[GHC] #13316: Bad inlining cascade leads to slow optimisation

[GHC]

#13316: Bad inlining cascade leads to slow optimisation
-------------------------------------+-------------------------------------
        Reporter:  simonpj           |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.0.1
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by simonpj):

 Here's my work-in-progress
 {{{
         Modified compiler/simplCore/SimplUtils.hs
 diff --git a/compiler/simplCore/SimplUtils.hs
 b/compiler/simplCore/SimplUtils.hs
 index 7deaf5b..2f9e2e9 100644
 --- a/compiler/simplCore/SimplUtils.hs
 +++ b/compiler/simplCore/SimplUtils.hs
 @@ -1200,7 +1200,7 @@ postInlineUnconditionally dflags env top_lvl bndr
 occ_info rhs unfolding
                          -- PRINCIPLE: when we've already simplified an
 expression once,
                          -- make sure that we only inline it if it's
 reasonably small.

 -           && (not in_lam ||
 +           && ( (not in_lam && not (isValueUnfolding unfolding)) ||
                          -- Outside a lambda, we want to be reasonably
 aggressive
                          -- about inlining into multiple branches of case
                          -- e.g. let x = <non-value>
 @@ -1209,7 +1209,7 @@ postInlineUnconditionally dflags env top_lvl bndr
 occ_info rhs unfolding
                          -- the uses in C1, C2 are not 'interesting'
                          -- An example that gets worse if you add int_cxt
 here is 'clausify'

 -                (isCheapUnfolding unfolding && int_cxt))
 +                 (int_cxt && isCheapUnfolding unfolding))
                          -- isCheap => acceptable work duplication; in_lam
 may be true
                          -- int_cxt to prevent us inlining inside a lambda
 without some
                          -- good reason.  See the notes on int_cxt in
 preInlineUnconditionally
 }}}

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13316#comment:2>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler