[GHC] #4012: Compilation results are not deterministic

GHC ghc-devs at haskell.org
Wed Jun 3 12:00:59 UTC 2015


#4012: Compilation results are not deterministic
-------------------------------------+-------------------------------------
        Reporter:  kili              |                   Owner:
            Type:  bug               |                  Status:  new
        Priority:  high              |               Milestone:  7.12.1
       Component:  Compiler          |                 Version:  6.12.2
      Resolution:                    |                Keywords:
Operating System:  Unknown/Multiple  |            Architecture:
 Type of failure:  Other             |  Unknown/Multiple
      Blocked By:                    |               Test Case:
 Related Tickets:                    |                Blocking:
                                     |  Differential Revisions:
-------------------------------------+-------------------------------------

Comment (by thoughtpolice):

 Also, just a random tidbit: the (excellent!) test in comment:76 can in
 fact be solved by simply compiling with `-fno-cse` - the nondeterministic
 ordering of bindings is the culprit in that case, and CSE chooses `return`
 or `pure` based on what shirt you're wearing.

 If determinism here is hurting NixOS, it may be preferrable to simply
 switch off CSE by default or always default to `-fno-cse` for every
 compilation. That sucks, but in the mean time it may help while we work
 towards a real fix (having a deterministic ordering for Uniques).

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4012#comment:95>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list