==2562== Cachegrind, a cache and branch-prediction profiler
==2562== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2562== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2562== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/t9300-fast-import.sh
==2562== Parent PID: 2555
==2562== 
--2562-- warning: L3 cache found, using its data for the LL simulation.
==2562== 
==2562== I   refs:      13,911,876
==2562== I1  misses:        56,647
==2562== LLi misses:         8,052
==2562== I1  miss rate:       0.41%
==2562== LLi miss rate:       0.06%
==2562== 
==2562== D   refs:       6,260,677  (3,518,178 rd   + 2,742,499 wr)
==2562== D1  misses:        62,133  (   29,685 rd   +    32,448 wr)
==2562== LLd misses:        43,428  (   13,228 rd   +    30,200 wr)
==2562== D1  miss rate:        1.0% (      0.8%     +       1.2%  )
==2562== LLd miss rate:        0.7% (      0.4%     +       1.1%  )
==2562== 
==2562== LL refs:          118,780  (   86,332 rd   +    32,448 wr)
==2562== LL misses:         51,480  (   21,280 rd   +    30,200 wr)
==2562== LL miss rate:         0.3% (      0.1%     +       1.1%  )