gcc 4.6.3. The benchmarks for -msse3 and -msse4 appear identical (at least for degg+optnz). This is fortunate since there seems to be a bug in -msse4 on 2.33 GHz (T7600) Intel Core 2 Duo Merom (http://llvm.org/bugs/show_bug.cgi?id=12434). I've added two additional entries to the table. The first, degg+novect+optnz, should show the optimizations achieved by -fplugin-arg-dragonegg-ena...

Attached are the Polyhedron 2005 benchmark results for current llvm/dragonegg svn on x86_64-apple-darwin11 built against Xcode 4.3.2 and FSF gcc 4.6.3. The benchmarks for -msse3 and -msse4 appear identical (at least for degg+optnz). This is fortunate since there seems to be a bug in -msse4 on 2.33 GHz (T7600) Intel Core 2 Duo Merom (http://llvm.org/bugs/show_bug.cgi?id=12434). Jack llvm/dragonegg r153877 dragonegg: de-gfortran46 -msse3 -ffast-math -funroll-loops -O3 %n.f90 -o %n degg+vectorize: de-gfort...

...need to pass any additional flags to actually achieve autovectorization via llvm (in absence of -ftree-vectorize and -fplugin-arg-dragonegg-enable-gcc-optzns)? Jack > > Ciao, Duncan. > > The benchmarks >> for -msse3 and -msse4 appear identical (at least for degg+optnz). This is fortunate >> since there seems to be a bug in -msse4 on 2.33 GHz (T7600) Intel Core 2 Duo Merom >> (http://llvm.org/bugs/show_bug.cgi?id=12434). >> Jack >> >> llvm/dragonegg r153877 >> >> dragonegg: >> de-gfortran46 -msse...

...ain-depth=3) will cause a lot more vectorization. -Hal (in > absence of -ftree-vectorize and > -fplugin-arg-dragonegg-enable-gcc-optzns)? Jack > > > > > Ciao, Duncan. > > > > The benchmarks > >> for -msse3 and -msse4 appear identical (at least for degg+optnz). > >> This is fortunate since there seems to be a bug in -msse4 on 2.33 > >> GHz (T7600) Intel Core 2 Duo Merom > >> (http://llvm.org/bugs/show_bug.cgi?id=12434). Jack > >> > >> llvm/dragonegg r153877 > >> > >> dragonegg: > >>...

...results for current llvm/dragonegg svn > on x86_64-apple-darwin11 built against Xcode 4.3.2 and FSF gcc 4.6.3. thanks for the numbers. How does this compare to LLVM 3.0 - were there any regressions? Ciao, Duncan. The benchmarks > for -msse3 and -msse4 appear identical (at least for degg+optnz). This is fortunate > since there seems to be a bug in -msse4 on 2.33 GHz (T7600) Intel Core 2 Duo Merom > (http://llvm.org/bugs/show_bug.cgi?id=12434). > Jack > > llvm/dragonegg r153877 > > dragonegg: > de-gfortran46 -msse3 -ffast-math -funroll-loops -O3...

Below are the results for the Polyhedron 2005 benchmarks compiled with the llvm/compiler-rt/dragonegg 3.3 release and 3.4svn against FSF gcc 4.8.2. The *-stock-de runs omit the -fplugin-arg-dragonegg-enable-gcc-optzns flag and the *-de-optnz runs include it. There seems to be little improvement between llvm 3.3 and 3.4 in the stock case which relies entirely on llvm for vectorization. Duncan, has dragonegg been updated to enable the same llvm vectorization options as clang for the 3.4 release? If not, we may be leaving some potential...

Hi Chris, >> PS: With -fplugin-arg-dragonegg-enable-gcc-optzns the LLVM optimizers are run at >> the following levels: >> >> Command line option LLVM optimizers run at >> ------------------- ---------------------- >> -O1 tiny amount of optimization >> -O2 or -O3 -O1 >> -O4 or -O5

On Oct 8, 2011, at 12:05 PM, Duncan Sands wrote: > PS: With -fplugin-arg-dragonegg-enable-gcc-optzns the LLVM optimizers are run at > the following levels: > > Command line option LLVM optimizers run at > ------------------- ---------------------- > -O1 tiny amount of optimization > -O2 or -O3 -O1 > -O4 or -O5

Hi Jack >               dragonegg degg+vectorize degg+optnz  gfortran > ac               12.45       12.45         8.85       8.80 > gas_dyn          11.72       11.80         4.47       4.26 > induct           24.02       24.91        12.08      13.65 > rnflow           32.25       32.35        26.47      24.06 Any idea what might cause such di...

Hi Anton, >> dragonegg degg+vectorize degg+optnz gfortran >> ac 12.45 12.45 8.85 8.80 >> gas_dyn 11.72 11.80 4.47 4.26 >> induct 24.02 24.91 12.08 13.65 >> rnflow 32.25 32.35 26.47 24.06 > Any idea wha...

...-Hal > > (in > > absence of -ftree-vectorize and > > -fplugin-arg-dragonegg-enable-gcc-optzns)? Jack > > > > > > > > Ciao, Duncan. > > > > > > The benchmarks > > >> for -msse3 and -msse4 appear identical (at least for degg+optnz). > > >> This is fortunate since there seems to be a bug in -msse4 on 2.33 > > >> GHz (T7600) Intel Core 2 Duo Merom > > >> (http://llvm.org/bugs/show_bug.cgi?id=12434). Jack > > >> > > >> llvm/dragonegg r153877 > > >> > &gt...

...flags... -msse4 -ffast-math -funroll-loops -O3 in all cases. The use of -fplugin-arg-dragonegg-enable-gcc-optzns to allow for autovectorization from the FSF gcc front-end only produces a single run-time regression, fatigue, which is PR10892. Run time Benchmark gfortran dragonegg dragonegg+optnz ------------------------------------------------ ac 8.81 10.83 8.89 aermod 18.21 16.77 15.79 air 5.51 7.12 6.66 capacita 32.59 42.30 36.53 channel 1.84 2.52 1.95 doduc 26.78 30.24 27.9...