OV - exec - Loop 123

0x41fde8 LD1D	{Z6.D}, P3/Z, [X4, X0,LSL #3]    [7]

0x41fdec LD1D	{Z7.D}, P3/Z, [X3, X0,LSL #3]    [3]

0x41fdf0 LD1D	{Z16.D}, P3/Z, [X8, X0,LSL #3]    [2]

0x41fdf4 LD1D	{Z18.D}, P3/Z, [X10, X0,LSL #3]    [5]

0x41fdf8 LD1D	{Z20.D}, P3/Z, [X9, X0,LSL #3]    [1]

0x41fdfc FSUB	Z17.D, Z6.D, Z7.D

0x41fe00 FADD	Z19.D, Z17.D, Z16.D

0x41fe04 FSUB	Z21.D, Z19.D, Z18.D

0x41fe08 FADD	Z22.D, Z21.D, Z20.D

0x41fe0c ST1D	{Z22.D}, P3, [X1, X0,LSL #3]    [6]

0x41fe10 LD1D	{Z23.D}, P3/Z, [X4, X0,LSL #3]    [7]

0x41fe14 LD1D	{Z24.D}, P3/Z, [X3, X0,LSL #3]    [3]

0x41fe18 FSUB	Z25.D, Z23.D, Z24.D

0x41fe1c FSUB	Z26.D, Z22.D, Z25.D

0x41fe20 ST1D	{Z26.D}, P3, [X2, X0,LSL #3]    [4]

0x41fe24 ADD	X0, X0, X12

0x41fe28 WHILELO	P3.D, X0, X28

0x41fe2c B.NE	41fde8

/home/hbollore/qaas/qaas-runs/169-814-4157/intel/CloverLeaf/build/CloverLeaf/CloverLeaf_ref/kernels/advec_cell_kernel.f90: 182 - 183

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

182:             pre_vol(j,k)=volume(j,k)+(vol_flux_y(j  ,k+1)-vol_flux_y(j,k)+vol_flux_x(j+1,k  )-vol_flux_x(j,k))

183:             post_vol(j,k)=pre_vol(j,k)-(vol_flux_y(j  ,k+1)-vol_flux_y(j,k))

Coverage (%)	Name	Source Location	Module
►100.00+	GOMP_parallel		libgomp.so.1.0.0
○	advec_cell_kernel	advec_cell_kernel.f90:270	exec
○	advec_cell_driver	advec_cell_driver.f90:84	exec
○	advection	advection.f90:57	exec
○	hydro	hydro.f90:64	exec
○	main	clover_leaf.f90:41	exec
○	__libc_start_main		libc-2.31.so
○	_start	clover_leaf.f90:41	exec

Path /

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	1.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.50
Bottlenecks	P10, P11,
Function	advec_cell_kernel._omp_fn.0
Source	advec_cell_kernel.f90:182-183
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	4.50
CQA cycles if no scalar integer	4.50
CQA cycles if FP arith vectorized	4.50
CQA cycles if fully vectorized	4.50
Front-end cycles	2.25
DIV/SQRT cycles	0.50
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.25
P3 cycles	1.00
P4 cycles	0.25
P5 cycles	3.00
P6 cycles	3.00
P7 cycles	3.00
P8 cycles	3.00
P9 cycles	4.50
P10 cycles	4.50
P11 cycles	0.00
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	18.00
Nb uops	18.00
Nb loads	NA
Nb stores	2.00
Nb stack references	0.00
FLOP/cycle	5.33
Nb FLOP add-sub	24.00
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	64.00
Bytes prefetched	0.00
Bytes loaded	224.00
Bytes stored	64.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	100.00
Vectorization ratio load	100.00
Vectorization ratio store	100.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	100.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	NA
Vector-efficiency ratio all	100.00
Vector-efficiency ratio load	100.00
Vector-efficiency ratio store	100.00
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	100.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	NA

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	1.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.50
Bottlenecks	P10, P11,
Function	advec_cell_kernel._omp_fn.0
Source	advec_cell_kernel.f90:182-183
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	4.50
CQA cycles if no scalar integer	4.50
CQA cycles if FP arith vectorized	4.50
CQA cycles if fully vectorized	4.50
Front-end cycles	2.25
DIV/SQRT cycles	0.50
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.25
P3 cycles	1.00
P4 cycles	0.25
P5 cycles	3.00
P6 cycles	3.00
P7 cycles	3.00
P8 cycles	3.00
P9 cycles	4.50
P10 cycles	4.50
P11 cycles	0.00
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	18.00
Nb uops	18.00
Nb loads	NA
Nb stores	2.00
Nb stack references	0.00
FLOP/cycle	5.33
Nb FLOP add-sub	24.00
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	64.00
Bytes prefetched	0.00
Bytes loaded	224.00
Bytes stored	64.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	100.00
Vectorization ratio load	100.00
Vectorization ratio store	100.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	100.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	NA
Vector-efficiency ratio all	100.00
Vector-efficiency ratio load	100.00
Vector-efficiency ratio store	100.00
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	100.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	NA

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.f90:182-183
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/169-814-4157/intel/CloverLeaf/build/CloverLeaf/CloverLeaf_ref/kernels/advec_cell_kernel.f90:182-183.

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

gain
potential
hint
expert

Vectorization

Your loop is fully vectorized, using full register length.

Details

All VPU instructions are used in vector version (process two or more data elements in vector registers).

No data for this section

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

The binary loop is composed of 24 FP arithmetical operations:

24: addition or subtraction

The binary loop is loading 224 bytes. The binary loop is storing 64 bytes.

Arithmetic intensity

Arithmetic intensity is 0.08 FP operations per loaded or stored byte.

General properties

nb instructions	18
loop length	72
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

2.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	0.50	0.50	0.50	0.25	1.00	0.25	3.00	3.00	1.00	1.00	4.50	4.50	0.00	0.00	0.00
cycles	0.50	0.50	0.50	0.25	1.00	0.25	3.00	3.00	3.00	3.00	4.50	4.50	0.00	0.00	0.00

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	2.25
Data deps.	1.00
Overall L1	4.50

Vectorization ratios

INT

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

FP

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	100%
load	100%
store	100%
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)

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 4.50 cycles. At this rate:

51% of peak load performance is reached (49.78 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
29% of peak store performance is reached (14.22 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41fde8

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput
LD1D {Z6.D}, P3/Z, [X4, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z7.D}, P3/Z, [X3, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z16.D}, P3/Z, [X8, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z18.D}, P3/Z, [X10, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z20.D}, P3/Z, [X9, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
FSUB Z17.D, Z6.D, Z7.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FADD Z19.D, Z17.D, Z16.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FSUB Z21.D, Z19.D, Z18.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FADD Z22.D, Z21.D, Z20.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
ST1D {Z22.D}, P3, [X1, X0,LSL #3]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	2	0.50
LD1D {Z23.D}, P3/Z, [X4, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z24.D}, P3/Z, [X3, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
FSUB Z25.D, Z23.D, Z24.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FSUB Z26.D, Z22.D, Z25.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
ST1D {Z26.D}, P3, [X2, X0,LSL #3]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	2	0.50
ADD X0, X0, X12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.25
WHILELO P3.D, X0, X28	1	0	0	0	0	1	0	0	0	0	0	0	0	3	2
B.NE 41fde8 <__advec_cell_kernel_module_MOD_advec_cell_kernel._omp_fn.0+0xd08>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50

Function	advec_cell_kernel._omp_fn.0
Source file and lines	advec_cell_kernel.f90:182-183
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/169-814-4157/intel/CloverLeaf/build/CloverLeaf/CloverLeaf_ref/kernels/advec_cell_kernel.f90:182-183.

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

gain
potential
hint
expert

Vectorization

Your loop is fully vectorized, using full register length.

Details

All VPU instructions are used in vector version (process two or more data elements in vector registers).

No data for this section

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

The binary loop is composed of 24 FP arithmetical operations:

24: addition or subtraction

The binary loop is loading 224 bytes. The binary loop is storing 64 bytes.

Arithmetic intensity

Arithmetic intensity is 0.08 FP operations per loaded or stored byte.

General properties

nb instructions	18
loop length	72
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

2.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	0.50	0.50	0.50	0.25	1.00	0.25	3.00	3.00	1.00	1.00	4.50	4.50	0.00	0.00	0.00
cycles	0.50	0.50	0.50	0.25	1.00	0.25	3.00	3.00	3.00	3.00	4.50	4.50	0.00	0.00	0.00

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	2.25
Data deps.	1.00
Overall L1	4.50

Vectorization ratios

INT

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

FP

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	100%
load	100%
store	100%
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)

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 4.50 cycles. At this rate:

51% of peak load performance is reached (49.78 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
29% of peak store performance is reached (14.22 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41fde8

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput
LD1D {Z6.D}, P3/Z, [X4, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z7.D}, P3/Z, [X3, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z16.D}, P3/Z, [X8, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z18.D}, P3/Z, [X10, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z20.D}, P3/Z, [X9, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
FSUB Z17.D, Z6.D, Z7.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FADD Z19.D, Z17.D, Z16.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FSUB Z21.D, Z19.D, Z18.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FADD Z22.D, Z21.D, Z20.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
ST1D {Z22.D}, P3, [X1, X0,LSL #3]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	2	0.50
LD1D {Z23.D}, P3/Z, [X4, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z24.D}, P3/Z, [X3, X0,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
FSUB Z25.D, Z23.D, Z24.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FSUB Z26.D, Z22.D, Z25.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
ST1D {Z26.D}, P3, [X2, X0,LSL #3]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	2	0.50
ADD X0, X0, X12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.25
WHILELO P3.D, X0, X28	1	0	0	0	0	1	0	0	0	0	0	0	0	3	2
B.NE 41fde8 <__advec_cell_kernel_module_MOD_advec_cell_kernel._omp_fn.0+0xd08>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50

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