OV - exec - Loop 147

0x41ec28 ADD	X17, X30, X28,LSL #3

0x41ec2c ORR	X0, XZR, X21

  (146) 0x41ec30 LDR	D28, [X17, X0,LSL #3]

  (146) 0x41ec34 SUB	W11, W0, #2

  (146) 0x41ec38 SUB	X2, X0, #1

  (146) 0x41ec3c ORR	X3, XZR, X0

  (146) 0x41ec40 ADD	W10, W0, #1

  (146) 0x41ec44 SBFM	X1, X11, #0, #31

  (146) 0x41ec48 ORR	X4, XZR, X2

  (146) 0x41ec4c FCMPE	D28, #0

  (146) 0x41ec50 B.GT	41ec6c

  (146) 0x41ec54 CMP	W10, W13

  (146) 0x41ec58 CSEL	W3, W10, W13, #13

  (146) 0x41ec5c SBFM	X4, X3, #0, #31

  (146) 0x41ec60 ORR	X1, XZR, X4

  (146) 0x41ec64 ORR	X3, XZR, X2

  (146) 0x41ec68 ORR	X2, XZR, X0

  (146) 0x41ec6c ADD	X6, X16, X2

  (146) 0x41ec70 FABS	D3, D28

  (146) 0x41ec74 ADD	X4, X26, X4

  (146) 0x41ec78 ADD	X10, X7, X2

  (146) 0x41ec7c ADD	X11, X7, X3

  (146) 0x41ec80 LDR	D4, [X25, X0,LSL #3]

  (146) 0x41ec84 MOVI	D0, #0

  (146) 0x41ec88 LDR	D22, [X12, X6,LSL #3]

  (146) 0x41ec8c LDR	D20, [X18, X4,LSL #3]

  (146) 0x41ec90 ADD	X4, X7, X1

  (146) 0x41ec94 LDR	D19, [X5, X10,LSL #3]

  (146) 0x41ec98 FDIV	D1, D3, D22

  (146) 0x41ec9c LDR	D27, [X5, X11,LSL #3]

  (146) 0x41eca0 LDR	D5, [X5, X4,LSL #3]

  (146) 0x41eca4 FDIV	D24, D4, D20

  (146) 0x41eca8 FSUB	D23, D27, S19

  (146) 0x41ecac FSUB	D25, D19, S5

  (146) 0x41ecb0 FCMPE	D23, D0

  (146) 0x41ecb4 FABS	D29, D23

  (146) 0x41ecb8 FMUL	D30, D25, D23

  (146) 0x41ecbc FABS	D21, D25

  (146) 0x41ecc0 FADD	D31, D1, D6

  (146) 0x41ecc4 FSUB	D26, D17, S1

  (146) 0x41ecc8 FCSEL	D2, D7, D6, #9

  (146) 0x41eccc FCMPE	D30, D0

  (146) 0x41ecd0 FMUL	D4, D24, D31

  (146) 0x41ecd4 B.GT	41f0b8

  (146) 0x41ecd8 FADD	D19, D0, D19

  (146) 0x41ecdc ADD	X2, X8, X2

  (146) 0x41ece0 ADD	X3, X8, X3

  (146) 0x41ece4 ADD	X1, X8, X1

  (146) 0x41ece8 FMUL	D28, D19, D28

  (146) 0x41ecec STR	D28, [X14, X0,LSL #3]

  (146) 0x41ecf0 FABS	D24, D28

  (146) 0x41ecf4 LDR	D25, [X9, X2,LSL #3]

  (146) 0x41ecf8 LDR	D23, [X9, X3,LSL #3]

  (146) 0x41ecfc LDR	D29, [X9, X1,LSL #3]

  (146) 0x41ed00 LDR	D21, [X5, X10,LSL #3]

  (146) 0x41ed04 FSUB	D30, D23, S25

  (146) 0x41ed08 LDR	D31, [X12, X6,LSL #3]

  (146) 0x41ed0c FSUB	D2, D25, S29

  (146) 0x41ed10 FCMPE	D30, #0

  (146) 0x41ed14 FABS	D20, D30

  (146) 0x41ed18 FMUL	D3, D2, D30

  (146) 0x41ed1c FABS	D22, D2

  (146) 0x41ed20 FCSEL	D1, D7, D6, #9

  (146) 0x41ed24 FCMPE	D3, #0

  (146) 0x41ed28 B.GT	41f0dc

  (146) 0x41ed2c FADD	D25, D25, D18

  (146) 0x41ed30 FMUL	D28, D25, D28

  (146) 0x41ed34 STR	D28, [X15, X0,LSL #3]

  (146) 0x41ed38 ADD	X0, X0, #1

  (146) 0x41ed3c CMP	W13, W0

  (146) 0x41ed40 B.GE	41ec30

0x41ed44 LDR	W6, [SP, #112]

0x41ed48 ADD	W27, W27, #1

0x41ed4c ADD	X15, X15, X19

0x41ed50 ADD	X14, X14, X20

0x41ed54 ADD	X8, X8, X22

0x41ed58 LDR	X17, [SP, #104]

0x41ed5c ADD	X16, X16, X24

0x41ed60 ADD	X7, X7, X23

0x41ed64 ADD	X28, X28, X17

0x41ed68 CMP	W6, W27

0x41ed6c B.GT	41ec28

  (146) 0x41f0b8 FMUL	D0, D26, D29

  (146) 0x41f0bc FSUB	D3, D6, S1

  (146) 0x41f0c0 FMINNM	D22, D29, D21

  (146) 0x41f0c4 FMADD	D20, D4, D21, D0

  (146) 0x41f0c8 FMUL	D1, D3, D2

  (146) 0x41f0cc FMUL	D27, D20, D16

  (146) 0x41f0d0 FMINNM	D5, D27, D22

  (146) 0x41f0d4 FMUL	D0, D5, D1

  (146) 0x41f0d8 B	41ecd8

  (146) 0x41f0dc FMUL	D27, D21, D31

  (146) 0x41f0e0 FMUL	D26, D26, D20

  (146) 0x41f0e4 FMINNM	D5, D20, D22

  (146) 0x41f0e8 FMADD	D4, D4, D22, D26

  (146) 0x41f0ec FDIV	D0, D24, D27

  (146) 0x41f0f0 FMUL	D19, D4, D16

  (146) 0x41f0f4 FMINNM	D24, D19, D5

  (146) 0x41f0f8 FSUB	D23, D6, S0

  (146) 0x41f0fc FMUL	D29, D23, D1

  (146) 0x41f100 FMADD	D25, D24, D29, D25

  (146) 0x41f104 B	41ed30

/home/hbollore/qaas-runs/170-309-0949/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/advec_cell_kernel.f90: 110 - 155

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

110:         DO j=x_min,x_max+2

111:

112:           IF(vol_flux_x(j,k).GT.0.0)THEN

[...]

118:             upwind   =MIN(j+1,x_max+2)

[...]

124:           sigmat=ABS(vol_flux_x(j,k))/pre_vol(donor,k)

125:           sigma3=(1.0_8+sigmat)*(vertexdx(j)/vertexdx(dif))

126:           sigma4=2.0_8-sigmat

127:

128:           sigma=sigmat

129:           sigmav=sigmat

130:

131:           diffuw=density1(donor,k)-density1(upwind,k)

132:           diffdw=density1(downwind,k)-density1(donor,k)

133:           wind=1.0_8

134:           IF(diffdw.LE.0.0) wind=-1.0_8

135:           IF(diffuw*diffdw.GT.0.0)THEN

136:             limiter=(1.0_8-sigmav)*wind*MIN(ABS(diffuw),ABS(diffdw)&

137:               ,one_by_six*(sigma3*ABS(diffuw)+sigma4*ABS(diffdw)))

138:           ELSE

139:             limiter=0.0

140:           ENDIF

141:           mass_flux_x(j,k)=vol_flux_x(j,k)*(density1(donor,k)+limiter)

142:

143:           sigmam=ABS(mass_flux_x(j,k))/(density1(donor,k)*pre_vol(donor,k))

144:           diffuw=energy1(donor,k)-energy1(upwind,k)

145:           diffdw=energy1(downwind,k)-energy1(donor,k)

146:           wind=1.0_8

147:           IF(diffdw.LE.0.0) wind=-1.0_8

148:           IF(diffuw*diffdw.GT.0.0)THEN

149:             limiter=(1.0_8-sigmam)*wind*MIN(ABS(diffuw),ABS(diffdw)&

150:               ,one_by_six*(sigma3*ABS(diffuw)+sigma4*ABS(diffdw)))

151:           ELSE

152:             limiter=0.0

153:           ENDIF

154:

155:           ener_flux(j,k)=mass_flux_x(j,k)*(energy1(donor,k)+limiter)

Path /

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	4.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.54
Bottlenecks	P2, P3, P4, P5,
Function	advec_cell_kernel._omp_fn.0
Source	advec_cell_kernel.f90:110-110,advec_cell_kernel.f90:137-137
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	2.50
CQA cycles if no scalar integer	2.50
CQA cycles if FP arith vectorized	2.50
CQA cycles if fully vectorized	0.63
Front-end cycles	1.63
DIV/SQRT cycles	0.50
P0 cycles	0.50
P1 cycles	2.50
P2 cycles	2.50
P3 cycles	2.50
P4 cycles	2.50
P5 cycles	0.00
P6 cycles	0.00
P7 cycles	0.00
P8 cycles	0.00
P9 cycles	0.67
P10 cycles	0.67
P11 cycles	0.67
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	13.00
Nb uops	13.00
Nb loads	NA
Nb stores	0.00
Nb stack references	0.00
FLOP/cycle	0.00
Nb FLOP add-sub	0.00
Nb FLOP mul	0.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	4.80
Bytes prefetched	0.00
Bytes loaded	12.00
Bytes stored	0.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	NA
Vectorization ratio store	NA
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	25.00
Vector-efficiency ratio load	NA
Vector-efficiency ratio store	NA
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	25.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	25.00

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	4.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.54
Bottlenecks	P2, P3, P4, P5,
Function	advec_cell_kernel._omp_fn.0
Source	advec_cell_kernel.f90:110-110,advec_cell_kernel.f90:137-137
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	2.50
CQA cycles if no scalar integer	2.50
CQA cycles if FP arith vectorized	2.50
CQA cycles if fully vectorized	0.63
Front-end cycles	1.63
DIV/SQRT cycles	0.50
P0 cycles	0.50
P1 cycles	2.50
P2 cycles	2.50
P3 cycles	2.50
P4 cycles	2.50
P5 cycles	0.00
P6 cycles	0.00
P7 cycles	0.00
P8 cycles	0.00
P9 cycles	0.67
P10 cycles	0.67
P11 cycles	0.67
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	13.00
Nb uops	13.00
Nb loads	NA
Nb stores	0.00
Nb stack references	0.00
FLOP/cycle	0.00
Nb FLOP add-sub	0.00
Nb FLOP mul	0.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	4.80
Bytes prefetched	0.00
Bytes loaded	12.00
Bytes stored	0.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	NA
Vectorization ratio store	NA
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	25.00
Vector-efficiency ratio load	NA
Vector-efficiency ratio store	NA
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	25.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	25.00

Path /

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

Function	advec_cell_kernel._omp_fn.0
Source file and lines	advec_cell_kernel.f90:110-155
Module	exec

The loop is defined in /home/hbollore/qaas-runs/170-309-0949/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/advec_cell_kernel.f90:110-112,118,124-155.

Analyzed code is defined in /home/hbollore/qaas-runs/170-309-0949/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/advec_cell_kernel.f90:110,137.

Warnings:

Non-innermost loop: analyzing only self part (ignoring child loops).
Ignoring paths for analysis
Too many paths. Rerun with max-paths=8
RecMII not computed since number of paths is unknown or > max_paths

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

This loop has 8 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 = max(0,x) (Fortran instrinsic procedure)

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. 4 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 2.50 to 0.62 cycles (4.00x speedup).

Details

All VPU instructions are used in scalar version (process only one data element in vector registers). Since your execution units are vector units, only a vectorized loop can use their full power.

Workaround

Try another compiler or update/tune your current one:
- 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: Fortran storage order is column-major: do i do j a(i,j) = b(i,j) (slow, non stride 1) => do i do j a(j,i) = b(i,j) (fast, stride 1)
- If your loop streams arrays of structures (AoS), try to use structures of arrays instead (SoA): do i a(i)%x = b(i)%x (slow, non stride 1) => do i a%x(i) = b%x(i) (fast, stride 1)

No data for this section

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

The binary loop does not contain any FP arithmetical operations. The binary loop is loading 12 bytes.

General properties

nb instructions	13
loop length	52
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

1.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	0.50	0.50	2.50	2.50	2.50	2.50	0.00	0.00	0.00	0.00	0.67	0.67	0.67	0.00	0.00
cycles	0.50	0.50	2.50	2.50	2.50	2.50	0.00	0.00	0.00	0.00	0.67	0.67	0.67	0.00	0.00

Execution ports to units layout:

P0:
P1:
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

NA

Cycles summary

Front-end	1.63
Overall L1	2.50

Vectorization ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 2.50 cycles. At this rate:

5% of peak load performance is reached (4.80 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41ec28

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P10	P11	P12	Latency	Recip. throughput
ADD X17, X30, X28,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ORR X0, XZR, X21	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
LDR W6, [SP, #112]	1	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
ADD W27, W27, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ADD X15, X15, X19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ADD X14, X14, X20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ADD X8, X8, X22	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
LDR X17, [SP, #104]	1	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
ADD X16, X16, X24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ADD X7, X7, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ADD X28, X28, X17	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
CMP W6, W27	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.33
B.GT 41ec28 <__advec_cell_kernel_module_MOD_advec_cell_kernel._omp_fn.0+0x8e8>	1	0.50	0.50	0	0	0	0	0	0	0	1	0.50

Function	advec_cell_kernel._omp_fn.0
Source file and lines	advec_cell_kernel.f90:110-155
Module	exec

The loop is defined in /home/hbollore/qaas-runs/170-309-0949/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/advec_cell_kernel.f90:110-112,118,124-155.

Analyzed code is defined in /home/hbollore/qaas-runs/170-309-0949/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/advec_cell_kernel.f90:110,137.

Warnings:

Non-innermost loop: analyzing only self part (ignoring child loops).
Ignoring paths for analysis
Too many paths. Rerun with max-paths=8
RecMII not computed since number of paths is unknown or > max_paths

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

This loop has 8 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 = max(0,x) (Fortran instrinsic procedure)

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. 4 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 2.50 to 0.62 cycles (4.00x speedup).

Details

All VPU instructions are used in scalar version (process only one data element in vector registers). Since your execution units are vector units, only a vectorized loop can use their full power.

Workaround

Try another compiler or update/tune your current one:
- 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: Fortran storage order is column-major: do i do j a(i,j) = b(i,j) (slow, non stride 1) => do i do j a(j,i) = b(i,j) (fast, stride 1)
- If your loop streams arrays of structures (AoS), try to use structures of arrays instead (SoA): do i a(i)%x = b(i)%x (slow, non stride 1) => do i a%x(i) = b%x(i) (fast, stride 1)

No data for this section

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

The binary loop does not contain any FP arithmetical operations. The binary loop is loading 12 bytes.

General properties

nb instructions	13
loop length	52
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

1.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	0.50	0.50	2.50	2.50	2.50	2.50	0.00	0.00	0.00	0.00	0.67	0.67	0.67	0.00	0.00
cycles	0.50	0.50	2.50	2.50	2.50	2.50	0.00	0.00	0.00	0.00	0.67	0.67	0.67	0.00	0.00

Execution ports to units layout:

P0:
P1:
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

NA

Cycles summary

Front-end	1.63
Overall L1	2.50

Vectorization ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 2.50 cycles. At this rate:

5% of peak load performance is reached (4.80 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41ec28

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P10	P11	P12	Latency	Recip. throughput
ADD X17, X30, X28,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ORR X0, XZR, X21	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
LDR W6, [SP, #112]	1	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
ADD W27, W27, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ADD X15, X15, X19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ADD X14, X14, X20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ADD X8, X8, X22	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
LDR X17, [SP, #104]	1	0	0	0	0	0	0	0.33	0.33	0.33	4	0.33
ADD X16, X16, X24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ADD X7, X7, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
ADD X28, X28, X17	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.25
CMP W6, W27	1	0	0	0.25	0.25	0.25	0.25	0	0	0	1	0.33
B.GT 41ec28 <__advec_cell_kernel_module_MOD_advec_cell_kernel._omp_fn.0+0x8e8>	1	0.50	0.50	0	0	0	0	0	0	0	1	0.50

Report Configuration

Vectorization

Details

Workaround

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code

Vectorization

Details

Workaround

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code