I think the program fails in the final GC that runs right before the
#15300: Unboxed Sums Crash -------------------------------------+------------------------------------- Reporter: andrewthad | Owner: (none) Type: task | Status: new Priority: normal | Milestone: 8.6.1 Component: Compiler | Version: 8.4.3 Resolution: | Keywords: UnboxedSums Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by osa1): Thanks for reporting. I managed to reproduce this with GHC HEAD. program
terminates. We can see that it produces a correct result of (Nothing,Just "\rhas\narrived"), but something on the heap has definitely been corrupted.
This program does not do any GC at all. Try this: {{{ $ gdb --args ./eol Reading symbols from ./eol...done.
break GarbageCollect Breakpoint 1 at 0x8b3ca7: file rts/sm/GC.c, line 224. r }}}
you'll see that the breakpoint is never triggered. So this is definitely a code generation bug. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/15300#comment:1 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler