There's no special syntax or functionality to do this. So you have to do it yourself. And it depends whether you want the nth non-overlapping, or the nth overlapping match. For instance, the line "hahahaha" has 2 non-overlapping matches of "haha", but three overlapping. For non overlapping, this may work for you:

($str =~ /hai/g)[3];   # Fourth match
[download]

For overlapping, try:

($str =~ /(?=(hai))/g)[3]; # Fourth match
[download]