[GHC] #9989: GHCI is slow for precompiled code

[GHC]

#9989: GHCI is slow for precompiled code
-------------------------------------+-------------------------------------
        Reporter:  remdezx           |                   Owner:
            Type:  bug               |                  Status:  new
        Priority:  normal            |               Milestone:
       Component:  GHC API           |                 Version:  7.8.3
      Resolution:                    |                Keywords:
Operating System:  Linux             |            Architecture:
 Type of failure:  Runtime           |  Unknown/Multiple
  performance bug                    |               Test Case:
      Blocked By:                    |                Blocking:
 Related Tickets:                    |  Differential Revisions:
-------------------------------------+-------------------------------------

Comment (by remdezx):

 Yes, it is still slower when it is run on CPU.

 When I change `import Data.Array.Accelerate.Interpreter as C` to `import
 Data.Array.Accelerate.CUDA as C` compiled code for matrix size `10000`
 executes `0.007s` and precompiled and run under ghci is `0.035s` (5x
 slower). For `1000000` matrix size execution times are `0.57s` (compiled)
 vs `0.90s` (precompiled).

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9989#comment:4>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler