OV - exec - Loop 157

	Loop Id: 157	Module: exec	Source: advec_cell.cpp:55-58 [...]	Coverage: 1.01%

0x41b2c4 ADD	X6, X18, X1

0x41b2c8 ADD	W25, W5, W1

0x41b2cc ADD	W26, W4, W1

0x41b2d0 ADD	X1, X1, #2

0x41b2d4 ADD	X7, X6, #1

0x41b2d8 SDIV	X6, X6, X24

0x41b2dc ADD	W20, W23, W6

0x41b2e0 SDIV	X7, X7, X24

0x41b2e4 SBFM	X20, X20, #0, #31

0x41b2e8 ADD	W21, W23, W7

0x41b2ec MADD	X27, X9, X20, XZR

0x41b2f0 MADD	X19, X11, X20, XZR

0x41b2f4 SBFM	X21, X21, #0, #31

0x41b2f8 MADD	X28, X9, X21, XZR

0x41b2fc ADD	X27, X8, X27,LSL #3

0x41b300 ADD	X19, X10, X19,LSL #3

0x41b304 ADD	X28, X8, X28,LSL #3

0x41b308 MADD	W25, W3, W6, W25

0x41b30c LDR	D0, [X27, X25,SXTW #3]    [8]

0x41b310 MADD	W6, W3, W6, W26

0x41b314 LDR	D1, [X19, X6,SXTW #3]    [10]

0x41b318 MADD	X6, X11, X21, XZR

0x41b31c ADD	X6, X10, X6,LSL #3

0x41b320 MADD	W7, W3, W7, W26

0x41b324 ADD	W26, W7, #1

0x41b328 SBFM	X30, X7, #61, #31

0x41b32c ADD	X0, X28, X30

0x41b330 ADD	X26, X6, W26,SXTW #3

0x41b334 LD1	{V0.D[1]}, [X0]    [5]

0x41b338 LD1	{V1.D[1]}, [X26]    [6]

0x41b33c ADD	X0, X6, X30

0x41b340 MADD	X6, X13, X21, XZR

0x41b344 ADD	X6, X12, X6,LSL #3

0x41b348 FADD	V0.2D, V1.2D, V0.2D

0x41b34c LDR	D1, [X19, X25,SXTW #3]    [10]

0x41b350 LD1	{V1.D[1]}, [X0]    [4]

0x41b354 MADD	X0, X13, X20, XZR

0x41b358 ADD	X0, X12, X0,LSL #3

0x41b35c FSUB	V0.2D, V0.2D, V1.2D

0x41b360 STR	D0, [X0, X25,SXTW #3]    [3]

0x41b364 ADD	X0, X6, X30

0x41b368 MADD	X6, X15, X21, XZR

0x41b36c ST1	{V0.D[1]}, [X0]    [2]

0x41b370 MADD	X0, X15, X20, XZR

0x41b374 LDR	D0, [X27, X25,SXTW #3]    [8]

0x41b378 LDR	D1, [X28, X7,SXTW #3]    [9]

0x41b37c ADD	X6, X14, X6,LSL #3

0x41b380 ADD	X0, X14, X0,LSL #3

0x41b384 STR	D0, [X0, X25,SXTW #3]    [1]

0x41b388 STR	D1, [X6, X7,SXTW #3]    [7]

0x41b38c CMP	X2, X1

0x41b390 B.NE	41b2c4

/home/hbollore/qaas/qaas-runs/174-169-1878/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_cell.cpp: 55 - 58

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

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

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

57:           pre_vol(i, j) = volume(i, j) + vol_flux_x(i + 1, j + 0) - vol_flux_x(i, j);

58:           post_vol(i, j) = volume(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	3.25 - 10.00
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	3.08
Bottlenecks	P8, P10,
Function	advec_cell_kernel(int, int, int, int, int, int, clover::Buffer1D&, clover::Buffer1D&, 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::Buffer2D&) [clone .omp_outlined.2]
Source	advec_cell.cpp:55-58,context.h:69-69
Source loop unroll info	unrolled by 2
Source loop unroll confidence level	medium
Unroll/vectorization loop type	NA
Unroll factor	2
CQA cycles	13.00 - 40.00
CQA cycles if no scalar integer	4.00
CQA cycles if FP arith vectorized	13.00 - 40.00
CQA cycles if fully vectorized	13.00 - 40.00
Front-end cycles	6.50
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	4.83
P3 cycles	4.83
P4 cycles	4.83
P5 cycles	4.83
P6 cycles	13.00
P7 cycles	4.67
P8 cycles	2.25
P9 cycles	2.25
P10 cycles	2.25
P11 cycles	2.25
P12 cycles	4.00
P13 cycles	4.00
P14 cycles	4.00
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	10.00 - 40.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	52.00
Nb uops	52.00
Nb loads	NA
Nb stores	4.00
Nb stack references	0.00
FLOP/cycle	0.31 - 0.10
Nb FLOP add-sub	4.00
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	2.60 - 8.00
Bytes prefetched	0.00
Bytes loaded	64.00
Bytes stored	40.00
Stride 0	0.00
Stride 1	0.00
Stride n	0.00
Stride unknown	4.00
Stride indirect	6.00
Vectorization ratio all	10.53
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	50.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	60.53
Vector-efficiency ratio load	50.00
Vector-efficiency ratio store	50.00
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	62.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	100.00

Metric	Value
CQA speedup if no scalar integer	3.25 - 10.00
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	3.08
Bottlenecks	P8, P10,
Function	advec_cell_kernel(int, int, int, int, int, int, clover::Buffer1D&, clover::Buffer1D&, 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::Buffer2D&) [clone .omp_outlined.2]
Source	advec_cell.cpp:55-58,context.h:69-69
Source loop unroll info	unrolled by 2
Source loop unroll confidence level	medium
Unroll/vectorization loop type	NA
Unroll factor	2
CQA cycles	13.00 - 40.00
CQA cycles if no scalar integer	4.00
CQA cycles if FP arith vectorized	13.00 - 40.00
CQA cycles if fully vectorized	13.00 - 40.00
Front-end cycles	6.50
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	4.83
P3 cycles	4.83
P4 cycles	4.83
P5 cycles	4.83
P6 cycles	13.00
P7 cycles	4.67
P8 cycles	2.25
P9 cycles	2.25
P10 cycles	2.25
P11 cycles	2.25
P12 cycles	4.00
P13 cycles	4.00
P14 cycles	4.00
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	10.00 - 40.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	52.00
Nb uops	52.00
Nb loads	NA
Nb stores	4.00
Nb stack references	0.00
FLOP/cycle	0.31 - 0.10
Nb FLOP add-sub	4.00
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	2.60 - 8.00
Bytes prefetched	0.00
Bytes loaded	64.00
Bytes stored	40.00
Stride 0	0.00
Stride 1	0.00
Stride n	0.00
Stride unknown	4.00
Stride indirect	6.00
Vectorization ratio all	10.53
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	50.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	60.53
Vector-efficiency ratio load	50.00
Vector-efficiency ratio store	50.00
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	62.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	100.00

Path /

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

Function	advec_cell_kernel(int, int, int, int, int, int, clover::Buffer1D&, clover::Buffer1D&, 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::Buffer2D&) [clone .omp_outlined.2]
Source file and lines	advec_cell.cpp:55-58
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_cell.cpp:55-58
/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

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 40.00 to 4.00 cycles (10.00x 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 60% of vector register length is used (average across all VPU instructions).

Details

10% of VPU instructions are used in vector version (process two or more data elements in vector registers):

0% of VPU loads are used in vector version.
0% of VPU stores are used in vector version.
50% of VPU addition or subtraction 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.

No data for this section

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

The binary loop is composed of 4 FP arithmetical operations:

4: addition or subtraction

The binary loop is loading 64 bytes. The binary loop is storing 40 bytes.

Arithmetic intensity

Arithmetic intensity is 0.04 FP operations per loaded or stored byte.

General properties

nb instructions	52
loop length	208
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

6.50 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	4.83	4.83	4.83	4.83	13.00	4.67	2.25	2.25	2.25	2.25	4.00	4.00	4.00	0.00	0.00
cycles	0.50	0.50	4.83	4.83	4.83	4.83	13.00	4.67	2.25	2.25	2.25	2.25	4.00	4.00	4.00	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	10.00-40.00
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	6.50
Data deps.	1.00
Overall L1	13.00-40.00

Vectorization ratios

INT

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

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	10%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	50%
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 40.00 cycles. At this rate:

1% of peak load performance is reached (1.60 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (1.00 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41b2c4

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
ADD X6, X18, X1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD W25, W5, W1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD W26, W4, W1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X1, X1, #2	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X7, X6, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
SDIV X6, X6, X24	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	5-20	5-20	N/A
ADD W20, W23, W6	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	scal (25.0%)
SDIV X7, X7, X24	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	5-20	5-20	N/A
SBFM X20, X20, #0, #31	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	scal (100.0%)
ADD W21, W23, W7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	scal (25.0%)
MADD X27, X9, X20, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
MADD X19, X11, X20, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
SBFM X21, X21, #0, #31	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	scal (100.0%)
MADD X28, X9, X21, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ADD X27, X8, X27,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X19, X10, X19,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X28, X8, X28,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
MADD W25, W3, W6, W25	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
LDR D0, [X27, X25,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
MADD W6, W3, W6, W26	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
LDR D1, [X19, X6,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
MADD X6, X11, X21, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ADD X6, X10, X6,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
MADD W7, W3, W7, W26	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ADD W26, W7, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
SBFM X30, X7, #61, #31	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	scal (100.0%)
ADD X0, X28, X30	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X26, X6, W26,SXTW #3	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50	N/A
LD1 {V0.D[1]}, [X0]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	8	0.33	scal (50.0%)
LD1 {V1.D[1]}, [X26]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	8	0.33	scal (50.0%)
ADD X0, X6, X30	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
MADD X6, X13, X21, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ADD X6, X12, X6,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
FADD V0.2D, V1.2D, V0.2D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	vect (100.0%)
LDR D1, [X19, X25,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
LD1 {V1.D[1]}, [X0]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	8	0.33	scal (50.0%)
MADD X0, X13, X20, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ADD X0, X12, X0,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
FSUB V0.2D, V0.2D, V1.2D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	vect (100.0%)
STR D0, [X0, X25,SXTW #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (50.0%)
ADD X0, X6, X30	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
MADD X6, X15, X21, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ST1 {V0.D[1]}, [X0]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	4	1	scal (50.0%)
MADD X0, X15, X20, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
LDR D0, [X27, X25,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
LDR D1, [X28, X7,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
ADD X6, X14, X6,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X0, X14, X0,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
STR D0, [X0, X25,SXTW #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (50.0%)
STR D1, [X6, X7,SXTW #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (50.0%)
CMP X2, X1	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	1	0.33	N/A
B.NE 41b2c4 <_Z17advec_cell_kerneliiiiiiRN6clover8Buffer1DIdEES2_RNS_8Buffer2DIdEES5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_.omp_outlined.2+0x114>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	advec_cell_kernel(int, int, int, int, int, int, clover::Buffer1D&, clover::Buffer1D&, 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::Buffer2D&) [clone .omp_outlined.2]
Source file and lines	advec_cell.cpp:55-58
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_cell.cpp:55-58
/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

Code clean check

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 60% of vector register length is used (average across all VPU instructions).

Details

10% of VPU instructions are used in vector version (process two or more data elements in vector registers):

0% of VPU loads are used in vector version.
0% of VPU stores are used in vector version.
50% of VPU addition or subtraction 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.

No data for this section

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

The binary loop is composed of 4 FP arithmetical operations:

4: addition or subtraction

The binary loop is loading 64 bytes. The binary loop is storing 40 bytes.

Arithmetic intensity

Arithmetic intensity is 0.04 FP operations per loaded or stored byte.

General properties

nb instructions	52
loop length	208
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

6.50 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	4.83	4.83	4.83	4.83	13.00	4.67	2.25	2.25	2.25	2.25	4.00	4.00	4.00	0.00	0.00
cycles	0.50	0.50	4.83	4.83	4.83	4.83	13.00	4.67	2.25	2.25	2.25	2.25	4.00	4.00	4.00	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	10.00-40.00
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	6.50
Data deps.	1.00
Overall L1	13.00-40.00

Vectorization ratios

INT

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

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	10%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	50%
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 40.00 cycles. At this rate:

1% of peak load performance is reached (1.60 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (1.00 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 41b2c4

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
ADD X6, X18, X1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD W25, W5, W1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD W26, W4, W1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X1, X1, #2	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X7, X6, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
SDIV X6, X6, X24	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	5-20	5-20	N/A
ADD W20, W23, W6	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	scal (25.0%)
SDIV X7, X7, X24	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	5-20	5-20	N/A
SBFM X20, X20, #0, #31	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	scal (100.0%)
ADD W21, W23, W7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	scal (25.0%)
MADD X27, X9, X20, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
MADD X19, X11, X20, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
SBFM X21, X21, #0, #31	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	scal (100.0%)
MADD X28, X9, X21, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ADD X27, X8, X27,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X19, X10, X19,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X28, X8, X28,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
MADD W25, W3, W6, W25	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
LDR D0, [X27, X25,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
MADD W6, W3, W6, W26	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
LDR D1, [X19, X6,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
MADD X6, X11, X21, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ADD X6, X10, X6,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
MADD W7, W3, W7, W26	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ADD W26, W7, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
SBFM X30, X7, #61, #31	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	scal (100.0%)
ADD X0, X28, X30	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X26, X6, W26,SXTW #3	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50	N/A
LD1 {V0.D[1]}, [X0]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	8	0.33	scal (50.0%)
LD1 {V1.D[1]}, [X26]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	8	0.33	scal (50.0%)
ADD X0, X6, X30	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
MADD X6, X13, X21, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ADD X6, X12, X6,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
FADD V0.2D, V1.2D, V0.2D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	vect (100.0%)
LDR D1, [X19, X25,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
LD1 {V1.D[1]}, [X0]	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.33	0.33	0.33	8	0.33	scal (50.0%)
MADD X0, X13, X20, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ADD X0, X12, X0,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
FSUB V0.2D, V0.2D, V1.2D	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	vect (100.0%)
STR D0, [X0, X25,SXTW #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (50.0%)
ADD X0, X6, X30	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
MADD X6, X15, X21, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
ST1 {V0.D[1]}, [X0]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	4	1	scal (50.0%)
MADD X0, X15, X20, XZR	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	N/A
LDR D0, [X27, X25,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
LDR D1, [X28, X7,SXTW #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	scal (50.0%)
ADD X6, X14, X6,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
ADD X0, X14, X0,LSL #3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	1	0.17	N/A
STR D0, [X0, X25,SXTW #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (50.0%)
STR D1, [X6, X7,SXTW #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	2	0.50	scal (50.0%)
CMP X2, X1	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	1	0.33	N/A
B.NE 41b2c4 <_Z17advec_cell_kerneliiiiiiRN6clover8Buffer1DIdEES2_RNS_8Buffer2DIdEES5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_.omp_outlined.2+0x114>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Report Configuration

Code clean check

Workaround

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

Code clean check

Workaround

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