==3214== Cachegrind, a cache and branch-prediction profiler
==3214== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==3214== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==3214== Command: mksh -n benchmarks/testdata/functions
==3214== Parent PID: 3209
==3214== 
--3214-- warning: L3 cache found, using its data for the LL simulation.
==3214== 
==3214== I   refs:      14,776,029
==3214== I1  misses:        32,126
==3214== LLi misses:         2,108
==3214== I1  miss rate:       0.22%
==3214== LLi miss rate:       0.01%
==3214== 
==3214== D   refs:       5,427,439  (3,412,490 rd   + 2,014,949 wr)
==3214== D1  misses:        21,029  (   15,113 rd   +     5,916 wr)
==3214== LLd misses:         5,635  (    2,780 rd   +     2,855 wr)
==3214== D1  miss rate:        0.4% (      0.4%     +       0.3%  )
==3214== LLd miss rate:        0.1% (      0.1%     +       0.1%  )
==3214== 
==3214== LL refs:           53,155  (   47,239 rd   +     5,916 wr)
==3214== LL misses:          7,743  (    4,888 rd   +     2,855 wr)
==3214== LL miss rate:         0.0% (      0.0%     +       0.1%  )