[GHC] #15832: returnIO/bindIO destroys runtime explicit stack information in some cases
GHC
ghc-devs at haskell.org
Fri Nov 2 02:49:57 UTC 2018
#15832: returnIO/bindIO destroys runtime explicit stack information in some cases
-------------------------------------+-------------------------------------
Reporter: infinity0 | Owner: (none)
Type: bug | Status: new
Priority: normal | Milestone:
Component: Compiler | Version: 8.6.1
Resolution: | Keywords:
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 infinity0):
OK, but then how do I get a proper "evaluated by" stack trace? From
[https://github.com/ghc/ghc/blob/f877d9cc99dd1ba0c038e70527031e9ba0934cd3/rts/Profiling.c#L929
here] it seems that only CAF stacks can be traversed to the "evaluated by"
stacks. Stacks that don't come from a CAF, can't be traversed further to
see what caused the evaluation of an error/bottom value.
In both cases, I am creating an `X` with a bottom inside it. This is a
decently common pattern used in Haskell, the intent is that I never
evaluate the bottom. However, programmers make mistakes and it's nice to
be able to get a stack trace showing me what was trying to evaluate the
bottom.
In the good case, I can see both (a) the stack that created the bottom
value and (b) the stack that caused the evaluation of bottom. In the bad
case, I can only see (a) and I can't see (b), making things very hard to
debug.
--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15832#comment:5>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
More information about the ghc-tickets
mailing list