[GHC] #15068: Small number of test case failures on Ubuntu Bionic (GCC 7.3)

[GHC]

#15068: Small number of test case failures on Ubuntu Bionic (GCC 7.3)
-------------------------------------+-------------------------------------
        Reporter:  jrp               |                Owner:  (none)
            Type:  bug               |               Status:  merge
        Priority:  highest           |            Milestone:  8.6.1
       Component:  Compiler          |              Version:  8.4.1
  (CodeGen)                          |
      Resolution:  fixed             |             Keywords:
Operating System:  Linux             |         Architecture:  x86_64
                                     |  (amd64)
 Type of failure:  Incorrect result  |            Test Case:  T13658
  at runtime                         |  T14779a T14779b T14868 debug
                                     |  parsing001
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):  Phab:D4654
       Wiki Page:                    |
-------------------------------------+-------------------------------------
Changes (by svenpanne):

 * milestone:  8.4.3 => 8.6.1


Comment:

 Replying to [comment:29 hvr]:
 > Replying to [comment:27 svenpanne]:
 >
 > Sven, I'm not sure you're seeing the full picture here. The
 (questionable) popularity argument is completely off-topic and irrelevant
 to the issue at hand.

 Why on earth is popularity off-topic? Each and every bug is weighed by its
 impact. Why do we e.g. focus on x86_64 CPUs and not on e.g. ARM64?

 > Releasing a quickfixed GHC 8.4.3 would still undermine the whole idea of
 Stack's "LTS" snapshot releases and its (imperfect) "reproducability"
 concept, as this problem isn't limited to GHC 8.4.* but rather extends
 back to GHC 8.2.1.  As such I see no reason for yet another rushed
 quickfix release of GHC, for something that *must* be fixed on Stack's end
 anyway in order to be effective.

 I fail to see why `stack` should be changed.

 > And it would be quite laughable if Stack wasn't in a better position to
 release a quick fix than for GHC HQ to make a full compiler release (which
 I have to emphasize wouldn't be sufficient anyway for users that are
 forced to use Stack and their GHC 8.[24].[12] snapshots to keep working
 with the affected GCC versions)... ;-)

 It's not laughable at all: You can provoke the same bug with `cabal`, too,
 just pass '-g' to GHC somehow. I somehow doubt that Ubuntu will release a
 new version of binutils soon (they would have to recompile each and every
 package if they take testing/QA seriously), so the only sane way to fix
 this would be in GHC itself.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15068#comment:32>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler