[GHC] #14777: panic when using a function defined in terms of `error`

[GHC]

#14777: panic when using a function defined in terms of `error`
-------------------------------------+-------------------------------------
        Reporter:  zilinc            |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.2.2
      Resolution:                    |             Keywords:
Operating System:  Linux             |         Architecture:
 Type of failure:  Compile-time      |  Unknown/Multiple
  crash or panic                     |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by tdammers):

 I'm a bit reluctant to brush this off as "the panic doesn't happen anymore
 on 8.4, so we must have fixed it somehow", because it could just be that
 we're now optimizing differently and the panic doesn't happen in this
 particular case anymore, but still might pop up in other situations.

 So I'm wondering whether we should:

 a) Just write up a suitable regression test, commit that, and see if it
 ever triggers again in the future.

 b) Actively go hunting for the source of this. Upside is that we would
 know whether the bug has been fixed or just masked; downside is that this
 is potentially very time consuming.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14777#comment:25>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler