==2360== Cachegrind, a cache and branch-prediction profiler
==2360== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2360== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2360== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/t9300-fast-import.sh
==2360== Parent PID: 2353
==2360== 
--2360-- warning: L3 cache found, using its data for the LL simulation.
==2360== 
==2360== I   refs:      14,034,678
==2360== I1  misses:        70,514
==2360== LLi misses:         7,754
==2360== I1  miss rate:       0.50%
==2360== LLi miss rate:       0.06%
==2360== 
==2360== D   refs:       6,366,009  (3,565,876 rd   + 2,800,133 wr)
==2360== D1  misses:        71,720  (   38,186 rd   +    33,534 wr)
==2360== LLd misses:        43,648  (   13,366 rd   +    30,282 wr)
==2360== D1  miss rate:        1.1% (      1.1%     +       1.2%  )
==2360== LLd miss rate:        0.7% (      0.4%     +       1.1%  )
==2360== 
==2360== LL refs:          142,234  (  108,700 rd   +    33,534 wr)
==2360== LL misses:         51,402  (   21,120 rd   +    30,282 wr)
==2360== LL miss rate:         0.3% (      0.1%     +       1.1%  )