OV - exec - Loop 95

	Loop Id: 95	Module: exec	Source: ljForce.c:191-216 [...]	Coverage: 89.06%

0x418100 LDP	D23, D24, [X0]    [5]

0x418104 LDP	D25, D3, [X1]    [8]

0x418108 LDR	D2, [X0, #16]    [5]

0x41810c LDR	D27, [X1, #16]    [8]

0x418110 FSUB	D30, D25, S23

0x418114 FSUB	D31, D3, S24

0x418118 FSUB	D1, D27, S2

0x41811c FMADD	D26, D30, D30, D6

0x418120 FMADD	D4, D31, D31, D26

0x418124 FMADD	D0, D1, D1, D4

0x418128 FCMPE	D5, D0

0x41812c B.GE	418248

0x418130 LDP	D26, D3, [X1]    [8]

0x418134 ADD	X2, X0, #24

0x418138 LDR	D25, [X0, #24]    [5]

0x41813c LDP	D0, D28, [X2, #8]    [10]

0x418140 LDR	D27, [X1, #16]    [8]

0x418144 FSUB	D4, D26, S25

0x418148 FSUB	D29, D3, S0

0x41814c FSUB	D30, D27, S28

0x418150 FMADD	D23, D4, D4, D6

0x418154 FMADD	D24, D29, D29, D23

0x418158 FMADD	D2, D30, D30, D24

0x41815c FCMPE	D5, D2

0x418160 B.GE	4181d8

0x418164 ADD	X0, X2, #24

0x418168 CMP	X0, X3

0x41816c B.NE	418100

0x4181d8 FCMPE	D2, #0

0x4181dc B.GT	4181e4

0x4181e0 B	418164

0x4181e4 FDIV	D31, D17, D2

0x4181e8 LDP	X7, X26, [X6, #40]    [1]

0x4181ec FMUL	D1, D8, D22

0x4181f0 LDR	D2, [X26, X4]    [7]

0x4181f4 ADD	X0, X7, X5

0x4181f8 FMUL	D25, D31, D31

0x4181fc FMUL	D26, D25, D31

0x418200 FMUL	D3, D26, D19

0x418204 FMADD	D27, D3, D21, D20

0x418208 FSUB	D0, D3, S17

0x41820c FMUL	D28, D1, D3

0x418210 FNMSUB	D23, D0, D3, D18

0x418214 FMUL	D24, D28, D31

0x418218 FMADD	D1, D23, D16, D2

0x41821c FMADD	D7, D23, D16, D7

0x418220 FMUL	D31, D27, D24

0x418224 STR	D1, [X26, X4]    [7]

0x418228 LDR	D25, [X7, X5]    [3]

0x41822c FMSUB	D4, D31, D4, D25

0x418230 STR	D4, [X7, X5]    [3]

0x418234 LDP	D26, D27, [X0, #8]    [4]

0x418238 FMSUB	D29, D31, D29, D26

0x41823c FMSUB	D30, D31, D30, D27

0x418240 STP	D29, D30, [X0, #8]    [4]

0x418244 B	418164

0x418248 FCMPE	D0, #0

0x41824c B.GT	418254

0x418250 B	418130

0x418254 FDIV	D28, D17, D0

0x418258 FMUL	D29, D8, D22

0x41825c LDP	X7, X26, [X6, #40]    [1]

0x418260 ADD	X2, X7, X5

0x418264 FMUL	D23, D28, D28

0x418268 FMUL	D24, D23, D28

0x41826c FMUL	D2, D24, D19

0x418270 FMADD	D3, D2, D21, D20

0x418274 FMUL	D26, D29, D2

0x418278 FSUB	D25, D2, S17

0x41827c FMUL	D27, D26, D28

0x418280 LDR	D28, [X26, X4]    [6]

0x418284 FNMSUB	D4, D25, D2, D18

0x418288 FMADD	D7, D4, D16, D7

0x41828c FMADD	D29, D4, D16, D28

0x418290 FMUL	D0, D3, D27

0x418294 STR	D29, [X26, X4]    [6]

0x418298 LDR	D23, [X7, X5]    [2]

0x41829c FMSUB	D30, D0, D30, D23

0x4182a0 STR	D30, [X7, X5]    [2]

0x4182a4 LDP	D24, D2, [X2, #8]    [9]

0x4182a8 FMSUB	D31, D0, D31, D24

0x4182ac FMSUB	D1, D0, D1, D2

0x4182b0 STP	D31, D1, [X2, #8]    [9]

0x4182b4 B	418130

/home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/build/CoMD/CoMD/src-openmp/ljForce.c: 191 - 216

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

191:             for (int jOff=jBox*MAXATOMS; jOff<(jBox*MAXATOMS+nJBox); jOff++)

[...]

197:                   dr[m] = s->atoms->r[iOff][m]-s->atoms->r[jOff][m];

198:                   r2+=dr[m]*dr[m];

199:                }

200:

201:                if ( r2 <= rCut2 && r2 > 0.0)

202:                {

203:

204:                   // Important note:

205:                   // from this point on r actually refers to 1.0/r

206:                   r2 = 1.0/r2;

207:                   real_t r6 = s6 * (r2*r2*r2);

208:                   real_t eLocal = r6 * (r6 - 1.0) - eShift;

209:                   s->atoms->U[iOff] += 0.5*eLocal;

210:                   ePot += 0.5*eLocal;

211:

212:                   // different formulation to avoid sqrt computation

213:                   real_t fr = - 4.0*epsilon*r6*r2*(12.0*r6 - 6.0);

214:                   for (int m=0; m<3; m++)

215:                   {

216:                      s->atoms->f[iOff][m] -= dr[m]*fr;

Coverage (%)	Name	Source Location	Module
►99.16+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

Coverage (%)	Name	Source Location	Module
►49.95+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►49.63+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

Coverage (%)	Name	Source Location	Module
►75.00+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►24.79+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

Coverage (%)	Name	Source Location	Module
►87.37+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►12.52+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

Coverage (%)	Name	Source Location	Module
►93.68+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►6.27+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

Coverage (%)	Name	Source Location	Module
►95.79+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►4.18+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

Coverage (%)	Name	Source Location	Module
►97.04+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►2.93+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

Coverage (%)	Name	Source Location	Module
►97.66+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►2.32+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

Coverage (%)	Name	Source Location	Module
►98.07+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►1.92+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

Coverage (%)	Name	Source Location	Module
►98.38+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►1.61+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

Coverage (%)	Name	Source Location	Module
►98.57+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►1.42+	GOMP_parallel		libgomp.so.1.0.0
○	ljForce	ljForce.c:172	exec
○	timestep	timestep.c:51	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	CoMD.c:266	exec

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	2.02
CQA speedup if FP arith vectorized	1.98
CQA speedup if fully vectorized	3.86
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.29
Bottlenecks	P6, P7, P8, P9,
Function	ljForce._omp_fn.1
Source	ljForce.c:191-191,ljForce.c:197-216
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	13.50
CQA cycles if no scalar integer	6.67
CQA cycles if FP arith vectorized	6.83
CQA cycles if fully vectorized	3.50
Front-end cycles	10.50
P0 cycles	4.50
P1 cycles	4.50
P2 cycles	1.25
P3 cycles	1.25
P4 cycles	1.25
P5 cycles	1.25
P6 cycles	13.50
P7 cycles	13.50
P8 cycles	13.50
P9 cycles	13.50
P10 cycles	7.33
P11 cycles	7.33
P12 cycles	7.33
P13 cycles	0.00
P14 cycles	0.00
DIV/SQRT cycles	3.50 - 6.99
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	84.00
Nb uops	84.00
Nb loads	NA
Nb stores	6.00
Nb stack references	0.00
FLOP/cycle	4.74
Nb FLOP add-sub	8.00
Nb FLOP mul	14.00
Nb FLOP fma	20.00
Nb FLOP div	2.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	0.00
Bytes prefetched	0.00
Bytes loaded	0.00
Bytes stored	0.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	11.59
Vectorization ratio load	42.86
Vectorization ratio store	33.33
Vectorization ratio mul	0.00
Vectorization ratio add_sub	0.00
Vectorization ratio fma	0.00
Vectorization ratio div_sqrt	0.00
Vectorization ratio other	0.00
Vector-efficiency ratio all	27.90
Vector-efficiency ratio load	35.71
Vector-efficiency ratio store	33.33
Vector-efficiency ratio mul	25.00
Vector-efficiency ratio add_sub	25.00
Vector-efficiency ratio fma	25.00
Vector-efficiency ratio div_sqrt	25.00
Vector-efficiency ratio other	25.00

Metric	Value
CQA speedup if no scalar integer	2.02
CQA speedup if FP arith vectorized	1.98
CQA speedup if fully vectorized	3.86
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.29
Bottlenecks	P6, P7, P8, P9,
Function	ljForce._omp_fn.1
Source	ljForce.c:191-191,ljForce.c:197-216
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	13.50
CQA cycles if no scalar integer	6.67
CQA cycles if FP arith vectorized	6.83
CQA cycles if fully vectorized	3.50
Front-end cycles	10.50
P0 cycles	4.50
P1 cycles	4.50
P2 cycles	1.25
P3 cycles	1.25
P4 cycles	1.25
P5 cycles	1.25
P6 cycles	13.50
P7 cycles	13.50
P8 cycles	13.50
P9 cycles	13.50
P10 cycles	7.33
P11 cycles	7.33
P12 cycles	7.33
P13 cycles	0.00
P14 cycles	0.00
DIV/SQRT cycles	3.50 - 6.99
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	84.00
Nb uops	84.00
Nb loads	NA
Nb stores	6.00
Nb stack references	0.00
FLOP/cycle	4.74
Nb FLOP add-sub	8.00
Nb FLOP mul	14.00
Nb FLOP fma	20.00
Nb FLOP div	2.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	0.00
Bytes prefetched	0.00
Bytes loaded	0.00
Bytes stored	0.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	11.59
Vectorization ratio load	42.86
Vectorization ratio store	33.33
Vectorization ratio mul	0.00
Vectorization ratio add_sub	0.00
Vectorization ratio fma	0.00
Vectorization ratio div_sqrt	0.00
Vectorization ratio other	0.00
Vector-efficiency ratio all	27.90
Vector-efficiency ratio load	35.71
Vector-efficiency ratio store	33.33
Vector-efficiency ratio mul	25.00
Vector-efficiency ratio add_sub	25.00
Vector-efficiency ratio fma	25.00
Vector-efficiency ratio div_sqrt	25.00
Vector-efficiency ratio other	25.00

Path /

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

Function	ljForce._omp_fn.1
Source file and lines	ljForce.c:191-216
Module	exec

The loop is defined in /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/build/CoMD/CoMD/src-openmp/ljForce.c:191,197-216.

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

Ignoring paths for analysis
Too many paths. Rerun with max-paths=9
RecMII not computed since number of paths is unknown or > max_paths
Streams not analyzed 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 9 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 13.50 to 6.67 cycles (2.03x 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 27% of vector register length is used (average across all VPU instructions). By fully vectorizing your loop, you can lower the cost of an iteration from 13.50 to 3.50 cycles (3.86x speedup).

Details

11% of VPU instructions are used in vector version (process two or more data elements in vector registers):

42% of VPU loads are used in vector version.
33% of VPU stores are used in vector version.
0% of VPU addition or subtraction instructions are used in vector version.
0% of VPU multiply instructions are used in vector version.
0% of VPU fused multiply-add instructions are used in vector version.
0% of VPU divide and square root instructions are used in vector version.
0% of VPU instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

Since your execution units are vector units, only a fully vectorized loop can use their full power.

Workaround

Try another compiler or update/tune your current one:
- recompile with fassociative-math (included in Ofast or ffast-math) to extend loop vectorization to 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)

FMA

Detected 20 FMA (fused multiply-add) operations. Presence of both ADD/SUB and MUL operations.

Workaround

Try to change order in which elements are evaluated (using parentheses) in arithmetic expressions containing both ADD/SUB and MUL operations to enable your compiler to generate FMA instructions wherever possible. For instance a + b*c is a valid FMA (MUL then ADD). However (a+b)* c cannot be translated into an FMA (ADD then MUL).

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

The binary loop is composed of 64 FP arithmetical operations:

28: addition or subtraction (20 inside FMA instructions)
34: multiply (20 inside FMA instructions)
2: divide

The binary loop does not load or store any data.

General properties

nb instructions	84
nb uops	84
loop length	336
used w registers	0
used x registers	9
used b registers	0
used h registers	0
used s registers	7
used d registers	25
used q registers	0
used v registers	0
used z registers	0
nb stack references	0
ADD-SUB / MUL ratio	0.57

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	10.50 cycles
front end	10.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	4.50	4.50	1.25	1.25	1.25	1.25	13.50	13.50	13.50	13.50	7.33	7.33	7.33	0.00	0.00
cycles	4.50	4.50	1.25	1.25	1.25	1.25	13.50	13.50	13.50	13.50	7.33	7.33	7.33	0.00	0.00

Execution ports to units layout:

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

Cycles executing div or sqrt instructions

3.50-6.99

Cycles summary

Front-end	10.50
Dispatch	13.50
DIV/SQRT	3.50-6.99
Overall L1	13.50

Vectorization ratios

INT

all	38%
load	42%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	0%

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

INT+FP

all	11%
load	42%
store	33%
mul	0%
add-sub	0%
fma	0%
div/sqrt	0%
other	0%

Vector efficiency ratios

INT

all	34%
load	35%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	25%

all	25%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	25%
add-sub	25%
fma	25%
div/sqrt	25%
other	25%

INT+FP

all	27%
load	35%
store	33%
mul	25%
add-sub	25%
fma	25%
div/sqrt	25%
other	25%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 13.50 cycles.

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the loop is: 418100

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	Latency	Recip. throughput	Vectorization
LDP D23, D24, [X0]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
LDP D25, D3, [X1]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
LDR D2, [X0, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
LDR D27, [X1, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
FSUB D30, D25, S23	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FSUB D31, D3, S24	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FSUB D1, D27, S2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FMADD D26, D30, D30, D6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D4, D31, D31, D26	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D0, D1, D1, D4	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FCMPE D5, D0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	2	1	scal (25.0%)
B.GE 418248 <ljForce._omp_fn.1+0x2c0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDP D26, D3, [X1]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
ADD X2, X0, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
LDR D25, [X0, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
LDP D0, D28, [X2, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
LDR D27, [X1, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
FSUB D4, D26, S25	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FSUB D29, D3, S0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FSUB D30, D27, S28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FMADD D23, D4, D4, D6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D24, D29, D29, D23	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D2, D30, D30, D24	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FCMPE D5, D2	1	0	0	0	0	0	0	1	0	0	0	0	0	0	2	1	scal (25.0%)
B.GE 4181d8 <ljForce._omp_fn.1+0x250>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X0, X2, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
CMP X0, X3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.NE 418100 <ljForce._omp_fn.1+0x178>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
FCMPE D2, #0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	2	1	scal (25.0%)
B.GT 4181e4 <ljForce._omp_fn.1+0x25c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
B 418164 <ljForce._omp_fn.1+0x1dc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
FDIV D31, D17, D2	1	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	7-15	1.75-3.50	scal (25.0%)
LDP X7, X26, [X6, #40]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	1	N/A
FMUL D1, D8, D22	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
LDR D2, [X26, X4]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
ADD X0, X7, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
FMUL D25, D31, D31	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMUL D26, D25, D31	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMUL D3, D26, D19	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMADD D27, D3, D21, D20	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FSUB D0, D3, S17	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FMUL D28, D1, D3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FNMSUB D23, D0, D3, D18	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMUL D24, D28, D31	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMADD D1, D23, D16, D2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D7, D23, D16, D7	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMUL D31, D27, D24	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
STR D1, [X26, X4]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (25.0%)
LDR D25, [X7, X5]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
FMSUB D4, D31, D4, D25	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
STR D4, [X7, X5]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (25.0%)
LDP D26, D27, [X0, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
FMSUB D29, D31, D29, D26	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMSUB D30, D31, D30, D27	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
STP D29, D30, [X0, #8]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	vect (50.0%)
B 418164 <ljForce._omp_fn.1+0x1dc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
FCMPE D0, #0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	2	1	scal (25.0%)
B.GT 418254 <ljForce._omp_fn.1+0x2cc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
B 418130 <ljForce._omp_fn.1+0x1a8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
FDIV D28, D17, D0	1	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	7-15	1.75-3.50	scal (25.0%)
FMUL D29, D8, D22	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
LDP X7, X26, [X6, #40]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	1	N/A
ADD X2, X7, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
FMUL D23, D28, D28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMUL D24, D23, D28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMUL D2, D24, D19	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMADD D3, D2, D21, D20	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMUL D26, D29, D2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FSUB D25, D2, S17	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FMUL D27, D26, D28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
LDR D28, [X26, X4]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
FNMSUB D4, D25, D2, D18	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D7, D4, D16, D7	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D29, D4, D16, D28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMUL D0, D3, D27	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
STR D29, [X26, X4]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (25.0%)
LDR D23, [X7, X5]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
FMSUB D30, D0, D30, D23	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
STR D30, [X7, X5]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (25.0%)
LDP D24, D2, [X2, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
FMSUB D31, D0, D31, D24	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMSUB D1, D0, D1, D2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
STP D31, D1, [X2, #8]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	vect (50.0%)
B 418130 <ljForce._omp_fn.1+0x1a8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	ljForce._omp_fn.1
Source file and lines	ljForce.c:191-216
Module	exec

Ignoring paths for analysis
Too many paths. Rerun with max-paths=9
RecMII not computed since number of paths is unknown or > max_paths
Streams not analyzed 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 13.50 to 6.67 cycles (2.03x speedup).

Workaround

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

Vectorization

Details

11% of VPU instructions are used in vector version (process two or more data elements in vector registers):

42% of VPU loads are used in vector version.
33% of VPU stores are used in vector version.
0% of VPU addition or subtraction instructions are used in vector version.
0% of VPU multiply instructions are used in vector version.
0% of VPU fused multiply-add instructions are used in vector version.
0% of VPU divide and square root instructions are used in vector version.
0% of VPU instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

Since your execution units are vector units, only a fully vectorized loop can use their full power.

Workaround

Try another compiler or update/tune your current one:
- recompile with fassociative-math (included in Ofast or ffast-math) to extend loop vectorization to 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)

FMA

Detected 20 FMA (fused multiply-add) operations. Presence of both ADD/SUB and MUL operations.

Workaround

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

The binary loop is composed of 64 FP arithmetical operations:

28: addition or subtraction (20 inside FMA instructions)
34: multiply (20 inside FMA instructions)
2: divide

The binary loop does not load or store any data.

General properties

nb instructions	84
nb uops	84
loop length	336
used w registers	0
used x registers	9
used b registers	0
used h registers	0
used s registers	7
used d registers	25
used q registers	0
used v registers	0
used z registers	0
nb stack references	0
ADD-SUB / MUL ratio	0.57

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	10.50 cycles
front end	10.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	4.50	4.50	1.25	1.25	1.25	1.25	13.50	13.50	13.50	13.50	7.33	7.33	7.33	0.00	0.00
cycles	4.50	4.50	1.25	1.25	1.25	1.25	13.50	13.50	13.50	13.50	7.33	7.33	7.33	0.00	0.00

Execution ports to units layout:

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

Cycles executing div or sqrt instructions

3.50-6.99

Cycles summary

Front-end	10.50
Dispatch	13.50
DIV/SQRT	3.50-6.99
Overall L1	13.50

Vectorization ratios

INT

all	38%
load	42%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	0%

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

INT+FP

all	11%
load	42%
store	33%
mul	0%
add-sub	0%
fma	0%
div/sqrt	0%
other	0%

Vector efficiency ratios

INT

all	34%
load	35%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	25%

all	25%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	25%
add-sub	25%
fma	25%
div/sqrt	25%
other	25%

INT+FP

all	27%
load	35%
store	33%
mul	25%
add-sub	25%
fma	25%
div/sqrt	25%
other	25%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 13.50 cycles.

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the loop is: 418100

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	Latency	Recip. throughput	Vectorization
LDP D23, D24, [X0]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
LDP D25, D3, [X1]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
LDR D2, [X0, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
LDR D27, [X1, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
FSUB D30, D25, S23	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FSUB D31, D3, S24	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FSUB D1, D27, S2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FMADD D26, D30, D30, D6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D4, D31, D31, D26	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D0, D1, D1, D4	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FCMPE D5, D0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	2	1	scal (25.0%)
B.GE 418248 <ljForce._omp_fn.1+0x2c0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDP D26, D3, [X1]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
ADD X2, X0, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
LDR D25, [X0, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
LDP D0, D28, [X2, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
LDR D27, [X1, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
FSUB D4, D26, S25	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FSUB D29, D3, S0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FSUB D30, D27, S28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FMADD D23, D4, D4, D6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D24, D29, D29, D23	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D2, D30, D30, D24	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FCMPE D5, D2	1	0	0	0	0	0	0	1	0	0	0	0	0	0	2	1	scal (25.0%)
B.GE 4181d8 <ljForce._omp_fn.1+0x250>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X0, X2, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
CMP X0, X3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.NE 418100 <ljForce._omp_fn.1+0x178>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
FCMPE D2, #0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	2	1	scal (25.0%)
B.GT 4181e4 <ljForce._omp_fn.1+0x25c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
B 418164 <ljForce._omp_fn.1+0x1dc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
FDIV D31, D17, D2	1	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	7-15	1.75-3.50	scal (25.0%)
LDP X7, X26, [X6, #40]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	1	N/A
FMUL D1, D8, D22	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
LDR D2, [X26, X4]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
ADD X0, X7, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
FMUL D25, D31, D31	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMUL D26, D25, D31	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMUL D3, D26, D19	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMADD D27, D3, D21, D20	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FSUB D0, D3, S17	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FMUL D28, D1, D3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FNMSUB D23, D0, D3, D18	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMUL D24, D28, D31	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMADD D1, D23, D16, D2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D7, D23, D16, D7	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMUL D31, D27, D24	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
STR D1, [X26, X4]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (25.0%)
LDR D25, [X7, X5]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
FMSUB D4, D31, D4, D25	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
STR D4, [X7, X5]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (25.0%)
LDP D26, D27, [X0, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
FMSUB D29, D31, D29, D26	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMSUB D30, D31, D30, D27	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
STP D29, D30, [X0, #8]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	vect (50.0%)
B 418164 <ljForce._omp_fn.1+0x1dc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
FCMPE D0, #0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	2	1	scal (25.0%)
B.GT 418254 <ljForce._omp_fn.1+0x2cc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
B 418130 <ljForce._omp_fn.1+0x1a8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
FDIV D28, D17, D0	1	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	7-15	1.75-3.50	scal (25.0%)
FMUL D29, D8, D22	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
LDP X7, X26, [X6, #40]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	4	1	N/A
ADD X2, X7, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
FMUL D23, D28, D28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMUL D24, D23, D28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMUL D2, D24, D19	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FMADD D3, D2, D21, D20	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMUL D26, D29, D2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
FSUB D25, D2, S17	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (25.0%)
FMUL D27, D26, D28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
LDR D28, [X26, X4]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
FNMSUB D4, D25, D2, D18	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D7, D4, D16, D7	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMADD D29, D4, D16, D28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMUL D0, D3, D27	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (25.0%)
STR D29, [X26, X4]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (25.0%)
LDR D23, [X7, X5]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (25.0%)
FMSUB D30, D0, D30, D23	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
STR D30, [X7, X5]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (25.0%)
LDP D24, D2, [X2, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (50.0%)
FMSUB D31, D0, D31, D24	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
FMSUB D1, D0, D1, D2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (25.0%)
STP D31, D1, [X2, #8]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	vect (50.0%)
B 418130 <ljForce._omp_fn.1+0x1a8>	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: NARun Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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: 1Run Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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: 1Run Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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: 1Run Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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: 1Run Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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: 1Run Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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: 1Run Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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: 1Run Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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: 1Run Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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: 1Run Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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: 1Run Command: <executable> -x 100 -y 100 -z 100MPI Command: mpirun -n <number_processes> --bind-to core --map-by package:PE=64 --rank-by fill --report-bindings Dataset: Run Directory: /home/eoseret/qaas/qaas_runs/178-176-0594/intel/CoMD/run/oneview_runs/multicore/gcc/oneview_run_1781767694OMP_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

(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 (%)
1	0	0.99	0.87	0.97	2.3	0.94	5.6	0.88	10.34	0.85	13.38	0.85	12.67	0.82	15.63	0.79	17.54	0.77	19.58	0.74	21.84

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	580.75012207031	89.06209564209
1x2	2	0.99	1.98	2	291.42501831055	89.137756347656
1x4	4	0.97	3.9	4	145.99000549316	89.134185791016
1x8	8	0.94	7.5	8	73.994987487793	88.817207336426
1x16	16	0.88	14.12	16	37.334995269775	88.070991516113
1x24	24	0.85	20.33	24	26.269996643066	87.48233795166
1x32	32	0.85	27.32	32	20.109998703003	86.598968505859
1x40	40	0.82	32.72	40	16.76000213623	85.840217590332
1x48	48	0.79	38.12	48	13.859996795654	85.187431335449
1x56	56	0.77	43.01	56	12.090000152588	84.397216796875
1x64	64	0.74	47.31	64	11.18000125885	83.735420227051

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

FMA

Workaround

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

Code clean check

Workaround

Vectorization

Details

Workaround

FMA

Workaround

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

▶Scalability Runs Description

▼Columns Filter

Speed-Up and Efficiency

▶Detailed Speed-Up and Efficiency