Segfaults
This post is part 5 of the arduino-scripting series:
Today I did a little more work on my programming language TEHSSL. And in testing it, I discovered some anomalies.
The first one is that – because of a typo on my part – testing TEHSSL will occasionally segfault when it tries to print the type of the “next” object after freeing the last one – there is no “next object”, so it dereferences a null pointer and crashes. I fixed that, and it appears to work. However, it worked just fine when I pasted it into cpp.sh, and didn’t hang or anything, so I’m not sure what’s up with that.
Valgrind reports no memory leaks from the garpage collector after I fixed things (woohoo!), so that’s good. However, in testing those bugs, I discovered another anomaly: when the tests de-allocate and free everything at the end, it prints out how many objects it freed. This count varies based on the compiler – without me changing any code at all!!
When I run make test (i.e. running it natively on my computer, using g++), the output shows 56 objects freed at the end.
When I paste it into cpp.sh (which uses Emscripten), it says there are 51 objects freed.
What happened to the other 5 objects? Is it something to do with Emscripten vs g++? I’m not sure. But it has no memory leaks, and that’s what I am really happy about.
Related Posts
- A Hash-Mapped Mess
- The Lesser of Two Evils
- Order Up
- Continuations and the thunk queue
- Error: out of memory