| Loop Id: 4 | Module: kmeans-gcc-Ofast | Source: main.cpp:61-67 | Coverage: 92.30% |
|---|
| Loop Id: 4 | Module: kmeans-gcc-Ofast | Source: main.cpp:61-67 | Coverage: 92.30% |
|---|
0x401c20 LDP D31, D30, [X0], #16 [2] |
0x401c24 FSUB D30, D27, S30 |
0x401c28 FSUB D31, D28, S31 |
0x401c2c FMUL D30, D30, D30 |
0x401c30 FMADD D31, D31, D31, D30 |
0x401c34 FCMPE D29, D31 |
0x401c38 B.LS 401c44 |
0x401c3c FMOV D29, D31 |
0x401c40 STR W1, [X5, X3,LSL #2] [1] |
0x401c44 ADD W1, W1, #1 |
0x401c48 CMP W2, W1 |
0x401c4c B.NE 401c20 |
/home/fmusial/KMEANS_Benchmarks/kmeans/main.cpp: 61 - 67 |
-------------------------------------------------------------------------------- |
61: for (int j = 0; j < k; ++j) { |
62: double dist = |
63: (points[i].x - centroids[j].x) * (points[i].x - centroids[j].x) + |
64: (points[i].y - centroids[j].y) * (points[i].y - centroids[j].y); |
65: if (dist < optimal_dist) { |
66: optimal_dist = dist; |
67: assignment[i] = j; |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►100.00+ | GOMP_parallel | libgomp.h:980 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-Ofast |
| ○ | main | main.cpp:20 | kmeans-gcc-Ofast |
| ○ | __libc_start_call_main | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | new_allocator.h:104 | kmeans-gcc-Ofast |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►50.17+ | GOMP_parallel | libgomp.h:980 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-Ofast |
| ○ | main | main.cpp:20 | kmeans-gcc-Ofast |
| ○ | __libc_start_call_main | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | new_allocator.h:104 | kmeans-gcc-Ofast |
| ►49.83+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | start_thread | libc.so.6 | |
| ○ | thread_start | libc.so.6 |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►74.90+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | start_thread | libc.so.6 | |
| ○ | thread_start | libc.so.6 | |
| ►25.10+ | GOMP_parallel | libgomp.h:980 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-Ofast |
| ○ | main | main.cpp:20 | kmeans-gcc-Ofast |
| ○ | __libc_start_call_main | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | new_allocator.h:104 | kmeans-gcc-Ofast |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►87.28+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | start_thread | libc.so.6 | |
| ○ | thread_start | libc.so.6 | |
| ►12.72+ | GOMP_parallel | libgomp.h:980 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-Ofast |
| ○ | main | main.cpp:20 | kmeans-gcc-Ofast |
| ○ | __libc_start_call_main | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | new_allocator.h:104 | kmeans-gcc-Ofast |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►93.78+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | start_thread | libc.so.6 | |
| ○ | thread_start | libc.so.6 | |
| ►6.22+ | GOMP_parallel | libgomp.h:980 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-Ofast |
| ○ | main | main.cpp:20 | kmeans-gcc-Ofast |
| ○ | __libc_start_call_main | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | new_allocator.h:104 | kmeans-gcc-Ofast |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►96.92+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | start_thread | libc.so.6 | |
| ○ | thread_start | libc.so.6 | |
| ►3.08+ | GOMP_parallel | libgomp.h:980 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-Ofast |
| ○ | main | main.cpp:20 | kmeans-gcc-Ofast |
| ○ | __libc_start_call_main | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | new_allocator.h:104 | kmeans-gcc-Ofast |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►97.96+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | start_thread | libc.so.6 | |
| ○ | thread_start | libc.so.6 | |
| ►2.04+ | GOMP_parallel | libgomp.h:980 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-Ofast |
| ○ | main | main.cpp:20 | kmeans-gcc-Ofast |
| ○ | __libc_start_call_main | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | new_allocator.h:104 | kmeans-gcc-Ofast |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►98.46+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | start_thread | libc.so.6 | |
| ○ | thread_start | libc.so.6 | |
| ►1.54+ | GOMP_parallel | libgomp.h:980 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-Ofast |
| ○ | main | main.cpp:20 | kmeans-gcc-Ofast |
| ○ | __libc_start_call_main | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | new_allocator.h:104 | kmeans-gcc-Ofast |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►98.78+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | start_thread | libc.so.6 | |
| ○ | thread_start | libc.so.6 | |
| ►1.22+ | GOMP_parallel | libgomp.h:980 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-Ofast |
| ○ | main | main.cpp:20 | kmeans-gcc-Ofast |
| ○ | __libc_start_call_main | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | new_allocator.h:104 | kmeans-gcc-Ofast |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►98.97+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | start_thread | libc.so.6 | |
| ○ | thread_start | libc.so.6 | |
| ►1.03+ | GOMP_parallel | libgomp.h:980 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-Ofast |
| ○ | main | main.cpp:20 | kmeans-gcc-Ofast |
| ○ | __libc_start_call_main | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | new_allocator.h:104 | kmeans-gcc-Ofast |
| min | med | avg | max |
|---|---|---|---|
| Percentile Index | 10 | 20 | 30 | 40 | 50 | 60 | 70 | 80 | 90 | 100 |
|---|---|---|---|---|---|---|---|---|---|---|
| Value |
| min | med | avg | max |
|---|---|---|---|
| Percentile Index | 10 | 20 | 30 | 40 | 50 | 60 | 70 | 80 | 90 | 100 |
|---|---|---|---|---|---|---|---|---|---|---|
| Value |
| Path / |
| Metric | Value |
|---|---|
| CQA speedup if no scalar integer | 1.38 |
| CQA speedup if FP arith vectorized | 1.38 |
| CQA speedup if fully vectorized | 1.31 |
| CQA speedup if no inter-iteration dependency | NA |
| CQA speedup if next bottleneck killed | 1.38 |
| Bottlenecks | |
| Function | k_means(int, point_t*, point_t*, int*, point_t*, int, int) [clone ._omp_fn.0] |
| Source | main.cpp:61-67 |
| Source loop unroll info | not unrolled or unrolled with no peel/tail loop |
| Source loop unroll confidence level | max |
| Unroll/vectorization loop type | NA |
| Unroll factor | NA |
| CQA cycles | 1.38 |
| CQA cycles if no scalar integer | 1.00 |
| CQA cycles if FP arith vectorized | 1.00 |
| CQA cycles if fully vectorized | 1.05 |
| Front-end cycles | 1.38 |
| P0 cycles | 1.00 |
| P1 cycles | 1.00 |
| P2 cycles | 0.58 |
| P3 cycles | 0.58 |
| P4 cycles | 0.50 |
| P5 cycles | 0.50 |
| P6 cycles | 0.42 |
| P7 cycles | 0.42 |
| P8 cycles | 1.38 |
| P9 cycles | 1.38 |
| P10 cycles | 1.38 |
| P11 cycles | 1.38 |
| P12 cycles | 0.58 |
| P13 cycles | 0.42 |
| P14 cycles | 0.50 |
| P15 cycles | 0.25 |
| P16 cycles | 0.25 |
| DIV/SQRT cycles | 0.00 |
| Inter-iter dependencies cycles | 1 |
| FE+BE cycles (UFS) | NA |
| Stall cycles (UFS) | NA |
| Nb insns | 11.00 |
| Nb uops | 11.00 |
| Nb loads | NA |
| Nb stores | 0.50 |
| Nb stack references | 0.00 |
| FLOP/cycle | 3.64 |
| Nb FLOP add-sub | 2.00 |
| Nb FLOP mul | 1.00 |
| Nb FLOP fma | 1.00 |
| Nb FLOP div | 0.00 |
| Nb FLOP rcp | 0.00 |
| Nb FLOP sqrt | 0.00 |
| Nb FLOP rsqrt | 0.00 |
| Bytes/cycle | 13.07 |
| Bytes prefetched | 0.00 |
| Bytes loaded | 16.00 |
| Bytes stored | 2.00 |
| Stride 0 | 1.50 |
| Stride 1 | 0.00 |
| Stride n | 0.00 |
| Stride unknown | 0.00 |
| Stride indirect | 0.00 |
| Vectorization ratio all | 14.58 |
| Vectorization ratio load | 100.00 |
| Vectorization ratio store | 0.00 |
| Vectorization ratio mul | 0.00 |
| Vectorization ratio add_sub | 0.00 |
| Vectorization ratio fma | 0.00 |
| Vectorization ratio div_sqrt | NA |
| Vectorization ratio other | 0.00 |
| Vector-efficiency ratio all | 55.73 |
| Vector-efficiency ratio load | 100.00 |
| Vector-efficiency ratio store | 25.00 |
| Vector-efficiency ratio mul | 50.00 |
| Vector-efficiency ratio add_sub | 50.00 |
| Vector-efficiency ratio fma | 50.00 |
| Vector-efficiency ratio div_sqrt | NA |
| Vector-efficiency ratio other | 50.00 |
| Metric | Value |
|---|---|
| CQA speedup if no scalar integer | 1.50 |
| CQA speedup if FP arith vectorized | 1.50 |
| CQA speedup if fully vectorized | 1.37 |
| CQA speedup if no inter-iteration dependency | NA |
| CQA speedup if next bottleneck killed | 1.50 |
| Bottlenecks | micro-operation queue, P8, P9, P10, P11, |
| Function | k_means(int, point_t*, point_t*, int*, point_t*, int, int) [clone ._omp_fn.0] |
| Source | main.cpp:61-67 |
| Source loop unroll info | not unrolled or unrolled with no peel/tail loop |
| Source loop unroll confidence level | max |
| Unroll/vectorization loop type | NA |
| Unroll factor | NA |
| CQA cycles | 1.50 |
| CQA cycles if no scalar integer | 1.00 |
| CQA cycles if FP arith vectorized | 1.00 |
| CQA cycles if fully vectorized | 1.09 |
| Front-end cycles | 1.50 |
| P0 cycles | 1.00 |
| P1 cycles | 1.00 |
| P2 cycles | 0.58 |
| P3 cycles | 0.58 |
| P4 cycles | 0.50 |
| P5 cycles | 0.50 |
| P6 cycles | 0.42 |
| P7 cycles | 0.42 |
| P8 cycles | 1.50 |
| P9 cycles | 1.50 |
| P10 cycles | 1.50 |
| P11 cycles | 1.50 |
| P12 cycles | 0.83 |
| P13 cycles | 0.50 |
| P14 cycles | 0.67 |
| P15 cycles | 0.50 |
| P16 cycles | 0.50 |
| DIV/SQRT cycles | 0.00 |
| Inter-iter dependencies cycles | 1 |
| FE+BE cycles (UFS) | NA |
| Stall cycles (UFS) | NA |
| Nb insns | 12.00 |
| Nb uops | 12.00 |
| Nb loads | NA |
| Nb stores | 1.00 |
| Nb stack references | 0.00 |
| FLOP/cycle | 3.33 |
| Nb FLOP add-sub | 2.00 |
| Nb FLOP mul | 1.00 |
| Nb FLOP fma | 1.00 |
| Nb FLOP div | 0.00 |
| Nb FLOP rcp | 0.00 |
| Nb FLOP sqrt | 0.00 |
| Nb FLOP rsqrt | 0.00 |
| Bytes/cycle | 13.33 |
| Bytes prefetched | 0.00 |
| Bytes loaded | 16.00 |
| Bytes stored | 4.00 |
| Stride 0 | 2.00 |
| Stride 1 | 0.00 |
| Stride n | 0.00 |
| Stride unknown | 0.00 |
| Stride indirect | 0.00 |
| Vectorization ratio all | 12.50 |
| Vectorization ratio load | 100.00 |
| Vectorization ratio store | 0.00 |
| Vectorization ratio mul | 0.00 |
| Vectorization ratio add_sub | 0.00 |
| Vectorization ratio fma | 0.00 |
| Vectorization ratio div_sqrt | NA |
| Vectorization ratio other | 0.00 |
| Vector-efficiency ratio all | 53.13 |
| Vector-efficiency ratio load | 100.00 |
| Vector-efficiency ratio store | 25.00 |
| Vector-efficiency ratio mul | 50.00 |
| Vector-efficiency ratio add_sub | 50.00 |
| Vector-efficiency ratio fma | 50.00 |
| Vector-efficiency ratio div_sqrt | NA |
| Vector-efficiency ratio other | 50.00 |
| Metric | Value |
|---|---|
| CQA speedup if no scalar integer | 1.25 |
| CQA speedup if FP arith vectorized | 1.25 |
| CQA speedup if fully vectorized | 1.25 |
| CQA speedup if no inter-iteration dependency | NA |
| CQA speedup if next bottleneck killed | 1.25 |
| Bottlenecks | micro-operation queue, P8, P9, P10, P11, |
| Function | k_means(int, point_t*, point_t*, int*, point_t*, int, int) [clone ._omp_fn.0] |
| Source | main.cpp:61-67 |
| Source loop unroll info | not unrolled or unrolled with no peel/tail loop |
| Source loop unroll confidence level | max |
| Unroll/vectorization loop type | NA |
| Unroll factor | NA |
| CQA cycles | 1.25 |
| CQA cycles if no scalar integer | 1.00 |
| CQA cycles if FP arith vectorized | 1.00 |
| CQA cycles if fully vectorized | 1.00 |
| Front-end cycles | 1.25 |
| P0 cycles | 1.00 |
| P1 cycles | 1.00 |
| P2 cycles | 0.58 |
| P3 cycles | 0.58 |
| P4 cycles | 0.50 |
| P5 cycles | 0.50 |
| P6 cycles | 0.42 |
| P7 cycles | 0.42 |
| P8 cycles | 1.25 |
| P9 cycles | 1.25 |
| P10 cycles | 1.25 |
| P11 cycles | 1.25 |
| P12 cycles | 0.33 |
| P13 cycles | 0.33 |
| P14 cycles | 0.33 |
| P15 cycles | 0.00 |
| P16 cycles | 0.00 |
| DIV/SQRT cycles | 0.00 |
| Inter-iter dependencies cycles | 1 |
| FE+BE cycles (UFS) | NA |
| Stall cycles (UFS) | NA |
| Nb insns | 10.00 |
| Nb uops | 10.00 |
| Nb loads | NA |
| Nb stores | 0.00 |
| Nb stack references | 0.00 |
| FLOP/cycle | 4.00 |
| Nb FLOP add-sub | 2.00 |
| Nb FLOP mul | 1.00 |
| Nb FLOP fma | 1.00 |
| Nb FLOP div | 0.00 |
| Nb FLOP rcp | 0.00 |
| Nb FLOP sqrt | 0.00 |
| Nb FLOP rsqrt | 0.00 |
| Bytes/cycle | 12.80 |
| Bytes prefetched | 0.00 |
| Bytes loaded | 16.00 |
| Bytes stored | 0.00 |
| Stride 0 | 1.00 |
| Stride 1 | 0.00 |
| Stride n | 0.00 |
| Stride unknown | 0.00 |
| Stride indirect | 0.00 |
| Vectorization ratio all | 16.67 |
| Vectorization ratio load | 100.00 |
| Vectorization ratio store | NA |
| Vectorization ratio mul | 0.00 |
| Vectorization ratio add_sub | 0.00 |
| Vectorization ratio fma | 0.00 |
| Vectorization ratio div_sqrt | NA |
| Vectorization ratio other | 0.00 |
| Vector-efficiency ratio all | 58.33 |
| Vector-efficiency ratio load | 100.00 |
| Vector-efficiency ratio store | NA |
| Vector-efficiency ratio mul | 50.00 |
| Vector-efficiency ratio add_sub | 50.00 |
| Vector-efficiency ratio fma | 50.00 |
| Vector-efficiency ratio div_sqrt | NA |
| Vector-efficiency ratio other | 50.00 |
| Path / |
| Function | k_means(int, point_t*, point_t*, int*, point_t*, int, int) [clone ._omp_fn.0] |
| Source file and lines | main.cpp:61-67 |
| Module | kmeans-gcc-Ofast |
| nb instructions | 11 |
| nb uops | 11 |
| loop length | 44 |
| used w registers | 2 |
| used x registers | 2 |
| used b registers | 0 |
| used h registers | 0 |
| used s registers | 2 |
| used d registers | 5 |
| used q registers | 0 |
| used v registers | 0 |
| used z registers | 0 |
| nb stack references | 0 |
| ADD-SUB / MUL ratio | 2.00 |
| micro-operation queue | 1.38 cycles |
| front end | 1.38 cycles |
| P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | P12 | P13 | P14 | P15 | P16 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| uops | 1.00 | 1.00 | 0.58 | 0.58 | 0.50 | 0.50 | 0.42 | 0.42 | 1.38 | 1.38 | 1.38 | 1.38 | 0.58 | 0.42 | 0.50 | 0.25 | 0.25 |
| cycles | 1.00 | 1.00 | 0.58 | 0.58 | 0.50 | 0.50 | 0.42 | 0.42 | 1.38 | 1.38 | 1.38 | 1.38 | 0.58 | 0.42 | 0.50 | 0.25 | 0.25 |
| Cycles executing div or sqrt instructions | NA |
| Longest recurrence chain latency (RecMII) | 1.00 |
| Front-end | 1.38 |
| Dispatch | 1.38 |
| Data deps. | 1.00 |
| Overall L1 | 1.38 |
| all | 75% |
| load | 100% |
| store | 0% |
| mul | NA (no mul vectorizable/vectorized instructions) |
| add-sub | NA (no add-sub vectorizable/vectorized instructions) |
| fma | NA (no fma vectorizable/vectorized instructions) |
| other | NA (no other vectorizable/vectorized instructions) |
| all | 0% |
| load | NA (no load vectorizable/vectorized instructions) |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 0% |
| add-sub | 0% |
| fma | 0% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 0% |
| all | 14% |
| load | 100% |
| store | 0% |
| mul | 0% |
| add-sub | 0% |
| fma | 0% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 0% |
| all | 81% |
| load | 100% |
| store | 25% |
| mul | NA (no mul vectorizable/vectorized instructions) |
| add-sub | NA (no add-sub vectorizable/vectorized instructions) |
| fma | NA (no fma vectorizable/vectorized instructions) |
| other | NA (no other vectorizable/vectorized instructions) |
| all | 50% |
| load | NA (no load vectorizable/vectorized instructions) |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 50% |
| add-sub | 50% |
| fma | 50% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 50% |
| all | 55% |
| load | 100% |
| store | 25% |
| mul | 50% |
| add-sub | 50% |
| fma | 50% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 50% |
| Function | k_means(int, point_t*, point_t*, int*, point_t*, int, int) [clone ._omp_fn.0] |
| Source file and lines | main.cpp:61-67 |
| Module | kmeans-gcc-Ofast |
| nb instructions | 12 |
| nb uops | 12 |
| loop length | 48 |
| used w registers | 2 |
| used x registers | 3 |
| used b registers | 0 |
| used h registers | 0 |
| used s registers | 2 |
| used d registers | 5 |
| used q registers | 0 |
| used v registers | 0 |
| used z registers | 0 |
| nb stack references | 0 |
| ADD-SUB / MUL ratio | 2.00 |
| micro-operation queue | 1.50 cycles |
| front end | 1.50 cycles |
| P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | P12 | P13 | P14 | P15 | P16 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| uops | 1.00 | 1.00 | 0.58 | 0.58 | 0.50 | 0.50 | 0.42 | 0.42 | 1.50 | 1.50 | 1.50 | 1.50 | 0.83 | 0.50 | 0.67 | 0.50 | 0.50 |
| cycles | 1.00 | 1.00 | 0.58 | 0.58 | 0.50 | 0.50 | 0.42 | 0.42 | 1.50 | 1.50 | 1.50 | 1.50 | 0.83 | 0.50 | 0.67 | 0.50 | 0.50 |
| Cycles executing div or sqrt instructions | NA |
| Longest recurrence chain latency (RecMII) | 1.00 |
| Front-end | 1.50 |
| Dispatch | 1.50 |
| Data deps. | 1.00 |
| Overall L1 | 1.50 |
| all | 50% |
| load | 100% |
| store | 0% |
| mul | NA (no mul vectorizable/vectorized instructions) |
| add-sub | NA (no add-sub vectorizable/vectorized instructions) |
| fma | NA (no fma vectorizable/vectorized instructions) |
| other | NA (no other vectorizable/vectorized instructions) |
| all | 0% |
| load | NA (no load vectorizable/vectorized instructions) |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 0% |
| add-sub | 0% |
| fma | 0% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 0% |
| all | 12% |
| load | 100% |
| store | 0% |
| mul | 0% |
| add-sub | 0% |
| fma | 0% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 0% |
| all | 62% |
| load | 100% |
| store | 25% |
| mul | NA (no mul vectorizable/vectorized instructions) |
| add-sub | NA (no add-sub vectorizable/vectorized instructions) |
| fma | NA (no fma vectorizable/vectorized instructions) |
| other | NA (no other vectorizable/vectorized instructions) |
| all | 50% |
| load | NA (no load vectorizable/vectorized instructions) |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 50% |
| add-sub | 50% |
| fma | 50% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 50% |
| all | 53% |
| load | 100% |
| store | 25% |
| mul | 50% |
| add-sub | 50% |
| fma | 50% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 50% |
| Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | P12 | P13 | P14 | P15 | P16 | Latency | Recip. throughput | Vectorization |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| LDP D31, D30, [X0], #16 | 1 | 0 | 0 | 0.17 | 0.17 | 0.17 | 0.17 | 0.17 | 0.17 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 | vect (100.0%) |
| FSUB D30, D27, S30 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 | scal (50.0%) |
| FSUB D31, D28, S31 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 | scal (50.0%) |
| FMUL D30, D30, D30 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 3 | 0.25 | scal (50.0%) |
| FMADD D31, D31, D31, D30 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 4 | 0.25 | scal (50.0%) |
| FCMPE D29, D31 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 | scal (50.0%) |
| B.LS 401c44 <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0xa4> | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 | N/A |
| FMOV D29, D31 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 | scal (50.0%) |
| STR W1, [X5, X3,LSL #2] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0.50 | 0.50 | 1 | 0.50 | scal (25.0%) |
| ADD W1, W1, #1 | 1 | 0 | 0 | 0.17 | 0.17 | 0.17 | 0.17 | 0.17 | 0.17 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 | N/A |
| CMP W2, W1 | 1 | 0 | 0 | 0.25 | 0.25 | 0 | 0 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.33 | N/A |
| B.NE 401c20 <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x80> | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 | N/A |
| Function | k_means(int, point_t*, point_t*, int*, point_t*, int, int) [clone ._omp_fn.0] |
| Source file and lines | main.cpp:61-67 |
| Module | kmeans-gcc-Ofast |
| nb instructions | 10 |
| nb uops | 10 |
| loop length | 40 |
| used w registers | 2 |
| used x registers | 1 |
| used b registers | 0 |
| used h registers | 0 |
| used s registers | 2 |
| used d registers | 5 |
| used q registers | 0 |
| used v registers | 0 |
| used z registers | 0 |
| nb stack references | 0 |
| ADD-SUB / MUL ratio | 2.00 |
| micro-operation queue | 1.25 cycles |
| front end | 1.25 cycles |
| P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | P12 | P13 | P14 | P15 | P16 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| uops | 1.00 | 1.00 | 0.58 | 0.58 | 0.50 | 0.50 | 0.42 | 0.42 | 1.25 | 1.25 | 1.25 | 1.25 | 0.33 | 0.33 | 0.33 | 0.00 | 0.00 |
| cycles | 1.00 | 1.00 | 0.58 | 0.58 | 0.50 | 0.50 | 0.42 | 0.42 | 1.25 | 1.25 | 1.25 | 1.25 | 0.33 | 0.33 | 0.33 | 0.00 | 0.00 |
| Cycles executing div or sqrt instructions | NA |
| Longest recurrence chain latency (RecMII) | 1.00 |
| Front-end | 1.25 |
| Dispatch | 1.25 |
| Data deps. | 1.00 |
| Overall L1 | 1.25 |
| all | 100% |
| load | 100% |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | NA (no mul vectorizable/vectorized instructions) |
| add-sub | NA (no add-sub vectorizable/vectorized instructions) |
| fma | NA (no fma vectorizable/vectorized instructions) |
| other | NA (no other vectorizable/vectorized instructions) |
| all | 0% |
| load | NA (no load vectorizable/vectorized instructions) |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 0% |
| add-sub | 0% |
| fma | 0% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 0% |
| all | 16% |
| load | 100% |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 0% |
| add-sub | 0% |
| fma | 0% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 0% |
| all | 100% |
| load | 100% |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | NA (no mul vectorizable/vectorized instructions) |
| add-sub | NA (no add-sub vectorizable/vectorized instructions) |
| fma | NA (no fma vectorizable/vectorized instructions) |
| other | NA (no other vectorizable/vectorized instructions) |
| all | 50% |
| load | NA (no load vectorizable/vectorized instructions) |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 50% |
| add-sub | 50% |
| fma | 50% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 50% |
| all | 58% |
| load | 100% |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 50% |
| add-sub | 50% |
| fma | 50% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 50% |
| Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | P12 | P13 | P14 | P15 | P16 | Latency | Recip. throughput | Vectorization |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| LDP D31, D30, [X0], #16 | 1 | 0 | 0 | 0.17 | 0.17 | 0.17 | 0.17 | 0.17 | 0.17 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 | vect (100.0%) |
| FSUB D30, D27, S30 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 | scal (50.0%) |
| FSUB D31, D28, S31 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 | scal (50.0%) |
| FMUL D30, D30, D30 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 3 | 0.25 | scal (50.0%) |
| FMADD D31, D31, D31, D30 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 4 | 0.25 | scal (50.0%) |
| FCMPE D29, D31 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 | scal (50.0%) |
| B.LS 401c44 <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0xa4> | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 | N/A |
| ADD W1, W1, #1 | 1 | 0 | 0 | 0.17 | 0.17 | 0.17 | 0.17 | 0.17 | 0.17 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 | N/A |
| CMP W2, W1 | 1 | 0 | 0 | 0.25 | 0.25 | 0 | 0 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.33 | N/A |
| B.NE 401c20 <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x80> | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 | N/A |
| Run run_1_thread | Number processes: 1Number nodes: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: trueOMP_NUM_THREADS: 1 |
|---|---|
| Run run_2_threads | Number processes: 1Number nodes: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: trueOMP_NUM_THREADS: 2 |
| Run run_4_threads | Number processes: 1Number nodes: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: trueOMP_NUM_THREADS: 4 |
| Run run_8_threads | Number processes: 1Number nodes: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: trueOMP_NUM_THREADS: 8 |
| Run run_16_threads | Number processes: 1Number nodes: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: trueOMP_NUM_THREADS: 16 |
| Run run_32_threads | Number processes: 1Number nodes: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: trueOMP_NUM_THREADS: 32 |
| Run run_48_threads | Number processes: 1Number nodes: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: trueOMP_NUM_THREADS: 48 |
| Run run_64_threads | Number processes: 1Number nodes: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: trueOMP_NUM_THREADS: 64 |
| Run run_80_threads | Number processes: 1Number nodes: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: trueOMP_NUM_THREADS: 80 |
| Run run_96_threads | Number processes: 1Number nodes: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: trueOMP_NUM_THREADS: 96 |
| (run_1_thread) Efficiency | (run_1_thread) Potential Speed-Up (%) | (run_2_threads) Efficiency | (run_2_threads) Potential Speed-Up (%) | (run_4_threads) Efficiency | (run_4_threads) Potential Speed-Up (%) | (run_8_threads) Efficiency | (run_8_threads) Potential Speed-Up (%) | (run_16_threads) Efficiency | (run_16_threads) Potential Speed-Up (%) | (run_32_threads) Efficiency | (run_32_threads) Potential Speed-Up (%) | (run_48_threads) Efficiency | (run_48_threads) Potential Speed-Up (%) | (run_64_threads) Efficiency | (run_64_threads) Potential Speed-Up (%) | (run_80_threads) Efficiency | (run_80_threads) Potential Speed-Up (%) | (run_96_threads) Efficiency | (run_96_threads) Potential Speed-Up (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 0 | 0.94 | 5.92 | 0.83 | 15.74 | 0.67 | 30.78 | 0.49 | 47.06 | 0.32 | 62.89 | 0.23 | 70.6 | 0.18 | 75.1 | 0.15 | 78.39 | 0.13 | 80.91 |
| Run | Number of threads | Efficiency (ideal is 1) | Speedup | Ideal Speedup | Time (s) | Coverage (%) |
|---|---|---|---|---|---|---|
| run_1_thread | 1 | 1 | 1 | 1 | 98.719985961914 | 92.304817199707 |
| run_2_threads | 2 | 0.94 | 1.87 | 2 | 49.364994049072 | 92.17366027832 |
| run_4_threads | 4 | 0.83 | 3.32 | 4 | 24.780002593994 | 92.110054016113 |
| run_8_threads | 8 | 0.67 | 5.33 | 8 | 12.565000534058 | 92.337898254395 |
| run_16_threads | 16 | 0.49 | 7.83 | 16 | 6.3999991416931 | 92.216598510742 |
| run_32_threads | 32 | 0.32 | 10.16 | 32 | 3.2699997425079 | 92.135917663574 |
| run_48_threads | 48 | 0.23 | 11.23 | 48 | 2.2049999237061 | 92.168838500977 |
| run_64_threads | 64 | 0.18 | 11.84 | 64 | 1.6699999570847 | 92.147789001465 |
| run_80_threads | 80 | 0.15 | 12.15 | 80 | 1.4399996995926 | 92.41584777832 |
| run_96_threads | 96 | 0.13 | 12.25 | 96 | 1.2999999523163 | 92.755165100098 |
