OV - exec - Loop 2372

	Loop Id: 2372	Module: exec	Source: par_strength.c:2011-2048	Coverage: 0.15%

0x4b50c4 LDR	X5, [X26, X9,LSL #3]

0x4b50c8 UBFM	X16, X5, #61, #60

0x4b50cc LDR	X17, [X23, X16]

0x4b50d0 CMP	X17, #0

0x4b50d4 B.LE	4b5104

0x4b50d8 LDR	X22, [X7, X16]

0x4b50dc UBFM	X6, X22, #61, #60

0x4b50e0 LDR	X14, [X21, X6]

0x4b50e4 CMP	X14, X3

0x4b50e8 B.GE	4b5104

0x4b50ec LDR	X30, [SP, #336]

0x4b50f0 STR	X30, [X21, X6]

0x4b50f4 STR	X22, [X10, X30,LSL #3]

0x4b50f8 LDR	X5, [SP, #336]

0x4b50fc ADD	X17, X5, #1

0x4b5100 STR	X17, [SP, #336]

0x4b5104 ADD	X17, X16, #8

0x4b5108 LDR	X5, [X12, X16]

0x4b510c ADD	X22, X12, X17

0x4b5110 LDR	X14, [X12, X17]

0x4b5114 CMP	X14, X5

0x4b5118 B.LE	4b518c

0x4b511c ORR	X30, XZR, X22

0x4b5120 STR	X0, [SP, #96]

  (2374) 0x4b5124 LDR	X0, [X26, X5,LSL #3]

  (2374) 0x4b5128 ADD	X5, X5, #1

  (2374) 0x4b512c UBFM	X6, X0, #61, #60

  (2374) 0x4b5130 LDR	X22, [X23, X6]

  (2374) 0x4b5134 CMP	X22, #0

  (2374) 0x4b5138 B.LE	4b5180

  (2374) 0x4b513c LDR	X6, [X7, X6]

  (2374) 0x4b5140 UBFM	X22, X6, #61, #60

  (2374) 0x4b5144 CMP	X20, X6

  (2374) 0x4b5148 B.EQ	4b5180

  (2374) 0x4b514c LDR	X0, [X21, X22]

  (2374) 0x4b5150 CMP	X0, X3

  (2374) 0x4b5154 B.GE	4b5180

  (2374) 0x4b5158 LDR	X14, [SP, #336]

  (2374) 0x4b515c STR	X14, [X21, X22]

  (2374) 0x4b5160 STR	X6, [X10, X14,LSL #3]

  (2374) 0x4b5164 LDR	X6, [SP, #336]

  (2374) 0x4b5168 LDR	X14, [X30]

  (2374) 0x4b516c ADD	X22, X6, #1

  (2374) 0x4b5170 STR	X22, [SP, #336]

  (2374) 0x4b5174 HINT	#0

  (2374) 0x4b5178 HINT	#0

  (2374) 0x4b517c HINT	#0

  (2374) 0x4b5180 CMP	X5, X14

  (2374) 0x4b5184 B.LT	4b5124

0x4b5188 LDR	X0, [SP, #96]

0x4b518c ADD	X30, X11, X17

0x4b5190 LDR	X5, [X11, X16]

0x4b5194 LDR	X6, [X11, X17]

0x4b5198 CMP	X6, X5

0x4b519c B.LE	4b51f0

  (2373) 0x4b51a0 LDR	X16, [X25, X5,LSL #3]

  (2373) 0x4b51a4 ADD	X5, X5, #1

  (2373) 0x4b51a8 UBFM	X17, X16, #61, #60

  (2373) 0x4b51ac LDR	X14, [X24, X17]

  (2373) 0x4b51b0 CMP	X14, #0

  (2373) 0x4b51b4 B.LE	4b51e8

  (2373) 0x4b51b8 LDR	X22, [X18, X17]

  (2373) 0x4b51bc UBFM	X16, X22, #61, #60

  (2373) 0x4b51c0 LDR	X17, [X19, X16]

  (2373) 0x4b51c4 CMP	X17, X0

  (2373) 0x4b51c8 B.GE	4b51e8

  (2373) 0x4b51cc LDR	X6, [SP, #344]

  (2373) 0x4b51d0 STR	X6, [X19, X16]

  (2373) 0x4b51d4 STR	X22, [X8, X6,LSL #3]

  (2373) 0x4b51d8 LDR	X14, [SP, #344]

  (2373) 0x4b51dc LDR	X6, [X30]

  (2373) 0x4b51e0 ADD	X22, X14, #1

  (2373) 0x4b51e4 STR	X22, [SP, #344]

  (2373) 0x4b51e8 CMP	X5, X6

  (2373) 0x4b51ec B.LT	4b51a0

0x4b51f0 LDR	X30, [X15]

0x4b51f4 ADD	X9, X9, #1

0x4b51f8 CMP	X30, X9

0x4b51fc B.GT	4b50c4

/home/hbollore/qaas/qaas-runs/174-161-6712/intel/AMG/build/AMG/AMG/parcsr_ls/par_strength.c: 2011 - 2048

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

2011:              for (jj1 = S_diag_i[i1]; jj1 < S_diag_i[i1+1]; jj1++)

2012:              {

2013:                  i2 = S_diag_j[jj1];

2014:                  if (CF_marker[i2] > 0)

2015:                  {

2016:                     index = fine_to_coarse[i2];

2017:                     if (S_marker[index] < jj_row_begin_diag)

2018:                     {

2019:                        S_marker[index] = num_nonzeros_diag;

2020:                        C_diag_j[num_nonzeros_diag] = index;

2021:                        num_nonzeros_diag++;

2022:                     }

2023:                  }

2024:                  for (jj2 = S_diag_i[i2]; jj2 < S_diag_i[i2+1]; jj2++)

2025:                  {

2026:                     i3 = S_diag_j[jj2];

2027:                     if (CF_marker[i3] > 0)

2028:                     {

2029:                        index = fine_to_coarse[i3];

2030:                        if (index != ic && S_marker[index] < jj_row_begin_diag)

2031:                        {

2032:                           S_marker[index] = num_nonzeros_diag;

2033:                           C_diag_j[num_nonzeros_diag] = index;

2034:                           num_nonzeros_diag++;

2035:                        }

2036:                     }

2037:                  }

2038:                  for (jj2 = S_offd_i[i2]; jj2 < S_offd_i[i2+1]; jj2++)

2039:                  {

2040:                     i3 = S_offd_j[jj2];

2041:                     if (CF_marker_offd[i3] > 0)

2042:                     {

2043:                        index = map_S_to_C[i3];

2044:                        if (S_marker_offd[index] < jj_row_begin_offd)

2045:                        {

2046:                           S_marker_offd[index] = num_nonzeros_offd;

2047:                           C_offd_j[num_nonzeros_offd] = index;

2048:                           num_nonzeros_offd++;

Coverage (%)	Name	Source Location	Module
►99.43+	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	1.25
Bottlenecks	P12, P13, P14,
Function	hypre_BoomerAMGCreate2ndS._omp_fn.7
Source	par_strength.c:2011-2011,par_strength.c:2014-2021,par_strength.c:2024-2024,par_strength.c:2038-2038
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	5.33
CQA cycles if no scalar integer	5.33
CQA cycles if FP arith vectorized	5.33
CQA cycles if fully vectorized	2.67
Front-end cycles	4.25
P0 cycles	2.50
P1 cycles	2.50
P2 cycles	2.25
P3 cycles	2.25
P4 cycles	2.17
P5 cycles	2.17
P6 cycles	2.08
P7 cycles	2.08
P8 cycles	0.00
P9 cycles	0.00
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	5.33
P13 cycles	5.33
P14 cycles	5.33
P15 cycles	2.00
P16 cycles	2.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	34.00
Nb uops	34.00
Nb loads	NA
Nb stores	4.00
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	24.00
Bytes prefetched	0.00
Bytes loaded	96.00
Bytes stored	32.00
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	50.00
Vector-efficiency ratio load	50.00
Vector-efficiency ratio store	50.00
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	50.00
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	1.25
Bottlenecks	P12, P13, P14,
Function	hypre_BoomerAMGCreate2ndS._omp_fn.7
Source	par_strength.c:2011-2011,par_strength.c:2014-2021,par_strength.c:2024-2024,par_strength.c:2038-2038
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	5.33
CQA cycles if no scalar integer	5.33
CQA cycles if FP arith vectorized	5.33
CQA cycles if fully vectorized	2.67
Front-end cycles	4.25
P0 cycles	2.50
P1 cycles	2.50
P2 cycles	2.25
P3 cycles	2.25
P4 cycles	2.17
P5 cycles	2.17
P6 cycles	2.08
P7 cycles	2.08
P8 cycles	0.00
P9 cycles	0.00
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	5.33
P13 cycles	5.33
P14 cycles	5.33
P15 cycles	2.00
P16 cycles	2.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	34.00
Nb uops	34.00
Nb loads	NA
Nb stores	4.00
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	24.00
Bytes prefetched	0.00
Bytes loaded	96.00
Bytes stored	32.00
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	50.00
Vector-efficiency ratio load	50.00
Vector-efficiency ratio store	50.00
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	50.00
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_BoomerAMGCreate2ndS._omp_fn.7
Source file and lines	par_strength.c:2011-2048
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/174-161-6712/intel/AMG/build/AMG/AMG/parcsr_ls/par_strength.c:2011-2048.

Analyzed code is defined in /home/hbollore/qaas/qaas-runs/174-161-6712/intel/AMG/build/AMG/AMG/parcsr_ls/par_strength.c:2011,2014-2021,2024,2038.

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 60 paths individually, rerun with max-paths=60
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 60 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 5.33 to 2.67 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 96 bytes. The binary loop is storing 32 bytes.

General properties

nb instructions	34
loop length	136
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

4.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16
uops	2.50	2.50	2.25	2.25	2.17	2.17	2.08	2.08	0.00	0.00	0.00	0.00	5.33	5.33	5.33	2.00	2.00
cycles	2.50	2.50	2.25	2.25	2.17	2.17	2.08	2.08	0.00	0.00	0.00	0.00	5.33	5.33	5.33	2.00	2.00

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

Cycles summary

Front-end	4.25
Overall L1	5.33

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 5.33 cycles. At this rate:

18% of peak load performance is reached (18.00 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
12% of peak store performance is reached (6.00 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 4b50c4

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P12	P13	P14	P15	P16	Latency	Recip. throughput	Vectorization
LDR X5, [X26, X9,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
UBFM X16, X5, #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
LDR X17, [X23, X16]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CMP X17, #0	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.LE 4b5104 <hypre_BoomerAMGCreate2ndS._omp_fn.7+0xde0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X22, [X7, X16]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
UBFM X6, X22, #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
LDR X14, [X21, X6]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
CMP X14, X3	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 4b5104 <hypre_BoomerAMGCreate2ndS._omp_fn.7+0xde0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X30, [SP, #336]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
STR X30, [X21, X6]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
STR X22, [X10, X30,LSL #3]	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 X5, [SP, #336]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ADD X17, X5, #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
STR X17, [SP, #336]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X17, X16, #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 X5, [X12, X16]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ADD X22, X12, X17	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%)
LDR X14, [X12, X17]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
CMP X14, X5	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.LE 4b518c <hypre_BoomerAMGCreate2ndS._omp_fn.7+0xe68>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X30, XZR, X22	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, #96]	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, #96]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ADD X30, X11, X17	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 X5, [X11, X16]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
LDR X6, [X11, X17]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CMP X6, X5	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.LE 4b51f0 <hypre_BoomerAMGCreate2ndS._omp_fn.7+0xecc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X30, [X15]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X9, X9, #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 X30, X9	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	1	0.33	N/A
B.GT 4b50c4 <hypre_BoomerAMGCreate2ndS._omp_fn.7+0xda0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	hypre_BoomerAMGCreate2ndS._omp_fn.7
Source file and lines	par_strength.c:2011-2048
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 60 paths individually, rerun with max-paths=60
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

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 5.33 to 2.67 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 96 bytes. The binary loop is storing 32 bytes.

General properties

nb instructions	34
loop length	136
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

4.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16
uops	2.50	2.50	2.25	2.25	2.17	2.17	2.08	2.08	0.00	0.00	0.00	0.00	5.33	5.33	5.33	2.00	2.00
cycles	2.50	2.50	2.25	2.25	2.17	2.17	2.08	2.08	0.00	0.00	0.00	0.00	5.33	5.33	5.33	2.00	2.00

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

Cycles summary

Front-end	4.25
Overall L1	5.33

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 5.33 cycles. At this rate:

18% of peak load performance is reached (18.00 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
12% of peak store performance is reached (6.00 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 4b50c4

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P12	P13	P14	P15	P16	Latency	Recip. throughput	Vectorization
LDR X5, [X26, X9,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
UBFM X16, X5, #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
LDR X17, [X23, X16]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CMP X17, #0	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.LE 4b5104 <hypre_BoomerAMGCreate2ndS._omp_fn.7+0xde0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X22, [X7, X16]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
UBFM X6, X22, #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
LDR X14, [X21, X6]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
CMP X14, X3	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 4b5104 <hypre_BoomerAMGCreate2ndS._omp_fn.7+0xde0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X30, [SP, #336]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
STR X30, [X21, X6]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
STR X22, [X10, X30,LSL #3]	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 X5, [SP, #336]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ADD X17, X5, #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
STR X17, [SP, #336]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X17, X16, #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 X5, [X12, X16]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ADD X22, X12, X17	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%)
LDR X14, [X12, X17]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
CMP X14, X5	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.LE 4b518c <hypre_BoomerAMGCreate2ndS._omp_fn.7+0xe68>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X30, XZR, X22	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, #96]	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, #96]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
ADD X30, X11, X17	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 X5, [X11, X16]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (50.0%)
LDR X6, [X11, X17]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CMP X6, X5	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.LE 4b51f0 <hypre_BoomerAMGCreate2ndS._omp_fn.7+0xecc>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X30, [X15]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X9, X9, #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 X30, X9	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	1	0.33	N/A
B.GT 4b50c4 <hypre_BoomerAMGCreate2ndS._omp_fn.7+0xda0>	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