OV - exec - Loop 3091

0x4e6900 LDR	X25, [X6, X17,LSL #3]

0x4e6904 UBFM	X18, X25, #61, #60

0x4e6908 ADD	X30, X18, #8

0x4e690c LDR	X12, [X23, X18]

0x4e6910 ADD	X25, X23, X30

0x4e6914 LDR	X16, [X23, X30]

0x4e6918 CMP	X12, X16

0x4e691c B.GE	4e695c

0x4e6920 ORR	X13, XZR, X11

0x4e6924 STR	X0, [SP, #104]

  (3094) 0x4e6928 LDR	X11, [X21, X12,LSL #3]

  (3094) 0x4e692c UBFM	X11, X11, #61, #60

  (3094) 0x4e6930 LDR	X0, [X19, X11]

  (3094) 0x4e6934 CMP	X20, X0

  (3094) 0x4e6938 B.LE	4e6a84

  (3094) 0x4e693c STR	X3, [X19, X11]

  (3094) 0x4e6940 ADD	X12, X12, #1

  (3094) 0x4e6944 ADD	X3, X3, #1

  (3094) 0x4e6948 LDR	X16, [X25]

  (3094) 0x4e694c CMP	X16, X12

  (3094) 0x4e6950 B.GT	4e6928

0x4e6954 LDR	X0, [SP, #104]

0x4e6958 ORR	X11, XZR, X13

0x4e695c CBNZ	X5, 4e6acc

0x4e6960 LDR	X18, [X28, X15]

0x4e6964 ADD	X17, X17, #1

0x4e6968 CMP	X18, X17

0x4e696c B.GT	4e6900

  (3094) 0x4e6a84 ADD	X12, X12, #1

  (3094) 0x4e6a88 CMP	X12, X16

  (3094) 0x4e6a8c B.LT	4e6928

0x4e6a90 LDR	X0, [SP, #104]

0x4e6a94 ORR	X11, XZR, X13

0x4e6a98 B	4e695c

0x4e6acc LDR	X13, [X9, X18]

0x4e6ad0 ADD	X18, X9, X30

0x4e6ad4 LDR	X25, [X9, X30]

0x4e6ad8 CMP	X13, X25

0x4e6adc B.GE	4e6960

  (3092) 0x4e6ae0 LDR	X30, [X27, X13,LSL #3]

  (3092) 0x4e6ae4 LDR	X12, [X24, X30,LSL #3]

  (3092) 0x4e6ae8 ADD	X16, X22, X12

  (3092) 0x4e6aec UBFM	X30, X16, #61, #60

  (3092) 0x4e6af0 LDR	X12, [X19, X30]

  (3092) 0x4e6af4 CMP	X14, X12

  (3092) 0x4e6af8 B.LE	4e6b30

  (3093) 0x4e6afc STR	X11, [X19, X30]

  (3093) 0x4e6b00 ADD	X13, X13, #1

  (3093) 0x4e6b04 ADD	X11, X11, #1

  (3093) 0x4e6b08 LDR	X25, [X18]

  (3093) 0x4e6b0c CMP	X25, X13

  (3093) 0x4e6b10 B.LE	4e6960

  (3093) 0x4e6b14 LDR	X30, [X27, X13,LSL #3]

  (3093) 0x4e6b18 LDR	X12, [X24, X30,LSL #3]

  (3093) 0x4e6b1c ADD	X16, X22, X12

  (3093) 0x4e6b20 UBFM	X30, X16, #61, #60

  (3093) 0x4e6b24 LDR	X12, [X19, X30]

  (3093) 0x4e6b28 CMP	X14, X12

  (3093) 0x4e6b2c B.GT	4e6afc

  (3092) 0x4e6b30 ADD	X13, X13, #1

  (3092) 0x4e6b34 CMP	X25, X13

  (3092) 0x4e6b38 B.GT	4e6ae0

0x4e6b3c B	4e6960

/home/hbollore/qaas/qaas-runs/174-161-6712/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c: 187 - 231

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

187:       for (jj2 = A_diag_i[i1]; jj2 < A_diag_i[i1+1]; jj2++)

[...]

195:          for (jj3 = B_diag_i[i2]; jj3 < B_diag_i[i2+1]; jj3++)

[...]

205:             if (B_marker[i3] < jj_row_begin_diag)

206:             {

207:                B_marker[i3] = jj_count_diag;

208:                jj_count_diag++;

[...]

216:          if (num_cols_offd_B)

217:          {

218:             for (jj3 = B_offd_i[i2]; jj3 < B_offd_i[i2+1]; jj3++)

219:             {

220:                i3 = num_cols_diag_B+map_B_to_C[B_offd_j[jj3]];

[...]

228:                if (B_marker[i3] < jj_row_begin_offd)

229:                {

230:                   B_marker[i3] = jj_count_offd;

231:                   jj_count_offd++;

Coverage (%)	Name	Source Location	Module
►99.23+	gomp_thread_start	team.c:130	libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6

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	1.00
CQA speedup if fully vectorized	2.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.03
Bottlenecks	micro-operation queue,
Function	hypre_ParMatmul_RowSizes._omp_fn.0
Source	par_csr_matop.c:187-187,par_csr_matop.c:195-195,par_csr_matop.c:216-218
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	3.25
CQA cycles if no scalar integer	3.25
CQA cycles if FP arith vectorized	3.25
CQA cycles if fully vectorized	1.63
Front-end cycles	3.25
P0 cycles	3.00
P1 cycles	3.00
P2 cycles	1.92
P3 cycles	1.92
P4 cycles	1.83
P5 cycles	1.83
P6 cycles	1.75
P7 cycles	1.75
P8 cycles	0.00
P9 cycles	0.00
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	3.17
P13 cycles	2.83
P14 cycles	3.00
P15 cycles	0.50
P16 cycles	0.50
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	26.00
Nb uops	26.00
Nb loads	NA
Nb stores	1.00
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	22.15
Bytes prefetched	0.00
Bytes loaded	64.00
Bytes stored	8.00
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	NA
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	50.00
Vector-efficiency ratio load	50.00
Vector-efficiency ratio store	50.00
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	NA
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	50.00

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	2.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.03
Bottlenecks	micro-operation queue,
Function	hypre_ParMatmul_RowSizes._omp_fn.0
Source	par_csr_matop.c:187-187,par_csr_matop.c:195-195,par_csr_matop.c:216-218
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	3.25
CQA cycles if no scalar integer	3.25
CQA cycles if FP arith vectorized	3.25
CQA cycles if fully vectorized	1.63
Front-end cycles	3.25
P0 cycles	3.00
P1 cycles	3.00
P2 cycles	1.92
P3 cycles	1.92
P4 cycles	1.83
P5 cycles	1.83
P6 cycles	1.75
P7 cycles	1.75
P8 cycles	0.00
P9 cycles	0.00
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	3.17
P13 cycles	2.83
P14 cycles	3.00
P15 cycles	0.50
P16 cycles	0.50
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	26.00
Nb uops	26.00
Nb loads	NA
Nb stores	1.00
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	22.15
Bytes prefetched	0.00
Bytes loaded	64.00
Bytes stored	8.00
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	NA
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	50.00
Vector-efficiency ratio load	50.00
Vector-efficiency ratio store	50.00
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	NA
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	50.00

Path /

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

Function	hypre_ParMatmul_RowSizes._omp_fn.0
Source file and lines	par_csr_matop.c:187-231
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/174-161-6712/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c:187,195,205-208,216-220,228-231.

Analyzed code is defined in /home/hbollore/qaas/qaas-runs/174-161-6712/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c:187,195,216-218.

Warnings:

Non-innermost loop: analyzing only self part (ignoring child loops).
Ignoring paths for analysis
Too many paths. Rerun with max-paths=15
RecMII not computed since number of paths is unknown or > max_paths
Streams not analyzed since number of paths is unknown or > max_paths

Try to simplify control and/or increase the maximum number of paths per function/loop through the 'max-paths-nb' option.

This loop has 15 execution paths.

The presence of multiple execution paths is typically the main/first bottleneck.
Try to simplify control inside loop: ideally, try to remove all conditional expressions, for example by (if applicable):

hoisting them (moving them outside the loop)
turning them into conditional moves, MIN or MAX

Ex: if (x<0) x=0 => x = (x<0 ? 0 : x) (or MAX(0,x) after defining the corresponding macro)

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. 2 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 3.25 to 1.62 cycles (2.00x 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:
- recompile with fassociative-math (included in Ofast or ffast-math) 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)

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 64 bytes. The binary loop is storing 8 bytes.

General properties

nb instructions	26
loop length	104
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

3.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16
uops	3.00	3.00	1.92	1.92	1.83	1.83	1.75	1.75	0.00	0.00	0.00	0.00	3.17	2.83	3.00	0.50	0.50
cycles	3.00	3.00	1.92	1.92	1.83	1.83	1.75	1.75	0.00	0.00	0.00	0.00	3.17	2.83	3.00	0.50	0.50

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

Cycles summary

Front-end	3.25
Overall L1	3.25

Vectorization ratios

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

20% of peak load performance is reached (19.69 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (2.46 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 4e6900

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P12	P13	P14	P15	P16	Latency	Recip. throughput	Vectorization
LDR X25, [X6, X17,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
UBFM X18, X25, #61, #60	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
ADD X30, X18, #8	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
LDR X12, [X23, X18]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ADD X25, X23, X30	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
LDR X16, [X23, X30]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
CMP X12, X16	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	1	0.33	scal (50.0%)
B.GE 4e695c <hypre_ParMatmul_RowSizes._omp_fn.0+0x19c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X13, XZR, X11	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
STR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
LDR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ORR X11, XZR, X13	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	scal (50.0%)
CBNZ X5, 4e6acc <hypre_ParMatmul_RowSizes._omp_fn.0+0x30c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X18, [X28, X15]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X17, X17, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
CMP X18, X17	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	1	0.33	scal (50.0%)
B.GT 4e6900 <hypre_ParMatmul_RowSizes._omp_fn.0+0x140>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ORR X11, XZR, X13	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	scal (50.0%)
B 4e695c <hypre_ParMatmul_RowSizes._omp_fn.0+0x19c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X13, [X9, X18]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X18, X9, X30	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
LDR X25, [X9, X30]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CMP X13, X25	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	1	0.33	N/A
B.GE 4e6960 <hypre_ParMatmul_RowSizes._omp_fn.0+0x1a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
B 4e6960 <hypre_ParMatmul_RowSizes._omp_fn.0+0x1a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	hypre_ParMatmul_RowSizes._omp_fn.0
Source file and lines	par_csr_matop.c:187-231
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/174-161-6712/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c:187,195,205-208,216-220,228-231.

Analyzed code is defined in /home/hbollore/qaas/qaas-runs/174-161-6712/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c:187,195,216-218.

Warnings:

Non-innermost loop: analyzing only self part (ignoring child loops).
Ignoring paths for analysis
Too many paths. Rerun with max-paths=15
RecMII not computed since number of paths is unknown or > max_paths
Streams not analyzed since number of paths is unknown or > max_paths

Try to simplify control and/or increase the maximum number of paths per function/loop through the 'max-paths-nb' option.

This loop has 15 execution paths.

The presence of multiple execution paths is typically the main/first bottleneck.
Try to simplify control inside loop: ideally, try to remove all conditional expressions, for example by (if applicable):

hoisting them (moving them outside the loop)
turning them into conditional moves, MIN or MAX

Ex: if (x<0) x=0 => x = (x<0 ? 0 : x) (or MAX(0,x) after defining the corresponding macro)

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. 2 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 3.25 to 1.62 cycles (2.00x 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:
- recompile with fassociative-math (included in Ofast or ffast-math) 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)

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 64 bytes. The binary loop is storing 8 bytes.

General properties

nb instructions	26
loop length	104
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

3.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16
uops	3.00	3.00	1.92	1.92	1.83	1.83	1.75	1.75	0.00	0.00	0.00	0.00	3.17	2.83	3.00	0.50	0.50
cycles	3.00	3.00	1.92	1.92	1.83	1.83	1.75	1.75	0.00	0.00	0.00	0.00	3.17	2.83	3.00	0.50	0.50

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

Cycles summary

Front-end	3.25
Overall L1	3.25

Vectorization ratios

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

20% of peak load performance is reached (19.69 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (2.46 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 4e6900

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P12	P13	P14	P15	P16	Latency	Recip. throughput	Vectorization
LDR X25, [X6, X17,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
UBFM X18, X25, #61, #60	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
ADD X30, X18, #8	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
LDR X12, [X23, X18]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ADD X25, X23, X30	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
LDR X16, [X23, X30]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
CMP X12, X16	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	1	0.33	scal (50.0%)
B.GE 4e695c <hypre_ParMatmul_RowSizes._omp_fn.0+0x19c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X13, XZR, X11	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
STR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
LDR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ORR X11, XZR, X13	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	scal (50.0%)
CBNZ X5, 4e6acc <hypre_ParMatmul_RowSizes._omp_fn.0+0x30c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X18, [X28, X15]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X17, X17, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
CMP X18, X17	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	1	0.33	scal (50.0%)
B.GT 4e6900 <hypre_ParMatmul_RowSizes._omp_fn.0+0x140>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ORR X11, XZR, X13	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	scal (50.0%)
B 4e695c <hypre_ParMatmul_RowSizes._omp_fn.0+0x19c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X13, [X9, X18]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X18, X9, X30	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	1	0.17	N/A
LDR X25, [X9, X30]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CMP X13, X25	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	1	0.33	N/A
B.GE 4e6960 <hypre_ParMatmul_RowSizes._omp_fn.0+0x1a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
B 4e6960 <hypre_ParMatmul_RowSizes._omp_fn.0+0x1a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Report Configuration

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

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