Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
OK, well this is embarrassing. Despite asserting in the PR message of inits in terms of what it did, apparently I never actually ran with memleaks on to make sure that no memory was leaked when a postinit throws, and... Well, it was. (I had inspected the generated code and seen that we were emitting some code to handle its thrown errors, but didn't look closely enough to see that that handling didn't actually do any clean-up). Sloppy. Here, I'm fixing my issue by tucking the call to postinit() into the same try...catch block as the throwing initializer code is put into, which gives it the same outcome, which is to free the memory. This should resolve the memory leaks failures we saw on Dec 20, 2024. --- Signed-off-by: Brad Chamberlain <[email protected]>
- Loading branch information