==2593== Cachegrind, a cache and branch-prediction profiler
==2593== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2593== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2593== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/test-cmd-util.sh
==2593== Parent PID: 2586
==2593== 
--2593-- warning: L3 cache found, using its data for the LL simulation.
==2593== 
==2593== I   refs:      111,881,873
==2593== I1  misses:        877,356
==2593== LLi misses:          8,992
==2593== I1  miss rate:        0.78%
==2593== LLi miss rate:        0.01%
==2593== 
==2593== D   refs:       51,879,075  (27,505,808 rd   + 24,373,267 wr)
==2593== D1  misses:        639,934  (   432,821 rd   +    207,113 wr)
==2593== LLd misses:        111,020  (    14,548 rd   +     96,472 wr)
==2593== D1  miss rate:         1.2% (       1.6%     +        0.8%  )
==2593== LLd miss rate:         0.2% (       0.1%     +        0.4%  )
==2593== 
==2593== LL refs:         1,517,290  ( 1,310,177 rd   +    207,113 wr)
==2593== LL misses:         120,012  (    23,540 rd   +     96,472 wr)
==2593== LL miss rate:          0.1% (       0.0%     +        0.4%  )