OV - exec - Loop 16

	Loop Id: 16	Module: exec	Source: cg.cpp:86-90	Coverage: 37.18%

0x4027ec ADD	X25, X7, X2

0x4027f0 LDR	D5, [X5, X2]    [1]

0x4027f4 LDR	D6, [X3, X2]    [7]

0x4027f8 LDP	D3, D2, [X25, #1016]    [6]

0x4027fc ADD	X25, X20, X2

0x402800 LDUR	D7, [X25, #504]    [5]

0x402804 FADD	D16, D5, D6

0x402808 LDR	D17, [X4, X2]    [8]

0x40280c SUBS	X24, X24, #1

0x402810 FADD	D4, D2, D3

0x402814 FMUL	D3, D3, D7

0x402818 LDR	D7, [X25, #8]    [5]

0x40281c FADD	D4, D4, D1

0x402820 FMADD	D2, D2, D7, D3

0x402824 FMUL	D3, D6, D17

0x402828 LDR	D6, [X25]    [5]

0x40282c LDR	D7, [X6, X2]    [2]

0x402830 FADD	D4, D4, D16

0x402834 FMADD	D3, D5, D7, D3

0x402838 FNMSUB	D2, D4, D6, D2

0x40283c FSUB	D2, D2, S3

0x402840 STR	D2, [X23, X2]    [4]

0x402844 ADD	X2, X2, #8

0x402848 LDR	D3, [X25]    [5]

0x40284c FMADD	D0, D2, D3, D0

0x402850 STR	D0, [SP, #24]    [3]

0x402854 B.NE	4027ec

/home/hbollore/qaas/qaas-runs/174-118-5752/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp: 86 - 90

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

86:     for (int kk = halo_depth; kk < x - halo_depth; ++kk) {

87:       const int index = kk + jj * x;

88:       const double smvp = tealeaf_SMVP(p);

89:       w[index] = smvp;

90:       pw_temp += w[index] * p[index];

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	1.05
CQA speedup if fully vectorized	1.83
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.09
Bottlenecks	P12, P13, P14,
Function	cg_calc_w(int, int, int, double, double const, double, double const, double const*) [clone .omp_outlined]
Source	cg.cpp:86-90
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	3.67
CQA cycles if no scalar integer	3.67
CQA cycles if FP arith vectorized	3.50
CQA cycles if fully vectorized	2.00
Front-end cycles	3.38
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.75
P3 cycles	0.75
P4 cycles	0.67
P5 cycles	0.67
P6 cycles	0.58
P7 cycles	0.58
P8 cycles	3.25
P9 cycles	3.25
P10 cycles	3.25
P11 cycles	3.25
P12 cycles	3.67
P13 cycles	3.67
P14 cycles	3.67
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	2
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	27.00
Nb uops	27.00
Nb loads	NA
Nb stores	2.00
Nb stack references	0.00
FLOP/cycle	4.09
Nb FLOP add-sub	5.00
Nb FLOP mul	2.00
Nb FLOP fma	4.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	26.18
Bytes prefetched	0.00
Bytes loaded	80.00
Bytes stored	16.00
Stride 0	1.00
Stride 1	5.00
Stride n	0.00
Stride unknown	2.00
Stride indirect	0.00
Vectorization ratio all	4.35
Vectorization ratio load	11.11
Vectorization ratio store	0.00
Vectorization ratio mul	0.00
Vectorization ratio add_sub	0.00
Vectorization ratio fma	0.00
Vectorization ratio div_sqrt	NA
Vectorization ratio other	NA
Vector-efficiency ratio all	52.17
Vector-efficiency ratio load	55.56
Vector-efficiency ratio store	50.00
Vector-efficiency ratio mul	50.00
Vector-efficiency ratio add_sub	50.00
Vector-efficiency ratio fma	50.00
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	1.05
CQA speedup if fully vectorized	1.83
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.09
Bottlenecks	P12, P13, P14,
Function	cg_calc_w(int, int, int, double, double const, double, double const, double const*) [clone .omp_outlined]
Source	cg.cpp:86-90
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	3.67
CQA cycles if no scalar integer	3.67
CQA cycles if FP arith vectorized	3.50
CQA cycles if fully vectorized	2.00
Front-end cycles	3.38
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.75
P3 cycles	0.75
P4 cycles	0.67
P5 cycles	0.67
P6 cycles	0.58
P7 cycles	0.58
P8 cycles	3.25
P9 cycles	3.25
P10 cycles	3.25
P11 cycles	3.25
P12 cycles	3.67
P13 cycles	3.67
P14 cycles	3.67
P15 cycles	0.00
P16 cycles	0.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	2
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	27.00
Nb uops	27.00
Nb loads	NA
Nb stores	2.00
Nb stack references	0.00
FLOP/cycle	4.09
Nb FLOP add-sub	5.00
Nb FLOP mul	2.00
Nb FLOP fma	4.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	26.18
Bytes prefetched	0.00
Bytes loaded	80.00
Bytes stored	16.00
Stride 0	1.00
Stride 1	5.00
Stride n	0.00
Stride unknown	2.00
Stride indirect	0.00
Vectorization ratio all	4.35
Vectorization ratio load	11.11
Vectorization ratio store	0.00
Vectorization ratio mul	0.00
Vectorization ratio add_sub	0.00
Vectorization ratio fma	0.00
Vectorization ratio div_sqrt	NA
Vectorization ratio other	NA
Vector-efficiency ratio all	52.17
Vector-efficiency ratio load	55.56
Vector-efficiency ratio store	50.00
Vector-efficiency ratio mul	50.00
Vector-efficiency ratio add_sub	50.00
Vector-efficiency ratio fma	50.00
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	cg_calc_w(int, int, int, double, double const, double, double const, double const*) [clone .omp_outlined]
Source file and lines	cg.cpp:86-90
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/174-118-5752/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp:86-90.

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

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. Only 52% of vector register length is used (average across all VPU instructions). By fully vectorizing your loop, you can lower the cost of an iteration from 3.67 to 2.00 cycles (1.83x speedup).

Details

4% of VPU instructions are used in vector version (process two or more data elements in vector registers):

11% of VPU loads are used in vector version.
0% of VPU stores are used in vector version.
0% of VPU addition or subtraction instructions are used in vector version.
0% of VPU multiply instructions are used in vector version.
0% of VPU fused multiply-add instructions are used in vector version.

Since your execution units are vector units, only a fully vectorized loop can use their full power.

Workaround

Try another compiler or update/tune your current one:
- recompile with ffast-math (included in Ofast) 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: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop streams arrays of structures (AoS), try to use structures of arrays instead (SoA): for(i) a[i].x = b[i].x; (slow, non stride 1) => for(i) a.x[i] = b.x[i]; (fast, stride 1)

FMA

Detected 4 FMA (fused multiply-add) operations. Presence of both ADD/SUB and MUL operations.

Workaround

Try to change order in which elements are evaluated (using parentheses) in arithmetic expressions containing both ADD/SUB and MUL operations to enable your compiler to generate FMA instructions wherever possible. For instance a + b*c is a valid FMA (MUL then ADD). However (a+b)* c cannot be translated into an FMA (ADD then MUL).

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

The binary loop is composed of 15 FP arithmetical operations:

9: addition or subtraction (4 inside FMA instructions)
6: multiply (4 inside FMA instructions)

The binary loop is loading 80 bytes. The binary loop is storing 16 bytes.

Arithmetic intensity

Arithmetic intensity is 0.16 FP operations per loaded or stored byte.

General properties

nb instructions	27
loop length	108
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

3.38 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	0.75	0.75	0.67	0.67	0.58	0.58	3.25	3.25	3.25	3.25	3.67	3.67	3.67	0.00	0.00
cycles	0.50	0.50	0.75	0.75	0.67	0.67	0.58	0.58	3.25	3.25	3.25	3.25	3.67	3.67	3.67	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	NA
Longest recurrence chain latency (RecMII)	2.00

Cycles summary

Front-end	3.38
Data deps.	2.00
Overall L1	3.67

Vectorization ratios

INT

all	8%
load	11%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
other	NA (no other vectorizable/vectorized instructions)

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

INT+FP

all	4%
load	11%
store	0%
mul	0%
add-sub	0%
fma	0%
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 3.67 cycles. At this rate:

22% of peak load performance is reached (21.82 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
9% of peak store performance is reached (4.36 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 4027ec

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
ADD X25, X7, X2	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
LDR D5, [X5, X2]	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 D6, [X3, X2]	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%)
LDP D3, D2, [X25, #1016]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (100.0%)
ADD X25, X20, X2	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
LDUR D7, [X25, #504]	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%)
FADD D16, D5, D6	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (50.0%)
LDR D17, [X4, X2]	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%)
SUBS X24, X24, #1	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	1	0.33	scal (50.0%)
FADD D4, D2, D3	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (50.0%)
FMUL D3, D3, D7	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (50.0%)
LDR D7, [X25, #8]	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%)
FADD D4, D4, D1	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (50.0%)
FMADD D2, D2, D7, D3	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (50.0%)
FMUL D3, D6, D17	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (50.0%)
LDR D6, [X25]	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 D7, [X6, X2]	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%)
FADD D4, D4, D16	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (50.0%)
FMADD D3, D5, D7, D3	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (50.0%)
FNMSUB D2, D4, D6, D2	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (50.0%)
FSUB D2, D2, S3	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (50.0%)
STR D2, [X23, X2]	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 X2, X2, #8	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
LDR D3, [X25]	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%)
FMADD D0, D2, D3, D0	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (50.0%)
STR D0, [SP, #24]	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%)
B.NE 4027ec <_Z9cg_calc_wiiiPdPKdS_S1_S1_.omp_outlined+0x134>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	cg_calc_w(int, int, int, double, double const, double, double const, double const*) [clone .omp_outlined]
Source file and lines	cg.cpp:86-90
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/174-118-5752/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp:86-90.

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

gain
potential
hint
expert

Vectorization

Details

4% of VPU instructions are used in vector version (process two or more data elements in vector registers):

11% of VPU loads are used in vector version.
0% of VPU stores are used in vector version.
0% of VPU addition or subtraction instructions are used in vector version.
0% of VPU multiply instructions are used in vector version.
0% of VPU fused multiply-add instructions are used in vector version.

Since your execution units are vector units, only a fully vectorized loop can use their full power.

Workaround

Try another compiler or update/tune your current one:
- recompile with ffast-math (included in Ofast) 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: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop streams arrays of structures (AoS), try to use structures of arrays instead (SoA): for(i) a[i].x = b[i].x; (slow, non stride 1) => for(i) a.x[i] = b.x[i]; (fast, stride 1)

FMA

Detected 4 FMA (fused multiply-add) operations. Presence of both ADD/SUB and MUL operations.

Workaround

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

The binary loop is composed of 15 FP arithmetical operations:

9: addition or subtraction (4 inside FMA instructions)
6: multiply (4 inside FMA instructions)

The binary loop is loading 80 bytes. The binary loop is storing 16 bytes.

Arithmetic intensity

Arithmetic intensity is 0.16 FP operations per loaded or stored byte.

General properties

nb instructions	27
loop length	108
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

3.38 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	0.75	0.75	0.67	0.67	0.58	0.58	3.25	3.25	3.25	3.25	3.67	3.67	3.67	0.00	0.00
cycles	0.50	0.50	0.75	0.75	0.67	0.67	0.58	0.58	3.25	3.25	3.25	3.25	3.67	3.67	3.67	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	NA
Longest recurrence chain latency (RecMII)	2.00

Cycles summary

Front-end	3.38
Data deps.	2.00
Overall L1	3.67

Vectorization ratios

INT

all	8%
load	11%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
other	NA (no other vectorizable/vectorized instructions)

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

INT+FP

all	4%
load	11%
store	0%
mul	0%
add-sub	0%
fma	0%
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 3.67 cycles. At this rate:

22% of peak load performance is reached (21.82 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
9% of peak store performance is reached (4.36 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 4027ec

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
ADD X25, X7, X2	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
LDR D5, [X5, X2]	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 D6, [X3, X2]	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%)
LDP D3, D2, [X25, #1016]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	6	0.33	vect (100.0%)
ADD X25, X20, X2	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
LDUR D7, [X25, #504]	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%)
FADD D16, D5, D6	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (50.0%)
LDR D17, [X4, X2]	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%)
SUBS X24, X24, #1	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	1	0.33	scal (50.0%)
FADD D4, D2, D3	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (50.0%)
FMUL D3, D3, D7	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (50.0%)
LDR D7, [X25, #8]	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%)
FADD D4, D4, D1	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (50.0%)
FMADD D2, D2, D7, D3	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (50.0%)
FMUL D3, D6, D17	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	3	0.25	scal (50.0%)
LDR D6, [X25]	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 D7, [X6, X2]	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%)
FADD D4, D4, D16	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (50.0%)
FMADD D3, D5, D7, D3	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (50.0%)
FNMSUB D2, D4, D6, D2	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (50.0%)
FSUB D2, D2, S3	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	2	0.25	scal (50.0%)
STR D2, [X23, X2]	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 X2, X2, #8	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
LDR D3, [X25]	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%)
FMADD D0, D2, D3, D0	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	4	0.25	scal (50.0%)
STR D0, [SP, #24]	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%)
B.NE 4027ec <_Z9cg_calc_wiiiPdPKdS_S1_S1_.omp_outlined+0x134>	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

Vectorization

Details

Workaround

FMA

Workaround

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

Workaround

FMA

Workaround

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