[GHC] #8763: forM_ [1..N] does not get fused (10 times slower than go function)

[GHC]

#8763: forM_ [1..N] does not get fused (10 times slower than go function)
-------------------------------------+-------------------------------------
        Reporter:  nh2               |                Owner:  (none)
            Type:  bug               |               Status:  infoneeded
        Priority:  normal            |            Milestone:  8.6.1
       Component:  Compiler          |              Version:  7.6.3
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Runtime           |  Unknown/Multiple
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #7206             |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by osa1):

 > I have a very similar program where, with 8.4.1, the forM_ version
 allocates 50% more than the version with recursive go but both versions
 run in about the same time . Is it worth giving that code here in this bug
 or should I enter a new bug referencing this one?

 What is the first argument of `forM_` in you program? This ticket (and
 #7206) is specifically about `forM_`/`mapM_` when the foldable argument is
 of form `[n .. m]`. If your program is different than perhaps a new ticket
 would be better. In any case, it's not a big deal if you paste your
 program here, we can move it to a new ticket if it turns out to be
 something different than what we try to improve here.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8763#comment:40>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler