infix assert panics ghc
I was repeatedly getting this:
ghc-6.8.3: panic! (the 'impossible' happened)
(GHC version 6.8.3 for i386-unknown-linux):
rename/RnExpr.lhs:(139,26)-(150,42): Non-exhaustive patterns in lambda
After some narrowing down, I eventually found the culprit and got a repeatable error with this minimal snippet:
import Control.Exception (assert)
foo = True `assert` ()
The infixness is crucial; assert True () compiles without difficulty. I use infix notation elsewhere, so it appears to be something special about assert.
I was able to compile successfully if I used -O2. I see now that was likely because the asserts are simply eliminated early.
I compiled ghc from sources, so it's barely possible I did something wrong there, but the nature of the symptoms make this doubtful.
Trac metadata
| Trac field | Value |
|---|---|
| Version | 6.8.3 |
| Type | Bug |
| TypeOfFailure | OtherFailure |
| Priority | normal |
| Resolution | Unresolved |
| Component | Compiler |
| Test case | |
| Differential revisions | |
| BlockedBy | |
| Related | |
| Blocking | |
| CC | |
| Operating system | |
| Architecture |