OV - exec - Loop 581

0x469aa0 LD1D	{Z18.D}, P1/Z, [X26, X20,LSL #3]    [2]

0x469aa4 LD1D	{Z19.D}, P1/Z, [X22, X20,LSL #3]    [1]

0x469aa8 ADD	X20, X20, X19

0x469aac CMP	X5, X20

0x469ab0 FMINNM	Z6.D, P1/M, Z6.D, Z18.D

0x469ab4 FMINNM	Z17.D, P1/M, Z17.D, Z19.D

0x469ab8 FADD	Z7.D, Z18.D, Z7.D

0x469abc FADD	Z16.D, Z19.D, Z16.D

0x469ac0 B.NE	469aa0

/home/hbollore/qaas/qaas-runs/169-817-3176/intel/AMG/build/AMG/AMG/parcsr_ls/par_strength.c: 326 - 327

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

326:                row_scale = hypre_min(row_scale, A_diag_data[jA]);

327:                row_sum += A_diag_data[jA];

Coverage (%)	Name	Source Location	Module
○100.00	__kmp_invoke_microtask		libomp.so

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	1.78
Bottlenecks	P6, P7,
Function	.omp_outlined..2#0x469780
Source	par_strength.c:326-327
Source loop unroll info	unrolled by 8
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	8
CQA cycles	2.00
CQA cycles if no scalar integer	2.00
CQA cycles if FP arith vectorized	2.00
CQA cycles if fully vectorized	2.00
Front-end cycles	1.13
DIV/SQRT cycles	0.50
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.50
P3 cycles	0.50
P4 cycles	0.50
P5 cycles	2.00
P6 cycles	2.00
P7 cycles	1.00
P8 cycles	1.00
P9 cycles	1.00
P10 cycles	1.00
P11 cycles	0.00
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
Inter-iter dependencies cycles	2
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	9.00
Nb uops	9.00
Nb loads	NA
Nb stores	0.00
Nb stack references	0.00
FLOP/cycle	4.00
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	32.00
Bytes prefetched	0.00
Bytes loaded	64.00
Bytes stored	0.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	85.71
Vectorization ratio load	100.00
Vectorization ratio store	NA
Vectorization ratio mul	NA
Vectorization ratio add_sub	100.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	66.67
Vector-efficiency ratio all	89.29
Vector-efficiency ratio load	100.00
Vector-efficiency ratio store	NA
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	75.00

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	1.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.78
Bottlenecks	P6, P7,
Function	.omp_outlined..2#0x469780
Source	par_strength.c:326-327
Source loop unroll info	unrolled by 8
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	8
CQA cycles	2.00
CQA cycles if no scalar integer	2.00
CQA cycles if FP arith vectorized	2.00
CQA cycles if fully vectorized	2.00
Front-end cycles	1.13
DIV/SQRT cycles	0.50
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.50
P3 cycles	0.50
P4 cycles	0.50
P5 cycles	2.00
P6 cycles	2.00
P7 cycles	1.00
P8 cycles	1.00
P9 cycles	1.00
P10 cycles	1.00
P11 cycles	0.00
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
Inter-iter dependencies cycles	2
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	9.00
Nb uops	9.00
Nb loads	NA
Nb stores	0.00
Nb stack references	0.00
FLOP/cycle	4.00
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	32.00
Bytes prefetched	0.00
Bytes loaded	64.00
Bytes stored	0.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	85.71
Vectorization ratio load	100.00
Vectorization ratio store	NA
Vectorization ratio mul	NA
Vectorization ratio add_sub	100.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	66.67
Vector-efficiency ratio all	89.29
Vector-efficiency ratio load	100.00
Vector-efficiency ratio store	NA
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	75.00

Path /

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

Function	.omp_outlined..2#0x469780
Source file and lines	par_strength.c:326-327
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/169-817-3176/intel/AMG/build/AMG/AMG/parcsr_ls/par_strength.c:326-327.

The related source loop is unrolled by 8 (including vectorization).

gain
potential
hint
expert

Vectorization

Your loop is partially vectorized. 89% of vector register length is used (average across all VPU instructions).

Details

85% of VPU instructions are used in vector version (process two or more data elements in vector registers):

66% 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 64 bytes.

Arithmetic intensity

Arithmetic intensity is 0.12 FP operations per loaded or stored byte.

General properties

nb instructions	9
loop length	36
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

1.13 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	0.50	0.50	0.50	2.00	2.00	0.00	0.00	1.00	1.00	0.00	0.00	0.00
cycles	0.50	0.50	0.50	0.50	0.50	0.50	2.00	2.00	1.00	1.00	1.00	1.00	0.00	0.00	0.00

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	2.00

Cycles summary

Front-end	1.13
Data deps.	2.00
Overall L1	2.00

Vectorization ratios

INT

all	66%
load	100%
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	0%

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	100%

INT+FP

all	85%
load	100%
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	66%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 2.00 cycles. At this rate:

33% of peak load performance is reached (32.00 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 469aa0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P10	P11	Latency	Recip. throughput
LD1D {Z18.D}, P1/Z, [X26, X20,LSL #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z19.D}, P1/Z, [X22, X20,LSL #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
ADD X20, X20, X19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
CMP X5, X20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.33
FMINNM Z6.D, P1/M, Z6.D, Z18.D	1	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50
FMINNM Z17.D, P1/M, Z17.D, Z19.D	1	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50
FADD Z7.D, Z18.D, Z7.D	1	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50
FADD Z16.D, Z19.D, Z16.D	1	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50
B.NE 469aa0 <.omp_outlined..2+0x320>	1	0.50	0.50	0	0	0	0	0	0	0	0	1	0.50

Function	.omp_outlined..2#0x469780
Source file and lines	par_strength.c:326-327
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/169-817-3176/intel/AMG/build/AMG/AMG/parcsr_ls/par_strength.c:326-327.

The related source loop is unrolled by 8 (including vectorization).

gain
potential
hint
expert

Vectorization

Your loop is partially vectorized. 89% of vector register length is used (average across all VPU instructions).

Details

85% of VPU instructions are used in vector version (process two or more data elements in vector registers):

66% 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 64 bytes.

Arithmetic intensity

Arithmetic intensity is 0.12 FP operations per loaded or stored byte.

General properties

nb instructions	9
loop length	36
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

1.13 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	0.50	0.50	0.50	2.00	2.00	0.00	0.00	1.00	1.00	0.00	0.00	0.00
cycles	0.50	0.50	0.50	0.50	0.50	0.50	2.00	2.00	1.00	1.00	1.00	1.00	0.00	0.00	0.00

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	2.00

Cycles summary

Front-end	1.13
Data deps.	2.00
Overall L1	2.00

Vectorization ratios

INT

all	66%
load	100%
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	0%

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	100%

INT+FP

all	85%
load	100%
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	66%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 2.00 cycles. At this rate:

33% of peak load performance is reached (32.00 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 469aa0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P10	P11	Latency	Recip. throughput
LD1D {Z18.D}, P1/Z, [X26, X20,LSL #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z19.D}, P1/Z, [X22, X20,LSL #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
ADD X20, X20, X19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
CMP X5, X20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.33
FMINNM Z6.D, P1/M, Z6.D, Z18.D	1	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50
FMINNM Z17.D, P1/M, Z17.D, Z19.D	1	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50
FADD Z7.D, Z18.D, Z7.D	1	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50
FADD Z16.D, Z19.D, Z16.D	1	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50
B.NE 469aa0 <.omp_outlined..2+0x320>	1	0.50	0.50	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