==2324== Cachegrind, a cache and branch-prediction profiler
==2324== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2324== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2324== Command: mksh -n benchmarks/testdata/t9300-fast-import.sh
==2324== Parent PID: 2319
==2324== 
--2324-- warning: L3 cache found, using its data for the LL simulation.
==2324== 
==2324== I   refs:      10,203,022
==2324== I1  misses:         4,473
==2324== LLi misses:         1,932
==2324== I1  miss rate:       0.04%
==2324== LLi miss rate:       0.02%
==2324== 
==2324== D   refs:       3,503,290  (2,276,802 rd   + 1,226,488 wr)
==2324== D1  misses:         7,188  (    3,987 rd   +     3,201 wr)
==2324== LLd misses:         5,382  (    2,545 rd   +     2,837 wr)
==2324== D1  miss rate:        0.2% (      0.2%     +       0.3%  )
==2324== LLd miss rate:        0.2% (      0.1%     +       0.2%  )
==2324== 
==2324== LL refs:           11,661  (    8,460 rd   +     3,201 wr)
==2324== LL misses:          7,314  (    4,477 rd   +     2,837 wr)
==2324== LL miss rate:         0.1% (      0.0%     +       0.2%  )