OV - exec - Loop 178

	Loop Id: 178	Module: exec	Source: advec_mom.cpp:54-57 [...]	Coverage: 2.54%

0x41d160 MOVPRFX	Z25, Z24

0x41d164 SDIV	Z25.D, P0/M, Z25.D, Z1.D

0x41d168 DECD	X4, ALL

0x41d16c ADD	Z26.D, Z2.D, Z25.D

0x41d170 MOVPRFX	Z27, Z24

0x41d174 MLS	Z27.D, P0/M, Z25.D, Z1.D

0x41d178 ADD	Z25.D, Z16.D, Z25.D

0x41d17c ADD	Z24.D, Z24.D, Z0.D

0x41d180 SXTW	Z26.D, P0/M, Z26.D

0x41d184 SXTW	Z25.D, P0/M, Z25.D

0x41d188 MUL	Z29.D, Z4.D, Z26.D

0x41d18c MUL	Z30.D, Z6.D, Z26.D

0x41d190 MUL	Z25.D, Z17.D, Z25.D

0x41d194 ADD	Z28.D, Z3.D, Z27.D

0x41d198 ADD	Z27.D, Z5.D, Z27.D

0x41d19c ADR	Z29.D, [Z19, Z29.D,LSL #3]    [12]

0x41d1a0 ADR	Z30.D, [Z20, Z30.D,LSL #3]    [1]

0x41d1a4 ADR	Z25.D, [Z22, Z25.D,LSL #3]    [4]

0x41d1a8 ADR	Z29.D, [Z29, Z28.D,SXTW #3]    [17]

0x41d1ac ADR	Z27.D, [Z30, Z27.D,SXTW #3]    [2]

0x41d1b0 ADR	Z25.D, [Z25, Z28.D,SXTW #3]    [7]

0x41d1b4 LD1D	{Z29.D}, P0/Z, [V29.D]    [16]

0x41d1b8 LD1D	{Z27.D}, P0/Z, [V27.D]    [11]

0x41d1bc FADD	Z27.D, Z29.D, Z27.D

0x41d1c0 ADR	Z29.D, [Z30, Z28.D,SXTW #3]    [2]

0x41d1c4 LD1D	{Z29.D}, P0/Z, [V29.D]    [15]

0x41d1c8 FSUB	Z27.D, Z27.D, Z29.D

0x41d1cc MUL	Z29.D, Z7.D, Z26.D

0x41d1d0 ADR	Z29.D, [Z21, Z29.D,LSL #3]    [3]

0x41d1d4 ADR	Z29.D, [Z29, Z28.D,SXTW #3]    [14]

0x41d1d8 ST1D	{Z27.D}, P0, [V29.D]    [13]

0x41d1dc LD1D	{Z25.D}, P0/Z, [V25.D]    [6]

0x41d1e0 FADD	Z25.D, Z27.D, Z25.D

0x41d1e4 MUL	Z27.D, Z17.D, Z26.D

0x41d1e8 MUL	Z26.D, Z18.D, Z26.D

0x41d1ec ADR	Z27.D, [Z22, Z27.D,LSL #3]    [4]

0x41d1f0 ADR	Z26.D, [Z23, Z26.D,LSL #3]    [5]

0x41d1f4 ADR	Z27.D, [Z27, Z28.D,SXTW #3]    [10]

0x41d1f8 ADR	Z26.D, [Z26, Z28.D,SXTW #3]    [9]

0x41d1fc LD1D	{Z27.D}, P0/Z, [V27.D]    [18]

0x41d200 FSUB	Z25.D, Z25.D, Z27.D

0x41d204 ST1D	{Z25.D}, P0, [V26.D]    [8]

0x41d208 CBNZ	X4, 41d160

/home/hbollore/qaas/qaas-runs/174-169-1878/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp: 54 - 57

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

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

55:       for (int i = (x_min - 2 + 1); i < (x_max + 2 + 2); i++) {

56:         post_vol(i, j) = volume(i, j) + vol_flux_x(i + 1, j + 0) - vol_flux_x(i, j);

57:         pre_vol(i, j) = post_vol(i, j) + vol_flux_y(i + 0, j + 1) - vol_flux_y(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	1.23 - 1.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.89 - 2.29
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.2]
Source	advec_mom.cpp:54-57,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	8.75 - 20.00
CQA cycles if no scalar integer	8.75 - 20.00
CQA cycles if FP arith vectorized	8.75 - 20.00
CQA cycles if fully vectorized	7.13 - 20.00
Front-end cycles	4.63
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	0.50
P7 cycles	0.50
P8 cycles	8.75
P9 cycles	8.75
P10 cycles	8.75
P11 cycles	8.75
P12 cycles	2.33
P13 cycles	2.33
P14 cycles	2.33
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	0.87 - 20.00
Inter-iter dependencies cycles	2
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	43.00
Nb uops	37.00
Nb loads	NA
Nb stores	2.00
Nb stack references	0.00
FLOP/cycle	0.91 - 0.40
Nb FLOP add-sub	8.00
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	5.60 - 12.80
Bytes prefetched	0.00
Bytes loaded	80.00
Bytes stored	32.00
Stride 0	0.00
Stride 1	0.00
Stride n	0.00
Stride unknown	12.00
Stride indirect	6.00
Vectorization ratio all	61.90
Vectorization ratio load	27.78
Vectorization ratio store	100.00
Vectorization ratio mul	100.00
Vectorization ratio add_sub	41.67
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	100.00
Vectorization ratio other	50.00
Vector-efficiency ratio all	83.33
Vector-efficiency ratio load	63.89
Vector-efficiency ratio store	100.00
Vector-efficiency ratio mul	100.00
Vector-efficiency ratio add_sub	70.83
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	100.00
Vector-efficiency ratio other	100.00

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	1.23 - 1.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.89 - 2.29
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.2]
Source	advec_mom.cpp:54-57,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	8.75 - 20.00
CQA cycles if no scalar integer	8.75 - 20.00
CQA cycles if FP arith vectorized	8.75 - 20.00
CQA cycles if fully vectorized	7.13 - 20.00
Front-end cycles	4.63
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	0.50
P7 cycles	0.50
P8 cycles	8.75
P9 cycles	8.75
P10 cycles	8.75
P11 cycles	8.75
P12 cycles	2.33
P13 cycles	2.33
P14 cycles	2.33
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	0.87 - 20.00
Inter-iter dependencies cycles	2
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	43.00
Nb uops	37.00
Nb loads	NA
Nb stores	2.00
Nb stack references	0.00
FLOP/cycle	0.91 - 0.40
Nb FLOP add-sub	8.00
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	5.60 - 12.80
Bytes prefetched	0.00
Bytes loaded	80.00
Bytes stored	32.00
Stride 0	0.00
Stride 1	0.00
Stride n	0.00
Stride unknown	12.00
Stride indirect	6.00
Vectorization ratio all	61.90
Vectorization ratio load	27.78
Vectorization ratio store	100.00
Vectorization ratio mul	100.00
Vectorization ratio add_sub	41.67
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	100.00
Vectorization ratio other	50.00
Vector-efficiency ratio all	83.33
Vector-efficiency ratio load	63.89
Vector-efficiency ratio store	100.00
Vector-efficiency ratio mul	100.00
Vector-efficiency ratio add_sub	70.83
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	100.00
Vector-efficiency ratio other	100.00

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.2]
Source file and lines	advec_mom.cpp:54-57
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:54-57
/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. 83% of vector register length is used (average across all VPU instructions).

Details

61% of VPU instructions are used in vector version (process two or more data elements in vector registers):

27% of VPU loads are used in vector version.
41% of VPU addition or subtraction instructions are used in vector version.
50% of VPU instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

No data for this section

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

The binary loop is composed of 8 FP arithmetical operations:

8: addition or subtraction

The binary loop is loading 80 bytes. The binary loop is storing 32 bytes.

Arithmetic intensity

Arithmetic intensity is 0.07 FP operations per loaded or stored byte.

General properties

nb instructions	43
loop length	172
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

4.63 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	0.50	0.50	8.75	8.75	8.75	8.75	2.33	2.33	2.33	0.00	0.00
cycles	0.50	0.50	0.00	0.00	0.00	0.00	0.50	0.50	8.75	8.75	8.75	8.75	2.33	2.33	2.33	0.00	0.00

Execution ports to units layout:

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

Cycles executing div or sqrt instructions	0.87-20.00
Longest recurrence chain latency (RecMII)	2.00

Cycles summary

Front-end	4.63
Data deps.	2.00
Overall L1	8.75-20.00

Vectorization ratios

INT

all	57%
load	27%
store	100%
mul	100%
add-sub	30%
fma	NA (no fma vectorizable/vectorized instructions)
other	60%

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

INT+FP

all	61%
load	27%
store	100%
mul	100%
add-sub	41%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	100%
other	50%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 20.00 cycles. At this rate:

4% of peak load performance is reached (4.00 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
3% of peak store performance is reached (1.60 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41d160

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 X4, 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 Z26.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%)
MOVPRFX Z27, 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%)
MLS Z27.D, P0/M, Z25.D, Z1.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 Z25.D, Z16.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%)
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%)
SXTW Z26.D, P0/M, Z26.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%)
SXTW Z25.D, P0/M, Z25.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 Z29.D, Z4.D, Z26.D																					vect (100.0%)
MUL Z30.D, Z6.D, Z26.D																					vect (100.0%)
MUL Z25.D, Z17.D, Z25.D																					vect (100.0%)
ADD Z28.D, Z3.D, 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%)
ADD Z27.D, Z5.D, 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%)
ADR Z29.D, [Z19, Z29.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 Z25.D, [Z22, Z25.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 Z29.D, [Z29, Z28.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 Z27.D, [Z30, Z27.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 Z25.D, [Z25, Z28.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 {Z29.D}, P0/Z, [V29.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 {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%)
FADD Z27.D, Z29.D, 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%)
ADR Z29.D, [Z30, Z28.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 {Z29.D}, P0/Z, [V29.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%)
FSUB Z27.D, Z27.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%)
MUL Z29.D, Z7.D, Z26.D																					vect (100.0%)
ADR Z29.D, [Z21, Z29.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 Z29.D, [Z29, Z28.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%)
ST1D {Z27.D}, P0, [V29.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%)
LD1D {Z25.D}, P0/Z, [V25.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%)
FADD Z25.D, Z27.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%)
MUL Z27.D, Z17.D, Z26.D																					vect (100.0%)
MUL Z26.D, Z18.D, Z26.D																					vect (100.0%)
ADR Z27.D, [Z22, 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 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 Z27.D, [Z27, Z28.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 Z26.D, [Z26, Z28.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%)
FSUB Z25.D, Z25.D, 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%)
ST1D {Z25.D}, P0, [V26.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 X4, 41d160 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.2+0x160>	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.2]
Source file and lines	advec_mom.cpp:54-57
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:54-57
/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. 83% of vector register length is used (average across all VPU instructions).

Details

61% of VPU instructions are used in vector version (process two or more data elements in vector registers):

27% of VPU loads are used in vector version.
41% of VPU addition or subtraction instructions are used in vector version.
50% of VPU instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

No data for this section

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

The binary loop is composed of 8 FP arithmetical operations:

8: addition or subtraction

The binary loop is loading 80 bytes. The binary loop is storing 32 bytes.

Arithmetic intensity

Arithmetic intensity is 0.07 FP operations per loaded or stored byte.

General properties

nb instructions	43
loop length	172
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

4.63 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	0.50	0.50	8.75	8.75	8.75	8.75	2.33	2.33	2.33	0.00	0.00
cycles	0.50	0.50	0.00	0.00	0.00	0.00	0.50	0.50	8.75	8.75	8.75	8.75	2.33	2.33	2.33	0.00	0.00

Execution ports to units layout:

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

Cycles executing div or sqrt instructions	0.87-20.00
Longest recurrence chain latency (RecMII)	2.00

Cycles summary

Front-end	4.63
Data deps.	2.00
Overall L1	8.75-20.00

Vectorization ratios

INT

all	57%
load	27%
store	100%
mul	100%
add-sub	30%
fma	NA (no fma vectorizable/vectorized instructions)
other	60%

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

INT+FP

all	61%
load	27%
store	100%
mul	100%
add-sub	41%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	100%
other	50%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 20.00 cycles. At this rate:

4% of peak load performance is reached (4.00 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
3% of peak store performance is reached (1.60 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41d160

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 X4, 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 Z26.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%)
MOVPRFX Z27, 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%)
MLS Z27.D, P0/M, Z25.D, Z1.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 Z25.D, Z16.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%)
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%)
SXTW Z26.D, P0/M, Z26.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%)
SXTW Z25.D, P0/M, Z25.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 Z29.D, Z4.D, Z26.D																					vect (100.0%)
MUL Z30.D, Z6.D, Z26.D																					vect (100.0%)
MUL Z25.D, Z17.D, Z25.D																					vect (100.0%)
ADD Z28.D, Z3.D, 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%)
ADD Z27.D, Z5.D, 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%)
ADR Z29.D, [Z19, Z29.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 Z25.D, [Z22, Z25.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 Z29.D, [Z29, Z28.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 Z27.D, [Z30, Z27.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 Z25.D, [Z25, Z28.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 {Z29.D}, P0/Z, [V29.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 {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%)
FADD Z27.D, Z29.D, 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%)
ADR Z29.D, [Z30, Z28.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 {Z29.D}, P0/Z, [V29.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%)
FSUB Z27.D, Z27.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%)
MUL Z29.D, Z7.D, Z26.D																					vect (100.0%)
ADR Z29.D, [Z21, Z29.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 Z29.D, [Z29, Z28.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%)
ST1D {Z27.D}, P0, [V29.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%)
LD1D {Z25.D}, P0/Z, [V25.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%)
FADD Z25.D, Z27.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%)
MUL Z27.D, Z17.D, Z26.D																					vect (100.0%)
MUL Z26.D, Z18.D, Z26.D																					vect (100.0%)
ADR Z27.D, [Z22, 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 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 Z27.D, [Z27, Z28.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 Z26.D, [Z26, Z28.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%)
FSUB Z25.D, Z25.D, 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%)
ST1D {Z25.D}, P0, [V26.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 X4, 41d160 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.2+0x160>	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

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

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