Loop Id: 671 | Module: exec | Source: MultiBsplineRef.hpp:66-74 [...] | Coverage: 0.03% |
---|
Loop Id: 671 | Module: exec | Source: MultiBsplineRef.hpp:66-74 [...] | Coverage: 0.03% |
---|
0x4380e0 MOV 0x180(%RSP),%RDI |
0x4380e8 IMUL %R10,%RDI |
0x4380ec ADD %R13,%RDI |
0x4380ef VPBROADCASTQ %RAX,%ZMM7 |
0x4380f5 VPORQ %ZMM11,%ZMM7,%ZMM7 |
0x4380fb VPCMPLTUQ %ZMM4,%ZMM7,%K1 |
0x438102 MOV 0x1c0(%RSP),%RDX |
0x43810a ADD %RAX,%RDX |
0x43810d ADD %RDI,%RDX |
0x438110 VMOVUPD (%RCX,%RDX,8),%ZMM7{%K1}{z} |
0x438117 VMOVAPD %ZMM7,%ZMM16{%K1} |
0x43811d VMULPD %ZMM0,%ZMM16,%ZMM7 |
0x438123 MOV 0x140(%RSP),%RDX |
0x43812b ADD %RAX,%RDX |
0x43812e ADD %RDI,%RDX |
0x438131 VMOVUPD (%RCX,%RDX,8),%ZMM8{%K1}{z} |
0x438138 VMOVAPD %ZMM8,%ZMM15{%K1} |
0x43813e VFMADD231PD %ZMM1,%ZMM15,%ZMM7 |
0x438144 MOV 0x240(%RSP),%RDX |
0x43814c ADD %RAX,%RDX |
0x43814f ADD %RDI,%RDX |
0x438152 VMOVUPD (%RCX,%RDX,8),%ZMM8{%K1}{z} |
0x438159 VMOVAPD %ZMM8,%ZMM14{%K1} |
0x43815f VFMADD231PD %ZMM2,%ZMM14,%ZMM7 |
0x438165 MOV 0x200(%RSP),%RDX |
0x43816d ADD %RAX,%RDX |
0x438170 ADD %RDX,%RDI |
0x438173 VMOVUPD (%RCX,%RDI,8),%ZMM8{%K1}{z} |
0x43817a VMOVAPD %ZMM8,%ZMM13{%K1} |
0x438180 VFMADD231PD %ZMM13,%ZMM3,%ZMM7 |
0x438186 VMOVUPD (%RBX,%RAX,8),%ZMM8{%K1}{z} |
0x43818d VMOVAPD %ZMM8,%ZMM12{%K1} |
0x438193 VFMADD213PD %ZMM12,%ZMM6,%ZMM7 |
0x438199 VMOVUPD %ZMM7,(%RBX,%RAX,8){%K1} |
0x4381a0 LEA 0x1(%R10),%RAX |
0x4381a4 ADD %R14,%R9 |
0x4381a7 ADD %R14,%R15 |
0x4381aa ADD %R14,%R12 |
0x4381ad ADD %R14,%R8 |
0x4381b0 CMP $0x3,%R10 |
0x4381b4 MOV %RAX,%R10 |
0x4381b7 JE 438060 |
0x4381bd VMULSD 0x100(%RSP,%R10,8),%XMM5,%XMM6 |
0x4381c7 VBROADCASTSD %XMM6,%ZMM6 |
0x4381cd XOR %EAX,%EAX |
0x4381cf TEST %R11,%R11 |
0x4381d2 JE 4380e0 |
0x4381d8 NOPL (%RAX,%RAX,1) |
(672) 0x4381e0 VMULPD (%R8,%RAX,8),%ZMM0,%ZMM7 |
(672) 0x4381e7 VFMADD231PD (%R12,%RAX,8),%ZMM1,%ZMM7 |
(672) 0x4381ee VFMADD231PD (%R15,%RAX,8),%ZMM2,%ZMM7 |
(672) 0x4381f5 VFMADD231PD (%R9,%RAX,8),%ZMM3,%ZMM7 |
(672) 0x4381fc VFMADD213PD (%RBX,%RAX,8),%ZMM6,%ZMM7 |
(672) 0x438203 VMOVUPD %ZMM7,(%RBX,%RAX,8) |
(672) 0x43820a ADD $0x8,%RAX |
(672) 0x43820e CMP %RSI,%RAX |
(672) 0x438211 JBE 4381e0 |
0x438213 MOV %R11,%RAX |
0x438216 CMP %R11,0x80(%RSP) |
0x43821e JNE 4380e0 |
0x438224 JMP 4381a0 |
/home/kcamus/qaas_runs/169-451-1869/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVectorOps.h: 59 - 59 |
-------------------------------------------------------------------------------- |
59: for (unsigned d = 0; d < D; ++d) |
/home/kcamus/qaas_runs/169-451-1869/intel/miniqmc/build/miniqmc/src/Numerics/Spline2/MultiBsplineRef.hpp: 66 - 74 |
-------------------------------------------------------------------------------- |
66: for (size_t j = 0; j < 4; j++) |
67: { |
68: const T pre00 = a[i] * b[j]; |
69: const T* restrict coefs = spline_m->coefs + (ix + i) * xs + (iy + j) * ys + iz * zs; |
70: #pragma omp simd aligned(coefs: QMC_SIMD_ALIGNMENT) simdlen(simdlen_) |
71: for (size_t n = 0; n < num_splines; n++) |
72: vals[n] += pre00 * |
73: (c[0] * coefs[n] + c[1] * coefs[n + zs] + c[2] * coefs[n + 2 * zs] + |
74: c[3] * coefs[n + 3 * zs]); |
Coverage (%) | Name | Source Location | Module |
---|---|---|---|
►100.00+ | miniqmcreference::DiracDetermi[...] | DiracDeterminantRef.cpp:194 | exec |
○ | qmcplusplus::WaveFunction::rat[...] | WaveFunction.cpp:214 | 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 | 2.09 |
CQA speedup if FP arith vectorized | 1.10 |
CQA speedup if fully vectorized | 1.37 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.64 |
Bottlenecks | |
Function | miniqmcreference::einspline_spo_ref |
Source | TinyVectorOps.h:59-59,MultiBsplineRef.hpp:66-74 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 9.58 |
CQA cycles if no scalar integer | 4.58 |
CQA cycles if FP arith vectorized | 8.70 |
CQA cycles if fully vectorized | 6.98 |
Front-end cycles | 9.58 |
DIV/SQRT cycles | 5.83 |
P0 cycles | 5.83 |
P1 cycles | 4.17 |
P2 cycles | 4.17 |
P3 cycles | 0.67 |
P4 cycles | 5.83 |
P5 cycles | 5.83 |
P6 cycles | 0.67 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 9.88 |
Stall cycles (UFS) | 0.00 |
Nb insns | 38.67 |
Nb uops | 37.33 |
Nb loads | 8.33 |
Nb stores | 0.67 |
Nb stack references | 4.00 |
FLOP/cycle | 5.11 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 6.33 |
Nb FLOP fma | 21.33 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 25.38 |
Bytes prefetched | 0.00 |
Bytes loaded | 253.33 |
Bytes stored | 42.67 |
Stride 0 | 1.00 |
Stride 1 | 0.00 |
Stride n | 0.00 |
Stride unknown | 5.00 |
Stride indirect | 1.00 |
Vectorization ratio all | 45.30 |
Vectorization ratio load | 55.56 |
Vectorization ratio store | 100.00 |
Vectorization ratio mul | 33.33 |
Vectorization ratio add_sub | 0.00 |
Vectorization ratio fma | 100.00 |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 44.55 |
Vector-efficiency ratio all | 52.14 |
Vector-efficiency ratio load | 61.11 |
Vector-efficiency ratio store | 100.00 |
Vector-efficiency ratio mul | 41.67 |
Vector-efficiency ratio add_sub | 12.50 |
Vector-efficiency ratio fma | 100.00 |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 51.48 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 2.00 |
CQA speedup if FP arith vectorized | 1.01 |
CQA speedup if fully vectorized | 1.17 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.70 |
Bottlenecks | micro-operation queue, |
Function | miniqmcreference::einspline_spo_ref |
Source | TinyVectorOps.h:59-59,MultiBsplineRef.hpp:66-74 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 11.50 |
CQA cycles if no scalar integer | 5.75 |
CQA cycles if FP arith vectorized | 11.39 |
CQA cycles if fully vectorized | 9.79 |
Front-end cycles | 11.50 |
DIV/SQRT cycles | 6.75 |
P0 cycles | 6.75 |
P1 cycles | 5.50 |
P2 cycles | 5.50 |
P3 cycles | 1.00 |
P4 cycles | 6.75 |
P5 cycles | 6.75 |
P6 cycles | 1.00 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 11.86 |
Stall cycles (UFS) | 0.00 |
Nb insns | 47.00 |
Nb uops | 45.00 |
Nb loads | 11.00 |
Nb stores | 1.00 |
Nb stack references | 5.00 |
FLOP/cycle | 6.35 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 9.00 |
Nb FLOP fma | 32.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 37.57 |
Bytes prefetched | 0.00 |
Bytes loaded | 368.00 |
Bytes stored | 64.00 |
Stride 0 | 1.00 |
Stride 1 | 0.00 |
Stride n | 0.00 |
Stride unknown | 3.00 |
Stride indirect | 1.00 |
Vectorization ratio all | 69.23 |
Vectorization ratio load | 83.33 |
Vectorization ratio store | 100.00 |
Vectorization ratio mul | 50.00 |
Vectorization ratio add_sub | 0.00 |
Vectorization ratio fma | 100.00 |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 70.00 |
Vector-efficiency ratio all | 73.08 |
Vector-efficiency ratio load | 85.42 |
Vector-efficiency ratio store | 100.00 |
Vector-efficiency ratio mul | 56.25 |
Vector-efficiency ratio add_sub | 12.50 |
Vector-efficiency ratio fma | 100.00 |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 73.75 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 2.38 |
CQA speedup if FP arith vectorized | 2.04 |
CQA speedup if fully vectorized | 8.00 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.36 |
Bottlenecks | micro-operation queue, |
Function | miniqmcreference::einspline_spo_ref |
Source | TinyVectorOps.h:59-59,MultiBsplineRef.hpp:66-74 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 4.75 |
CQA cycles if no scalar integer | 2.00 |
CQA cycles if FP arith vectorized | 2.33 |
CQA cycles if fully vectorized | 0.59 |
Front-end cycles | 4.75 |
DIV/SQRT cycles | 3.50 |
P0 cycles | 3.50 |
P1 cycles | 1.00 |
P2 cycles | 1.00 |
P3 cycles | 0.00 |
P4 cycles | 3.50 |
P5 cycles | 3.50 |
P6 cycles | 0.00 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 4.91 |
Stall cycles (UFS) | 0.00 |
Nb insns | 18.00 |
Nb uops | 18.00 |
Nb loads | 2.00 |
Nb stores | 0.00 |
Nb stack references | 1.00 |
FLOP/cycle | 0.21 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 1.00 |
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 | 3.37 |
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 | 5.00 |
Stride indirect | 1.00 |
Vectorization ratio all | 0.00 |
Vectorization ratio load | 0.00 |
Vectorization ratio store | NA |
Vectorization ratio mul | 0.00 |
Vectorization ratio add_sub | 0.00 |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 0.00 |
Vector-efficiency ratio all | 12.50 |
Vector-efficiency ratio load | 12.50 |
Vector-efficiency ratio store | NA |
Vector-efficiency ratio mul | 12.50 |
Vector-efficiency ratio add_sub | 12.50 |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 12.50 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 2.08 |
CQA speedup if FP arith vectorized | 1.01 |
CQA speedup if fully vectorized | 1.18 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.72 |
Bottlenecks | micro-operation queue, |
Function | miniqmcreference::einspline_spo_ref |
Source | TinyVectorOps.h:59-59,MultiBsplineRef.hpp:66-74 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 12.50 |
CQA cycles if no scalar integer | 6.00 |
CQA cycles if FP arith vectorized | 12.39 |
CQA cycles if fully vectorized | 10.57 |
Front-end cycles | 12.50 |
DIV/SQRT cycles | 7.25 |
P0 cycles | 7.25 |
P1 cycles | 6.00 |
P2 cycles | 6.00 |
P3 cycles | 1.00 |
P4 cycles | 7.25 |
P5 cycles | 7.25 |
P6 cycles | 1.00 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 12.86 |
Stall cycles (UFS) | 0.00 |
Nb insns | 51.00 |
Nb uops | 49.00 |
Nb loads | 12.00 |
Nb stores | 1.00 |
Nb stack references | 6.00 |
FLOP/cycle | 5.84 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 9.00 |
Nb FLOP fma | 32.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 35.20 |
Bytes prefetched | 0.00 |
Bytes loaded | 376.00 |
Bytes stored | 64.00 |
Stride 0 | 1.00 |
Stride 1 | 0.00 |
Stride n | 0.00 |
Stride unknown | 7.00 |
Stride indirect | 1.00 |
Vectorization ratio all | 66.67 |
Vectorization ratio load | 83.33 |
Vectorization ratio store | 100.00 |
Vectorization ratio mul | 50.00 |
Vectorization ratio add_sub | 0.00 |
Vectorization ratio fma | 100.00 |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 63.64 |
Vector-efficiency ratio all | 70.83 |
Vector-efficiency ratio load | 85.42 |
Vector-efficiency ratio store | 100.00 |
Vector-efficiency ratio mul | 56.25 |
Vector-efficiency ratio add_sub | 12.50 |
Vector-efficiency ratio fma | 100.00 |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 68.18 |
Path / |
Function | miniqmcreference::einspline_spo_ref |
Source file and lines | MultiBsplineRef.hpp:66-74 |
Module | exec |
nb instructions | 38.67 |
nb uops | 37.33 |
loop length | 202.33 |
used x86 registers | 12.33 |
used mmx registers | 0 |
used xmm registers | 2 |
used ymm registers | 0 |
used zmm registers | 9.67 |
nb stack references | 4 |
micro-operation queue | 9.58 cycles |
front end | 9.58 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 5.83 | 5.83 | 4.17 | 4.17 | 0.67 | 5.83 | 5.83 | 0.67 |
cycles | 5.83 | 5.83 | 4.17 | 4.17 | 0.67 | 5.83 | 5.83 | 0.67 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 9.88 |
Stall cycles | 0.00 |
Front-end | 9.58 |
Dispatch | 5.83 |
Data deps. | 1.00 |
Overall L1 | 9.58 |
all | 15% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 0% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 30% |
all | 59% |
load | 55% |
store | 100% |
mul | 33% |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 55% |
all | 45% |
load | 55% |
store | 100% |
mul | 33% |
add-sub | 0% |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 44% |
all | 26% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 12% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 38% |
all | 64% |
load | 61% |
store | 100% |
mul | 41% |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 61% |
all | 52% |
load | 61% |
store | 100% |
mul | 41% |
add-sub | 12% |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 51% |
Function | miniqmcreference::einspline_spo_ref |
Source file and lines | MultiBsplineRef.hpp:66-74 |
Module | exec |
nb instructions | 47 |
nb uops | 45 |
loop length | 248 |
used x86 registers | 14 |
used mmx registers | 0 |
used xmm registers | 2 |
used ymm registers | 0 |
used zmm registers | 14 |
nb stack references | 5 |
micro-operation queue | 11.50 cycles |
front end | 11.50 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 6.75 | 6.75 | 5.50 | 5.50 | 1.00 | 6.75 | 6.75 | 1.00 |
cycles | 6.75 | 6.75 | 5.50 | 5.50 | 1.00 | 6.75 | 6.75 | 1.00 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 11.86 |
Stall cycles | 0.00 |
Front-end | 11.50 |
Dispatch | 6.75 |
Data deps. | 1.00 |
Overall L1 | 11.50 |
all | 25% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 0% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 50% |
all | 88% |
load | 83% |
store | 100% |
mul | 50% |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 83% |
all | 69% |
load | 83% |
store | 100% |
mul | 50% |
add-sub | 0% |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 70% |
all | 34% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 12% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 56% |
all | 90% |
load | 85% |
store | 100% |
mul | 56% |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 85% |
all | 73% |
load | 85% |
store | 100% |
mul | 56% |
add-sub | 12% |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 73% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|
MOV 0x180(%RSP),%RDI | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4-5 | 0.50 |
IMUL %R10,%RDI | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 |
ADD %R13,%RDI | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VPBROADCASTQ %RAX,%ZMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPORQ %ZMM11,%ZMM7,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPCMPLTUQ %ZMM4,%ZMM7,%K1 | |||||||||||
MOV 0x1c0(%RSP),%RDX | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4-5 | 0.50 |
ADD %RAX,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %RDI,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VMOVUPD (%RCX,%RDX,8),%ZMM7{%K1}{z} | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVAPD %ZMM7,%ZMM16{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VMULPD %ZMM0,%ZMM16,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 4 | 0.50 |
MOV 0x140(%RSP),%RDX | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4-5 | 0.50 |
ADD %RAX,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %RDI,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VMOVUPD (%RCX,%RDX,8),%ZMM8{%K1}{z} | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVAPD %ZMM8,%ZMM15{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VFMADD231PD %ZMM1,%ZMM15,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 4 | 0.50 |
MOV 0x240(%RSP),%RDX | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4-5 | 0.50 |
ADD %RAX,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %RDI,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VMOVUPD (%RCX,%RDX,8),%ZMM8{%K1}{z} | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVAPD %ZMM8,%ZMM14{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VFMADD231PD %ZMM2,%ZMM14,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 4 | 0.50 |
MOV 0x200(%RSP),%RDX | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4-5 | 0.50 |
ADD %RAX,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %RDX,%RDI | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VMOVUPD (%RCX,%RDI,8),%ZMM8{%K1}{z} | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVAPD %ZMM8,%ZMM13{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VFMADD231PD %ZMM13,%ZMM3,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 4 | 0.50 |
VMOVUPD (%RBX,%RAX,8),%ZMM8{%K1}{z} | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVAPD %ZMM8,%ZMM12{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VFMADD213PD %ZMM12,%ZMM6,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 4 | 0.50 |
VMOVUPD %ZMM7,(%RBX,%RAX,8){%K1} | 1 | 0 | 0 | 0.33 | 0.33 | 1 | 0 | 0 | 0.33 | 3 | 1 |
LEA 0x1(%R10),%RAX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
ADD %R14,%R9 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %R14,%R15 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %R14,%R12 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %R14,%R8 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
CMP $0x3,%R10 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
MOV %RAX,%R10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
JE 438060 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
VMULSD 0x100(%RSP,%R10,8),%XMM5,%XMM6 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VBROADCASTSD %XMM6,%ZMM6 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
XOR %EAX,%EAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
TEST %R11,%R11 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
JE 4380e0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
Function | miniqmcreference::einspline_spo_ref |
Source file and lines | MultiBsplineRef.hpp:66-74 |
Module | exec |
nb instructions | 18 |
nb uops | 18 |
loop length | 86 |
used x86 registers | 9 |
used mmx registers | 0 |
used xmm registers | 2 |
used ymm registers | 0 |
used zmm registers | 1 |
nb stack references | 1 |
micro-operation queue | 4.75 cycles |
front end | 4.75 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 3.50 | 3.50 | 1.00 | 1.00 | 0.00 | 3.50 | 3.50 | 0.00 |
cycles | 3.50 | 3.50 | 1.00 | 1.00 | 0.00 | 3.50 | 3.50 | 0.00 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 4.91 |
Stall cycles | 0.00 |
Front-end | 4.75 |
Dispatch | 3.50 |
Data deps. | 1.00 |
Overall L1 | 4.75 |
all | 0% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 0% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 0% |
all | 0% |
load | 0% |
store | NA (no store vectorizable/vectorized instructions) |
mul | 0% |
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 | 0% |
all | 0% |
load | 0% |
store | NA (no store vectorizable/vectorized instructions) |
mul | 0% |
add-sub | 0% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 0% |
all | 12% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 12% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 12% |
all | 12% |
load | 12% |
store | NA (no store vectorizable/vectorized instructions) |
mul | 12% |
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 | 12% |
all | 12% |
load | 12% |
store | NA (no store vectorizable/vectorized instructions) |
mul | 12% |
add-sub | 12% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 12% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|
LEA 0x1(%R10),%RAX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
ADD %R14,%R9 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %R14,%R15 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %R14,%R12 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %R14,%R8 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
CMP $0x3,%R10 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
MOV %RAX,%R10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
JE 438060 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
VMULSD 0x100(%RSP,%R10,8),%XMM5,%XMM6 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VBROADCASTSD %XMM6,%ZMM6 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
XOR %EAX,%EAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
TEST %R11,%R11 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
JE 4380e0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
NOPL (%RAX,%RAX,1) | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
MOV %R11,%RAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
CMP %R11,0x80(%RSP) | 1 | 0.25 | 0.25 | 0.50 | 0.50 | 0 | 0.25 | 0.25 | 0 | 1 | 0.50 |
JNE 4380e0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
JMP 4381a0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1-2 |
Function | miniqmcreference::einspline_spo_ref |
Source file and lines | MultiBsplineRef.hpp:66-74 |
Module | exec |
nb instructions | 51 |
nb uops | 49 |
loop length | 273 |
used x86 registers | 14 |
used mmx registers | 0 |
used xmm registers | 2 |
used ymm registers | 0 |
used zmm registers | 14 |
nb stack references | 6 |
micro-operation queue | 12.50 cycles |
front end | 12.50 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 7.25 | 7.25 | 6.00 | 6.00 | 1.00 | 7.25 | 7.25 | 1.00 |
cycles | 7.25 | 7.25 | 6.00 | 6.00 | 1.00 | 7.25 | 7.25 | 1.00 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 12.86 |
Stall cycles | 0.00 |
Front-end | 12.50 |
Dispatch | 7.25 |
Data deps. | 1.00 |
Overall L1 | 12.50 |
all | 22% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 0% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 40% |
all | 88% |
load | 83% |
store | 100% |
mul | 50% |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 83% |
all | 66% |
load | 83% |
store | 100% |
mul | 50% |
add-sub | 0% |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 63% |
all | 31% |
load | NA (no load vectorizable/vectorized instructions) |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 12% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 47% |
all | 90% |
load | 85% |
store | 100% |
mul | 56% |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 85% |
all | 70% |
load | 85% |
store | 100% |
mul | 56% |
add-sub | 12% |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 68% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|
MOV 0x180(%RSP),%RDI | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4-5 | 0.50 |
IMUL %R10,%RDI | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 |
ADD %R13,%RDI | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VPBROADCASTQ %RAX,%ZMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 |
VPORQ %ZMM11,%ZMM7,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
VPCMPLTUQ %ZMM4,%ZMM7,%K1 | |||||||||||
MOV 0x1c0(%RSP),%RDX | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4-5 | 0.50 |
ADD %RAX,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %RDI,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VMOVUPD (%RCX,%RDX,8),%ZMM7{%K1}{z} | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVAPD %ZMM7,%ZMM16{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VMULPD %ZMM0,%ZMM16,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 4 | 0.50 |
MOV 0x140(%RSP),%RDX | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4-5 | 0.50 |
ADD %RAX,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %RDI,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VMOVUPD (%RCX,%RDX,8),%ZMM8{%K1}{z} | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVAPD %ZMM8,%ZMM15{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VFMADD231PD %ZMM1,%ZMM15,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 4 | 0.50 |
MOV 0x240(%RSP),%RDX | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4-5 | 0.50 |
ADD %RAX,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %RDI,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VMOVUPD (%RCX,%RDX,8),%ZMM8{%K1}{z} | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVAPD %ZMM8,%ZMM14{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VFMADD231PD %ZMM2,%ZMM14,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 4 | 0.50 |
MOV 0x200(%RSP),%RDX | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4-5 | 0.50 |
ADD %RAX,%RDX | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %RDX,%RDI | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
VMOVUPD (%RCX,%RDI,8),%ZMM8{%K1}{z} | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVAPD %ZMM8,%ZMM13{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VFMADD231PD %ZMM13,%ZMM3,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 4 | 0.50 |
VMOVUPD (%RBX,%RAX,8),%ZMM8{%K1}{z} | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
VMOVAPD %ZMM8,%ZMM12{%K1} | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VFMADD213PD %ZMM12,%ZMM6,%ZMM7 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 4 | 0.50 |
VMOVUPD %ZMM7,(%RBX,%RAX,8){%K1} | 1 | 0 | 0 | 0.33 | 0.33 | 1 | 0 | 0 | 0.33 | 3 | 1 |
LEA 0x1(%R10),%RAX | 1 | 0 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 1 | 0.50 |
ADD %R14,%R9 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %R14,%R15 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %R14,%R12 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
ADD %R14,%R8 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
CMP $0x3,%R10 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
MOV %RAX,%R10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
JE 438060 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
VMULSD 0x100(%RSP,%R10,8),%XMM5,%XMM6 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VBROADCASTSD %XMM6,%ZMM6 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 3 | 1 |
XOR %EAX,%EAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
TEST %R11,%R11 | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
JE 4380e0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
NOPL (%RAX,%RAX,1) | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
MOV %R11,%RAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
CMP %R11,0x80(%RSP) | 1 | 0.25 | 0.25 | 0.50 | 0.50 | 0 | 0.25 | 0.25 | 0 | 1 | 0.50 |
JNE 4380e0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |