==2582== Cachegrind, a cache and branch-prediction profiler
==2582== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2582== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2582== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/test-cmd-util.sh
==2582== Parent PID: 2575
==2582== 
--2582-- warning: L3 cache found, using its data for the LL simulation.
==2582== 
==2582== I   refs:      111,865,696
==2582== I1  misses:        879,607
==2582== LLi misses:          8,933
==2582== I1  miss rate:        0.79%
==2582== LLi miss rate:        0.01%
==2582== 
==2582== D   refs:       51,872,293  (27,502,019 rd   + 24,370,274 wr)
==2582== D1  misses:        642,624  (   434,672 rd   +    207,952 wr)
==2582== LLd misses:        110,980  (    14,519 rd   +     96,461 wr)
==2582== D1  miss rate:         1.2% (       1.6%     +        0.9%  )
==2582== LLd miss rate:         0.2% (       0.1%     +        0.4%  )
==2582== 
==2582== LL refs:         1,522,231  ( 1,314,279 rd   +    207,952 wr)
==2582== LL misses:         119,913  (    23,452 rd   +     96,461 wr)
==2582== LL miss rate:          0.1% (       0.0%     +        0.4%  )