[inplace_function] benchmark about sg14

[user706]

Usage:

#!/usr/bin/env bash

git clone https://github.com/user706/CxxFunctionBenchmark.git
cd CxxFunctionBenchmark

mkdir build
cd    build

git checkout feature/sg14_inplace_function
cmake -D GNR=ON   \
      -D FOLLY=ON \
      -D BDE=ON    `# will clone into git submodule CxxFunctionBenchmark/bde/bde` \
      -D BOOST_ROOT=/my/path/to/boost/install/dir `# only needed if boost is in non-standard place`\
      ..
cmake --build .

./various   # run program

[Quuxplusone]

FYI @user706, you should rebase on https://github.com/jamboree/CxxFunctionBenchmark master, which has recently fixed an utter mess on non-case-sensitive filesystems (e.g. OS X and Windows).

EDIT: You should rebase your master branch. Or else change the git clone command-line above to clone directly into your feature branch.

[user706]

branch feature/sg14_inplace_function (https://github.com/user706/CxxFunctionBenchmark/tree/feature/sg14_inplace_function) is on top of the latest commit of jamboree's master (i.e. if there were any fixes by jamboree [as you imply above], they are in the branch feature/sg14_inplace_function as well)

[user706]

jamboree merged it into his repo.

Thus usage:

#!/usr/bin/env bash

git clone https://github.com/jamboree/CxxFunctionBenchmark.git
cd CxxFunctionBenchmark

mkdir build
cd    build

cmake -D GNR=ON   \
      -D FOLLY=ON \
      -D BDE=ON    `# will clone into git submodule CxxFunctionBenchmark/bde/bde` \
      -D BOOST_ROOT=/my/path/to/boost/install/dir `# only needed if boost is in non-standard place`\
      ..
cmake --build .

./various   # run program

What results do you get on your platforms? Is speed a goal? If yes: what kind of goal-> as fast as possible (unsurpassable) vs fast enough (what is fast enough?). Where should the main speed-focus be: construction, function-call or destruction? etc.

How does it correlate with your expectations? etc.