==3278== Cachegrind, a cache and branch-prediction profiler
==3278== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==3278== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==3278== Command: mksh -n benchmarks/testdata/t9300-fast-import.sh
==3278== Parent PID: 3273
==3278== 
--3278-- warning: L3 cache found, using its data for the LL simulation.
==3278== 
==3278== I   refs:      10,203,277
==3278== I1  misses:         9,194
==3278== LLi misses:         1,926
==3278== I1  miss rate:       0.09%
==3278== LLi miss rate:       0.02%
==3278== 
==3278== D   refs:       3,503,337  (2,276,844 rd   + 1,226,493 wr)
==3278== D1  misses:         7,518  (    4,177 rd   +     3,341 wr)
==3278== LLd misses:         5,385  (    2,548 rd   +     2,837 wr)
==3278== D1  miss rate:        0.2% (      0.2%     +       0.3%  )
==3278== LLd miss rate:        0.2% (      0.1%     +       0.2%  )
==3278== 
==3278== LL refs:           16,712  (   13,371 rd   +     3,341 wr)
==3278== LL misses:          7,311  (    4,474 rd   +     2,837 wr)
==3278== LL miss rate:         0.1% (      0.0%     +       0.2%  )