OV - exec - Function .omp

	Function: .omp_outlined..15	Module: exec	Source: par_coarsen.c:2327-2334	Coverage: 0.02%

/home/hbollore/qaas-runs/171-233-5044/intel/AMG/build/AMG/AMG/parcsr_ls/par_coarsen.c: 2327 - 2334

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

2327: #pragma omp parallel for private(ig, i) HYPRE_SMP_SCHEDULE

2328: #endif

2329:         for (ig = 0; ig < graph_size; ig++)

2330:         {

2331:            i = graph_array[ig];

2332:            if (measure_array[i] > 1)

2333:            {

2334:               CF_marker[i] = 1;

0x421860 SUB	SP, SP, #96

0x421864 STP	X29, X30, [SP, #32]

0x421868 STR	X23, [SP, #48]

0x42186c STP	X22, X21, [SP, #64]

0x421870 STP	X20, X19, [SP, #80]

0x421874 ADD	X29, SP, #32

0x421878 LDR	X8, [X2]

0x42187c SUBS	X23, X8, #1

0x421880 B.LT	421940

0x421884 LDR	W19, [X0]

0x421888 LDR	X22, [X3]

0x42188c LDR	X20, [X4]

0x421890 MOVZ	W8, #1

0x421894 ADRP	X0,

0x421898 ADD	X0, X0, #688

0x42189c ADD	X3, SP, #12

0x4218a0 ADD	X4, X29, #24

0x4218a4 LDR	X21, [X5]

0x4218a8 SUB	X5, X29, #8

0x4218ac ADD	X6, SP, #16

0x4218b0 MOVZ	W2, #34

0x4218b4 STR	XZR, [X29, #24]

0x4218b8 STUR	X23, [X29, #504]

0x4218bc STR	X8, [SP, #16]

0x4218c0 MOVZ	W7, #1

0x4218c4 ORR	W1, WZR, W19

0x4218c8 STR	WZR, [SP, #12]

0x4218cc STR	X8, [SP]

0x4218d0 BL	40f550

0x4218d4 LDUR	X8, [X29, #504]

0x4218d8 LDR	X9, [X29, #24]

0x4218dc CMP	X8, X23

0x4218e0 CSEL	X8, X8, X23, #11

0x4218e4 CMP	X8, X9

0x4218e8 B.LT	421930

0x4218ec SUB	W10, W8, W9

0x4218f0 ADD	W10, W10, #1

0x4218f4 TBNZ	W10, #0, 421908

0x4218f8 ORR	X10, XZR, X9

0x4218fc CMP	X8, X9

0x421900 B.EQ	421930

0x421904 B	421958

0x421908 LDR	X10, [X22, X9,LSL #3]

0x42190c FMOV	D1, #1.0000000

0x421910 LDR	D0, [X20, X10,LSL #3]

0x421914 FCMP	D0, D1

0x421918 B.LE	421924

0x42191c MOVZ	W11, #1

0x421920 STR	X11, [X21, X10,LSL #3]

0x421924 ADD	X10, X9, #1

0x421928 CMP	X8, X9

0x42192c B.NE	421958

  (528) 0x421930 ORR	W1, WZR, W19

  (528) 0x421934 ADRP	X0,

  (528) 0x421938 ADD	X0, X0, #712

  (528) 0x42193c BL	40f410

  (528) 0x421940 LDP	X20, X19, [SP, #80]

  (528) 0x421944 LDP	X22, X21, [SP, #64]

  (528) 0x421948 LDR	X23, [SP, #48]

  (528) 0x42194c LDP	X29, X30, [SP, #32]

  (528) 0x421950 ADD	SP, SP, #96

  (528) 0x421954 RET

  (528) 0x421958 SUB	X8, X8, X10

  (528) 0x42195c ADD	X9, X22, X10,LSL #3

  (528) 0x421960 FMOV	D0, #1.0000000

  (528) 0x421964 MOVZ	W10, #1

  (528) 0x421968 ADD	X8, X8, #1

  (528) 0x42196c ADD	X9, X9, #8

  (528) 0x421970 B	42198c

0x421974 HINT	#0

0x421978 HINT	#0

0x42197c HINT	#0

  (528) 0x421980 SUBS	X8, X8, #2

  (528) 0x421984 ADD	X9, X9, #16

  (528) 0x421988 B.EQ	421930

  (528) 0x42198c LDUR	X11, [X9, #504]

  (528) 0x421990 LDR	D1, [X20, X11,LSL #3]

  (528) 0x421994 FCMP	D1, D0

  (528) 0x421998 B.GT	4219c0

  (528) 0x42199c LDR	X11, [X9]

  (528) 0x4219a0 LDR	D1, [X20, X11,LSL #3]

  (528) 0x4219a4 FCMP	D1, D0

  (528) 0x4219a8 B.LE	421980

  (528) 0x4219ac B	4219d4

0x4219b0 HINT	#0

0x4219b4 HINT	#0

0x4219b8 HINT	#0

0x4219bc HINT	#0

(527) 0x4219c0 STR	X10, [X21, X11,LSL #3]

(527) 0x4219c4 LDR	X11, [X9]

(527) 0x4219c8 LDR	D1, [X20, X11,LSL #3]

(527) 0x4219cc FCMP	D1, D0

(527) 0x4219d0 B.LE	421980

  (528) 0x4219d4 STR	X10, [X21, X11,LSL #3]

  (528) 0x4219d8 B	421980

0x4219dc HINT	#0

Coverage (%)	Name	Source Location	Module

Path /

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

Source file and lines	par_coarsen.c:2327-2334
Module	exec

The function is defined in /home/hbollore/qaas-runs/171-233-5044/intel/AMG/build/AMG/AMG/parcsr_ls/par_coarsen.c:2327-2334.

Warnings:
Ignoring paths for analysis

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 6.50 to 1.33 cycles (4.87x speedup).

Workaround

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

Vectorization

Your function is not vectorized. Only 20% of vector register length is used (average across all VPU instructions). By vectorizing your function, you can lower the cost of an iteration from 6.50 to 0.91 cycles (7.17x 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 function can use their full power.

Workaround

Try another compiler or update/tune your current one
Make array accesses unit-stride:
- If your function 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)

Expensive FP math instructions/calls

Detected performance impact from expensive FP math instructions/calls. By removing/reexpressing them, you can lower the cost of an iteration from 6.50 to 0.50 cycles (13.00x speedup).

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

The binary function does not contain any FP arithmetical operations. The binary function is loading 68 bytes. The binary function is storing 100 bytes.

General properties

nb instructions	60
loop length	240
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

6.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	4.00	4.00	6.00	6.00	6.00	6.00	1.00	0.50	0.25	0.25	6.33	6.33	6.33	5.00	5.00
cycles	4.00	4.00	6.00	6.00	6.00	6.00	1.00	0.50	0.25	0.25	6.33	6.33	6.33	5.00	5.00

Execution ports to units layout:

P0:
P1:
P2: ALU
P3: ALU
P4: ALU
P5: ALU
P6 (128 bits): VPU, FP store data, ALU, DIV/SQRT
P7 (128 bits): VPU, ALU, FP store data
P8 (128 bits): VPU, ALU, DIV/SQRT
P9 (128 bits): ALU, VPU
P10 (256 bits): store address, load
P11 (256 bits): store address, load
P12 (256 bits): load
P13 (64 bits): store data
P14 (64 bits): store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	6.50
Overall L1	6.50

Vectorization ratios

all	0%
load	0%
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%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 6.50 cycles. At this rate:

10% of peak load performance is reached (10.46 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
32% of peak store performance is reached (15.38 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the function is: 421860

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput
SUB SP, SP, #96	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STP X29, X30, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR X23, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X22, X21, [SP, #64]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X20, X19, [SP, #80]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
ADD X29, SP, #32	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
LDR X8, [X2]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
SUBS X23, X8, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.LT 421940 <.omp_outlined..15+0xe0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR W19, [X0]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X22, [X3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X20, [X4]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
MOVZ W8, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADRP X0, <4eb894>	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X0, X0, #688	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X3, SP, #12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X4, X29, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
LDR X21, [X5]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
SUB X5, X29, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X6, SP, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ W2, #34	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STR XZR, [X29, #24]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STUR X23, [X29, #504]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR X8, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
MOVZ W7, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR W1, WZR, W19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STR WZR, [SP, #12]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR X8, [SP]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
BL 40f550 <@plt_start@+0x160>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDUR X8, [X29, #504]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X9, [X29, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X8, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
CSEL X8, X8, X23, #11	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X8, X9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.LT 421930 <.omp_outlined..15+0xd0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
SUB W10, W8, W9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD W10, W10, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
TBNZ W10, #0, 421908 <.omp_outlined..15+0xa8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X10, XZR, X9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X8, X9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.EQ 421930 <.omp_outlined..15+0xd0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
B 421958 <.omp_outlined..15+0xf8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X10, [X22, X9,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
FMOV D1, #1.0000000	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
LDR D0, [X20, X10,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
FCMP D0, D1	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1
B.LE 421924 <.omp_outlined..15+0xc4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
MOVZ W11, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STR X11, [X21, X10,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
ADD X10, X9, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X8, X9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.NE 421958 <.omp_outlined..15+0xf8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
HINT #0
HINT #0
HINT #0
HINT #0
HINT #0
HINT #0
HINT #0
HINT #0

Source file and lines	par_coarsen.c:2327-2334
Module	exec

The function is defined in /home/hbollore/qaas-runs/171-233-5044/intel/AMG/build/AMG/AMG/parcsr_ls/par_coarsen.c:2327-2334.

Warnings:
Ignoring paths for analysis

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 6.50 to 1.33 cycles (4.87x 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 function can use their full power.

Workaround

Try another compiler or update/tune your current one
Make array accesses unit-stride:
- If your function 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)

Expensive FP math instructions/calls

Detected performance impact from expensive FP math instructions/calls. By removing/reexpressing them, you can lower the cost of an iteration from 6.50 to 0.50 cycles (13.00x speedup).

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

The binary function does not contain any FP arithmetical operations. The binary function is loading 68 bytes. The binary function is storing 100 bytes.

General properties

nb instructions	60
loop length	240
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

6.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	4.00	4.00	6.00	6.00	6.00	6.00	1.00	0.50	0.25	0.25	6.33	6.33	6.33	5.00	5.00
cycles	4.00	4.00	6.00	6.00	6.00	6.00	1.00	0.50	0.25	0.25	6.33	6.33	6.33	5.00	5.00

Execution ports to units layout:

P0:
P1:
P2: ALU
P3: ALU
P4: ALU
P5: ALU
P6 (128 bits): VPU, FP store data, ALU, DIV/SQRT
P7 (128 bits): VPU, ALU, FP store data
P8 (128 bits): VPU, ALU, DIV/SQRT
P9 (128 bits): ALU, VPU
P10 (256 bits): store address, load
P11 (256 bits): store address, load
P12 (256 bits): load
P13 (64 bits): store data
P14 (64 bits): store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	6.50
Overall L1	6.50

Vectorization ratios

all	0%
load	0%
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%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 6.50 cycles. At this rate:

10% of peak load performance is reached (10.46 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
32% of peak store performance is reached (15.38 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the function is: 421860

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput
SUB SP, SP, #96	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STP X29, X30, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR X23, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X22, X21, [SP, #64]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X20, X19, [SP, #80]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
ADD X29, SP, #32	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
LDR X8, [X2]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
SUBS X23, X8, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.LT 421940 <.omp_outlined..15+0xe0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR W19, [X0]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X22, [X3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X20, [X4]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
MOVZ W8, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADRP X0, <4eb894>	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X0, X0, #688	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X3, SP, #12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X4, X29, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
LDR X21, [X5]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
SUB X5, X29, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X6, SP, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ W2, #34	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STR XZR, [X29, #24]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STUR X23, [X29, #504]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR X8, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
MOVZ W7, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR W1, WZR, W19	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STR WZR, [SP, #12]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR X8, [SP]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
BL 40f550 <@plt_start@+0x160>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDUR X8, [X29, #504]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X9, [X29, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X8, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
CSEL X8, X8, X23, #11	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X8, X9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.LT 421930 <.omp_outlined..15+0xd0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
SUB W10, W8, W9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD W10, W10, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
TBNZ W10, #0, 421908 <.omp_outlined..15+0xa8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X10, XZR, X9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X8, X9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.EQ 421930 <.omp_outlined..15+0xd0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
B 421958 <.omp_outlined..15+0xf8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X10, [X22, X9,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
FMOV D1, #1.0000000	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
LDR D0, [X20, X10,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
FCMP D0, D1	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1
B.LE 421924 <.omp_outlined..15+0xc4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
MOVZ W11, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STR X11, [X21, X10,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
ADD X10, X9, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X8, X9	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.NE 421958 <.omp_outlined..15+0xf8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
HINT #0
HINT #0
HINT #0
HINT #0
HINT #0
HINT #0
HINT #0
HINT #0

Name	Coverage (%)	Time (s)
▼.omp_outlined..15–	0.02	0.01
▼Loop 527 - par_coarsen.c:2327-2334 - exec–	0.01	0.01
○Loop 528 - par_coarsen.c:2327-2334 - exec	0.01	0.01

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Expensive FP math instructions/calls

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code

Code clean check

Workaround

Vectorization

Details

Workaround

Expensive FP math instructions/calls

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code