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: 0.11%

/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++)

0x439ae0 STP	X29, X30, [SP, #976]!

0x439ae4 STP	X22, X21, [SP, #16]

0x439ae8 STP	X20, X19, [SP, #32]

0x439aec ADD	X29, SP, #0

0x439af0 ORR	X19, XZR, X0

0x439af4 LDR	X0, [X0, #1128]

0x439af8 ORR	X20, XZR, X3

0x439afc ORR	W21, WZR, W2

0x439b00 ORR	X22, XZR, X1

0x439b04 BL	44ec60

0x439b08 LDR	X0, [X19, #1136]

0x439b0c ORR	X1, XZR, X22

0x439b10 ORR	W2, WZR, W21

0x439b14 ADD	X3, X19, #400

0x439b18 ADD	X4, X19, #440

0x439b1c ADD	X5, X19, #480

0x439b20 LDR	X8, [X0]

0x439b24 LDR	X8, [X8, #24]

0x439b28 BLR	X8

0x439b2c LDR	X0, [X19, #1128]

0x439b30 BL	44ed80

0x439b34 LDR	X0, [X19, #1096]

0x439b38 BL	44ec60

0x439b3c LDR	W8, [X19, #1144]

0x439b40 SUB	W2, W21, W8

0x439b44 MOVZ	W8, #2

0x439b48 STR	W8, [X19, #12]

0x439b4c LDR	W8, [X19, #1072]

0x439b50 CMP	W8, W2

0x439b54 B.EQ	439b6c

0x439b58 ADD	X0, X19, #520

0x439b5c ADD	X1, X19, #208

0x439b60 ADD	X3, X19, #1032

0x439b64 STR	W2, [X19, #1072]

0x439b68 BL	437c50

0x439b6c LDR	X9, [X19, #1040]

0x439b70 CMP	W9, #0

0x439b74 B.LE	439ba8

0x439b78 CNTW	X10, ALL

0x439b7c AND	X9, X9, #6080

0x439b80 LDR	X8, [X19, #1056]

0x439b84 LDR	X13, [X19, #424]

0x439b88 PTRUE	P0.D, ALL

0x439b8c CMP	X9, X10

0x439b90 UDIV	X12, X9, X10

0x439b94 ADDVL	X11, X8, #1

0x439b98 B.CS	439bc0

0x439b9c ORR	X14, XZR, XZR

0x439ba0 MOVI	D0, #0

0x439ba4 B	439c10

0x439ba8 MOVZ	X8, #32752

0x439bac STR	XZR, [X19, #1080]

0x439bb0 MOVI	V1.2D, #0

0x439bb4 MOVI	D2, #0

0x439bb8 FMOV	D0, X8

0x439bbc B	439d3c

0x439bc0 DUP	Z0.D, #0

0x439bc4 ORR	X16, XZR, XZR

0x439bc8 UMSUBL	X15, W12, W10, X9

0x439bcc ORR	Z1.D, Z0.D, Z0.D

0x439bd0 UMADDL	X14, W12, W10, XZR

0x439bd4 ADDVL	X17, X13, #1

0x439bd8 HINT	#0

0x439bdc HINT	#0

(1052) 0x439be0 LD1D	{Z2.D}, P0/Z, [X8, X16,LSL #3]

(1052) 0x439be4 LD1D	{Z3.D}, P0/Z, [X11, X16,LSL #3]

(1052) 0x439be8 LD1D	{Z4.D}, P0/Z, [X13, X16,LSL #3]

(1052) 0x439bec FMLA	Z0.D, P0/M, Z4.D, Z2.D

(1052) 0x439bf0 LD1D	{Z5.D}, P0/Z, [X17, X16,LSL #3]

(1052) 0x439bf4 FMLA	Z1.D, P0/M, Z5.D, Z3.D

(1052) 0x439bf8 ADD	X16, X16, X10

(1052) 0x439bfc CMP	X14, X16

(1052) 0x439c00 B.NE	439be0

0x439c04 FADD	Z0.D, Z1.D, Z0.D

0x439c08 FADDV	D0, P0, Z0.D

0x439c0c CBZ	X15, 439c34

0x439c10 SUB	X15, X9, X14

0x439c14 UBFM	X14, X14, #61, #60

0x439c18 ADD	X13, X13, X14

0x439c1c ADD	X14, X8, X14

(1055) 0x439c20 LDR	D1, [X14], #8

(1055) 0x439c24 LDR	D2, [X13], #8

(1055) 0x439c28 FMADD	D0, D2, D1, D0

(1055) 0x439c2c SUBS	X15, X15, #1

(1055) 0x439c30 B.NE	439c20

0x439c34 FMOV	D1, #1.0000000

0x439c38 STR	D0, [X19, #1080]

0x439c3c LDR	X13, [X19, #464]

0x439c40 CMP	X9, X10

0x439c44 FDIV	D0, D1, D0

0x439c48 B.CS	439c5c

0x439c4c ORR	X12, XZR, XZR

0x439c50 MOVI	V1.2D, #0

0x439c54 MOVI	D2, #0

0x439c58 B	439cfc

0x439c5c RDVL	X14, #1

0x439c60 DUP	Z2.D, #0

0x439c64 ORR	X15, XZR, XZR

0x439c68 UBFM	X16, X14, #4, #63

0x439c6c UMSUBL	X14, W12, W10, X9

0x439c70 UMADDL	X12, W12, W10, XZR

0x439c74 ORR	X18, XZR, X13

0x439c78 ORR	Z3.D, Z2.D, Z2.D

0x439c7c ORR	Z4.D, Z2.D, Z2.D

0x439c80 ORR	Z5.D, Z2.D, Z2.D

0x439c84 ORR	Z1.D, Z2.D, Z2.D

0x439c88 ORR	Z6.D, Z2.D, Z2.D

0x439c8c ADD	X17, X16, X16,LSL #1

0x439c90 UBFM	X16, X17, #59, #58

0x439c94 UBFM	X17, X17, #60, #59

0x439c98 HINT	#0

0x439c9c HINT	#0

(1053) 0x439ca0 ADD	X0, X18, X17

(1053) 0x439ca4 LD1D	{Z7.D}, P0/Z, [X8, X15,LSL #3]

(1053) 0x439ca8 LD1D	{Z16.D}, P0/Z, [X11, X15,LSL #3]

(1053) 0x439cac LD3D	{Z17.D, Z18.D, Z19.D}, P0/Z, [X18, MUL VL]

(1053) 0x439cb0 ADD	X15, X15, X10

(1053) 0x439cb4 ADD	X18, X18, X16

(1053) 0x439cb8 LD3D	{Z20.D, Z21.D, Z22.D}, P0/Z, [X0, MUL VL]

(1053) 0x439cbc CMP	X12, X15

(1053) 0x439cc0 FMLA	Z1.D, P0/M, Z17.D, Z7.D

(1053) 0x439cc4 FMLA	Z4.D, P0/M, Z18.D, Z7.D

(1053) 0x439cc8 FMLA	Z2.D, P0/M, Z19.D, Z7.D

(1053) 0x439ccc FMLA	Z6.D, P0/M, Z20.D, Z16.D

(1053) 0x439cd0 FMLA	Z5.D, P0/M, Z21.D, Z16.D

(1053) 0x439cd4 FMLA	Z3.D, P0/M, Z22.D, Z16.D

(1053) 0x439cd8 B.NE	439ca0

0x439cdc FADD	Z1.D, Z6.D, Z1.D

0x439ce0 FADD	Z4.D, Z5.D, Z4.D

0x439ce4 FADD	Z2.D, Z3.D, Z2.D

0x439ce8 FADDV	D1, P0, Z1.D

0x439cec FADDV	D4, P0, Z4.D

0x439cf0 FADDV	D2, P0, Z2.D

0x439cf4 MOV	V1.D[1], V4.D[0]

0x439cf8 CBZ	X14, 439d3c

0x439cfc MOVZ	W10, #24

0x439d00 SUB	X9, X9, X12

0x439d04 ADD	X8, X8, X12,LSL #3

0x439d08 MADD	X10, X12, X10, X13

0x439d0c ADD	X10, X10, #16

0x439d10 HINT	#0

0x439d14 HINT	#0

0x439d18 HINT	#0

0x439d1c HINT	#0

(1054) 0x439d20 LDUR	Q4, [X10, #496]

(1054) 0x439d24 LDR	D3, [X8], #8

(1054) 0x439d28 LDR	D5, [X10], #24

(1054) 0x439d2c FMADD	D2, D5, D3, D2

(1054) 0x439d30 SUBS	X9, X9, #1

(1054) 0x439d34 FMLA	V1.2D, V4.2D, V3.D[0]

(1054) 0x439d38 B.NE	439d20

0x439d3c LDR	Q3, [X20]

0x439d40 LDR	X0, [X19, #1096]

0x439d44 FMLA	V3.2D, V1.2D, V0.D[0]

0x439d48 LDR	D1, [X20, #16]

0x439d4c STR	Q3, [X20]

0x439d50 FMADD	D0, D2, D0, D1

0x439d54 STR	D0, [X20, #16]

0x439d58 BL	44ed80

0x439d5c LDR	D0, [X19, #1080]

0x439d60 LDP	X20, X19, [SP, #32]

0x439d64 LDP	X22, X21, [SP, #16]

0x439d68 LDP	X29, X30, [SP], #48

0x439d6c RET

0x439d70 HINT	#0

0x439d74 HINT	#0

0x439d78 HINT	#0

0x439d7c 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: 439ae0

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 44ec60 <_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 44ed80 <_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 44ec60 <_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 439b6c <_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 437c50 <_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 439ba8 <_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 439bc0 <_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 439c10 <_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 439d3c <_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, 439c34 <_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 439c5c <_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 439cfc <_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, 439d3c <_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 44ed80 <_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: 439ae0

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 44ec60 <_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 44ed80 <_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 44ec60 <_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 439b6c <_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 437c50 <_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 439ba8 <_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 439bc0 <_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 439c10 <_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 439d3c <_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, 439c34 <_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 439c5c <_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 439cfc <_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, 439d3c <_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 44ed80 <_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&)–	0.11	0.05
○Loop 1053 - inner_product.hpp:154-154 - exec	0.08	0.04
○Loop 1052 - inner_product.hpp:81-82 - exec	0.03	0.01
○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