[GHC] #8133: T7881 fails with hpc

[GHC]

#8133: T7881 fails with hpc
----------------------------------+--------------------------
       Reporter:  jstolarek       |             Owner:
           Type:  bug             |            Status:  new
       Priority:  normal          |         Milestone:
      Component:  Code Coverage   |           Version:  7.7
       Keywords:                  |  Operating System:  Linux
   Architecture:  x86_64 (amd64)  |   Type of failure:  Other
     Difficulty:  Unknown         |         Test Case:
     Blocked By:                  |          Blocking:
Related Tickets:                  |
----------------------------------+--------------------------
 I noticed that T7881 fails with `hpc` way. I investigated a bit and I
 noticed that if I compile the program without hpc it emits 3 warnings
 (which is intended):

 {{{
 [t-jastol at cam-05-unx : /5playpen/t-jastol/ghc-
 validate2/testsuite/tests/numeric/should_compile] '/5playpen/t-jastol/ghc-
 validate2/inplace/bin/ghc-stage2' -fforce-recomp -c T7881.hs -O

 T7881.hs:7:6: Warning: Enumeration is empty

 T7881.hs:25:6: Warning: Enumeration is empty

 T7881.hs:31:6: Warning: Enumeration is empty
 }}}

 But if I add `-fhpc` on the command line the warnings disappear:

 {{{
 [t-jastol at cam-05-unx : /5playpen/t-jastol/ghc-
 validate2/testsuite/tests/numeric/should_compile] '/5playpen/t-jastol/ghc-
 validate2/inplace/bin/ghc-stage2' -fforce-recomp -c T7881.hs -O -fhpc
 [t-jastol at cam-05-unx : /5playpen/t-jastol/ghc-
 validate2/testsuite/tests/numeric/should_compile]
 }}}

 That's the reason why this test fails (mismatch of stderr). I don't know
 whether this is expected behaviour of hpc or not. If the former, then test
 configuration should be fixed, if the latter then hpc itself should be
 fixed.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8133>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler