[GHC] #8945: GHC produces grouped declarations in a weird order

[GHC]

#8945: GHC produces grouped declarations in a weird order
-------------------------------------+------------------------------------
        Reporter:  Fuuzetsu          |            Owner:
            Type:  bug               |           Status:  patch
        Priority:  normal            |        Milestone:
       Component:  GHC API           |          Version:  7.9
      Resolution:                    |         Keywords:
Operating System:  Unknown/Multiple  |     Architecture:  Unknown/Multiple
 Type of failure:  None/Unknown      |       Difficulty:  Unknown
       Test Case:                    |       Blocked By:
        Blocking:                    |  Related Tickets:
-------------------------------------+------------------------------------
Changes (by haasn):

 * status:  new => patch


Comment:

 Fix located at
 https://github.com/haasn/ghc/commit/0b51dcae17cc283372304022db2873968a0c6059

 Note: I'm not sure how to handle this submodule stuff etc. correctly.
 Applying this commit will make the Haddock test suite fail, unless we also
 apply
 https://github.com/haasn/haddock/commit/bbcf5d385d6b8682884558500e17fdcddf8099af

 Would it be correct to push bbcf5d38 to Haddock's master, and then update
 GHC's master to include 0b51dcae which also changes the commit Haddock
 points to at the same time, to make sure the test suite is always
 consistent with the change?

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8945#comment:3>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler