Can't use --profile and --benchmark-arguments simultaneously.

[ldcasillas-progreso]

When I run the following command my benchmark suite runs with profiling enabled (except for the issue I reported in #1771):

% stack bench --profile 

When I run this command my benchmark runs without profiling (as it should), and the argument is passed to Critierion to select a subset of the tests:

% stack bench --benchmark-arguments='theo1'

But when I try to combine them both I get an error:

% stack bench --profile --benchmark-arguments='theo1'
tau-sigma-0.5.1: benchmarks
Running 1 benchmarks...
Benchmark bench: RUNNING...
Usage: bench ([-I|--ci CI] [-G|--no-gc] [-L|--time-limit SECS]
             [--resamples COUNT] [--regress RESP:PRED..] [--raw FILE]
             [-o|--output FILE] [--csv FILE] [--junit FILE]
             [-v|--verbosity LEVEL] [-t|--template FILE] [-m|--match MATCH]
             [NAME...] | [-n|--only-run ITERS] [-m|--match MATCH] [NAME...] |
Benchmark bench: ERROR
             [-l|--list] | [--version])

--  While building package tau-sigma-0.5.1 using:
      /Users/luis.casillas/.stack/setup-exe-cache/x86_64-osx/setup-Simple-Cabal-1.22.4.0-ghc-7.10.2 --builddir=.stack-work/dist/x86_64-osx/Cabal-1.22.4.0 bench bench "--benchmark-options=theo1+RTS -p"
    Process exited with code: ExitFailure 1

More info:

% stack --version
Version 1.0.3, Git revision 260ea31e787593fe27d002985f45739d8a04a498 x86_64

Resolver is lts-3.20.

[mgsloan]

Good catch! +RTS options were not being terminated with -RTS. Please reopen if not fixed in HEAD.

[ldcasillas-progreso]

I can confirm that the following command no longer gets an error:

 stack bench --profile --benchmark-arguments='theo1'

It also produces the profile report file, and that profile reports on cost centers from my library component. There's some residual weirdness, but not within the scope of this ticket.

Stack version:

Version 1.0.3, Git revision 3f7b50fe598ec83eda5d393fe5b08a99ce5ac1e9 x86_64