OV - exec - Loop 2175

	Loop Id: 2175	Module: exec	Source: ams.c:3382-3401 [...]	Coverage: 0.03%

0x4c06dc MOVI	D19, #0

0x4c06e0 CBNZ	X4, 4c1498

0x4c06e4 ADD	X12, X12, #1

0x4c06e8 CMP	X26, X12

0x4c06ec B.EQ	4bfa0c

0x4c06f0 LDR	X14, [X7, X12,LSL #3]

0x4c06f4 ORR	X3, XZR, X12

0x4c06f8 LDR	X0, [X24, X12,LSL #3]

0x4c06fc STR	XZR, [X19, X12,LSL #3]

0x4c0700 CMP	X14, X0

0x4c0704 B.LE	4c06dc

0x4c0708 ADD	X30, X21, X14,LSL #3

0x4c070c UBFM	X14, X0, #61, #60

0x4c0710 MOVI	D19, #0

0x4c0714 ADD	X0, X21, X0,LSL #3

0x4c0718 SUB	X16, X30, X0

0x4c071c SUB	X2, X16, #8

0x4c0720 UBFM	X18, X2, #3, #63

0x4c0724 ADD	X17, X18, #1

0x4c0728 ANDS	X11, X17, #0x7

0x4c072c B.EQ	4c07dc

0x4c0730 CMP	X11, #1

0x4c0734 B.EQ	4c07c4

0x4c0738 CMP	X11, #2

0x4c073c B.EQ	4c07b4

0x4c0740 CMP	X11, #3

0x4c0744 B.EQ	4c07a4

0x4c0748 CMP	X11, #4

0x4c074c B.EQ	4c0794

0x4c0750 CMP	X11, #5

0x4c0754 B.EQ	4c0784

0x4c0758 CMP	X11, #6

0x4c075c B.EQ	4c0774

0x4c0760 LDR	D26, [X21, X14]

0x4c0764 ADD	X0, X0, #8

0x4c0768 FABS	D29, D26

0x4c076c FADD	D19, D19, D29

0x4c0770 STR	D19, [X19, X3,LSL #3]

0x4c0774 LDR	D6, [X0], #8

0x4c0778 FABS	D0, D6

0x4c077c FADD	D19, D19, D0

0x4c0780 STR	D19, [X19, X3,LSL #3]

0x4c0784 LDR	D1, [X0], #8

0x4c0788 FABS	D31, D1

0x4c078c FADD	D19, D19, D31

0x4c0790 STR	D19, [X19, X3,LSL #3]

0x4c0794 LDR	D2, [X0], #8

0x4c0798 FABS	D3, D2

0x4c079c FADD	D19, D19, D3

0x4c07a0 STR	D19, [X19, X3,LSL #3]

0x4c07a4 LDR	D4, [X0], #8

0x4c07a8 FABS	D5, D4

0x4c07ac FADD	D19, D19, D5

0x4c07b0 STR	D19, [X19, X3,LSL #3]

0x4c07b4 LDR	D7, [X0], #8

0x4c07b8 FABS	D16, D7

0x4c07bc FADD	D19, D19, D16

0x4c07c0 STR	D19, [X19, X3,LSL #3]

0x4c07c4 LDR	D17, [X0], #8

0x4c07c8 FABS	D18, D17

0x4c07cc FADD	D19, D19, D18

0x4c07d0 STR	D19, [X19, X3,LSL #3]

0x4c07d4 CMP	X30, X0

0x4c07d8 B.EQ	4c06e0

  (2177) 0x4c07dc ORR	X14, XZR, X0

  (2177) 0x4c07e0 ADD	X0, X0, #64

  (2177) 0x4c07e4 LDR	D20, [X14], #8

  (2177) 0x4c07e8 FABS	D21, D20

  (2177) 0x4c07ec FADD	D27, D19, D21

  (2177) 0x4c07f0 STR	D27, [X19, X3,LSL #3]

  (2177) 0x4c07f4 LDUR	D28, [X0, #456]

  (2177) 0x4c07f8 FABS	D25, D28

  (2177) 0x4c07fc FADD	D22, D27, D25

  (2177) 0x4c0800 STR	D22, [X19, X3,LSL #3]

  (2177) 0x4c0804 LDR	D23, [X14, #8]

  (2177) 0x4c0808 FABS	D30, D23

  (2177) 0x4c080c FADD	D24, D22, D30

  (2177) 0x4c0810 STR	D24, [X19, X3,LSL #3]

  (2177) 0x4c0814 LDUR	D26, [X0, #472]

  (2177) 0x4c0818 FABS	D29, D26

  (2177) 0x4c081c FADD	D6, D24, D29

  (2177) 0x4c0820 STR	D6, [X19, X3,LSL #3]

  (2177) 0x4c0824 LDUR	D0, [X0, #480]

  (2177) 0x4c0828 FABS	D1, D0

  (2177) 0x4c082c FADD	D31, D6, D1

  (2177) 0x4c0830 STR	D31, [X19, X3,LSL #3]

  (2177) 0x4c0834 LDUR	D2, [X0, #488]

  (2177) 0x4c0838 FABS	D3, D2

  (2177) 0x4c083c FADD	D4, D31, D3

  (2177) 0x4c0840 STR	D4, [X19, X3,LSL #3]

  (2177) 0x4c0844 LDUR	D5, [X0, #496]

  (2177) 0x4c0848 FABS	D7, D5

  (2177) 0x4c084c FADD	D16, D4, D7

  (2177) 0x4c0850 STR	D16, [X19, X3,LSL #3]

  (2177) 0x4c0854 LDUR	D17, [X0, #504]

  (2177) 0x4c0858 FABS	D18, D17

  (2177) 0x4c085c FADD	D19, D16, D18

  (2177) 0x4c0860 STR	D19, [X19, X3,LSL #3]

  (2177) 0x4c0864 CMP	X30, X0

  (2177) 0x4c0868 B.NE	4c07dc

0x4c086c B	4c06e0

0x4c1498 LDR	X30, [X10, X3,LSL #3]

0x4c149c LDR	X16, [X15, X3,LSL #3]

0x4c14a0 CMP	X16, X30

0x4c14a4 B.LE	4c06e4

0x4c14a8 ADD	X18, X6, X16,LSL #3

0x4c14ac UBFM	X17, X30, #61, #60

0x4c14b0 ADD	X0, X6, X30,LSL #3

0x4c14b4 SUB	X2, X18, X0

0x4c14b8 SUB	X11, X2, #8

0x4c14bc UBFM	X14, X11, #3, #63

0x4c14c0 ADD	X30, X14, #1

0x4c14c4 ANDS	X16, X30, #0x7

0x4c14c8 B.EQ	4c1564

0x4c14cc CMP	X16, #1

0x4c14d0 B.EQ	4c154c

0x4c14d4 CMP	X16, #2

0x4c14d8 B.EQ	4c153c

0x4c14dc CMP	X16, #3

0x4c14e0 B.EQ	4c152c

0x4c14e4 CMP	X16, #4

0x4c14e8 B.EQ	4c151c

0x4c14ec CMP	X16, #5

0x4c14f0 B.EQ	4c150c

0x4c14f4 CMP	X16, #6

0x4c14f8 B.NE	4c3a84

0x4c14fc LDR	D27, [X0], #8

0x4c1500 FABS	D28, D27

0x4c1504 FADD	D19, D19, D28

0x4c1508 STR	D19, [X19, X3,LSL #3]

0x4c150c LDR	D25, [X0], #8

0x4c1510 FABS	D22, D25

0x4c1514 FADD	D19, D19, D22

0x4c1518 STR	D19, [X19, X3,LSL #3]

0x4c151c LDR	D23, [X0], #8

0x4c1520 FABS	D30, D23

0x4c1524 FADD	D19, D19, D30

0x4c1528 STR	D19, [X19, X3,LSL #3]

0x4c152c LDR	D24, [X0], #8

0x4c1530 FABS	D26, D24

0x4c1534 FADD	D19, D19, D26

0x4c1538 STR	D19, [X19, X3,LSL #3]

0x4c153c LDR	D29, [X0], #8

0x4c1540 FABS	D6, D29

0x4c1544 FADD	D19, D19, D6

0x4c1548 STR	D19, [X19, X3,LSL #3]

0x4c154c LDR	D0, [X0], #8

0x4c1550 FABS	D1, D0

0x4c1554 FADD	D19, D19, D1

0x4c1558 STR	D19, [X19, X3,LSL #3]

0x4c155c CMP	X0, X18

0x4c1560 B.EQ	4c06e4

  (2176) 0x4c1564 ORR	X17, XZR, X0

  (2176) 0x4c1568 ADD	X0, X0, #64

  (2176) 0x4c156c LDR	D31, [X17], #8

  (2176) 0x4c1570 FABS	D2, D31

  (2176) 0x4c1574 FADD	D19, D19, D2

  (2176) 0x4c1578 STR	D19, [X19, X3,LSL #3]

  (2176) 0x4c157c LDUR	D3, [X0, #456]

  (2176) 0x4c1580 FABS	D4, D3

  (2176) 0x4c1584 FADD	D5, D19, D4

  (2176) 0x4c1588 STR	D5, [X19, X3,LSL #3]

  (2176) 0x4c158c LDR	D7, [X17, #8]

  (2176) 0x4c1590 FABS	D16, D7

  (2176) 0x4c1594 FADD	D17, D5, D16

  (2176) 0x4c1598 STR	D17, [X19, X3,LSL #3]

  (2176) 0x4c159c LDUR	D18, [X0, #472]

  (2176) 0x4c15a0 FABS	D20, D18

  (2176) 0x4c15a4 FADD	D21, D17, D20

  (2176) 0x4c15a8 STR	D21, [X19, X3,LSL #3]

  (2176) 0x4c15ac LDUR	D27, [X0, #480]

  (2176) 0x4c15b0 FABS	D28, D27

  (2176) 0x4c15b4 FADD	D25, D21, D28

  (2176) 0x4c15b8 STR	D25, [X19, X3,LSL #3]

  (2176) 0x4c15bc LDUR	D22, [X0, #488]

  (2176) 0x4c15c0 FABS	D23, D22

  (2176) 0x4c15c4 FADD	D24, D25, D23

  (2176) 0x4c15c8 STR	D24, [X19, X3,LSL #3]

  (2176) 0x4c15cc LDUR	D30, [X0, #496]

  (2176) 0x4c15d0 FABS	D26, D30

  (2176) 0x4c15d4 FADD	D29, D24, D26

  (2176) 0x4c15d8 STR	D29, [X19, X3,LSL #3]

  (2176) 0x4c15dc LDUR	D6, [X0, #504]

  (2176) 0x4c15e0 FABS	D0, D6

  (2176) 0x4c15e4 FADD	D19, D29, D0

  (2176) 0x4c15e8 STR	D19, [X19, X3,LSL #3]

  (2176) 0x4c15ec CMP	X0, X18

  (2176) 0x4c15f0 B.NE	4c1564

0x4c15f4 B	4c06e4

0x4c3a84 LDR	D20, [X6, X17]

0x4c3a88 ADD	X0, X0, #8

0x4c3a8c FABS	D21, D20

0x4c3a90 FADD	D19, D19, D21

0x4c3a94 STR	D19, [X19, X3,LSL #3]

0x4c3a98 B	4c14fc

/home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c: 3382 - 3401

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

3382:          for (i = ns; i < ne; i++)

3383:          {

3384:             l1_norm[i] = 0.0;

3385:             if (cf_marker == NULL)

3386:             {

3387:                /* Add the l1 norm of the diag part of the ith row */

3388:                for (j = A_diag_I[i]; j < A_diag_I[i+1]; j++)

3389:                   l1_norm[i] += fabs(A_diag_data[j]);

3390:                /* Add the l1 norm of the offd part of the ith row */

3391:                if (num_cols_offd)

3392:                {

3393:                   for (j = A_offd_I[i]; j < A_offd_I[i+1]; j++)

3394:                      l1_norm[i] += fabs(A_offd_data[j]);

[...]

3401:                for (j = A_diag_I[i]; j < A_diag_I[i+1]; j++)

Coverage (%)	Name	Module
►99.05+	omp_fulfill_event	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	1.65
CQA speedup if FP arith vectorized	1.45
CQA speedup if fully vectorized	4.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.26
Bottlenecks	micro-operation queue,
Function	hypre_ParCSRComputeL1NormsThreads._omp_fn.0
Source	ams.c:3382-3384,ams.c:3388-3394,ams.c:3401-3401
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	15.38
CQA cycles if no scalar integer	9.33
CQA cycles if FP arith vectorized	10.59
CQA cycles if fully vectorized	3.84
Front-end cycles	15.38
P0 cycles	11.50
P1 cycles	11.50
P2 cycles	12.25
P3 cycles	12.25
P4 cycles	12.25
P5 cycles	12.25
P6 cycles	11.00
P7 cycles	11.00
P8 cycles	11.00
P9 cycles	11.00
P10 cycles	11.17
P11 cycles	10.83
P12 cycles	11.00
P13 cycles	0.50
P14 cycles	0.50
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	123.00
Nb uops	123.00
Nb loads	NA
Nb stores	15.00
Nb stack references	0.00
FLOP/cycle	0.91
Nb FLOP add-sub	14.00
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	0.00
Bytes prefetched	0.00
Bytes loaded	0.00
Bytes stored	0.00
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	1.65
CQA speedup if FP arith vectorized	1.45
CQA speedup if fully vectorized	4.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.26
Bottlenecks	micro-operation queue,
Function	hypre_ParCSRComputeL1NormsThreads._omp_fn.0
Source	ams.c:3382-3384,ams.c:3388-3394,ams.c:3401-3401
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	15.38
CQA cycles if no scalar integer	9.33
CQA cycles if FP arith vectorized	10.59
CQA cycles if fully vectorized	3.84
Front-end cycles	15.38
P0 cycles	11.50
P1 cycles	11.50
P2 cycles	12.25
P3 cycles	12.25
P4 cycles	12.25
P5 cycles	12.25
P6 cycles	11.00
P7 cycles	11.00
P8 cycles	11.00
P9 cycles	11.00
P10 cycles	11.17
P11 cycles	10.83
P12 cycles	11.00
P13 cycles	0.50
P14 cycles	0.50
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	123.00
Nb uops	123.00
Nb loads	NA
Nb stores	15.00
Nb stack references	0.00
FLOP/cycle	0.91
Nb FLOP add-sub	14.00
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	0.00
Bytes prefetched	0.00
Bytes loaded	0.00
Bytes stored	0.00
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	hypre_ParCSRComputeL1NormsThreads._omp_fn.0
Source file and lines	ams.c:3382-3401
Module	exec

The loop is defined in /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c:3382-3394,3401.

Analyzed code is defined in /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c:3382-3384,3388-3394,3401.

Warnings:

Non-innermost loop: analyzing only self part (ignoring child loops).
Ignoring paths for analysis
Too many paths. If you really need to analyze all of the 272 paths individually, rerun with max-paths=272
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 272 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

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 15.38 to 9.33 cycles (1.65x 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 15.38 to 3.84 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:
- 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 is composed of 14 FP arithmetical operations:

14: addition or subtraction

The binary loop does not load or store any data.

General properties

nb instructions	123
nb uops	123
loop length	492
used w registers	0
used x registers	20
used b registers	0
used h registers	0
used s registers	0
used d registers	24
used q registers	0
used v registers	0
used z registers	0
nb stack references	0

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	15.38 cycles
front end	15.38 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	11.50	11.50	12.25	12.25	12.25	12.25	11.00	11.00	11.00	11.00	11.17	10.83	11.00	0.50	0.50
cycles	11.50	11.50	12.25	12.25	12.25	12.25	11.00	11.00	11.00	11.00	11.17	10.83	11.00	0.50	0.50

Execution ports to units layout:

P0: BRU
P1: BRU
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	15.38
Dispatch	12.25
Overall L1	15.38

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	0%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
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%

INT+FP

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%

Vector efficiency ratios

INT

all	25%
load	25%
store	25%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	NA (no fma vectorizable/vectorized instructions)
other	25%

all	25%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	25%

INT+FP

all	25%
load	25%
store	25%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	25%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 15.38 cycles.

Front-end bottlenecks

Performance is limited by instruction throughput (loading/decoding program instructions to execution core) (front-end is a bottleneck). By removing all these bottlenecks, you can lower the cost of an iteration from 15.38 to 12.25 cycles (1.26x speedup).

ASM code

In the binary file, the address of the loop is: 4c0700

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
MOVI D19, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	N/A
CBNZ X4, 4c1498 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1f18>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X12, X12, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
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	scal (25.0%)
B.EQ 4bfa0c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x48c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X14, [X7, X12,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ORR X3, XZR, X12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
LDR X0, [X24, X12,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
STR XZR, [X19, X12,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	scal (25.0%)
CMP X14, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.LE 4c06dc <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x115c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X30, X21, X14,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
UBFM X14, X0, #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	N/A
MOVI D19, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	N/A
ADD X0, X21, X0,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
SUB X16, X30, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
SUB X2, X16, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
UBFM X18, X2, #3, #63	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ADD X17, X18, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ANDS X11, X17, #0x7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c07dc <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x125c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c07c4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1244>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c07b4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1234>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c07a4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1224>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c0794 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1214>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c0784 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1204>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #6	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c0774 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x11f4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR D26, [X21, X14]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
ADD X0, X0, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
FABS D29, D26	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D29	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D6, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D0, D6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D1, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D31, D1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D31	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D2, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D3, D2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D4, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D5, D4	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D5	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D7, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D16, D7	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D16	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D17, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D18, D17	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D18	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
CMP X30, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c06e0 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1160>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
B 4c06e0 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1160>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X30, [X10, X3,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDR X16, [X15, X3,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
CMP X16, X30	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.LE 4c06e4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1164>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X18, X6, X16,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
UBFM X17, X30, #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	N/A
ADD X0, X6, X30,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
SUB X2, X18, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
SUB X11, X2, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
UBFM X14, X11, #3, #63	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ADD X30, X14, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ANDS X16, X30, #0x7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c1564 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1fe4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c154c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1fcc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c153c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1fbc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c152c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1fac>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c151c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1f9c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c150c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1f8c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #6	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.NE 4c3a84 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x4504>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR D27, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D28, D27	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D25, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D22, D25	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D22	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D23, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D30, D23	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D30	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D24, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D26, D24	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D26	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D29, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D6, D29	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D0, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D1, D0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
CMP X0, X18	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c06e4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1164>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
B 4c06e4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1164>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR D20, [X6, X17]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
ADD X0, X0, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
FABS D21, D20	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D21	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
B 4c14fc <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1f7c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	hypre_ParCSRComputeL1NormsThreads._omp_fn.0
Source file and lines	ams.c:3382-3401
Module	exec

Non-innermost loop: analyzing only self part (ignoring child loops).
Ignoring paths for analysis
Too many paths. If you really need to analyze all of the 272 paths individually, rerun with max-paths=272
RecMII not computed since number of paths is unknown or > max_paths
Streams not analyzed since number of paths is unknown or > max_paths

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

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 15.38 to 9.33 cycles (1.65x speedup).

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)

Vectorization

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 is composed of 14 FP arithmetical operations:

14: addition or subtraction

The binary loop does not load or store any data.

General properties

nb instructions	123
nb uops	123
loop length	492
used w registers	0
used x registers	20
used b registers	0
used h registers	0
used s registers	0
used d registers	24
used q registers	0
used v registers	0
used z registers	0
nb stack references	0

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	15.38 cycles
front end	15.38 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	11.50	11.50	12.25	12.25	12.25	12.25	11.00	11.00	11.00	11.00	11.17	10.83	11.00	0.50	0.50
cycles	11.50	11.50	12.25	12.25	12.25	12.25	11.00	11.00	11.00	11.00	11.17	10.83	11.00	0.50	0.50

Execution ports to units layout:

P0: BRU
P1: BRU
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	15.38
Dispatch	12.25
Overall L1	15.38

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	0%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
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%

INT+FP

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%

Vector efficiency ratios

INT

all	25%
load	25%
store	25%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	NA (no fma vectorizable/vectorized instructions)
other	25%

all	25%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	25%

INT+FP

all	25%
load	25%
store	25%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	25%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 15.38 cycles.

Front-end bottlenecks

ASM code

In the binary file, the address of the loop is: 4c0700

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
MOVI D19, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	N/A
CBNZ X4, 4c1498 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1f18>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X12, X12, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
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	scal (25.0%)
B.EQ 4bfa0c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x48c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X14, [X7, X12,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ORR X3, XZR, X12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
LDR X0, [X24, X12,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
STR XZR, [X19, X12,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	scal (25.0%)
CMP X14, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.LE 4c06dc <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x115c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X30, X21, X14,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
UBFM X14, X0, #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	N/A
MOVI D19, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	N/A
ADD X0, X21, X0,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
SUB X16, X30, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
SUB X2, X16, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
UBFM X18, X2, #3, #63	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ADD X17, X18, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ANDS X11, X17, #0x7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c07dc <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x125c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c07c4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1244>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c07b4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1234>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c07a4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1224>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c0794 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1214>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c0784 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1204>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X11, #6	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c0774 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x11f4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR D26, [X21, X14]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
ADD X0, X0, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
FABS D29, D26	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D29	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D6, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D0, D6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D1, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D31, D1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D31	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D2, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D3, D2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D4, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D5, D4	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D5	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D7, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D16, D7	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D16	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D17, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D18, D17	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D18	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
CMP X30, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c06e0 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1160>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
B 4c06e0 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1160>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X30, [X10, X3,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDR X16, [X15, X3,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
CMP X16, X30	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.LE 4c06e4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1164>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X18, X6, X16,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
UBFM X17, X30, #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	N/A
ADD X0, X6, X30,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
SUB X2, X18, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
SUB X11, X2, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
UBFM X14, X11, #3, #63	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ADD X30, X14, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ANDS X16, X30, #0x7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c1564 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1fe4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c154c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1fcc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c153c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1fbc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c152c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1fac>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c151c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1f9c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c150c <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1f8c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X16, #6	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.NE 4c3a84 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x4504>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR D27, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D28, D27	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D28	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D25, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D22, D25	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D22	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D23, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D30, D23	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D30	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D24, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D26, D24	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D26	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D29, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D6, D29	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
LDR D0, [X0], #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
FABS D1, D0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
CMP X0, X18	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.EQ 4c06e4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1164>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
B 4c06e4 <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1164>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR D20, [X6, X17]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
ADD X0, X0, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
FABS D21, D20	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
FADD D19, D19, D21	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
STR D19, [X19, X3,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	scal (25.0%)
B 4c14fc <hypre_ParCSRComputeL1NormsThreads._omp_fn.0+0x1f7c>	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

Workaround

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

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

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code