OV - exec - Function miniqmcreference::DiracDeterminantRef<qmcplusplus::DelayedUpdate<double, double> >::ratioGrad(qmcplusplus::ParticleSet&, int, qmcplusplus::TinyVector<double, 3u>&)

	Function: miniqmcreference::DiracDeterminantRef<qmcplusplus::DelayedUpdate<double, double> >::ratioG ...	Module: exec	Source: DiracDeterminantRef.cpp:97-118 [...]	Coverage: 0.2%

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/DiracDeterminantRef.cpp: 97 - 118

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

97: {

98:   SPOVGLTimer->start();

99:   Phi->evaluate(P, iat, psiV, dpsiV, d2psiV);

100:   SPOVGLTimer->stop();

101:   RatioTimer->start();

102:   const int WorkingIndex = iat - FirstIndex;

103:   UpdateMode             = ORB_PBYP_PARTIAL;

[...]

109:   if (invRow_id != WorkingIndex)

110:   {

111:     invRow_id = WorkingIndex;

112:     updateEng.getInvRow(psiM, WorkingIndex, invRow);

113:   }

114:   curRatio = simd::dot(invRow.data(), psiV.data(), invRow.size());

115:   grad_iat += ((RealType)1.0 / curRatio) * simd::dot(invRow.data(), dpsiV.data(), invRow.size());

116:   RatioTimer->stop();

117:   return curRatio;

118: }

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h: 63 - 94

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

63:     return (a * b);

[...]

94:     (const_cast<T1&>(a) += b);

/usr/lib/gcc/aarch64-linux-gnu/11/../../../../include/c++/11/bits/unique_ptr.h: 173 - 173

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

173:       pointer    _M_ptr() const { return std::get<0>(_M_t); }

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h: 178 - 248

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

178:   inline size_t size() const { return nLocal; }

[...]

248:   inline pointer data() { return X; }

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Platforms/CPU/SIMD/inner_product.hpp: 81 - 154

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

81:   for (int i = 0; i < n; i++)

82:     res += a[i] * b[i];

[...]

154:   for (int i = 0; i < n; i++)

0x439bc0 STP	X29, X30, [SP, #976]!

0x439bc4 STP	X22, X21, [SP, #16]

0x439bc8 STP	X20, X19, [SP, #32]

0x439bcc ADD	X29, SP, #0

0x439bd0 ORR	X19, XZR, X0

0x439bd4 LDR	X0, [X0, #1128]

0x439bd8 ORR	X20, XZR, X3

0x439bdc ORR	W21, WZR, W2

0x439be0 ORR	X22, XZR, X1

0x439be4 BL	44ed40

0x439be8 LDR	X0, [X19, #1136]

0x439bec ORR	X1, XZR, X22

0x439bf0 ORR	W2, WZR, W21

0x439bf4 ADD	X3, X19, #400

0x439bf8 ADD	X4, X19, #440

0x439bfc ADD	X5, X19, #480

0x439c00 LDR	X8, [X0]

0x439c04 LDR	X8, [X8, #24]

0x439c08 BLR	X8

0x439c0c LDR	X0, [X19, #1128]

0x439c10 BL	44ee60

0x439c14 LDR	X0, [X19, #1096]

0x439c18 BL	44ed40

0x439c1c LDR	W8, [X19, #1144]

0x439c20 SUB	W2, W21, W8

0x439c24 MOVZ	W8, #2

0x439c28 STR	W8, [X19, #12]

0x439c2c LDR	W8, [X19, #1072]

0x439c30 CMP	W8, W2

0x439c34 B.EQ	439c4c

0x439c38 ADD	X0, X19, #520

0x439c3c ADD	X1, X19, #208

0x439c40 ADD	X3, X19, #1032

0x439c44 STR	W2, [X19, #1072]

0x439c48 BL	437d30

0x439c4c LDR	X9, [X19, #1040]

0x439c50 CMP	W9, #0

0x439c54 B.LE	439c88

0x439c58 CNTW	X10, ALL

0x439c5c AND	X9, X9, #6080

0x439c60 LDR	X8, [X19, #1056]

0x439c64 LDR	X13, [X19, #424]

0x439c68 PTRUE	P0.D, ALL

0x439c6c CMP	X9, X10

0x439c70 UDIV	X12, X9, X10

0x439c74 ADDVL	X11, X8, #1

0x439c78 B.CS	439ca0

0x439c7c ORR	X14, XZR, XZR

0x439c80 MOVI	D0, #0

0x439c84 B	439cf0

0x439c88 MOVZ	X8, #32752

0x439c8c STR	XZR, [X19, #1080]

0x439c90 MOVI	V1.2D, #0

0x439c94 MOVI	D2, #0

0x439c98 FMOV	D0, X8

0x439c9c B	439e1c

0x439ca0 DUP	Z0.D, #0

0x439ca4 ORR	X16, XZR, XZR

0x439ca8 UMSUBL	X15, W12, W10, X9

0x439cac ORR	Z1.D, Z0.D, Z0.D

0x439cb0 UMADDL	X14, W12, W10, XZR

0x439cb4 ADDVL	X17, X13, #1

0x439cb8 HINT	#0

0x439cbc HINT	#0

(1052) 0x439cc0 LD1D	{Z2.D}, P0/Z, [X8, X16,LSL #3]

(1052) 0x439cc4 LD1D	{Z3.D}, P0/Z, [X11, X16,LSL #3]

(1052) 0x439cc8 LD1D	{Z4.D}, P0/Z, [X13, X16,LSL #3]

(1052) 0x439ccc FMLA	Z0.D, P0/M, Z4.D, Z2.D

(1052) 0x439cd0 LD1D	{Z5.D}, P0/Z, [X17, X16,LSL #3]

(1052) 0x439cd4 FMLA	Z1.D, P0/M, Z5.D, Z3.D

(1052) 0x439cd8 ADD	X16, X16, X10

(1052) 0x439cdc CMP	X14, X16

(1052) 0x439ce0 B.NE	439cc0

0x439ce4 FADD	Z0.D, Z1.D, Z0.D

0x439ce8 FADDV	D0, P0, Z0.D

0x439cec CBZ	X15, 439d14

0x439cf0 SUB	X15, X9, X14

0x439cf4 UBFM	X14, X14, #61, #60

0x439cf8 ADD	X13, X13, X14

0x439cfc ADD	X14, X8, X14

(1055) 0x439d00 LDR	D1, [X14], #8

(1055) 0x439d04 LDR	D2, [X13], #8

(1055) 0x439d08 FMADD	D0, D2, D1, D0

(1055) 0x439d0c SUBS	X15, X15, #1

(1055) 0x439d10 B.NE	439d00

0x439d14 FMOV	D1, #1.0000000

0x439d18 STR	D0, [X19, #1080]

0x439d1c LDR	X13, [X19, #464]

0x439d20 CMP	X9, X10

0x439d24 FDIV	D0, D1, D0

0x439d28 B.CS	439d3c

0x439d2c ORR	X12, XZR, XZR

0x439d30 MOVI	V1.2D, #0

0x439d34 MOVI	D2, #0

0x439d38 B	439ddc

0x439d3c RDVL	X14, #1

0x439d40 DUP	Z2.D, #0

0x439d44 ORR	X15, XZR, XZR

0x439d48 UBFM	X16, X14, #4, #63

0x439d4c UMSUBL	X14, W12, W10, X9

0x439d50 UMADDL	X12, W12, W10, XZR

0x439d54 ORR	X18, XZR, X13

0x439d58 ORR	Z3.D, Z2.D, Z2.D

0x439d5c ORR	Z4.D, Z2.D, Z2.D

0x439d60 ORR	Z5.D, Z2.D, Z2.D

0x439d64 ORR	Z1.D, Z2.D, Z2.D

0x439d68 ORR	Z6.D, Z2.D, Z2.D

0x439d6c ADD	X17, X16, X16,LSL #1

0x439d70 UBFM	X16, X17, #59, #58

0x439d74 UBFM	X17, X17, #60, #59

0x439d78 HINT	#0

0x439d7c HINT	#0

(1053) 0x439d80 ADD	X0, X18, X17

(1053) 0x439d84 LD1D	{Z7.D}, P0/Z, [X8, X15,LSL #3]

(1053) 0x439d88 LD1D	{Z16.D}, P0/Z, [X11, X15,LSL #3]

(1053) 0x439d8c LD3D	{Z17.D, Z18.D, Z19.D}, P0/Z, [X18, MUL VL]

(1053) 0x439d90 ADD	X15, X15, X10

(1053) 0x439d94 ADD	X18, X18, X16

(1053) 0x439d98 LD3D	{Z20.D, Z21.D, Z22.D}, P0/Z, [X0, MUL VL]

(1053) 0x439d9c CMP	X12, X15

(1053) 0x439da0 FMLA	Z1.D, P0/M, Z17.D, Z7.D

(1053) 0x439da4 FMLA	Z4.D, P0/M, Z18.D, Z7.D

(1053) 0x439da8 FMLA	Z2.D, P0/M, Z19.D, Z7.D

(1053) 0x439dac FMLA	Z6.D, P0/M, Z20.D, Z16.D

(1053) 0x439db0 FMLA	Z5.D, P0/M, Z21.D, Z16.D

(1053) 0x439db4 FMLA	Z3.D, P0/M, Z22.D, Z16.D

(1053) 0x439db8 B.NE	439d80

0x439dbc FADD	Z1.D, Z6.D, Z1.D

0x439dc0 FADD	Z4.D, Z5.D, Z4.D

0x439dc4 FADD	Z2.D, Z3.D, Z2.D

0x439dc8 FADDV	D1, P0, Z1.D

0x439dcc FADDV	D4, P0, Z4.D

0x439dd0 FADDV	D2, P0, Z2.D

0x439dd4 MOV	V1.D[1], V4.D[0]

0x439dd8 CBZ	X14, 439e1c

0x439ddc MOVZ	W10, #24

0x439de0 SUB	X9, X9, X12

0x439de4 ADD	X8, X8, X12,LSL #3

0x439de8 MADD	X10, X12, X10, X13

0x439dec ADD	X10, X10, #16

0x439df0 HINT	#0

0x439df4 HINT	#0

0x439df8 HINT	#0

0x439dfc HINT	#0

(1054) 0x439e00 LDUR	Q4, [X10, #496]

(1054) 0x439e04 LDR	D3, [X8], #8

(1054) 0x439e08 LDR	D5, [X10], #24

(1054) 0x439e0c FMADD	D2, D5, D3, D2

(1054) 0x439e10 SUBS	X9, X9, #1

(1054) 0x439e14 FMLA	V1.2D, V4.2D, V3.D[0]

(1054) 0x439e18 B.NE	439e00

0x439e1c LDR	Q3, [X20]

0x439e20 LDR	X0, [X19, #1096]

0x439e24 FMLA	V3.2D, V1.2D, V0.D[0]

0x439e28 LDR	D1, [X20, #16]

0x439e2c STR	Q3, [X20]

0x439e30 FMADD	D0, D2, D0, D1

0x439e34 STR	D0, [X20, #16]

0x439e38 BL	44ee60

0x439e3c LDR	D0, [X19, #1080]

0x439e40 LDP	X20, X19, [SP, #32]

0x439e44 LDP	X22, X21, [SP, #16]

0x439e48 LDP	X29, X30, [SP], #48

0x439e4c RET

0x439e50 HINT	#0

0x439e54 HINT	#0

0x439e58 HINT	#0

0x439e5c HINT	#0

Coverage (%)	Name	Source Location	Module
►100.00+	qmcplusplus::WaveFunction::rat[...]	stl_vector.h:919	exec
○	.omp_outlined..62	miniqmc.cpp:437	exec
○	__kmp_invoke_microtask		libomp.so

Path /

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

Source file and lines	DiracDeterminantRef.cpp:97-118
Module	exec

The function is defined in:

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/DiracDeterminantRef.cpp:97-103,109-118
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:63,94
/usr/lib/gcc/aarch64-linux-gnu/11/../../../../include/c++/11/bits/unique_ptr.h:173
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:178,248
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Platforms/CPU/SIMD/inner_product.hpp:81-82,154

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 15.00 to 9.50 cycles (1.58x speedup).

Workaround

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

Vectorization

Your function is poorly vectorized. Only 42% 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 15.00 to 11.38 cycles (1.32x speedup).

Details

26% of VPU instructions are used in vector version (process two or more data elements in vector registers):

33% of VPU loads are used in vector version.
33% of VPU stores are used in vector version.
36% of VPU addition or subtraction instructions are used in vector version.
20% of VPU fused multiply-add instructions are used in vector version.
0% of VPU divide and square root instructions are used in vector version.
20% 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)

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 15.00 to 3.00 cycles (5.00x speedup).

FMA

Detected 3 FMA (fused multiply-add) operations.

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

The binary function is composed of 39 FP arithmetical operations:

35: addition or subtraction (3 inside FMA instructions)
3: multiply (all inside FMA instructions)
1: divide

The binary function is loading 176 bytes. The binary function is storing 96 bytes.

Arithmetic intensity

Arithmetic intensity is 0.14 FP operations per loaded or stored byte.

General properties

nb instructions	132
loop length	528
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

15.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	8.00	8.00	14.00	14.00	14.00	14.00	9.50	9.50	7.50	7.50	9.50	9.17	9.33	3.00	3.00
cycles	8.00	8.00	14.00	14.00	14.00	14.00	9.50	9.50	7.50	7.50	9.50	9.17	9.33	3.00	3.00

Cycles executing div or sqrt instructions

2.00-1.00

Cycles summary

Front-end	15.00
Overall L1	15.00

Vectorization ratios

INT

all	14%
load	33%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	0%
other	20%

all	100%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	100%
fma	100%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	NA (no other vectorizable/vectorized instructions)

INT+FP

all	26%
load	33%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	36%
fma	20%
div/sqrt	0%
other	20%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 15.00 cycles. At this rate:

12% of peak load performance is reached (11.73 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
13% of peak store performance is reached (6.40 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the function is: 439bc0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput
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
STP X22, X21, [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
STP X20, X19, [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
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
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
LDR X0, [X0, #1128]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
ORR X20, XZR, X3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR W21, WZR, W2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR X22, XZR, X1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
BL 44ed40 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE5startEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X0, [X19, #1136]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
ORR X1, XZR, X22	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR W2, WZR, W21	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, X19, #400	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, X19, #440	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X5, X19, #480	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, [X0]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X8, [X8, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
BLR X8	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X0, [X19, #1128]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
BL 44ee60 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X0, [X19, #1096]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
BL 44ed40 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE5startEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR W8, [X19, #1144]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
SUB W2, W21, W8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ 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
STR W8, [X19, #12]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
LDR W8, [X19, #1072]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
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
B.EQ 439c4c <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x8c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ADD X0, X19, #520	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X1, X19, #208	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, X19, #1032	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STR W2, [X19, #1072]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
BL 437d30 <_ZN11qmcplusplus13DelayedUpdateIddE9getInvRowINS_6VectorIdSaIdEEEEEvRKNS_6MatrixIdS4_EEiRT_>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X9, [X19, #1040]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP W9, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.LE 439c88 <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0xc8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
CNTW X10, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
AND X9, X9, #6080	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, [X19, #1056]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X13, [X19, #424]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
PTRUE P0.D, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
CMP X9, X10	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
UDIV X12, X9, X10	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	5-20	1-0.50
ADDVL X11, X8, #1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
B.CS 439ca0 <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0xe0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X14, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVI D0, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
B 439cf0 <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x130>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
MOVZ X8, #32752	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, [X19, #1080]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
MOVI V1.2D, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
MOVI D2, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
FMOV D0, X8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
B 439e1c <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x25c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
DUP Z0.D, #0	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR X16, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
UMSUBL X15, W12, W10, X9	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
ORR Z1.D, Z0.D, Z0.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
UMADDL X14, W12, W10, XZR	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
ADDVL X17, X13, #1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
HINT #0
HINT #0
FADD Z0.D, Z1.D, Z0.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
FADDV D0, P0, Z0.D	1	0	0	0	0	0	0	0.75	0.75	0.25	0.25	0	0	0	0	0	9	2
CBZ X15, 439d14 <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x154>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
SUB X15, X9, X14	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
UBFM X14, X14, #61, #60	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X13, X13, X14	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X14, X8, X14	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
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
STR D0, [X19, #1080]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
LDR X13, [X19, #464]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X9, X10	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
FDIV D0, D1, D0	1	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	1-0.50
B.CS 439d3c <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x17c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X12, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVI V1.2D, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
MOVI D2, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
B 439ddc <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x21c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
RDVL X14, #1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
DUP Z2.D, #0	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR X15, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
UBFM X16, X14, #4, #63	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
UMSUBL X14, W12, W10, X9	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
UMADDL X12, W12, W10, XZR	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
ORR X18, XZR, X13	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR Z3.D, Z2.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR Z4.D, Z2.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR Z5.D, Z2.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR Z1.D, Z2.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR Z6.D, Z2.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ADD X17, X16, X16,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
UBFM X16, X17, #59, #58	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
UBFM X17, X17, #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
HINT #0
HINT #0
FADD Z1.D, Z6.D, Z1.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
FADD Z4.D, Z5.D, Z4.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
FADD Z2.D, Z3.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
FADDV D1, P0, Z1.D	1	0	0	0	0	0	0	0.75	0.75	0.25	0.25	0	0	0	0	0	9	2
FADDV D4, P0, Z4.D	1	0	0	0	0	0	0	0.75	0.75	0.25	0.25	0	0	0	0	0	9	2
FADDV D2, P0, Z2.D	1	0	0	0	0	0	0	0.75	0.75	0.25	0.25	0	0	0	0	0	9	2
MOV V1.D[1], V4.D[0]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
CBZ X14, 439e1c <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x25c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
MOVZ W10, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
SUB X9, X9, X12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X8, X8, X12,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
MADD X10, X12, X10, X13	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
ADD X10, X10, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
HINT #0
HINT #0
HINT #0
HINT #0
LDR Q3, [X20]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
LDR X0, [X19, #1096]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
FMLA V3.2D, V1.2D, V0.D[0]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	4	0.25
LDR D1, [X20, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
STR Q3, [X20]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
FMADD D0, D2, D0, D1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	4	0.25
STR D0, [X20, #16]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
BL 44ee60 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR D0, [X19, #1080]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
LDP X20, X19, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
LDP X22, X21, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
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
RET	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

Source file and lines	DiracDeterminantRef.cpp:97-118
Module	exec

The function is defined in:

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/DiracDeterminantRef.cpp:97-103,109-118
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:63,94
/usr/lib/gcc/aarch64-linux-gnu/11/../../../../include/c++/11/bits/unique_ptr.h:173
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:178,248
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Platforms/CPU/SIMD/inner_product.hpp:81-82,154

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 15.00 to 9.50 cycles (1.58x speedup).

Workaround

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

Vectorization

Details

26% of VPU instructions are used in vector version (process two or more data elements in vector registers):

33% of VPU loads are used in vector version.
33% of VPU stores are used in vector version.
36% of VPU addition or subtraction instructions are used in vector version.
20% of VPU fused multiply-add instructions are used in vector version.
0% of VPU divide and square root instructions are used in vector version.
20% 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)

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 15.00 to 3.00 cycles (5.00x speedup).

FMA

Detected 3 FMA (fused multiply-add) operations.

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

The binary function is composed of 39 FP arithmetical operations:

35: addition or subtraction (3 inside FMA instructions)
3: multiply (all inside FMA instructions)
1: divide

The binary function is loading 176 bytes. The binary function is storing 96 bytes.

Arithmetic intensity

Arithmetic intensity is 0.14 FP operations per loaded or stored byte.

General properties

nb instructions	132
loop length	528
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

15.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	8.00	8.00	14.00	14.00	14.00	14.00	9.50	9.50	7.50	7.50	9.50	9.17	9.33	3.00	3.00
cycles	8.00	8.00	14.00	14.00	14.00	14.00	9.50	9.50	7.50	7.50	9.50	9.17	9.33	3.00	3.00

Cycles executing div or sqrt instructions

2.00-1.00

Cycles summary

Front-end	15.00
Overall L1	15.00

Vectorization ratios

INT

all	14%
load	33%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	0%
other	20%

all	100%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	100%
fma	100%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	NA (no other vectorizable/vectorized instructions)

INT+FP

all	26%
load	33%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	36%
fma	20%
div/sqrt	0%
other	20%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 15.00 cycles. At this rate:

12% of peak load performance is reached (11.73 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
13% of peak store performance is reached (6.40 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the function is: 439bc0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput
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
STP X22, X21, [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
STP X20, X19, [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
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
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
LDR X0, [X0, #1128]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
ORR X20, XZR, X3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR W21, WZR, W2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR X22, XZR, X1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
BL 44ed40 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE5startEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X0, [X19, #1136]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
ORR X1, XZR, X22	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR W2, WZR, W21	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, X19, #400	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, X19, #440	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X5, X19, #480	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, [X0]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X8, [X8, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
BLR X8	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X0, [X19, #1128]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
BL 44ee60 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X0, [X19, #1096]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
BL 44ed40 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE5startEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR W8, [X19, #1144]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
SUB W2, W21, W8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ 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
STR W8, [X19, #12]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
LDR W8, [X19, #1072]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
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
B.EQ 439c4c <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x8c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ADD X0, X19, #520	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X1, X19, #208	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, X19, #1032	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STR W2, [X19, #1072]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
BL 437d30 <_ZN11qmcplusplus13DelayedUpdateIddE9getInvRowINS_6VectorIdSaIdEEEEEvRKNS_6MatrixIdS4_EEiRT_>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR X9, [X19, #1040]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP W9, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.LE 439c88 <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0xc8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
CNTW X10, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
AND X9, X9, #6080	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, [X19, #1056]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X13, [X19, #424]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
PTRUE P0.D, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
CMP X9, X10	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
UDIV X12, X9, X10	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	5-20	1-0.50
ADDVL X11, X8, #1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
B.CS 439ca0 <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0xe0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X14, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVI D0, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
B 439cf0 <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x130>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
MOVZ X8, #32752	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, [X19, #1080]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
MOVI V1.2D, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
MOVI D2, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
FMOV D0, X8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
B 439e1c <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x25c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
DUP Z0.D, #0	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR X16, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
UMSUBL X15, W12, W10, X9	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
ORR Z1.D, Z0.D, Z0.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
UMADDL X14, W12, W10, XZR	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
ADDVL X17, X13, #1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
HINT #0
HINT #0
FADD Z0.D, Z1.D, Z0.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
FADDV D0, P0, Z0.D	1	0	0	0	0	0	0	0.75	0.75	0.25	0.25	0	0	0	0	0	9	2
CBZ X15, 439d14 <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x154>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
SUB X15, X9, X14	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
UBFM X14, X14, #61, #60	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X13, X13, X14	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X14, X8, X14	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
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
STR D0, [X19, #1080]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
LDR X13, [X19, #464]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X9, X10	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
FDIV D0, D1, D0	1	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	1-0.50
B.CS 439d3c <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x17c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X12, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVI V1.2D, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
MOVI D2, #0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
B 439ddc <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x21c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
RDVL X14, #1	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
DUP Z2.D, #0	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR X15, XZR, XZR	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
UBFM X16, X14, #4, #63	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
UMSUBL X14, W12, W10, X9	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
UMADDL X12, W12, W10, XZR	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
ORR X18, XZR, X13	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR Z3.D, Z2.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR Z4.D, Z2.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR Z5.D, Z2.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR Z1.D, Z2.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ORR Z6.D, Z2.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
ADD X17, X16, X16,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
UBFM X16, X17, #59, #58	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
UBFM X17, X17, #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
HINT #0
HINT #0
FADD Z1.D, Z6.D, Z1.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
FADD Z4.D, Z5.D, Z4.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
FADD Z2.D, Z3.D, Z2.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.50
FADDV D1, P0, Z1.D	1	0	0	0	0	0	0	0.75	0.75	0.25	0.25	0	0	0	0	0	9	2
FADDV D4, P0, Z4.D	1	0	0	0	0	0	0	0.75	0.75	0.25	0.25	0	0	0	0	0	9	2
FADDV D2, P0, Z2.D	1	0	0	0	0	0	0	0.75	0.75	0.25	0.25	0	0	0	0	0	9	2
MOV V1.D[1], V4.D[0]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25
CBZ X14, 439e1c <_ZN16miniqmcreference19DiracDeterminantRefIN11qmcplusplus13DelayedUpdateIddEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x25c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
MOVZ W10, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
SUB X9, X9, X12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X8, X8, X12,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
MADD X10, X12, X10, X13	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1
ADD X10, X10, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
HINT #0
HINT #0
HINT #0
HINT #0
LDR Q3, [X20]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
LDR X0, [X19, #1096]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
FMLA V3.2D, V1.2D, V0.D[0]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	4	0.25
LDR D1, [X20, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
STR Q3, [X20]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
FMADD D0, D2, D0, D1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	4	0.25
STR D0, [X20, #16]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
BL 44ee60 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR D0, [X19, #1080]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
LDP X20, X19, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
LDP X22, X21, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
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
RET	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

Name	Coverage (%)	Time (s)
▼miniqmcreference::DiracDeterminantRef >::ratioGrad(qmcplusplus::ParticleSet&, int, qmcplusplus::TinyVector&)–	0.2	0.1
○Loop 1053 - inner_product.hpp:154-154 - exec	0.14	0.07
○Loop 1052 - inner_product.hpp:81-82 - exec	0.06	0.03
○Loop 1054 - inner_product.hpp:154-154 - exec	0	0
○Loop 1055 - inner_product.hpp:81-82 - exec	0	0

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Expensive FP math instructions/calls

FMA

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

Arithmetic intensity

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

FMA

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

Arithmetic intensity

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code