==2524== Cachegrind, a cache and branch-prediction profiler
==2524== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2524== Using Valgrind-3.19.0 and LibVEX; rerun with -h for copyright info
==2524== Command: bash -n benchmarks/testdata/Build.sh
==2524== Parent PID: 2517
==2524== 
--2524-- warning: L3 cache found, using its data for the LL simulation.
==2524== 
==2524== I   refs:      19,403,577
==2524== I1  misses:        12,188
==2524== LLi misses:         3,180
==2524== I1  miss rate:       0.06%
==2524== LLi miss rate:       0.02%
==2524== 
==2524== D   refs:       6,699,147  (4,701,766 rd   + 1,997,381 wr)
==2524== D1  misses:        17,897  (   12,824 rd   +     5,073 wr)
==2524== LLd misses:         7,416  (    4,199 rd   +     3,217 wr)
==2524== D1  miss rate:        0.3% (      0.3%     +       0.3%  )
==2524== LLd miss rate:        0.1% (      0.1%     +       0.2%  )
==2524== 
==2524== LL refs:           30,085  (   25,012 rd   +     5,073 wr)
==2524== LL misses:         10,596  (    7,379 rd   +     3,217 wr)
==2524== LL miss rate:         0.0% (      0.0%     +       0.2%  )