OV - exec - Loop 531

	Loop Id: 531	Module: exec	Source: IJMatrix_parcsr.c:3383-3454 [...]	Coverage: 3.37%

0x492d84 LDR	D0, [X17, X24,LSL #3]

0x492d88 LDP	X5, X6, [X29, #984]

0x492d8c STR	D0, [X3]

0x492d90 ADD	X10, X10, #1

0x492d94 ADD	X24, X24, #1

0x492d98 CMP	X10, X20

0x492d9c B.EQ	493044

0x492da0 LDP	X3, X4, [X29, #1000]

0x492da4 LDR	X2, [X15, X24,LSL #3]

0x492da8 LDR	X3, [X3]

0x492dac LDR	X4, [X4]

0x492db0 CMP	X2, X3

0x492db4 CCMP	X2, X4, #0, #10

0x492db8 B.LE	492e20

0x492dbc LDR	X4, [X13, X28,LSL #3]

0x492dc0 CMP	X4, X8

0x492dc4 B.GE	492e00

0x492dc8 UBFM	X5, X4, #61, #60

0x492dcc SUB	X4, X8, X4

0x492dd0 ADD	X3, X1, X5

0x492dd4 ADD	X5, X0, X5

0x492dd8 HINT	#0

0x492ddc HINT	#0

  (533) 0x492de0 LDR	X6, [X5]

  (533) 0x492de4 CMP	X6, X2

  (533) 0x492de8 B.EQ	492d84

  (533) 0x492dec ADD	X3, X3, #8

  (533) 0x492df0 SUBS	X4, X4, #1

  (533) 0x492df4 ADD	X5, X5, #8

  (533) 0x492df8 B.NE	492de0

0x492dfc LDP	X5, X6, [X29, #984]

0x492e00 CMP	X21, X14

0x492e04 B.GE	49314c

0x492e08 LDR	D0, [X17, X24,LSL #3]

0x492e0c STR	X2, [X0, X21,LSL #3]

0x492e10 STR	D0, [X1, X21,LSL #3]

0x492e14 ADD	X21, X21, #1

0x492e18 B	492d90

0x492e20 LDR	X4, [X11, X28,LSL #3]

0x492e24 CMP	X4, X9

0x492e28 B.GE	492e60

0x492e2c UBFM	X5, X4, #61, #60

0x492e30 SUB	X4, X9, X4

0x492e34 ADD	X3, X18, X5

0x492e38 ADD	X5, X16, X5

0x492e3c HINT	#0

  (532) 0x492e40 LDR	X6, [X5]

  (532) 0x492e44 CMP	X6, X2

  (532) 0x492e48 B.EQ	492d84

  (532) 0x492e4c ADD	X3, X3, #8

  (532) 0x492e50 SUBS	X4, X4, #1

  (532) 0x492e54 ADD	X5, X5, #8

  (532) 0x492e58 B.NE	492e40

0x492e5c LDP	X5, X6, [X29, #984]

0x492e60 CMP	X26, X12

0x492e64 B.GE	4931c8

0x492e68 LDR	D0, [X17, X24,LSL #3]

0x492e6c STR	X2, [X16, X26,LSL #3]

0x492e70 STR	D0, [X18, X26,LSL #3]

0x492e74 ADD	X26, X26, #1

0x492e78 B	492d90

/home/hbollore/qaas-runs/171-233-5044/intel/AMG/build/AMG/AMG/IJ_mv/IJMatrix_parcsr.c: 3383 - 3454

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

3383:   	          for (i=0; i < n; i++)

3384: 	          {

3385: 	             if (cols[indx] < col_0 || cols[indx] > col_n)

3386:                      /* insert into offd */

3387: 	             {

3388: 	                for (j=offd_i[row_local]; j < offd_indx; j++)

3389: 	                {

3390: 		           if (offd_j[j] == cols[indx])

[...]

3399: 	                   if (cnt_offd < offd_space)

3400: 	                   {

3401: 	                      offd_j[cnt_offd] = cols[indx];

3402: 	                      offd_data[cnt_offd++] = values[indx];

[...]

3422: 	                for (j=diag_i[row_local]; j < diag_indx; j++)

3423: 	                {

3424: 		           if (diag_j[j] == cols[indx])

[...]

3433: 	                   if (cnt_diag < diag_space)

3434: 	                   {

3435: 	                      diag_j[cnt_diag] = cols[indx];

3436: 	                      diag_data[cnt_diag++] = values[indx];

[...]

3454: 	             indx++;

Coverage (%)	Name	Source Location	Module

Path /

Metric	Value
CQA speedup if no scalar integer	2.69
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	1.03
Bottlenecks	P10,
Function	.omp_outlined..21#0x4928b0
Source	IJMatrix_parcsr.c:3383-3385,IJMatrix_parcsr.c:3388-3388,IJMatrix_parcsr.c:3399-3402,IJMatrix_parcsr.c:3422-3422,IJMatrix_parcsr.c:3433-3436,IJMatrix_parcsr.c:3454-3454
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	5.83
CQA cycles if no scalar integer	2.17
CQA cycles if FP arith vectorized	5.83
CQA cycles if fully vectorized	1.46
Front-end cycles	5.50
DIV/SQRT cycles	4.00
P0 cycles	4.00
P1 cycles	4.75
P2 cycles	4.75
P3 cycles	4.75
P4 cycles	4.75
P5 cycles	1.50
P6 cycles	1.50
P7 cycles	0.00
P8 cycles	0.00
P9 cycles	5.83
P10 cycles	5.50
P11 cycles	5.67
P12 cycles	1.00
P13 cycles	1.00
P14 cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	47.00
Nb uops	44.00
Nb loads	NA
Nb stores	5.00
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	28.80
Bytes prefetched	0.00
Bytes loaded	128.00
Bytes stored	40.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
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	25.00
Vector-efficiency ratio store	25.00
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	2.69
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	1.03
Bottlenecks	P10,
Function	.omp_outlined..21#0x4928b0
Source	IJMatrix_parcsr.c:3383-3385,IJMatrix_parcsr.c:3388-3388,IJMatrix_parcsr.c:3399-3402,IJMatrix_parcsr.c:3422-3422,IJMatrix_parcsr.c:3433-3436,IJMatrix_parcsr.c:3454-3454
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	5.83
CQA cycles if no scalar integer	2.17
CQA cycles if FP arith vectorized	5.83
CQA cycles if fully vectorized	1.46
Front-end cycles	5.50
DIV/SQRT cycles	4.00
P0 cycles	4.00
P1 cycles	4.75
P2 cycles	4.75
P3 cycles	4.75
P4 cycles	4.75
P5 cycles	1.50
P6 cycles	1.50
P7 cycles	0.00
P8 cycles	0.00
P9 cycles	5.83
P10 cycles	5.50
P11 cycles	5.67
P12 cycles	1.00
P13 cycles	1.00
P14 cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	47.00
Nb uops	44.00
Nb loads	NA
Nb stores	5.00
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	28.80
Bytes prefetched	0.00
Bytes loaded	128.00
Bytes stored	40.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
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	25.00
Vector-efficiency ratio store	25.00
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	.omp_outlined..21#0x4928b0
Source file and lines	IJMatrix_parcsr.c:3383-3454
Module	exec

The loop is defined in /home/hbollore/qaas-runs/171-233-5044/intel/AMG/build/AMG/AMG/IJ_mv/IJMatrix_parcsr.c:3383-3390,3399-3402,3422-3424,3433-3436,3454.

Analyzed code is defined in /home/hbollore/qaas-runs/171-233-5044/intel/AMG/build/AMG/AMG/IJ_mv/IJMatrix_parcsr.c:3383-3385,3388,3399-3402,3422,3433-3436,3454.

Warnings:

Non-innermost loop: analyzing only self part (ignoring child loops).
Ignoring paths for analysis
Failed to get the number of paths
RecMII not computed since number of paths is unknown or > max_paths

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 5.83 to 2.17 cycles (2.69x 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. 4 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 5.83 to 1.46 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
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)

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 128 bytes. The binary loop is storing 40 bytes.

General properties

nb instructions	47
loop length	188
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

5.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	4.00	4.00	4.75	4.75	4.75	4.75	1.50	1.50	0.00	0.00	5.83	5.50	5.67	1.00	1.00
cycles	4.00	4.00	4.75	4.75	4.75	4.75	1.50	1.50	0.00	0.00	5.83	5.50	5.67	1.00	1.00

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

Cycles summary

Front-end	5.50
Overall L1	5.83

Vectorization ratios

all	0%
load	0%
store	0%
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 5.83 cycles. At this rate:

22% of peak load performance is reached (21.94 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
14% of peak store performance is reached (6.86 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 492da0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput
LDR D0, [X17, X24,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
LDP X5, X6, [X29, #984]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
STR D0, [X3]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
ADD X10, X10, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X24, X24, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X10, X20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.EQ 493044 <.omp_outlined..21+0x794>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDP X3, X4, [X29, #1000]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
LDR X2, [X15, X24,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X3, [X3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X4, [X4]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X2, X3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
CCMP X2, X4, #0, #10	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
B.LE 492e20 <.omp_outlined..21+0x570>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X4, [X13, X28,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X4, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.GE 492e00 <.omp_outlined..21+0x550>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
UBFM X5, X4, #61, #60	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
SUB X4, X8, X4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X3, X1, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X5, X0, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
HINT #0
HINT #0
LDP X5, X6, [X29, #984]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
CMP X21, X14	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.GE 49314c <.omp_outlined..21+0x89c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR D0, [X17, X24,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
STR X2, [X0, X21,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR D0, [X1, X21,LSL #3]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
ADD X21, X21, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
B 492d90 <.omp_outlined..21+0x4e0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X4, [X11, X28,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X4, X9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.GE 492e60 <.omp_outlined..21+0x5b0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
UBFM X5, X4, #61, #60	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
SUB X4, X9, X4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X3, X18, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X5, X16, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
HINT #0
LDP X5, X6, [X29, #984]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
CMP X26, X12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.GE 4931c8 <.omp_outlined..21+0x918>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR D0, [X17, X24,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
STR X2, [X16, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR D0, [X18, X26,LSL #3]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
ADD X26, X26, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
B 492d90 <.omp_outlined..21+0x4e0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50

Function	.omp_outlined..21#0x4928b0
Source file and lines	IJMatrix_parcsr.c:3383-3454
Module	exec

Non-innermost loop: analyzing only self part (ignoring child loops).
Ignoring paths for analysis
Failed to get the number of paths
RecMII not computed since number of paths is unknown or > max_paths

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 5.83 to 2.17 cycles (2.69x 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. 4 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 5.83 to 1.46 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
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)

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 128 bytes. The binary loop is storing 40 bytes.

General properties

nb instructions	47
loop length	188
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

5.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	4.00	4.00	4.75	4.75	4.75	4.75	1.50	1.50	0.00	0.00	5.83	5.50	5.67	1.00	1.00
cycles	4.00	4.00	4.75	4.75	4.75	4.75	1.50	1.50	0.00	0.00	5.83	5.50	5.67	1.00	1.00

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

Cycles summary

Front-end	5.50
Overall L1	5.83

Vectorization ratios

all	0%
load	0%
store	0%
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 5.83 cycles. At this rate:

22% of peak load performance is reached (21.94 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
14% of peak store performance is reached (6.86 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 492da0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput
LDR D0, [X17, X24,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
LDP X5, X6, [X29, #984]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
STR D0, [X3]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
ADD X10, X10, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X24, X24, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X10, X20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.EQ 493044 <.omp_outlined..21+0x794>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDP X3, X4, [X29, #1000]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
LDR X2, [X15, X24,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X3, [X3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X4, [X4]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X2, X3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
CCMP X2, X4, #0, #10	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
B.LE 492e20 <.omp_outlined..21+0x570>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X4, [X13, X28,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X4, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.GE 492e00 <.omp_outlined..21+0x550>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
UBFM X5, X4, #61, #60	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
SUB X4, X8, X4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X3, X1, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X5, X0, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
HINT #0
HINT #0
LDP X5, X6, [X29, #984]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
CMP X21, X14	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.GE 49314c <.omp_outlined..21+0x89c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR D0, [X17, X24,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
STR X2, [X0, X21,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR D0, [X1, X21,LSL #3]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
ADD X21, X21, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
B 492d90 <.omp_outlined..21+0x4e0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X4, [X11, X28,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X4, X9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.GE 492e60 <.omp_outlined..21+0x5b0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
UBFM X5, X4, #61, #60	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
SUB X4, X9, X4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X3, X18, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X5, X16, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
HINT #0
LDP X5, X6, [X29, #984]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
CMP X26, X12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.GE 4931c8 <.omp_outlined..21+0x918>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR D0, [X17, X24,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
STR X2, [X16, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR D0, [X18, X26,LSL #3]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
ADD X26, X26, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
B 492d90 <.omp_outlined..21+0x4e0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50

Report Configuration

Code clean check

Workaround

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

Code clean check

Workaround

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