OV - exec - Loop 199

	Loop Id: 199	Module: exec	Source: advec_mom.cpp:181-211 [...]	Coverage: 5.05%

0x41eb40 MOVPRFX	Z25, Z24

0x41eb44 SDIV	Z25.D, P0/M, Z25.D, Z1.D

0x41eb48 DECD	X2, ALL

0x41eb4c ADD	Z28.D, Z2.D, Z25.D

0x41eb50 MSB	Z25.D, P0/M, Z1.D, Z24.D

0x41eb54 ADD	Z24.D, Z24.D, Z0.D

0x41eb58 MOVPRFX	Z26, Z28

0x41eb5c SXTW	Z26.D, P0/M, Z28.D

0x41eb60 ORR	Z29.D, Z28.D, Z28.D

0x41eb64 ORR	Z10.D, Z28.D, Z28.D

0x41eb68 ADD	Z31.D, Z28.D, Z18.D

0x41eb6c MUL	Z27.D, Z4.D, Z26.D

0x41eb70 ADD	Z29.D, Z29.D, #1

0x41eb74 ADD	Z10.D, Z10.D, #2

0x41eb78 ADD	Z25.D, Z3.D, Z25.D

0x41eb7c MOVPRFX	Z30, Z29

0x41eb80 SXTW	Z30.D, P0/M, Z29.D

0x41eb84 ADR	Z27.D, [Z16, Z27.D,LSL #3]    [12]

0x41eb88 ADR	Z27.D, [Z27, Z25.D,SXTW #3]    [15]

0x41eb8c LD1D	{Z27.D}, P0/Z, [V27.D]    [14]

0x41eb90 FCMLT	P1.D, P0/Z, Z27.D, #0

0x41eb94 SEL	Z8.D, P1, Z30.D, Z26.D

0x41eb98 SEL	Z10.D, P1, Z10.D, Z31.D

0x41eb9c SEL	Z30.D, P1, Z26.D, Z30.D

0x41eba0 SEL	Z29.D, P1, Z29.D, Z31.D

0x41eba4 MOVPRFX	Z31, Z27

0x41eba8 FABS	Z31.D, P0/M, Z27.D

0x41ebac MUL	Z26.D, Z7.D, Z26.D

0x41ebb0 MUL	Z9.D, Z5.D, Z8.D

0x41ebb4 SXTW	Z10.D, P0/M, Z10.D

0x41ebb8 MUL	Z8.D, Z6.D, Z8.D

0x41ebbc MUL	Z30.D, Z6.D, Z30.D

0x41ebc0 MUL	Z10.D, Z6.D, Z10.D

0x41ebc4 ADR	Z26.D, [Z23, Z26.D,LSL #3]    [10]

0x41ebc8 ADR	Z9.D, [Z19, Z9.D,LSL #3]    [16]

0x41ebcc ADR	Z8.D, [Z20, Z8.D,LSL #3]    [3]

0x41ebd0 ADR	Z30.D, [Z20, Z30.D,LSL #3]    [3]

0x41ebd4 ADR	Z9.D, [Z9, Z25.D,SXTW #3]    [9]

0x41ebd8 ADR	Z10.D, [Z20, Z10.D,LSL #3]    [3]

0x41ebdc ADR	Z8.D, [Z8, Z25.D,SXTW #3]    [7]

0x41ebe0 ADR	Z30.D, [Z30, Z25.D,SXTW #3]    [5]

0x41ebe4 ADR	Z10.D, [Z10, Z25.D,SXTW #3]    [2]

0x41ebe8 LD1D	{Z9.D}, P0/Z, [V9.D]    [8]

0x41ebec LD1D	{Z8.D}, P0/Z, [V8.D]    [6]

0x41ebf0 LD1D	{Z30.D}, P0/Z, [V30.D]    [4]

0x41ebf4 ADR	Z25.D, [Z26, Z25.D,SXTW #3]    [13]

0x41ebf8 LD1D	{Z10.D}, P0/Z, [V10.D]    [1]

0x41ebfc FDIV	Z31.D, P0/M, Z31.D, Z9.D

0x41ec00 FSUB	Z11.D, Z30.D, Z8.D

0x41ec04 FABD	Z30.D, P0/M, Z30.D, Z8.D

0x41ec08 FSUB	Z9.D, Z8.D, Z10.D

0x41ec0c FCMLE	P2.D, P0/Z, Z11.D, #0

0x41ec10 FMUL	Z9.D, Z9.D, Z11.D

0x41ec14 EOR	P2.B, P0/Z, P2.B, P0.B

0x41ec18 FCMGT	P1.D, P0/Z, Z9.D, #0

0x41ec1c MOVPRFX	Z9, Z8

0x41ec20 FABD	Z9.D, P0/M, Z9.D, Z10.D

0x41ec24 FSUB	Z10.D, Z21.D, Z31.D

0x41ec28 MOVPRFX	Z11, Z31

0x41ec2c FADD	Z11.D, P0/M, Z11.D, #1

0x41ec30 LD1D	{Z28.D}, P1/Z, [X14, Z28.D,SXTW #3]    [17]

0x41ec34 LD1D	{Z29.D}, P1/Z, [X14, Z29.D,SXTW #3]    [17]

0x41ec38 FMUL	Z10.D, Z10.D, Z30.D

0x41ec3c FMUL	Z11.D, Z11.D, Z9.D

0x41ec40 FDIV	Z10.D, P0/M, Z10.D, Z28.D

0x41ec44 FDIVR	Z29.D, P0/M, Z29.D, Z11.D

0x41ec48 FADD	Z29.D, Z10.D, Z29.D

0x41ec4c FMUL	Z28.D, Z28.D, Z29.D

0x41ec50 FDIV	Z28.D, P0/M, Z28.D, Z22.D

0x41ec54 FMINNM	Z28.D, P0/M, Z28.D, Z9.D

0x41ec58 FMINNM	Z28.D, P0/M, Z28.D, Z30.D

0x41ec5c MOVPRFX	Z29, Z28

0x41ec60 FNEG	Z29.D, P0/M, Z28.D

0x41ec64 SEL	Z28.D, P2, Z28.D, Z29.D

0x41ec68 MOVPRFX	Z29, Z31

0x41ec6c FSUBR	Z29.D, P0/M, Z29.D, #1

0x41ec70 SEL	Z28.D, P1, Z28.D, Z17.D

0x41ec74 FMAD	Z28.D, P0/M, Z29.D, Z8.D

0x41ec78 FMUL	Z27.D, Z27.D, Z28.D

0x41ec7c ST1D	{Z27.D}, P0, [V25.D]    [11]

0x41ec80 CBNZ	X2, 41eb40

/home/hbollore/qaas/qaas-runs/174-169-1878/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp: 181 - 211

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

181:     for (int j = (y_min - 1 + 1); j < (y_max + 1 + 2); j++) {

182:       for (int i = (x_min + 1); i < (x_max + 1 + 2); i++)

183:         ({

184:           int upwind, donor, downwind, dif;

185:           double sigma, width, limiter, vdiffuw, vdiffdw, auw, adw, wind, advec_vel_s;

186:           if (node_flux(i, j) < 0.0) {

187:             upwind = j + 2;

188:             donor = j + 1;

189:             downwind = j;

190:             dif = donor;

191:           } else {

192:             upwind = j - 1;

193:             donor = j;

194:             downwind = j + 1;

195:             dif = upwind;

196:           }

197:           sigma = std::fabs(node_flux(i, j)) / (node_mass_pre(i, donor));

198:           width = celldy[j];

199:           vdiffuw = vel1(i, donor) - vel1(i, upwind);

200:           vdiffdw = vel1(i, downwind) - vel1(i, donor);

201:           limiter = 0.0;

202:           if (vdiffuw * vdiffdw > 0.0) {

203:             auw = std::fabs(vdiffuw);

204:             adw = std::fabs(vdiffdw);

205:             wind = 1.0;

206:             if (vdiffdw <= 0.0) wind = -1.0;

207:             limiter =

208:                 wind * std::fmin(std::fmin(width * ((2.0 - sigma) * adw / width + (1.0 + sigma) * auw / celldy[dif]) / 6.0, auw), adw);

209:           }

210:           advec_vel_s = vel1(i, donor) + (1.0 - sigma) * limiter;

211:           mom_flux(i, j) = advec_vel_s * node_flux(i, j);

/home/hbollore/qaas/qaas-runs/174-169-1878/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h: 69 - 69

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

69:   T &operator()(size_t i, size_t j) const { return data[i + j * sizeX]; }

Coverage (%)	Name	Source Location	Module
○100.00	__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	1.09 - 1.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	2.74
Bottlenecks	P8, P10,
Function	advec_mom_kernel(int, int, int, int, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer1D&, clover::Buffer1D&, int, int, int) [clone .omp_outlined.20]
Source	advec_mom.cpp:181-211,context.h:69-69
Source loop unroll info	unrolled by 2
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	2
CQA cycles	17.50 - 47.97
CQA cycles if no scalar integer	17.50 - 47.97
CQA cycles if FP arith vectorized	17.50 - 47.97
CQA cycles if fully vectorized	16.00 - 47.97
Front-end cycles	9.38
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.00
P3 cycles	0.00
P4 cycles	0.00
P5 cycles	0.00
P6 cycles	1.00
P7 cycles	1.00
P8 cycles	17.50
P9 cycles	17.50
P10 cycles	17.50
P11 cycles	17.50
P12 cycles	2.83
P13 cycles	2.50
P14 cycles	2.67
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	14.86 - 47.97
Inter-iter dependencies cycles	2
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	81.00
Nb uops	75.00
Nb loads	NA
Nb stores	1.00
Nb stack references	0.00
FLOP/cycle	1.94 - 0.71
Nb FLOP add-sub	12.00
Nb FLOP mul	10.00
Nb FLOP fma	2.00
Nb FLOP div	8.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	2.67 - 7.31
Bytes prefetched	0.00
Bytes loaded	112.00
Bytes stored	16.00
Stride 0	0.00
Stride 1	0.00
Stride n	0.00
Stride unknown	11.00
Stride indirect	6.00
Vectorization ratio all	73.75
Vectorization ratio load	36.84
Vectorization ratio store	100.00
Vectorization ratio mul	100.00
Vectorization ratio add_sub	48.00
Vectorization ratio fma	100.00
Vectorization ratio div_sqrt	100.00
Vectorization ratio other	72.41
Vector-efficiency ratio all	90.78
Vector-efficiency ratio load	68.42
Vector-efficiency ratio store	100.00
Vector-efficiency ratio mul	100.00
Vector-efficiency ratio add_sub	74.00
Vector-efficiency ratio fma	100.00
Vector-efficiency ratio div_sqrt	100.00
Vector-efficiency ratio other	96.98

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	1.09 - 1.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	2.74
Bottlenecks	P8, P10,
Function	advec_mom_kernel(int, int, int, int, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer1D&, clover::Buffer1D&, int, int, int) [clone .omp_outlined.20]
Source	advec_mom.cpp:181-211,context.h:69-69
Source loop unroll info	unrolled by 2
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	2
CQA cycles	17.50 - 47.97
CQA cycles if no scalar integer	17.50 - 47.97
CQA cycles if FP arith vectorized	17.50 - 47.97
CQA cycles if fully vectorized	16.00 - 47.97
Front-end cycles	9.38
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.00
P3 cycles	0.00
P4 cycles	0.00
P5 cycles	0.00
P6 cycles	1.00
P7 cycles	1.00
P8 cycles	17.50
P9 cycles	17.50
P10 cycles	17.50
P11 cycles	17.50
P12 cycles	2.83
P13 cycles	2.50
P14 cycles	2.67
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	14.86 - 47.97
Inter-iter dependencies cycles	2
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	81.00
Nb uops	75.00
Nb loads	NA
Nb stores	1.00
Nb stack references	0.00
FLOP/cycle	1.94 - 0.71
Nb FLOP add-sub	12.00
Nb FLOP mul	10.00
Nb FLOP fma	2.00
Nb FLOP div	8.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	2.67 - 7.31
Bytes prefetched	0.00
Bytes loaded	112.00
Bytes stored	16.00
Stride 0	0.00
Stride 1	0.00
Stride n	0.00
Stride unknown	11.00
Stride indirect	6.00
Vectorization ratio all	73.75
Vectorization ratio load	36.84
Vectorization ratio store	100.00
Vectorization ratio mul	100.00
Vectorization ratio add_sub	48.00
Vectorization ratio fma	100.00
Vectorization ratio div_sqrt	100.00
Vectorization ratio other	72.41
Vector-efficiency ratio all	90.78
Vector-efficiency ratio load	68.42
Vector-efficiency ratio store	100.00
Vector-efficiency ratio mul	100.00
Vector-efficiency ratio add_sub	74.00
Vector-efficiency ratio fma	100.00
Vector-efficiency ratio div_sqrt	100.00
Vector-efficiency ratio other	96.98

Path /

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

Function	advec_mom_kernel(int, int, int, int, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer1D&, clover::Buffer1D&, int, int, int) [clone .omp_outlined.20]
Source file and lines	advec_mom.cpp:181-211
Module	exec

The loop is defined in:

/home/hbollore/qaas/qaas-runs/174-169-1878/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:181-211
/home/hbollore/qaas/qaas-runs/174-169-1878/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h:69

The related source loop is unrolled by 2 (including vectorization).

gain
potential
hint
expert

Vectorization

Your loop is partially vectorized. 90% of vector register length is used (average across all VPU instructions).

Details

73% of VPU instructions are used in vector version (process two or more data elements in vector registers):

36% of VPU loads are used in vector version.
48% of VPU addition or subtraction instructions are used in vector version.
72% of VPU instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

FMA

Detected 2 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 34 FP arithmetical operations:

14: addition or subtraction (2 inside FMA instructions)
12: multiply (2 inside FMA instructions)
8: divide

The binary loop is loading 112 bytes. The binary loop is storing 16 bytes.

Arithmetic intensity

Arithmetic intensity is 0.27 FP operations per loaded or stored byte.

General properties

nb instructions	81
loop length	324
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

9.38 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	0.00	0.00	0.00	0.00	1.00	1.00	17.50	17.50	17.50	17.50	2.83	2.50	2.67	0.00	0.00
cycles	0.50	0.50	0.00	0.00	0.00	0.00	1.00	1.00	17.50	17.50	17.50	17.50	2.83	2.50	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	14.86-47.97
Longest recurrence chain latency (RecMII)	2.00

Cycles summary

Front-end	9.38
Data deps.	2.00
Overall L1	17.50-47.97

Vectorization ratios

INT

all	61%
load	36%
store	100%
mul	100%
add-sub	31%
fma	100%
other	61%

all	100%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	100%
add-sub	100%
fma	100%
div/sqrt	100%
other	100%

INT+FP

all	73%
load	36%
store	100%
mul	100%
add-sub	48%
fma	100%
div/sqrt	100%
other	72%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 47.97 cycles. At this rate:

2% of peak load performance is reached (2.33 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
0% of peak store performance is reached (0.33 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41eb40

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16	Latency	Recip. throughput	Vectorization
MOVPRFX Z25, Z24	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
SDIV Z25.D, P0/M, Z25.D, Z1.D	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	7-20	0.87-20	vect (100.0%)
DECD X2, ALL	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	scal (50.0%)
ADD Z28.D, Z2.D, Z25.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MSB Z25.D, P0/M, Z1.D, Z24.D	1	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	5	1	vect (100.0%)
ADD Z24.D, Z24.D, Z0.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z26, Z28	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
SXTW Z26.D, P0/M, Z28.D	1	0	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	2	0.50	vect (100.0%)
ORR Z29.D, Z28.D, Z28.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
ORR Z10.D, Z28.D, Z28.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
ADD Z31.D, Z28.D, Z18.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MUL Z27.D, Z4.D, Z26.D																					vect (100.0%)
ADD Z29.D, Z29.D, #1	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
ADD Z10.D, Z10.D, #2	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
ADD Z25.D, Z3.D, Z25.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z30, Z29	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
SXTW Z30.D, P0/M, Z29.D	1	0	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	2	0.50	vect (100.0%)
ADR Z27.D, [Z16, Z27.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z27.D, [Z27, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
LD1D {Z27.D}, P0/Z, [V27.D]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	0	0	9	1	vect (100.0%)
FCMLT P1.D, P0/Z, Z27.D, #0	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	vect (100.0%)
SEL Z8.D, P1, Z30.D, Z26.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
SEL Z10.D, P1, Z10.D, Z31.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
SEL Z30.D, P1, Z26.D, Z30.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
SEL Z29.D, P1, Z29.D, Z31.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z31, Z27	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
FABS Z31.D, P0/M, Z27.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MUL Z26.D, Z7.D, Z26.D																					vect (100.0%)
MUL Z9.D, Z5.D, Z8.D																					vect (100.0%)
SXTW Z10.D, P0/M, Z10.D	1	0	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	2	0.50	vect (100.0%)
MUL Z8.D, Z6.D, Z8.D																					vect (100.0%)
MUL Z30.D, Z6.D, Z30.D																					vect (100.0%)
MUL Z10.D, Z6.D, Z10.D																					vect (100.0%)
ADR Z26.D, [Z23, Z26.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z9.D, [Z19, Z9.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z8.D, [Z20, Z8.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z30.D, [Z20, Z30.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z9.D, [Z9, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z10.D, [Z20, Z10.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z8.D, [Z8, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z30.D, [Z30, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z10.D, [Z10, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
LD1D {Z9.D}, P0/Z, [V9.D]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	0	0	9	1	vect (100.0%)
LD1D {Z8.D}, P0/Z, [V8.D]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	0	0	9	1	vect (100.0%)
LD1D {Z30.D}, P0/Z, [V30.D]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	0	0	9	1	vect (100.0%)
ADR Z25.D, [Z26, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
LD1D {Z10.D}, P0/Z, [V10.D]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	0	0	9	1	vect (100.0%)
FDIV Z31.D, P0/M, Z31.D, Z9.D	1	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	3.50-6.99	vect (100.0%)
FSUB Z11.D, Z30.D, Z8.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FABD Z30.D, P0/M, Z30.D, Z8.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FSUB Z9.D, Z8.D, Z10.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FCMLE P2.D, P0/Z, Z11.D, #0	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	vect (100.0%)
FMUL Z9.D, Z9.D, Z11.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	vect (100.0%)
EOR P2.B, P0/Z, P2.B, P0.B	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	1	1	vect (12.5%)
FCMGT P1.D, P0/Z, Z9.D, #0	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	vect (100.0%)
MOVPRFX Z9, Z8	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
FABD Z9.D, P0/M, Z9.D, Z10.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FSUB Z10.D, Z21.D, Z31.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z11, Z31	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
FADD Z11.D, P0/M, Z11.D, #1	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
LD1D {Z28.D}, P1/Z, [X14, Z28.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (100.0%)
LD1D {Z29.D}, P1/Z, [X14, Z29.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (100.0%)
FMUL Z10.D, Z10.D, Z30.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	vect (100.0%)
FMUL Z11.D, Z11.D, Z9.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	vect (100.0%)
FDIV Z10.D, P0/M, Z10.D, Z28.D	1	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	3.50-6.99	vect (100.0%)
FDIVR Z29.D, P0/M, Z29.D, Z11.D	1	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	3.50-6.99	vect (100.0%)
FADD Z29.D, Z10.D, Z29.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FMUL Z28.D, Z28.D, Z29.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	vect (100.0%)
FDIV Z28.D, P0/M, Z28.D, Z22.D	1	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	3.50-6.99	vect (100.0%)
FMINNM Z28.D, P0/M, Z28.D, Z9.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FMINNM Z28.D, P0/M, Z28.D, Z30.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z29, Z28	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
FNEG Z29.D, P0/M, Z28.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
SEL Z28.D, P2, Z28.D, Z29.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z29, Z31	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
FSUBR Z29.D, P0/M, Z29.D, #1	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
SEL Z28.D, P1, Z28.D, Z17.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FMAD Z28.D, P0/M, Z29.D, Z8.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	4	0.25	vect (100.0%)
FMUL Z27.D, Z27.D, Z28.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	vect (100.0%)
ST1D {Z27.D}, P0, [V25.D]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	1	vect (100.0%)
CBNZ X2, 41eb40 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.20+0x190>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	advec_mom_kernel(int, int, int, int, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer1D&, clover::Buffer1D&, int, int, int) [clone .omp_outlined.20]
Source file and lines	advec_mom.cpp:181-211
Module	exec

The loop is defined in:

/home/hbollore/qaas/qaas-runs/174-169-1878/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:181-211
/home/hbollore/qaas/qaas-runs/174-169-1878/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h:69

The related source loop is unrolled by 2 (including vectorization).

gain
potential
hint
expert

Vectorization

Your loop is partially vectorized. 90% of vector register length is used (average across all VPU instructions).

Details

73% of VPU instructions are used in vector version (process two or more data elements in vector registers):

36% of VPU loads are used in vector version.
48% of VPU addition or subtraction instructions are used in vector version.
72% of VPU instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

FMA

Detected 2 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 34 FP arithmetical operations:

14: addition or subtraction (2 inside FMA instructions)
12: multiply (2 inside FMA instructions)
8: divide

The binary loop is loading 112 bytes. The binary loop is storing 16 bytes.

Arithmetic intensity

Arithmetic intensity is 0.27 FP operations per loaded or stored byte.

General properties

nb instructions	81
loop length	324
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

9.38 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	0.00	0.00	0.00	0.00	1.00	1.00	17.50	17.50	17.50	17.50	2.83	2.50	2.67	0.00	0.00
cycles	0.50	0.50	0.00	0.00	0.00	0.00	1.00	1.00	17.50	17.50	17.50	17.50	2.83	2.50	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	14.86-47.97
Longest recurrence chain latency (RecMII)	2.00

Cycles summary

Front-end	9.38
Data deps.	2.00
Overall L1	17.50-47.97

Vectorization ratios

INT

all	61%
load	36%
store	100%
mul	100%
add-sub	31%
fma	100%
other	61%

all	100%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	100%
add-sub	100%
fma	100%
div/sqrt	100%
other	100%

INT+FP

all	73%
load	36%
store	100%
mul	100%
add-sub	48%
fma	100%
div/sqrt	100%
other	72%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 47.97 cycles. At this rate:

2% of peak load performance is reached (2.33 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
0% of peak store performance is reached (0.33 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41eb40

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16	Latency	Recip. throughput	Vectorization
MOVPRFX Z25, Z24	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
SDIV Z25.D, P0/M, Z25.D, Z1.D	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	7-20	0.87-20	vect (100.0%)
DECD X2, ALL	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	scal (50.0%)
ADD Z28.D, Z2.D, Z25.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MSB Z25.D, P0/M, Z1.D, Z24.D	1	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	5	1	vect (100.0%)
ADD Z24.D, Z24.D, Z0.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z26, Z28	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
SXTW Z26.D, P0/M, Z28.D	1	0	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	2	0.50	vect (100.0%)
ORR Z29.D, Z28.D, Z28.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
ORR Z10.D, Z28.D, Z28.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
ADD Z31.D, Z28.D, Z18.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MUL Z27.D, Z4.D, Z26.D																					vect (100.0%)
ADD Z29.D, Z29.D, #1	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
ADD Z10.D, Z10.D, #2	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
ADD Z25.D, Z3.D, Z25.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z30, Z29	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
SXTW Z30.D, P0/M, Z29.D	1	0	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	2	0.50	vect (100.0%)
ADR Z27.D, [Z16, Z27.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z27.D, [Z27, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
LD1D {Z27.D}, P0/Z, [V27.D]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	0	0	9	1	vect (100.0%)
FCMLT P1.D, P0/Z, Z27.D, #0	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	vect (100.0%)
SEL Z8.D, P1, Z30.D, Z26.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
SEL Z10.D, P1, Z10.D, Z31.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
SEL Z30.D, P1, Z26.D, Z30.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
SEL Z29.D, P1, Z29.D, Z31.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z31, Z27	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
FABS Z31.D, P0/M, Z27.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MUL Z26.D, Z7.D, Z26.D																					vect (100.0%)
MUL Z9.D, Z5.D, Z8.D																					vect (100.0%)
SXTW Z10.D, P0/M, Z10.D	1	0	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	2	0.50	vect (100.0%)
MUL Z8.D, Z6.D, Z8.D																					vect (100.0%)
MUL Z30.D, Z6.D, Z30.D																					vect (100.0%)
MUL Z10.D, Z6.D, Z10.D																					vect (100.0%)
ADR Z26.D, [Z23, Z26.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z9.D, [Z19, Z9.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z8.D, [Z20, Z8.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z30.D, [Z20, Z30.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z9.D, [Z9, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z10.D, [Z20, Z10.D,LSL #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z8.D, [Z8, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z30.D, [Z30, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
ADR Z10.D, [Z10, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
LD1D {Z9.D}, P0/Z, [V9.D]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	0	0	9	1	vect (100.0%)
LD1D {Z8.D}, P0/Z, [V8.D]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	0	0	9	1	vect (100.0%)
LD1D {Z30.D}, P0/Z, [V30.D]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	0	0	9	1	vect (100.0%)
ADR Z25.D, [Z26, Z25.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (50.0%)
LD1D {Z10.D}, P0/Z, [V10.D]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	0	0	9	1	vect (100.0%)
FDIV Z31.D, P0/M, Z31.D, Z9.D	1	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	3.50-6.99	vect (100.0%)
FSUB Z11.D, Z30.D, Z8.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FABD Z30.D, P0/M, Z30.D, Z8.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FSUB Z9.D, Z8.D, Z10.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FCMLE P2.D, P0/Z, Z11.D, #0	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	vect (100.0%)
FMUL Z9.D, Z9.D, Z11.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	vect (100.0%)
EOR P2.B, P0/Z, P2.B, P0.B	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	1	1	vect (12.5%)
FCMGT P1.D, P0/Z, Z9.D, #0	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	vect (100.0%)
MOVPRFX Z9, Z8	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
FABD Z9.D, P0/M, Z9.D, Z10.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FSUB Z10.D, Z21.D, Z31.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z11, Z31	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
FADD Z11.D, P0/M, Z11.D, #1	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
LD1D {Z28.D}, P1/Z, [X14, Z28.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (100.0%)
LD1D {Z29.D}, P1/Z, [X14, Z29.D,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (100.0%)
FMUL Z10.D, Z10.D, Z30.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	vect (100.0%)
FMUL Z11.D, Z11.D, Z9.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	vect (100.0%)
FDIV Z10.D, P0/M, Z10.D, Z28.D	1	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	3.50-6.99	vect (100.0%)
FDIVR Z29.D, P0/M, Z29.D, Z11.D	1	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	3.50-6.99	vect (100.0%)
FADD Z29.D, Z10.D, Z29.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FMUL Z28.D, Z28.D, Z29.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	vect (100.0%)
FDIV Z28.D, P0/M, Z28.D, Z22.D	1	0	0	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	3.50-6.99	vect (100.0%)
FMINNM Z28.D, P0/M, Z28.D, Z9.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FMINNM Z28.D, P0/M, Z28.D, Z30.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z29, Z28	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
FNEG Z29.D, P0/M, Z28.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
SEL Z28.D, P2, Z28.D, Z29.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
MOVPRFX Z29, Z31	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (100.0%)
FSUBR Z29.D, P0/M, Z29.D, #1	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
SEL Z28.D, P1, Z28.D, Z17.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	vect (100.0%)
FMAD Z28.D, P0/M, Z29.D, Z8.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	4	0.25	vect (100.0%)
FMUL Z27.D, Z27.D, Z28.D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	vect (100.0%)
ST1D {Z27.D}, P0, [V25.D]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	1	vect (100.0%)
CBNZ X2, 41eb40 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.20+0x190>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Report Configuration

Vectorization

Details

FMA

Workaround

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

Arithmetic intensity

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code

Vectorization

Details

FMA

Workaround

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

Arithmetic intensity

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code