Re: [Haskell-cafe] For one cabal'ised library: running specific benchmarks & tests, and obtaining multiple Travis reports