OV - exec - Loop 50

0x409000 LDR	W9, [X26, #7744]

0x409004 LDR	X22, [SP, #64]

  (49) 0x409008 ADD	X19, X19, #1

  (49) 0x40900c CMP	X19, W9,SXTW

  (49) 0x409010 B.GE	409160

  (49) 0x409014 CMP	W8, #1

  (49) 0x409018 B.LT	409008

0x40901c LDR	X11, [SP, #120]

0x409020 ORR	X25, XZR, XZR

0x409024 LDR	X10, [X11, #24]

0x409028 LDR	X9, [X11, #64]

0x40902c MADD	X10, X10, X19, XZR

0x409030 ADD	X21, X9, X10,LSL #3

0x409034 LDR	X9, [X11, #80]

0x409038 MOVZ	W10, #40

0x40903c MADD	X9, X19, X10, X9

0x409040 ADD	X22, X9, #24

0x409044 ADD	X20, X9, #8

0x409048 LDR	X9, [SP, #128]

0x40904c LDR	W9, [X9]

0x409050 B	40906c

0x409060 ADD	X25, X25, #1

0x409064 CMP	X25, W8,SXTW

0x409068 B.GE	409000

0x40906c LDR	X10, [SP, #112]

0x409070 LDR	D0, [X21, X25,LSL #3]

0x409074 LDR	D1, [X10]

0x409078 FCMP	D0, D1

0x40907c CCMP	W9, #1, #8, #11

0x409080 B.LT	409060

0x409084 ORR	X28, XZR, XZR

0x409088 ORR	X24, XZR, XZR

  (51) 0x40908c LDR	X10, [X22]

  (51) 0x409090 UBFM	X9, X25, #61, #60

  (51) 0x409094 LDUR	X8, [X29, #440]

  (51) 0x409098 LDR	D0, [X21, X9]

  (51) 0x40909c ADD	X9, X10, X9

  (51) 0x4090a0 LDR	X10, [X20]

  (51) 0x4090a4 ADD	X8, X8, X28

  (51) 0x4090a8 LDR	D4, [X9]

  (51) 0x4090ac LDR	Q3, [X8]

  (51) 0x4090b0 LDR	D1, [X8, #16]

  (51) 0x4090b4 UBFM	X11, X10, #32, #31

  (51) 0x4090b8 SBFM	X10, X10, #60, #31

  (51) 0x4090bc ADD	X11, X9, X11,ASR #29

  (51) 0x4090c0 LDR	D2, [X9, X10]

  (51) 0x4090c4 LD1	{V4.D[1]}, [X11]

  (51) 0x4090c8 FMUL	V3.2D, V3.2D, V0.D[0]

  (51) 0x4090cc FNMSUB	D0, D1, D0, D2

  (51) 0x4090d0 STUR	D0, [X29, #384]

  (51) 0x4090d4 FSUB	V3.2D, V3.2D, V4.2D

  (51) 0x4090d8 STUR	Q3, [X29, #368]

  (51) 0x4090dc ORR	X0, XZR, X27

  (51) 0x4090e0 ORR	W1, WZR, W19

  (51) 0x4090e4 SUB	X2, X29, #144

  (51) 0x4090e8 BL	445d30

  (51) 0x4090ec LDR	X8, [SP, #144]

  (51) 0x4090f0 LDR	X8, [X8]

  (51) 0x4090f4 LDR	X0, [X8, #32]

  (51) 0x4090f8 BL	44ed40

  (51) 0x4090fc LDR	X0, [SP, #136]

  (51) 0x409100 ORR	X1, XZR, X27

  (51) 0x409104 ORR	W2, WZR, W19

  (51) 0x409108 BL	40ceb0

  (51) 0x40910c LDR	X8, [SP, #144]

  (51) 0x409110 LDR	X8, [X8]

  (51) 0x409114 LDR	X0, [X8, #32]

  (51) 0x409118 BL	44ee60

  (51) 0x40911c ORR	X0, XZR, X27

  (51) 0x409120 ORR	W1, WZR, W19

  (51) 0x409124 BL	446480

  (51) 0x409128 LDR	X8, [SP, #128]

  (51) 0x40912c ADD	X24, X24, #1

  (51) 0x409130 ADD	X28, X28, #24

  (51) 0x409134 LDRSW	X9, [X8]

  (51) 0x409138 CMP	X24, X9

  (51) 0x40913c B.LT	40908c

0x409140 LDR	X8, [SP, #104]

0x409144 MOVZ	W24, #22144

0x409148 MOVZ	W28, #45279

0x40914c MOVK	W24, #40236

0x409150 MOVK	W28, #39176

0x409154 LDR	W8, [X8]

0x409158 B	409060

/usr/lib/gcc/aarch64-linux-gnu/11/../../../../include/c++/11/bits/stl_vector.h: 1046 - 1064

--------------------------------------------------------------------------------

1046: 	return *(this->_M_impl._M_start + __n);

[...]

1064: 	return *(this->_M_impl._M_start + __n);

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h: 53 - 63

--------------------------------------------------------------------------------

53:     return (a - b);

[...]

63:     return (a * b);

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Drivers/miniqmc.cpp: 472 - 488

--------------------------------------------------------------------------------

472:       for (int jel = 0; jel < els.getTotalNum(); ++jel)

473:       {

474:         const auto& dist  = d_ie->Distances[jel];

475:         const auto& displ = d_ie->Displacements[jel];

476:         for (int iat = 0; iat < nions; ++iat)

477:           if (dist[iat] < Rmax)

478:             for (int k = 0; k < nknots; k++)

479:             {

480:               PosType deltar(dist[iat] * rOnSphere[k] - displ[iat]);

481:

482:               els.makeMove(jel, deltar);

483:

484:               Timers[Timer_Value].get().start();

485:               wavefunction.ratio(els, jel);

486:               Timers[Timer_Value].get().stop();

487:

488:               els.rejectMove(jel);

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h: 145 - 146

--------------------------------------------------------------------------------

145:     for (int i = 0; i < D; ++i)

146:       X[i] = base[i * offset];

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h: 223 - 223

--------------------------------------------------------------------------------

223:     return X[i];

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h: 203 - 203

--------------------------------------------------------------------------------

203:   const Type_t operator[](size_t i) const { return Type_t(myData + i, nGhosts); }

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVectorOps.h: 85 - 85

--------------------------------------------------------------------------------

85:       ret[d] = op(lhs[d], rhs[d]);

/usr/lib/gcc/aarch64-linux-gnu/11/../../../../include/c++/11/bits/refwrap.h: 338 - 338

--------------------------------------------------------------------------------

338:       { return *_M_data; }

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h: 263 - 263

--------------------------------------------------------------------------------

263:   inline int getTotalNum() const { return TotalNum; }

Coverage (%)	Name	Source Location	Module
○100.00	__kmp_invoke_microtask		libomp.so

Path /

Metric	Value
CQA speedup if no scalar integer	3.25
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	6.60
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.02
Bottlenecks	P10, P11, P12,
Function	.omp_outlined..62
Source	stl_vector.h:1064-1064,miniqmc.cpp:476-478,ParticleSet.h:263-263
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	4.33
CQA cycles if no scalar integer	1.33
CQA cycles if FP arith vectorized	4.33
CQA cycles if fully vectorized	0.66
Front-end cycles	4.25
DIV/SQRT cycles	2.00
P0 cycles	2.00
P1 cycles	4.00
P2 cycles	4.00
P3 cycles	4.00
P4 cycles	4.00
P5 cycles	1.00
P6 cycles	0.00
P7 cycles	0.00
P8 cycles	0.00
P9 cycles	4.33
P10 cycles	4.33
P11 cycles	4.33
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	34.00
Nb uops	34.00
Nb loads	NA
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	21.23
Bytes prefetched	0.00
Bytes loaded	92.00
Bytes stored	0.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	NA
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	20.45
Vector-efficiency ratio load	25.00
Vector-efficiency ratio store	NA
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	25.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	17.86

Metric	Value
CQA speedup if no scalar integer	3.25
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	6.60
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.02
Bottlenecks	P10, P11, P12,
Function	.omp_outlined..62
Source	stl_vector.h:1064-1064,miniqmc.cpp:476-478,ParticleSet.h:263-263
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	4.33
CQA cycles if no scalar integer	1.33
CQA cycles if FP arith vectorized	4.33
CQA cycles if fully vectorized	0.66
Front-end cycles	4.25
DIV/SQRT cycles	2.00
P0 cycles	2.00
P1 cycles	4.00
P2 cycles	4.00
P3 cycles	4.00
P4 cycles	4.00
P5 cycles	1.00
P6 cycles	0.00
P7 cycles	0.00
P8 cycles	0.00
P9 cycles	4.33
P10 cycles	4.33
P11 cycles	4.33
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	34.00
Nb uops	34.00
Nb loads	NA
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	21.23
Bytes prefetched	0.00
Bytes loaded	92.00
Bytes stored	0.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	NA
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	20.45
Vector-efficiency ratio load	25.00
Vector-efficiency ratio store	NA
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	25.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	17.86

Path /

Average path: Display a virtual path defined by average values of all real paths

Function	.omp_outlined..62
Source file and lines	miniqmc.cpp:472-488
Module	exec

The loop is defined in:

/usr/lib/gcc/aarch64-linux-gnu/11/../../../../include/c++/11/bits/stl_vector.h:1046,1064
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:53,63
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Drivers/miniqmc.cpp:472-488
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h:145-146
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:203
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVectorOps.h:85
/usr/lib/gcc/aarch64-linux-gnu/11/../../../../include/c++/11/bits/refwrap.h:338
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h:263

Warnings:

Non-innermost loop: analyzing only self part (ignoring child loops).
Ignoring paths for analysis
Failed to get the number of paths
RecMII not computed since number of paths is unknown or > max_paths

gain
potential
hint
expert

Code clean check

Detected a slowdown caused by scalar integer instructions (typically used for address computation). By removing them, you can lower the cost of an iteration from 4.33 to 1.33 cycles (3.25x speedup).

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)

Vectorization

Your loop is not vectorized. Only 20% of vector register length is used (average across all VPU instructions). By vectorizing your loop, you can lower the cost of an iteration from 4.33 to 0.66 cycles (6.60x speedup).

Details

All VPU instructions are used in scalar version (process only one data element in vector registers). Since your execution units are vector units, only a vectorized loop can use their full power.

Workaround

Try another compiler or update/tune your current one
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop streams arrays of structures (AoS), try to use structures of arrays instead (SoA): for(i) a[i].x = b[i].x; (slow, non stride 1) => for(i) a.x[i] = b.x[i]; (fast, stride 1)

No data for this section

Matching between your loop (in the source code) and the binary loop

The binary loop does not contain any FP arithmetical operations. The binary loop is loading 92 bytes.

General properties

nb instructions	34
loop length	136
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

4.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	2.00	2.00	4.00	4.00	4.00	4.00	1.00	0.00	0.00	0.00	4.33	4.33	4.33	0.00	0.00
cycles	2.00	2.00	4.00	4.00	4.00	4.00	1.00	0.00	0.00	0.00	4.33	4.33	4.33	0.00	0.00

Cycles executing div or sqrt instructions

NA

Cycles summary

Front-end	4.25
Overall L1	4.33

Vectorization ratios

all	0%
load	0%
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 4.33 cycles. At this rate:

22% of peak load performance is reached (21.23 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 409060

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P10	P11	P12	Latency	Recip. throughput
LDR W9, [X26, #7744]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
LDR X22, [SP, #64]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
LDR X11, [SP, #120]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
ORR X25, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LDR X10, [X11, #24]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
LDR X9, [X11, #64]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
MADD X10, X10, X19, XZR	1	0	0	0	0	1	0	0	0	0	0	2	1
ADD X21, X9, X10,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LDR X9, [X11, #80]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
MOVZ W10, #40	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MADD X9, X19, X10, X9	1	0	0	0	0	1	0	0	0	0	0	2	1
ADD X22, X9, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
ADD X20, X9, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LDR X9, [SP, #128]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
LDR W9, [X9]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
B 40906c <.omp_outlined..62+0xa9c>	1	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50
ADD X25, X25, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
CMP X25, W8,SXTW	1	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
B.GE 409000 <.omp_outlined..62+0xa30>	1	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50
LDR X10, [SP, #112]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
LDR D0, [X21, X25,LSL #3]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
LDR D1, [X10]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
FCMP D0, D1	1	0	0	0	0	0	0	1	0	0	0	2	1
CCMP W9, #1, #8, #11	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
B.LT 409060 <.omp_outlined..62+0xa90>	1	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50
ORR X28, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
ORR X24, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LDR X8, [SP, #104]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
MOVZ W24, #22144	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOVZ W28, #45279	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOVK W24, #40236	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOVK W28, #39176	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LDR W8, [X8]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
B 409060 <.omp_outlined..62+0xa90>	1	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50

Function	.omp_outlined..62
Source file and lines	miniqmc.cpp:472-488
Module	exec

The loop is defined in:

/usr/lib/gcc/aarch64-linux-gnu/11/../../../../include/c++/11/bits/stl_vector.h:1046,1064
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:53,63
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Drivers/miniqmc.cpp:472-488
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h:145-146
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:203
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVectorOps.h:85
/usr/lib/gcc/aarch64-linux-gnu/11/../../../../include/c++/11/bits/refwrap.h:338
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h:263

Warnings:

Non-innermost loop: analyzing only self part (ignoring child loops).
Ignoring paths for analysis
Failed to get the number of paths
RecMII not computed since number of paths is unknown or > max_paths

gain
potential
hint
expert

Code clean check

Detected a slowdown caused by scalar integer instructions (typically used for address computation). By removing them, you can lower the cost of an iteration from 4.33 to 1.33 cycles (3.25x speedup).

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)

Vectorization

Your loop is not vectorized. Only 20% of vector register length is used (average across all VPU instructions). By vectorizing your loop, you can lower the cost of an iteration from 4.33 to 0.66 cycles (6.60x speedup).

Details

All VPU instructions are used in scalar version (process only one data element in vector registers). Since your execution units are vector units, only a vectorized loop can use their full power.

Workaround

Try another compiler or update/tune your current one
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop streams arrays of structures (AoS), try to use structures of arrays instead (SoA): for(i) a[i].x = b[i].x; (slow, non stride 1) => for(i) a.x[i] = b.x[i]; (fast, stride 1)

No data for this section

Matching between your loop (in the source code) and the binary loop

The binary loop does not contain any FP arithmetical operations. The binary loop is loading 92 bytes.

General properties

nb instructions	34
loop length	136
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

4.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	2.00	2.00	4.00	4.00	4.00	4.00	1.00	0.00	0.00	0.00	4.33	4.33	4.33	0.00	0.00
cycles	2.00	2.00	4.00	4.00	4.00	4.00	1.00	0.00	0.00	0.00	4.33	4.33	4.33	0.00	0.00

Cycles executing div or sqrt instructions

NA

Cycles summary

Front-end	4.25
Overall L1	4.33

Vectorization ratios

all	0%
load	0%
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 4.33 cycles. At this rate:

22% of peak load performance is reached (21.23 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 409060

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P10	P11	P12	Latency	Recip. throughput
LDR W9, [X26, #7744]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
LDR X22, [SP, #64]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
LDR X11, [SP, #120]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
ORR X25, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LDR X10, [X11, #24]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
LDR X9, [X11, #64]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
MADD X10, X10, X19, XZR	1	0	0	0	0	1	0	0	0	0	0	2	1
ADD X21, X9, X10,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LDR X9, [X11, #80]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
MOVZ W10, #40	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MADD X9, X19, X10, X9	1	0	0	0	0	1	0	0	0	0	0	2	1
ADD X22, X9, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
ADD X20, X9, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LDR X9, [SP, #128]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
LDR W9, [X9]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
B 40906c <.omp_outlined..62+0xa9c>	1	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50
ADD X25, X25, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
CMP X25, W8,SXTW	1	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
B.GE 409000 <.omp_outlined..62+0xa30>	1	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50
LDR X10, [SP, #112]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
LDR D0, [X21, X25,LSL #3]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
LDR D1, [X10]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
FCMP D0, D1	1	0	0	0	0	0	0	1	0	0	0	2	1
CCMP W9, #1, #8, #11	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
B.LT 409060 <.omp_outlined..62+0xa90>	1	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50
ORR X28, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
ORR X24, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LDR X8, [SP, #104]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
MOVZ W24, #22144	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOVZ W28, #45279	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOVK W24, #40236	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOVK W28, #39176	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LDR W8, [X8]	1	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
B 409060 <.omp_outlined..62+0xa90>	1	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Matching between your loop (in the source code) and the binary loop

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code

Code clean check

Workaround

Vectorization

Details

Workaround

Matching between your loop (in the source code) and the binary loop

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code