==2620== Cachegrind, a cache and branch-prediction profiler
==2620== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2620== Using Valgrind-3.19.0 and LibVEX; rerun with -h for copyright info
==2620== Command: bash -n benchmarks/testdata/ltmain.sh
==2620== Parent PID: 2613
==2620== 
--2620-- warning: L3 cache found, using its data for the LL simulation.
==2620== 
==2620== I   refs:      80,225,385
==2620== I1  misses:        33,346
==2620== LLi misses:         3,153
==2620== I1  miss rate:       0.04%
==2620== LLi miss rate:       0.00%
==2620== 
==2620== D   refs:      27,266,152  (19,288,157 rd   + 7,977,995 wr)
==2620== D1  misses:        76,685  (    51,632 rd   +    25,053 wr)
==2620== LLd misses:        22,415  (     4,207 rd   +    18,208 wr)
==2620== D1  miss rate:        0.3% (       0.3%     +       0.3%  )
==2620== LLd miss rate:        0.1% (       0.0%     +       0.2%  )
==2620== 
==2620== LL refs:          110,031  (    84,978 rd   +    25,053 wr)
==2620== LL misses:         25,568  (     7,360 rd   +    18,208 wr)
==2620== LL miss rate:         0.0% (       0.0%     +       0.2%  )