OV - exec - Loop 61

0x40ac00 LDR	X9, [X20, #608]

  (60) 0x40ac04 ADD	X23, X23, #1

  (60) 0x40ac08 CMP	X9, X23

  (60) 0x40ac0c B.LS	40ad60

  (60) 0x40ac10 LDR	X10, [X8, #608]

  (60) 0x40ac14 CBZ	X10, 40ac04

0x40ac18 LDR	X11, [SP]

0x40ac1c ADD	X10, X23, X23,LSL #2

0x40ac20 ORR	X24, XZR, XZR

0x40ac24 LDR	X9, [X11, #96]

0x40ac28 ADD	X28, X9, X10,LSL #3

0x40ac2c LDR	X9, [X11, #72]

0x40ac30 ADD	X22, X9, X10,LSL #3

0x40ac34 B	40ac50

0x40ac40 LDR	X9, [X8, #608]

0x40ac44 ADD	X24, X24, #1

0x40ac48 CMP	X9, X24

0x40ac4c B.LS	40ac00

0x40ac50 LDUR	X9, [X22, #24]

0x40ac54 LDR	D1, [X21]

0x40ac58 LDR	D0, [X9, X24,LSL #3]

0x40ac5c FCMP	D0, D1

0x40ac60 B.GE	40ac40

0x40ac64 LDR	X9, [X21, #5104]

0x40ac68 LDR	X10, [X21, #5096]

0x40ac6c SUB	X9, X9, X10

0x40ac70 UBFM	X9, X9, #3, #63

0x40ac74 MADD	W9, W9, W26, WZR

0x40ac78 CMP	W9, #1

0x40ac7c B.LT	40ad1c

0x40ac80 ORR	X8, XZR, XZR

0x40ac84 ORR	X9, XZR, XZR

  (62) 0x40ac88 LDUR	X10, [X22, #24]

  (62) 0x40ac8c LDUR	X11, [X29, #496]

  (62) 0x40ac90 LDR	X12, [X28, #8]

  (62) 0x40ac94 ADD	X9, X9, #1

  (62) 0x40ac98 ADD	X11, X11, X8

  (62) 0x40ac9c LDR	D0, [X10, X24,LSL #3]

  (62) 0x40aca0 LDR	X10, [X28, #24]

  (62) 0x40aca4 UBFM	X13, X12, #32, #31

  (62) 0x40aca8 SBFM	X12, X12, #60, #31

  (62) 0x40acac LDR	D1, [X11, #16]

  (62) 0x40acb0 ADD	X10, X10, X24,LSL #3

  (62) 0x40acb4 LDR	D2, [X10]

  (62) 0x40acb8 ADD	X13, X10, X13,ASR #29

  (62) 0x40acbc LDR	D3, [X10, X12]

  (62) 0x40acc0 LDR	X10, [X20, #64]

  (62) 0x40acc4 LDR	X12, [SP, #32]

  (62) 0x40acc8 MADD	X10, X23, X27, X10

  (62) 0x40accc ADD	X12, X12, X8

  (62) 0x40acd0 ADD	X8, X8, #24

  (62) 0x40acd4 FMADD	D1, D1, D0, D3

  (62) 0x40acd8 LDR	D3, [X10, #16]

  (62) 0x40acdc LD1	{V2.D[1]}, [X13]

  (62) 0x40ace0 FADD	D1, D1, D3

  (62) 0x40ace4 LDR	Q3, [X11]

  (62) 0x40ace8 FMLA	V2.2D, V3.2D, V0.D[0]

  (62) 0x40acec LDR	Q0, [X10]

  (62) 0x40acf0 STR	D1, [X12, #16]

  (62) 0x40acf4 FADD	V0.2D, V2.2D, V0.2D

  (62) 0x40acf8 STR	Q0, [X12]

  (62) 0x40acfc LDR	X10, [X21, #5104]

  (62) 0x40ad00 LDR	X11, [X21, #5096]

  (62) 0x40ad04 SUB	X10, X10, X11

  (62) 0x40ad08 UBFM	X10, X10, #3, #63

  (62) 0x40ad0c MADD	W10, W10, W26, WZR

  (62) 0x40ad10 CMP	X9, W10,SXTW

  (62) 0x40ad14 B.LT	40ac88

0x40ad18 LDR	X8, [X21, #5144]

0x40ad1c LDR	X8, [X8, #24]

0x40ad20 LDR	X9, [X21, #5120]

0x40ad24 LDRSW	X8, [X8, X24,LSL #2]

0x40ad28 LDR	X25, [X9, X8,LSL #3]

0x40ad2c ADD	X3, SP, #32

0x40ad30 MOVZ	W4, #1

0x40ad34 ORR	X0, XZR, X25

0x40ad38 ORR	X1, XZR, X20

0x40ad3c ORR	W2, WZR, W23

0x40ad40 ORR	W5, WZR, W24

0x40ad44 BL	44ab90

0x40ad48 ADD	X2, SP, #8

0x40ad4c ORR	X0, XZR, X19

0x40ad50 ORR	X1, XZR, X25

0x40ad54 BL	40dee0

0x40ad58 LDR	X8, [X21, #5144]

0x40ad5c B	40ac40

/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Drivers/NonLocalPP.hpp: 122 - 135

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

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

123:     {

124:       const auto& dist  = d_ie.getDistRow(jel);

125:       const auto& displ = d_ie.getDisplRow(jel);

126:       for (int iat = 0; iat < ions_ref.getTotalNum(); ++iat)

127:       {

128:         //due to < Rmax condition, the actually iteration iat is [0,2] in a real simulation

129:         if (dist[iat] < Rmax)

130:         {

131:           for (int k = 0; k < size(); k++)

132:             virtualPos[k] = dist[iat] * rOnSphere[k] + displ[iat] + els.R[jel];

133:           auto& VP = *VPs[ions_ref.GroupID[iat]];

134:           VP.makeMoves(els, jel, virtualPos, true, iat);

135:           wf.evaluateRatios(VP, ratios);

/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h: 223 - 229

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

223:     return X[i];

[...]

229:     return X[i];

/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h: 277 - 277

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

277:   inline size_t getTotalNum() const { return TotalNum; }

/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h: 43 - 63

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

43:     return (a + b);

[...]

63:     return (a * b);

/opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/../../../../include/c++/14.2.0/bits/stl_vector.h: 993 - 1150

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

993:       { return size_type(this->_M_impl._M_finish - this->_M_impl._M_start); }

[...]

1131: 	return *(this->_M_impl._M_start + __n);

[...]

1150: 	return *(this->_M_impl._M_start + __n);

/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h: 231 - 231

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

231:   inline const AoSElement_t operator[](size_t i) const { return AoSElement_t(myData + i, nGhosts); }

/opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/../../../../include/c++/14.2.0/bits/unique_ptr.h: 193 - 193

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

193:       pointer    _M_ptr() const noexcept { return std::get<0>(_M_t); }

/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h: 144 - 145

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

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

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

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►100.00+	main.omp_outlined.63	NewTimer.h:249	exec
○	__kmp_invoke_microtask		libomp.so

min	med	avg	max

Percentile Index	10	20	30	40	50	60	70	80	90	100
Value

min	med	avg	max

Percentile Index	10	20	30	40	50	60	70	80	90	100
Value

Path /

Metric	Value
CQA speedup if no scalar integer	8.44
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	3.10
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.05
Bottlenecks	micro-operation queue,
Function	qmcplusplus::NonLocalPP::evaluate(qmcplusplus::ParticleSet const&, qmcplusplus::WaveFunction&)
Source	NonLocalPP.hpp:126-126,NonLocalPP.hpp:129-135,OhmmsVector.h:229-229,ParticleSet.h:277-277,stl_vector.h:993-993,stl_vector.h:1131-1131,stl_vector.h:1150-1150,unique_ptr.h:193-193
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	5.63
CQA cycles if no scalar integer	0.67
CQA cycles if FP arith vectorized	5.63
CQA cycles if fully vectorized	1.81
Front-end cycles	5.63
P0 cycles	3.50
P1 cycles	3.50
P2 cycles	3.92
P3 cycles	3.92
P4 cycles	3.83
P5 cycles	3.83
P6 cycles	3.75
P7 cycles	3.75
P8 cycles	1.00
P9 cycles	0.00
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	5.33
P13 cycles	5.33
P14 cycles	5.33
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	45.00
Nb uops	45.00
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	22.04
Bytes prefetched	0.00
Bytes loaded	124.00
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	45.31
Vector-efficiency ratio load	50.00
Vector-efficiency ratio store	NA
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	50.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	41.67

Metric	Value
CQA speedup if no scalar integer	8.44
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	3.10
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.05
Bottlenecks	micro-operation queue,
Function	qmcplusplus::NonLocalPP::evaluate(qmcplusplus::ParticleSet const&, qmcplusplus::WaveFunction&)
Source	NonLocalPP.hpp:126-126,NonLocalPP.hpp:129-135,OhmmsVector.h:229-229,ParticleSet.h:277-277,stl_vector.h:993-993,stl_vector.h:1131-1131,stl_vector.h:1150-1150,unique_ptr.h:193-193
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	5.63
CQA cycles if no scalar integer	0.67
CQA cycles if FP arith vectorized	5.63
CQA cycles if fully vectorized	1.81
Front-end cycles	5.63
P0 cycles	3.50
P1 cycles	3.50
P2 cycles	3.92
P3 cycles	3.92
P4 cycles	3.83
P5 cycles	3.83
P6 cycles	3.75
P7 cycles	3.75
P8 cycles	1.00
P9 cycles	0.00
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	5.33
P13 cycles	5.33
P14 cycles	5.33
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	45.00
Nb uops	45.00
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	22.04
Bytes prefetched	0.00
Bytes loaded	124.00
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	45.31
Vector-efficiency ratio load	50.00
Vector-efficiency ratio store	NA
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	50.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	41.67

Path /

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

Function	qmcplusplus::NonLocalPP::evaluate(qmcplusplus::ParticleSet const&, qmcplusplus::WaveFunction&)
Source file and lines	NonLocalPP.hpp:122-135
Module	exec

The loop is defined in:

/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Drivers/NonLocalPP.hpp:122-135
/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223,229
/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h:277
/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:43,63
/opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/../../../../include/c++/14.2.0/bits/stl_vector.h:993,1131,1150
/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:231
/opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/../../../../include/c++/14.2.0/bits/unique_ptr.h:193
/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h:144-145

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
Streams not analyzed 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 5.63 to 0.67 cycles (8.44x 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 45% of vector register length is used (average across all VPU instructions). By vectorizing your loop, you can lower the cost of an iteration from 5.63 to 1.81 cycles (3.10x 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)

Expensive FP math instructions/calls

Detected performance impact from expensive FP math instructions/calls. By removing/reexpressing them, you can lower the cost of an iteration from 5.63 to 1.00 cycles (5.62x speedup).

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 124 bytes.

General properties

nb instructions	45
loop length	180
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

5.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16
uops	3.50	3.50	3.92	3.92	3.83	3.83	3.75	3.75	1.00	0.00	0.00	0.00	5.33	5.33	5.33	0.00	0.00
cycles	3.50	3.50	3.92	3.92	3.83	3.83	3.75	3.75	1.00	0.00	0.00	0.00	5.33	5.33	5.33	0.00	0.00

Execution ports to units layout:

P0: BRU
P1: BRU
P2: ALU
P3: ALU
P4: ALU
P5: ALU
P6: ALU
P7: ALU
P8 (128 bits): VPU, FP store data, ALU, DIV/SQRT
P9 (128 bits): VPU, ALU, FP store data
P10 (128 bits): VPU, ALU, DIV/SQRT
P11 (128 bits): ALU, VPU
P12 (256 bits): store address, load
P13 (256 bits): store address, load
P14 (256 bits): load
P15 (64 bits): store data
P16 (64 bits): store data

Cycles executing div or sqrt instructions

NA

Cycles summary

Front-end	5.63
Overall L1	5.63

Vectorization ratios

INT

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)
other	0%

FP

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	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%

INT+FP

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 5.63 cycles. At this rate:

22% of peak load performance is reached (22.04 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 40ac40

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P12	P13	P14	Latency	Recip. throughput	Vectorization
LDR X9, [X20, #608]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR X11, [SP]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
ADD X10, X23, X23,LSL #2	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
ORR X24, XZR, XZR	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LDR X9, [X11, #96]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
ADD X28, X9, X10,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
LDR X9, [X11, #72]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
ADD X22, X9, X10,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
B 40ac50 <_ZN11qmcplusplus10NonLocalPPIdE8evaluateERKNS_11ParticleSetERNS_12WaveFunctionE+0x1f0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X9, [X8, #608]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
ADD X24, X24, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP X9, X24	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	1	0.33	scal (50.0%)
B.LS 40ac00 <_ZN11qmcplusplus10NonLocalPPIdE8evaluateERKNS_11ParticleSetERNS_12WaveFunctionE+0x1a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDUR X9, [X22, #24]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR D1, [X21]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
LDR D0, [X9, X24,LSL #3]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
FCMP D0, D1	1	0	0	0	0	0	0	0	0	1	0	0	0	2	1	scal (50.0%)
B.GE 40ac40 <_ZN11qmcplusplus10NonLocalPPIdE8evaluateERKNS_11ParticleSetERNS_12WaveFunctionE+0x1e0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X9, [X21, #5104]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR X10, [X21, #5096]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	scal (50.0%)
SUB X9, X9, X10	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
UBFM X9, X9, #3, #63	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MADD W9, W9, W26, WZR	1	0	0	0	0	0	0	1	0	0	0	0	0	2	1	N/A
CMP W9, #1	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	1	0.33	scal (25.0%)
B.LT 40ad1c <_ZN11qmcplusplus10NonLocalPPIdE8evaluateERKNS_11ParticleSetERNS_12WaveFunctionE+0x2bc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X8, XZR, XZR	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ORR X9, XZR, XZR	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LDR X8, [X21, #5144]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR X8, [X8, #24]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR X9, [X21, #5120]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDRSW X8, [X8, X24,LSL #2]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR X25, [X9, X8,LSL #3]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	scal (50.0%)
ADD X3, SP, #32	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
MOVZ W4, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (25.0%)
ORR X0, XZR, X25	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
ORR X1, XZR, X20	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
ORR W2, WZR, W23	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ORR W5, WZR, W24	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (25.0%)
BL 44ab90 <_ZN11qmcplusplus18VirtualParticleSet9makeMovesERKNS_11ParticleSetEiRKSt6vectorINS_10TinyVectorIdLj3EEESaIS6_EEbi>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X2, SP, #8	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ORR X0, XZR, X19	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
ORR X1, XZR, X25	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
BL 40dee0 <_ZN11qmcplusplus12WaveFunction14evaluateRatiosERNS_18VirtualParticleSetERSt6vectorIdSaIdEE>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X8, [X21, #5144]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
B 40ac40 <_ZN11qmcplusplus10NonLocalPPIdE8evaluateERKNS_11ParticleSetERNS_12WaveFunctionE+0x1e0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	qmcplusplus::NonLocalPP::evaluate(qmcplusplus::ParticleSet const&, qmcplusplus::WaveFunction&)
Source file and lines	NonLocalPP.hpp:122-135
Module	exec

The loop is defined in:

/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Drivers/NonLocalPP.hpp:122-135
/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223,229
/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h:277
/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:43,63
/opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/../../../../include/c++/14.2.0/bits/stl_vector.h:993,1131,1150
/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:231
/opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/../../../../include/c++/14.2.0/bits/unique_ptr.h:193
/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h:144-145

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
Streams not analyzed 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 5.63 to 0.67 cycles (8.44x 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 45% of vector register length is used (average across all VPU instructions). By vectorizing your loop, you can lower the cost of an iteration from 5.63 to 1.81 cycles (3.10x 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)

Expensive FP math instructions/calls

Detected performance impact from expensive FP math instructions/calls. By removing/reexpressing them, you can lower the cost of an iteration from 5.63 to 1.00 cycles (5.62x speedup).

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 124 bytes.

General properties

nb instructions	45
loop length	180
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

5.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16
uops	3.50	3.50	3.92	3.92	3.83	3.83	3.75	3.75	1.00	0.00	0.00	0.00	5.33	5.33	5.33	0.00	0.00
cycles	3.50	3.50	3.92	3.92	3.83	3.83	3.75	3.75	1.00	0.00	0.00	0.00	5.33	5.33	5.33	0.00	0.00

Execution ports to units layout:

P0: BRU
P1: BRU
P2: ALU
P3: ALU
P4: ALU
P5: ALU
P6: ALU
P7: ALU
P8 (128 bits): VPU, FP store data, ALU, DIV/SQRT
P9 (128 bits): VPU, ALU, FP store data
P10 (128 bits): VPU, ALU, DIV/SQRT
P11 (128 bits): ALU, VPU
P12 (256 bits): store address, load
P13 (256 bits): store address, load
P14 (256 bits): load
P15 (64 bits): store data
P16 (64 bits): store data

Cycles executing div or sqrt instructions

NA

Cycles summary

Front-end	5.63
Overall L1	5.63

Vectorization ratios

INT

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)
other	0%

FP

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	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%

INT+FP

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 5.63 cycles. At this rate:

22% of peak load performance is reached (22.04 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 40ac40

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P12	P13	P14	Latency	Recip. throughput	Vectorization
LDR X9, [X20, #608]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR X11, [SP]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
ADD X10, X23, X23,LSL #2	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
ORR X24, XZR, XZR	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LDR X9, [X11, #96]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
ADD X28, X9, X10,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
LDR X9, [X11, #72]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
ADD X22, X9, X10,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
B 40ac50 <_ZN11qmcplusplus10NonLocalPPIdE8evaluateERKNS_11ParticleSetERNS_12WaveFunctionE+0x1f0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X9, [X8, #608]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
ADD X24, X24, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP X9, X24	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	1	0.33	scal (50.0%)
B.LS 40ac00 <_ZN11qmcplusplus10NonLocalPPIdE8evaluateERKNS_11ParticleSetERNS_12WaveFunctionE+0x1a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDUR X9, [X22, #24]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR D1, [X21]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
LDR D0, [X9, X24,LSL #3]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
FCMP D0, D1	1	0	0	0	0	0	0	0	0	1	0	0	0	2	1	scal (50.0%)
B.GE 40ac40 <_ZN11qmcplusplus10NonLocalPPIdE8evaluateERKNS_11ParticleSetERNS_12WaveFunctionE+0x1e0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X9, [X21, #5104]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR X10, [X21, #5096]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	scal (50.0%)
SUB X9, X9, X10	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
UBFM X9, X9, #3, #63	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MADD W9, W9, W26, WZR	1	0	0	0	0	0	0	1	0	0	0	0	0	2	1	N/A
CMP W9, #1	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	1	0.33	scal (25.0%)
B.LT 40ad1c <_ZN11qmcplusplus10NonLocalPPIdE8evaluateERKNS_11ParticleSetERNS_12WaveFunctionE+0x2bc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X8, XZR, XZR	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ORR X9, XZR, XZR	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LDR X8, [X21, #5144]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR X8, [X8, #24]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR X9, [X21, #5120]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDRSW X8, [X8, X24,LSL #2]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
LDR X25, [X9, X8,LSL #3]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	scal (50.0%)
ADD X3, SP, #32	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
MOVZ W4, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (25.0%)
ORR X0, XZR, X25	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
ORR X1, XZR, X20	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
ORR W2, WZR, W23	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ORR W5, WZR, W24	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (25.0%)
BL 44ab90 <_ZN11qmcplusplus18VirtualParticleSet9makeMovesERKNS_11ParticleSetEiRKSt6vectorINS_10TinyVectorIdLj3EEESaIS6_EEbi>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X2, SP, #8	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ORR X0, XZR, X19	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
ORR X1, XZR, X25	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (50.0%)
BL 40dee0 <_ZN11qmcplusplus12WaveFunction14evaluateRatiosERNS_18VirtualParticleSetERSt6vectorIdSaIdEE>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X8, [X21, #5144]	1	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33	N/A
B 40ac40 <_ZN11qmcplusplus10NonLocalPPIdE8evaluateERKNS_11ParticleSetERNS_12WaveFunctionE+0x1e0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

▶Scalability Runs Description

Run 1x1	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_NUM_THREADS: 1OMP_PLACES: threads
Run 1x2	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 2OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x4	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 4OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x8	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 8OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x16	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 16OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x24	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 24OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x32	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 32OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x40	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 40OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x48	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 48OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x56	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 56OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x64	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 64OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x72	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 72OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x80	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 80OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x88	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 88OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x96	Number processes: 1Number nodes: 1Run Command: <executable> -g "4 2 2" -bMPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=96 --rank-by fill --report-bindings Dataset: Run Directory: /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/run/oneview_runs/multicore/armclang/oneview_run_1741388363OMP_NUM_THREADS: 96OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

(1x1) Efficiency	(1x1) Potential Speed-Up (%)	(1x2) Efficiency	(1x2) Potential Speed-Up (%)	(1x4) Efficiency	(1x4) Potential Speed-Up (%)	(1x8) Efficiency	(1x8) Potential Speed-Up (%)	(1x16) Efficiency	(1x16) Potential Speed-Up (%)	(1x24) Efficiency	(1x24) Potential Speed-Up (%)	(1x32) Efficiency	(1x32) Potential Speed-Up (%)	(1x40) Efficiency	(1x40) Potential Speed-Up (%)	(1x48) Efficiency	(1x48) Potential Speed-Up (%)	(1x56) Efficiency	(1x56) Potential Speed-Up (%)	(1x64) Efficiency	(1x64) Potential Speed-Up (%)	(1x72) Efficiency	(1x72) Potential Speed-Up (%)	(1x80) Efficiency	(1x80) Potential Speed-Up (%)	(1x88) Efficiency	(1x88) Potential Speed-Up (%)	(1x96) Efficiency	(1x96) Potential Speed-Up (%)
1	0	0.2	0.05	0.24	0.04	0.15	0.06	0.13	0.06	0.13	0.05	0.09	0.07	0.08	0.07	0.08	0.07	0.07	0.07	0.06	0.07	0.06	0.08	0.05	0.08	0.05	0.08	0.04	0.09

Speed-Up and Efficiency

▶Detailed Speed-Up and Efficiency

Run	Number of threads	Efficiency (ideal is 1)	Speedup	Ideal Speedup	Time (s)	Coverage (%)
1x1	1	1	1	1	0.0049999998882413	0.011618477292359
1x2	2	0.2	0.2	2	0.029999999329448	0.057191465049982
1x4	4	0.24	0.24	4	0.034999996423721	0.047339260578156
1x8	8	0.15	0.15	8	0.059999994933605	0.067744880914688
1x16	16	0.13	0.13	16	0.070000000298023	0.066849663853645
1x24	24	0.13	0.13	24	0.064999990165234	0.060636106878519
1x32	32	0.09	0.09	32	0.08500000089407	0.078296475112438
1x40	40	0.08	0.08	40	0.089999988675117	0.076299145817757
1x48	48	0.08	0.08	48	0.099999986588955	0.07219760119915
1x56	56	0.07	0.07	56	0.114999987185	0.075229540467262
1x64	64	0.06	0.06	64	0.12499997764826	0.079423032701015
1x72	72	0.06	0.06	72	0.16500000655651	0.079798579216003
1x80	80	0.05	0.05	80	0.16499997675419	0.08203237503767
1x88	88	0.05	0.05	88	0.1499999910593	0.082229025661945
1x96	96	0.04	0.04	96	0.18999998271465	0.089050710201263

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Expensive FP math instructions/calls

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

Expensive FP math instructions/calls

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

▶Scalability Runs Description

▼Columns Filter

Speed-Up and Efficiency

▶Detailed Speed-Up and Efficiency