==2362== Cachegrind, a cache and branch-prediction profiler
==2362== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2362== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2362== Command: dash -n benchmarks/testdata/t9300-fast-import.sh
==2362== Parent PID: 2355
==2362== 
--2362-- warning: L3 cache found, using its data for the LL simulation.
==2362== 
==2362== I   refs:      4,358,947
==2362== I1  misses:        1,394
==2362== LLi misses:        1,288
==2362== I1  miss rate:      0.03%
==2362== LLi miss rate:      0.03%
==2362== 
==2362== D   refs:      1,803,232  (1,243,773 rd   + 559,459 wr)
==2362== D1  misses:        3,857  (    3,022 rd   +     835 wr)
==2362== LLd misses:        3,251  (    2,486 rd   +     765 wr)
==2362== D1  miss rate:       0.2% (      0.2%     +     0.1%  )
==2362== LLd miss rate:       0.2% (      0.2%     +     0.1%  )
==2362== 
==2362== LL refs:           5,251  (    4,416 rd   +     835 wr)
==2362== LL misses:         4,539  (    3,774 rd   +     765 wr)
==2362== LL miss rate:        0.1% (      0.1%     +     0.1%  )