OV - exec - Function miniqmcreference::OneBodyJastrowRef<qmcplusplus::BsplineFunctor<double> >::computeU3(qmcplusplus::ParticleSet&, int, double const*) [clone .isra.0]

	Function: miniqmcreference::OneBodyJastrowRef<qmcplusplus::BsplineFunctor<double> >::computeU3(qmcpl ...	Module: exec	Source: OneBodyJastrowRef.h:205-241 [...]	Coverage: 0.01%

/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/OneBodyJastrowRef.h: 205 - 241

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

205:   inline void computeU3(ParticleSet& P, int iat, const valT* dist)

206:   {

207:     if (NumGroups > 0)

[...]

214:       for (int jg = 0; jg < NumGroups; ++jg)

215:       {

216:         if (F[jg] == nullptr)

217:           continue;

218:         F[jg]->evaluateVGL(-1,

[...]

231:       for (int c = 0; c < Nions; ++c)

232:       {

233:         int gid = Ions.GroupID[c];

234:         if (F[gid] != nullptr)

235:         {

236:           U[c] = F[gid]->evaluate(dist[c], dU[c], d2U[c]);

237:           dU[c] /= dist[c];

238:         }

239:       }

240:     }

241:   }

/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/BsplineFunctor.h: 186 - 191

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

186:   inline real_type evaluate(real_type r, real_type& dudr, real_type& d2udr2)

187:   {

188:     if (r >= cutoff_radius)

189:     {

190:       dudr = d2udr2 = 0.0;

191:       return 0.0;

/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h: 316 - 316

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

316:   inline int last(int igroup) const { return (*group_offsets_)[igroup + 1]; }

/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h: 223 - 229

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

223:     return X[i];

[...]

229:     return X[i];

/home/hbollore/gnu/unknow_version/include/c++/14.0.1/bits/stl_algobase.h: 952 - 1151

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

952: 	*__first = __tmp;

[...]

1146:       if (__n <= 0)

1147: 	return __first;

1148:

1149:       __glibcxx_requires_can_increment(__first, __n);

1150:

1151:       std::__fill_a(__first, __first + __n, __value);

/home/hbollore/gnu/unknow_version/include/c++/14.0.1/bits/stl_vector.h: 1128 - 1263

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

1128:       operator[](size_type __n) _GLIBCXX_NOEXCEPT

1129:       {

1130: 	__glibcxx_requires_subscript(__n);

1131: 	return *(this->_M_impl._M_start + __n);

[...]

1263:       { return _M_data_ptr(this->_M_impl._M_start); }

/home/hbollore/gnu/unknow_version/include/c++/14.0.1/bits/shared_ptr_base.h: 1667 - 1667

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

1667:       { return _M_ptr; }

0x41560c SUB	SP, SP, #96

0x415610 STP	X29, X30, [SP, #16]

0x415614 ADD	X29, SP, #16

0x415618 STP	X19, X20, [SP, #32]

0x41561c ORR	X19, XZR, X0

0x415620 STP	X21, X22, [SP, #48]

0x415624 ORR	X21, XZR, X1

0x415628 LDR	W22, [X0, #152]

0x41562c LDR	W2, [X0, #144]

0x415630 CMP	W22, #0

0x415634 B.LE	41571c

0x415638 STP	X23, X24, [SP, #64]

0x41563c STP	X25, X26, [SP, #80]

0x415640 LDR	X23, [X0, #256]

0x415644 LDR	X24, [X0, #280]

0x415648 LDR	X25, [X0, #304]

0x41564c CMP	W2, #0

0x415650 B.LE	415688

0x415654 UBFM	X26, X2, #61, #31

0x415658 MOVZ	W1, #0

0x41565c ORR	X2, XZR, X26

0x415660 ORR	X0, XZR, X23

0x415664 BL	403bd0

0x415668 ORR	X2, XZR, X26

0x41566c MOVZ	W1, #0

0x415670 ORR	X0, XZR, X24

0x415674 BL	403bd0

0x415678 ORR	X2, XZR, X26

0x41567c ORR	X0, XZR, X25

0x415680 MOVZ	W1, #0

0x415684 BL	403bd0

0x415688 MOVZ	X20, #0

0x41568c ADD	X26, X19, #456

    (265) 0x415690 LDR	X12, [X26]

    (265) 0x415694 ORR	W10, WZR, W20

    (265) 0x415698 ORR	X7, XZR, X25

    (265) 0x41569c ORR	X6, XZR, X24

    (265) 0x4156a0 ORR	X5, XZR, X23

    (265) 0x4156a4 ORR	X4, XZR, X21

    (265) 0x4156a8 MOVN	W1, #0

    (265) 0x4156ac UBFM	X11, X10, #62, #61

    (265) 0x4156b0 LDR	X0, [X12, X10,LSL #3]

    (265) 0x4156b4 CBZ	X0, 41570c

    (265) 0x4156b8 LDR	X13, [X19, #160]

    (265) 0x4156bc ADD	X20, X20, #1

    (265) 0x4156c0 LDR	X14, [X19, #328]

    (265) 0x4156c4 LDR	X15, [X19, #352]

    (265) 0x4156c8 LDR	X16, [X13, #616]

    (265) 0x4156cc LDR	X17, [X16, #24]

    (265) 0x4156d0 ADD	X18, X17, X11

    (265) 0x4156d4 LDR	W2, [X17, X11]

    (265) 0x4156d8 LDR	W3, [X18, #4]

    (265) 0x4156dc STP	X14, X15, [SP]

    (265) 0x4156e0 BL	4457e0

    (265) 0x4156e4 LDR	W22, [X19, #152]

    (265) 0x4156e8 CMP	W22, W20

    (265) 0x4156ec B.GT	415690

(266) 0x4156f0 LDP	X23, X24, [SP, #64]

(266) 0x4156f4 LDP	X25, X26, [SP, #80]

(266) 0x4156f8 LDP	X29, X30, [SP, #16]

(266) 0x4156fc LDP	X19, X20, [SP, #32]

(266) 0x415700 LDP	X21, X22, [SP, #48]

(266) 0x415704 ADD	SP, SP, #96

(266) 0x415708 RET

  (264) 0x41570c ADD	X20, X20, #1

  (264) 0x415710 CMP	W22, W20

  (264) 0x415714 B.GT	415690

(266) 0x415718 B	4156f0

0x41571c CMP	W2, #0

0x415720 B.LE	4156f8

0x415724 STP	X23, X24, [SP, #64]

0x415728 SBFM	X24, X2, #0, #31

0x41572c MOVZ	X20, #0

0x415730 STP	X25, X26, [SP, #80]

0x415734 LDR	X0, [X0, #160]

0x415738 LDR	X25, [X19, #456]

0x41573c LDR	X26, [X0, #24]

0x415740 TBZ	W24, #0, 415828

0x415744 LDRSW	X1, [X26]

0x415748 LDR	X0, [X25, X1,LSL #3]

0x41574c CBZ	X0, 415790

0x415750 LDR	D0, [X21]

0x415754 MOVZ	X22, #0

0x415758 LDR	D31, [X0, #8]

0x41575c LDR	X20, [X19, #280]

0x415760 LDR	X2, [X19, #304]

0x415764 FCMPE	D0, D31

0x415768 B.MI	415868

0x41576c MOVI	D0, #0

0x415770 STR	XZR, [X2]

0x415774 STR	D0, [X20]

0x415778 LDR	X3, [X19, #256]

0x41577c STR	D0, [X3, X22]

0x415780 LDR	D30, [X20]

0x415784 LDR	D0, [X21]

0x415788 FDIV	D1, D30, D0

0x41578c STR	D1, [X20]

0x415790 MOVZ	X20, #1

0x415794 CMP	X24, X20

0x415798 B.NE	415828

0x41579c B	4156f0

(267) 0x4157a0 MOVI	D0, #0

(267) 0x4157a4 STR	XZR, [X5, X22]

(267) 0x4157a8 STR	D0, [X23, X22]

(267) 0x4157ac LDR	X6, [X19, #256]

(267) 0x4157b0 STR	D0, [X6, X22]

(267) 0x4157b4 LDR	D29, [X21, X20,LSL #3]

(267) 0x4157b8 LDR	D28, [X23, X22]

(267) 0x4157bc FDIV	D3, D28, D29

(267) 0x4157c0 STR	D3, [X23, X22]

(267) 0x4157c4 ADD	X20, X20, #1

(267) 0x4157c8 LDRSW	X7, [X26, X20,LSL #2]

(267) 0x4157cc LDR	X0, [X25, X7,LSL #3]

(267) 0x4157d0 CBZ	X0, 41581c

(267) 0x4157d4 LDR	D0, [X21, X20,LSL #3]

(267) 0x4157d8 UBFM	X22, X20, #61, #31

(267) 0x4157dc LDR	D4, [X0, #8]

(267) 0x4157e0 LDR	X23, [X19, #280]

(267) 0x4157e4 LDR	X8, [X19, #304]

(267) 0x4157e8 FCMPE	D0, D4

(267) 0x4157ec ADD	X1, X23, X22

(267) 0x4157f0 ADD	X2, X8, X22

(267) 0x4157f4 B.MI	415860

(267) 0x4157f8 MOVI	D0, #0

(267) 0x4157fc STR	XZR, [X8, X22]

(267) 0x415800 STR	D0, [X23, X22]

(267) 0x415804 LDR	X9, [X19, #256]

(267) 0x415808 STR	D0, [X9, X22]

(267) 0x41580c LDR	D27, [X21, X20,LSL #3]

(267) 0x415810 LDR	D26, [X23, X22]

(267) 0x415814 FDIV	D5, D26, D27

(267) 0x415818 STR	D5, [X23, X22]

(267) 0x41581c ADD	X20, X20, #1

(267) 0x415820 CMP	X24, X20

(267) 0x415824 B.EQ	4156f0

(267) 0x415828 LDRSW	X4, [X26, X20,LSL #2]

(267) 0x41582c LDR	X0, [X25, X4,LSL #3]

(267) 0x415830 CBZ	X0, 4157c4

(267) 0x415834 LDR	D0, [X21, X20,LSL #3]

(267) 0x415838 UBFM	X22, X20, #61, #31

(267) 0x41583c LDR	D2, [X0, #8]

(267) 0x415840 LDR	X23, [X19, #280]

(267) 0x415844 LDR	X5, [X19, #304]

(267) 0x415848 FCMPE	D0, D2

(267) 0x41584c ADD	X1, X23, X22

(267) 0x415850 ADD	X2, X5, X22

(267) 0x415854 B.GE	4157a0

(267) 0x415858 BL	40d760

(267) 0x41585c B	4157ac

(267) 0x415860 BL	40d760

(267) 0x415864 B	415804

0x415868 ORR	X1, XZR, X20

0x41586c BL	40d760

0x415870 B	415778

0x415874 HINT	#0

0x415878 HINT	#0

0x41587c HINT	#0

Coverage (%)	Name	Source Location	Module

Path /

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

Source file and lines	OneBodyJastrowRef.h:205-241
Module	exec

The function is defined in:

/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/OneBodyJastrowRef.h:205-207,214-218,231-241
/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/BsplineFunctor.h:186-191
/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h:316
/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223,229
/home/hbollore/gnu/unknow_version/include/c++/14.0.1/bits/stl_algobase.h:952,1146-1151
/home/hbollore/gnu/unknow_version/include/c++/14.0.1/bits/stl_vector.h:1128-1131,1263
/home/hbollore/gnu/unknow_version/include/c++/14.0.1/bits/shared_ptr_base.h:1667

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 9.50 to 3.17 cycles (3.00x 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 21% of vector register length is used (average across all VPU instructions). By vectorizing your function, you can lower the cost of an iteration from 9.50 to 1.75 cycles (5.43x speedup).

Details

All VPU instructions are used in scalar version (process only one data element in vector registers). Since your execution units are vector units, only a vectorized function can use their full power.

Workaround

Try another compiler or update/tune your current one
Make array accesses unit-stride:
- If your function streams arrays of structures (AoS), try to use structures of arrays instead (SoA): for(i) a[i].x = b[i].x; (slow, non stride 1) => for(i) a.x[i] = b.x[i]; (fast, stride 1)

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 9.50 to 2.00 cycles (4.75x speedup).

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

The binary function is composed of 1 FP arithmetical operations:

1: divide

The binary function is loading 124 bytes. The binary function is storing 144 bytes.

Arithmetic intensity

Arithmetic intensity is 0.00 FP operations per loaded or stored byte.

General properties

nb instructions	72
loop length	288
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

8.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	6.50	6.50	7.25	7.25	7.25	7.25	2.00	2.00	1.00	1.00	9.50	9.17	9.33	4.00	4.00
cycles	6.50	6.50	7.25	7.25	7.25	7.25	2.00	2.00	1.00	1.00	9.50	9.17	9.33	4.00	4.00

Execution ports to units layout:

P0:
P1:
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

1.75-3.50

Cycles summary

Front-end	8.63
Overall L1	9.50

Vectorization ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 9.50 cycles. At this rate:

13% of peak load performance is reached (13.05 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
31% of peak store performance is reached (15.16 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the function is: 41560c

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput
SUB SP, SP, #96	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STP X29, X30, [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
ADD X29, SP, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STP X19, X20, [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
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
STP X21, X22, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
ORR X21, 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
LDR W22, [X0, #152]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR W2, [X0, #144]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP W22, #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 41571c <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x110>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
STP X23, X24, [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
STP X25, X26, [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
LDR X23, [X0, #256]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X24, [X0, #280]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X25, [X0, #304]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP W2, #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 415688 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x7c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
UBFM X26, X2, #61, #31	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ W1, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR X2, XZR, X26	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR X0, XZR, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
BL 403bd0 <@plt_start@+0x1e0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X2, XZR, X26	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ W1, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR X0, XZR, X24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
BL 403bd0 <@plt_start@+0x1e0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X2, XZR, X26	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR X0, XZR, X25	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ W1, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
BL 403bd0 <@plt_start@+0x1e0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
MOVZ X20, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X26, X19, #456	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP W2, #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 4156f8 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0xec>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
STP X23, X24, [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
SBFM X24, X2, #0, #31	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ X20, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STP X25, X26, [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
LDR X0, [X0, #160]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X25, [X19, #456]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X26, [X0, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
TBZ W24, #0, 415828 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x21c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDRSW X1, [X26]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X0, [X25, X1,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CBZ X0, 415790 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x184>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR D0, [X21]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
MOVZ X22, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
LDR D31, [X0, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
LDR X20, [X19, #280]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X2, [X19, #304]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
FCMPE D0, D31	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1
B.MI 415868 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x25c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
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
STR XZR, [X2]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR D0, [X20]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
LDR X3, [X19, #256]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
STR D0, [X3, X22]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
LDR D30, [X20]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
LDR D0, [X21]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
FDIV D1, D30, D0	1	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	1.75-3.50
STR D1, [X20]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
MOVZ X20, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X24, X20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.NE 415828 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x21c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
B 4156f0 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0xe4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X1, XZR, X20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
BL 40d760 <_ZN11qmcplusplus14BsplineFunctorIdE8evaluateEdRdS2_.part.0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
B 415778 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x16c>	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

Source file and lines	OneBodyJastrowRef.h:205-241
Module	exec

The function is defined in:

/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/OneBodyJastrowRef.h:205-207,214-218,231-241
/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/BsplineFunctor.h:186-191
/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h:316
/home/hbollore/qaas-runs/171-275-7410/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223,229
/home/hbollore/gnu/unknow_version/include/c++/14.0.1/bits/stl_algobase.h:952,1146-1151
/home/hbollore/gnu/unknow_version/include/c++/14.0.1/bits/stl_vector.h:1128-1131,1263
/home/hbollore/gnu/unknow_version/include/c++/14.0.1/bits/shared_ptr_base.h:1667

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 9.50 to 3.17 cycles (3.00x speedup).

Workaround

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

Vectorization

Details

All VPU instructions are used in scalar version (process only one data element in vector registers). Since your execution units are vector units, only a vectorized function can use their full power.

Workaround

Try another compiler or update/tune your current one
Make array accesses unit-stride:
- If your function streams arrays of structures (AoS), try to use structures of arrays instead (SoA): for(i) a[i].x = b[i].x; (slow, non stride 1) => for(i) a.x[i] = b.x[i]; (fast, stride 1)

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 9.50 to 2.00 cycles (4.75x speedup).

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

The binary function is composed of 1 FP arithmetical operations:

1: divide

The binary function is loading 124 bytes. The binary function is storing 144 bytes.

Arithmetic intensity

Arithmetic intensity is 0.00 FP operations per loaded or stored byte.

General properties

nb instructions	72
loop length	288
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

8.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	6.50	6.50	7.25	7.25	7.25	7.25	2.00	2.00	1.00	1.00	9.50	9.17	9.33	4.00	4.00
cycles	6.50	6.50	7.25	7.25	7.25	7.25	2.00	2.00	1.00	1.00	9.50	9.17	9.33	4.00	4.00

Execution ports to units layout:

P0:
P1:
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

1.75-3.50

Cycles summary

Front-end	8.63
Overall L1	9.50

Vectorization ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 9.50 cycles. At this rate:

13% of peak load performance is reached (13.05 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
31% of peak store performance is reached (15.16 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the function is: 41560c

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput
SUB SP, SP, #96	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STP X29, X30, [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
ADD X29, SP, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STP X19, X20, [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
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
STP X21, X22, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
ORR X21, 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
LDR W22, [X0, #152]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR W2, [X0, #144]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP W22, #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 41571c <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x110>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
STP X23, X24, [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
STP X25, X26, [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
LDR X23, [X0, #256]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X24, [X0, #280]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X25, [X0, #304]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP W2, #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 415688 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x7c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
UBFM X26, X2, #61, #31	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ W1, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR X2, XZR, X26	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR X0, XZR, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
BL 403bd0 <@plt_start@+0x1e0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X2, XZR, X26	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ W1, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR X0, XZR, X24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
BL 403bd0 <@plt_start@+0x1e0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X2, XZR, X26	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ORR X0, XZR, X25	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ W1, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
BL 403bd0 <@plt_start@+0x1e0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
MOVZ X20, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X26, X19, #456	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP W2, #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 4156f8 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0xec>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
STP X23, X24, [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
SBFM X24, X2, #0, #31	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
MOVZ X20, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STP X25, X26, [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
LDR X0, [X0, #160]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X25, [X19, #456]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X26, [X0, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
TBZ W24, #0, 415828 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x21c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDRSW X1, [X26]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X0, [X25, X1,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CBZ X0, 415790 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x184>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDR D0, [X21]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
MOVZ X22, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
LDR D31, [X0, #8]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
LDR X20, [X19, #280]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X2, [X19, #304]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
FCMPE D0, D31	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1
B.MI 415868 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x25c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
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
STR XZR, [X2]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STR D0, [X20]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
LDR X3, [X19, #256]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
STR D0, [X3, X22]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
LDR D30, [X20]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
LDR D0, [X21]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	6	0.33
FDIV D1, D30, D0	1	0	0	0	0	0	0	0.50	0	0.50	0	0	0	0	0	0	7-15	1.75-3.50
STR D1, [X20]	1	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50
MOVZ X20, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X24, X20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.NE 415828 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x21c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
B 4156f0 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0xe4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
ORR X1, XZR, X20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
BL 40d760 <_ZN11qmcplusplus14BsplineFunctorIdE8evaluateEdRdS2_.part.0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50
B 415778 <_ZN16miniqmcreference17OneBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9computeU3ERNS1_11ParticleSetEiPKd.isra.0+0x16c>	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

Name	Coverage (%)	Time (s)
▼miniqmcreference::OneBodyJastrowRef >::computeU3(qmcplusplus::ParticleSet&, int, double const*) [clone .isra.0]–	0.01	0.01
▼Loop 266 - OneBodyJastrowRef.h:214-241 - exec–	0	0.01
▼Loop 264 - OneBodyJastrowRef.h:214-218 - exec–	0	0
○Loop 265 - OneBodyJastrowRef.h:214-218 - exec	0	0.01
○Loop 267 - BsplineFunctor.h:188-191 - exec	0	0

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Expensive FP math instructions/calls

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

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