OV - exec - Loop 2781

	Loop Id: 2781	Module: exec	Source: par_csr_matop.c:865-937 [...]	Coverage: 0.01%

0x486000 LDUR	X8, [X29, #432]

0x486004 ORR	X19, XZR, X2

0x486008 ORR	X22, XZR, X1

0x48600c CMP	X26, X8

0x486010 B.GE	486350

0x486014 LDUR	X8, [X29, #424]

0x486018 LDR	X8, [X8]

0x48601c CBZ	X8, 48603c

0x486020 LDUR	X8, [X29, #408]

0x486024 ADD	X1, X22, #1

0x486028 STR	X22, [X0, X26,LSL #3]

0x48602c LDR	D0, [X8]

0x486030 LDP	X8, X9, [SP, #104]

0x486034 STR	X26, [X8, X22,LSL #3]

0x486038 STR	D0, [X9, X22,LSL #3]

0x48603c LDUR	X8, [X29, #416]

0x486040 LDR	X8, [X8]

0x486044 CBZ	X8, 4861c0

0x486048 LDR	X8, [SP, #96]

0x48604c ADD	X3, X8, X26,LSL #3

0x486050 LDP	X4, X8, [X3]

0x486054 CMP	X4, X8

0x486058 B.GE	4861e0

0x48605c LDP	X9, X8, [SP, #48]

0x486060 ORR	X2, XZR, X19

0x486064 LDR	X5, [X9]

0x486068 LDR	X6, [X8]

0x48606c B	486090

  (2785) 0x486080 LDR	X8, [X3, #8]

  (2785) 0x486084 ADD	X4, X4, #1

  (2785) 0x486088 CMP	X4, X8

  (2785) 0x48608c B.GE	4861c4

  (2785) 0x486090 LDP	X8, X9, [X29, #984]

  (2785) 0x486094 LDR	X30, [X9, X4,LSL #3]

  (2785) 0x486098 LDR	D0, [X8, X4,LSL #3]

  (2785) 0x48609c ADD	X7, X30, #1

  (2785) 0x4860a0 LDR	X8, [X13, X30,LSL #3]

  (2785) 0x4860a4 LDR	X23, [X13, X7,LSL #3]

  (2785) 0x4860a8 CMP	X8, X23

  (2785) 0x4860ac B.GE	48612c

  (2785) 0x4860b0 LDUR	X9, [X29, #440]

  (2785) 0x4860b4 LDP	X10, X11, [X29, #1000]

  (2785) 0x4860b8 LDR	X9, [X9]

  (2785) 0x4860bc LDR	X10, [X10]

  (2785) 0x4860c0 LDR	X27, [X11]

  (2785) 0x4860c4 B	4860e4

    (2787) 0x4860c8 LDR	D1, [X9, X8,LSL #3]

    (2787) 0x4860cc LDR	D2, [X10, X11,LSL #3]

    (2787) 0x4860d0 FMADD	D1, D0, D1, D2

    (2787) 0x4860d4 STR	D1, [X10, X11,LSL #3]

    (2787) 0x4860d8 ADD	X8, X8, #1

    (2787) 0x4860dc CMP	X8, X23

    (2787) 0x4860e0 B.GE	48612c

    (2787) 0x4860e4 LDR	X11, [X20]

    (2787) 0x4860e8 LDR	X12, [X5, X8,LSL #3]

    (2787) 0x4860ec ADD	X24, X12, X11

    (2787) 0x4860f0 LDR	X11, [X0, X24,LSL #3]

    (2787) 0x4860f4 CMP	X11, X19

    (2787) 0x4860f8 B.GE	4860c8

    (2787) 0x4860fc STR	X2, [X0, X24,LSL #3]

    (2787) 0x486100 LDR	D1, [X9, X8,LSL #3]

    (2787) 0x486104 LDR	X11, [X20]

    (2787) 0x486108 SUB	X11, X24, X11

    (2787) 0x48610c STR	X11, [X27, X2,LSL #3]

    (2787) 0x486110 FMUL	D1, D0, D1

    (2787) 0x486114 LDR	X23, [X13, X7,LSL #3]

    (2787) 0x486118 STR	D1, [X10, X2,LSL #3]

    (2787) 0x48611c ADD	X2, X2, #1

    (2787) 0x486120 ADD	X8, X8, #1

    (2787) 0x486124 CMP	X8, X23

    (2787) 0x486128 B.LT	4860e4

  (2785) 0x48612c LDR	X8, [X14, X30,LSL #3]

  (2785) 0x486130 LDR	X23, [X14, X7,LSL #3]

  (2785) 0x486134 CMP	X8, X23

  (2785) 0x486138 B.GE	486080

  (2785) 0x48613c LDUR	X9, [X29, #448]

  (2785) 0x486140 LDR	X10, [X25]

  (2785) 0x486144 LDR	X27, [X21]

  (2785) 0x486148 LDR	X9, [X9]

  (2785) 0x48614c B	48617c

    (2786) 0x486160 LDR	D1, [X9, X8,LSL #3]

    (2786) 0x486164 LDR	D2, [X10, X11,LSL #3]

    (2786) 0x486168 FMADD	D1, D0, D1, D2

    (2786) 0x48616c STR	D1, [X10, X11,LSL #3]

    (2786) 0x486170 ADD	X8, X8, #1

    (2786) 0x486174 CMP	X8, X23

    (2786) 0x486178 B.GE	486080

    (2786) 0x48617c LDR	X24, [X6, X8,LSL #3]

    (2786) 0x486180 LDR	X11, [X0, X24,LSL #3]

    (2786) 0x486184 CMP	X11, X22

    (2786) 0x486188 B.GE	486160

    (2786) 0x48618c LDR	D1, [X9, X8,LSL #3]

    (2786) 0x486190 STR	X1, [X0, X24,LSL #3]

    (2786) 0x486194 STR	X24, [X27, X1,LSL #3]

    (2786) 0x486198 LDR	X23, [X14, X7,LSL #3]

    (2786) 0x48619c FMUL	D1, D0, D1

    (2786) 0x4861a0 STR	D1, [X10, X1,LSL #3]

    (2786) 0x4861a4 ADD	X1, X1, #1

    (2786) 0x4861a8 ADD	X8, X8, #1

    (2786) 0x4861ac CMP	X8, X23

    (2786) 0x4861b0 B.LT	48617c

  (2785) 0x4861b4 B	486080

0x4861c0 ORR	X2, XZR, X19

0x4861c4 LDR	X3, [X15, X26,LSL #3]

0x4861c8 ADD	X26, X26, #1

0x4861cc LDR	X8, [X15, X26,LSL #3]

0x4861d0 CMP	X3, X8

0x4861d4 B.GE	486000

0x4861e0 ORR	X2, XZR, X19

0x4861e4 LDR	X3, [X15, X26,LSL #3]

0x4861e8 ADD	X26, X26, #1

0x4861ec LDR	X8, [X15, X26,LSL #3]

0x4861f0 CMP	X3, X8

0x4861f4 B.GE	486000

/home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c: 865 - 937

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

865:     for (i1 = ns; i1 < ne; i1++)

[...]

874:       if ( allsquare )

875:       {

876:          B_marker[i1] = jj_count_diag;

877:          C_diag_data[jj_count_diag] = zero;

878:          C_diag_j[jj_count_diag] = i1;

879:          jj_count_diag++;

[...]

886:       if (num_cols_offd_A)

887:       {

888:          for (jj2 = A_offd_i[i1]; jj2 < A_offd_i[i1+1]; jj2++)

889:          {

890:             i2 = A_offd_j[jj2];

891:             a_entry = A_offd_data[jj2];

[...]

897:             for (jj3 = B_ext_offd_i[i2]; jj3 < B_ext_offd_i[i2+1]; jj3++)

898:             {

899:                i3 = num_cols_diag_B+B_ext_offd_j[jj3];

[...]

907:                if (B_marker[i3] < jj_row_begin_offd)

908:                {

909:                   B_marker[i3] = jj_count_offd;

910:                   C_offd_data[jj_count_offd] = a_entry*B_ext_offd_data[jj3];

911:                   C_offd_j[jj_count_offd] = i3-num_cols_diag_B;

912:                   jj_count_offd++;

913:                }

914:                else

915:                   C_offd_data[B_marker[i3]] += a_entry*B_ext_offd_data[jj3];

916:             }

917:             for (jj3 = B_ext_diag_i[i2]; jj3 < B_ext_diag_i[i2+1]; jj3++)

918:             {

919:                i3 = B_ext_diag_j[jj3];

920:                if (B_marker[i3] < jj_row_begin_diag)

921:                {

922:                   B_marker[i3] = jj_count_diag;

923:                   C_diag_data[jj_count_diag] = a_entry*B_ext_diag_data[jj3];

924:                   C_diag_j[jj_count_diag] = i3;

925:                   jj_count_diag++;

926:                }

927:                else

928:                   C_diag_data[B_marker[i3]] += a_entry*B_ext_diag_data[jj3];

[...]

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

Coverage (%)	Name	Module
►100.00+	__kmp_invoke_microtask	libomp.so
○	__kmp_invoke_task_func	libomp.so
○	__kmp_launch_thread	libomp.so
○	__kmp_launch_worker(void*)	libomp.so
○	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	1.24
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	1.02
Bottlenecks	P10,
Function	hypre_ParMatmul.omp_outlined.6
Source	par_csr_matop.c:865-865,par_csr_matop.c:874-879,par_csr_matop.c:886-888,par_csr_matop.c:917-917,par_csr_matop.c:937-937
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	6.83
CQA cycles if no scalar integer	5.50
CQA cycles if FP arith vectorized	6.83
CQA cycles if fully vectorized	1.71
Front-end cycles	5.00
P0 cycles	3.50
P1 cycles	3.50
P2 cycles	3.25
P3 cycles	3.25
P4 cycles	3.25
P5 cycles	3.25
P6 cycles	0.50
P7 cycles	0.50
P8 cycles	0.00
P9 cycles	0.00
P10 cycles	6.83
P11 cycles	6.50
P12 cycles	6.67
P13 cycles	1.00
P14 cycles	1.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	40.00
Nb uops	40.00
Nb loads	NA
Nb stores	3.00
Nb stack references	3.00
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	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	1.24
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	1.02
Bottlenecks	P10,
Function	hypre_ParMatmul.omp_outlined.6
Source	par_csr_matop.c:865-865,par_csr_matop.c:874-879,par_csr_matop.c:886-888,par_csr_matop.c:917-917,par_csr_matop.c:937-937
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	6.83
CQA cycles if no scalar integer	5.50
CQA cycles if FP arith vectorized	6.83
CQA cycles if fully vectorized	1.71
Front-end cycles	5.00
P0 cycles	3.50
P1 cycles	3.50
P2 cycles	3.25
P3 cycles	3.25
P4 cycles	3.25
P5 cycles	3.25
P6 cycles	0.50
P7 cycles	0.50
P8 cycles	0.00
P9 cycles	0.00
P10 cycles	6.83
P11 cycles	6.50
P12 cycles	6.67
P13 cycles	1.00
P14 cycles	1.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	40.00
Nb uops	40.00
Nb loads	NA
Nb stores	3.00
Nb stack references	3.00
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	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_ParMatmul.omp_outlined.6
Source file and lines	par_csr_matop.c:865-937
Module	exec

The loop is defined in /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c:865,874-879,886-891,897-899,907-928,937.

Analyzed code is defined in /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c:865,874-879,886-888,917,937.

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
Streams not analyzed 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 6.83 to 5.50 cycles (1.24x 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 6.83 to 1.71 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 ffast-math (included in Ofast) to extend loop vectorization to FP reductions.
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop streams arrays of structures (AoS), try to use structures of arrays instead (SoA): for(i) a[i].x = b[i].x; (slow, non stride 1) => for(i) a.x[i] = b.x[i]; (fast, stride 1)

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 does not load or store any data.

General properties

nb instructions	40
nb uops	40
loop length	160
used w registers	0
used x registers	15
used b registers	0
used h registers	0
used s registers	1
used d registers	1
used q registers	0
used v registers	0
used z registers	0
nb stack references	3

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	5.00 cycles
front end	5.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	3.50	3.50	3.25	3.25	3.25	3.25	0.50	0.50	0.00	0.00	6.83	6.50	6.67	1.00	1.00
cycles	3.50	3.50	3.25	3.25	3.25	3.25	0.50	0.50	0.00	0.00	6.83	6.50	6.67	1.00	1.00

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	5.00
Dispatch	6.83
Overall L1	6.83

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%

Vector efficiency ratios

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 6.83 cycles.

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the loop is: 486000

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
LDUR X8, [X29, #432]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ORR X19, XZR, X2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ORR X22, XZR, X1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
CMP X26, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.GE 486350 <hypre_ParMatmul.omp_outlined.6+0x560>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDUR X8, [X29, #424]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X8, [X8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CBZ X8, 48603c <hypre_ParMatmul.omp_outlined.6+0x24c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDUR X8, [X29, #408]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X1, X22, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
STR X22, [X0, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
LDR D0, [X8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
LDP X8, X9, [SP, #104]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
STR X26, [X8, X22,LSL #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
STR D0, [X9, X22,LSL #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	0	0	2	0.50	scal (25.0%)
LDUR X8, [X29, #416]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X8, [X8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CBZ X8, 4861c0 <hypre_ParMatmul.omp_outlined.6+0x3d0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X8, [SP, #96]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X3, X8, X26,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
LDP X4, X8, [X3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
CMP X4, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.GE 4861e0 <hypre_ParMatmul.omp_outlined.6+0x3f0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDP X9, X8, [SP, #48]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
ORR X2, XZR, X19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X5, [X9]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDR X6, [X8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
B 486090 <hypre_ParMatmul.omp_outlined.6+0x2a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X2, XZR, X19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X3, [X15, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X26, X26, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
LDR X8, [X15, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CMP X3, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.GE 486000 <hypre_ParMatmul.omp_outlined.6+0x210>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X2, XZR, X19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X3, [X15, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X26, X26, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
LDR X8, [X15, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CMP X3, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.33	N/A
B.GE 486000 <hypre_ParMatmul.omp_outlined.6+0x210>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	hypre_ParMatmul.omp_outlined.6
Source file and lines	par_csr_matop.c:865-937
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
Streams not analyzed 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 6.83 to 5.50 cycles (1.24x 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 6.83 to 1.71 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 ffast-math (included in Ofast) to extend loop vectorization to FP reductions.
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop streams arrays of structures (AoS), try to use structures of arrays instead (SoA): for(i) a[i].x = b[i].x; (slow, non stride 1) => for(i) a.x[i] = b.x[i]; (fast, stride 1)

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 does not load or store any data.

General properties

nb instructions	40
nb uops	40
loop length	160
used w registers	0
used x registers	15
used b registers	0
used h registers	0
used s registers	1
used d registers	1
used q registers	0
used v registers	0
used z registers	0
nb stack references	3

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	5.00 cycles
front end	5.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	3.50	3.50	3.25	3.25	3.25	3.25	0.50	0.50	0.00	0.00	6.83	6.50	6.67	1.00	1.00
cycles	3.50	3.50	3.25	3.25	3.25	3.25	0.50	0.50	0.00	0.00	6.83	6.50	6.67	1.00	1.00

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	5.00
Dispatch	6.83
Overall L1	6.83

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%

Vector efficiency ratios

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 6.83 cycles.

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the loop is: 486000

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
LDUR X8, [X29, #432]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ORR X19, XZR, X2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ORR X22, XZR, X1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
CMP X26, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.GE 486350 <hypre_ParMatmul.omp_outlined.6+0x560>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDUR X8, [X29, #424]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X8, [X8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CBZ X8, 48603c <hypre_ParMatmul.omp_outlined.6+0x24c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDUR X8, [X29, #408]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X1, X22, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
STR X22, [X0, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
LDR D0, [X8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	scal (25.0%)
LDP X8, X9, [SP, #104]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
STR X26, [X8, X22,LSL #3]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
STR D0, [X9, X22,LSL #3]	1	0	0	0	0	0	0	0.50	0.50	0.50	0.50	0	0	0	2	0.50	scal (25.0%)
LDUR X8, [X29, #416]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X8, [X8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CBZ X8, 4861c0 <hypre_ParMatmul.omp_outlined.6+0x3d0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X8, [SP, #96]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X3, X8, X26,LSL #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
LDP X4, X8, [X3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
CMP X4, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.GE 4861e0 <hypre_ParMatmul.omp_outlined.6+0x3f0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDP X9, X8, [SP, #48]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
ORR X2, XZR, X19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X5, [X9]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDR X6, [X8]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
B 486090 <hypre_ParMatmul.omp_outlined.6+0x2a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X2, XZR, X19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X3, [X15, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X26, X26, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
LDR X8, [X15, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CMP X3, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.GE 486000 <hypre_ParMatmul.omp_outlined.6+0x210>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X2, XZR, X19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X3, [X15, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X26, X26, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.25	N/A
LDR X8, [X15, X26,LSL #3]	1	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CMP X3, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	1	0.33	N/A
B.GE 486000 <hypre_ParMatmul.omp_outlined.6+0x210>	1	0.50	0.50	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