==1888== Cachegrind, a cache and branch-prediction profiler
==1888== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==1888== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==1888== Command: bash -n benchmarks/testdata/configure-helper.sh
==1888== Parent PID: 1881
==1888== 
--1888-- warning: L3 cache found, using its data for the LL simulation.
==1888== 
==1888== I   refs:      47,119,375
==1888== I1  misses:        12,143
==1888== LLi misses:         2,815
==1888== I1  miss rate:       0.03%
==1888== LLi miss rate:       0.01%
==1888== 
==1888== D   refs:      19,152,716  (12,083,678 rd   + 7,069,038 wr)
==1888== D1  misses:        28,209  (    21,471 rd   +     6,738 wr)
==1888== LLd misses:         8,550  (     4,843 rd   +     3,707 wr)
==1888== D1  miss rate:        0.1% (       0.2%     +       0.1%  )
==1888== LLd miss rate:        0.0% (       0.0%     +       0.1%  )
==1888== 
==1888== LL refs:           40,352  (    33,614 rd   +     6,738 wr)
==1888== LL misses:         11,365  (     7,658 rd   +     3,707 wr)
==1888== LL miss rate:         0.0% (       0.0%     +       0.1%  )