[GHC] #9233: Compiler performance regression

[GHC]

#9233: Compiler performance regression
---------------------------------+------------------------------------
        Reporter:  augustss      |            Owner:
            Type:  bug           |           Status:  new
        Priority:  low           |        Milestone:
       Component:  Compiler      |          Version:  7.8.2
      Resolution:                |         Keywords:
Operating System:  Windows       |     Architecture:  Unknown/Multiple
 Type of failure:  None/Unknown  |       Difficulty:  Unknown
       Test Case:                |       Blocked By:
        Blocking:                |  Related Tickets:
---------------------------------+------------------------------------

Comment (by goldfire):

 Hrm -- I'm responsible for the offending code.

 If `coercionRole` calling `coercionKind` is the hotspot that's taking the
 trouble, that only seems to happen from the `NthCo` case in
 `coercionRole`. It would be straightforward to add a `Role` field to
 `NthCo` to make `coercionRole` on an `NthCo` run in constant time. Would
 that solve this problem? I'm 95% sure that this change would be very
 straightforward. However, it seems that the change you made already
 mitigates the problem significantly.

 Let me know if I can help further!

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9233#comment:11>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler