# Issue type % of bandwidth % of fetches % of write-backs Fetch utilization Write-back utilization
27 Inefficient loop nesting8.7%6.4%13.7%56.3%46.9%
44 Loop fusion8.7%6.4%13.7%56.3%46.9%
46 Temporal blocking8.7%6.4%13.7%56.3%46.9%
31 Inefficient loop nesting3.8%2.8%5.9%89.2%91.6%
49 Temporal blocking3.8%2.8%5.9%89.2%91.6%
28 Inefficient loop nesting11.0%16.1%0.0%66.1%100.0%
47 Spat/temp blocking11.0%16.1%0.0%66.1%100.0%
36 Inefficient loop nesting1.8%1.5%2.6%88.0%100.0%
52 Temporal blocking1.8%1.5%2.6%88.0%100.0%
32 Inefficient loop nesting1.0%1.5%0.0%26.9%100.0%
50 Spat/temp blocking1.0%1.5%0.0%26.9%100.0%
37 Inefficient loop nesting2.6%1.9%4.1%47.8%50.2%
54 Temporal blocking2.6%1.9%4.1%47.8%50.2%
29 Inefficient loop nesting7.7%11.3%0.0%54.3%100.0%
48 Spat/temp blocking7.7%11.3%0.0%54.3%100.0%
30 Inefficient loop nesting1.9%1.4%3.0%100.0%100.0%
40 Inefficient loop nesting0.7%0.0%2.1%100.0%18.4%
41 Inefficient loop nesting4.0%2.9%6.2%0.0%51.0%
33 Inefficient loop nesting0.9%0.7%1.5%0.0%51.0%
34 Inefficient loop nesting0.9%0.7%1.4%0.0%40.5%
45 Loop fusion0.9%0.7%1.4%0.0%40.5%
42 Inefficient loop nesting1.6%1.2%2.5%0.0%26.8%
43 Inefficient loop nesting1.3%0.9%2.0%0.0%32.3%
35 Inefficient loop nesting0.9%1.3%0.0%13.4%100.0%
51 Spat/temp blocking0.9%1.3%0.0%13.4%100.0%
23 Fetch utilization1.6%2.4%0.0%5.4%100.0%
53 Spatial blocking1.6%2.4%0.0%5.4%100.0%
38 Inefficient loop nesting1.7%2.1%1.0%27.7%72.8%
55 Spat/temp blocking1.7%2.1%1.0%27.7%72.8%
39 Inefficient loop nesting1.0%0.7%1.5%0.0%45.6%
56 Spat/temp blocking1.0%0.5%2.0%12.3%6.1%

Copyright (c) 2006-2012 Rogue Wave Software, Inc. All Rights Reserved.
Patents pending.

# Issue type % of misses HW-Prefetch Randomness Fetch utilization
27 Inefficient loop nesting6.4%0.0%Low56.3%
44 Loop fusion6.4%0.0%Low56.3%
46 Temporal blocking6.4%0.0%Low56.3%
31 Inefficient loop nesting2.8%0.0%Low89.2%
49 Temporal blocking2.8%0.0%Low89.2%
28 Inefficient loop nesting16.1%0.0%Low66.1%
47 Spat/temp blocking16.1%0.0%Low66.1%
36 Inefficient loop nesting1.5%0.0%Low88.0%
52 Temporal blocking1.5%0.0%Low88.0%
32 Inefficient loop nesting1.5%0.0%Low26.9%
50 Spat/temp blocking1.5%0.0%Low26.9%
37 Inefficient loop nesting1.9%0.0%Low47.8%
54 Temporal blocking1.9%0.0%Low47.8%
29 Inefficient loop nesting11.3%0.0%Low54.3%
48 Spat/temp blocking11.3%0.0%Low54.3%
30 Inefficient loop nesting1.4%0.0%Low100.0%
40 Inefficient loop nesting0.0%0.0%Low100.0%
41 Inefficient loop nesting2.9%0.0%Low0.0%
33 Inefficient loop nesting0.7%0.0%Low0.0%
34 Inefficient loop nesting0.7%0.0%Low0.0%
45 Loop fusion0.7%0.0%Low0.0%
42 Inefficient loop nesting1.2%0.0%Low0.0%
43 Inefficient loop nesting0.9%0.0%Low0.0%
35 Inefficient loop nesting1.3%0.0%Low13.4%
51 Spat/temp blocking1.3%0.0%Low13.4%
23 Fetch utilization2.4%0.0%Low5.4%
53 Spatial blocking2.4%0.0%Low5.4%
38 Inefficient loop nesting2.1%0.0%Low27.7%
55 Spat/temp blocking2.1%0.0%Low27.7%
39 Inefficient loop nesting0.7%0.0%Low0.0%
56 Spat/temp blocking0.5%0.0%Low12.3%

Copyright (c) 2006-2012 Rogue Wave Software, Inc. All Rights Reserved.
Patents pending.

# Issue type % of communication Communication utilization False sharing
70 Communication utilization23.1%44.4%0.0%
59 Communication hot-spot12.6%98.5%0.0%
72 Communication utilization6.5%79.1%0.0%
73 Communication utilization9.9%35.7%0.0%
79 False sharing0.0%100.0%0.4%
75 Communication utilization3.3%0.0%3.2%
80 False sharing3.3%0.0%3.2%
77 Communication utilization1.3%0.0%1.5%
82 False sharing1.3%0.0%1.5%
71 Communication utilization1.6%0.0%0.0%
62 Communication hot-spot1.6%49.2%1.1%
78 False sharing1.6%49.2%1.1%
60 Communication hot-spot1.6%94.7%0.0%
83 False sharing0.0%100.0%3.3%
81 False sharing0.0%100.0%14.3%
76 Communication utilization1.6%0.0%0.0%
74 Communication utilization1.6%25.0%0.0%
68 Communication hot-spot1.6%38.2%0.0%
69 Communication hot-spot1.6%54.6%0.0%

Copyright (c) 2006-2012 Rogue Wave Software, Inc. All Rights Reserved.
Patents pending.

# Issue type % of fetches Required cache size
84 Non-temporal data2.4%24M
85 Non-temporal data1.4%20M

Copyright (c) 2006-2012 Rogue Wave Software, Inc. All Rights Reserved.
Patents pending.

Loop % of misses % of fetches Fetch utilization Write-back utilization Issues
1 6.7% 6.7% 54.5% 46.9%
2 2.8% 2.8% 92.4% 91.6%
4 16.3% 16.3% 68.4% 100.0%
6 1.6% 1.6% 87.6% 100.0%
7 1.5% 1.5% 27.0% 66.0%
8 1.9% 1.9% 51.3% 50.2%
9 11.8% 11.8% 56.0% 73.1%
11 1.8% 1.8% 100.0% 100.0%
12 0.1% 0.1% 100.0% 27.8%
24 0.7% 0.7% 13.3% 100.0%
25 3.7% 3.7% 16.1% 51.0%
31 0.0% 0.0% 100.0% 12.5%
63 0.3% 0.3% 43.2% 72.9%
64 0.7% 0.7% 0.0% 51.0%
68 0.7% 0.7% 0.7% 40.5%
70 3.0% 3.0% 7.5% 35.0%
76 1.8% 1.8% 35.9% 35.8%
77 0.0% 0.0% 100.0% 100.0%
83 0.5% 0.5% 91.6% 100.0%
87 0.0% 0.0% 18.8% 100.0%
109 2.4% 2.4% 5.5% 100.0%
150 0.2% 0.2% 12.7% 100.0%
152 0.0% 0.0% 50.7% 12.5%
153 2.1% 2.1% 27.7% 72.8%
156 0.7% 0.7% 4.2% 45.6%
166 0.2% 0.2% 25.6% 25.0%
210 0.5% 0.5% 12.3% 6.1%
267 0.2% 0.2% 67.4% 100.0%
280 0.0% 0.0% 100.0% 100.0%

Copyright (c) 2006-2012 Rogue Wave Software, Inc. All Rights Reserved.
Patents pending.

Global statistics
Miss/Fetch ratio
Write-back ratio
Utilization
Accesses3.73e+11
Misses3.04e+09
Fetches3.04e+09
Write-backs1.42e+09
Upgrades2.85e+08
Miss ratio0.8%
Fetch ratio0.8%
Write-back ratio0.4%
Upgrade ratio0.1%
Communication ratio0.2%
Fetch utilization51.0%
Write-back utilization64.0%
Communication utilization46.7%
Analysis parameters
Processor modelintel/sandy_bridge_ep_6_12_15360
Number of CPUs2
Number of caches2
Cache level3
Cache size15M
Line size64
Replacement policy random
Software prefetches active Yes

Copyright (c) 2006-2012 Rogue Wave Software, Inc. All Rights Reserved.
Patents pending.

Source code files

File
/home/users/khuck/src/hpx-lsu/apex/src/apex/apex.cpp
/home/users/khuck/src/hpx-lsu/apex/src/apex/profiler.hpp
/home/users/khuck/src/hpx-lsu/apex/src/apex/profiler_listener.cpp
/home/users/khuck/src/hpx-lsu/apex/src/apex/task_identifier.hpp
/home/users/khuck/src/hpx-lsu/hpx/lcos/async.hpp
/home/users/khuck/src/hpx-lsu/hpx/lcos/detail/future_data.hpp
/home/users/khuck/src/hpx-lsu/hpx/lcos/future.hpp
/home/users/khuck/src/hpx-lsu/hpx/lcos/local/futures_factory.hpp
/home/users/khuck/src/hpx-lsu/hpx/lcos/local/packaged_continuation.hpp
/home/users/khuck/src/hpx-lsu/hpx/lcos/local/spinlock.hpp
/home/users/khuck/src/hpx-lsu/hpx/lcos/promise.hpp
/home/users/khuck/src/hpx-lsu/hpx/lcos/wait_all.hpp
/home/users/khuck/src/hpx-lsu/hpx/lcos/when_all.hpp
/home/users/khuck/src/hpx-lsu/hpx/parallel/executors/executor_parameter_traits.hpp
/home/users/khuck/src/hpx-lsu/hpx/parallel/util/partitioner.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/actions/component_action.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/actions/plain_action.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/actions/trigger.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/agas/interface.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/applier/detail/apply_implementations.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/components/component_factory.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/components/server/managed_component_base.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/components/server/runtime_support.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/threads/coroutines/detail/context_linux_x86.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/threads/detail/create_thread.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/threads/detail/scheduling_loop.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/threads/policies/local_priority_queue_scheduler.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/threads/policies/thread_queue.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/threads/thread_helpers.hpp
/home/users/khuck/src/hpx-lsu/hpx/runtime/trigger_lco.hpp
/home/users/khuck/src/hpx-lsu/hpx/util/apex.hpp
/home/users/khuck/src/hpx-lsu/hpx/util/detail/basic_function.hpp
/home/users/khuck/src/hpx-lsu/hpx/util/invoke.hpp
/home/users/khuck/src/hpx-lsu/hpx/util/unlock_guard.hpp
/home/users/khuck/src/hpx-lsu/src/runtime.cpp
/home/users/khuck/src/hpx-lsu/src/runtime/agas/addressing_service.cpp
/home/users/khuck/src/hpx-lsu/src/runtime/agas/primary_namespace.cpp
/home/users/khuck/src/hpx-lsu/src/runtime/applier/applier.cpp
/home/users/khuck/src/hpx-lsu/src/runtime/threads/detail/thread_pool.cpp
/home/users/khuck/src/hpx-lsu/src/runtime/threads/thread_helpers.cpp
/home/users/khuck/src/hpx-lsu/src/util/runtime_configuration.cpp
/home/users/khuck/src/jemalloc-3.5.1/include/jemalloc/internal/tcache.h
/home/users/khuck/src/jemalloc-3.5.1/src/arena.c
/home/users/khuck/src/jemalloc-3.5.1/src/tcache.c
/home/users/khuck/src/octotiger/src/grid.cpp
/home/users/khuck/src/octotiger/src/grid_fmm.cpp
/home/users/khuck/src/octotiger/src/grid_output.cpp
/home/users/khuck/src/octotiger/src/node_client.cpp
/home/users/khuck/src/octotiger/src/node_server.cpp
/home/users/khuck/src/octotiger/src/node_server_actions_1.cpp
/home/users/khuck/src/octotiger/src/node_server_actions_2.cpp
/home/users/khuck/src/octotiger/src/node_server_actions_3.cpp
/home/users/khuck/src/operation_gordon_bell/Vc/include/Vc/common/simdarray.h
/packages/src/gcc-5.3.0-build/x86_64-unknown-linux-gnu/libstdc++-v3/libsupc++/../../../../gcc-5.3.0/libstdc++-v3/libsupc++/hash_bytes.cc
/storage/packages/boost/1.61_gcc-5.3/include/boost/atomic/detail/ops_gcc_atomic.hpp
/storage/packages/boost/1.61_gcc-5.3/include/boost/thread/pthread/mutex.hpp
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/basic_string.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/char_traits.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/functional_hash.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/shared_ptr_base.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/stl_algobase.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/stl_construct.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/stl_set.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/stl_tree.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/stl_vector.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/unique_ptr.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/bits/unordered_map.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/cmath
/storage/packages/gcc/5.3.0/include/c++/5.3.0/ext/new_allocator.h
/storage/packages/gcc/5.3.0/include/c++/5.3.0/mutex
/storage/packages/gcc/5.3.0/lib/gcc/x86_64-unknown-linux-gnu/5.3.0/include/avxintrin.h

Shared libraries

Filename Symbols Low address High address
/lib64/ld-2.12.so Only symbols found 0x34d6000000 0x34d6020000
/lib64/libc-2.12.so Only symbols found 0x34d6400000 0x34d6793928
/lib64/libdl-2.12.so Only symbols found 0x34d7000000 0x34d7203100
/lib64/libm-2.12.so Only symbols found 0x34d6800000 0x34d6a830b8
/lib64/libpthread-2.12.so Only symbols found 0x34d6c00000 0x34d6e1c3f0
/lib64/librt-2.12.so Only symbols found 0x34d7400000 0x34d7607bb0
/storage/packages/boost/1.61_gcc-5.3/lib/libboost_atomic.so.1.61.0 Only symbols found 0x7faa7dadd000 0x7faa7dcded00
/storage/packages/boost/1.61_gcc-5.3/lib/libboost_chrono.so.1.61.0 Only symbols found 0x7faa7eebb000 0x7faa7f0c1618
/storage/packages/boost/1.61_gcc-5.3/lib/libboost_context.so.1.61.0 Only symbols found 0x7faa7dee5000 0x7faa7e0f1b00
/storage/packages/boost/1.61_gcc-5.3/lib/libboost_date_time.so.1.61.0 Only symbols found 0x7faa7ecaa000 0x7faa7eebaec0
/storage/packages/boost/1.61_gcc-5.3/lib/libboost_filesystem.so.1.61.0 Only symbols found 0x7faa7ea90000 0x7faa7eca8ac0
/storage/packages/boost/1.61_gcc-5.3/lib/libboost_program_options.so.1.61.0 Only symbols found 0x7faa7e80f000 0x7faa7ea8fa00
/storage/packages/boost/1.61_gcc-5.3/lib/libboost_random.so.1.61.0 Only symbols found 0x7faa7dcdf000 0x7faa7dee45f0
/storage/packages/boost/1.61_gcc-5.3/lib/libboost_regex.so.1.61.0 Only symbols found 0x7faa7e51e000 0x7faa7e80e840
/storage/packages/boost/1.61_gcc-5.3/lib/libboost_system.so.1.61.0 Only symbols found 0x7faa7e31a000 0x7faa7e51c6a0
/storage/packages/boost/1.61_gcc-5.3/lib/libboost_thread.so.1.61.0 Only symbols found 0x7faa7e0f3000 0x7faa7e319f10
/storage/packages/gcc/5.3.0/lib64/libgcc_s.so.1 Debug info found 0x7faa7c8ab000 0x7faa7cac0860
/storage/packages/gcc/5.3.0/lib64/libgfortran.so.3.0.0 Debug info found 0x7faa7d08e000 0x7faa7d3ade58
/storage/packages/gcc/5.3.0/lib64/libquadmath.so.0.0.0 Debug info found 0x7faa7ce50000 0x7faa7d08dd18
/storage/packages/gcc/5.3.0/lib64/libstdc++.so.6.0.21 Debug info found 0x7faa7cac2000 0x7faa7ce4f5c0
/storage/users/khuck/install/hwloc-1.8/lib/libhwloc.so.5.4.0 Debug info found 0x7faa7d66f000 0x7faa7d89db78
/storage/users/khuck/install/jemalloc-3.5.1/lib/libjemalloc.so.1 Debug info found 0x7faa7d89e000 0x7faa7dadbec8
/storage/users/khuck/install/otf2/2.0/lib/libotf2.so.7.0.0 Debug info found 0x7faa7d3af000 0x7faa7d66e9c0
/storage/users/khuck/src/operation_gordon_bell/build-x86-octotiger/octotiger Debug info found 0x400000 0xb2a000
/storage/users/khuck/src/operation_gordon_bell/build-x86/lib/hpx/libhpx_io_counters.so.1.0.0 Debug info found 0x7faa7b218000 0x7faa7b432af0
/storage/users/khuck/src/operation_gordon_bell/build-x86/lib/hpx/libhpx_memory.so.1.0.0 Debug info found 0x7faa7b473000 0x7faa7b68ffd8
/storage/users/khuck/src/operation_gordon_bell/build-x86/lib/hpx/libhpx_parcel_coalescing.so.1.0.0 Debug info found 0x7faa7afc6000 0x7faa7b207ec8
/storage/users/khuck/src/operation_gordon_bell/build-x86/lib/libhpx.so.1.0.0 Debug info found 0x7faa7f0ff000 0x7faa80116d48

Copyright (c) 2006-2012 Rogue Wave Software, Inc. All Rights Reserved.
Patents pending.

Command line ./build-x86-octotiger/octotiger -Disableoutput -Problem=moving_star -Max_level=6 -Stopstep=0 -Xscale=32 -Odt=0.5 --hpx:threads=24 -Ihpx.stacks.use_guard_pages=0
Sampling start time Tue Dec 13 13:51:49 2016
Sampling end time Tue Dec 13 13:59:00 2016
User khuck
Sample file /storage/users/khuck/src/operation_gordon_bell/sample.smp
Number of samples 51770
Analysis parameters report_ts --cpu intel/sandy_bridge_ep_6_12_15360 -i sample.smp
Analysis time Tue Dec 13 14:02:33 2016
Effective thread binding
Cache # Thread id Virtual Thread id
0 23399 0
23401 1
23404 4
23406 6
23410 10
23412 12
23413 13
23414 14
23415 15
23416 16
23418 18
23420 20
23422 22
23424 24
23425 25
23431 31
23432 32
1 23402 2
23403 3
23405 5
23407 7
23408 8
23409 9
23411 11
23417 17
23419 19
23421 21
23423 23
23426 26
23427 27
23428 28
23429 29
23430 30

Copyright (c) 2006-2012 Rogue Wave Software, Inc. All Rights Reserved.
Patents pending.

ThreadSpotterâ„¢

Version: 1.3.10, Build: unknown

View the manual.

Legend

Slowspot issues Opportunity issues
Fetch utilization
Write-back utilization
Communication utilization
Inefficient loop nesting
Random access
Prefetch: too close
Prefetch: too distant
Prefetch: unnecessary
False sharing
Spatial blocking
Temporal blocking
Spat/temp blocking
Loop fusion
Non-temporal data
Non-temporal store possible
Fetch hot-spot
Write-back hot-spot
Communication hot-spot

Contact information:

ParaTools, Inc.
Address

2836 Kincaid St.

Eugene, OR 97405 USA

Phone +1 541-913-8797
Phone +1 443-543-5475
Fax +1 541-343-6086
Support Email threadspotter@paratools.com
Web http://www.paratools.com

Copyright (c) 2006-2012 Rogue Wave Software, Inc. All Rights Reserved.
Patents pending.