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

	Function: miniqmcreference::TwoBodyJastrowRef<qmcplusplus::BsplineFunctor<double> >::ratioGrad(qmcpl ...	Module: exec	Source: TwoBodyJastrowRef.h:155-306 [...]	Coverage (incl. loops): 0.38% \| (excl. loops): 0.01%

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h: 94 - 94

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

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

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h: 273 - 273

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

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

/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_numeric.h: 140 - 141

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

140:       for (; __first != __last; ++__first)

141: 	__init = _GLIBCXX_MOVE_IF_20(__init) + *__first;

/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/shared_ptr_base.h: 1296 - 1296

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

1296:       { return _M_ptr; }

/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_vector.h: 1169 - 1169

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

1169:       { return _M_data_ptr(this->_M_impl._M_start); }

/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algobase.h: 923 - 1115

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

923:       for (; __first != __last; ++__first)

924: 	*__first = __tmp;

[...]

1115:       if (__n <= 0)

/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_iterator.h: 1028 - 1182

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

1028:       : _M_current(__i) { }

[...]

1182:     { return __lhs.base() != __rhs.base(); }

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h: 155 - 306

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

155:       for (int jat = 0; jat < N; ++jat)

156:         s += du[jat] * dX[jat];

157:       grad[idim] = s;

[...]

262:   const int jelmax = triangle ? iat : N;

[...]

268:   const int igt = P.GroupID[iat] * NumGroups;

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

270:   {

271:     const FuncType& f2(*F[igt + jg]);

272:     int iStart = P.first(jg);

273:     int iEnd   = std::min(jelmax, P.last(jg));

274:     f2.evaluateVGL(iat, iStart, iEnd, dist, u, du, d2u, DistCompressed.data(), DistIndice.data());

[...]

299: {

300:   UpdateMode = ORB_PBYP_PARTIAL;

301:

302:   computeU3(P, iat, P.getDistTableAA(myTableID).getTempDists().data(), cur_u.data(), cur_du.data(), cur_d2u.data());

303:   cur_Uat = std::accumulate(cur_u.begin(), cur_u.begin() + N, valT());

304:   DiffVal = Uat[iat] - cur_Uat;

305:   grad_iat += accumulateG(cur_du.data(), P.getDistTableAA(myTableID).getTempDispls());

306:   return std::exp(DiffVal);

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h: 223 - 249

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

223:     return X[i];

[...]

229:     return X[i];

[...]

249:   inline const_pointer data() const { return X; }

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h: 313 - 316

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

313:   inline int first(int igroup) const { return (*group_offsets_)[igroup]; }

314:

315:   ///return the last index of a group i

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

0x421f30 PUSH	%RBP

0x421f31 MOV	%RSP,%RBP

0x421f34 PUSH	%R15

0x421f36 PUSH	%R14

0x421f38 PUSH	%R13

0x421f3a PUSH	%R12

0x421f3c PUSH	%RBX

0x421f3d SUB	$0x38,%RSP

0x421f41 MOV	%RCX,-0x60(%RBP)

0x421f45 MOV	%EDX,%R14D

0x421f48 MOV	%RSI,%R13

0x421f4b MOV	%RDI,%RBX

0x421f4e MOVL	$0x2,0xc(%RDI)

0x421f55 MOV	0x248(%RDI),%ESI

0x421f5b MOV	%R13,%RDI

0x421f5e CALL	460b50 <_ZNK11qmcplusplus11ParticleSet14getDistTableAAEi>

0x421f63 MOV	%RBX,%R10

0x421f66 MOV	0x90(%RAX),%R8

0x421f6d MOV	0x90(%RBX),%RBX

0x421f74 MOV	0x140(%R10),%R12

0x421f7b MOV	0x158(%R10),%R11

0x421f82 MOV	0x170(%R10),%R15

0x421f89 TEST	%EBX,%EBX

0x421f8b MOV	%R10,-0x48(%RBP)

0x421f8f MOV	%R11,-0x40(%RBP)

0x421f93 JLE	421fe8

0x421f95 MOV	%RBX,%RAX

0x421f98 AND	$0x7fffffff,%RAX

0x421f9e JE	421fe8

0x421fa0 SAL	$0x3,%RAX

0x421fa4 MOV	%R15,-0x30(%RBP)

0x421fa8 MOV	%R12,%RDI

0x421fab XOR	%ESI,%ESI

0x421fad MOV	%RAX,%R15

0x421fb0 MOV	%RAX,%RDX

0x421fb3 MOV	%R8,-0x38(%RBP)

0x421fb7 CALL	4efda0 <_intel_fast_memset>

0x421fbc MOV	-0x40(%RBP),%RDI

0x421fc0 XOR	%ESI,%ESI

0x421fc2 MOV	%R15,%RDX

0x421fc5 CALL	4efda0 <_intel_fast_memset>

0x421fca MOV	-0x30(%RBP),%RDI

0x421fce XOR	%ESI,%ESI

0x421fd0 MOV	%R15,%RDX

0x421fd3 CALL	4efda0 <_intel_fast_memset>

0x421fd8 MOV	-0x30(%RBP),%R15

0x421fdc MOV	-0x40(%RBP),%R11

0x421fe0 MOV	-0x38(%RBP),%R8

0x421fe4 MOV	-0x48(%RBP),%R10

0x421fe8 MOVSXD	%R14D,%RDX

0x421feb MOV	0xa0(%R10),%RAX

0x421ff2 TEST	%RAX,%RAX

0x421ff5 JE	42209e

0x421ffb MOV	0x18(%R13),%RCX

0x421fff MOV	%RDX,-0x58(%RBP)

0x422003 IMUL	(%RCX,%RDX,4),%EAX

0x422007 CLTQ

0x422009 SAL	$0x3,%RAX

0x42200d MOV	%RAX,-0x30(%RBP)

0x422011 XOR	%EAX,%EAX

0x422013 MOV	%R12,-0x38(%RBP)

0x422017 MOV	%R14D,-0x4c(%RBP)

0x42201b MOV	%R15,%R12

0x42201e XCHG	%AX,%AX

(370) 0x422020 MOV	0x200(%R10),%RCX

(370) 0x422027 ADD	-0x30(%RBP),%RCX

(370) 0x42202b MOV	(%RCX,%RAX,8),%RDI

(370) 0x42202f MOV	0x268(%R13),%RCX

(370) 0x422036 MOV	0x18(%RCX),%RCX

(370) 0x42203a MOV	%R13,%R14

(370) 0x42203d LEA	0x1(%RAX),%R13

(370) 0x422041 MOV	(%RCX,%RAX,4),%EDX

(370) 0x422044 MOV	0x4(%RCX,%RAX,4),%ECX

(370) 0x422048 CMP	%EBX,%ECX

(370) 0x42204a CMOVGE	%EBX,%ECX

(370) 0x42204d MOV	-0x4c(%RBP),%ESI

(370) 0x422050 MOV	%R8,%R15

(370) 0x422053 MOV	-0x38(%RBP),%R9

(370) 0x422057 PUSHQ	0x1e8(%R10)

(370) 0x42205e PUSHQ	0x1d0(%R10)

(370) 0x422065 PUSH	%R12

(370) 0x422067 PUSH	%R11

(370) 0x422069 CALL	41fba0 <_ZNK11qmcplusplus14BsplineFunctorIdE11evaluateVGLEiiiPKdPdS4_S4_S4_Pi>

(370) 0x42206e MOV	-0x40(%RBP),%R11

(370) 0x422072 MOV	%R15,%R8

(370) 0x422075 MOV	-0x48(%RBP),%R10

(370) 0x422079 ADD	$0x20,%RSP

(370) 0x42207d MOV	%R13,%RAX

(370) 0x422080 CMP	%R13,0xa0(%R10)

(370) 0x422087 MOV	%R14,%R13

(370) 0x42208a JA	422020

0x42208c MOV	0x90(%R10),%RBX

0x422093 MOV	0x140(%R10),%R12

0x42209a MOV	-0x58(%RBP),%RDX

0x42209e LEA	(,%RBX,8),%RAX

0x4220a6 VXORPD	%XMM0,%XMM0,%XMM0

0x4220aa TEST	%RAX,%RAX

0x4220ad JE	422114

0x4220af MOV	$0x1fffffffffffffff,%RAX

0x4220b9 ADD	%RAX,%RBX

0x4220bc AND	%RAX,%RBX

0x4220bf INC	%RBX

0x4220c2 MOV	$0x3ffffffffffffffc,%RAX

0x4220cc AND	%RBX,%RAX

0x4220cf JE	422104

0x4220d1 VXORPD	%XMM0,%XMM0,%XMM0

0x4220d5 XOR	%ECX,%ECX

0x4220d7 NOPW	(%RAX,%RAX,1)

(369) 0x4220e0 VADDPD	(%R12,%RCX,8),%YMM0,%YMM0

(369) 0x4220e6 ADD	$0x4,%RCX

(369) 0x4220ea CMP	%RAX,%RCX

(369) 0x4220ed JB	4220e0

0x4220ef VEXTRACTF128	$0x1,%YMM0,%XMM1

0x4220f5 VADDPD	%XMM1,%XMM0,%XMM0

0x4220f9 VSHUFPD	$0x1,%XMM0,%XMM0,%XMM1

0x4220fe VADDSD	%XMM1,%XMM0,%XMM0

0x422102 JMP	42210f

0x422104 XOR	%EAX,%EAX

(368) 0x422106 VADDSD	(%R12,%RAX,8),%XMM0,%XMM0

(368) 0x42210c INC	%RAX

(368) 0x42210f CMP	%RAX,%RBX

(368) 0x422112 JNE	422106

0x422114 VMOVSD	%XMM0,0x138(%R10)

0x42211d MOV	0xd8(%R10),%RAX

0x422124 MOV	0x158(%R10),%RBX

0x42212b VMOVSD	(%RAX,%RDX,8),%XMM1

0x422130 VSUBSD	%XMM0,%XMM1,%XMM0

0x422134 VMOVSD	%XMM0,0xb0(%R10)

0x42213d MOV	0x248(%R10),%ESI

0x422144 MOV	%R13,%RDI

0x422147 VZEROUPPER

0x42214a CALL	460b50 <_ZNK11qmcplusplus11ParticleSet14getDistTableAAEi>

0x42214f MOV	-0x48(%RBP),%R10

0x422153 MOV	0x90(%R10),%RCX

0x42215a TEST	%RCX,%RCX

0x42215d JE	4221aa

0x42215f MOV	0xb8(%RAX),%RDX

0x422166 MOV	%RCX,%RSI

0x422169 AND	$-0x4,%RSI

0x42216d JE	4221b7

0x42216f LEA	-0x1(%RSI),%RDI

0x422173 VXORPD	%XMM0,%XMM0,%XMM0

0x422177 XOR	%R8D,%R8D

0x42217a NOPW	(%RAX,%RAX,1)

(367) 0x422180 VMOVUPD	(%RDX,%R8,8),%YMM1

(367) 0x422186 VFMADD231PD	(%RBX,%R8,8),%YMM1,%YMM0

(367) 0x42218c ADD	$0x4,%R8

(367) 0x422190 CMP	%RDI,%R8

(367) 0x422193 JLE	422180

0x422195 VEXTRACTF128	$0x1,%YMM0,%XMM1

0x42219b VADDPD	%XMM1,%XMM0,%XMM0

0x42219f VSHUFPD	$0x1,%XMM0,%XMM0,%XMM1

0x4221a4 VADDSD	%XMM1,%XMM0,%XMM0

0x4221a8 JMP	4221cb

0x4221aa VXORPD	%XMM0,%XMM0,%XMM0

0x4221ae VXORPD	%XMM2,%XMM2,%XMM2

0x4221b2 JMP	4222c3

0x4221b7 XOR	%ESI,%ESI

0x4221b9 VXORPD	%XMM0,%XMM0,%XMM0

(362) 0x4221bd VMOVSD	(%RDX,%RSI,8),%XMM1

(362) 0x4221c2 VFMADD231SD	(%RBX,%RSI,8),%XMM1,%XMM0

(362) 0x4221c8 INC	%RSI

(362) 0x4221cb CMP	%RSI,%RCX

(362) 0x4221ce JNE	4221bd

0x4221d0 MOV	0xa8(%RAX),%RAX

0x4221d7 MOV	%RCX,%RSI

0x4221da AND	$-0x4,%RSI

0x4221de JE	42221f

0x4221e0 LEA	-0x1(%RSI),%RDI

0x4221e4 LEA	(%RDX,%RAX,8),%R8

0x4221e8 VXORPD	%XMM1,%XMM1,%XMM1

0x4221ec XOR	%R9D,%R9D

0x4221ef NOP

(366) 0x4221f0 VMOVUPD	(%R8,%R9,8),%YMM2

(366) 0x4221f6 VFMADD231PD	(%RBX,%R9,8),%YMM2,%YMM1

(366) 0x4221fc ADD	$0x4,%R9

(366) 0x422200 CMP	%RDI,%R9

(366) 0x422203 JLE	4221f0

0x422205 VEXTRACTF128	$0x1,%YMM1,%XMM2

0x42220b VADDPD	%XMM2,%XMM1,%XMM1

0x42220f VSHUFPD	$0x1,%XMM1,%XMM1,%XMM2

0x422214 VADDSD	%XMM2,%XMM1,%XMM1

0x422218 CMP	%RSI,%RCX

0x42221b JNE	422225

0x42221d JMP	422243

0x42221f XOR	%ESI,%ESI

0x422221 VXORPD	%XMM1,%XMM1,%XMM1

0x422225 LEA	(%RDX,%RAX,8),%RDI

0x422229 NOPL	(%RAX)

(363) 0x422230 VMOVSD	(%RDI,%RSI,8),%XMM2

(363) 0x422235 VFMADD231SD	(%RBX,%RSI,8),%XMM2,%XMM1

(363) 0x42223b INC	%RSI

(363) 0x42223e CMP	%RSI,%RCX

(363) 0x422241 JNE	422230

0x422243 VPUNPCKLQDQ	%XMM1,%XMM0,%XMM0

0x422247 MOV	%RCX,%RSI

0x42224a AND	$-0x4,%RSI

0x42224e JE	42229f

0x422250 LEA	-0x1(%RSI),%RDI

0x422254 MOV	%RAX,%R8

0x422257 SAL	$0x4,%R8

0x42225b ADD	%RDX,%R8

0x42225e VXORPD	%XMM1,%XMM1,%XMM1

0x422262 XOR	%R9D,%R9D

0x422265 NOPW	%CS:(%RAX,%RAX,1)

(365) 0x422270 VMOVUPD	(%R8,%R9,8),%YMM2

(365) 0x422276 VFMADD231PD	(%RBX,%R9,8),%YMM2,%YMM1

(365) 0x42227c ADD	$0x4,%R9

(365) 0x422280 CMP	%RDI,%R9

(365) 0x422283 JLE	422270

0x422285 VEXTRACTF128	$0x1,%YMM1,%XMM2

0x42228b VADDPD	%XMM2,%XMM1,%XMM1

0x42228f VSHUFPD	$0x1,%XMM1,%XMM1,%XMM2

0x422294 VADDSD	%XMM2,%XMM1,%XMM2

0x422298 CMP	%RSI,%RCX

0x42229b JNE	4222a5

0x42229d JMP	4222c3

0x42229f XOR	%ESI,%ESI

0x4222a1 VXORPD	%XMM2,%XMM2,%XMM2

0x4222a5 SAL	$0x4,%RAX

0x4222a9 ADD	%RAX,%RDX

0x4222ac NOPL	(%RAX)

(364) 0x4222b0 VMOVSD	(%RDX,%RSI,8),%XMM1

(364) 0x4222b5 VFMADD231SD	(%RBX,%RSI,8),%XMM1,%XMM2

(364) 0x4222bb INC	%RSI

(364) 0x4222be CMP	%RSI,%RCX

(364) 0x4222c1 JNE	4222b0

0x4222c3 MOV	-0x60(%RBP),%RAX

0x4222c7 VADDPD	(%RAX),%XMM0,%XMM0

0x4222cb VMOVUPD	%XMM0,(%RAX)

0x4222cf VADDSD	0x10(%RAX),%XMM2,%XMM0

0x4222d4 VMOVSD	%XMM0,0x10(%RAX)

0x4222d9 VMOVSD	0xb0(%R10),%XMM0

0x4222e2 ADD	$0x38,%RSP

0x4222e6 POP	%RBX

0x4222e7 POP	%R12

0x4222e9 POP	%R13

0x4222eb POP	%R14

0x4222ed POP	%R15

0x4222ef POP	%RBP

0x4222f0 VZEROUPPER

0x4222f3 JMP	4df280

0x4222f8 NOPL	(%RAX,%RAX,1)

Coverage (%)	Name	Source Location	Module
►99.87+	qmcplusplus::WaveFunction::rat[...]	WaveFunction.cpp:201	exec
○	main.extracted.107	refwrap.h:338	exec
○	__kmp_invoke_microtask		libiomp5.so
○	__kmp_invoke_task_func		libiomp5.so

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 0.01% of application time for run orig_0

Source file and lines	TwoBodyJastrowRef.h:155-306
Module	exec

The function is defined in:

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:94
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:273
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_numeric.h:140-141
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/shared_ptr_base.h:1296
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_vector.h:1169
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algobase.h:923-924,1115
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_iterator.h:1028,1182
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:155-157,262,268-274,299-306
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223,229,249
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h:313-316

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 30.83 to 10.00 cycles (3.08x 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 16% of vector register length is used (average across all SSE/AVX instructions).

Details

36% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

8% of SSE/AVX loads are used in vector version.
7% of SSE/AVX stores are used in vector version.
41% of SSE/AVX addition or subtraction instructions are used in vector version.
55% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

Workaround

Read the "512-bits vectorization" report at "Potential" confidence level.

Execution units bottlenecks

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

512-bits vectorization

On some x86 processors supporting 512-bits vectorization, compilers are often too conservative and limit vectorization to 256 bits. Performance can then be improved by enforcing 512-bits vectorization, especially with many vectorized and high trip count loops. 512-bits vectorization performance overhead (compared to 256-bits) is generally lower on newer processors.

Workaround

Recompile with -mprefer-vector-width=512

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.

_intel_fast_memset: 3 occurrences►
- /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algobase.h:924
- /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algobase.h:924
- /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algobase.h:924
qmcplusplus::ParticleSet::getDistTableAA(int) const: 2 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:302
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:305

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.

VZEROUPPER: 2 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:305
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:306

Vector unaligned load/store instructions

Detected 4 suboptimal vector unaligned load/store instructions.

Details

VEXTRACTF128: 4 occurrences►
- /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_numeric.h:141
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:156
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:156
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:156

Workaround

Pass to your compiler a micro-architecture specialization option:
- use march=native
Use vector aligned instructions:
1. align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
2. inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.

Conversion instructions

Detected expensive conversion instructions.

Details

CLTQ: 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:269

Workaround

Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.

Type of elements and instruction set

6 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in scalar mode (one at a time). 5 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 16 FP arithmetical operations:

16: addition or subtraction

The binary function is loading 236 bytes (29 double precision FP elements). The binary function is storing 112 bytes (14 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.05 FP operations per loaded or stored byte.

General properties

nb instructions	178
nb uops	185
loop length	719
used x86 registers	16
used mmx registers	0
used xmm registers	3
used ymm registers	2
used zmm registers	0
nb stack references	7

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	30.83 cycles
front end	30.83 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	11.10	11.00	12.00	12.00	12.50	11.00	10.90	12.50	12.50	12.50	11.00	12.00
cycles	11.10	11.00	12.00	12.00	12.50	11.00	10.90	12.50	12.50	12.50	11.00	12.00

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	30.83
Dispatch	12.50
Overall L1	30.83

Vectorization ratios

INT

all	5%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
other	11%

all	68%
load	25%
store	25%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	45%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	100%

INT+FP

all	36%
load	8%
store	7%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	41%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	55%

Vector efficiency ratios

INT

all	11%
load	12%
store	11%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
other	12%

all	21%
load	15%
store	15%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	18%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	25%

INT+FP

all	16%
load	13%
store	12%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	17%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	18%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 30.83 cycles. At this rate:

3% of peak load performance is reached (7.65 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (3.63 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 30.83 to 12.50 cycles (2.47x speedup).

ASM code

In the binary file, the address of the function is: 421f30

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
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
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
SUB $0x38,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RCX,-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 %EDX,%R14D	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV %RSI,%R13	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RDI,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOVL $0x2,0xc(%RDI)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOV 0x248(%RDI),%ESI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %R13,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
CALL 460b50 <_ZNK11qmcplusplus11ParticleSet14getDistTableAAEi>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV %RBX,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x90(%RAX),%R8	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x90(%RBX),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x140(%R10),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x158(%R10),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x170(%R10),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
TEST %EBX,%EBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (6.3%)
MOV %R10,-0x48(%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 %R11,-0x40(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
JLE 421fe8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0xb8>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %RBX,%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $0x7fffffff,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 421fe8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0xb8>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
SAL $0x3,%RAX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
MOV %R15,-0x30(%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 %R12,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RAX,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RAX,%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %R8,-0x38(%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 4efda0 <_intel_fast_memset>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV -0x40(%RBP),%RDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R15,%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
CALL 4efda0 <_intel_fast_memset>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV -0x30(%RBP),%RDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R15,%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
CALL 4efda0 <_intel_fast_memset>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV -0x30(%RBP),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x40(%RBP),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x38(%RBP),%R8	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x48(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOVSXD %R14D,%RDX	1	0	0.33	0	0	0	0.33	0	0	0	0	0.33	0	1	0.33	N/A
MOV 0xa0(%R10),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
TEST %RAX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (12.5%)
JE 42209e <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x16e>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV 0x18(%R13),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %RDX,-0x58(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
IMUL (%RCX,%RDX,4),%EAX	1	0	1	0.33	0.33	0	0	0	0	0	0	0	0.33	3	1	N/A
CLTQ	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.33	scal (12.5%)
SAL $0x3,%RAX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
MOV %RAX,-0x30(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
XOR %EAX,%EAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R12,-0x38(%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 %R14D,-0x4c(%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 %R15,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
XCHG %AX,%AX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV 0x90(%R10),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x140(%R10),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x58(%RBP),%RDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
LEA (,%RBX,8),%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM0,%XMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
TEST %RAX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (12.5%)
JE 422114 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x1e4>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV $0x1fffffffffffffff,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
ADD %RAX,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
AND %RAX,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
INC %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV $0x3ffffffffffffffc,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
AND %RBX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 422104 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x1d4>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
VXORPD %XMM0,%XMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
XOR %ECX,%ECX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
NOPW (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VEXTRACTF128 $0x1,%YMM0,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
VADDPD %XMM1,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
VSHUFPD $0x1,%XMM0,%XMM0,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
VADDSD %XMM1,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
JMP 42210f <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x1df>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
XOR %EAX,%EAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VMOVSD %XMM0,0x138(%R10)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV 0xd8(%R10),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x158(%R10),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
VMOVSD (%RAX,%RDX,8),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VSUBSD %XMM0,%XMM1,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
VMOVSD %XMM0,0xb0(%R10)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV 0x248(%R10),%ESI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %R13,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VZEROUPPER	2	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 460b50 <_ZNK11qmcplusplus11ParticleSet14getDistTableAAEi>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV -0x48(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x90(%R10),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
TEST %RCX,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (12.5%)
JE 4221aa <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x27a>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV 0xb8(%RAX),%RDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %RCX,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $-0x4,%RSI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 4221b7 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x287>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
LEA -0x1(%RSI),%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM0,%XMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
NOPW (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VEXTRACTF128 $0x1,%YMM0,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
VADDPD %XMM1,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
VSHUFPD $0x1,%XMM0,%XMM0,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
VADDSD %XMM1,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
JMP 4221cb <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x29b>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
VXORPD %XMM0,%XMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
VXORPD %XMM2,%XMM2,%XMM2	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
JMP 4222c3 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x393>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM0,%XMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
MOV 0xa8(%RAX),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %RCX,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $-0x4,%RSI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 42221f <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x2ef>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
LEA -0x1(%RSI),%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA (%RDX,%RAX,8),%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM1,%XMM1,%XMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
XOR %R9D,%R9D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
NOP	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VEXTRACTF128 $0x1,%YMM1,%XMM2	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
VADDPD %XMM2,%XMM1,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
VSHUFPD $0x1,%XMM1,%XMM1,%XMM2	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
VADDSD %XMM2,%XMM1,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
CMP %RSI,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JNE 422225 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x2f5>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
JMP 422243 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x313>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM1,%XMM1,%XMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
LEA (%RDX,%RAX,8),%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VPUNPCKLQDQ %XMM1,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	scal (12.5%)
MOV %RCX,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $-0x4,%RSI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 42229f <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x36f>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
LEA -0x1(%RSI),%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RAX,%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
SAL $0x4,%R8	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	scal (12.5%)
ADD %RDX,%R8	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
VXORPD %XMM1,%XMM1,%XMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
XOR %R9D,%R9D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
NOPW %CS:(%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VEXTRACTF128 $0x1,%YMM1,%XMM2	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
VADDPD %XMM2,%XMM1,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
VSHUFPD $0x1,%XMM1,%XMM1,%XMM2	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
VADDSD %XMM2,%XMM1,%XMM2	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
CMP %RSI,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JNE 4222a5 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x375>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
JMP 4222c3 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x393>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM2,%XMM2,%XMM2	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
SAL $0x4,%RAX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
ADD %RAX,%RDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV -0x60(%RBP),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
VADDPD (%RAX),%XMM0,%XMM0	1	0	0.50	0.33	0.33	0	0.50	0	0	0	0	0	0.33	3	0.50	vect (25.0%)
VMOVUPD %XMM0,(%RAX)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
VADDSD 0x10(%RAX),%XMM2,%XMM0	1	0	0.50	0.33	0.33	0	0.50	0	0	0	0	0	0.33	3	0.50	scal (12.5%)
VMOVSD %XMM0,0x10(%RAX)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD 0xb0(%R10),%XMM0	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
ADD $0x38,%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
VZEROUPPER	2	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
JMP 4df280 <exp>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPL (%RAX,%RAX,1)	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 0.01% of application time for run orig_0

Source file and lines	TwoBodyJastrowRef.h:155-306
Module	exec

The function is defined in:

/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:94
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:273
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_numeric.h:140-141
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/shared_ptr_base.h:1296
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_vector.h:1169
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algobase.h:923-924,1115
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_iterator.h:1028,1182
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:155-157,262,268-274,299-306
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223,229,249
/home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.h:313-316

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Code clean check

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 16% of vector register length is used (average across all SSE/AVX instructions).

Details

36% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

8% of SSE/AVX loads are used in vector version.
7% of SSE/AVX stores are used in vector version.
41% of SSE/AVX addition or subtraction instructions are used in vector version.
55% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

Workaround

Read the "512-bits vectorization" report at "Potential" confidence level.

Execution units bottlenecks

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

512-bits vectorization

Workaround

Recompile with -mprefer-vector-width=512

CALL instructions

Detected function call instructions.

Details

_intel_fast_memset: 3 occurrences►
- /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algobase.h:924
- /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algobase.h:924
- /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algobase.h:924
qmcplusplus::ParticleSet::getDistTableAA(int) const: 2 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:302
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:305

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.

VZEROUPPER: 2 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:305
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:306

Vector unaligned load/store instructions

Detected 4 suboptimal vector unaligned load/store instructions.

Details

VEXTRACTF128: 4 occurrences►
- /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_numeric.h:141
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:156
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:156
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:156

Workaround

Pass to your compiler a micro-architecture specialization option:
- use march=native
Use vector aligned instructions:
1. align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
2. inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.

Conversion instructions

Detected expensive conversion instructions.

Details

CLTQ: 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-773-5521/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:269

Workaround

Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.

Type of elements and instruction set

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

The binary function is composed of 16 FP arithmetical operations:

16: addition or subtraction

The binary function is loading 236 bytes (29 double precision FP elements). The binary function is storing 112 bytes (14 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.05 FP operations per loaded or stored byte.

General properties

nb instructions	178
nb uops	185
loop length	719
used x86 registers	16
used mmx registers	0
used xmm registers	3
used ymm registers	2
used zmm registers	0
nb stack references	7

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	30.83 cycles
front end	30.83 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	11.10	11.00	12.00	12.00	12.50	11.00	10.90	12.50	12.50	12.50	11.00	12.00
cycles	11.10	11.00	12.00	12.00	12.50	11.00	10.90	12.50	12.50	12.50	11.00	12.00

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	30.83
Dispatch	12.50
Overall L1	30.83

Vectorization ratios

INT

all	5%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
other	11%

all	68%
load	25%
store	25%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	45%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	100%

INT+FP

all	36%
load	8%
store	7%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	41%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	55%

Vector efficiency ratios

INT

all	11%
load	12%
store	11%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
other	12%

all	21%
load	15%
store	15%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	18%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	25%

INT+FP

all	16%
load	13%
store	12%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	17%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	18%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 30.83 cycles. At this rate:

3% of peak load performance is reached (7.65 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (3.63 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: 421f30

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
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
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
SUB $0x38,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RCX,-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 %EDX,%R14D	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV %RSI,%R13	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RDI,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOVL $0x2,0xc(%RDI)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOV 0x248(%RDI),%ESI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %R13,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
CALL 460b50 <_ZNK11qmcplusplus11ParticleSet14getDistTableAAEi>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV %RBX,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x90(%RAX),%R8	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x90(%RBX),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x140(%R10),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x158(%R10),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x170(%R10),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
TEST %EBX,%EBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (6.3%)
MOV %R10,-0x48(%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 %R11,-0x40(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
JLE 421fe8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0xb8>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %RBX,%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $0x7fffffff,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 421fe8 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0xb8>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
SAL $0x3,%RAX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
MOV %R15,-0x30(%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 %R12,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RAX,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RAX,%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %R8,-0x38(%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 4efda0 <_intel_fast_memset>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV -0x40(%RBP),%RDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R15,%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
CALL 4efda0 <_intel_fast_memset>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV -0x30(%RBP),%RDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R15,%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
CALL 4efda0 <_intel_fast_memset>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV -0x30(%RBP),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x40(%RBP),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x38(%RBP),%R8	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x48(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOVSXD %R14D,%RDX	1	0	0.33	0	0	0	0.33	0	0	0	0	0.33	0	1	0.33	N/A
MOV 0xa0(%R10),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
TEST %RAX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (12.5%)
JE 42209e <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x16e>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV 0x18(%R13),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %RDX,-0x58(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
IMUL (%RCX,%RDX,4),%EAX	1	0	1	0.33	0.33	0	0	0	0	0	0	0	0.33	3	1	N/A
CLTQ	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.33	scal (12.5%)
SAL $0x3,%RAX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
MOV %RAX,-0x30(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
XOR %EAX,%EAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R12,-0x38(%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 %R14D,-0x4c(%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 %R15,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
XCHG %AX,%AX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV 0x90(%R10),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x140(%R10),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x58(%RBP),%RDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
LEA (,%RBX,8),%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM0,%XMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
TEST %RAX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (12.5%)
JE 422114 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x1e4>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV $0x1fffffffffffffff,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
ADD %RAX,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
AND %RAX,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
INC %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV $0x3ffffffffffffffc,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
AND %RBX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 422104 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x1d4>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
VXORPD %XMM0,%XMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
XOR %ECX,%ECX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
NOPW (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VEXTRACTF128 $0x1,%YMM0,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
VADDPD %XMM1,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
VSHUFPD $0x1,%XMM0,%XMM0,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
VADDSD %XMM1,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
JMP 42210f <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x1df>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
XOR %EAX,%EAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VMOVSD %XMM0,0x138(%R10)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV 0xd8(%R10),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x158(%R10),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
VMOVSD (%RAX,%RDX,8),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VSUBSD %XMM0,%XMM1,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
VMOVSD %XMM0,0xb0(%R10)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV 0x248(%R10),%ESI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %R13,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VZEROUPPER	2	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 460b50 <_ZNK11qmcplusplus11ParticleSet14getDistTableAAEi>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV -0x48(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x90(%R10),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
TEST %RCX,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (12.5%)
JE 4221aa <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x27a>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV 0xb8(%RAX),%RDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %RCX,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $-0x4,%RSI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 4221b7 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x287>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
LEA -0x1(%RSI),%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM0,%XMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
NOPW (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VEXTRACTF128 $0x1,%YMM0,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
VADDPD %XMM1,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
VSHUFPD $0x1,%XMM0,%XMM0,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
VADDSD %XMM1,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
JMP 4221cb <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x29b>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
VXORPD %XMM0,%XMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
VXORPD %XMM2,%XMM2,%XMM2	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
JMP 4222c3 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x393>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM0,%XMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
MOV 0xa8(%RAX),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %RCX,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $-0x4,%RSI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 42221f <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x2ef>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
LEA -0x1(%RSI),%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA (%RDX,%RAX,8),%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM1,%XMM1,%XMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
XOR %R9D,%R9D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
NOP	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VEXTRACTF128 $0x1,%YMM1,%XMM2	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
VADDPD %XMM2,%XMM1,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
VSHUFPD $0x1,%XMM1,%XMM1,%XMM2	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
VADDSD %XMM2,%XMM1,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
CMP %RSI,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JNE 422225 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x2f5>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
JMP 422243 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x313>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM1,%XMM1,%XMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
LEA (%RDX,%RAX,8),%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VPUNPCKLQDQ %XMM1,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	scal (12.5%)
MOV %RCX,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $-0x4,%RSI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 42229f <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x36f>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
LEA -0x1(%RSI),%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RAX,%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
SAL $0x4,%R8	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	scal (12.5%)
ADD %RDX,%R8	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
VXORPD %XMM1,%XMM1,%XMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
XOR %R9D,%R9D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
NOPW %CS:(%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VEXTRACTF128 $0x1,%YMM1,%XMM2	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	vect (25.0%)
VADDPD %XMM2,%XMM1,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
VSHUFPD $0x1,%XMM1,%XMM1,%XMM2	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
VADDSD %XMM2,%XMM1,%XMM2	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
CMP %RSI,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JNE 4222a5 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x375>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
JMP 4222c3 <_ZN16miniqmcreference17TwoBodyJastrowRefIN11qmcplusplus14BsplineFunctorIdEEE9ratioGradERNS1_11ParticleSetEiRNS1_10TinyVectorIdLj3EEE+0x393>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
XOR %ESI,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VXORPD %XMM2,%XMM2,%XMM2	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	vect (25.0%)
SAL $0x4,%RAX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
ADD %RAX,%RDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV -0x60(%RBP),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
VADDPD (%RAX),%XMM0,%XMM0	1	0	0.50	0.33	0.33	0	0.50	0	0	0	0	0	0.33	3	0.50	vect (25.0%)
VMOVUPD %XMM0,(%RAX)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
VADDSD 0x10(%RAX),%XMM2,%XMM0	1	0	0.50	0.33	0.33	0	0.50	0	0	0	0	0	0.33	3	0.50	scal (12.5%)
VMOVSD %XMM0,0x10(%RAX)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD 0xb0(%R10),%XMM0	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
ADD $0x38,%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
VZEROUPPER	2	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
JMP 4df280 <exp>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A

Name	Coverage (%)	Time (s)
▼miniqmcreference::TwoBodyJastrowRef >::ratioGrad(qmcplusplus::ParticleSet&, int, qmcplusplus::TinyVector&)–	0.38	0.33
○Loop 367 - TwoBodyJastrowRef.h:155-156 - exec	0.11	0.10
○Loop 365 - TwoBodyJastrowRef.h:155-156 - exec	0.11	0.09
○Loop 366 - TwoBodyJastrowRef.h:155-156 - exec	0.11	0.09
○Loop 369 - stl_numeric.h:140-141 - exec	0.04	0.03
○Loop 370 - TwoBodyJastrowRef.h:269-274 - exec	0.00	0.01
○Loop 362 - TwoBodyJastrowRef.h:155-156 - exec	0.00	0.00
○Loop 364 - TwoBodyJastrowRef.h:155-156 - exec	0.00	0.00
○Loop 363 - TwoBodyJastrowRef.h:155-156 - exec	0.00	0.00
○Loop 368 - stl_numeric.h:140-141 - exec	0.00	0.00

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

512-bits vectorization

Workaround

CALL instructions

Details

Complex instructions

Details

Vector unaligned load/store instructions

Details

Workaround

Conversion instructions

Details

Workaround

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

512-bits vectorization

Workaround

CALL instructions

Details

Complex instructions

Details

Vector unaligned load/store instructions

Details

Workaround

Conversion instructions

Details

Workaround

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