==2580== Cachegrind, a cache and branch-prediction profiler
==2580== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2580== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2580== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/test-cmd-util.sh
==2580== Parent PID: 2573
==2580== 
--2580-- warning: L3 cache found, using its data for the LL simulation.
==2580== 
==2580== I   refs:      111,865,790
==2580== I1  misses:        893,642
==2580== LLi misses:          8,963
==2580== I1  miss rate:        0.80%
==2580== LLi miss rate:        0.01%
==2580== 
==2580== D   refs:       51,872,527  (27,502,219 rd   + 24,370,308 wr)
==2580== D1  misses:        648,159  (   439,161 rd   +    208,998 wr)
==2580== LLd misses:        110,984  (    14,525 rd   +     96,459 wr)
==2580== D1  miss rate:         1.2% (       1.6%     +        0.9%  )
==2580== LLd miss rate:         0.2% (       0.1%     +        0.4%  )
==2580== 
==2580== LL refs:         1,541,801  ( 1,332,803 rd   +    208,998 wr)
==2580== LL misses:         119,947  (    23,488 rd   +     96,459 wr)
==2580== LL miss rate:          0.1% (       0.0%     +        0.4%  )