Difference in performance of compiled accelerate code ran from ghci and shell
I investigated accelerate and accelerate-cuda and put some debug code to measure a time both under ghci and in a compiled, optimised version. Results are below, you can see stack trace and execution times. ghci run $ ghc -O2 -dynamic -c -threaded Main.hs && ghci GHCi, version 7.8.3: http://www.haskell.org/ghc/ 😕 for help … Loading package … Read more