OV - exec - Function update_left(int, int, int, int, double*, bool) [clone ._omp

	Function: update_left(int, int, int, int, double*, bool) [clone ._omp_fn.0]	Module: exec	Source: local_halos.cpp:10-15	Coverage (incl. loops): 0.04% \| (excl. loops): 0.01%

/home/eoseret/qaas/qaas_runs/178-237-4322/intel/TeaLeaf/build/TeaLeaf/src/omp/local_halos.cpp: 10 - 15

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

10:   #pragma omp parallel for

11: #endif

12:   for (int jj = halo_depth; jj < y - halo_depth; ++jj) {

13:     for (int kk = 0; kk < depth; ++kk) {

14:       int base = jj * x;

15:       buffer[base + (halo_depth - kk - 1)] = buffer[base + (halo_depth + kk)];

0x414bc0 STP	X29, X30, [SP, #976]!

0x414bc4 ADD	X29, SP, #0

0x414bc8 STP	X19, X20, [SP, #16]

0x414bcc LDR	W19, [X0, #16]

0x414bd0 STR	X21, [SP, #32]

0x414bd4 ORR	X21, XZR, X0

0x414bd8 BL	410320

0x414bdc ORR	W20, WZR, W0

0x414be0 BL	4102b0

0x414be4 LDR	W1, [X21, #12]

0x414be8 ORR	W8, WZR, W0

0x414bec SUB	W0, W1, W19,LSL #1

0x414bf0 SDIV	W2, W0, W20

0x414bf4 MSUB	W1, W2, W20, W0

0x414bf8 CMP	W8, W1

0x414bfc B.LT	414d00

0x414c00 MADD	W8, W2, W8, W1

0x414c04 ADD	W2, W2, W8

0x414c08 CMP	W8, W2

0x414c0c B.GE	414cf0

0x414c10 LDR	W7, [X21, #20]

0x414c14 ADD	W8, W19, W8

0x414c18 ADD	W9, W19, W2

0x414c1c LDR	X13, [X21]

0x414c20 LDR	W12, [X21, #8]

0x414c24 CMP	W7, #0

0x414c28 B.LE	414cf0

0x414c2c SUB	W0, W7, #1

0x414c30 CNTD	X14, ALL

0x414c34 CMP	W0, #2

0x414c38 SUB	W1, W14, #1

0x414c3c CCMP	W0, W1, #0, #8

0x414c40 B.CC	414d0c

0x414c44 MUL	W11, W12, W8

0x414c48 SBFM	X3, X19, #0, #31

0x414c4c RDVL	X4, #63

0x414c50 SUB	X16, X3, #1

0x414c54 SBFM	X15, X12, #0, #31

0x414c58 PTRUE	P7.B, ALL

0x414c5c SBFM	X12, X12, #61, #31

0x414c60 ADDPL	X6, X7, #63

0x414c64 SBFM	X11, X11, #0, #31

0x414c68 CNTD	X5, ALL

0x414c6c ADD	X3, X3, X11

0x414c70 ADD	X10, X4, X3,LSL #3

0x414c74 ADD	X3, X13, X3,LSL #3

0x414c78 ADD	X10, X13, X10

0x414c7c HINT	#0

(83) 0x414c80 ORR	X2, XZR, X10

(83) 0x414c84 MOVZ	X0, #0

  (82) 0x414c88 LD1D	{Z31.D}, P7/Z, [X3, X0,LSL #3]

  (82) 0x414c8c ORR	X1, XZR, X0

  (82) 0x414c90 REV	Z31.D, Z31.D

  (82) 0x414c94 ADD	X0, X0, X5

  (82) 0x414c98 ST1D	{Z31.D}, P7, [X2, MUL VL]

  (82) 0x414c9c ADD	X2, X2, X4

  (82) 0x414ca0 CMP	W6, W0

  (82) 0x414ca4 B.CS	414c88

(83) 0x414ca8 ADD	W0, W14, W1

(83) 0x414cac CMP	W7, W0

(83) 0x414cb0 B.EQ	414cd8

(83) 0x414cb4 SBFM	X0, X0, #0, #31

(83) 0x414cb8 SUB	X1, X11, X0

(83) 0x414cbc ADD	X1, X1, X16

(83) 0x414cc0 ADD	X1, X13, X1,LSL #3

  (84) 0x414cc4 LDR	D31, [X3, X0,LSL #3]

  (84) 0x414cc8 ADD	X0, X0, #1

  (84) 0x414ccc STR	D31, [X1], #504

  (84) 0x414cd0 CMP	W7, W0

  (84) 0x414cd4 B.GT	414cc4

(83) 0x414cd8 ADD	W8, W8, #1

(83) 0x414cdc ADD	X3, X3, X12

(83) 0x414ce0 ADD	X10, X10, X12

(83) 0x414ce4 ADD	X11, X11, X15

(83) 0x414ce8 CMP	W9, W8

(83) 0x414cec B.GT	414c80

0x414cf0 LDR	X21, [SP, #32]

0x414cf4 LDP	X19, X20, [SP, #16]

0x414cf8 LDP	X29, X30, [SP], #48

0x414cfc RET

0x414d00 ADD	W2, W2, #1

0x414d04 MOVZ	W1, #0

0x414d08 B	414c00

0x414d0c MUL	W0, W12, W8

0x414d10 SBFM	X2, X7, #0, #31

0x414d14 SBFM	X19, X19, #0, #31

0x414d18 ADD	X4, X19, X2

0x414d1c SUB	X2, XZR, X2,LSL #3

0x414d20 SBFM	X7, X12, #60, #31

0x414d24 UBFM	X6, X13, #63, #62

0x414d28 SBFM	X12, X12, #61, #31

0x414d2c SBFM	X0, X0, #0, #31

0x414d30 ADD	X4, X4, X0

0x414d34 ADD	X19, X19, X0

0x414d38 ADD	X4, X13, X4,LSL #3

0x414d3c UBFM	X0, X19, #60, #59

(81) 0x414d40 ADD	X1, X4, X2

(81) 0x414d44 SUB	X5, X0, #8

  (80) 0x414d48 LDR	D31, [X1]

  (80) 0x414d4c SUB	X3, X5, X1

  (80) 0x414d50 ADD	X1, X1, #8

  (80) 0x414d54 STR	D31, [X3, X6]

  (80) 0x414d58 CMP	X1, X4

  (80) 0x414d5c B.NE	414d48

(81) 0x414d60 ADD	W8, W8, #1

(81) 0x414d64 ADD	X0, X0, X7

(81) 0x414d68 ADD	X4, X4, X12

(81) 0x414d6c CMP	W9, W8

(81) 0x414d70 B.GT	414d40

0x414d74 B	414cf0

0x414d78 HINT	#0

0x414d7c HINT	#0

Coverage (%)	Name	Module
►93.92+	omp_fulfill_event	libgomp.so.1.0.0
○	start_thread	libc.so.6
○	thread_start	libc.so.6
►5.08+	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 /

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 0.01% of application time for run gcc_0

Source file and lines	local_halos.cpp:10-15
Module	exec

The function is defined in /home/eoseret/qaas/qaas_runs/178-237-4322/intel/TeaLeaf/build/TeaLeaf/src/omp/local_halos.cpp:10-15.

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 12.75 to 2.00 cycles (6.37x 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 35% of vector register length is used (average across all VPU instructions). By vectorizing your function, you can lower the cost of an iteration from 12.75 to 6.87 cycles (1.85x 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)

No data for this section

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/178-237-4322/intel/TeaLeaf/build/TeaLeaf/src/omp/local_halos.cpp:10
- /home/eoseret/qaas/qaas_runs/178-237-4322/intel/TeaLeaf/build/TeaLeaf/src/omp/local_halos.cpp:10

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 12 bytes.

General properties

nb instructions	71
nb uops	68
loop length	284
used w registers	12
used x registers	20
used b registers	0
used h registers	0
used s registers	1
used d registers	0
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	8.50 cycles
front end	8.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	4.50	4.50	12.75	12.75	12.75	12.75	0.00	0.00	0.00	0.00	3.83	3.50	3.67	1.50	1.50
cycles	4.50	4.50	12.75	12.75	12.75	12.75	0.00	0.00	0.00	0.00	3.83	3.50	3.67	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

5.00-12.50

Cycles summary

Front-end	8.50
Dispatch	12.75
DIV/SQRT	5.00-12.50
Overall L1	12.75

Vectorization ratios

all	0%
load	0%
store	0%
mul	0%
add-sub	0%
fma	0%
div/sqrt	0%
other	0%

Vector efficiency ratios

all	35%
load	25%
store	41%
mul	12%
add-sub	20%
fma	12%
div/sqrt	12%
other	53%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 12.75 cycles. At this rate:

0% of peak load performance is reached (0.94 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 414bc0

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%)
LDR W19, [X0, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (12.5%)
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%)
ORR X21, 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
BL 410320 <@plt_start@+0x300>	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
ORR W20, WZR, W0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
BL 4102b0 <@plt_start@+0x290>	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
LDR W1, [X21, #12]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (12.5%)
ORR W8, WZR, W0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
SUB W0, W1, W19,LSL #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
SDIV W2, W0, W20	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	5-12	5-12.50	scal (12.5%)
MSUB W1, W2, W20, W0	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (12.5%)
CMP W8, W1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (12.5%)
B.LT 414d00 <_Z11update_leftiiiiPdb._omp_fn.0+0x140>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MADD W8, W2, W8, W1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (12.5%)
ADD W2, W2, W8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
CMP W8, W2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (12.5%)
B.GE 414cf0 <_Z11update_leftiiiiPdb._omp_fn.0+0x130>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR W7, [X21, #20]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (12.5%)
ADD W8, W19, W8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
ADD W9, W19, W2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
LDR X13, [X21]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR W12, [X21, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (12.5%)
CMP W7, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (12.5%)
B.LE 414cf0 <_Z11update_leftiiiiPdb._omp_fn.0+0x130>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SUB W0, 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	N/A
CNTD X14, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
CMP W0, #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 (12.5%)
SUB W1, W14, #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 (12.5%)
CCMP W0, W1, #0, #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 (12.5%)
B.CC 414d0c <_Z11update_leftiiiiPdb._omp_fn.0+0x14c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MUL W11, W12, W8	1	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	scal (12.5%)
SBFM X3, X19, #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%)
RDVL X4, #63	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	N/A
SUB X16, X3, #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%)
SBFM X15, X12, #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%)
PTRUE P7.B, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (100.0%)
SBFM X12, X12, #61, #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%)
ADDPL X6, X7, #63	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
SBFM X11, X11, #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%)
CNTD X5, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
ADD X3, X3, X11	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 X10, X4, X3,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 X3, X13, X3,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 X10, X13, X10	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%)
HINT #0																			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	N/A
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	scal (50.0%)
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 W2, W2, #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 (12.5%)
MOVZ W1, #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 (12.5%)
B 414c00 <_Z11update_leftiiiiPdb._omp_fn.0+0x40>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MUL W0, W12, W8	1	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
SBFM X2, X7, #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%)
SBFM X19, X19, #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%)
ADD X4, X19, X2	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, XZR, X2,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%)
SBFM X7, X12, #60, #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%)
UBFM X6, X13, #63, #62	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%)
SBFM X12, X12, #61, #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%)
SBFM X0, 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	N/A
ADD X4, X4, 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
ADD X19, X19, 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%)
ADD X4, X13, 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	N/A
UBFM X0, X19, #60, #59	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
B 414cf0 <_Z11update_leftiiiiPdb._omp_fn.0+0x130>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
HINT #0																			N/A
HINT #0																			N/A

The code analyzed by CQA in that panel excludes loops and represents 0.01% of application time for run gcc_0

Source file and lines	local_halos.cpp:10-15
Module	exec

The function is defined in /home/eoseret/qaas/qaas_runs/178-237-4322/intel/TeaLeaf/build/TeaLeaf/src/omp/local_halos.cpp:10-15.

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 12.75 to 2.00 cycles (6.37x 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)

No data for this section

CALL instructions

Detected function call instructions.

Details

unknown: 2 occurrences►
- /home/eoseret/qaas/qaas_runs/178-237-4322/intel/TeaLeaf/build/TeaLeaf/src/omp/local_halos.cpp:10
- /home/eoseret/qaas/qaas_runs/178-237-4322/intel/TeaLeaf/build/TeaLeaf/src/omp/local_halos.cpp:10

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 12 bytes.

General properties

nb instructions	71
nb uops	68
loop length	284
used w registers	12
used x registers	20
used b registers	0
used h registers	0
used s registers	1
used d registers	0
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	8.50 cycles
front end	8.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	4.50	4.50	12.75	12.75	12.75	12.75	0.00	0.00	0.00	0.00	3.83	3.50	3.67	1.50	1.50
cycles	4.50	4.50	12.75	12.75	12.75	12.75	0.00	0.00	0.00	0.00	3.83	3.50	3.67	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

5.00-12.50

Cycles summary

Front-end	8.50
Dispatch	12.75
DIV/SQRT	5.00-12.50
Overall L1	12.75

Vectorization ratios

all	0%
load	0%
store	0%
mul	0%
add-sub	0%
fma	0%
div/sqrt	0%
other	0%

Vector efficiency ratios

all	35%
load	25%
store	41%
mul	12%
add-sub	20%
fma	12%
div/sqrt	12%
other	53%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 12.75 cycles. At this rate:

0% of peak load performance is reached (0.94 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 414bc0

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%)
LDR W19, [X0, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (12.5%)
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%)
ORR X21, 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
BL 410320 <@plt_start@+0x300>	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
ORR W20, WZR, W0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
BL 4102b0 <@plt_start@+0x290>	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
LDR W1, [X21, #12]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (12.5%)
ORR W8, WZR, W0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
SUB W0, W1, W19,LSL #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
SDIV W2, W0, W20	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	5-12	5-12.50	scal (12.5%)
MSUB W1, W2, W20, W0	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (12.5%)
CMP W8, W1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (12.5%)
B.LT 414d00 <_Z11update_leftiiiiPdb._omp_fn.0+0x140>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MADD W8, W2, W8, W1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (12.5%)
ADD W2, W2, W8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
CMP W8, W2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (12.5%)
B.GE 414cf0 <_Z11update_leftiiiiPdb._omp_fn.0+0x130>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR W7, [X21, #20]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (12.5%)
ADD W8, W19, W8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
ADD W9, W19, W2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
LDR X13, [X21]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR W12, [X21, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (12.5%)
CMP W7, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (12.5%)
B.LE 414cf0 <_Z11update_leftiiiiPdb._omp_fn.0+0x130>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SUB W0, 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	N/A
CNTD X14, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
CMP W0, #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 (12.5%)
SUB W1, W14, #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 (12.5%)
CCMP W0, W1, #0, #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 (12.5%)
B.CC 414d0c <_Z11update_leftiiiiPdb._omp_fn.0+0x14c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MUL W11, W12, W8	1	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	scal (12.5%)
SBFM X3, X19, #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%)
RDVL X4, #63	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	N/A
SUB X16, X3, #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%)
SBFM X15, X12, #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%)
PTRUE P7.B, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (100.0%)
SBFM X12, X12, #61, #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%)
ADDPL X6, X7, #63	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
SBFM X11, X11, #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%)
CNTD X5, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
ADD X3, X3, X11	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 X10, X4, X3,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 X3, X13, X3,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 X10, X13, X10	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%)
HINT #0																			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	N/A
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	scal (50.0%)
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 W2, W2, #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 (12.5%)
MOVZ W1, #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 (12.5%)
B 414c00 <_Z11update_leftiiiiPdb._omp_fn.0+0x40>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MUL W0, W12, W8	1	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
SBFM X2, X7, #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%)
SBFM X19, X19, #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%)
ADD X4, X19, X2	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, XZR, X2,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%)
SBFM X7, X12, #60, #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%)
UBFM X6, X13, #63, #62	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%)
SBFM X12, X12, #61, #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%)
SBFM X0, 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	N/A
ADD X4, X4, 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
ADD X19, X19, 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%)
ADD X4, X13, 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	N/A
UBFM X0, X19, #60, #59	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
B 414cf0 <_Z11update_leftiiiiPdb._omp_fn.0+0x130>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
HINT #0																			N/A
HINT #0																			N/A

Name	Coverage (%)	Time (s)
▼update_left(int, int, int, int, double*, bool) [clone ._omp_fn.0]–	0.04	0.10
▼Loop 81 - local_halos.cpp:13-15 - exec–	0.00	0.00
○Loop 80 - local_halos.cpp:13-15 - exec	0.03	0.09
▼Loop 83 - local_halos.cpp:12-15 - exec–	0.00	0.00
○Loop 84 - local_halos.cpp:13-15 - exec	0.00	0.00
○Loop 82 - local_halos.cpp:13-15 - exec	0.00	0.00

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

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

Code clean check

Workaround

Vectorization

Details

Workaround

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