Crash when finalising lazy values of the wrong type. #6664
Comments
|
Comment author: @damiendoligez After thinking about it and discussing with some users, it seems clear that Gc.finalise should raise Invalid_argument when called on a lazy or a float, because these types are not (always) represented by heap blocks. Unfortunately, the run-time test for lazy is not exact: some lazy values (the ones that are already forwarded) cannot be detected as such. This will have to be documented. |
|
Comment author: @damiendoligez fixed in trunk (commit 15876) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Original bug ID: 6664
Reporter: @damiendoligez
Assigned to: @damiendoligez
Status: closed (set by @damiendoligez on 2015-03-04T19:59:26Z)
Resolution: fixed
Priority: normal
Severity: minor
Platform: all
OS: all
OS Version: all
Version: 4.03.0+dev / +beta1
Target version: 4.03.0+dev / +beta1
Fixed in version: 4.03.0+dev / +beta1
Category: runtime system and C interface
Bug description
When you finalise a value of type [int lazy], you get a crash.
The problem is in caml_final_update, at finalise.c:94, short-circuiting the pointer makes a finalised int, which violates the invariants.
I'm not sure what the behavior should be in that case.
Steps to reproduce
With ocamlc, ocamlopt, or the toplevel:
let f x =
let r = ref (lazy (1+x)) in
let fin x = Printf.eprintf "finalisation %d\n%!" (Lazy.force x) in
Gc.finalise fin !r;
ignore (Lazy.force !r);
r := lazy 0;
Gc.full_major ();
;;
f 3;;
The text was updated successfully, but these errors were encountered: