==2392== Cachegrind, a cache and branch-prediction profiler
==2392== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==2392== Using Valgrind-3.14.0 and LibVEX; rerun with -h for copyright info
==2392== Command: _bin/cxx-opt/osh -n --ast-format none benchmarks/testdata/ltmain.sh
==2392== Parent PID: 2385
==2392== 
--2392-- warning: L3 cache found, using its data for the LL simulation.
==2392== brk segment overflow in thread #1: can't grow to 0x482f000
==2392== (see section Limitations in user manual)
==2392== NOTE: further instances of this message will not be shown
==2392== 
==2392== I   refs:      114,671,858
==2392== I1  misses:        681,076
==2392== LLi misses:          8,659
==2392== I1  miss rate:        0.59%
==2392== LLi miss rate:        0.01%
==2392== 
==2392== D   refs:       54,939,623  (28,499,380 rd   + 26,440,243 wr)
==2392== D1  misses:        609,043  (   336,290 rd   +    272,753 wr)
==2392== LLd misses:        157,634  (    14,226 rd   +    143,408 wr)
==2392== D1  miss rate:         1.1% (       1.2%     +        1.0%  )
==2392== LLd miss rate:         0.3% (       0.0%     +        0.5%  )
==2392== 
==2392== LL refs:         1,290,119  ( 1,017,366 rd   +    272,753 wr)
==2392== LL misses:         166,293  (    22,885 rd   +    143,408 wr)
==2392== LL miss rate:          0.1% (       0.0%     +        0.5%  )