[GHC] #13193: Integer (gmp) performance regression?

[GHC]

#13193: Integer (gmp) performance regression?
-------------------------------------+-------------------------------------
        Reporter:  j.waldmann        |                Owner:
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.0.1
      Resolution:                    |             Keywords:  newcomer
Operating System:  Linux             |         Architecture:  x86_64
 Type of failure:  Runtime           |  (amd64)
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by rwbarton):

 Replying to [comment:9 simonpj]:
 > Thanks for working on this.  If you think that HEAD is compiling code
 that is sub-optimal, can you boil it out to an example?   But if it is
 just that the original Haskell forces GHC to generate bad code, then yes
 let's just improve the implementation of `minusInteger`.
 >
 > I couldn't quite figure out which is the case from the notes above.

 Please disregard the second bullet point of comment:6. I was comparing
 compiled code from 7.10 with the source code from HEAD, but there was a
 change that I didn't know about in the source between those versions.

 My hope is that this change (Phab:D2278 from this past May) will bring
 HEAD's performance back in line with 7.8's, but I don't have a suitable
 setup to test that at the moment. If it does so then we can close this
 ticket (optionally: after landing Phab:D3034, improving the implementation
 of `minusInteger` slightly).

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13193#comment:10>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler