OV - exec - Function hypre_BoomerAMGCoarsenPMIS._omp

	Function: hypre_BoomerAMGCoarsenPMIS._omp_fn.6	Module: exec	Source: par_coarsen.c:2354-2381	Coverage (incl. loops): 0.72% \| (excl. loops): 0.00%

/home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/par_coarsen.c: 2354 - 2381

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

2354: #pragma omp parallel for private(ig, i, jS, j, jj) HYPRE_SMP_SCHEDULE

2355: #endif

2356:         for (ig = 0; ig < graph_size; ig++)

2357:         {

2358:            i = graph_array[ig];

2359:            if (measure_array[i] > 1)

2360:            {

2361:               for (jS = S_diag_i[i]; jS < S_diag_i[i+1]; jS++)

2362:               {

2363:                  j = S_diag_j[jS];

2364:                  if (measure_array[j] > 1)

2365:                  {

2366:                     if (measure_array[i] > measure_array[j])

2367:                        CF_marker[j] = 0;

2368:                     else if (measure_array[j] > measure_array[i])

2369:                        CF_marker[i] = 0;

2370:                  }

2371:               } /* for each local neighbor j of i */

2372:               for (jS = S_offd_i[i]; jS < S_offd_i[i+1]; jS++)

2373:               {

2374:                  jj = S_offd_j[jS];

2375:                  j = num_variables+jj;

2376:                  if (measure_array[j] > 1)

2377:                  {

2378:                     if (measure_array[i] > measure_array[j])

2379:                        CF_marker_offd[jj] = 0;

2380:                     else if (measure_array[j] > measure_array[i])

2381:                        CF_marker[i] = 0;

0x426f60 STP	X29, X30, [SP, #976]!

0x426f64 ADD	X29, SP, #0

0x426f68 STP	X19, X20, [SP, #16]

0x426f6c ORR	X19, XZR, X0

0x426f70 LDR	X0, [X0, #72]

0x426f74 STR	X21, [SP, #32]

0x426f78 LDR	X21, [X0]

0x426f7c BL	4101b0

0x426f80 SBFM	X20, X0, #0, #31

0x426f84 BL	4101c0

0x426f88 SBFM	X3, X0, #0, #31

0x426f8c SDIV	X10, X21, X20

0x426f90 MSUB	X1, X10, X20, X21

0x426f94 CMP	X3, X1

0x426f98 B.LT	4270c4

0x426f9c MADD	X4, X10, X3, X1

0x426fa0 ADD	X5, X10, X4

0x426fa4 CMP	X4, X5

0x426fa8 B.GE	4270b4

0x426fac LDR	X6, [X19, #64]

0x426fb0 FMOV	D31, #1.0000000

0x426fb4 LDP	X14, X9, [X19]

0x426fb8 LDP	X13, X8, [X19, #16]

0x426fbc ADD	X3, X6, X4,LSL #3

0x426fc0 ADD	X16, X6, X5,LSL #3

0x426fc4 LDP	X7, X11, [X19, #32]

0x426fc8 LDP	X15, X2, [X19, #48]

0x426fcc B	426fdc

(110) 0x426fd0 ADD	X3, X3, #8

(110) 0x426fd4 CMP	X16, X3

(110) 0x426fd8 B.EQ	4270b4

(110) 0x426fdc LDR	X12, [X3]

(110) 0x426fe0 UBFM	X17, X12, #61, #60

(110) 0x426fe4 LDR	D30, [X2, X17]

(110) 0x426fe8 FCMPE	D30, D31

(110) 0x426fec B.LS	426fd0

(110) 0x426ff0 ADD	X18, X17, #8

(110) 0x426ff4 LDR	X20, [X14, X17]

(110) 0x426ff8 ADD	X30, X14, X18

(110) 0x426ffc LDR	X21, [X14, X18]

(110) 0x427000 CMP	X20, X21

(110) 0x427004 B.LT	427024

(110) 0x427008 B	427054

  (112) 0x42700c B.GE	427018

  (112) 0x427010 STR	XZR, [X11, X17]

  (112) 0x427014 LDR	X21, [X30]

  (112) 0x427018 ADD	X20, X20, #1

  (112) 0x42701c CMP	X20, X21

  (112) 0x427020 B.GE	427054

  (112) 0x427024 LDR	X19, [X9, X20,LSL #3]

  (112) 0x427028 UBFM	X0, X19, #61, #60

  (112) 0x42702c LDR	D0, [X2, X0]

  (112) 0x427030 FCMPE	D0, D31

  (112) 0x427034 B.LS	427018

  (112) 0x427038 FCMPE	D30, D0

  (112) 0x42703c B.LS	42700c

  (112) 0x427040 STR	XZR, [X11, X0]

  (112) 0x427044 ADD	X20, X20, #1

  (112) 0x427048 LDR	X21, [X30]

  (112) 0x42704c CMP	X20, X21

  (112) 0x427050 B.LT	427024

(110) 0x427054 LDR	X6, [X13, X17]

(110) 0x427058 ADD	X10, X13, X18

(110) 0x42705c LDR	X5, [X13, X18]

(110) 0x427060 CMP	X6, X5

(110) 0x427064 B.LT	42708c

(110) 0x427068 B	426fd0

  (111) 0x42706c B.GE	427080

  (111) 0x427070 STR	XZR, [X11, X17]

  (111) 0x427074 LDR	X5, [X10]

  (111) 0x427078 HINT	#0

  (111) 0x42707c HINT	#0

  (111) 0x427080 ADD	X6, X6, #1

  (111) 0x427084 CMP	X6, X5

  (111) 0x427088 B.GE	426fd0

  (111) 0x42708c LDR	X1, [X8, X6,LSL #3]

  (111) 0x427090 ADD	X4, X7, X1

  (111) 0x427094 LDR	D29, [X2, X4,LSL #3]

  (111) 0x427098 FCMPE	D29, D31

  (111) 0x42709c B.LS	427080

  (111) 0x4270a0 FCMPE	D30, D29

  (111) 0x4270a4 B.LS	42706c

  (111) 0x4270a8 STR	XZR, [X15, X1,LSL #3]

  (111) 0x4270ac LDR	X5, [X10]

  (111) 0x4270b0 B	427080

0x4270b4 LDR	X21, [SP, #32]

0x4270b8 LDP	X19, X20, [SP, #16]

0x4270bc LDP	X29, X30, [SP], #48

0x4270c0 RET

0x4270c4 ADD	X10, X10, #1

0x4270c8 MOVZ	X1, #0

0x4270cc B	426f9c

Coverage (%)	Name	Source Location	Module
►98.71+	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►1.19+	GOMP_parallel		libgomp.so.1.0.0
○	hypre_BoomerAMGCoarsenPMIS	par_coarsen.c:2393	exec
○	hypre_BoomerAMGSetup	par_amg_setup.c:612	exec
○	hypre_PCGSetup	pcg.c:234	exec
○	main	amg.c:398	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	amg.c:253	exec

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 /

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

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run gcc_1

Source file and lines	par_coarsen.c:2354-2381
Module	exec

The function is defined in /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/par_coarsen.c:2354-2381.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Your function is not vectorized. Only 39% of vector register length is used (average across all VPU instructions).

Details

All VPU instructions are used in scalar version (process only one data element in vector registers).

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 20.00 to 4.50 cycles (4.44x speedup).

CALL instructions

Detected function call instructions.

Details

Calling (and then returning from) a function prevents many compiler optimizations (like vectorization), breaks control flow (which reduces pipeline performance) and executes extra instructions to save/restore the registers used inside it, which is very expensive (dozens of cycles). Consider to inline small functions.

unknown: 2 occurrences►
- /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/par_coarsen.c:2354
- /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/par_coarsen.c:2354

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 does not load or store any data.

General properties

nb instructions	35
nb uops	35
loop length	140
used w registers	0
used x registers	22
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	6

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	4.38 cycles
front end	4.38 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	3.50	3.50	4.25	4.25	4.25	4.25	0.25	0.25	0.25	0.25	4.50	4.17	4.33	1.50	1.50
cycles	3.50	3.50	4.25	4.25	4.25	4.25	0.25	0.25	0.25	0.25	4.50	4.17	4.33	1.50	1.50

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

5.00-20.00

Cycles summary

Front-end	4.38
Dispatch	4.50
DIV/SQRT	5.00-20.00
Overall L1	5.00-20.00

Vectorization ratios

INT

all	0%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	0%
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	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

INT+FP

all	0%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	0%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

Vector efficiency ratios

INT

all	39%
load	40%
store	41%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	25%
other	55%

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

INT+FP

all	39%
load	40%
store	41%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	25%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	50%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 20.00 cycles.

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 426f60

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
STP X29, X30, [SP, #976]!	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X29, SP, #0	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%)
STP X19, X20, [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	scal (50.0%)
ORR X19, XZR, X0	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, [X0, #72]	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 X21, [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	scal (25.0%)
LDR X21, [X0]	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%)
BL 4101b0 <@plt_start@+0x190>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SBFM X20, X0, #0, #31	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (100.0%)
BL 4101c0 <@plt_start@+0x1a0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SBFM X3, X0, #0, #31	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (100.0%)
SDIV X10, X21, X20	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	5-20	5-20	N/A
MSUB X1, X10, X20, X21	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
CMP X3, X1	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.LT 4270c4 <hypre_BoomerAMGCoarsenPMIS._omp_fn.6+0x164>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MADD X4, X10, X3, X1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
ADD X5, X10, X4	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%)
CMP X4, X5	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.GE 4270b4 <hypre_BoomerAMGCoarsenPMIS._omp_fn.6+0x154>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X6, [X19, #64]	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%)
FMOV D31, #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	scal (25.0%)
LDP X14, X9, [X19]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
LDP X13, X8, [X19, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
ADD X3, X6, X4,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%)
ADD X16, X6, X5,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%)
LDP X7, X11, [X19, #32]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
LDP X15, X2, [X19, #48]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
B 426fdc <hypre_BoomerAMGCoarsenPMIS._omp_fn.6+0x7c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X21, [SP, #32]	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%)
LDP X19, X20, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X29, X30, [SP], #48	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
RET	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X10, X10, #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
MOVZ X1, #0	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%)
B 426f9c <hypre_BoomerAMGCoarsenPMIS._omp_fn.6+0x3c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run gcc_1

Source file and lines	par_coarsen.c:2354-2381
Module	exec

The function is defined in /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/par_coarsen.c:2354-2381.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Your function is not vectorized. Only 39% of vector register length is used (average across all VPU instructions).

Details

All VPU instructions are used in scalar version (process only one data element in vector registers).

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 20.00 to 4.50 cycles (4.44x speedup).

CALL instructions

Detected function call instructions.

Details

unknown: 2 occurrences►
- /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/par_coarsen.c:2354
- /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/par_coarsen.c:2354

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 does not load or store any data.

General properties

nb instructions	35
nb uops	35
loop length	140
used w registers	0
used x registers	22
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	6

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	4.38 cycles
front end	4.38 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	3.50	3.50	4.25	4.25	4.25	4.25	0.25	0.25	0.25	0.25	4.50	4.17	4.33	1.50	1.50
cycles	3.50	3.50	4.25	4.25	4.25	4.25	0.25	0.25	0.25	0.25	4.50	4.17	4.33	1.50	1.50

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

5.00-20.00

Cycles summary

Front-end	4.38
Dispatch	4.50
DIV/SQRT	5.00-20.00
Overall L1	5.00-20.00

Vectorization ratios

INT

all	0%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	0%
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	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

INT+FP

all	0%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	0%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

Vector efficiency ratios

INT

all	39%
load	40%
store	41%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	25%
other	55%

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

INT+FP

all	39%
load	40%
store	41%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	25%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	50%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 20.00 cycles.

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 426f60

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
STP X29, X30, [SP, #976]!	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X29, SP, #0	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%)
STP X19, X20, [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	scal (50.0%)
ORR X19, XZR, X0	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, [X0, #72]	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 X21, [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	scal (25.0%)
LDR X21, [X0]	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%)
BL 4101b0 <@plt_start@+0x190>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SBFM X20, X0, #0, #31	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (100.0%)
BL 4101c0 <@plt_start@+0x1a0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SBFM X3, X0, #0, #31	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (100.0%)
SDIV X10, X21, X20	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	5-20	5-20	N/A
MSUB X1, X10, X20, X21	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
CMP X3, X1	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.LT 4270c4 <hypre_BoomerAMGCoarsenPMIS._omp_fn.6+0x164>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MADD X4, X10, X3, X1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
ADD X5, X10, X4	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%)
CMP X4, X5	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.GE 4270b4 <hypre_BoomerAMGCoarsenPMIS._omp_fn.6+0x154>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X6, [X19, #64]	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%)
FMOV D31, #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	scal (25.0%)
LDP X14, X9, [X19]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
LDP X13, X8, [X19, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
ADD X3, X6, X4,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%)
ADD X16, X6, X5,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%)
LDP X7, X11, [X19, #32]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
LDP X15, X2, [X19, #48]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
B 426fdc <hypre_BoomerAMGCoarsenPMIS._omp_fn.6+0x7c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X21, [SP, #32]	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%)
LDP X19, X20, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X29, X30, [SP], #48	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
RET	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADD X10, X10, #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
MOVZ X1, #0	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%)
B 426f9c <hypre_BoomerAMGCoarsenPMIS._omp_fn.6+0x3c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Name	Coverage (%)	Time (s)
▼hypre_BoomerAMGCoarsenPMIS._omp_fn.6–	0.72	0.30
▼Loop 110 - par_coarsen.c:2354-2381 - exec–	0.05	0.02
○Loop 112 - par_coarsen.c:2361-2369 - exec	0.66	0.21
○Loop 111 - par_coarsen.c:2372-2381 - exec	0.00	0.00

Report Configuration

Vectorization

Details

Expensive FP math instructions/calls

CALL instructions

Details

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

Vectorization

Details

Expensive FP math instructions/calls

CALL instructions

Details

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code