6 lines (5 loc) · 373 Bytes

Thread Racing

was reviewing concurrency and wanted to make the threads race
first one uses a barrier, second one just interleaves them. avoids the println! lock by acquiring and writing/reading the Rwlock in a snapshot scope. Race3 and 4 are the same as 1/2 but without the snapshot.
barrier approach seems to be the best
def has some memory leaks or smth