OV - exec - Function advec_cell_kernel(int, int, int, int, int, int, clover::Buffer1D<double>&, clover::Buffer1D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&) [clone .omp

	Function: advec_cell_kernel(int, int, int, int, int, int, clover::Buffer1D<double>&, clover::Buffer1 ...	Module: exec	Source: advec_cell.cpp:65-110 [...]	Coverage (incl. loops): 2.92% \| (excl. loops): 0.00%

/home/eoseret/qaas/qaas_runs/178-219-7589/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_cell.cpp: 65 - 110

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

65: #pragma omp parallel for simd collapse(2)

66:     for (int j = (y_min + 1); j < (y_max + 2); j++) {

67:       for (int i = (x_min + 1); i < (x_max + 2 + 2); i++)

68:         ({

69:           int upwind, donor, downwind, dif;

70:           double sigmat, sigma3, sigma4, sigmav, sigmam, diffuw, diffdw, limiter, wind;

71:           if (vol_flux_x(i, j) > 0.0) {

72:             upwind = i - 2;

73:             donor = i - 1;

74:             downwind = i;

75:             dif = donor;

76:           } else {

77:             upwind = std::min(i + 1, x_max + 2);

78:             donor = i;

79:             downwind = i - 1;

80:             dif = upwind;

81:           }

82:           sigmat = std::fabs(vol_flux_x(i, j)) / pre_vol(donor, j);

83:           sigma3 = (1.0 + sigmat) * (vertexdx[i] / vertexdx[dif]);

84:           sigma4 = 2.0 - sigmat;

85:           sigmav = sigmat;

86:           diffuw = density1(donor, j) - density1(upwind, j);

87:           diffdw = density1(downwind, j) - density1(donor, j);

88:           wind = 1.0;

89:           if (diffdw <= 0.0) wind = -1.0;

90:           if (diffuw * diffdw > 0.0) {

91:             limiter = (1.0 - sigmav) * wind *

92:                       std::fmin(std::fmin(std::fabs(diffuw), std::fabs(diffdw)),

93:                                 one_by_six * (sigma3 * std::fabs(diffuw) + sigma4 * std::fabs(diffdw)));

94:           } else {

95:             limiter = 0.0;

96:           }

97:           mass_flux_x(i, j) = vol_flux_x(i, j) * (density1(donor, j) + limiter);

98:           sigmam = std::fabs(mass_flux_x(i, j)) / (density1(donor, j) * pre_vol(donor, j));

99:           diffuw = energy1(donor, j) - energy1(upwind, j);

100:           diffdw = energy1(downwind, j) - energy1(donor, j);

101:           wind = 1.0;

102:           if (diffdw <= 0.0) wind = -1.0;

103:           if (diffuw * diffdw > 0.0) {

104:             limiter = (1.0 - sigmam) * wind *

105:                       std::fmin(std::fmin(std::fabs(diffuw), std::fabs(diffdw)),

106:                                 one_by_six * (sigma3 * std::fabs(diffuw) + sigma4 * std::fabs(diffdw)));

107:           } else {

108:             limiter = 0.0;

109:           }

110:           ener_flux(i, j) = mass_flux_x(i, j) * (energy1(donor, j) + limiter);

/home/eoseret/qaas/qaas_runs/178-219-7589/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h: 69 - 69

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

69:   T &operator()(size_t i, size_t j) const { return data[i + j * sizeX]; }

0x427f30 STP	D15, D14, [SP, #864]!

0x427f34 STP	D13, D12, [SP, #16]

0x427f38 STP	D11, D10, [SP, #32]

0x427f3c STP	D9, D8, [SP, #48]

0x427f40 STP	X29, X30, [SP, #64]

0x427f44 STP	X28, X27, [SP, #80]

0x427f48 STP	X26, X25, [SP, #96]

0x427f4c STP	X24, X23, [SP, #112]

0x427f50 STP	X22, X21, [SP, #128]

0x427f54 STP	X20, X19, [SP, #144]

0x427f58 ADD	X29, SP, #64

0x427f5c SUB	SP, SP, #48

0x427f60 ADDVL	SP, SP, #61

0x427f64 LDR	W8, [X2]

0x427f68 LDR	W23, [X4]

0x427f6c LDR	W9, [X5]

0x427f70 ADD	W24, W8, #1

0x427f74 LDR	W8, [X3]

0x427f78 ADD	W25, W23, #1

0x427f7c ADD	W9, W9, #4

0x427f80 ADD	W8, W8, #2

0x427f84 SUBS	W8, W8, W24

0x427f88 CCMP	W9, W25, #4, #12

0x427f8c B.LE	4282cc

0x427f90 LDR	W19, [X0]

0x427f94 SUB	W26, W9, W25

0x427f98 MOVN	X9, #0

0x427f9c ORR	X20, XZR, X7

0x427fa0 ORR	X21, XZR, X6

0x427fa4 ORR	X22, XZR, X5

0x427fa8 ADRP	X0,

0x427fac ADD	X0, X0, #3192

0x427fb0 UMADDL	X27, W26, W8, X9

0x427fb4 MOVZ	W8, #1

0x427fb8 ADD	X3, SP, #20

0x427fbc ADD	X4, SP, #40

0x427fc0 STR	WZR, [SP, #20]

0x427fc4 STR	X8, [SP, #24]

0x427fc8 ADD	X5, SP, #32

0x427fcc ADD	X6, SP, #24

0x427fd0 MOVZ	W2, #34

0x427fd4 MOVZ	W7, #1

0x427fd8 STR	X8, [SP]

0x427fdc STP	X27, XZR, [SP, #32]

0x427fe0 ORR	W1, WZR, W19

0x427fe4 BL	410680

0x427fe8 LDP	X8, X12, [SP, #32]

0x427fec CMP	X8, X27

0x427ff0 CSEL	X10, X8, X27, #11

0x427ff4 CMP	X12, X10

0x427ff8 STR	X10, [SP, #32]

0x427ffc B.GT	4282bc

0x428000 PTRUE	P1.D, ALL

0x428004 LDR	X16, [X21, #16]

0x428008 LDP	X13, X11, [X29, #128]

0x42800c LDR	X14, [X29, #120]

0x428010 CNTD	X8, ALL

0x428014 LDP	X9, X15, [X29, #96]

0x428018 SUB	X10, X10, X12

0x42801c LDR	W3, [X22]

0x428020 ADD	X18, X10, #1

0x428024 INDEX	Z30.D, X12, #1

0x428028 LDR	X17, [X20, #16]

0x42802c LD1RD	{Z0.D}, P1/Z, [X21]

0x428030 SUB	X21, X29, #64

0x428034 SUBS	X10, X18, X8

0x428038 CSEL	X10, XZR, X10, #3

0x42803c LDR	X0, [X14, #16]

0x428040 WHILELO	P0.D, XZR, X18

0x428044 LDR	X18, [X15, #16]

0x428048 LDR	X1, [X13, #16]

0x42804c LDR	X2, [X11, #16]

0x428050 ADD	W3, W3, #2

0x428054 LD1RD	{Z3.D}, P1/Z, [X14]

0x428058 LD1RD	{Z5.D}, P1/Z, [X13]

0x42805c LD1RD	{Z6.D}, P1/Z, [X11]

0x428060 ORR	X13, XZR, #0x55

0x428064 LDR	X9, [X9, #8]

0x428068 SUB	X11, XZR, X8

0x42806c MOVK	X13, #16325

0x428070 STR	Z0, [X21, #511, MUL VL]

0x428074 LD1RD	{Z0.D}, P1/Z, [X20]

0x428078 SUB	X20, X29, #64

0x42807c DUP	Z21.D, X3

0x428080 DUP	Z4.D, X26

0x428084 DUP	Z7.D, X24

0x428088 SXTW	Z21.D, P1/M, Z21.D

0x42808c STR	Z0, [X20, #510, MUL VL]

0x428090 LD1RD	{Z0.D}, P1/Z, [X15]

0x428094 DUP	Z16.D, X25

0x428098 DUP	Z17.D, X23

0x42809c DUP	Z18.D, X8

0x4280a0 DUP	Z20.D, X16

0x4280a4 DUP	Z23.D, X17

0x4280a8 DUP	Z25.D, X18

0x4280ac SUB	X15, X29, #64

0x4280b0 STR	Z0, [X15, #509, MUL VL]

0x4280b4 DUP	Z26.D, X13

0x4280b8 DUP	Z27.D, X0

0x4280bc DUP	Z28.D, X1

0x4280c0 DUP	Z29.D, X2

(150) 0x4280c4 DUP	Z0.D, #1

(150) 0x4280c8 SUB	X12, X29, #64

(150) 0x4280cc ADD	X11, X11, X8

(150) 0x4280d0 SEL	Z31.D, P0, Z4.D, Z0.D

(150) 0x4280d4 LDR	Z0, [X12, #511, MUL VL]

(150) 0x4280d8 SDIVR	Z31.D, P1/M, Z31.D, Z30.D

(150) 0x4280dc MOVPRFX	Z10, Z30

(150) 0x4280e0 MLS	Z10.D, P1/M, Z31.D, Z4.D

(150) 0x4280e4 ADD	Z8.D, Z7.D, Z31.D

(150) 0x4280e8 ADD	Z30.D, Z30.D, Z18.D

(150) 0x4280ec SXTW	Z8.D, P1/M, Z8.D

(150) 0x4280f0 MOVPRFX	Z9, Z0

(150) 0x4280f4 MUL	Z9.D, P1/M, Z9.D, Z8.D

(150) 0x4280f8 DUPM	Z0.D, #0xfffffffe

(150) 0x4280fc ADD	Z31.D, Z16.D, Z10.D

(150) 0x428100 ADD	Z10.D, Z17.D, Z10.D

(150) 0x428104 ADR	Z12.D, [Z20, Z31.D,SXTW #3]

(150) 0x428108 ORR	Z14.D, Z31.D, Z31.D

(150) 0x42810c ADD	Z13.D, Z31.D, Z0.D

(150) 0x428110 LDR	Z0, [X12, #510, MUL VL]

(150) 0x428114 MOVPRFX	Z11, Z31

(150) 0x428118 SXTW	Z11.D, P1/M, Z31.D

(150) 0x42811c MOVPRFX	Z15, Z10

(150) 0x428120 SXTW	Z15.D, P1/M, Z10.D

(150) 0x428124 ADR	Z9.D, [Z12, Z9.D,LSL #3]

(150) 0x428128 ADD	Z14.D, Z14.D, #1

(150) 0x42812c SXTW	Z14.D, P1/M, Z14.D

(150) 0x428130 SMIN	Z14.D, P1/M, Z14.D, Z21.D

(150) 0x428134 LD1D	{Z1.D}, P0/Z, [V9.D]

(150) 0x428138 LD1D	{Z12.D}, P0/Z, [X9, Z31.D,SXTW #3]

(150) 0x42813c FCMGT	P2.D, P1/Z, Z1.D, #0.0000000

(150) 0x428140 MOVPRFX	Z9, Z1

(150) 0x428144 FABS	Z9.D, P1/M, Z1.D

(150) 0x428148 SEL	Z19.D, P2, Z11.D, Z15.D

(150) 0x42814c SEL	Z11.D, P2, Z15.D, Z11.D

(150) 0x428150 SEL	Z15.D, P2, Z10.D, Z14.D

(150) 0x428154 MOVPRFX	Z10, Z0

(150) 0x428158 MUL	Z10.D, P1/M, Z10.D, Z8.D

(150) 0x42815c SEL	Z13.D, P2, Z13.D, Z14.D

(150) 0x428160 LDR	Z0, [X12, #509, MUL VL]

(150) 0x428164 ADR	Z14.D, [Z23, Z11.D,LSL #3]

(150) 0x428168 LD1D	{Z15.D}, P0/Z, [X9, Z15.D,SXTW #3]

(150) 0x42816c ADR	Z24.D, [Z25, Z11.D,LSL #3]

(150) 0x428170 ADR	Z10.D, [Z14, Z10.D,LSL #3]

(150) 0x428174 MOVPRFX	Z2, Z0

(150) 0x428178 MUL	Z2.D, P1/M, Z2.D, Z8.D

(150) 0x42817c LD1D	{Z14.D}, P0/Z, [V10.D]

(150) 0x428180 ADR	Z24.D, [Z24, Z2.D,LSL #3]

(150) 0x428184 FDIV	Z9.D, P1/M, Z9.D, Z14.D

(150) 0x428188 ADR	Z14.D, [Z25, Z13.D,SXTW #3]

(150) 0x42818c ADR	Z14.D, [Z14, Z2.D,LSL #3]

(150) 0x428190 LD1D	{Z0.D}, P0/Z, [V24.D]

(150) 0x428194 LD1D	{Z14.D}, P0/Z, [V14.D]

(150) 0x428198 FMAD	Z12.D, P1/M, Z9.D, Z12.D

(150) 0x42819c FDIV	Z12.D, P1/M, Z12.D, Z15.D

(150) 0x4281a0 ADR	Z15.D, [Z25, Z19.D,LSL #3]

(150) 0x4281a4 ADR	Z19.D, [Z28, Z19.D,LSL #3]

(150) 0x4281a8 ADR	Z2.D, [Z15, Z2.D,LSL #3]

(150) 0x4281ac FSUB	Z15.D, Z0.D, Z14.D

(150) 0x4281b0 FABD	Z14.D, P1/M, Z14.D, Z0.D

(150) 0x4281b4 LD1D	{Z2.D}, P0/Z, [V2.D]

(150) 0x4281b8 FSUB	Z22.D, Z2.D, Z0.D

(150) 0x4281bc FABD	Z2.D, P1/M, Z2.D, Z0.D

(150) 0x4281c0 FMUL	Z15.D, Z22.D, Z15.D

(150) 0x4281c4 FCMGT	P3.D, P1/Z, Z22.D, #0.0000000

(150) 0x4281c8 FCMGT	P2.D, P1/Z, Z15.D, #0.0000000

(150) 0x4281cc FDUP	Z15.D, #0

(150) 0x4281d0 FSUB	Z15.D, Z15.D, Z9.D

(150) 0x4281d4 FSUBR	Z9.D, P1/M, Z9.D, #0.0000000

(150) 0x4281d8 MOVPRFX	Z22, Z9

(150) 0x4281dc FNEG	Z22.D, P1/M, Z9.D

(150) 0x4281e0 SEL	Z22.D, P3, Z9.D, Z22.D

(150) 0x4281e4 MOVPRFX	Z9, Z14

(150) 0x4281e8 FMINNM	Z9.D, P1/M, Z9.D, Z2.D

(150) 0x4281ec FMUL	Z14.D, Z14.D, Z12.D

(150) 0x4281f0 FMAD	Z2.D, P1/M, Z15.D, Z14.D

(150) 0x4281f4 FMUL	Z2.D, Z2.D, Z26.D

(150) 0x4281f8 FMINNM	Z2.D, P1/M, Z2.D, Z9.D

(150) 0x4281fc FMLA	Z0.D, P2/M, Z2.D, Z22.D

(150) 0x428200 ADR	Z2.D, [Z27, Z31.D,SXTW #3]

(150) 0x428204 ADR	Z22.D, [Z28, Z13.D,SXTW #3]

(150) 0x428208 FMUL	Z0.D, Z0.D, Z1.D

(150) 0x42820c MOVPRFX	Z1, Z3

(150) 0x428210 MUL	Z1.D, P1/M, Z1.D, Z8.D

(150) 0x428214 ADR	Z1.D, [Z2, Z1.D,LSL #3]

(150) 0x428218 ADR	Z2.D, [Z28, Z11.D,LSL #3]

(150) 0x42821c ST1D	{Z0.D}, P0, [V1.D]

(150) 0x428220 MOVPRFX	Z1, Z5

(150) 0x428224 MUL	Z1.D, P1/M, Z1.D, Z8.D

(150) 0x428228 MUL	Z8.D, P1/M, Z8.D, Z6.D

(150) 0x42822c ADR	Z2.D, [Z2, Z1.D,LSL #3]

(150) 0x428230 ADR	Z22.D, [Z22, Z1.D,LSL #3]

(150) 0x428234 ADR	Z1.D, [Z19, Z1.D,LSL #3]

(150) 0x428238 LD1D	{Z2.D}, P0/Z, [V2.D]

(150) 0x42823c LD1D	{Z22.D}, P0/Z, [V22.D]

(150) 0x428240 LD1D	{Z1.D}, P0/Z, [V1.D]

(150) 0x428244 FSUB	Z9.D, Z2.D, Z22.D

(150) 0x428248 FSUB	Z19.D, Z1.D, Z2.D

(150) 0x42824c FABD	Z22.D, P1/M, Z22.D, Z2.D

(150) 0x428250 FABD	Z1.D, P1/M, Z1.D, Z2.D

(150) 0x428254 FMUL	Z9.D, Z19.D, Z9.D

(150) 0x428258 FCMGT	P4.D, P1/Z, Z19.D, #0.0000000

(150) 0x42825c FCMGT	P2.D, P1/Z, Z9.D, #0.0000000

(150) 0x428260 AND	P3.B, P0/Z, P0.B, P2.B

(150) 0x428264 LD1D	{Z24.D}, P3/Z, [V24.D]

(150) 0x428268 LD1D	{Z19.D}, P3/Z, [V10.D]

(150) 0x42826c FMUL	Z19.D, Z19.D, Z24.D

(150) 0x428270 MOVPRFX	Z24, Z0

(150) 0x428274 FABS	Z24.D, P1/M, Z0.D

(150) 0x428278 FDIVR	Z19.D, P1/M, Z19.D, Z24.D

(150) 0x42827c FSUBR	Z19.D, P1/M, Z19.D, #0.0000000

(150) 0x428280 MOVPRFX	Z24, Z19

(150) 0x428284 FNEG	Z24.D, P1/M, Z19.D

(150) 0x428288 SEL	Z19.D, P4, Z19.D, Z24.D

(150) 0x42828c FMUL	Z24.D, Z22.D, Z12.D

(150) 0x428290 FMLA	Z24.D, P1/M, Z1.D, Z15.D

(150) 0x428294 FMINNM	Z1.D, P1/M, Z1.D, Z22.D

(150) 0x428298 FMUL	Z22.D, Z24.D, Z26.D

(150) 0x42829c FMINNM	Z1.D, P1/M, Z1.D, Z22.D

(150) 0x4282a0 FMLA	Z2.D, P2/M, Z19.D, Z1.D

(150) 0x4282a4 ADR	Z1.D, [Z29, Z31.D,SXTW #3]

(150) 0x4282a8 ADR	Z1.D, [Z1, Z8.D,LSL #3]

(150) 0x4282ac FMUL	Z0.D, Z2.D, Z0.D

(150) 0x4282b0 ST1D	{Z0.D}, P0, [V1.D]

(150) 0x4282b4 WHILELO	P0.D, X11, X10

(150) 0x4282b8 B.MI	4280c4

0x4282bc ADRP	X0,

0x4282c0 ADD	X0, X0, #3216

0x4282c4 ORR	W1, WZR, W19

0x4282c8 BL	4105b0

0x4282cc ADDVL	SP, SP, #3

0x4282d0 ADD	SP, SP, #48

0x4282d4 LDP	D9, D8, [SP, #48]

0x4282d8 LDP	D11, D10, [SP, #32]

0x4282dc LDP	D13, D12, [SP, #16]

0x4282e0 LDP	X20, X19, [SP, #144]

0x4282e4 LDP	X22, X21, [SP, #128]

0x4282e8 LDP	X24, X23, [SP, #112]

0x4282ec LDP	X26, X25, [SP, #96]

0x4282f0 LDP	X28, X27, [SP, #80]

0x4282f4 LDP	X29, X30, [SP, #64]

0x4282f8 LDP	D15, D14, [SP], #160

0x4282fc RET

Coverage (%)	Name	Module
►98.42+	__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 /

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 armclang_1

Source file and lines	advec_cell.cpp:65-110
Module	exec

The function is defined in:

/home/eoseret/qaas/qaas_runs/178-219-7589/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_cell.cpp:65-110
/home/eoseret/qaas/qaas_runs/178-219-7589/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h:69

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 19.00 to 16.00 cycles (1.19x 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 33% of vector register length is used (average across all VPU instructions). By fully vectorizing your function, you can lower the cost of an iteration from 19.00 to 9.75 cycles (1.95x speedup).

Details

16% of VPU instructions are used in vector version (process two or more data elements in vector registers):

18% of VPU loads are used in vector version.
38% of VPU stores are used in vector version.
0% of VPU addition or subtraction instructions are used in vector version.
0% of VPU fused multiply-add instructions are used in vector version.
4% of VPU instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

Since your execution units are vector units, only a fully 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/178-219-7589/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_cell.cpp:65
- /home/eoseret/qaas/qaas_runs/178-219-7589/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_cell.cpp:65

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 18 bytes. The binary function is storing 99 bytes.

General properties

nb instructions	118
nb uops	118
loop length	472
used w registers	12
used x registers	32
used b registers	0
used h registers	0
used s registers	1
used d registers	8
used q registers	0
used v registers	0
used z registers	18
nb stack references	26

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	14.75 cycles
front end	14.75 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	2.50	2.50	15.00	15.00	19.00	15.00	3.00	3.00	1.50	1.50	17.00	17.00	17.00	5.50	5.50
cycles	2.50	2.50	15.00	15.00	19.00	15.00	3.00	3.00	1.50	1.50	17.00	17.00	17.00	5.50	5.50

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

Cycles summary

Front-end	14.75
Dispatch	19.00
Overall L1	19.00

Vectorization ratios

all	16%
load	18%
store	38%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	0%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	4%

Vector efficiency ratios

all	33%
load	31%
store	52%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	18%
fma	25%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	28%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 19.00 cycles. At this rate:

0% of peak load performance is reached (0.95 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
10% of peak store performance is reached (5.21 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 427f30

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
STP D15, D14, [SP, #864]!	1	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	vect (50.0%)
STP D13, D12, [SP, #16]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	vect (50.0%)
STP D11, D10, [SP, #32]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	vect (50.0%)
STP D9, D8, [SP, #48]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	vect (50.0%)
STP X29, X30, [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	scal (50.0%)
STP X28, X27, [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	scal (50.0%)
STP X26, X25, [SP, #96]	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%)
STP X24, X23, [SP, #112]	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%)
STP X22, X21, [SP, #128]	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%)
STP X20, X19, [SP, #144]	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%)
ADD X29, SP, #64	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 SP, SP, #48	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
ADDVL SP, SP, #61	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	N/A
LDR W8, [X2]	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%)
LDR W23, [X4]	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%)
LDR W9, [X5]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD W24, 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	scal (12.5%)
LDR W8, [X3]	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 W25, W23, #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%)
ADD W9, W9, #4	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 W8, W8, #2	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%)
SUBS W8, W8, W24	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%)
CCMP W9, W25, #4, #12	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.LE 4282cc <_Z17advec_cell_kerneliiiiiiRN6clover8Buffer1DIdEES2_RNS_8Buffer2DIdEES5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_.omp_outlined.3+0x39c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
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	scal (12.5%)
SUB W26, W9, W25	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%)
MOVN X9, #0	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
ORR X20, XZR, X7	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
ORR X21, XZR, X6	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
ORR X22, XZR, X5	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
ADRP X0, <47cfa8>	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 X0, X0, #3192	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
UMADDL X27, W26, W8, X9	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
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	scal (12.5%)
ADD X3, SP, #20	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, SP, #40	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
STR WZR, [SP, #20]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (12.5%)
STR X8, [SP, #24]	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%)
ADD X5, 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	N/A
ADD X6, SP, #24	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%)
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	N/A
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	scal (12.5%)
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	scal (25.0%)
STP X27, XZR, [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 (50.0%)
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	scal (12.5%)
BL 410680 <@plt_start@+0x660>	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
LDP X8, X12, [SP, #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%)
CMP X8, X27	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%)
CSEL X10, X8, X27, #11	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 X12, X10	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%)
STR X10, [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%)
B.GT 4282bc <_Z17advec_cell_kerneliiiiiiRN6clover8Buffer1DIdEES2_RNS_8Buffer2DIdEES5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_.omp_outlined.3+0x38c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PTRUE P1.D, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (100.0%)
LDR X16, [X21, #16]	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 X13, X11, [X29, #128]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDR X14, [X29, #120]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CNTD X8, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
LDP X9, X15, [X29, #96]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
SUB X10, X10, X12	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%)
LDR W3, [X22]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X18, 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	scal (25.0%)
INDEX Z30.D, X12, #1	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	5	2	N/A
LDR X17, [X20, #16]	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%)
LD1RD {Z0.D}, P1/Z, [X21]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
SUB X21, X29, #64	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
SUBS X10, X18, X8	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%)
CSEL X10, XZR, X10, #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%)
LDR X0, [X14, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
WHILELO P0.D, XZR, X18	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	2	N/A
LDR X18, [X15, #16]	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%)
LDR X1, [X13, #16]	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%)
LDR X2, [X11, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD W3, W3, #2	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
LD1RD {Z3.D}, P1/Z, [X14]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
LD1RD {Z5.D}, P1/Z, [X13]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
LD1RD {Z6.D}, P1/Z, [X11]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
ORR X13, XZR, #0x55	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 X9, [X9, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
SUB X11, XZR, X8	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
MOVK X13, #16325	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
STR Z0, [X21, #511, MUL VL]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	0	2	0.50	vect (100.0%)
LD1RD {Z0.D}, P1/Z, [X20]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
SUB X20, X29, #64	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
DUP Z21.D, X3	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z4.D, X26	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z7.D, X24	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
SXTW Z21.D, P1/M, Z21.D	1	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	2	1	vect (100.0%)
STR Z0, [X20, #510, MUL VL]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	0	2	0.50	vect (100.0%)
LD1RD {Z0.D}, P1/Z, [X15]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
DUP Z16.D, X25	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z17.D, X23	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z18.D, X8	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z20.D, X16	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z23.D, X17	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z25.D, X18	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
SUB X15, X29, #64	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
STR Z0, [X15, #509, MUL VL]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	0	2	0.50	vect (100.0%)
DUP Z26.D, X13	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z27.D, X0	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z28.D, X1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z29.D, X2	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
ADRP X0, <47c2bc>	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 X0, X0, #3216	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
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	scal (12.5%)
BL 4105b0 <@plt_start@+0x590>	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
ADDVL SP, SP, #3	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	N/A
ADD SP, SP, #48	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
LDP D9, D8, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (50.0%)
LDP D11, D10, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (50.0%)
LDP D13, D12, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (50.0%)
LDP X20, X19, [SP, #144]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X22, X21, [SP, #128]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X24, X23, [SP, #112]	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 X26, X25, [SP, #96]	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 X28, X27, [SP, #80]	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, #64]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP D15, D14, [SP], #160	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (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

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run armclang_1

Source file and lines	advec_cell.cpp:65-110
Module	exec

The function is defined in:

/home/eoseret/qaas/qaas_runs/178-219-7589/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_cell.cpp:65-110
/home/eoseret/qaas/qaas_runs/178-219-7589/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h:69

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Code clean check

Workaround

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

Vectorization

Details

16% of VPU instructions are used in vector version (process two or more data elements in vector registers):

18% of VPU loads are used in vector version.
38% of VPU stores are used in vector version.
0% of VPU addition or subtraction instructions are used in vector version.
0% of VPU fused multiply-add instructions are used in vector version.
4% of VPU instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

Since your execution units are vector units, only a fully 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/178-219-7589/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_cell.cpp:65
- /home/eoseret/qaas/qaas_runs/178-219-7589/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_cell.cpp:65

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 18 bytes. The binary function is storing 99 bytes.

General properties

nb instructions	118
nb uops	118
loop length	472
used w registers	12
used x registers	32
used b registers	0
used h registers	0
used s registers	1
used d registers	8
used q registers	0
used v registers	0
used z registers	18
nb stack references	26

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	14.75 cycles
front end	14.75 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	2.50	2.50	15.00	15.00	19.00	15.00	3.00	3.00	1.50	1.50	17.00	17.00	17.00	5.50	5.50
cycles	2.50	2.50	15.00	15.00	19.00	15.00	3.00	3.00	1.50	1.50	17.00	17.00	17.00	5.50	5.50

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

Cycles summary

Front-end	14.75
Dispatch	19.00
Overall L1	19.00

Vectorization ratios

all	16%
load	18%
store	38%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	0%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	4%

Vector efficiency ratios

all	33%
load	31%
store	52%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	18%
fma	25%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	28%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 19.00 cycles. At this rate:

0% of peak load performance is reached (0.95 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
10% of peak store performance is reached (5.21 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 427f30

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
STP D15, D14, [SP, #864]!	1	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	vect (50.0%)
STP D13, D12, [SP, #16]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	vect (50.0%)
STP D11, D10, [SP, #32]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	vect (50.0%)
STP D9, D8, [SP, #48]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	vect (50.0%)
STP X29, X30, [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	scal (50.0%)
STP X28, X27, [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	scal (50.0%)
STP X26, X25, [SP, #96]	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%)
STP X24, X23, [SP, #112]	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%)
STP X22, X21, [SP, #128]	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%)
STP X20, X19, [SP, #144]	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%)
ADD X29, SP, #64	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 SP, SP, #48	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
ADDVL SP, SP, #61	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	N/A
LDR W8, [X2]	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%)
LDR W23, [X4]	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%)
LDR W9, [X5]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD W24, 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	scal (12.5%)
LDR W8, [X3]	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 W25, W23, #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%)
ADD W9, W9, #4	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 W8, W8, #2	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%)
SUBS W8, W8, W24	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%)
CCMP W9, W25, #4, #12	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.LE 4282cc <_Z17advec_cell_kerneliiiiiiRN6clover8Buffer1DIdEES2_RNS_8Buffer2DIdEES5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_.omp_outlined.3+0x39c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
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	scal (12.5%)
SUB W26, W9, W25	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%)
MOVN X9, #0	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
ORR X20, XZR, X7	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
ORR X21, XZR, X6	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
ORR X22, XZR, X5	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
ADRP X0, <47cfa8>	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 X0, X0, #3192	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
UMADDL X27, W26, W8, X9	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
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	scal (12.5%)
ADD X3, SP, #20	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, SP, #40	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
STR WZR, [SP, #20]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (12.5%)
STR X8, [SP, #24]	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%)
ADD X5, 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	N/A
ADD X6, SP, #24	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%)
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	N/A
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	scal (12.5%)
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	scal (25.0%)
STP X27, XZR, [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 (50.0%)
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	scal (12.5%)
BL 410680 <@plt_start@+0x660>	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
LDP X8, X12, [SP, #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%)
CMP X8, X27	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%)
CSEL X10, X8, X27, #11	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 X12, X10	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%)
STR X10, [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%)
B.GT 4282bc <_Z17advec_cell_kerneliiiiiiRN6clover8Buffer1DIdEES2_RNS_8Buffer2DIdEES5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_S5_.omp_outlined.3+0x38c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PTRUE P1.D, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (100.0%)
LDR X16, [X21, #16]	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 X13, X11, [X29, #128]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDR X14, [X29, #120]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
CNTD X8, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
LDP X9, X15, [X29, #96]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
SUB X10, X10, X12	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%)
LDR W3, [X22]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X18, 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	scal (25.0%)
INDEX Z30.D, X12, #1	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	5	2	N/A
LDR X17, [X20, #16]	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%)
LD1RD {Z0.D}, P1/Z, [X21]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
SUB X21, X29, #64	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
SUBS X10, X18, X8	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%)
CSEL X10, XZR, X10, #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%)
LDR X0, [X14, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
WHILELO P0.D, XZR, X18	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	2	N/A
LDR X18, [X15, #16]	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%)
LDR X1, [X13, #16]	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%)
LDR X2, [X11, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD W3, W3, #2	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
LD1RD {Z3.D}, P1/Z, [X14]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
LD1RD {Z5.D}, P1/Z, [X13]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
LD1RD {Z6.D}, P1/Z, [X11]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
ORR X13, XZR, #0x55	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 X9, [X9, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
SUB X11, XZR, X8	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
MOVK X13, #16325	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
STR Z0, [X21, #511, MUL VL]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	0	2	0.50	vect (100.0%)
LD1RD {Z0.D}, P1/Z, [X20]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
SUB X20, X29, #64	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
DUP Z21.D, X3	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z4.D, X26	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z7.D, X24	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
SXTW Z21.D, P1/M, Z21.D	1	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	2	1	vect (100.0%)
STR Z0, [X20, #510, MUL VL]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	0	2	0.50	vect (100.0%)
LD1RD {Z0.D}, P1/Z, [X15]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	6	0.50	scal (25.0%)
DUP Z16.D, X25	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z17.D, X23	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z18.D, X8	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z20.D, X16	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z23.D, X17	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z25.D, X18	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
SUB X15, X29, #64	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
STR Z0, [X15, #509, MUL VL]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	0	2	0.50	vect (100.0%)
DUP Z26.D, X13	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z27.D, X0	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z28.D, X1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
DUP Z29.D, X2	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (25.0%)
ADRP X0, <47c2bc>	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 X0, X0, #3216	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
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	scal (12.5%)
BL 4105b0 <@plt_start@+0x590>	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
ADDVL SP, SP, #3	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	N/A
ADD SP, SP, #48	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
LDP D9, D8, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (50.0%)
LDP D11, D10, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (50.0%)
LDP D13, D12, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (50.0%)
LDP X20, X19, [SP, #144]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X22, X21, [SP, #128]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X24, X23, [SP, #112]	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 X26, X25, [SP, #96]	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 X28, X27, [SP, #80]	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, #64]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP D15, D14, [SP], #160	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33	vect (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

Name	Coverage (%)	Time (s)
▼advec_cell_kernel(int, int, int, int, int, int, clover::Buffer1D&, clover::Buffer1D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D–	2.92	4.06
○Loop 150 - advec_cell.cpp:66-110 - exec	2.92	3.97

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