OV - exec - Function qmcplusplus::SoaDistanceTableAAOMPTarget<double, 3u, 40>::evaluate(qmcplusplus::ParticleSet&)

	Function: qmcplusplus::SoaDistanceTableAAOMPTarget<double, 3u, 40>::evaluate(qmcplusplus::ParticleSe ...	Module: exec	Source: SoaDistanceTableAAOMPTarget.h:179-187 [...]	Coverage (incl. loops): 0.07% \| (excl. loops): 0.00%

/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Particle/SoaDistanceTableAAOMPTarget.h: 179 - 187

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

179:   inline void evaluate(ParticleSet& P) override

180:   {

181:     ScopedTimer local_timer(evaluate_timer_);

182:

183:     constexpr T BigR = std::numeric_limits<T>::max();

184:     for (int iat = 1; iat < num_targets_; ++iat)

185:       DTD_BConds<T, D, SC>::computeDistances(P.R[iat], P.getCoordinates().getAllParticlePos(), distances_[iat].data(),

186:                                              displacements_[iat], 0, iat, iat);

187:   }

/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Particle/RealSpacePositionsOMPTarget.h: 163 - 163

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

163:   const PosVectorSoa& getAllParticlePos() const override { return RSoA_hostview; }

/usr/include/c++/14/bits/stl_vector.h: 1131 - 1131

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

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

/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h: 223 - 248

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

223:     return X[i];

[...]

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

/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h: 271 - 273

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

271:   inline T* restrict data(size_t i) { return myData + i * nGhosts; }

272:   ///return the const pointer of the i-th components

273:   inline const T* restrict data(size_t i) const { return myData + i * nGhosts; }

/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Particle/Lattice/ParticleBConds3DSoa.h: 234 - 255

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

234: #pragma omp simd aligned(temp_r, px, py, pz, dx, dy, dz: QMC_SIMD_ALIGNMENT)

235:     for (int iat = first; iat < last; ++iat)

236:     {

237:       T displ_0 = px[iat] - x0;

238:       T displ_1 = py[iat] - y0;

239:       T displ_2 = pz[iat] - z0;

240:

241:       T ar_0 = displ_0 * g00 + displ_1 * g10 + displ_2 * g20;

242:       T ar_1 = displ_0 * g01 + displ_1 * g11 + displ_2 * g21;

243:       T ar_2 = displ_0 * g02 + displ_1 * g12 + displ_2 * g22;

244:

245:       //put them in the box

246:       ar_0 -= round(ar_0);

247:       ar_1 -= round(ar_1);

248:       ar_2 -= round(ar_2);

249:

250:       //unit2cart

251:       dx[iat] = ar_0 * r00 + ar_1 * r10 + ar_2 * r20;

252:       dy[iat] = ar_0 * r01 + ar_1 * r11 + ar_2 * r21;

253:       dz[iat] = ar_0 * r02 + ar_1 * r12 + ar_2 * r22;

254:

255:       temp_r[iat] = std::sqrt(dx[iat] * dx[iat] + dy[iat] * dy[iat] + dz[iat] * dz[iat]);

/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h: 242 - 249

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

242:   ScopeGuard(TIMER& t) : timer(t) { timer.start(); }

[...]

249:   ~ScopeGuard() { timer.stop(); }

/usr/include/c++/14/bits/unique_ptr.h: 193 - 193

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

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

0x4ffa80 STP	X29, X30, [SP, #912]!

0x4ffa84 ADD	X29, SP, #0

0x4ffa88 STP	X23, X24, [SP, #48]

0x4ffa8c ORR	X24, XZR, X0

0x4ffa90 STP	X25, X26, [SP, #64]

0x4ffa94 ORR	X25, XZR, X1

0x4ffa98 LDR	X0, [X0, #632]

0x4ffa9c STR	X0, [SP, #104]

0x4ffaa0 BL	4cf640

0x4ffaa4 LDR	X13, [X24, #24]

0x4ffaa8 CMP	X13, #1

0x4ffaac B.LS	4ffc74

0x4ffab0 ADRP	X26,

0x4ffab4 MOVZ	X23, #24

0x4ffab8 STP	X19, X20, [SP, #16]

0x4ffabc ADD	X26, X26, #2848

0x4ffac0 ADD	X19, X24, #280

0x4ffac4 PTRUE	P7.B, ALL

0x4ffac8 STP	X21, X22, [SP, #32]

0x4ffacc MOVZ	X22, #1

0x4ffad0 CNTD	X21, ALL

0x4ffad4 STP	X27, X28, [SP, #80]

0x4ffad8 ADD	X28, X25, #40

0x4ffadc MOVZ	W27, #40

(2532) 0x4ffae0 LDR	X0, [X25, #632]

(2532) 0x4ffae4 LDR	X4, [X28, #24]

(2532) 0x4ffae8 LDR	X1, [X0]

(2532) 0x4ffaec ADD	X20, X4, X23

(2532) 0x4ffaf0 LDR	X2, [X1, #72]

(2532) 0x4ffaf4 CMP	X2, X26

(2532) 0x4ffaf8 B.NE	4ffc8c

(2532) 0x4ffafc ADD	X0, X0, #64

(2532) 0x4ffb00 LDR	X5, [X24, #96]

(2532) 0x4ffb04 UMULL	X3, W22, W27

(2532) 0x4ffb08 MOVZ	X2, #0

(2532) 0x4ffb0c ORR	W7, WZR, W22

(2532) 0x4ffb10 LD1RD	{Z31.D}, P7/Z, [X20]

(2532) 0x4ffb14 LD1RD	{Z30.D}, P7/Z, [X20, #1]

(2532) 0x4ffb18 LDR	X6, [X24, #72]

(2532) 0x4ffb1c LD1RD	{Z29.D}, P7/Z, [X20, #2]

(2532) 0x4ffb20 WHILELO	P6.D, WZR, W22

(2532) 0x4ffb24 LDR	X8, [X0, #8]

(2532) 0x4ffb28 ADD	X10, X5, X3

(2532) 0x4ffb2c LDR	X15, [X10, #8]

(2532) 0x4ffb30 ADD	X14, X6, X3

(2532) 0x4ffb34 LDR	X9, [X0, #24]

(2532) 0x4ffb38 LDR	X16, [X10, #24]

(2532) 0x4ffb3c LDR	X18, [X14, #24]

(2532) 0x4ffb40 ADD	X11, X9, X8,LSL #3

(2532) 0x4ffb44 ADD	X0, X9, X8,LSL #4

(2532) 0x4ffb48 ADD	X12, X16, X15,LSL #3

(2532) 0x4ffb4c ADD	X17, X16, X15,LSL #4

  (2531) 0x4ffb50 LD1D	{Z27.D}, P6/Z, [X11, X2,LSL #3]

  (2531) 0x4ffb54 LD1D	{Z28.D}, P6/Z, [X9, X2,LSL #3]

  (2531) 0x4ffb58 LD1D	{Z26.D}, P6/Z, [X0, X2,LSL #3]

  (2531) 0x4ffb5c LD1RD	{Z1.D}, P7/Z, [X19, #13]

  (2531) 0x4ffb60 LD1RD	{Z2.D}, P7/Z, [X19, #12]

  (2531) 0x4ffb64 LD1RD	{Z24.D}, P7/Z, [X19, #10]

  (2531) 0x4ffb68 LD1RD	{Z25.D}, P7/Z, [X19, #9]

  (2531) 0x4ffb6c LD1RD	{Z17.D}, P7/Z, [X19, #16]

  (2531) 0x4ffb70 LD1RD	{Z18.D}, P7/Z, [X19, #15]

  (2531) 0x4ffb74 FSUB	Z3.D, Z27.D, Z30.D

  (2531) 0x4ffb78 FSUB	Z5.D, Z28.D, Z31.D

  (2531) 0x4ffb7c FSUB	Z6.D, Z26.D, Z29.D

  (2531) 0x4ffb80 LD1RD	{Z0.D}, P7/Z, [X19, #14]

  (2531) 0x4ffb84 LD1RD	{Z23.D}, P7/Z, [X19, #11]

  (2531) 0x4ffb88 LD1RD	{Z16.D}, P7/Z, [X19, #17]

  (2531) 0x4ffb8c FMUL	Z4.D, Z1.D, Z3.D

  (2531) 0x4ffb90 FMUL	Z7.D, Z24.D, Z3.D

  (2531) 0x4ffb94 FMLA	Z4.D, P7/M, Z2.D, Z5.D

  (2531) 0x4ffb98 FMLA	Z7.D, P7/M, Z25.D, Z5.D

  (2531) 0x4ffb9c FMAD	Z0.D, P7/M, Z6.D, Z4.D

  (2531) 0x4ffba0 FMAD	Z23.D, P7/M, Z6.D, Z7.D

  (2531) 0x4ffba4 FMUL	Z19.D, Z17.D, Z3.D

  (2531) 0x4ffba8 FMLA	Z19.D, P7/M, Z18.D, Z5.D

  (2531) 0x4ffbac FMAD	Z16.D, P7/M, Z6.D, Z19.D

  (2531) 0x4ffbb0 LD1RD	{Z20.D}, P7/Z, [X19, #1]

  (2531) 0x4ffbb4 LD1RD	{Z3.D}, P7/Z, [X19]

  (2531) 0x4ffbb8 LD1RD	{Z5.D}, P7/Z, [X19, #2]

  (2531) 0x4ffbbc ADD	X30, X16, X2,LSL #3

  (2531) 0x4ffbc0 MOVPRFX	Z21, Z0

  (2531) 0x4ffbc4 FRINTA	Z21.D, P7/M, Z0.D

  (2531) 0x4ffbc8 MOVPRFX	Z22, Z23

  (2531) 0x4ffbcc FRINTA	Z22.D, P7/M, Z23.D

  (2531) 0x4ffbd0 FSUB	Z27.D, Z0.D, Z21.D

  (2531) 0x4ffbd4 FSUB	Z28.D, Z23.D, Z22.D

  (2531) 0x4ffbd8 ADD	X1, X12, X2,LSL #3

  (2531) 0x4ffbdc MOVPRFX	Z4, Z16

  (2531) 0x4ffbe0 FRINTA	Z4.D, P7/M, Z16.D

  (2531) 0x4ffbe4 FMUL	Z1.D, Z20.D, Z27.D

  (2531) 0x4ffbe8 FSUB	Z26.D, Z16.D, Z4.D

  (2531) 0x4ffbec FMLA	Z1.D, P7/M, Z3.D, Z28.D

  (2531) 0x4ffbf0 FMAD	Z5.D, P7/M, Z26.D, Z1.D

  (2531) 0x4ffbf4 ST1D	{Z5.D}, P6, [X30, MUL VL]

  (2531) 0x4ffbf8 LD1RD	{Z6.D}, P7/Z, [X19, #3]

  (2531) 0x4ffbfc LD1RD	{Z0.D}, P7/Z, [X19, #4]

  (2531) 0x4ffc00 LD1RD	{Z2.D}, P7/Z, [X19, #5]

  (2531) 0x4ffc04 FMUL	Z24.D, Z0.D, Z27.D

  (2531) 0x4ffc08 FMLA	Z24.D, P7/M, Z6.D, Z28.D

  (2531) 0x4ffc0c FMAD	Z2.D, P7/M, Z26.D, Z24.D

  (2531) 0x4ffc10 ST1D	{Z2.D}, P6, [X1, MUL VL]

  (2531) 0x4ffc14 LD1RD	{Z7.D}, P7/Z, [X19, #7]

  (2531) 0x4ffc18 LD1RD	{Z25.D}, P7/Z, [X19, #6]

  (2531) 0x4ffc1c LD1RD	{Z17.D}, P7/Z, [X19, #8]

  (2531) 0x4ffc20 FMUL	Z23.D, Z7.D, Z27.D

  (2531) 0x4ffc24 FMLA	Z23.D, P7/M, Z25.D, Z28.D

  (2531) 0x4ffc28 FMAD	Z17.D, P7/M, Z26.D, Z23.D

  (2531) 0x4ffc2c ST1D	{Z17.D}, P6, [X17, X2,LSL #3]

  (2531) 0x4ffc30 LD1D	{Z18.D}, P6/Z, [X1, MUL VL]

  (2531) 0x4ffc34 LD1D	{Z19.D}, P6/Z, [X30, MUL VL]

  (2531) 0x4ffc38 FMUL	Z16.D, Z18.D, Z18.D

  (2531) 0x4ffc3c FMLA	Z16.D, P7/M, Z19.D, Z19.D

  (2531) 0x4ffc40 FMLA	Z16.D, P7/M, Z17.D, Z17.D

  (2531) 0x4ffc44 FSQRT	Z16.D, P7/M, Z16.D

  (2531) 0x4ffc48 ST1D	{Z16.D}, P6, [X18, X2,LSL #3]

  (2531) 0x4ffc4c ADD	X2, X2, X21

  (2531) 0x4ffc50 WHILELO	P6.D, W2, W7

  (2531) 0x4ffc54 B.NE	4ffb50

(2532) 0x4ffc58 ADD	X22, X22, #1

(2532) 0x4ffc5c ADD	X23, X23, #24

(2532) 0x4ffc60 CMP	X22, X13

(2532) 0x4ffc64 B.CC	4ffae0

0x4ffc68 LDP	X19, X20, [SP, #16]

0x4ffc6c LDP	X21, X22, [SP, #32]

0x4ffc70 LDP	X27, X28, [SP, #80]

0x4ffc74 LDR	X0, [SP, #104]

0x4ffc78 BL	4cf8e0

0x4ffc7c LDP	X23, X24, [SP, #48]

0x4ffc80 LDP	X25, X26, [SP, #64]

0x4ffc84 LDP	X29, X30, [SP], #112

0x4ffc88 RET

(2532) 0x4ffc8c BLR	X2

(2532) 0x4ffc90 LDR	X13, [X24, #24]

(2532) 0x4ffc94 PTRUE	P7.B, ALL

(2532) 0x4ffc98 B	4ffb00

0x4ffc9c ORR	X22, XZR, X0

0x4ffca0 LDR	X0, [SP, #104]

0x4ffca4 BL	4cf8e0

0x4ffca8 ORR	X0, XZR, X22

0x4ffcac BL	4104a0

0x4ffcb0 HINT	#0

0x4ffcb4 HINT	#0

0x4ffcb8 HINT	#0

0x4ffcbc HINT	#0

Coverage (%)	Name	Source Location	Module
►98.69+	qmcplusplus::ParticleSet::upda[...]	stl_vector.h:993	exec
○	main._omp_fn.0	miniqmc.cpp:397	exec
○	omp_fulfill_event		libgomp.so.1.0.0
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►1.31+	qmcplusplus::ParticleSet::upda[...]	stl_vector.h:993	exec
○	main._omp_fn.0	miniqmc.cpp:397	exec
○	GOMP_parallel		libgomp.so.1.0.0
○	main	refwrap.h:350	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	miniqmc.cpp:529	exec

min	med	avg	max

Percentile Index	10	20	30	40	50	60	70	80	90	100
Value

min	med	avg	max

Percentile Index	10	20	30	40	50	60	70	80	90	100
Value

Path /

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

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run gcc_0

Source file and lines	SoaDistanceTableAAOMPTarget.h:179-187
Module	exec

The function is defined in:

/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Particle/SoaDistanceTableAAOMPTarget.h:179-187
/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Particle/RealSpacePositionsOMPTarget.h:163
/usr/include/c++/14/bits/stl_vector.h:1131
/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223,248
/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:271-273
/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Particle/Lattice/ParticleBConds3DSoa.h:234-255
/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:242,249
/usr/include/c++/14/bits/unique_ptr.h:193

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 5.83 to 1.00 cycles (5.83x 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 38% of vector register length is used (average across all VPU instructions). By vectorizing your function, you can lower the cost of an iteration from 5.83 to 3.75 cycles (1.56x speedup).

Details

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

Workaround

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

No data for this section

CALL instructions

Detected function call instructions.

Details

Calling (and then returning from) a function prevents many compiler optimizations (like vectorization), breaks control flow (which reduces pipeline performance) and executes extra instructions to save/restore the registers used inside it, which is very expensive (dozens of cycles). Consider to inline small functions.

qmcplusplus::TimerType
::start(): 1 occurrences►
- /home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:242
qmcplusplus::TimerType
::stop(): 2 occurrences►
- /home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:249
- /home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:249
unknown: 1 occurrences►
- /home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:249

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

The binary function does not contain any FP arithmetical operations. The binary function does not load or store any data.

General properties

nb instructions	42
nb uops	38
loop length	168
used w registers	1
used x registers	16
used b registers	0
used h registers	0
used s registers	1
used d registers	0
used q registers	0
used v registers	0
used z registers	0
nb stack references	15

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	4.75 cycles
front end	4.75 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	3.00	3.00	5.00	5.00	5.00	5.00	0.00	0.00	0.00	0.00	5.83	5.50	5.67	3.50	3.50
cycles	3.00	3.00	5.00	5.00	5.00	5.00	0.00	0.00	0.00	0.00	5.83	5.50	5.67	3.50	3.50

Execution ports to units layout:

P0: BRU
P1: BRU
P2: ALU
P3: ALU
P4: ALU
P5: ALU
P6 (128 bits): VPU, FP store data, ALU, DIV/SQRT
P7 (128 bits): VPU, ALU, FP store data
P8 (128 bits): VPU, ALU, DIV/SQRT
P9 (128 bits): ALU, VPU
P10 (256 bits): store address, load
P11 (256 bits): store address, load
P12 (256 bits): load
P13 (64 bits): store data
P14 (64 bits): store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	4.75
Dispatch	5.83
Overall L1	5.83

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%

Vector efficiency ratios

all	38%
load	43%
store	46%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	33%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 5.83 cycles.

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 4ffa80

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
STP X29, X30, [SP, #912]!	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X29, SP, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
STP X23, X24, [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	scal (50.0%)
ORR X24, XZR, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
STP X25, X26, [SP, #64]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ORR X25, 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	N/A
LDR X0, [X0, #632]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
STR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
BL 4cf640 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE5startEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X13, [X24, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
CMP X13, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.LS 4ffc74 <_ZN11qmcplusplus27SoaDistanceTableAAOMPTargetIdLj3ELi40EE8evaluateERNS_11ParticleSetE+0x1f4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADRP X26, <2004ccab0>	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
MOVZ X23, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
STP X19, X20, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X26, X26, #2848	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ADD X19, X24, #280	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
PTRUE P7.B, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (100.0%)
STP X21, X22, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
MOVZ X22, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
CNTD X21, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
STP X27, X28, [SP, #80]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X28, X25, #40	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
MOVZ W27, #40	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
LDP X19, X20, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
LDP X21, X22, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
LDP X27, X28, [SP, #80]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
BL 4cf8e0 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDP X23, X24, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X25, X26, [SP, #64]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X29, X30, [SP], #112	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
RET	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X22, 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	scal (25.0%)
LDR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
BL 4cf8e0 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X0, 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	N/A
BL 4104a0 <@plt_start@+0x480>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
HINT #0																			N/A
HINT #0																			N/A
HINT #0																			N/A
HINT #0																			N/A

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run gcc_0

Source file and lines	SoaDistanceTableAAOMPTarget.h:179-187
Module	exec

The function is defined in:

/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Particle/SoaDistanceTableAAOMPTarget.h:179-187
/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Particle/RealSpacePositionsOMPTarget.h:163
/usr/include/c++/14/bits/stl_vector.h:1131
/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223,248
/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:271-273
/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Particle/Lattice/ParticleBConds3DSoa.h:234-255
/home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:242,249
/usr/include/c++/14/bits/unique_ptr.h:193

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 5.83 to 1.00 cycles (5.83x speedup).

Workaround

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

Vectorization

Details

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

Workaround

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

No data for this section

CALL instructions

Detected function call instructions.

Details

qmcplusplus::TimerType
::start(): 1 occurrences►
- /home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:242
qmcplusplus::TimerType
::stop(): 2 occurrences►
- /home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:249
- /home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:249
unknown: 1 occurrences►
- /home/eoseret/qaas/qaas_runs/178-212-9071/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:249

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

The binary function does not contain any FP arithmetical operations. The binary function does not load or store any data.

General properties

nb instructions	42
nb uops	38
loop length	168
used w registers	1
used x registers	16
used b registers	0
used h registers	0
used s registers	1
used d registers	0
used q registers	0
used v registers	0
used z registers	0
nb stack references	15

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	4.75 cycles
front end	4.75 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	3.00	3.00	5.00	5.00	5.00	5.00	0.00	0.00	0.00	0.00	5.83	5.50	5.67	3.50	3.50
cycles	3.00	3.00	5.00	5.00	5.00	5.00	0.00	0.00	0.00	0.00	5.83	5.50	5.67	3.50	3.50

Execution ports to units layout:

P0: BRU
P1: BRU
P2: ALU
P3: ALU
P4: ALU
P5: ALU
P6 (128 bits): VPU, FP store data, ALU, DIV/SQRT
P7 (128 bits): VPU, ALU, FP store data
P8 (128 bits): VPU, ALU, DIV/SQRT
P9 (128 bits): ALU, VPU
P10 (256 bits): store address, load
P11 (256 bits): store address, load
P12 (256 bits): load
P13 (64 bits): store data
P14 (64 bits): store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	4.75
Dispatch	5.83
Overall L1	5.83

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%

Vector efficiency ratios

all	38%
load	43%
store	46%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	33%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 5.83 cycles.

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 4ffa80

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
STP X29, X30, [SP, #912]!	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X29, SP, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
STP X23, X24, [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	scal (50.0%)
ORR X24, XZR, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
STP X25, X26, [SP, #64]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ORR X25, 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	N/A
LDR X0, [X0, #632]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
STR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
BL 4cf640 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE5startEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X13, [X24, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
CMP X13, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.LS 4ffc74 <_ZN11qmcplusplus27SoaDistanceTableAAOMPTargetIdLj3ELi40EE8evaluateERNS_11ParticleSetE+0x1f4>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ADRP X26, <2004ccab0>	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
MOVZ X23, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
STP X19, X20, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X26, X26, #2848	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ADD X19, X24, #280	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
PTRUE P7.B, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (100.0%)
STP X21, X22, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
MOVZ X22, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
CNTD X21, ALL	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
STP X27, X28, [SP, #80]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X28, X25, #40	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
MOVZ W27, #40	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
LDP X19, X20, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
LDP X21, X22, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
LDP X27, X28, [SP, #80]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
BL 4cf8e0 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDP X23, X24, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X25, X26, [SP, #64]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X29, X30, [SP], #112	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
RET	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X22, 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	scal (25.0%)
LDR X0, [SP, #104]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
BL 4cf8e0 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
ORR X0, 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	N/A
BL 4104a0 <@plt_start@+0x480>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
HINT #0																			N/A
HINT #0																			N/A
HINT #0																			N/A
HINT #0																			N/A

Name	Coverage (%)	Time (s)
▼qmcplusplus::SoaDistanceTableAAOMPTarget::evaluate(qmcplusplus::ParticleSet&)–	0.07	0.10
▼Loop 2532 - SoaDistanceTableAAOMPTarget.h:184-185 - exec–	0.00	0.00
○Loop 2531 - ParticleBConds3DSoa.h:237-255 - exec	0.07	0.10

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

CALL instructions

Details

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

Code clean check

Workaround

Vectorization

Details

Workaround

CALL instructions

Details

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code