New UTF-8 stride function

[Kiith-Sa]

You don't need any instrumentation for perf and can get similar 
output (even in real time). (Don't intend to start a profiler 
war, but recommend looking at perf before messings with DLLs and 
the like) (although perf is is Linux-only)
I think the Windows version of AMD CodeAnalyst might have similar 
features, again without instrumentation, but never tried it 
(CodeAnalyst is gratis). Intel VTune definitely does, but it's 
not gratis.