Comment by duskwuff
2 years ago
There's a better workaround, incidentally, which is to map the same memory to two different address ranges with different access permissions. This allows code in the JIT region to be updated while threads may be running in it.
An even better workaround is to make that switch faster.