| Loop Id: 1 | Module: kmeans-gcc-O3-funroll | Source: main.cpp:61-67 | Coverage: 87.41% |
|---|
| Loop Id: 1 | Module: kmeans-gcc-O3-funroll | Source: main.cpp:61-67 | Coverage: 87.41% |
|---|
0x2ae9 VSUBSD 0x8(%RBX),%XMM2,%XMM14 [1] |
0x2aee VSUBSD (%RBX),%XMM3,%XMM13 [1] |
0x2af2 VMULSD %XMM14,%XMM14,%XMM15 |
0x2af7 VFMADD132SD %XMM13,%XMM15,%XMM13 |
0x2afc VCOMISD %XMM13,%XMM0 |
0x2b01 JBE 2b09 |
0x2b03 MOV %EDX,(%RSI) [3] |
0x2b05 VMOVAPD %XMM13,%XMM0 |
0x2b09 VSUBSD 0x18(%RBX),%XMM2,%XMM4 [1] |
0x2b0e VSUBSD 0x10(%RBX),%XMM3,%XMM1 [1] |
0x2b13 INC %EDX |
0x2b15 LEA 0x10(%RBX),%R11 |
0x2b19 VMULSD %XMM4,%XMM4,%XMM6 |
0x2b1d VFMADD132SD %XMM1,%XMM6,%XMM1 |
0x2b22 VCOMISD %XMM1,%XMM0 |
0x2b26 JBE 2b2e |
0x2b28 MOV %EDX,(%RSI) [3] |
0x2b2a VMOVAPD %XMM1,%XMM0 |
0x2b2e VSUBSD 0x18(%R11),%XMM2,%XMM8 [2] |
0x2b34 VSUBSD 0x10(%R11),%XMM3,%XMM7 [2] |
0x2b3a LEA 0x1(%RDX),%EAX |
0x2b3d VMULSD %XMM8,%XMM8,%XMM9 |
0x2b42 VFMADD132SD %XMM7,%XMM9,%XMM7 |
0x2b47 VCOMISD %XMM7,%XMM0 |
0x2b4b JBE 2b53 |
0x2b4d MOV %EAX,(%RSI) [3] |
0x2b4f VMOVAPD %XMM7,%XMM0 |
0x2b53 VSUBSD 0x28(%R11),%XMM2,%XMM11 [2] |
0x2b59 VSUBSD 0x20(%R11),%XMM3,%XMM10 [2] |
0x2b5f LEA 0x2(%RDX),%R12D |
0x2b63 VMULSD %XMM11,%XMM11,%XMM12 |
0x2b68 VFMADD132SD %XMM10,%XMM12,%XMM10 |
0x2b6d VCOMISD %XMM10,%XMM0 |
0x2b72 JBE 2b7b |
0x2b74 MOV %R12D,(%RSI) [3] |
0x2b77 VMOVAPD %XMM10,%XMM0 |
0x2b7b ADD $0x3,%EDX |
0x2b7e LEA 0x30(%R11),%RBX |
0x2b82 CMP %EDX,%R8D |
0x2b85 JNE 2ae9 |
/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 |
|---|---|---|---|
| ►98.61+ | GOMP_parallel | parallel.c:140 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-O3-funroll |
| ○ | main | main.cpp:20 | kmeans-gcc-O3-funroll |
| ○ | __libc_init_first | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | basic_string.h:809 | kmeans-gcc-O3-funroll |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►48.94+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | pthread_condattr_setpshared | libc.so.6 | |
| ○ | __clone | libc.so.6 | |
| ►48.31+ | GOMP_parallel | parallel.c:140 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-O3-funroll |
| ○ | main | main.cpp:20 | kmeans-gcc-O3-funroll |
| ○ | __libc_init_first | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | basic_string.h:809 | kmeans-gcc-O3-funroll |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►69.73+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | pthread_condattr_setpshared | libc.so.6 | |
| ○ | __clone | libc.so.6 | |
| ►23.69+ | GOMP_parallel | parallel.c:140 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-O3-funroll |
| ○ | main | main.cpp:20 | kmeans-gcc-O3-funroll |
| ○ | __libc_init_first | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | basic_string.h:809 | kmeans-gcc-O3-funroll |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►86.48+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | pthread_condattr_setpshared | libc.so.6 | |
| ○ | __clone | libc.so.6 | |
| ►12.53+ | GOMP_parallel | parallel.c:140 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-O3-funroll |
| ○ | main | main.cpp:20 | kmeans-gcc-O3-funroll |
| ○ | __libc_init_first | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | basic_string.h:809 | kmeans-gcc-O3-funroll |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►93.47+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | pthread_condattr_setpshared | libc.so.6 | |
| ○ | __clone | libc.so.6 | |
| ►6.27+ | GOMP_parallel | parallel.c:140 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-O3-funroll |
| ○ | main | main.cpp:20 | kmeans-gcc-O3-funroll |
| ○ | __libc_init_first | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | basic_string.h:809 | kmeans-gcc-O3-funroll |
| Coverage (%) | Name | Source Location | Module |
|---|---|---|---|
| ►96.03+ | gomp_thread_start | team.c:130 | libgomp.so.1.0.0 |
| ○ | pthread_condattr_setpshared | libc.so.6 | |
| ○ | __clone | libc.so.6 | |
| ►3.84+ | GOMP_parallel | parallel.c:140 | libgomp.so.1.0.0 |
| ○ | k_means(int, point_t*, point_t[...] | main.cpp:73 | kmeans-gcc-O3-funroll |
| ○ | main | main.cpp:20 | kmeans-gcc-O3-funroll |
| ○ | __libc_init_first | libc.so.6 | |
| ○ | __libc_start_main | libc.so.6 | |
| ○ | _start | basic_string.h:809 | kmeans-gcc-O3-funroll |
| 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 | 1.00 |
| CQA speedup if FP arith vectorized | 2.50 |
| CQA speedup if fully vectorized | 8.00 |
| CQA speedup if no inter-iteration dependency | NA |
| CQA speedup if next bottleneck killed | 1.03 |
| Bottlenecks | P0, P1, |
| 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 | 10.00 |
| CQA cycles if no scalar integer | 10.00 |
| CQA cycles if FP arith vectorized | 4.00 |
| CQA cycles if fully vectorized | 1.25 |
| Front-end cycles | 9.75 |
| P0 cycles | 10.00 |
| P1 cycles | 10.00 |
| P2 cycles | 4.00 |
| P3 cycles | 4.00 |
| P4 cycles | 4.00 |
| P5 cycles | 5.50 |
| P6 cycles | 5.50 |
| P7 cycles | 4.00 |
| DIV/SQRT cycles | 0.00 |
| Inter-iter dependencies cycles | NA |
| FE+BE cycles (UFS) | 10.37 |
| Stall cycles (UFS) | 0.07 |
| Nb insns | 40.00 |
| Nb uops | 39.00 |
| Nb loads | 8.00 |
| Nb stores | 4.00 |
| Nb stack references | 0.00 |
| FLOP/cycle | 2.00 |
| Nb FLOP add-sub | 8.00 |
| Nb FLOP mul | 4.00 |
| Nb FLOP fma | 4.00 |
| Nb FLOP div | 0.00 |
| Nb FLOP rcp | 0.00 |
| Nb FLOP sqrt | 0.00 |
| Nb FLOP rsqrt | 0.00 |
| Bytes/cycle | 8.00 |
| Bytes prefetched | 0.00 |
| Bytes loaded | 64.00 |
| Bytes stored | 16.00 |
| Stride 0 | NA |
| Stride 1 | NA |
| Stride n | NA |
| Stride unknown | NA |
| Stride indirect | NA |
| Vectorization ratio all | 14.29 |
| Vectorization ratio load | 0.00 |
| 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 | 50.00 |
| Vector-efficiency ratio all | 13.39 |
| Vector-efficiency ratio load | 12.50 |
| Vector-efficiency ratio store | 6.25 |
| Vector-efficiency ratio mul | 12.50 |
| Vector-efficiency ratio add_sub | 12.50 |
| Vector-efficiency ratio fma | 12.50 |
| Vector-efficiency ratio div_sqrt | NA |
| Vector-efficiency ratio other | 18.75 |
| Metric | Value |
|---|---|
| CQA speedup if no scalar integer | 1.00 |
| CQA speedup if FP arith vectorized | 2.50 |
| CQA speedup if fully vectorized | 8.00 |
| CQA speedup if no inter-iteration dependency | NA |
| CQA speedup if next bottleneck killed | 1.03 |
| Bottlenecks | P0, P1, |
| 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 | 10.00 |
| CQA cycles if no scalar integer | 10.00 |
| CQA cycles if FP arith vectorized | 4.00 |
| CQA cycles if fully vectorized | 1.25 |
| Front-end cycles | 9.75 |
| P0 cycles | 10.00 |
| P1 cycles | 10.00 |
| P2 cycles | 4.00 |
| P3 cycles | 4.00 |
| P4 cycles | 4.00 |
| P5 cycles | 5.50 |
| P6 cycles | 5.50 |
| P7 cycles | 4.00 |
| DIV/SQRT cycles | 0.00 |
| Inter-iter dependencies cycles | NA |
| FE+BE cycles (UFS) | 10.37 |
| Stall cycles (UFS) | 0.07 |
| Nb insns | 40.00 |
| Nb uops | 39.00 |
| Nb loads | 8.00 |
| Nb stores | 4.00 |
| Nb stack references | 0.00 |
| FLOP/cycle | 2.00 |
| Nb FLOP add-sub | 8.00 |
| Nb FLOP mul | 4.00 |
| Nb FLOP fma | 4.00 |
| Nb FLOP div | 0.00 |
| Nb FLOP rcp | 0.00 |
| Nb FLOP sqrt | 0.00 |
| Nb FLOP rsqrt | 0.00 |
| Bytes/cycle | 8.00 |
| Bytes prefetched | 0.00 |
| Bytes loaded | 64.00 |
| Bytes stored | 16.00 |
| Stride 0 | NA |
| Stride 1 | NA |
| Stride n | NA |
| Stride unknown | NA |
| Stride indirect | NA |
| Vectorization ratio all | 14.29 |
| Vectorization ratio load | 0.00 |
| 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 | 50.00 |
| Vector-efficiency ratio all | 13.39 |
| Vector-efficiency ratio load | 12.50 |
| Vector-efficiency ratio store | 6.25 |
| Vector-efficiency ratio mul | 12.50 |
| Vector-efficiency ratio add_sub | 12.50 |
| Vector-efficiency ratio fma | 12.50 |
| Vector-efficiency ratio div_sqrt | NA |
| Vector-efficiency ratio other | 18.75 |
| 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-O3-funroll |
| nb instructions | 40 |
| nb uops | 39 |
| loop length | 162 |
| used x86 registers | 7 |
| used mmx registers | 0 |
| used xmm registers | 15 |
| used ymm registers | 0 |
| used zmm registers | 0 |
| nb stack references | 0 |
| ADD-SUB / MUL ratio | 2.00 |
| micro-operation queue | 9.75 cycles |
| front end | 9.75 cycles |
| P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
|---|---|---|---|---|---|---|---|---|
| uops | 10.00 | 10.00 | 4.00 | 4.00 | 4.00 | 5.50 | 5.50 | 4.00 |
| cycles | 10.00 | 10.00 | 4.00 | 4.00 | 4.00 | 5.50 | 5.50 | 4.00 |
| Cycles executing div or sqrt instructions | NA |
| FE+BE cycles | 10.37 |
| Stall cycles | 0.07 |
| RS full (events) | 0.29 |
| Front-end | 9.75 |
| Dispatch | 10.00 |
| Overall L1 | 10.00 |
| all | 0% |
| load | NA (no load vectorizable/vectorized instructions) |
| 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 | 16% |
| load | 0% |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 0% |
| add-sub | 0% |
| fma | 0% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 50% |
| all | 14% |
| load | 0% |
| store | 0% |
| mul | 0% |
| add-sub | 0% |
| fma | 0% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 50% |
| all | 6% |
| load | NA (no load vectorizable/vectorized instructions) |
| store | 6% |
| 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 | 14% |
| load | 12% |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 12% |
| add-sub | 12% |
| fma | 12% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 18% |
| all | 13% |
| load | 12% |
| store | 6% |
| mul | 12% |
| add-sub | 12% |
| fma | 12% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 18% |
| Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | Latency | Recip. throughput | Vectorization |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| VSUBSD 0x8(%RBX),%XMM2,%XMM14 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VSUBSD (%RBX),%XMM3,%XMM13 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VMULSD %XMM14,%XMM14,%XMM15 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VFMADD132SD %XMM13,%XMM15,%XMM13 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VCOMISD %XMM13,%XMM0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 | scal (12.5%) |
| JBE 2b09 <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x149> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 | N/A |
| MOV %EDX,(%RSI) | 1 | 0 | 0 | 0.33 | 0.33 | 1 | 0 | 0 | 0.33 | 3 | 1 | scal (6.3%) |
| VMOVAPD %XMM13,%XMM0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | vect (25.0%) |
| VSUBSD 0x18(%RBX),%XMM2,%XMM4 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VSUBSD 0x10(%RBX),%XMM3,%XMM1 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| INC %EDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 | N/A |
| LEA 0x10(%RBX),%R11 | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
| VMULSD %XMM4,%XMM4,%XMM6 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VFMADD132SD %XMM1,%XMM6,%XMM1 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VCOMISD %XMM1,%XMM0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 | scal (12.5%) |
| JBE 2b2e <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x16e> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 | N/A |
| MOV %EDX,(%RSI) | 1 | 0 | 0 | 0.33 | 0.33 | 1 | 0 | 0 | 0.33 | 3 | 1 | scal (6.3%) |
| VMOVAPD %XMM1,%XMM0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | vect (25.0%) |
| VSUBSD 0x18(%R11),%XMM2,%XMM8 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VSUBSD 0x10(%R11),%XMM3,%XMM7 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| LEA 0x1(%RDX),%EAX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
| VMULSD %XMM8,%XMM8,%XMM9 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VFMADD132SD %XMM7,%XMM9,%XMM7 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VCOMISD %XMM7,%XMM0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 | scal (12.5%) |
| JBE 2b53 <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x193> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 | N/A |
| MOV %EAX,(%RSI) | 1 | 0 | 0 | 0.33 | 0.33 | 1 | 0 | 0 | 0.33 | 3 | 1 | scal (6.3%) |
| VMOVAPD %XMM7,%XMM0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | vect (25.0%) |
| VSUBSD 0x28(%R11),%XMM2,%XMM11 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VSUBSD 0x20(%R11),%XMM3,%XMM10 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| LEA 0x2(%RDX),%R12D | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
| VMULSD %XMM11,%XMM11,%XMM12 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VFMADD132SD %XMM10,%XMM12,%XMM10 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VCOMISD %XMM10,%XMM0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 | scal (12.5%) |
| JBE 2b7b <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x1bb> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 | N/A |
| MOV %R12D,(%RSI) | 1 | 0 | 0 | 0.33 | 0.33 | 1 | 0 | 0 | 0.33 | 3 | 1 | scal (6.3%) |
| VMOVAPD %XMM10,%XMM0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | vect (25.0%) |
| ADD $0x3,%EDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 | N/A |
| LEA 0x30(%R11),%RBX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
| CMP %EDX,%R8D | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 | N/A |
| JNE 2ae9 <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x129> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 | 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-O3-funroll |
| nb instructions | 40 |
| nb uops | 39 |
| loop length | 162 |
| used x86 registers | 7 |
| used mmx registers | 0 |
| used xmm registers | 15 |
| used ymm registers | 0 |
| used zmm registers | 0 |
| nb stack references | 0 |
| ADD-SUB / MUL ratio | 2.00 |
| micro-operation queue | 9.75 cycles |
| front end | 9.75 cycles |
| P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
|---|---|---|---|---|---|---|---|---|
| uops | 10.00 | 10.00 | 4.00 | 4.00 | 4.00 | 5.50 | 5.50 | 4.00 |
| cycles | 10.00 | 10.00 | 4.00 | 4.00 | 4.00 | 5.50 | 5.50 | 4.00 |
| Cycles executing div or sqrt instructions | NA |
| FE+BE cycles | 10.37 |
| Stall cycles | 0.07 |
| RS full (events) | 0.29 |
| Front-end | 9.75 |
| Dispatch | 10.00 |
| Overall L1 | 10.00 |
| all | 0% |
| load | NA (no load vectorizable/vectorized instructions) |
| 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 | 16% |
| load | 0% |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 0% |
| add-sub | 0% |
| fma | 0% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 50% |
| all | 14% |
| load | 0% |
| store | 0% |
| mul | 0% |
| add-sub | 0% |
| fma | 0% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 50% |
| all | 6% |
| load | NA (no load vectorizable/vectorized instructions) |
| store | 6% |
| 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 | 14% |
| load | 12% |
| store | NA (no store vectorizable/vectorized instructions) |
| mul | 12% |
| add-sub | 12% |
| fma | 12% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 18% |
| all | 13% |
| load | 12% |
| store | 6% |
| mul | 12% |
| add-sub | 12% |
| fma | 12% |
| div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
| other | 18% |
| Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | Latency | Recip. throughput | Vectorization |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| VSUBSD 0x8(%RBX),%XMM2,%XMM14 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VSUBSD (%RBX),%XMM3,%XMM13 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VMULSD %XMM14,%XMM14,%XMM15 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VFMADD132SD %XMM13,%XMM15,%XMM13 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VCOMISD %XMM13,%XMM0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 | scal (12.5%) |
| JBE 2b09 <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x149> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 | N/A |
| MOV %EDX,(%RSI) | 1 | 0 | 0 | 0.33 | 0.33 | 1 | 0 | 0 | 0.33 | 3 | 1 | scal (6.3%) |
| VMOVAPD %XMM13,%XMM0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | vect (25.0%) |
| VSUBSD 0x18(%RBX),%XMM2,%XMM4 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VSUBSD 0x10(%RBX),%XMM3,%XMM1 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| INC %EDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 | N/A |
| LEA 0x10(%RBX),%R11 | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
| VMULSD %XMM4,%XMM4,%XMM6 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VFMADD132SD %XMM1,%XMM6,%XMM1 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VCOMISD %XMM1,%XMM0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 | scal (12.5%) |
| JBE 2b2e <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x16e> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 | N/A |
| MOV %EDX,(%RSI) | 1 | 0 | 0 | 0.33 | 0.33 | 1 | 0 | 0 | 0.33 | 3 | 1 | scal (6.3%) |
| VMOVAPD %XMM1,%XMM0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | vect (25.0%) |
| VSUBSD 0x18(%R11),%XMM2,%XMM8 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VSUBSD 0x10(%R11),%XMM3,%XMM7 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| LEA 0x1(%RDX),%EAX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
| VMULSD %XMM8,%XMM8,%XMM9 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VFMADD132SD %XMM7,%XMM9,%XMM7 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VCOMISD %XMM7,%XMM0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 | scal (12.5%) |
| JBE 2b53 <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x193> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 | N/A |
| MOV %EAX,(%RSI) | 1 | 0 | 0 | 0.33 | 0.33 | 1 | 0 | 0 | 0.33 | 3 | 1 | scal (6.3%) |
| VMOVAPD %XMM7,%XMM0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | vect (25.0%) |
| VSUBSD 0x28(%R11),%XMM2,%XMM11 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VSUBSD 0x20(%R11),%XMM3,%XMM10 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| LEA 0x2(%RDX),%R12D | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
| VMULSD %XMM11,%XMM11,%XMM12 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VFMADD132SD %XMM10,%XMM12,%XMM10 | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0.50 | scal (12.5%) |
| VCOMISD %XMM10,%XMM0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 | scal (12.5%) |
| JBE 2b7b <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x1bb> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 | N/A |
| MOV %R12D,(%RSI) | 1 | 0 | 0 | 0.33 | 0.33 | 1 | 0 | 0 | 0.33 | 3 | 1 | scal (6.3%) |
| VMOVAPD %XMM10,%XMM0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | vect (25.0%) |
| ADD $0x3,%EDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 | N/A |
| LEA 0x30(%R11),%RBX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
| CMP %EDX,%R8D | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 | N/A |
| JNE 2ae9 <_Z7k_meansiP7point_tS0_PiS0_ii._omp_fn.0+0x129> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 | N/A |
| Run run_1_thread | Number processes: 1Number nodes: 1Number processes per node: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: closeOMP_NUM_THREADS: 1 |
|---|---|
| Run run_2_threads | Number processes: 1Number nodes: 1Number processes per node: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: closeOMP_NUM_THREADS: 2 |
| Run run_4_threads | Number processes: 1Number nodes: 1Number processes per node: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: closeOMP_NUM_THREADS: 4 |
| Run run_8_threads | Number processes: 1Number nodes: 1Number processes per node: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: closeOMP_NUM_THREADS: 8 |
| Run run_16_threads | Number processes: 1Number nodes: 1Number processes per node: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: closeOMP_NUM_THREADS: 16 |
| Run run_26_threads | Number processes: 1Number nodes: 1Number processes per node: 1Run Command: <executable> input/100000000.in 1000 100000000 50 25MPI Command: Dataset: Run Directory: /home/fmusial/KMEANS_BenchmarksOMP_PROC_BIND: closeOMP_NUM_THREADS: 26 |
| (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_26_threads) Efficiency | (run_26_threads) Potential Speed-Up (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 0 | 0.94 | 5.29 | 0.84 | 13.97 | 0.7 | 25.74 | 0.53 | 39.39 | 0.41 | 47.67 |
| Run | Number of threads | Efficiency (ideal is 1) | Speedup | Ideal Speedup | Time (s) | Coverage (%) |
|---|---|---|---|---|---|---|
| run_1_thread | 1 | 1 | 1 | 1 | 142.26498413086 | 87.405151367188 |
| run_2_threads | 2 | 0.94 | 1.88 | 2 | 71.31999206543 | 87.243278503418 |
| run_4_threads | 4 | 0.84 | 3.36 | 4 | 35.814994812012 | 86.764709472656 |
| run_8_threads | 8 | 0.7 | 5.58 | 8 | 17.834999084473 | 85.270050048828 |
| run_16_threads | 16 | 0.53 | 8.41 | 16 | 8.9849996566772 | 83.025085449219 |
| run_26_threads | 26 | 0.41 | 10.58 | 26 | 5.5699996948242 | 80.374389648438 |
