OV - exec - Loop 322

	Loop Id: 322	Module: exec	Source: BsplineFunctor.h:236-241	Coverage: 2.41%

0x41a4f8 LDR	D8, [X1]    [6]

0x41a4fc FCMPE	D8, D0

0x41a500 B.GE	41a514

0x41a504 CMP	W2, W7

0x41a508 B.EQ	41a514

0x41a50c STR	D8, [X9, X5,SXTW #3]    [10]

0x41a510 ADD	W5, W5, #1

0x41a514 LDR	D9, [X1, #8]    [6]

0x41a518 ADD	W7, W7, #1

0x41a51c ADD	X1, X1, #8

0x41a520 FCMPE	D9, D0

0x41a524 B.GE	41a538

0x41a528 CMP	W2, W7

0x41a52c B.EQ	41a538

0x41a530 STR	D9, [X9, X5,SXTW #3]    [9]

0x41a534 ADD	W5, W5, #1

0x41a538 LDR	D10, [X1, #8]    [7]

0x41a53c ADD	W8, W7, #1

0x41a540 FCMPE	D10, D0

0x41a544 B.GE	41a558

0x41a548 CMP	W2, W8

0x41a54c B.EQ	41a558

0x41a550 STR	D10, [X9, X5,SXTW #3]    [8]

0x41a554 ADD	W5, W5, #1

0x41a558 LDR	D11, [X1, #16]    [7]

0x41a55c ADD	W17, W7, #2

0x41a560 FCMPE	D11, D0

0x41a564 B.GE	41a578

0x41a568 CMP	W2, W17

0x41a56c B.EQ	41a578

0x41a570 STR	D11, [X9, X5,SXTW #3]    [5]

0x41a574 ADD	W5, W5, #1

0x41a578 LDR	D12, [X1, #24]    [7]

0x41a57c ADD	W3, W7, #3

0x41a580 FCMPE	D12, D0

0x41a584 B.GE	41a598

0x41a588 CMP	W2, W3

0x41a58c B.EQ	41a598

0x41a590 STR	D12, [X9, X5,SXTW #3]    [4]

0x41a594 ADD	W5, W5, #1

0x41a598 LDR	D13, [X1, #32]    [7]

0x41a59c ADD	W4, W7, #4

0x41a5a0 FCMPE	D13, D0

0x41a5a4 B.GE	41a5b8

0x41a5a8 CMP	W2, W4

0x41a5ac B.EQ	41a5b8

0x41a5b0 STR	D13, [X9, X5,SXTW #3]    [3]

0x41a5b4 ADD	W5, W5, #1

0x41a5b8 LDR	D14, [X1, #40]    [7]

0x41a5bc ADD	W6, W7, #5

0x41a5c0 FCMPE	D14, D0

0x41a5c4 B.GE	41a5d8

0x41a5c8 CMP	W2, W6

0x41a5cc B.EQ	41a5d8

0x41a5d0 STR	D14, [X9, X5,SXTW #3]    [2]

0x41a5d4 ADD	W5, W5, #1

0x41a5d8 LDR	D15, [X1, #48]    [7]

0x41a5dc ADD	W8, W7, #6

0x41a5e0 FCMPE	D15, D0

0x41a5e4 B.GE	41a5f8

0x41a5e8 CMP	W2, W8

0x41a5ec B.EQ	41a5f8

0x41a5f0 STR	D15, [X9, X5,SXTW #3]    [1]

0x41a5f4 ADD	W5, W5, #1

0x41a5f8 ADD	W7, W7, #7

0x41a5fc ADD	X1, X1, #56

0x41a600 CMP	W10, W7

0x41a604 B.NE	41a4f8

/home/hbollore/qaas/qaas-runs/169-816-7419/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/BsplineFunctor.h: 236 - 241

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

236:   for (int jat = 0; jat < iLimit; jat++)

237:   {

238:     real_type r = distArray[jat];

239:     // pick the distances smaller than the cutoff and avoid the reference atom

240:     if (r < cutoff_radius && iStart + jat != iat)

241:       distArrayCompressed[iCount++] = distArray[jat];

Coverage (%)	Name	Source Location	Module
►75.00+	main._omp_fn.1	refwrap.h:337	exec
○	GOMP_parallel		libomp.so
○	main	miniqmc.cpp:404	exec
○	__libc_start_main		libc-2.31.so
○	_start	miniqmc.cpp:546	exec
►25.00+	main._omp_fn.1	refwrap.h:337	exec
○	GOMP_parallel		libomp.so
○	main	miniqmc.cpp:402	exec
○	__libc_start_main		libc-2.31.so
○	_start	miniqmc.cpp:546	exec

Path /

Metric	Value
CQA speedup if no scalar integer	1.59
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	5.67
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.06
Bottlenecks	micro-operation queue, P0, P1,
Function	miniqmcreference::TwoBodyJastrowRef >::ratio(qmcplusplus::ParticleSet&, int)
Source	BsplineFunctor.h:236-241
Source loop unroll info	not unrolled or unrolled with no peel/tail loop
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	8.50
CQA cycles if no scalar integer	5.33
CQA cycles if FP arith vectorized	8.50
CQA cycles if fully vectorized	1.50
Front-end cycles	8.50
DIV/SQRT cycles	8.50
P0 cycles	8.50
P1 cycles	6.75
P2 cycles	6.75
P3 cycles	6.75
P4 cycles	6.75
P5 cycles	8.00
P6 cycles	8.00
P7 cycles	0.00
P8 cycles	0.00
P9 cycles	5.33
P10 cycles	5.33
P11 cycles	5.33
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	68.00
Nb uops	68.00
Nb loads	NA
Nb stores	8.00
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	15.06
Bytes prefetched	0.00
Bytes loaded	64.00
Bytes stored	64.00
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	0.00
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	19.17
Vector-efficiency ratio load	25.00
Vector-efficiency ratio store	25.00
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	12.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	12.50

Metric	Value
CQA speedup if no scalar integer	1.59
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	5.67
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.06
Bottlenecks	micro-operation queue, P0, P1,
Function	miniqmcreference::TwoBodyJastrowRef >::ratio(qmcplusplus::ParticleSet&, int)
Source	BsplineFunctor.h:236-241
Source loop unroll info	not unrolled or unrolled with no peel/tail loop
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	8.50
CQA cycles if no scalar integer	5.33
CQA cycles if FP arith vectorized	8.50
CQA cycles if fully vectorized	1.50
Front-end cycles	8.50
DIV/SQRT cycles	8.50
P0 cycles	8.50
P1 cycles	6.75
P2 cycles	6.75
P3 cycles	6.75
P4 cycles	6.75
P5 cycles	8.00
P6 cycles	8.00
P7 cycles	0.00
P8 cycles	0.00
P9 cycles	5.33
P10 cycles	5.33
P11 cycles	5.33
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	68.00
Nb uops	68.00
Nb loads	NA
Nb stores	8.00
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	15.06
Bytes prefetched	0.00
Bytes loaded	64.00
Bytes stored	64.00
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	0.00
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	19.17
Vector-efficiency ratio load	25.00
Vector-efficiency ratio store	25.00
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	12.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	12.50

Path /

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

Function	miniqmcreference::TwoBodyJastrowRef >::ratio(qmcplusplus::ParticleSet&, int)
Source file and lines	BsplineFunctor.h:236-241
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/169-816-7419/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/BsplineFunctor.h:236-241.

The related source loop is not unrolled or unrolled with no peel/tail loop.
Warnings:

Ignoring paths for analysis
Too many paths. If you really need to analyze all of the 6561 paths individually, rerun with max-paths=6561
RecMII not computed since number of paths is unknown or > max_paths

Try to simplify control and/or increase the maximum number of paths per function/loop through the 'max-paths-nb' option.

This loop has 6561 execution paths.

The presence of multiple execution paths is typically the main/first bottleneck.
Try to simplify control inside loop: ideally, try to remove all conditional expressions, for example by (if applicable):

hoisting them (moving them outside the loop)
turning them into conditional moves, MIN or MAX

Ex: if (x<0) x=0 => x = (x<0 ? 0 : x) (or MAX(0,x) after defining the corresponding macro)

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 8.50 to 5.33 cycles (1.59x 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 19% of vector register length is used (average across all VPU instructions). By vectorizing your loop, you can lower the cost of an iteration from 8.50 to 1.50 cycles (5.67x 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:
- recompile with ftree-vectorize (included in O3) to enable loop vectorization (including FP reductions).
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 64 bytes. The binary loop is storing 64 bytes.

General properties

nb instructions	68
loop length	272
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

8.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	8.50	8.50	6.75	6.75	6.75	6.75	8.00	8.00	0.00	0.00	5.33	5.33	5.33	0.00	0.00
cycles	8.50	8.50	6.75	6.75	6.75	6.75	8.00	8.00	0.00	0.00	5.33	5.33	5.33	0.00	0.00

Cycles executing div or sqrt instructions

Cycles summary

Front-end	8.50
Overall L1	8.50

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	0%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 8.50 cycles. At this rate:

7% of peak load performance is reached (7.53 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
15% of peak store performance is reached (7.53 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41a4f8

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P10	P11	P12	Latency	Recip. throughput
LDR D8, [X1]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
FCMPE D8, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a514 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x1d4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a514 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x1d4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D8, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D9, [X1, #8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W7, W7, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ADD X1, X1, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D9, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a538 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x1f8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a538 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x1f8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D9, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D10, [X1, #8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W8, W7, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D10, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a558 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x218>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a558 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x218>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D10, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D11, [X1, #16]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W17, W7, #2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D11, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a578 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x238>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W17	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a578 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x238>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D11, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D12, [X1, #24]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W3, W7, #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D12, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a598 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x258>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a598 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x258>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D12, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D13, [X1, #32]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W4, W7, #4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D13, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a5b8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x278>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a5b8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x278>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D13, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D14, [X1, #40]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W6, W7, #5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D14, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a5d8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x298>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W6	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a5d8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x298>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D14, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D15, [X1, #48]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W8, W7, #6	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D15, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a5f8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x2b8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a5f8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x2b8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D15, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ADD W7, W7, #7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ADD X1, X1, #56	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
CMP W10, W7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.NE 41a4f8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x1b8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50

Function	miniqmcreference::TwoBodyJastrowRef >::ratio(qmcplusplus::ParticleSet&, int)
Source file and lines	BsplineFunctor.h:236-241
Module	exec

Ignoring paths for analysis
Too many paths. If you really need to analyze all of the 6561 paths individually, rerun with max-paths=6561
RecMII not computed since number of paths is unknown or > max_paths

hoisting them (moving them outside the loop)
turning them into conditional moves, MIN or MAX

Ex: if (x<0) x=0 => x = (x<0 ? 0 : x) (or MAX(0,x) after defining the corresponding macro)

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 8.50 to 5.33 cycles (1.59x speedup).

Workaround

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

Vectorization

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:
- recompile with ftree-vectorize (included in O3) to enable loop vectorization (including FP reductions).
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 64 bytes. The binary loop is storing 64 bytes.

General properties

nb instructions	68
loop length	272
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

8.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	8.50	8.50	6.75	6.75	6.75	6.75	8.00	8.00	0.00	0.00	5.33	5.33	5.33	0.00	0.00
cycles	8.50	8.50	6.75	6.75	6.75	6.75	8.00	8.00	0.00	0.00	5.33	5.33	5.33	0.00	0.00

Cycles executing div or sqrt instructions

Cycles summary

Front-end	8.50
Overall L1	8.50

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	0%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 8.50 cycles. At this rate:

7% of peak load performance is reached (7.53 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
15% of peak store performance is reached (7.53 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41a4f8

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P10	P11	P12	Latency	Recip. throughput
LDR D8, [X1]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
FCMPE D8, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a514 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x1d4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a514 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x1d4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D8, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D9, [X1, #8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W7, W7, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ADD X1, X1, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D9, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a538 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x1f8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a538 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x1f8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D9, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D10, [X1, #8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W8, W7, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D10, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a558 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x218>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a558 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x218>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D10, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D11, [X1, #16]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W17, W7, #2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D11, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a578 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x238>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W17	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a578 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x238>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D11, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D12, [X1, #24]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W3, W7, #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D12, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a598 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x258>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a598 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x258>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D12, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D13, [X1, #32]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W4, W7, #4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D13, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a5b8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x278>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a5b8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x278>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D13, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D14, [X1, #40]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W6, W7, #5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D14, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a5d8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x298>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W6	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a5d8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x298>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D14, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR D15, [X1, #48]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33
ADD W8, W7, #6	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
FCMPE D15, D0	1	0	0	0	0	0	0	1	0	0	0	0	2	1
B.GE 41a5f8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x2b8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
CMP W2, W8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 41a5f8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x2b8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
STR D15, [X9, X5,SXTW #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	2	0.50
ADD W5, W5, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ADD W7, W7, #7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ADD X1, X1, #56	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
CMP W10, W7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.NE 41a4f8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE5ratioERNS1_11ParticleSetEi+0x1b8>	1	0.50	0.50	0	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