[GHC] #10711: Defining mapM_ in terms of traverse_ causes substantial blow-up in ByteCodeAsm

[GHC]

#10711: Defining mapM_ in terms of traverse_ causes substantial blow-up in
ByteCodeAsm
-------------------------------------+-------------------------------------
        Reporter:  bgamari           |                   Owner:  bgamari
            Type:  bug               |                  Status:  new
        Priority:  normal            |               Milestone:
       Component:  Compiler          |                 Version:
      Resolution:                    |                Keywords:
Operating System:  Unknown/Multiple  |            Architecture:
 Type of failure:  Compile-time      |  Unknown/Multiple
  performance bug                    |               Test Case:  ghcirun004
      Blocked By:                    |                Blocking:
 Related Tickets:                    |  Differential Revisions:
-------------------------------------+-------------------------------------

Comment (by rwbarton):

 I have to say that I still don't understand exactly why
 {{{
    a1 *> a2
 == (a1 >>= (\_ -> return id)) >>= (\x2 -> a2 >>= (\x3 -> return (x2 x3)))
 }}}
 is more than a constant (say 10 times) slower than `a1 >> a2` for this
 `Assembler` monad.

 Experimentally bgamari's test program does ~n^2^ allocations and takes
 ~n^3^ total time in the Applicative version, while the Monad version runs
 in linear allocations and time.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10711#comment:5>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler