OV - exec - Function miniqmcreference::einspline_spo_ref<double>::evaluate(qmcplusplus::ParticleSet const&, int, qmcplusplus::Vector<double, std::allocator<double> >&, qmcplusplus::Vector<qmcplusplus::TinyVector<double, 3u>, std::allocator<qmcplusplus::TinyVector<double, 3u> > >&, qmcplusplus::Vector<double, std::allocator<double> >&)

	Function: miniqmcreference::einspline_spo_ref<double>::evaluate(qmcplusplus::ParticleSet const&, int ...	Module: exec	Source: einspline_spo_ref.hpp:203-230 [...]	Coverage (incl. loops): 0.59% \| (excl. loops): 0.00%

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h: 231 - 271

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

231:   inline const AoSElement_t operator[](size_t i) const { return AoSElement_t(myData + i, nGhosts); }

[...]

265:   inline T* data() { return myData; }

[...]

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

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVectorTensorOps.h: 150 - 152

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

150:     return TinyVector<Type_t, 3>(lhs[0] * rhs[0] + lhs[1] * rhs[3] + lhs[2] * rhs[6],

151:                                  lhs[0] * rhs[1] + lhs[1] * rhs[4] + lhs[2] * rhs[7],

152:                                  lhs[0] * rhs[2] + lhs[1] * rhs[5] + lhs[2] * rhs[8]);

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h: 242 - 249

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

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

[...]

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

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/einspline_spo_ref.hpp: 203 - 230

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

203:     ScopedTimer local_timer(timer);

204:

205:     auto u = Lattice.toUnit_floor(P.activeR(iat));

206:     for (int i = 0; i < nBlocks; ++i)

207:       MultiBsplineEvalRef::evaluate_vgh(einsplines[i], u[0], u[1], u[2], psi[i].data(), grad[i].data(), hess[i].data(),

208:                                         nSplinesPerBlock);

209:   }

210:

211:   inline void evaluate(const ParticleSet& P,

[...]

219:     for (int i = 0; i < nBlocks; ++i)

220:     {

221:       // in real simulation, phase needs to be applied. Here just fake computation

222:       const int first = i * nBlocks;

223:       for (int j = first; j < std::min((i + 1) * nSplinesPerBlock, OrbitalSetSize); j++)

224:       {

225:         psi_v[j]   = psi[i][j - first];

226:         dpsi_v[j]  = grad[i][j - first];

227:         d2psi_v[j] = hess[i].data(0)[j - first];

228:       }

229:     }

230:   }

/cluster/comp/gcc/14.2.0/include/c++/14.2.0/bits/stl_vector.h: 1131 - 1263

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

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

[...]

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

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/Lattice/CrystalLattice.h: 170 - 173

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

170:       if (-std::numeric_limits<T1>::epsilon() < val_dot[i] && val_dot[i] < 0)

171:         val_dot[i] = T1(0.0);

172:       else

173:         val_dot[i] -= std::floor(val_dot[i]);

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/Tensor.h: 213 - 213

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

213:   inline Type_t operator[](unsigned int i) const { return X[i]; }

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h: 145 - 145

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

145:       X[i] = base[i * offset];

/cluster/comp/gcc/14.2.0/include/c++/14.2.0/bits/stl_algobase.h: 238 - 238

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

238:       if (__b < __a)

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h: 223 - 229

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

223:     return X[i];

[...]

229:     return X[i];

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h: 143 - 143

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

143:     return (active_ptcl_ == iat) ? active_pos_ : R[iat];

0x4871c0 PUSH	%RBP

0x4871c1 MOV	%RSP,%RBP

0x4871c4 PUSH	%R15

0x4871c6 PUSH	%R14

0x4871c8 PUSH	%R13

0x4871ca MOVSXD	%EDX,%R13

0x4871cd PUSH	%R12

0x4871cf MOV	%RSI,%R12

0x4871d2 PUSH	%RBX

0x4871d3 MOV	%RDI,%RBX

0x4871d6 SUB	$0x68,%RSP

0x4871da MOV	0x358(%RDI),%R14

0x4871e1 MOV	%RCX,-0x78(%RBP)

0x4871e5 MOV	%R14,%RDI

0x4871e8 MOV	%R8,-0x80(%RBP)

0x4871ec MOV	%R9,-0x88(%RBP)

0x4871f3 CALL	52c850 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE5startEv>

0x4871f8 LEA	0x128(%R12),%RCX

0x487200 CMP	0x124(%R12),%R13D

0x487208 JE	487218

0x48720a MOV	0x40(%R12),%RAX

0x48720f LEA	(%R13,%R13,2),%RDX

0x487214 LEA	(%RAX,%RDX,8),%RCX

0x487218 VMOVUPD	(%RCX),%XMM0

0x48721c VMOVSD	0x10(%RCX),%XMM1

0x487221 VPERMILPD	$0x3,%XMM0,%XMM4

0x487227 VMULPD	0xf0(%RBX),%XMM4,%XMM8

0x48722f VPERMILPD	$0,%XMM0,%XMM3

0x487235 VMULSD	0x118(%RBX),%XMM1,%XMM6

0x48723d VMOVDDUP	%XMM1,%XMM2

0x487241 VMOVSD	(%RCX),%XMM5

0x487245 VFMADD231PD	0xd8(%RBX),%XMM3,%XMM8

0x48724e VXORPD	%XMM9,%XMM9,%XMM9

0x487253 VFMADD231SD	0xe8(%RBX),%XMM5,%XMM6

0x48725c VMOVDDUP	0x1304bc(%RIP),%XMM10

0x487264 VUNPCKHPD	%XMM0,%XMM0,%XMM7

0x487268 VFMADD132PD	0x108(%RBX),%XMM8,%XMM2

0x487271 VFMADD132SD	0x100(%RBX),%XMM6,%XMM7

0x48727a VCMPPD	$0x1,%XMM9,%XMM2,%K1

0x487281 VRNDSCALEPD	$0x9,%XMM2,%XMM11

0x487288 VCMPPD	$0xe,%XMM10,%XMM2,%K0{%K1}

0x48728f KMOVB	%K0,%ESI

0x487293 NOT	%ESI

0x487295 AND	$0x3,%ESI

0x487298 VCOMISD	0x130480(%RIP),%XMM7

0x4872a0 KMOVB	%ESI,%K2

0x4872a4 VSUBPD	%XMM11,%XMM2,%XMM12{%K2}{z}

0x4872aa JBE	4872bb

0x4872ac VXORPD	%XMM13,%XMM13,%XMM13

0x4872b1 VCOMISD	%XMM7,%XMM13

0x4872b5 JA	4876aa

0x4872bb VRNDSCALESD	$0x9,%XMM7,%XMM7,%XMM14

0x4872c2 VSUBSD	%XMM14,%XMM7,%XMM15

0x4872c7 VMOVQ	%XMM15,%RDI

0x4872cc MOV	0x30(%RBX),%R9D

0x4872d0 VUNPCKHPD	%XMM12,%XMM12,%XMM0

0x4872d5 VMOVLPD	%XMM12,-0x38(%RBP)

0x4872da XOR	%R15D,%R15D

0x4872dd VMOVQ	%RDI,%XMM2

0x4872e2 VMOVAPD	%XMM0,%XMM1

0x4872e6 TEST	%R9D,%R9D

0x4872e9 JLE	4873c9

(1026) 0x4872ef MOV	0x310(%RBX),%RAX

(1026) 0x4872f6 MOV	0x340(%RBX),%R11

(1026) 0x4872fd MOV	0x328(%RBX),%R12

(1026) 0x487304 MOV	0x2f8(%RBX),%RDI

(1026) 0x48730b LEA	(%R15,%R15,4),%R10

(1026) 0x48730f LEA	(%R15,%R15,2),%R13

(1026) 0x487313 SAL	$0x3,%R10

(1026) 0x487317 LEA	(%RAX,%R13,8),%RSI

(1026) 0x48731b MOVSXD	0x40(%RBX),%R8

(1026) 0x48731f MOV	0x18(%R11,%R10,1),%RCX

(1026) 0x487324 MOV	0x18(%R12,%R10,1),%RDX

(1026) 0x487329 MOV	(%RSI),%RSI

(1026) 0x48732c MOV	(%RDI,%R15,8),%RDI

(1026) 0x487330 VMOVSD	-0x38(%RBP),%XMM0

(1026) 0x487335 VMOVSD	%XMM2,-0x48(%RBP)

(1026) 0x48733a VMOVSD	%XMM1,-0x40(%RBP)

(1026) 0x48733f CALL	486540 <_ZN16miniqmcreference19MultiBsplineEvalRef12evaluate_vghIdEEvPKN11qmcplusplus14bspline_traitsIT_Lj3EE10SplineTypeES4_S4_S4_PS4_S9_S9_m>

(1026) 0x487344 LEA	0x1(%R15),%R8

(1026) 0x487348 CMP	%R8D,0x30(%RBX)

(1026) 0x48734c VMOVSD	-0x40(%RBP),%XMM1

(1026) 0x487351 VMOVSD	-0x48(%RBP),%XMM2

(1026) 0x487356 JLE	4873c9

(1026) 0x487358 MOV	0x328(%RBX),%R9

(1026) 0x48735f MOV	0x310(%RBX),%R12

(1026) 0x487366 LEA	(%R8,%R8,4),%RDX

(1026) 0x48736a SAL	$0x3,%RDX

(1026) 0x48736e LEA	(%R8,%R8,2),%R11

(1026) 0x487372 MOV	0x340(%RBX),%RCX

(1026) 0x487379 MOV	0x2f8(%RBX),%RAX

(1026) 0x487380 MOV	0x18(%R9,%RDX,1),%R10

(1026) 0x487385 LEA	(%R12,%R11,8),%R13

(1026) 0x487389 MOV	0x18(%RCX,%RDX,1),%RCX

(1026) 0x48738e MOV	(%RAX,%R8,8),%RDI

(1026) 0x487392 MOV	(%R13),%RSI

(1026) 0x487396 MOVSXD	0x40(%RBX),%R8

(1026) 0x48739a VMOVSD	-0x38(%RBP),%XMM0

(1026) 0x48739f MOV	%R10,%RDX

(1026) 0x4873a2 ADD	$0x2,%R15

(1026) 0x4873a6 VMOVSD	%XMM2,-0x48(%RBP)

(1026) 0x4873ab VMOVSD	%XMM1,-0x40(%RBP)

(1026) 0x4873b0 CALL	486540 <_ZN16miniqmcreference19MultiBsplineEvalRef12evaluate_vghIdEEvPKN11qmcplusplus14bspline_traitsIT_Lj3EE10SplineTypeES4_S4_S4_PS4_S9_S9_m>

(1026) 0x4873b5 CMP	%R15D,0x30(%RBX)

(1026) 0x4873b9 VMOVSD	-0x40(%RBP),%XMM1

(1026) 0x4873be VMOVSD	-0x48(%RBP),%XMM2

(1026) 0x4873c3 JG	4872ef

0x4873c9 MOV	%R14,%RDI

0x4873cc CALL	52ca80 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE4stopEv>

0x4873d1 MOV	0x30(%RBX),%R14D

0x4873d5 MOV	%R14D,-0x54(%RBP)

0x4873d9 TEST	%R14D,%R14D

0x4873dc JLE	48769b

0x4873e2 MOVSXD	-0x54(%RBP),%R8

0x4873e6 MOV	0x40(%RBX),%EDI

0x4873e9 LEA	(%R8,%R8,2),%RCX

0x4873ed LEA	(,%R8,8),%RDX

0x4873f5 SAL	$0x3,%RCX

0x4873f9 MOV	0x8(%RBX),%R15D

0x4873fd MOV	%EDI,-0x64(%RBP)

0x487400 MOV	%RDX,-0x60(%RBP)

0x487404 MOV	%RCX,-0x70(%RBP)

0x487408 MOV	%EDI,-0x40(%RBP)

0x48740b MOVQ	$0,-0x48(%RBP)

0x487413 MOVQ	$0,-0x50(%RBP)

0x48741b MOVL	$0,-0x38(%RBP)

0x487422 MOV	%R15D,-0x58(%RBP)

0x487426 XOR	%R14D,%R14D

0x487429 XOR	%R15D,%R15D

0x48742c NOPL	(%RAX)

(1024) 0x487430 MOV	-0x40(%RBP),%R9D

(1024) 0x487434 MOV	-0x58(%RBP),%R12D

(1024) 0x487438 CMP	%R12D,%R9D

(1024) 0x48743b CMOVLE	%R9D,%R12D

(1024) 0x48743f CMP	%R12D,-0x38(%RBP)

(1024) 0x487443 JGE	48766a

(1024) 0x487449 MOV	0x310(%RBX),%R10

(1024) 0x487450 MOV	0x328(%RBX),%R13

(1024) 0x487457 MOV	-0x80(%RBP),%RDI

(1024) 0x48745b MOV	-0x38(%RBP),%R9D

(1024) 0x48745f MOV	(%R10,%R14,1),%R11

(1024) 0x487463 MOV	-0x48(%RBP),%RDX

(1024) 0x487467 MOV	-0x78(%RBP),%R10

(1024) 0x48746b ADD	%R15,%R13

(1024) 0x48746e MOV	0x18(%R13),%RSI

(1024) 0x487472 MOVSXD	0x8(%R13),%RCX

(1024) 0x487476 SUB	%R9D,%R12D

(1024) 0x487479 MOV	-0x88(%RBP),%R13

(1024) 0x487480 MOV	0x18(%R10),%R10

(1024) 0x487484 ADD	0x18(%RDI),%RDX

(1024) 0x487488 MOV	-0x50(%RBP),%RDI

(1024) 0x48748c SAL	$0x3,%R12

(1024) 0x487490 ADD	%RDI,%R10

(1024) 0x487493 ADD	0x18(%R13),%RDI

(1024) 0x487497 LEA	-0x8(%R12),%R13

(1024) 0x48749c MOV	0x340(%RBX),%RAX

(1024) 0x4874a3 SHR	$0x3,%R13

(1024) 0x4874a7 LEA	(%RSI,%RCX,8),%R9

(1024) 0x4874ab INC	%R13

(1024) 0x4874ae SAL	$0x4,%RCX

(1024) 0x4874b2 MOV	0x18(%RAX,%R15,1),%R8

(1024) 0x4874b7 ADD	%RSI,%RCX

(1024) 0x4874ba XOR	%EAX,%EAX

(1024) 0x4874bc AND	$0x3,%R13D

(1024) 0x4874c0 JE	487580

(1024) 0x4874c6 CMP	$0x1,%R13

(1024) 0x4874ca JE	48753e

(1024) 0x4874cc CMP	$0x2,%R13

(1024) 0x4874d0 JE	487505

(1024) 0x4874d2 VMOVSD	(%R11),%XMM3

(1024) 0x4874d7 ADD	$0x18,%RDX

(1024) 0x4874db VMOVSD	%XMM3,(%R10)

(1024) 0x4874e0 VMOVSD	(%RSI),%XMM1

(1024) 0x4874e4 VMOVSD	(%RCX),%XMM4

(1024) 0x4874e8 VMOVHPD	(%R9),%XMM1,%XMM2

(1024) 0x4874ed VMOVUPD	%XMM2,-0x18(%RDX)

(1024) 0x4874f2 VMOVSD	%XMM4,-0x8(%RDX)

(1024) 0x4874f7 VMOVSD	(%R8),%XMM6

(1024) 0x4874fc MOV	$0x8,%EAX

(1024) 0x487501 VMOVSD	%XMM6,(%RDI)

(1024) 0x487505 VMOVSD	(%R11,%RAX,1),%XMM5

(1024) 0x48750b ADD	$0x18,%RDX

(1024) 0x48750f VMOVSD	%XMM5,(%R10,%RAX,1)

(1024) 0x487515 VMOVSD	(%RSI,%RAX,1),%XMM8

(1024) 0x48751a VMOVSD	(%RCX,%RAX,1),%XMM7

(1024) 0x48751f VMOVHPD	(%R9,%RAX,1),%XMM8,%XMM9

(1024) 0x487525 VMOVUPD	%XMM9,-0x18(%RDX)

(1024) 0x48752a VMOVSD	%XMM7,-0x8(%RDX)

(1024) 0x48752f VMOVSD	(%R8,%RAX,1),%XMM10

(1024) 0x487535 VMOVSD	%XMM10,(%RDI,%RAX,1)

(1024) 0x48753a ADD	$0x8,%RAX

(1024) 0x48753e VMOVSD	(%R11,%RAX,1),%XMM11

(1024) 0x487544 ADD	$0x18,%RDX

(1024) 0x487548 VMOVSD	%XMM11,(%R10,%RAX,1)

(1024) 0x48754e VMOVSD	(%RSI,%RAX,1),%XMM13

(1024) 0x487553 VMOVSD	(%RCX,%RAX,1),%XMM12

(1024) 0x487558 VMOVHPD	(%R9,%RAX,1),%XMM13,%XMM14

(1024) 0x48755e VMOVUPD	%XMM14,-0x18(%RDX)

(1024) 0x487563 VMOVSD	%XMM12,-0x8(%RDX)

(1024) 0x487568 VMOVSD	(%R8,%RAX,1),%XMM15

(1024) 0x48756e VMOVSD	%XMM15,(%RDI,%RAX,1)

(1024) 0x487573 ADD	$0x8,%RAX

(1024) 0x487577 CMP	%R12,%RAX

(1024) 0x48757a JE	48766a

  (1025) 0x487580 VMOVSD	(%R11,%RAX,1),%XMM0

  (1025) 0x487586 ADD	$0x60,%RDX

  (1025) 0x48758a VMOVSD	%XMM0,(%R10,%RAX,1)

  (1025) 0x487590 VMOVSD	(%RSI,%RAX,1),%XMM4

  (1025) 0x487595 VMOVSD	(%RCX,%RAX,1),%XMM3

  (1025) 0x48759a VMOVHPD	(%R9,%RAX,1),%XMM4,%XMM1

  (1025) 0x4875a0 VMOVUPD	%XMM1,-0x60(%RDX)

  (1025) 0x4875a5 VMOVSD	%XMM3,-0x50(%RDX)

  (1025) 0x4875aa VMOVSD	(%R8,%RAX,1),%XMM2

  (1025) 0x4875b0 VMOVSD	%XMM2,(%RDI,%RAX,1)

  (1025) 0x4875b5 VMOVSD	0x8(%R11,%RAX,1),%XMM6

  (1025) 0x4875bc VMOVSD	%XMM6,0x8(%RAX,%R10,1)

  (1025) 0x4875c3 VMOVSD	0x8(%RSI,%RAX,1),%XMM7

  (1025) 0x4875c9 VMOVSD	0x8(%RCX,%RAX,1),%XMM5

  (1025) 0x4875cf VMOVHPD	0x8(%R9,%RAX,1),%XMM7,%XMM8

  (1025) 0x4875d6 VMOVUPD	%XMM8,-0x48(%RDX)

  (1025) 0x4875db VMOVSD	%XMM5,-0x38(%RDX)

  (1025) 0x4875e0 VMOVSD	0x8(%R8,%RAX,1),%XMM9

  (1025) 0x4875e7 VMOVSD	%XMM9,0x8(%RAX,%RDI,1)

  (1025) 0x4875ed VMOVSD	0x10(%R11,%RAX,1),%XMM10

  (1025) 0x4875f4 VMOVSD	%XMM10,0x10(%RAX,%R10,1)

  (1025) 0x4875fb VMOVSD	0x10(%RSI,%RAX,1),%XMM12

  (1025) 0x487601 VMOVSD	0x10(%RCX,%RAX,1),%XMM11

  (1025) 0x487607 VMOVHPD	0x10(%R9,%RAX,1),%XMM12,%XMM13

  (1025) 0x48760e VMOVUPD	%XMM13,-0x30(%RDX)

  (1025) 0x487613 VMOVSD	%XMM11,-0x20(%RDX)

  (1025) 0x487618 VMOVSD	0x10(%R8,%RAX,1),%XMM14

  (1025) 0x48761f VMOVSD	%XMM14,0x10(%RAX,%RDI,1)

  (1025) 0x487625 VMOVSD	0x18(%R11,%RAX,1),%XMM15

  (1025) 0x48762c VMOVSD	%XMM15,0x18(%RAX,%R10,1)

  (1025) 0x487633 VMOVSD	0x18(%RSI,%RAX,1),%XMM0

  (1025) 0x487639 VMOVSD	0x18(%RCX,%RAX,1),%XMM3

  (1025) 0x48763f VMOVHPD	0x18(%R9,%RAX,1),%XMM0,%XMM4

  (1025) 0x487646 VMOVUPD	%XMM4,-0x18(%RDX)

  (1025) 0x48764b VMOVSD	%XMM3,-0x8(%RDX)

  (1025) 0x487650 VMOVSD	0x18(%R8,%RAX,1),%XMM1

  (1025) 0x487657 ADD	$0x20,%RAX

  (1025) 0x48765b VMOVSD	%XMM1,-0x8(%RAX,%RDI,1)

  (1025) 0x487661 CMP	%R12,%RAX

  (1025) 0x487664 JNE	487580

(1024) 0x48766a MOV	-0x64(%RBP),%R12D

(1024) 0x48766e MOV	-0x54(%RBP),%R11D

(1024) 0x487672 MOV	-0x60(%RBP),%RSI

(1024) 0x487676 MOV	-0x70(%RBP),%RCX

(1024) 0x48767a ADD	$0x18,%R14

(1024) 0x48767e ADD	%R12D,-0x40(%RBP)

(1024) 0x487682 ADD	%R11D,-0x38(%RBP)

(1024) 0x487686 ADD	%RSI,-0x50(%RBP)

(1024) 0x48768a ADD	%RCX,-0x48(%RBP)

(1024) 0x48768e ADD	$0x28,%R15

(1024) 0x487692 CMP	%R14,%RCX

(1024) 0x487695 JNE	487430

0x48769b ADD	$0x68,%RSP

0x48769f POP	%RBX

0x4876a0 POP	%R12

0x4876a2 POP	%R13

0x4876a4 POP	%R14

0x4876a6 POP	%R15

0x4876a8 POP	%RBP

0x4876a9 RET

0x4876aa VMOVQ	%XMM13,%RDI

0x4876af JMP	4872cc

0x4876b4 NOPW	%CS:(%RAX,%RAX,1)

0x4876be XCHG	%AX,%AX

Coverage (%)	Name	Source Location	Module
►73.74+	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:100	exec
○	qmcplusplus::WaveFunction::rat[...]	WaveFunction.cpp:198	exec
○	main._omp_fn.1	refwrap.h:350	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0
►11.88+	miniqmcreference::DiracDetermi[...]	OhmmsVector.h:144	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main._omp_fn.0	miniqmc.cpp:397	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0
►11.75+	miniqmcreference::DiracDetermi[...]	OhmmsVector.h:144	exec
○	qmcplusplus::WaveFunction::eva[...]	WaveFunction.cpp:171	exec
○	main._omp_fn.0	miniqmc.cpp:397	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0
►1.79+	miniqmcreference::DiracDetermi[...]	DiracDeterminantRef.cpp:100	exec
○	qmcplusplus::WaveFunction::rat[...]	WaveFunction.cpp:198	exec
○	main._omp_fn.1	refwrap.h:350	exec
○	GOMP_parallel	libgomp.h:982	libgomp.so.1.0.0

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_3

Source file and lines	einspline_spo_ref.hpp:203-230
Module	exec

The function is defined in:

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:231,265,271
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVectorTensorOps.h:150-152
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:242,249
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/einspline_spo_ref.hpp:203-211,219-230
/cluster/comp/gcc/14.2.0/include/c++/14.2.0/bits/stl_vector.h:1131,1263
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/Lattice/CrystalLattice.h:170-173
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/Tensor.h:213
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h:145
/cluster/comp/gcc/14.2.0/include/c++/14.2.0/bits/stl_algobase.h:238
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223,229
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h:143

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 17.00 to 8.50 cycles (2.00x speedup).

Workaround

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

Vectorization

Your function is poorly vectorized. Only 13% of vector register length is used (average across all SSE/AVX instructions). By fully vectorizing your function, you can lower the cost of an iteration from 17.00 to 1.60 cycles (10.63x speedup).

Details

24% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

25% of SSE/AVX loads are used in vector version.
0% of SSE/AVX stores are used in vector version.
50% of SSE/AVX addition or subtraction instructions are used in vector version.
50% of SSE/AVX multiply instructions are used in vector version.
50% of SSE/AVX fused multiply-add instructions are used in vector version.
32% of SSE/AVX 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)

Execution units bottlenecks

Found no such bottlenecks but see expert reports for more complex bottlenecks.

Masked instructions

Detected masked instructions.

Details

Vector registers are partially exploited, which is expected if your function is irregular or mixes elements of different sizes.

Workaround

If your function is irregular, try to remove or hoist conditional structures out of your function. If it mixes elements of different sizes, try to uniformize them.

FMA

Detected 6 FMA (fused multiply-add) operations. Presence of both ADD/SUB and MUL operations.

Workaround

Try to change order in which elements are evaluated (using parentheses) in arithmetic expressions containing both ADD/SUB and MUL operations to enable your compiler to generate FMA instructions wherever possible. For instance a + b*c is a valid FMA (MUL then ADD). However (a+b)* c cannot be translated into an FMA (ADD then MUL).

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_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:242
qmcplusplus::TimerType
::stop(): 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:249

Complex instructions

Detected COMPLEX INSTRUCTIONS.

Details

These instructions generate more than one micro-operation and only one of them can be decoded during a cycle and the extra micro-operations increase pressure on execution units.

VCOMISD: 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/Lattice/CrystalLattice.h:170
VRNDSCALEPD: 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/Lattice/CrystalLattice.h:173
VRNDSCALESD: 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/Lattice/CrystalLattice.h:173

Type of elements and instruction set

7 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in scalar mode (one at a time). 7 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (two at a time).

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

The binary function is composed of 18 FP arithmetical operations:

9: addition or subtraction (6 inside FMA instructions)
9: multiply (6 inside FMA instructions)

The binary function is loading 160 bytes (20 double precision FP elements). The binary function is storing 84 bytes (10 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.07 FP operations per loaded or stored byte.

General properties

nb instructions	97
nb uops	102
loop length	443
used x86 registers	14
used mmx registers	0
used xmm registers	16
used ymm registers	0
used zmm registers	0
nb stack references	12
ADD-SUB / MUL ratio	1.00

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	17.00 cycles
front end	17.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	8.00	9.00	8.67	8.67	10.50	9.00	7.00	10.50	10.50	10.50	6.00	8.67
cycles	8.00	9.00	8.67	8.67	10.50	9.00	7.00	10.50	10.50	10.50	6.00	8.67

Execution ports to units layout:

P0 (256 bits): VPU, BRU, ALU, DIV/SQRT
P1 (256 bits): ALU, VPU
P2 (512 bits): load
P3 (512 bits): load
P4 (512 bits): store data
P5 (512 bits): ALU, VPU
P6: ALU, BRU
P7: store address
P8: store address
P9 (512 bits): store data
P10: ALU
P11 (512 bits): load

Cycles executing div or sqrt instructions

Cycles summary

Front-end	17.00
Dispatch	10.50
Overall L1	17.00

Vectorization ratios

INT

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

all	48%
load	36%
store	0%
mul	50%
add-sub	50%
fma	50%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	57%

INT+FP

all	24%
load	25%
store	0%
mul	50%
add-sub	50%
fma	50%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	32%

Vector efficiency ratios

INT

all	8%
load	7%
store	8%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	7%

all	18%
load	17%
store	12%
mul	18%
add-sub	18%
fma	18%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	19%

INT+FP

all	13%
load	14%
store	9%
mul	18%
add-sub	18%
fma	18%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	14%

Cycles and memory resources usage

Detected masked instructions: assuming all mask elements are active. Assuming all data fit into the L1 cache, each call to the function takes 17.00 cycles. At this rate:

4% of peak load performance is reached (9.41 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
3% of peak store performance is reached (4.94 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Performance is limited by instruction throughput (loading/decoding program instructions to execution core) (front-end is a bottleneck). By removing all these bottlenecks, you can lower the cost of an iteration from 17.00 to 10.50 cycles (1.62x speedup).

ASM code

In the binary file, the address of the function is: 4871c0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
PUSH %R15	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %R14	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %R13	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOVSXD %EDX,%R13	1	0	0.33	0	0	0	0.33	0	0	0	0	0.33	0	1	0.33	N/A
PUSH %R12	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %RSI,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
PUSH %RBX	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %RDI,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
SUB $0x68,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x358(%RDI),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV %RCX,-0x78(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %R14,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %R8,-0x80(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %R9,-0x88(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
CALL 52c850 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE5startEv>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
LEA 0x128(%R12),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMP 0x124(%R12),%R13D	1	0.20	0.20	0.33	0.33	0	0.20	0.20	0	0	0	0.20	0.33	1	0.33	scal (6.3%)
JE 487218 <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0x58>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV 0x40(%R12),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
LEA (%R13,%R13,2),%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA (%RAX,%RDX,8),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VMOVUPD (%RCX),%XMM0	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (25.0%)
VMOVSD 0x10(%RCX),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VPERMILPD $0x3,%XMM0,%XMM4	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	vect (25.0%)
VMULPD 0xf0(%RBX),%XMM4,%XMM8	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	vect (25.0%)
VPERMILPD $0,%XMM0,%XMM3	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	vect (25.0%)
VMULSD 0x118(%RBX),%XMM1,%XMM6	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVDDUP %XMM1,%XMM2	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
VMOVSD (%RCX),%XMM5	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VFMADD231PD 0xd8(%RBX),%XMM3,%XMM8	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	vect (25.0%)
VXORPD %XMM9,%XMM9,%XMM9	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
VFMADD231SD 0xe8(%RBX),%XMM5,%XMM6	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVDDUP 0x1304bc(%RIP),%XMM10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VUNPCKHPD %XMM0,%XMM0,%XMM7	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
VFMADD132PD 0x108(%RBX),%XMM8,%XMM2	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	vect (25.0%)
VFMADD132SD 0x100(%RBX),%XMM6,%XMM7	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VCMPPD $0x1,%XMM9,%XMM2,%K1	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
VRNDSCALEPD $0x9,%XMM2,%XMM11	2	1	1	0	0	0	0	0	0	0	0	0	0	8	1	vect (25.0%)
VCMPPD $0xe,%XMM10,%XMM2,%K0{%K1}	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
KMOVB %K0,%ESI	1	1	0	0	0	0	0	0	0	0	0	0	0	3	1	N/A
NOT %ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
AND $0x3,%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	scal (6.3%)
VCOMISD 0x130480(%RIP),%XMM7	2	1	0	0.33	0.33	0	0	0	0	0	0	0	0.33	3	1	scal (12.5%)
KMOVB %ESI,%K2	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	N/A
VSUBPD %XMM11,%XMM2,%XMM12{%K2}{z}	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
JBE 4872bb <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0xfb>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
VXORPD %XMM13,%XMM13,%XMM13	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
VCOMISD %XMM7,%XMM13	1	1	0	0	0	0	0	0	0	0	0	0	0	3	1	scal (12.5%)
JA 4876aa <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0x4ea>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
VRNDSCALESD $0x9,%XMM7,%XMM7,%XMM14	2	1	1	0	0	0	0	0	0	0	0	0	0	8	1	scal (12.5%)
VSUBSD %XMM14,%XMM7,%XMM15	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
VMOVQ %XMM15,%RDI	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
MOV 0x30(%RBX),%R9D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
VUNPCKHPD %XMM12,%XMM12,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
VMOVLPD %XMM12,-0x38(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	4-12	0.50	scal (12.5%)
XOR %R15D,%R15D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
VMOVQ %RDI,%XMM2	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
VMOVAPD %XMM0,%XMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
TEST %R9D,%R9D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (6.3%)
JLE 4873c9 <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0x209>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %R14,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
CALL 52ca80 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE4stopEv>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV 0x30(%RBX),%R14D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
MOV %R14D,-0x54(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
TEST %R14D,%R14D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (6.3%)
JLE 48769b <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0x4db>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOVSXD -0x54(%RBP),%R8	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x40(%RBX),%EDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
LEA (%R8,%R8,2),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA (,%R8,8),%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SAL $0x3,%RCX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
MOV 0x8(%RBX),%R15D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
MOV %EDI,-0x64(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOV %RDX,-0x60(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %RCX,-0x70(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %EDI,-0x40(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOVQ $0,-0x48(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOVQ $0,-0x50(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOVL $0,-0x38(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOV %R15D,-0x58(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
XOR %R14D,%R14D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
XOR %R15D,%R15D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
ADD $0x68,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R13	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %RBP	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
RET	1	0.50	0	0.33	0.33	0	0	0.50	0	0	0	0	0.33	0	2.13	N/A
VMOVQ %XMM13,%RDI	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
JMP 4872cc <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0x10c>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPW %CS:(%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
XCHG %AX,%AX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run gcc_3

Source file and lines	einspline_spo_ref.hpp:203-230
Module	exec

The function is defined in:

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:231,265,271
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVectorTensorOps.h:150-152
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:242,249
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/einspline_spo_ref.hpp:203-211,219-230
/cluster/comp/gcc/14.2.0/include/c++/14.2.0/bits/stl_vector.h:1131,1263
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/Lattice/CrystalLattice.h:170-173
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/Tensor.h:213
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h:145
/cluster/comp/gcc/14.2.0/include/c++/14.2.0/bits/stl_algobase.h:238
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223,229
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h:143

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 17.00 to 8.50 cycles (2.00x speedup).

Workaround

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

Vectorization

Details

24% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

25% of SSE/AVX loads are used in vector version.
0% of SSE/AVX stores are used in vector version.
50% of SSE/AVX addition or subtraction instructions are used in vector version.
50% of SSE/AVX multiply instructions are used in vector version.
50% of SSE/AVX fused multiply-add instructions are used in vector version.
32% of SSE/AVX 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)

Execution units bottlenecks

Found no such bottlenecks but see expert reports for more complex bottlenecks.

Masked instructions

Detected masked instructions.

Details

Vector registers are partially exploited, which is expected if your function is irregular or mixes elements of different sizes.

Workaround

If your function is irregular, try to remove or hoist conditional structures out of your function. If it mixes elements of different sizes, try to uniformize them.

FMA

Detected 6 FMA (fused multiply-add) operations. Presence of both ADD/SUB and MUL operations.

Workaround

CALL instructions

Detected function call instructions.

Details

qmcplusplus::TimerType
::start(): 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:242
qmcplusplus::TimerType
::stop(): 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:249

Complex instructions

Detected COMPLEX INSTRUCTIONS.

Details

These instructions generate more than one micro-operation and only one of them can be decoded during a cycle and the extra micro-operations increase pressure on execution units.

VCOMISD: 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/Lattice/CrystalLattice.h:170
VRNDSCALEPD: 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/Lattice/CrystalLattice.h:173
VRNDSCALESD: 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/Lattice/CrystalLattice.h:173

Type of elements and instruction set

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

The binary function is composed of 18 FP arithmetical operations:

9: addition or subtraction (6 inside FMA instructions)
9: multiply (6 inside FMA instructions)

The binary function is loading 160 bytes (20 double precision FP elements). The binary function is storing 84 bytes (10 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.07 FP operations per loaded or stored byte.

General properties

nb instructions	97
nb uops	102
loop length	443
used x86 registers	14
used mmx registers	0
used xmm registers	16
used ymm registers	0
used zmm registers	0
nb stack references	12
ADD-SUB / MUL ratio	1.00

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	17.00 cycles
front end	17.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	8.00	9.00	8.67	8.67	10.50	9.00	7.00	10.50	10.50	10.50	6.00	8.67
cycles	8.00	9.00	8.67	8.67	10.50	9.00	7.00	10.50	10.50	10.50	6.00	8.67

Execution ports to units layout:

P0 (256 bits): VPU, BRU, ALU, DIV/SQRT
P1 (256 bits): ALU, VPU
P2 (512 bits): load
P3 (512 bits): load
P4 (512 bits): store data
P5 (512 bits): ALU, VPU
P6: ALU, BRU
P7: store address
P8: store address
P9 (512 bits): store data
P10: ALU
P11 (512 bits): load

Cycles executing div or sqrt instructions

Cycles summary

Front-end	17.00
Dispatch	10.50
Overall L1	17.00

Vectorization ratios

INT

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

all	48%
load	36%
store	0%
mul	50%
add-sub	50%
fma	50%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	57%

INT+FP

all	24%
load	25%
store	0%
mul	50%
add-sub	50%
fma	50%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	32%

Vector efficiency ratios

INT

all	8%
load	7%
store	8%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	7%

all	18%
load	17%
store	12%
mul	18%
add-sub	18%
fma	18%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	19%

INT+FP

all	13%
load	14%
store	9%
mul	18%
add-sub	18%
fma	18%
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	14%

Cycles and memory resources usage

Detected masked instructions: assuming all mask elements are active. Assuming all data fit into the L1 cache, each call to the function takes 17.00 cycles. At this rate:

4% of peak load performance is reached (9.41 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
3% of peak store performance is reached (4.94 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

ASM code

In the binary file, the address of the function is: 4871c0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
PUSH %R15	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %R14	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %R13	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOVSXD %EDX,%R13	1	0	0.33	0	0	0	0.33	0	0	0	0	0.33	0	1	0.33	N/A
PUSH %R12	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %RSI,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
PUSH %RBX	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %RDI,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
SUB $0x68,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x358(%RDI),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV %RCX,-0x78(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %R14,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %R8,-0x80(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %R9,-0x88(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
CALL 52c850 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE5startEv>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
LEA 0x128(%R12),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMP 0x124(%R12),%R13D	1	0.20	0.20	0.33	0.33	0	0.20	0.20	0	0	0	0.20	0.33	1	0.33	scal (6.3%)
JE 487218 <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0x58>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV 0x40(%R12),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
LEA (%R13,%R13,2),%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA (%RAX,%RDX,8),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VMOVUPD (%RCX),%XMM0	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (25.0%)
VMOVSD 0x10(%RCX),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VPERMILPD $0x3,%XMM0,%XMM4	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	vect (25.0%)
VMULPD 0xf0(%RBX),%XMM4,%XMM8	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	vect (25.0%)
VPERMILPD $0,%XMM0,%XMM3	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	vect (25.0%)
VMULSD 0x118(%RBX),%XMM1,%XMM6	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVDDUP %XMM1,%XMM2	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
VMOVSD (%RCX),%XMM5	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VFMADD231PD 0xd8(%RBX),%XMM3,%XMM8	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	vect (25.0%)
VXORPD %XMM9,%XMM9,%XMM9	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
VFMADD231SD 0xe8(%RBX),%XMM5,%XMM6	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVDDUP 0x1304bc(%RIP),%XMM10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VUNPCKHPD %XMM0,%XMM0,%XMM7	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
VFMADD132PD 0x108(%RBX),%XMM8,%XMM2	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	vect (25.0%)
VFMADD132SD 0x100(%RBX),%XMM6,%XMM7	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VCMPPD $0x1,%XMM9,%XMM2,%K1	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
VRNDSCALEPD $0x9,%XMM2,%XMM11	2	1	1	0	0	0	0	0	0	0	0	0	0	8	1	vect (25.0%)
VCMPPD $0xe,%XMM10,%XMM2,%K0{%K1}	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
KMOVB %K0,%ESI	1	1	0	0	0	0	0	0	0	0	0	0	0	3	1	N/A
NOT %ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
AND $0x3,%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	scal (6.3%)
VCOMISD 0x130480(%RIP),%XMM7	2	1	0	0.33	0.33	0	0	0	0	0	0	0	0.33	3	1	scal (12.5%)
KMOVB %ESI,%K2	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	N/A
VSUBPD %XMM11,%XMM2,%XMM12{%K2}{z}	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
JBE 4872bb <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0xfb>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
VXORPD %XMM13,%XMM13,%XMM13	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
VCOMISD %XMM7,%XMM13	1	1	0	0	0	0	0	0	0	0	0	0	0	3	1	scal (12.5%)
JA 4876aa <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0x4ea>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
VRNDSCALESD $0x9,%XMM7,%XMM7,%XMM14	2	1	1	0	0	0	0	0	0	0	0	0	0	8	1	scal (12.5%)
VSUBSD %XMM14,%XMM7,%XMM15	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
VMOVQ %XMM15,%RDI	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
MOV 0x30(%RBX),%R9D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
VUNPCKHPD %XMM12,%XMM12,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
VMOVLPD %XMM12,-0x38(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	4-12	0.50	scal (12.5%)
XOR %R15D,%R15D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
VMOVQ %RDI,%XMM2	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
VMOVAPD %XMM0,%XMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
TEST %R9D,%R9D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (6.3%)
JLE 4873c9 <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0x209>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %R14,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
CALL 52ca80 <_ZN11qmcplusplus9TimerTypeINSt6chrono3_V212system_clockEE4stopEv>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV 0x30(%RBX),%R14D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
MOV %R14D,-0x54(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
TEST %R14D,%R14D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (6.3%)
JLE 48769b <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0x4db>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOVSXD -0x54(%RBP),%R8	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x40(%RBX),%EDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
LEA (%R8,%R8,2),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA (,%R8,8),%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SAL $0x3,%RCX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
MOV 0x8(%RBX),%R15D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
MOV %EDI,-0x64(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOV %RDX,-0x60(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %RCX,-0x70(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %EDI,-0x40(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOVQ $0,-0x48(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOVQ $0,-0x50(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOVL $0,-0x38(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOV %R15D,-0x58(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
XOR %R14D,%R14D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
XOR %R15D,%R15D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
ADD $0x68,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R13	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %RBP	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
RET	1	0.50	0	0.33	0.33	0	0	0.50	0	0	0	0	0.33	0	2.13	N/A
VMOVQ %XMM13,%RDI	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
JMP 4872cc <_ZN16miniqmcreference17einspline_spo_refIdE8evaluateERKN11qmcplusplus11ParticleSetEiRNS2_6VectorIdSaIdEEERNS6_INS2_10TinyVectorIdLj3EEESaISB_EEES9_+0x10c>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPW %CS:(%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
XCHG %AX,%AX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A

Name	Coverage (%)	Time (s)
▼miniqmcreference::einspline_spo_ref::evaluate(qmcplusplus::ParticleSet const&, int, qmcplusplus::Vector >&, qmcplusplus::Vector, std::allocator >...–	0.59	0.51
▼Loop 1024 - einspline_spo_ref.hpp:219-227 - exec–	0.00	0.01
○Loop 1025 - einspline_spo_ref.hpp:223-227 - exec	0.58	0.49
○Loop 1026 - einspline_spo_ref.hpp:206-207 - exec	0.00	0.01

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

Masked instructions

Details

Workaround

FMA

Workaround

CALL instructions

Details

Complex instructions

Details

Type of elements and instruction set

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

Arithmetic intensity

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

Execution units bottlenecks

Masked instructions

Details

Workaround

FMA

Workaround

CALL instructions

Details

Complex instructions

Details

Type of elements and instruction set

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

Arithmetic intensity

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code