Loop Id: 230 | Module: exec | Source: BsplineFunctor.h:236-241 | Coverage: 2.21% |
---|
Loop Id: 230 | Module: exec | Source: BsplineFunctor.h:236-241 | Coverage: 2.21% |
---|
0x41b8d0 VMOVDQA32 %ZMM10,%ZMM7{%K1} |
0x41b8d6 VEXTRACTI64X4 $0x1,%ZMM7,%YMM8 |
0x41b8dd VPADDD %ZMM8,%ZMM7,%ZMM7 |
0x41b8e3 VEXTRACTI128 $0x1,%YMM7,%XMM8 |
0x41b8e9 VPADDD %ZMM8,%ZMM7,%ZMM7 |
0x41b8ef VPSHUFD $-0x12,%XMM7,%XMM8 |
0x41b8f4 VPADDD %ZMM8,%ZMM7,%ZMM7 |
0x41b8fa VPSHUFD $0x55,%XMM7,%XMM8 |
0x41b8ff VPADDD %ZMM8,%ZMM7,%ZMM8 |
0x41b905 VPBLENDW $0x3,%XMM8,%XMM5,%XMM7 |
0x41b90b ADD $0x10,%RAX |
0x41b90f CMP %RCX,%RAX |
0x41b912 JAE 41b98d |
0x41b914 VMOVUPD (%R15,%RAX,8),%ZMM9 [1] |
0x41b91b VMOVUPD 0x40(%R15,%RAX,8),%ZMM8 [1] |
0x41b923 LEA (%R14,%RAX,1),%EBX |
0x41b927 VPBROADCASTD %EBX,%ZMM10 |
0x41b92d VPADDD %ZMM4,%ZMM10,%ZMM10 |
0x41b933 VCMPPD $0x1,%ZMM3,%ZMM9,%K0 |
0x41b93a VCMPPD $0x1,%ZMM3,%ZMM8,%K1 |
0x41b941 KUNPCKBW %K0,%K1,%K1 |
0x41b945 VPCMPNEQD %ZMM2,%ZMM10,%K1{%K1} |
0x41b94c KORTESTW %K1,%K1 |
0x41b950 VPXOR %XMM10,%XMM10,%XMM10 |
0x41b955 JE 41b8d0 |
0x41b95b VMOVD %XMM7,%EBX |
0x41b95f MOVSXD %EBX,%RBX |
0x41b962 LEA (%R9,%RBX,8),%R12 |
0x41b966 VCOMPRESSPD %ZMM9,(%R9,%RBX,8){%K1} [2] |
0x41b96d KSHIFTRW $0x8,%K1,%K2 |
0x41b973 KMOVB %K1,%EBX |
0x41b977 POPCNT %EBX,%EBX |
0x41b97b VCOMPRESSPD %ZMM8,(%R12,%RBX,8){%K2} [3] |
0x41b982 VPSUBD %ZMM6,%ZMM7,%ZMM10 |
0x41b988 JMP 41b8d0 |
/home/kcamus/qaas_runs/169-390-4082/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/BsplineFunctor.h: 236 - 241 |
-------------------------------------------------------------------------------- |
236: for (int jat = 0; jat < iLimit; jat++) |
237: { |
238: real_type r = distArray[jat]; |
239: // pick the distances smaller than the cutoff and avoid the reference atom |
240: if (r < cutoff_radius && iStart + jat != iat) |
241: distArrayCompressed[iCount++] = distArray[jat]; |
Coverage (%) | Name | Source Location | Module |
---|---|---|---|
►100.00+ | miniqmcreference::TwoBodyJastr[...] | TwoBodyJastrowRef.h:130 | exec |
○ | qmcplusplus::WaveFunction::rat[...] | WaveFunction.cpp:219 | exec |
○ | main.extracted.104 | stl_vector.h:1126 | exec |
○ | __kmp_invoke_microtask | libiomp5.so | |
○ | __kmp_fork_call | libiomp5.so | |
○ | __kmpc_fork_call | libiomp5.so | |
○ | main | miniqmc.cpp:404 | exec |
○ | __libc_init_first | libc.so.6 |
Path / |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.15 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 1.53 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.48 |
Bottlenecks | |
Function | qmcplusplus::BsplineFunctor |
Source | BsplineFunctor.h:236-241 |
Source loop unroll info | unrolled by 16 |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | main |
Unroll factor | 16 |
CQA cycles | 11.50 |
CQA cycles if no scalar integer | 10.00 |
CQA cycles if FP arith vectorized | 11.50 |
CQA cycles if fully vectorized | 7.53 |
Front-end cycles | 7.75 |
DIV/SQRT cycles | 7.50 |
P0 cycles | 3.50 |
P1 cycles | 1.17 |
P2 cycles | 1.17 |
P3 cycles | 1.00 |
P4 cycles | 11.50 |
P5 cycles | 3.50 |
P6 cycles | 0.67 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 12.15 |
Stall cycles (UFS) | 3.96 |
Nb insns | 30.00 |
Nb uops | 31.00 |
Nb loads | 2.00 |
Nb stores | 1.00 |
Nb stack references | 0.00 |
FLOP/cycle | 0.00 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 0.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 16.25 |
Bytes prefetched | 0.00 |
Bytes loaded | 128.00 |
Bytes stored | 64.00 |
Stride 0 | 0.00 |
Stride 1 | 1.00 |
Stride n | 0.50 |
Stride unknown | 0.50 |
Stride indirect | 0.00 |
Vectorization ratio all | 92.68 |
Vectorization ratio load | 100.00 |
Vectorization ratio store | 100.00 |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | 100.00 |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 87.12 |
Vector-efficiency ratio all | 71.67 |
Vector-efficiency ratio load | 100.00 |
Vector-efficiency ratio store | 100.00 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | 100.00 |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 50.90 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.17 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 1.46 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.47 |
Bottlenecks | P5, |
Function | qmcplusplus::BsplineFunctor |
Source | BsplineFunctor.h:236-241 |
Source loop unroll info | unrolled by 16 |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | main |
Unroll factor | 16 |
CQA cycles | 14.00 |
CQA cycles if no scalar integer | 12.00 |
CQA cycles if FP arith vectorized | 14.00 |
CQA cycles if fully vectorized | 9.56 |
Front-end cycles | 9.50 |
DIV/SQRT cycles | 9.00 |
P0 cycles | 4.50 |
P1 cycles | 1.33 |
P2 cycles | 1.33 |
P3 cycles | 2.00 |
P4 cycles | 14.00 |
P5 cycles | 4.50 |
P6 cycles | 1.33 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 15.17 |
Stall cycles (UFS) | 5.19 |
Nb insns | 35.00 |
Nb uops | 38.00 |
Nb loads | 2.00 |
Nb stores | 2.00 |
Nb stack references | 0.00 |
FLOP/cycle | 0.00 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 0.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 18.29 |
Bytes prefetched | 0.00 |
Bytes loaded | 128.00 |
Bytes stored | 128.00 |
Stride 0 | 0.00 |
Stride 1 | 1.00 |
Stride n | 1.00 |
Stride unknown | 1.00 |
Stride indirect | 0.00 |
Vectorization ratio all | 90.91 |
Vectorization ratio load | 100.00 |
Vectorization ratio store | 100.00 |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | 100.00 |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 83.33 |
Vector-efficiency ratio all | 72.16 |
Vector-efficiency ratio load | 100.00 |
Vector-efficiency ratio store | 100.00 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | 100.00 |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 48.96 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.13 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 1.64 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.50 |
Bottlenecks | P5, |
Function | qmcplusplus::BsplineFunctor |
Source | BsplineFunctor.h:236-241 |
Source loop unroll info | unrolled by 16 |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | main |
Unroll factor | 16 |
CQA cycles | 9.00 |
CQA cycles if no scalar integer | 8.00 |
CQA cycles if FP arith vectorized | 9.00 |
CQA cycles if fully vectorized | 5.50 |
Front-end cycles | 6.00 |
DIV/SQRT cycles | 6.00 |
P0 cycles | 2.50 |
P1 cycles | 1.00 |
P2 cycles | 1.00 |
P3 cycles | 0.00 |
P4 cycles | 9.00 |
P5 cycles | 2.50 |
P6 cycles | 0.00 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 9.14 |
Stall cycles (UFS) | 2.73 |
Nb insns | 25.00 |
Nb uops | 24.00 |
Nb loads | 2.00 |
Nb stores | 0.00 |
Nb stack references | 0.00 |
FLOP/cycle | 0.00 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 0.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 14.22 |
Bytes prefetched | 0.00 |
Bytes loaded | 128.00 |
Bytes stored | 0.00 |
Stride 0 | 0.00 |
Stride 1 | 1.00 |
Stride n | 0.00 |
Stride unknown | 0.00 |
Stride indirect | 0.00 |
Vectorization ratio all | 94.44 |
Vectorization ratio load | 100.00 |
Vectorization ratio store | NA |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | 100.00 |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 90.91 |
Vector-efficiency ratio all | 71.18 |
Vector-efficiency ratio load | 100.00 |
Vector-efficiency ratio store | NA |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | 100.00 |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 52.84 |
Path / |
Function | qmcplusplus::BsplineFunctor |
Source file and lines | BsplineFunctor.h:236-241 |
Module | exec |
nb instructions | 30 |
nb uops | 31 |
loop length | 164 |
used x86 registers | 6 |
used mmx registers | 0 |
used xmm registers | 4 |
used ymm registers | 2 |
used zmm registers | 7.50 |
nb stack references | 0 |
micro-operation queue | 7.75 cycles |
front end | 7.75 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 7.50 | 3.50 | 1.17 | 1.17 | 1.00 | 11.50 | 3.50 | 0.67 |
cycles | 7.50 | 3.50 | 1.17 | 1.17 | 1.00 | 11.50 | 3.50 | 0.67 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 12.16 |
Stall cycles | 3.96 |
RS full (events) | 8.85 |
Front-end | 7.75 |
Dispatch | 11.50 |
Data deps. | 1.00 |
Overall L1 | 11.50 |
all | 90% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 84% |
all | 100% |
load | 100% |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 100% |
all | 92% |
load | 100% |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 87% |
all | 62% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 40% |
all | 100% |
load | 100% |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 100% |
all | 71% |
load | 100% |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 50% |
Function | qmcplusplus::BsplineFunctor |
Source file and lines | BsplineFunctor.h:236-241 |
Module | exec |
nb instructions | 35 |
nb uops | 38 |
loop length | 189 |
used x86 registers | 7 |
used mmx registers | 0 |
used xmm registers | 4 |
used ymm registers | 2 |
used zmm registers | 8 |
nb stack references | 0 |
micro-operation queue | 9.50 cycles |
front end | 9.50 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 9.00 | 4.50 | 1.33 | 1.33 | 2.00 | 14.00 | 4.50 | 1.33 |
cycles | 9.00 | 4.50 | 1.33 | 1.33 | 2.00 | 14.00 | 4.50 | 1.33 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 15.17 |
Stall cycles | 5.19 |
RS full (events) | 11.32 |
Front-end | 9.50 |
Dispatch | 14.00 |
Data deps. | 1.00 |
Overall L1 | 14.00 |
all | 87% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 80% |
all | 100% |
load | 100% |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 100% |
all | 90% |
load | 100% |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 83% |
all | 61% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 38% |
all | 100% |
load | 100% |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 100% |
all | 72% |
load | 100% |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 48% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|
VMOVDQA32 %ZMM10,%ZMM7{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VEXTRACTI64X4 $0x1,%ZMM7,%YMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VPADDD %ZMM8,%ZMM7,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VEXTRACTI128 $0x1,%YMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VPADDD %ZMM8,%ZMM7,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPSHUFD $-0x12,%XMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM8,%ZMM7,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPSHUFD $0x55,%XMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM8,%ZMM7,%ZMM8 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPBLENDW $0x3,%XMM8,%XMM5,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
ADD $0x10,%RAX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
CMP %RCX,%RAX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
JAE 41b98d | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
VMOVUPD (%R15,%RAX,8),%ZMM9 | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVUPD 0x40(%R15,%RAX,8),%ZMM8 | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
LEA (%R14,%RAX,1),%EBX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPBROADCASTD %EBX,%ZMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM4,%ZMM10,%ZMM10 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VCMPPD $0x1,%ZMM3,%ZMM9,%K0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VCMPPD $0x1,%ZMM3,%ZMM8,%K1 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
KUNPCKBW %K0,%K1,%K1 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 4 | 1 |
VPCMPNEQD %ZMM2,%ZMM10,%K1{%K1} | |||||||||||
KORTESTW %K1,%K1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 |
VPXOR %XMM10,%XMM10,%XMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
JE 41b8d0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
VMOVD %XMM7,%EBX | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
MOVSXD %EBX,%RBX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
LEA (%R9,%RBX,8),%R12 | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VCOMPRESSPD %ZMM9,(%R9,%RBX,8){%K1} | 3 | 0 | 0 | 0.33 | 0.33 | 1 | 2 | 0 | 0.33 | 3 | 2 |
KSHIFTRW $0x8,%K1,%K2 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 4 | 1 |
KMOVB %K1,%EBX | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
POPCNT %EBX,%EBX | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 |
VCOMPRESSPD %ZMM8,(%R12,%RBX,8){%K2} | 3 | 0 | 0 | 0.33 | 0.33 | 1 | 2 | 0 | 0.33 | 3 | 2 |
VPSUBD %ZMM6,%ZMM7,%ZMM10 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
JMP 41b8d0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1-2 |
Function | qmcplusplus::BsplineFunctor |
Source file and lines | BsplineFunctor.h:236-241 |
Module | exec |
nb instructions | 25 |
nb uops | 24 |
loop length | 139 |
used x86 registers | 5 |
used mmx registers | 0 |
used xmm registers | 4 |
used ymm registers | 2 |
used zmm registers | 7 |
nb stack references | 0 |
micro-operation queue | 6.00 cycles |
front end | 6.00 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 6.00 | 2.50 | 1.00 | 1.00 | 0.00 | 9.00 | 2.50 | 0.00 |
cycles | 6.00 | 2.50 | 1.00 | 1.00 | 0.00 | 9.00 | 2.50 | 0.00 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 9.14 |
Stall cycles | 2.73 |
RS full (events) | 6.37 |
Front-end | 6.00 |
Dispatch | 9.00 |
Data deps. | 1.00 |
Overall L1 | 9.00 |
all | 92% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 88% |
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) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 100% |
all | 94% |
load | 100% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 90% |
all | 62% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 42% |
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) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 100% |
all | 71% |
load | 100% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 52% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|
VMOVDQA32 %ZMM10,%ZMM7{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VEXTRACTI64X4 $0x1,%ZMM7,%YMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VPADDD %ZMM8,%ZMM7,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VEXTRACTI128 $0x1,%YMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VPADDD %ZMM8,%ZMM7,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPSHUFD $-0x12,%XMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM8,%ZMM7,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPSHUFD $0x55,%XMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM8,%ZMM7,%ZMM8 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPBLENDW $0x3,%XMM8,%XMM5,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
ADD $0x10,%RAX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
CMP %RCX,%RAX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
JAE 41b98d | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
VMOVUPD (%R15,%RAX,8),%ZMM9 | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVUPD 0x40(%R15,%RAX,8),%ZMM8 | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
LEA (%R14,%RAX,1),%EBX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPBROADCASTD %EBX,%ZMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM4,%ZMM10,%ZMM10 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VCMPPD $0x1,%ZMM3,%ZMM9,%K0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VCMPPD $0x1,%ZMM3,%ZMM8,%K1 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
KUNPCKBW %K0,%K1,%K1 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 4 | 1 |
VPCMPNEQD %ZMM2,%ZMM10,%K1{%K1} | |||||||||||
KORTESTW %K1,%K1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 |
VPXOR %XMM10,%XMM10,%XMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
JE 41b8d0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |