==8166== Cachegrind, a cache and branch-prediction profiler
==8166== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==8166== Using Valgrind-3.13.0 and LibVEX; rerun with -h for copyright info
==8166== Command: mksh -n benchmarks/testdata/test-cmd-util.sh
==8166== Parent PID: 8159
==8166== 
--8166-- warning: L3 cache found, using its data for the LL simulation.
--8166-- warning: specified LL cache: line_size 64  assoc 11  total_size 34,603,008
--8166-- warning: simulated LL cache: line_size 64  assoc 17  total_size 35,651,584
==8166== 
==8166== I   refs:      17,183,883
==8166== I1  misses:         8,802
==8166== LLi misses:         2,168
==8166== I1  miss rate:       0.05%
==8166== LLi miss rate:       0.01%
==8166== 
==8166== D   refs:       6,060,541  (3,880,669 rd   + 2,179,872 wr)
==8166== D1  misses:        57,448  (   46,176 rd   +    11,272 wr)
==8166== LLd misses:         9,289  (    3,125 rd   +     6,164 wr)
==8166== D1  miss rate:        0.9% (      1.2%     +       0.5%  )
==8166== LLd miss rate:        0.2% (      0.1%     +       0.3%  )
==8166== 
==8166== LL refs:           66,250  (   54,978 rd   +    11,272 wr)
==8166== LL misses:         11,457  (    5,293 rd   +     6,164 wr)
==8166== LL miss rate:         0.0% (      0.0%     +       0.3%  )