Loop Id: 227 | Module: exec | Source: BsplineFunctor.h:291-298 | Coverage: 0.43% |
---|
Loop Id: 227 | Module: exec | Source: BsplineFunctor.h:291-298 | Coverage: 0.43% |
---|
0x41aad0 VMOVDQA32 %ZMM9,%ZMM6{%K1} |
0x41aad6 VEXTRACTI64X4 $0x1,%ZMM6,%YMM7 |
0x41aadd VPADDD %ZMM7,%ZMM6,%ZMM6 |
0x41aae3 VEXTRACTI128 $0x1,%YMM6,%XMM7 |
0x41aae9 VPADDD %ZMM7,%ZMM6,%ZMM6 |
0x41aaef VPSHUFD $-0x12,%XMM6,%XMM7 |
0x41aaf4 VPADDD %ZMM7,%ZMM6,%ZMM6 |
0x41aafa VPSHUFD $0x55,%XMM6,%XMM7 |
0x41aaff VPADDD %ZMM7,%ZMM6,%ZMM7 |
0x41ab05 VPBLENDW $0x3,%XMM7,%XMM4,%XMM6 |
0x41ab0b ADD $0x10,%RAX |
0x41ab0f CMP %RBX,%RAX |
0x41ab12 JAE 41abaf |
0x41ab18 VMOVUPD (%R12,%RAX,8),%ZMM7 [1] |
0x41ab1f VMOVUPD 0x40(%R12,%RAX,8),%ZMM8 [1] |
0x41ab27 LEA (%R15,%RAX,1),%ECX |
0x41ab2b VPBROADCASTD %ECX,%ZMM9 |
0x41ab31 VPADDD %ZMM3,%ZMM9,%ZMM9 |
0x41ab37 VCMPPD $0x1,%ZMM2,%ZMM7,%K0 |
0x41ab3e VCMPPD $0x1,%ZMM2,%ZMM8,%K1 |
0x41ab45 KUNPCKBW %K0,%K1,%K1 |
0x41ab49 VPCMPNEQD %ZMM1,%ZMM9,%K1{%K1} |
0x41ab50 KORTESTW %K1,%K1 |
0x41ab54 VPXOR %XMM9,%XMM9,%XMM9 |
0x41ab59 JE 41aad0 |
0x41ab5f VMOVD %XMM6,%ECX |
0x41ab63 MOVSXD %ECX,%R10 |
0x41ab66 VPBROADCASTD %EAX,%ZMM9 |
0x41ab6c VPORD %ZMM3,%ZMM9,%ZMM9 |
0x41ab72 VPCOMPRESSD %ZMM9,(%R13,%R10,4){%K1} [4] |
0x41ab7a MOV %R13,%R8 |
0x41ab7d LEA (%R14,%R10,8),%R13 |
0x41ab81 KSHIFTRW $0x8,%K1,%K2 |
0x41ab87 KMOVB %K1,%R11D |
0x41ab8b XOR %ECX,%ECX |
0x41ab8d POPCNT %R11D,%ECX |
0x41ab92 VCOMPRESSPD %ZMM8,(%R13,%RCX,8){%K2} [2] |
0x41ab9a MOV %R8,%R13 |
0x41ab9d VCOMPRESSPD %ZMM7,(%R14,%R10,8){%K1} [3] |
0x41aba4 VPSUBD %ZMM5,%ZMM6,%ZMM9 |
0x41abaa JMP 41aad0 |
/home/kcamus/qaas_runs/169-451-1869/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/BsplineFunctor.h: 291 - 298 |
-------------------------------------------------------------------------------- |
291: for (int jat = 0; jat < iLimit; jat++) |
292: { |
293: real_type r = distArray[jat]; |
294: if (r < cutoff_radius && iStart + jat != iat) |
295: { |
296: distIndices[iCount] = jat; |
297: distArrayCompressed[iCount] = r; |
298: iCount++; |
Coverage (%) | Name | Source Location | Module |
---|---|---|---|
►49.02+ | miniqmcreference::TwoBodyJastr[...] | TwoBodyJastrowRef.h:271 | exec |
○ | qmcplusplus::WaveFunction::rat[...] | WaveFunction.cpp:207 | 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 | |
►33.33+ | miniqmcreference::TwoBodyJastr[...] | TwoBodyJastrowRef.h:271 | exec |
○ | qmcplusplus::WaveFunction::acc[...] | NewTimer.h:249 | 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 | |
►11.76+ | miniqmcreference::TwoBodyJastr[...] | TwoBodyJastrowRef.h:271 | exec |
○ | miniqmcreference::TwoBodyJastr[...] | TwoBodyJastrowRef.h:409 | exec |
○ | qmcplusplus::WaveFunction::eva[...] | WaveFunction.cpp:181 | exec |
○ | main.extracted.107 | miniqmc.cpp:375 | exec |
○ | __kmp_invoke_microtask | libiomp5.so | |
○ | __kmp_fork_call | libiomp5.so | |
○ | __kmpc_fork_call | libiomp5.so | |
○ | main | miniqmc.cpp:374 | exec |
○ | __libc_init_first | libc.so.6 | |
►5.88+ | miniqmcreference::OneBodyJastr[...] | OneBodyJastrowRef.h:222 | exec |
○ | miniqmcreference::OneBodyJastr[...] | stl_vector.h:1258 | exec |
○ | qmcplusplus::WaveFunction::rat[...] | WaveFunction.cpp:207 | 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 | 1.13 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 1.52 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.49 |
Bottlenecks | |
Function | qmcplusplus::BsplineFunctor |
Source | BsplineFunctor.h:291-298 |
Source loop unroll info | unrolled by 16 |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | main |
Unroll factor | 16 |
CQA cycles | 13.00 |
CQA cycles if no scalar integer | 11.50 |
CQA cycles if FP arith vectorized | 13.00 |
CQA cycles if fully vectorized | 8.56 |
Front-end cycles | 8.75 |
DIV/SQRT cycles | 8.00 |
P0 cycles | 3.50 |
P1 cycles | 1.33 |
P2 cycles | 1.33 |
P3 cycles | 1.50 |
P4 cycles | 13.00 |
P5 cycles | 3.50 |
P6 cycles | 0.83 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 13.15 - 13.16 |
Stall cycles (UFS) | 4.02 - 4.02 |
Nb insns | 33.00 |
Nb uops | 35.00 |
Nb loads | 2.00 |
Nb stores | 1.50 |
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 | 16.52 |
Bytes prefetched | 0.00 |
Bytes loaded | 128.00 |
Bytes stored | 96.00 |
Stride 0 | 0.00 |
Stride 1 | 1.00 |
Stride n | 0.00 |
Stride unknown | 1.00 |
Stride indirect | 0.50 |
Vectorization ratio all | 89.53 |
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 | 82.12 |
Vector-efficiency ratio all | 70.33 |
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 | 49.96 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.13 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 1.46 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.48 |
Bottlenecks | P5, |
Function | qmcplusplus::BsplineFunctor |
Source | BsplineFunctor.h:291-298 |
Source loop unroll info | unrolled by 16 |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | main |
Unroll factor | 16 |
CQA cycles | 17.00 |
CQA cycles if no scalar integer | 15.00 |
CQA cycles if FP arith vectorized | 17.00 |
CQA cycles if fully vectorized | 11.63 |
Front-end cycles | 11.50 |
DIV/SQRT cycles | 10.00 |
P0 cycles | 4.50 |
P1 cycles | 1.67 |
P2 cycles | 1.67 |
P3 cycles | 3.00 |
P4 cycles | 17.00 |
P5 cycles | 4.50 |
P6 cycles | 1.67 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 17.16 - 17.19 |
Stall cycles (UFS) | 5.31 - 5.32 |
Nb insns | 41.00 |
Nb uops | 46.00 |
Nb loads | 2.00 |
Nb stores | 3.00 |
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 | 18.82 |
Bytes prefetched | 0.00 |
Bytes loaded | 128.00 |
Bytes stored | 192.00 |
Stride 0 | 0.00 |
Stride 1 | 1.00 |
Stride n | 0.00 |
Stride unknown | 2.00 |
Stride indirect | 1.00 |
Vectorization ratio all | 84.62 |
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 | 73.33 |
Vector-efficiency ratio all | 69.47 |
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 | 47.08 |
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:291-298 |
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:291-298 |
Module | exec |
nb instructions | 33 |
nb uops | 35 |
loop length | 183 |
used x86 registers | 7.50 |
used mmx registers | 0 |
used xmm registers | 4 |
used ymm registers | 2 |
used zmm registers | 7.50 |
nb stack references | 0 |
micro-operation queue | 8.75 cycles |
front end | 8.75 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 8.00 | 3.50 | 1.33 | 1.33 | 1.50 | 13.00 | 3.50 | 0.83 |
cycles | 8.00 | 3.50 | 1.33 | 1.33 | 1.50 | 13.00 | 3.50 | 0.83 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 13.15-13.17 |
Stall cycles | 4.02-4.02 |
RS full (events) | 8.49-8.98 |
Front-end | 8.75 |
Dispatch | 13.00 |
Data deps. | 1.00 |
Overall L1 | 13.00 |
all | 86% |
load | NA (no load vectorizable/vectorized instructions) |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 79% |
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 | 89% |
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 | 82% |
all | 61% |
load | NA (no load vectorizable/vectorized instructions) |
store | 100% |
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 | 70% |
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 | 49% |
Function | qmcplusplus::BsplineFunctor |
Source file and lines | BsplineFunctor.h:291-298 |
Module | exec |
nb instructions | 41 |
nb uops | 46 |
loop length | 223 |
used x86 registers | 10 |
used mmx registers | 0 |
used xmm registers | 4 |
used ymm registers | 2 |
used zmm registers | 8 |
nb stack references | 0 |
micro-operation queue | 11.50 cycles |
front end | 11.50 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 10.00 | 4.50 | 1.67 | 1.67 | 3.00 | 17.00 | 4.50 | 1.67 |
cycles | 10.00 | 4.50 | 1.67 | 1.67 | 3.00 | 17.00 | 4.50 | 1.67 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 17.16-17.19 |
Stall cycles | 5.31-5.32 |
RS full (events) | 10.62-11.60 |
Front-end | 11.50 |
Dispatch | 17.00 |
Data deps. | 1.00 |
Overall L1 | 17.00 |
all | 80% |
load | NA (no load vectorizable/vectorized instructions) |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 69% |
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 | 84% |
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 | 73% |
all | 60% |
load | NA (no load vectorizable/vectorized instructions) |
store | 100% |
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 | 69% |
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 | 47% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|
VMOVDQA32 %ZMM9,%ZMM6{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VEXTRACTI64X4 $0x1,%ZMM6,%YMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VPADDD %ZMM7,%ZMM6,%ZMM6 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VEXTRACTI128 $0x1,%YMM6,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VPADDD %ZMM7,%ZMM6,%ZMM6 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPSHUFD $-0x12,%XMM6,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM7,%ZMM6,%ZMM6 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPSHUFD $0x55,%XMM6,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM7,%ZMM6,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPBLENDW $0x3,%XMM7,%XMM4,%XMM6 | 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 %RBX,%RAX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
JAE 41abaf | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
VMOVUPD (%R12,%RAX,8),%ZMM7 | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVUPD 0x40(%R12,%RAX,8),%ZMM8 | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
LEA (%R15,%RAX,1),%ECX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPBROADCASTD %ECX,%ZMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM3,%ZMM9,%ZMM9 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VCMPPD $0x1,%ZMM2,%ZMM7,%K0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VCMPPD $0x1,%ZMM2,%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 %ZMM1,%ZMM9,%K1{%K1} | |||||||||||
KORTESTW %K1,%K1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 |
VPXOR %XMM9,%XMM9,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
JE 41aad0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
VMOVD %XMM6,%ECX | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
MOVSXD %ECX,%R10 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VPBROADCASTD %EAX,%ZMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPORD %ZMM3,%ZMM9,%ZMM9 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPCOMPRESSD %ZMM9,(%R13,%R10,4){%K1} | 3 | 0 | 0 | 0.33 | 0.33 | 1 | 2 | 0 | 0.33 | 3-6 | 2 |
MOV %R13,%R8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
LEA (%R14,%R10,8),%R13 | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
KSHIFTRW $0x8,%K1,%K2 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 4 | 1 |
KMOVB %K1,%R11D | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
XOR %ECX,%ECX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
POPCNT %R11D,%ECX | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 |
VCOMPRESSPD %ZMM8,(%R13,%RCX,8){%K2} | 3 | 0 | 0 | 0.33 | 0.33 | 1 | 2 | 0 | 0.33 | 3 | 2 |
MOV %R8,%R13 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VCOMPRESSPD %ZMM7,(%R14,%R10,8){%K1} | 3 | 0 | 0 | 0.33 | 0.33 | 1 | 2 | 0 | 0.33 | 3 | 2 |
VPSUBD %ZMM5,%ZMM6,%ZMM9 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
JMP 41aad0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1-2 |
Function | qmcplusplus::BsplineFunctor |
Source file and lines | BsplineFunctor.h:291-298 |
Module | exec |
nb instructions | 25 |
nb uops | 24 |
loop length | 143 |
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 %ZMM9,%ZMM6{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VEXTRACTI64X4 $0x1,%ZMM6,%YMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VPADDD %ZMM7,%ZMM6,%ZMM6 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VEXTRACTI128 $0x1,%YMM6,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VPADDD %ZMM7,%ZMM6,%ZMM6 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPSHUFD $-0x12,%XMM6,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM7,%ZMM6,%ZMM6 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPSHUFD $0x55,%XMM6,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM7,%ZMM6,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPBLENDW $0x3,%XMM7,%XMM4,%XMM6 | 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 %RBX,%RAX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
JAE 41abaf | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
VMOVUPD (%R12,%RAX,8),%ZMM7 | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVUPD 0x40(%R12,%RAX,8),%ZMM8 | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
LEA (%R15,%RAX,1),%ECX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPBROADCASTD %ECX,%ZMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPADDD %ZMM3,%ZMM9,%ZMM9 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VCMPPD $0x1,%ZMM2,%ZMM7,%K0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
VCMPPD $0x1,%ZMM2,%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 %ZMM1,%ZMM9,%K1{%K1} | |||||||||||
KORTESTW %K1,%K1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 |
VPXOR %XMM9,%XMM9,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
JE 41aad0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |