eigen/bench/benchmark_suite

#!/bin/bash
echo "Fixed size 3x3, column-major, -DNDEBUG"
$CXX -O3 -I .. -DNDEBUG benchmark.cpp -o benchmark && time ./benchmark >/dev/null
echo "Fixed size 3x3, column-major, with asserts"
$CXX -O3 -I .. benchmark.cpp -o benchmark && time ./benchmark >/dev/null
echo "Fixed size 3x3, row-major, -DNDEBUG"
$CXX -O3 -I .. -DEIGEN_DEFAULT_TO_ROW_MAJOR -DNDEBUG benchmark.cpp -o benchmark && time ./benchmark >/dev/null
echo "Fixed size 3x3, row-major, with asserts"
$CXX -O3 -I .. -DEIGEN_DEFAULT_TO_ROW_MAJOR benchmark.cpp -o benchmark && time ./benchmark >/dev/null
echo "Dynamic size 20x20, column-major, -DNDEBUG"
$CXX -O3 -I .. -DNDEBUG benchmarkX.cpp -o benchmarkX && time ./benchmarkX >/dev/null
echo "Dynamic size 20x20, column-major, with asserts"
$CXX -O3 -I .. benchmarkX.cpp -o benchmarkX && time ./benchmarkX >/dev/null
echo "Dynamic size 20x20, row-major, -DNDEBUG"
$CXX -O3 -I .. -DEIGEN_DEFAULT_TO_ROW_MAJOR -DNDEBUG benchmarkX.cpp -o benchmarkX && time ./benchmarkX >/dev/null
echo "Dynamic size 20x20, row-major, with asserts"
$CXX -O3 -I .. -DEIGEN_DEFAULT_TO_ROW_MAJOR benchmarkX.cpp -o benchmarkX && time ./benchmarkX >/dev/null
* coefficient wise operators are more generic, with controllable result type. - compatible with current STL's functors as well as with the extention proposal (TR1) * thanks to the above, Cast and ScalarMultiple have been removed * benchmark_suite is more flexible (compiler and matrix size) 2008-03-06 19:36:27 +08:00			`#!/bin/bash`
Make use of the LazyBit, introduce .lazy(), remove lazyProduct. 2008-04-01 00:20:06 +08:00			`echo "Fixed size 3x3, column-major, -DNDEBUG"`
* coefficient wise operators are more generic, with controllable result type. - compatible with current STL's functors as well as with the extention proposal (TR1) * thanks to the above, Cast and ScalarMultiple have been removed * benchmark_suite is more flexible (compiler and matrix size) 2008-03-06 19:36:27 +08:00			`$CXX -O3 -I .. -DNDEBUG benchmark.cpp -o benchmark && time ./benchmark >/dev/null`
Make use of the LazyBit, introduce .lazy(), remove lazyProduct. 2008-04-01 00:20:06 +08:00			`echo "Fixed size 3x3, column-major, with asserts"`
* coefficient wise operators are more generic, with controllable result type. - compatible with current STL's functors as well as with the extention proposal (TR1) * thanks to the above, Cast and ScalarMultiple have been removed * benchmark_suite is more flexible (compiler and matrix size) 2008-03-06 19:36:27 +08:00			`$CXX -O3 -I .. benchmark.cpp -o benchmark && time ./benchmark >/dev/null`
Make use of the LazyBit, introduce .lazy(), remove lazyProduct. 2008-04-01 00:20:06 +08:00			`echo "Fixed size 3x3, row-major, -DNDEBUG"`
* coefficient wise operators are more generic, with controllable result type. - compatible with current STL's functors as well as with the extention proposal (TR1) * thanks to the above, Cast and ScalarMultiple have been removed * benchmark_suite is more flexible (compiler and matrix size) 2008-03-06 19:36:27 +08:00			`$CXX -O3 -I .. -DEIGEN_DEFAULT_TO_ROW_MAJOR -DNDEBUG benchmark.cpp -o benchmark && time ./benchmark >/dev/null`
Make use of the LazyBit, introduce .lazy(), remove lazyProduct. 2008-04-01 00:20:06 +08:00			`echo "Fixed size 3x3, row-major, with asserts"`
* coefficient wise operators are more generic, with controllable result type. - compatible with current STL's functors as well as with the extention proposal (TR1) * thanks to the above, Cast and ScalarMultiple have been removed * benchmark_suite is more flexible (compiler and matrix size) 2008-03-06 19:36:27 +08:00			`$CXX -O3 -I .. -DEIGEN_DEFAULT_TO_ROW_MAJOR benchmark.cpp -o benchmark && time ./benchmark >/dev/null`
Make use of the LazyBit, introduce .lazy(), remove lazyProduct. 2008-04-01 00:20:06 +08:00			`echo "Dynamic size 20x20, column-major, -DNDEBUG"`
* coefficient wise operators are more generic, with controllable result type. - compatible with current STL's functors as well as with the extention proposal (TR1) * thanks to the above, Cast and ScalarMultiple have been removed * benchmark_suite is more flexible (compiler and matrix size) 2008-03-06 19:36:27 +08:00			`$CXX -O3 -I .. -DNDEBUG benchmarkX.cpp -o benchmarkX && time ./benchmarkX >/dev/null`
Make use of the LazyBit, introduce .lazy(), remove lazyProduct. 2008-04-01 00:20:06 +08:00			`echo "Dynamic size 20x20, column-major, with asserts"`
* coefficient wise operators are more generic, with controllable result type. - compatible with current STL's functors as well as with the extention proposal (TR1) * thanks to the above, Cast and ScalarMultiple have been removed * benchmark_suite is more flexible (compiler and matrix size) 2008-03-06 19:36:27 +08:00			`$CXX -O3 -I .. benchmarkX.cpp -o benchmarkX && time ./benchmarkX >/dev/null`
Make use of the LazyBit, introduce .lazy(), remove lazyProduct. 2008-04-01 00:20:06 +08:00			`echo "Dynamic size 20x20, row-major, -DNDEBUG"`
* coefficient wise operators are more generic, with controllable result type. - compatible with current STL's functors as well as with the extention proposal (TR1) * thanks to the above, Cast and ScalarMultiple have been removed * benchmark_suite is more flexible (compiler and matrix size) 2008-03-06 19:36:27 +08:00			`$CXX -O3 -I .. -DEIGEN_DEFAULT_TO_ROW_MAJOR -DNDEBUG benchmarkX.cpp -o benchmarkX && time ./benchmarkX >/dev/null`
Make use of the LazyBit, introduce .lazy(), remove lazyProduct. 2008-04-01 00:20:06 +08:00			`echo "Dynamic size 20x20, row-major, with asserts"`
* coefficient wise operators are more generic, with controllable result type. - compatible with current STL's functors as well as with the extention proposal (TR1) * thanks to the above, Cast and ScalarMultiple have been removed * benchmark_suite is more flexible (compiler and matrix size) 2008-03-06 19:36:27 +08:00			`$CXX -O3 -I .. -DEIGEN_DEFAULT_TO_ROW_MAJOR benchmarkX.cpp -o benchmarkX && time ./benchmarkX >/dev/null`