==2352== Cachegrind, a cache and branch-prediction profiler
==2352== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2352== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2352== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/t9300-fast-import.sh
==2352== Parent PID: 2345
==2352== 
--2352-- warning: L3 cache found, using its data for the LL simulation.
==2352== 
==2352== I   refs:      14,086,195
==2352== I1  misses:        41,378
==2352== LLi misses:         7,840
==2352== I1  miss rate:       0.29%
==2352== LLi miss rate:       0.06%
==2352== 
==2352== D   refs:       6,381,145  (3,575,964 rd   + 2,805,181 wr)
==2352== D1  misses:        64,309  (   31,458 rd   +    32,851 wr)
==2352== LLd misses:        43,936  (   13,431 rd   +    30,505 wr)
==2352== D1  miss rate:        1.0% (      0.9%     +       1.2%  )
==2352== LLd miss rate:        0.7% (      0.4%     +       1.1%  )
==2352== 
==2352== LL refs:          105,687  (   72,836 rd   +    32,851 wr)
==2352== LL misses:         51,776  (   21,271 rd   +    30,505 wr)
==2352== LL miss rate:         0.3% (      0.1%     +       1.1%  )