OV - exec - Loop 904

0x435f08 LDR	D0, [X24, X5]    [2]

0x435f0c SUBS	X3, X3, #4

0x435f10 STR	D0, [X7]    [1]

0x435f14 LDR	D0, [X24, X4]    [2]

0x435f18 STR	D0, [X7, #8]    [1]

0x435f1c LDR	D0, [X24, X23]    [2]

0x435f20 STR	D0, [X7, #16]    [1]

0x435f24 LDR	D0, [X24, X6]    [2]

0x435f28 ADD	X24, X24, X0

0x435f2c STR	D0, [X7, #24]    [1]

0x435f30 ADD	X7, X7, #32

0x435f34 B.NE	435f08

/home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Platforms/CPU/SIMD/inner_product.hpp: 211 - 212

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

211:     for (size_t j = 0; j < m; ++j)

212:       B[i * ldb + j] = A[j * lda + i];

Coverage (%)	Name	Source Location	Module
►50.00+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►50.00+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.00+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►50.00+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.00+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►50.00+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.46+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.54+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.22+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.78+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.29+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.71+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.43+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.57+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.00+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►50.00+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.40+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.60+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.17+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.83+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.05+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.95+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.16+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.84+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.25+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.75+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.13+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.87+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so

Coverage (%)	Name	Source Location	Module
►50.11+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	exec
○	__kmp_invoke_microtask		libomp.so
►49.89+	miniqmcreference::DiracDetermi[...]	DelayedUpdate.h:79	exec
○	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:238	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main.omp_outlined	miniqmc.cpp:379	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	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	2.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.33
Bottlenecks	P12, P13, P14,
Function	qmcplusplus::DiracMatrix::invert_transpose(qmcplusplus::Matrix > const&, qmcplusplus::Matrix >&, double&, double&)
Source	inner_product.hpp:211-212
Source loop unroll info	unrolled by 4
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	4
CQA cycles	2.67
CQA cycles if no scalar integer	2.67
CQA cycles if FP arith vectorized	2.67
CQA cycles if fully vectorized	1.33
Front-end cycles	1.50
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.58
P3 cycles	0.58
P4 cycles	0.50
P5 cycles	0.50
P6 cycles	0.42
P7 cycles	0.42
P8 cycles	2.00
P9 cycles	2.00
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	2.67
P13 cycles	2.67
P14 cycles	2.67
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	12.00
Nb uops	12.00
Nb loads	NA
Nb stores	4.00
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	24.00
Bytes prefetched	0.00
Bytes loaded	32.00
Bytes stored	32.00
Stride 0	0.00
Stride 1	1.00
Stride n	0.00
Stride unknown	1.00
Stride indirect	0.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	NA
Vector-efficiency ratio all	50.00
Vector-efficiency ratio load	50.00
Vector-efficiency ratio store	50.00
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	NA

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	2.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.33
Bottlenecks	P12, P13, P14,
Function	qmcplusplus::DiracMatrix::invert_transpose(qmcplusplus::Matrix > const&, qmcplusplus::Matrix >&, double&, double&)
Source	inner_product.hpp:211-212
Source loop unroll info	unrolled by 4
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	4
CQA cycles	2.67
CQA cycles if no scalar integer	2.67
CQA cycles if FP arith vectorized	2.67
CQA cycles if fully vectorized	1.33
Front-end cycles	1.50
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.58
P3 cycles	0.58
P4 cycles	0.50
P5 cycles	0.50
P6 cycles	0.42
P7 cycles	0.42
P8 cycles	2.00
P9 cycles	2.00
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	2.67
P13 cycles	2.67
P14 cycles	2.67
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	12.00
Nb uops	12.00
Nb loads	NA
Nb stores	4.00
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	24.00
Bytes prefetched	0.00
Bytes loaded	32.00
Bytes stored	32.00
Stride 0	0.00
Stride 1	1.00
Stride n	0.00
Stride unknown	1.00
Stride indirect	0.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	NA
Vector-efficiency ratio all	50.00
Vector-efficiency ratio load	50.00
Vector-efficiency ratio store	50.00
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	NA

Path /

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

Function	qmcplusplus::DiracMatrix::invert_transpose(qmcplusplus::Matrix > const&, qmcplusplus::Matrix >&, double&, double&)
Source file and lines	inner_product.hpp:211-212
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Platforms/CPU/SIMD/inner_product.hpp:211-212.

The related source loop is unrolled by 4 (including vectorization).

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. 2 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 2.67 to 1.33 cycles (2.00x 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 32 bytes. The binary loop is storing 32 bytes.

General properties

nb instructions	12
loop length	48
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

1.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16
uops	0.50	0.50	0.58	0.58	0.50	0.50	0.42	0.42	2.00	2.00	0.00	0.00	2.67	2.67	2.67	0.00	0.00
cycles	0.50	0.50	0.58	0.58	0.50	0.50	0.42	0.42	2.00	2.00	0.00	0.00	2.67	2.67	2.67	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
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	1.50
Data deps.	1.00
Overall L1	2.67

Vectorization ratios

all	0%
load	0%
store	0%
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	NA (no other vectorizable/vectorized instructions)

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 2.67 cycles. At this rate:

12% of peak load performance is reached (12.00 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
24% of peak store performance is reached (12.00 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 435f08

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P12	P13	P14	Latency	Recip. throughput	Vectorization
LDR D0, [X24, X5]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
SUBS X3, X3, #4	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	1	0.33	scal (50.0%)
STR D0, [X7]	1	0	0	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50	scal (50.0%)
LDR D0, [X24, X4]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
STR D0, [X7, #8]	1	0	0	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50	scal (50.0%)
LDR D0, [X24, X23]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
STR D0, [X7, #16]	1	0	0	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50	scal (50.0%)
LDR D0, [X24, X6]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
ADD X24, X24, X0	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
STR D0, [X7, #24]	1	0	0	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50	scal (50.0%)
ADD X7, X7, #32	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
B.NE 435f08 <_ZN11qmcplusplus11DiracMatrixIddE16invert_transposeERKNS_6MatrixIdSaIdEEERS4_RdS8_+0x188>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	qmcplusplus::DiracMatrix::invert_transpose(qmcplusplus::Matrix > const&, qmcplusplus::Matrix >&, double&, double&)
Source file and lines	inner_product.hpp:211-212
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/174-135-6342/intel/miniqmc/build/miniqmc/src/Platforms/CPU/SIMD/inner_product.hpp:211-212.

The related source loop is unrolled by 4 (including vectorization).

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. 2 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 2.67 to 1.33 cycles (2.00x 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 32 bytes. The binary loop is storing 32 bytes.

General properties

nb instructions	12
loop length	48
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

1.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16
uops	0.50	0.50	0.58	0.58	0.50	0.50	0.42	0.42	2.00	2.00	0.00	0.00	2.67	2.67	2.67	0.00	0.00
cycles	0.50	0.50	0.58	0.58	0.50	0.50	0.42	0.42	2.00	2.00	0.00	0.00	2.67	2.67	2.67	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
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	1.50
Data deps.	1.00
Overall L1	2.67

Vectorization ratios

all	0%
load	0%
store	0%
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	NA (no other vectorizable/vectorized instructions)

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 2.67 cycles. At this rate:

12% of peak load performance is reached (12.00 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
24% of peak store performance is reached (12.00 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 435f08

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P12	P13	P14	Latency	Recip. throughput	Vectorization
LDR D0, [X24, X5]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
SUBS X3, X3, #4	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	1	0.33	scal (50.0%)
STR D0, [X7]	1	0	0	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50	scal (50.0%)
LDR D0, [X24, X4]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
STR D0, [X7, #8]	1	0	0	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50	scal (50.0%)
LDR D0, [X24, X23]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
STR D0, [X7, #16]	1	0	0	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50	scal (50.0%)
LDR D0, [X24, X6]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
ADD X24, X24, X0	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
STR D0, [X7, #24]	1	0	0	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50	scal (50.0%)
ADD X7, X7, #32	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
B.NE 435f08 <_ZN11qmcplusplus11DiracMatrixIddE16invert_transposeERKNS_6MatrixIdSaIdEEERS4_RdS8_+0x188>	1	0.50	0.50	0	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	1	0	1	0	1.03	0	1	0	0.97	0	0.95	0	0.92	0.01	0.9	0.01	0.88	0.01	0.86	0.01	0.82	0.01	0.8	0.01	0.77	0.02	0.74	0.02

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	0.070000000298023	0.16265869140625
1x2	2	1	1	2	0.070000000298023	0.1601360887289
1x4	4	1	1	4	0.074999995529652	0.15594109892845
1x8	8	1.03	1.03	8	0.070000000298023	0.1420037150383
1x16	16	1	1	16	0.070000000298023	0.12320227175951
1x24	24	0.97	0.97	24	0.079999983310699	0.11460890620947
1x32	32	0.95	0.95	32	0.079999998211861	0.10248920321465
1x40	40	0.92	0.92	40	0.07999999076128	0.096092231571674
1x48	48	0.9	0.9	48	0.08500000089407	0.091042406857014
1x56	56	0.88	0.88	56	0.084999993443489	0.08513829857111
1x64	64	0.86	0.86	64	0.089999981224537	0.079962804913521
1x72	72	0.82	0.82	72	0.090000011026859	0.076889917254448
1x80	80	0.8	0.8	80	0.094999976456165	0.07278336584568
1x88	88	0.77	0.77	88	0.099999994039536	0.071097582578659
1x96	96	0.74	0.74	96	0.10499998927116	0.068641662597656

Report Configuration

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

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

▶Scalability Runs Description

▼Columns Filter

Speed-Up and Efficiency

▶Detailed Speed-Up and Efficiency