OV - exec - Loop 773

	Loop Id: 773	Module: exec	Source: For.hpp:137-137 [...]	Coverage: 0.05%

0x259130 MOV	-0xd8(%RBP),%RCX

0x259137 INC	%RCX

0x25913a CMP	-0x38(%RBP),%RCX

0x25913e MOV	-0xd0(%RBP),%RDX

0x259145 JE	259090

0x25914b MOV	-0x58(%RBP),%RAX

0x25914f MOV	%RCX,-0xd8(%RBP)

0x259156 LEA	(%RCX,%RAX,1),%R10

0x25915a LEA	(%RDX,%R10,1),%RAX

0x25915e IMUL	-0xc0(%RBP),%R10

0x259166 ADD	-0xc8(%RBP),%R10

0x25916d XOR	%ECX,%ECX

0x25916f JMP	2591aa

  (774) 0x259180 XORPD	%XMM0,%XMM0

  (774) 0x259184 LEA	(%R10,%RDI,1),%RDX

  (774) 0x259188 ADD	%R8,%RDI

  (774) 0x25918b MOV	-0x100(%RBP),%RSI

  (774) 0x259192 MULSD	(%RSI,%RDX,8),%XMM0

  (774) 0x259197 ADDSD	(%RBX,%RDI,8),%XMM0

  (774) 0x25919c MOVSD	%XMM0,(%RBX,%RDI,8)

  (774) 0x2591a1 INC	%RCX

  (774) 0x2591a4 CMP	-0x40(%RBP),%RCX

  (774) 0x2591a8 JE	259130

  (774) 0x2591aa MOV	-0x60(%RBP),%RDX

  (774) 0x2591ae LEA	(%RCX,%RDX,1),%RDI

  (774) 0x2591b2 MOV	-0x108(%RBP),%RDX

  (774) 0x2591b9 MOVSXD	(%RDX,%RDI,4),%R9

  (774) 0x2591bd TEST	%R9,%R9

  (774) 0x2591c0 JLE	259180

  (774) 0x2591c2 MOV	-0xe8(%RBP),%RDX

  (774) 0x2591c9 MOV	(%RDX,%RDI,8),%RDX

  (774) 0x2591cd ADD	%RDX,%R9

  (774) 0x2591d0 MOV	-0xf0(%RBP),%RSI

  (774) 0x2591d7 MOV	-0xf8(%RBP),%R11

  (774) 0x2591de MOV	(%RSI,%R11,8),%RSI

  (774) 0x2591e2 IMUL	-0xe0(%RBP),%RSI

  (774) 0x2591ea ADD	%RAX,%RSI

  (774) 0x2591ed XORPD	%XMM0,%XMM0

  (774) 0x2591f1 NOPW	%CS:(%RAX,%RAX,1)

    (775) 0x259200 MOV	(%R12,%RDX,8),%R11

    (775) 0x259204 IMUL	%R13,%R11

    (775) 0x259208 ADD	%RSI,%R11

    (775) 0x25920b MOVSD	(%R14,%R11,8),%XMM1

    (775) 0x259211 MULSD	(%R15,%RDX,8),%XMM1

    (775) 0x259217 ADDSD	%XMM1,%XMM0

    (775) 0x25921b INC	%RDX

    (775) 0x25921e CMP	%R9,%RDX

    (775) 0x259221 JL	259200

  (774) 0x259223 JMP	259184

/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/index/IndexValue.hpp: 87 - 221

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

87:     value++;

[...]

221:     return (value < x.value);

/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/pattern/kernel/For.hpp: 137 - 137

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

137:     for (decltype(distance_it) i = 0; i < distance_it; ++i) {

/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/src/Kripke/Kernel/Scattering.cpp: 88 - 97

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

88:             MixElem mix_stop = mix_start + zone_to_num_mixelem(z);

89:

90:             double sigs_z = 0.0;

91:             for(MixElem mix = mix_start;mix < mix_stop;++ mix){

92:               Material mat = mixelem_to_material(mix);

93:               double fraction = mixelem_to_fraction(mix);

94:

95:               sigs_z += sigs(mat, n, global_g, global_gp) * fraction;

96:             }

97:             phi_out(nm, g, z) += sigs_z * phi(nm, gp, z);

/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Layout.hpp: 190 - 190

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

190:          strides[RangeInts]*indices // it's not stride one

/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/internal/Iterators.hpp: 269 - 269

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

269:     return value_type(val + rhs);

/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/policy/openmp/kernel/Collapse.hpp: 76 - 76

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

76:     for (i0 = 0; i0 < l0; ++i0) {

/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Operators.hpp: 334 - 334

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

334:     return Ret{lhs} + rhs;

Coverage (%)	Name	Source Location	Module

min	med	avg	max

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

min	med	avg	max

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

Path /

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	8.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.08
Bottlenecks	P6, P7, P8, P9,
Function	void RAJA::internal::StatementExecutor, RAJA::statement::For<2l, RAJA::policy::sequential::seq_exec, RAJA::statement::For<3l, RAJA::policy::sequential::seq_exec, RAJA::statement::Lambda<0l> > > >, RAJA::internal::LoopTypes, camp::list > >::exec, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, ScatteringSdom::operator() >(Kripke::ArchLayoutT, Kripke::SdomId, Kripke::SdomId, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&) const::{lambda(Kripke::Moment, Kripke::Group, Kripke::Group, Kripke::Zone)#1}>&>(RAJA::internal::LoopData, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, ScatteringSdom::operator() >(Kripke::ArchLayoutT, Kripke::SdomId, Kripke::SdomId, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&) const::{lambda(Kripke::Moment, Kripke::Group, Kripke::Group, Kripke::Zone)#1}>&) [clone .omp_outlined]
Source	For.hpp:137-137,Collapse.hpp:76-76
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	1.75
CQA cycles if no scalar integer	1.75
CQA cycles if FP arith vectorized	1.75
CQA cycles if fully vectorized	0.22
Front-end cycles	1.63
P0 cycles	1.33
P1 cycles	1.33
P2 cycles	1.33
P3 cycles	1.33
P4 cycles	1.33
P5 cycles	1.33
P6 cycles	1.75
P7 cycles	1.75
P8 cycles	1.75
P9 cycles	1.75
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
P15 cycles	0.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	0
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	13.00
Nb uops	13.00
Nb loads	6.00
Nb stores	1.00
Nb stack references	6.00
FLOP/cycle	0.00
Nb FLOP add-sub	0.00
Nb FLOP mul	0.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	32.00
Bytes prefetched	0.00
Bytes loaded	48.00
Bytes stored	8.00
Stride 0	0.00
Stride 1	0.00
Stride n	0.00
Stride unknown	3.00
Stride indirect	0.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	NA
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	12.50
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	12.50
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	NA
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	12.50

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	8.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.08
Bottlenecks	P6, P7, P8, P9,
Function	void RAJA::internal::StatementExecutor, RAJA::statement::For<2l, RAJA::policy::sequential::seq_exec, RAJA::statement::For<3l, RAJA::policy::sequential::seq_exec, RAJA::statement::Lambda<0l> > > >, RAJA::internal::LoopTypes, camp::list > >::exec, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, ScatteringSdom::operator() >(Kripke::ArchLayoutT, Kripke::SdomId, Kripke::SdomId, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&) const::{lambda(Kripke::Moment, Kripke::Group, Kripke::Group, Kripke::Zone)#1}>&>(RAJA::internal::LoopData, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, ScatteringSdom::operator() >(Kripke::ArchLayoutT, Kripke::SdomId, Kripke::SdomId, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&) const::{lambda(Kripke::Moment, Kripke::Group, Kripke::Group, Kripke::Zone)#1}>&) [clone .omp_outlined]
Source	For.hpp:137-137,Collapse.hpp:76-76
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	1.75
CQA cycles if no scalar integer	1.75
CQA cycles if FP arith vectorized	1.75
CQA cycles if fully vectorized	0.22
Front-end cycles	1.63
P0 cycles	1.33
P1 cycles	1.33
P2 cycles	1.33
P3 cycles	1.33
P4 cycles	1.33
P5 cycles	1.33
P6 cycles	1.75
P7 cycles	1.75
P8 cycles	1.75
P9 cycles	1.75
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
P15 cycles	0.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	0
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	13.00
Nb uops	13.00
Nb loads	6.00
Nb stores	1.00
Nb stack references	6.00
FLOP/cycle	0.00
Nb FLOP add-sub	0.00
Nb FLOP mul	0.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	32.00
Bytes prefetched	0.00
Bytes loaded	48.00
Bytes stored	8.00
Stride 0	0.00
Stride 1	0.00
Stride n	0.00
Stride unknown	3.00
Stride indirect	0.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	NA
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	12.50
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	12.50
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	NA
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	12.50

Path /

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

Function	void RAJA::internal::StatementExecutor, RAJA::statement::For<2l, RAJA::policy::sequential::seq_exec, RAJA::statement::For<3l, RAJA::policy::sequential::seq_exec, RAJA::statement::Lambda<0l> > > >, RAJA::internal::LoopTypes, camp::list > >::exec, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, ScatteringSdom::operator() >(Kripke::ArchLayoutT, Kripke::SdomId, Kripke::SdomId, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&) const::{lambda(Kripke::Moment, Kripke::Group, Kripke::Group, Kripke::Zone)#1}>&>(RAJA::internal::LoopData, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, ScatteringSdom::operator() >(Kripke::ArchLayoutT, Kripke::SdomId, Kripke::SdomId, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&) const::{lambda(Kripke::Moment, Kripke::Group, Kripke::Group, Kripke::Zone)#1}>&) [clone .omp_outlined]
Source file and lines	For.hpp:137-137
Module	exec

The loop is defined in:

/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/index/IndexValue.hpp:87,221
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/pattern/kernel/For.hpp:137
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/src/Kripke/Kernel/Scattering.cpp:88-97
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Layout.hpp:190
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/internal/Iterators.hpp:269
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/policy/openmp/kernel/Collapse.hpp:76
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Operators.hpp:334

Warnings:
Non-innermost loop: analyzing only self part (ignoring child loops).
The structure of this loop is probably <if then [else] end>.

The presence of multiple execution paths is typically the main/first bottleneck.
Try to simplify control inside loop: ideally, try to remove all conditional expressions, for example by (if applicable):

hoisting them (moving them outside the loop)
turning them into conditional moves, MIN or MAX

Ex: if (x<0) x=0 => x = (x<0 ? 0 : x) (or MAX(0,x) after defining the corresponding macro)

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. 8 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 1.75 to 0.22 cycles (8.00x speedup).

Details

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

Workaround

Try another compiler or update/tune your current one:
- recompile with ffast-math (included in Ofast) to extend loop vectorization to FP reductions.
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop 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

Performance is limited by:

reading data from caches/RAM (load units are a bottleneck)
writing data to caches/RAM (the store unit is a bottleneck)

By removing all these bottlenecks, you can lower the cost of an iteration from 1.75 to 1.63 cycles (1.08x speedup).

Workaround

Read less array elements
Write less array elements
Provide more information to your compiler:
- hardcode the bounds of the corresponding 'for' loop

No data for this section

Slow data structures access

Detected data structures (typically arrays) that cannot be efficiently read/written

Details

Constant unknown stride: 3 occurrence(s)

Non-unit stride (uncontiguous) accesses are not efficiently using data caches

Workaround

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

Type of elements and instruction set

No instructions are processing arithmetic or math operations on FP elements. This loop is probably writing/copying data or processing integer elements.

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

The binary loop does not contain any FP arithmetical operations. The binary loop is loading 48 bytes. The binary loop is storing 8 bytes.

General properties

nb instructions	13
nb uops	13
loop length	65
used x86 registers	5
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	6

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	1.63 cycles
front end	1.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	1.33	1.33	1.33	1.33	1.33	1.33	1.75	1.75	1.75	1.75	0.00	0.00	0.00	0.00	0.00	0.00
cycles	1.33	1.33	1.33	1.33	1.33	1.33	1.75	1.75	1.75	1.75	0.00	0.00	0.00	0.00	0.00	0.00

Execution ports to units layout:

P0: ALU
P1: ALU
P2: ALU
P3: ALU, BRU
P4: ALU, BRU
P5: ALU, BRU
P6 (512 bits): store address, load
P7 (512 bits): store address, load
P8 (512 bits): store address, load
P9 (512 bits): store address, load
P10 (512 bits): VPU, DIV/SQRT
P11 (512 bits): VPU, DIV/SQRT
P12 (512 bits): VPU
P13 (512 bits): VPU
P14 (512 bits): FP store data
P15 (512 bits): FP store data

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	0.00

Cycles summary

Front-end	1.63
Dispatch	1.75
Data deps.	0.00
Overall L1	1.75

Vectorization ratios

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)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

Vector efficiency ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 1.75 cycles. At this rate:

10% of peak load performance is reached (27.43 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
3% of peak store performance is reached (4.57 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the loop is: 25914b

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
MOV -0xd8(%RBP),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
INC %RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP -0x38(%RBP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
MOV -0xd0(%RBP),%RDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
JE 259090 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl1EEEEJNS2_3ForILl2ENS_6policy10sequential8seq_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEENS0_9LoopTypesINS5_4listIJvvvvEEESJ_EEE4execIRNS0_8LoopDataINS5_5tupleIJNS_4SpanINS_9Iterators16numeric_iteratorIN6Kripke6MomentElPST_EElEENSP_INSR_INSS_5GroupElPSX_EElEES10_NSP_INSR_INSS_4ZoneElPS11_EElEEEEENSO_IJEEENS5_9resources2v14HostEJZNK14ScatteringSdomclINSS_11ArchLayoutTINSS_12ArchT_OpenMPENSS_11LayoutT_DGZEEEEEvT_NSS_6SdomIdES1H_RKNSS_4Core3SetES1L_S1L_RNS1I_5FieldIdJST_SX_S11_EEES1O_RNS1M_IdJNSS_8MaterialENSS_8LegendreENSS_11GlobalGroupES1R_EEERNS1M_INSS_7MixElemEJS11_EEERNS1M_IiJS11_EEERNS1M_IS1P_JS1U_EEERNS1M_IdJS1U_EEERNS1M_IS1Q_JST_EEEEUlST_SX_SX_S11_E_EEEEEvOS1G_.omp_outlined+0x200>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV -0x58(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %RCX,-0xd8(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
LEA (%RCX,%RAX,1),%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA (%RDX,%R10,1),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
IMUL -0xc0(%RBP),%R10	1	0.33	0.33	0.33	0	0	0	0.25	0.25	0.25	0.25	3	0.33	N/A
ADD -0xc8(%RBP),%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
XOR %ECX,%ECX	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
JMP 2591aa <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl1EEEEJNS2_3ForILl2ENS_6policy10sequential8seq_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEENS0_9LoopTypesINS5_4listIJvvvvEEESJ_EEE4execIRNS0_8LoopDataINS5_5tupleIJNS_4SpanINS_9Iterators16numeric_iteratorIN6Kripke6MomentElPST_EElEENSP_INSR_INSS_5GroupElPSX_EElEES10_NSP_INSR_INSS_4ZoneElPS11_EElEEEEENSO_IJEEENS5_9resources2v14HostEJZNK14ScatteringSdomclINSS_11ArchLayoutTINSS_12ArchT_OpenMPENSS_11LayoutT_DGZEEEEEvT_NSS_6SdomIdES1H_RKNSS_4Core3SetES1L_S1L_RNS1I_5FieldIdJST_SX_S11_EEES1O_RNS1M_IdJNSS_8MaterialENSS_8LegendreENSS_11GlobalGroupES1R_EEERNS1M_INSS_7MixElemEJS11_EEERNS1M_IiJS11_EEERNS1M_IS1P_JS1U_EEERNS1M_IdJS1U_EEERNS1M_IS1Q_JST_EEEEUlST_SX_SX_S11_E_EEEEEvOS1G_.omp_outlined+0x31a>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A

Function	void RAJA::internal::StatementExecutor, RAJA::statement::For<2l, RAJA::policy::sequential::seq_exec, RAJA::statement::For<3l, RAJA::policy::sequential::seq_exec, RAJA::statement::Lambda<0l> > > >, RAJA::internal::LoopTypes, camp::list > >::exec, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, ScatteringSdom::operator() >(Kripke::ArchLayoutT, Kripke::SdomId, Kripke::SdomId, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&) const::{lambda(Kripke::Moment, Kripke::Group, Kripke::Group, Kripke::Zone)#1}>&>(RAJA::internal::LoopData, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, ScatteringSdom::operator() >(Kripke::ArchLayoutT, Kripke::SdomId, Kripke::SdomId, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Set const&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&, Kripke::Core::Field&) const::{lambda(Kripke::Moment, Kripke::Group, Kripke::Group, Kripke::Zone)#1}>&) [clone .omp_outlined]
Source file and lines	For.hpp:137-137
Module	exec

The loop is defined in:

/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/index/IndexValue.hpp:87,221
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/pattern/kernel/For.hpp:137
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/src/Kripke/Kernel/Scattering.cpp:88-97
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Layout.hpp:190
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/internal/Iterators.hpp:269
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/policy/openmp/kernel/Collapse.hpp:76
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Operators.hpp:334

hoisting them (moving them outside the loop)
turning them into conditional moves, MIN or MAX

Ex: if (x<0) x=0 => x = (x<0 ? 0 : x) (or MAX(0,x) after defining the corresponding macro)

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. 8 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 1.75 to 0.22 cycles (8.00x speedup).

Details

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

Workaround

Try another compiler or update/tune your current one:
- recompile with ffast-math (included in Ofast) to extend loop vectorization to FP reductions.
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop 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

Performance is limited by:

reading data from caches/RAM (load units are a bottleneck)
writing data to caches/RAM (the store unit is a bottleneck)

By removing all these bottlenecks, you can lower the cost of an iteration from 1.75 to 1.63 cycles (1.08x speedup).

Workaround

Read less array elements
Write less array elements
Provide more information to your compiler:
- hardcode the bounds of the corresponding 'for' loop

No data for this section

Slow data structures access

Detected data structures (typically arrays) that cannot be efficiently read/written

Details

Constant unknown stride: 3 occurrence(s)

Non-unit stride (uncontiguous) accesses are not efficiently using data caches

Workaround

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

Type of elements and instruction set

No instructions are processing arithmetic or math operations on FP elements. This loop is probably writing/copying data or processing integer elements.

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

The binary loop does not contain any FP arithmetical operations. The binary loop is loading 48 bytes. The binary loop is storing 8 bytes.

General properties

nb instructions	13
nb uops	13
loop length	65
used x86 registers	5
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	6

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	1.63 cycles
front end	1.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	1.33	1.33	1.33	1.33	1.33	1.33	1.75	1.75	1.75	1.75	0.00	0.00	0.00	0.00	0.00	0.00
cycles	1.33	1.33	1.33	1.33	1.33	1.33	1.75	1.75	1.75	1.75	0.00	0.00	0.00	0.00	0.00	0.00

Execution ports to units layout:

P0: ALU
P1: ALU
P2: ALU
P3: ALU, BRU
P4: ALU, BRU
P5: ALU, BRU
P6 (512 bits): store address, load
P7 (512 bits): store address, load
P8 (512 bits): store address, load
P9 (512 bits): store address, load
P10 (512 bits): VPU, DIV/SQRT
P11 (512 bits): VPU, DIV/SQRT
P12 (512 bits): VPU
P13 (512 bits): VPU
P14 (512 bits): FP store data
P15 (512 bits): FP store data

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	0.00

Cycles summary

Front-end	1.63
Dispatch	1.75
Data deps.	0.00
Overall L1	1.75

Vectorization ratios

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)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

Vector efficiency ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 1.75 cycles. At this rate:

10% of peak load performance is reached (27.43 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
3% of peak store performance is reached (4.57 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the loop is: 25914b

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
MOV -0xd8(%RBP),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
INC %RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP -0x38(%RBP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
MOV -0xd0(%RBP),%RDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
JE 259090 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl1EEEEJNS2_3ForILl2ENS_6policy10sequential8seq_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEENS0_9LoopTypesINS5_4listIJvvvvEEESJ_EEE4execIRNS0_8LoopDataINS5_5tupleIJNS_4SpanINS_9Iterators16numeric_iteratorIN6Kripke6MomentElPST_EElEENSP_INSR_INSS_5GroupElPSX_EElEES10_NSP_INSR_INSS_4ZoneElPS11_EElEEEEENSO_IJEEENS5_9resources2v14HostEJZNK14ScatteringSdomclINSS_11ArchLayoutTINSS_12ArchT_OpenMPENSS_11LayoutT_DGZEEEEEvT_NSS_6SdomIdES1H_RKNSS_4Core3SetES1L_S1L_RNS1I_5FieldIdJST_SX_S11_EEES1O_RNS1M_IdJNSS_8MaterialENSS_8LegendreENSS_11GlobalGroupES1R_EEERNS1M_INSS_7MixElemEJS11_EEERNS1M_IiJS11_EEERNS1M_IS1P_JS1U_EEERNS1M_IdJS1U_EEERNS1M_IS1Q_JST_EEEEUlST_SX_SX_S11_E_EEEEEvOS1G_.omp_outlined+0x200>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV -0x58(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %RCX,-0xd8(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
LEA (%RCX,%RAX,1),%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA (%RDX,%R10,1),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
IMUL -0xc0(%RBP),%R10	1	0.33	0.33	0.33	0	0	0	0.25	0.25	0.25	0.25	3	0.33	N/A
ADD -0xc8(%RBP),%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
XOR %ECX,%ECX	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
JMP 2591aa <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl1EEEEJNS2_3ForILl2ENS_6policy10sequential8seq_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEENS0_9LoopTypesINS5_4listIJvvvvEEESJ_EEE4execIRNS0_8LoopDataINS5_5tupleIJNS_4SpanINS_9Iterators16numeric_iteratorIN6Kripke6MomentElPST_EElEENSP_INSR_INSS_5GroupElPSX_EElEES10_NSP_INSR_INSS_4ZoneElPS11_EElEEEEENSO_IJEEENS5_9resources2v14HostEJZNK14ScatteringSdomclINSS_11ArchLayoutTINSS_12ArchT_OpenMPENSS_11LayoutT_DGZEEEEEvT_NSS_6SdomIdES1H_RKNSS_4Core3SetES1L_S1L_RNS1I_5FieldIdJST_SX_S11_EEES1O_RNS1M_IdJNSS_8MaterialENSS_8LegendreENSS_11GlobalGroupES1R_EEERNS1M_INSS_7MixElemEJS11_EEERNS1M_IiJS11_EEERNS1M_IS1P_JS1U_EEERNS1M_IdJS1U_EEERNS1M_IS1Q_JST_EEEEUlST_SX_SX_S11_E_EEEEEvOS1G_.omp_outlined+0x31a>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A

Report Configuration

Vectorization

Details

Workaround

Execution units bottlenecks

Workaround

Slow data structures access

Details

Workaround

Type of elements and instruction set

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

Vectorization

Details

Workaround

Execution units bottlenecks

Workaround

Slow data structures access

Details

Workaround

Type of elements and instruction set

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code