As for the other (non FFI) crash I reported about, it turned out to be my mistake - just non-synchronized access to a file stream from several threads. Therefore only the original FFI-callback crash remains. Best regards, - Anton