[Haskell-cafe] debugging <<loop>> exception

[Brandon Allbery]

<<loop>> doesn't produce line numbers, since it happens at STG level
while evaluating a thunk: it overwrites the entry point with the
<<loop>> exception before evaluating it, so if the thunk is
immediately re-entered, the <<loop>> is thrown. This is better than
entering a tight loop, but not much easier to debug.

On Sat, Aug 26, 2023 at 9:44 PM Ivan Perez <ivanperezdominguez at gmail.com> wrote:
>
> Sounds like a Heisenbug (turning profiling or debugging on may affect the reproducibility of the bug).
>
> Run the program for a while with profiling enabled, and the functions involved in the infinite loop might be apparent.
>
> Ivan
>
> On Sat, 26 Aug 2023 at 18:33, Dennis Raddle <dennis.raddle at gmail.com> wrote:
>>
>> I'm getting Exception: <<loop>>.
>>
>> It doesn't print the line number. When I load the program into ghci, it still doesn't print the line number. When I trace the program in ghci it never gets to the point of emitting the exception.
>>
>> I set up my program in stack. I tried both "stack ghci" and just "ghci".. the latter to make sure it wasn't using any compiled code (at least in my own code). If it doesn't print a line number does this mean it's happening in compiled code?
>>
>> Any ideas how I can go about debugging this?
>>
>> _______________________________________________
>> Haskell-Cafe mailing list
>> To (un)subscribe, modify options or view archives go to:
>> http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe
>> Only members subscribed via the mailman list are allowed to post.
>
> _______________________________________________
> Haskell-Cafe mailing list
> To (un)subscribe, modify options or view archives go to:
> http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe
> Only members subscribed via the mailman list are allowed to post.



-- 
brandon s allbery kf8nh
allbery.b at gmail.com