First up, thanks a ton for making me aware of "onThreadDestruction", I was very
much not aware and am a huge fan of more explicit hooks like that. I'll include
that going forward for sure.
Next up, regarding the ThreadDispatcher: > Looking at the log you provided in
the GitHub issue, the internal per-thread dispatcher (which is a ref object) is
what causes the leak, as .threadvar's aren't destroyed on thread exit.
The problem here is that I seemingly can't clean this up properly. Trying to
nil it will... okay I just tried that attempt out again for like the 5th time
or so, and found the solution.
Where previously it lead to problems over and over again, I believe a
combination of your hint (that cleanupThread must be in a different scope than
the rest of the code) and some fiddling with the proc itself provided the
solution.
Literally doing this change:
proc cleanupThread*() {.gcsafe, raises: [].}=
{.cast(gcsafe).}:
try:
`=destroy`(getThreadDispatcher())
when defined(gcOrc):
GC_fullCollect() # from orc.nim. Has no destructor.
except Exception as e:
echo "Exception during cleanup: ", e.repr
Run
Fixed the leak problems I had.
