[GHC] #9203: Perf regression in 7.8.2 relative to 7.6.3, possibly related to HashMap

[GHC]

#9203: Perf regression in 7.8.2 relative to 7.6.3, possibly related to HashMap
--------------------------------------------+------------------------------
        Reporter:  simonmar                 |            Owner:
            Type:  bug                      |           Status:  merge
        Priority:  high                     |        Milestone:  7.8.3
       Component:  Compiler                 |          Version:  7.8.2
      Resolution:                           |         Keywords:
Operating System:  Unknown/Multiple         |     Architecture:
 Type of failure:  Runtime performance bug  |  Unknown/Multiple
       Test Case:                           |       Difficulty:  Unknown
        Blocking:                           |       Blocked By:
                                            |  Related Tickets:
--------------------------------------------+------------------------------

Comment (by simonmar):

 Simon: I don't think you added a comment on Phabricator (you probably need
 to click the big button at the bottom of the page).

 I've been digging a bit more into this: it turns out we had fixed this
 once before, see #3245, and there was even a big Note in the code that was
 *removed* by Pedro in 3d53407f9c6e87bcff30d51f0a048e92b2745a32.

 There's also a perf test (`testsuite/tests/perf/should_run/T3245.hs`), but
 that didn't trigger because (a) there are no bounds defined for it in
 all.T, and (b) that test doesn't trigger the bug any more anyway.  I'm
 going to add a new perf test, and some comments to `Internal.hs`.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9203#comment:8>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler