Currently encountering failing tests in a wasm-tools
branch in the roundtrip test suite. Seems to be a different that is running when the overflow occurs each time. Curious if anybody has encountered something similar before and/or has some intuition for identifying what's responsible?
Specifically hitting thread '<unknown>' has overflowed its stack
fatal runtime error: stack overflow
on roundtrip tests
For this I think the best route to debug would be to open up in a debugger and see what the stack says. It's probably some deeply nested WIT parser and your system may be configured with a slightly smaller stack than CI has or something like that. Either wast
needs to be optimized a bit or a test needs to be scaled back about how eager it is to have recursive definitions
Last updated: Dec 23 2024 at 12:05 UTC