#10712: Regression: make TEST=exceptionsrun001 WAY=optasm is failing -------------------------------------+------------------------------------- Reporter: thomie | Owner: Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler | Version: 7.11 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: | base/tests/exceptionsrun001 Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D1616 Wiki Page: | -------------------------------------+------------------------------------- Comment (by simonpj): Are you sure? exceptionsrun001 seems OK to me. It turns out that the other three all relied on imprecise exceptions being precise. Eg conc012 has {{{ forkIO $ Control.Exception.catch (x `seq` putMVar result Finished) $ }}} If you want to be sure that `x` will only be evaluated under the `catch` you must use `evaluate` thus: {{{ forkIO $ Control.Exception.catch (evaluate x >> putMVar result Finished) $ }}} Similarly conc014 had {{{ error "wibble" `Control.Exception.catch` ... }}} But the `error "wibble"` is a pure bottom value and we make no guarantees about when it is evaluated. It should be more like {{{ throwIO (ErrorCall "wibble") `Control.Exception.catch` .... }}} which raises the exception in the IO monad (i.e. precisely). Same with `T3279`, which had {{{ error "foo" `catch` \(SomeException e) -> do }}} Again that `error "foo"` should be `throwIO (ErrorCall "foo")`. Incidentally, is there a function for `throwIO . ErrorCall`? I'll make a patch for these three tests. Then I claim we are done. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/10712#comment:17 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler