==2528== Cachegrind, a cache and branch-prediction profiler
==2528== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2528== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2528== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/Build.sh
==2528== Parent PID: 2521
==2528== 
--2528-- warning: L3 cache found, using its data for the LL simulation.
==2528== 
==2528== I   refs:      37,382,956
==2528== I1  misses:       324,440
==2528== LLi misses:         8,534
==2528== I1  miss rate:       0.87%
==2528== LLi miss rate:       0.02%
==2528== 
==2528== D   refs:      17,238,867  (9,207,598 rd   + 8,031,269 wr)
==2528== D1  misses:       227,021  (  143,304 rd   +    83,717 wr)
==2528== LLd misses:        54,787  (   13,319 rd   +    41,468 wr)
==2528== D1  miss rate:        1.3% (      1.6%     +       1.0%  )
==2528== LLd miss rate:        0.3% (      0.1%     +       0.5%  )
==2528== 
==2528== LL refs:          551,461  (  467,744 rd   +    83,717 wr)
==2528== LL misses:         63,321  (   21,853 rd   +    41,468 wr)
==2528== LL miss rate:         0.1% (      0.0%     +       0.5%  )