OV - exec - Loop 1114

	Loop Id: 1114	Module: exec	Source: For.hpp:137-137 [...]	Coverage: 0.01%

0x26f620 MOV	-0x50(%RBP),%RAX

0x26f624 MOV	-0x60(%RBP),%R10

0x26f628 ADD	-0x1d8(%RBP),%R13

0x26f62f INC	%R10

0x26f632 ADD	%RAX,%R9

0x26f635 ADD	%RAX,%RDX

0x26f638 ADD	%RAX,%R8

0x26f63b MOV	%R10,%RAX

0x26f63e MOV	%R10,-0x60(%RBP)

0x26f642 CMP	%R10,-0x1e0(%RBP)

0x26f649 JE	26f550

0x26f64f CMPQ	$0,-0x1e8(%RBP)

0x26f657 MOV	-0x58(%RBP),%RBX

0x26f65b JLE	26f620

0x26f65d MOV	-0x60(%RBP),%RCX

0x26f661 MOV	-0x1d0(%RBP),%RAX

0x26f668 XOR	%R11D,%R11D

0x26f66b XOR	%R14D,%R14D

0x26f66e IMUL	-0x138(%RBP),%RCX

0x26f676 ADD	-0x140(%RBP),%RCX

0x26f67d ADD	%RCX,%RAX

0x26f680 MOV	%RCX,-0x78(%RBP)

0x26f684 MOV	%RAX,-0x70(%RBP)

0x26f688 MOV	-0x1c8(%RBP),%RAX

0x26f68f NOP

  (1115) 0x26f690 MOV	%RSI,%R15

  (1115) 0x26f693 MOV	%R9,%RSI

  (1115) 0x26f696 MOV	%RDX,%R9

  (1115) 0x26f699 MOV	%RSI,%R9

  (1115) 0x26f69c MOV	%R15,%RSI

  (1115) 0x26f69f MOV	-0x68(%RBP),%R10

  (1115) 0x26f6a3 MOV	-0x218(%RBP),%R15

  (1115) 0x26f6aa MOV	%R12,%RDI

  (1115) 0x26f6ad MOV	%RBX,%R12

  (1115) 0x26f6b0 MOV	-0x220(%RBP),%RBX

  (1115) 0x26f6b7 MOV	-0x78(%RBP),%RCX

  (1115) 0x26f6bb VMOVSD	(%R15,%R10,8),%XMM1

  (1115) 0x26f6c1 MOV	-0x210(%RBP),%R15

  (1115) 0x26f6c8 VMOVSD	(%RBX,%R10,8),%XMM0

  (1115) 0x26f6ce MOV	%R12,%RBX

  (1115) 0x26f6d1 MOV	-0x1f0(%RBP),%R12

  (1115) 0x26f6d8 VADDSD	%XMM1,%XMM1,%XMM1

  (1115) 0x26f6dc VDIVSD	(%R15,%RCX,8),%XMM1,%XMM1

  (1115) 0x26f6e2 MOV	-0x208(%RBP),%RCX

  (1115) 0x26f6e9 VADDSD	%XMM0,%XMM0,%XMM0

  (1115) 0x26f6ed VDIVSD	(%R12,%R14,1),%XMM0,%XMM0

  (1115) 0x26f6f3 MOV	-0x200(%RBP),%R12

  (1115) 0x26f6fa VMOVSD	(%RCX,%R10,8),%XMM2

  (1115) 0x26f700 MOV	-0x70(%RBP),%RCX

  (1115) 0x26f704 MOV	-0x30(%RBP),%R10

  (1115) 0x26f708 VMOVSD	(%RDI,%RCX,8),%XMM3

  (1115) 0x26f70d VADDSD	%XMM2,%XMM2,%XMM2

  (1115) 0x26f711 VDIVSD	(%R12,%RSI,8),%XMM2,%XMM2

  (1115) 0x26f717 VFMADD213SD	(%R8,%R11,1),%XMM0,%XMM3

  (1115) 0x26f71d MOV	%RDI,%R12

  (1115) 0x26f720 VFMADD231SD	(%R10,%R14,1),%XMM1,%XMM3

  (1115) 0x26f726 VADDSD	%XMM0,%XMM1,%XMM0

  (1115) 0x26f72a VFMADD231SD	(%R13,%R14,1),%XMM2,%XMM3

  (1115) 0x26f731 VADDSD	%XMM2,%XMM0,%XMM0

  (1115) 0x26f735 VADDSD	(%RDX,%R11,1),%XMM0,%XMM0

  (1115) 0x26f73b VDIVSD	%XMM0,%XMM3,%XMM0

  (1115) 0x26f73f VMOVSD	%XMM0,(%R9,%R11,1)

  (1115) 0x26f745 VADDSD	%XMM0,%XMM0,%XMM0

  (1115) 0x26f749 ADD	%RBX,%R11

  (1115) 0x26f74c VSUBSD	(%RDI,%RCX,8),%XMM0,%XMM1

  (1115) 0x26f751 VMOVSD	%XMM1,(%RDI,%RCX,8)

  (1115) 0x26f756 MOV	-0x1f8(%RBP),%RDI

  (1115) 0x26f75d VSUBSD	(%R10,%R14,1),%XMM0,%XMM1

  (1115) 0x26f763 VMOVSD	%XMM1,(%R10,%R14,1)

  (1115) 0x26f769 VSUBSD	(%R13,%R14,1),%XMM0,%XMM0

  (1115) 0x26f770 VMOVSD	%XMM0,(%R13,%R14,1)

  (1115) 0x26f777 ADD	%RDI,%R14

  (1115) 0x26f77a DEC	%RAX

  (1115) 0x26f77d JNE	26f690

0x26f783 JMP	26f620

/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/SweepSubdomain.cpp: 88 - 106

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

88:             double xcos_dxi = 2.0 * xcos(d) / dx(i);

89:             double ycos_dyj = 2.0 * ycos(d) / dy(j);

90:             double zcos_dzk = 2.0 * zcos(d) / dz(k);

[...]

96:                 + psi_lf(d, g, j, k) * xcos_dxi

97:                 + psi_fr(d, g, i, k) * ycos_dyj

98:                 + psi_bo(d, g, i, j) * zcos_dzk)

99:                 / (xcos_dxi + ycos_dyj + zcos_dzk + sigt(g, z));

100:

101:             psi(d, g, z) = psi_d_g_z;

102:

103:             /* Apply diamond-difference relationships */

104:             psi_lf(d, g, j, k) = 2.0 * psi_d_g_z - psi_lf(d, g, j, k);

105:             psi_fr(d, g, i, k) = 2.0 * psi_d_g_z - psi_fr(d, g, i, k);

106:             psi_bo(d, g, i, j) = 2.0 * psi_d_g_z - psi_bo(d, g, i, j);

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	9.33
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.14
Bottlenecks
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::For<4l, 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>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#1}>&>(RAJA::internal::LoopData, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#1}>&) [clone .omp_outlined]
Source	For.hpp:137-137
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	2.63
CQA cycles if no scalar integer	2.63
CQA cycles if FP arith vectorized	2.63
CQA cycles if fully vectorized	0.28
Front-end cycles	2.44
P0 cycles	1.83
P1 cycles	1.83
P2 cycles	1.83
P3 cycles	1.83
P4 cycles	1.83
P5 cycles	1.83
P6 cycles	2.63
P7 cycles	2.63
P8 cycles	2.63
P9 cycles	2.63
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	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	20.00
Nb uops	19.50
Nb loads	8.50
Nb stores	2.00
Nb stack references	9.00
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	68.00
Bytes stored	16.00
Stride 0	1.00
Stride 1	0.00
Stride n	0.00
Stride unknown	7.00
Stride indirect	5.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	12.05
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	12.50
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	12.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	10.94

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	10.18
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.12
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::For<4l, 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>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#1}>&>(RAJA::internal::LoopData, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#1}>&) [clone .omp_outlined]
Source	For.hpp:137-137
Source loop unroll info	NA
Source loop unroll confidence level	NA
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	3.50
CQA cycles if no scalar integer	3.50
CQA cycles if FP arith vectorized	3.50
CQA cycles if fully vectorized	0.34
Front-end cycles	3.13
P0 cycles	2.17
P1 cycles	2.17
P2 cycles	2.17
P3 cycles	2.17
P4 cycles	2.17
P5 cycles	2.17
P6 cycles	3.50
P7 cycles	3.50
P8 cycles	3.50
P9 cycles	3.50
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	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	26.00
Nb uops	25.00
Nb loads	11.00
Nb stores	3.00
Nb stack references	12.00
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	88.00
Bytes stored	24.00
Stride 0	1.00
Stride 1	0.00
Stride n	0.00
Stride unknown	7.00
Stride indirect	5.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	11.61
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	12.50
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	12.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	9.38

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	1.17
Bottlenecks	micro-operation queue, 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::For<4l, 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>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#1}>&>(RAJA::internal::LoopData, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#1}>&) [clone .omp_outlined]
Source	For.hpp:137-137
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	1.75
P0 cycles	1.50
P1 cycles	1.50
P2 cycles	1.50
P3 cycles	1.50
P4 cycles	1.50
P5 cycles	1.50
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	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	14.00
Nb uops	14.00
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	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
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	12.50
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::For<4l, 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>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#1}>&>(RAJA::internal::LoopData, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#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/pattern/kernel/For.hpp:137
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/src/Kripke/Kernel/SweepSubdomain.cpp:88-90,96-106

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

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. Only 12% of vector register length is used (average across all SSE/AVX instructions). By vectorizing your loop, you can lower the cost of an iteration from 2.63 to 0.28 cycles (9.33x 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
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)

No data for this section

Slow data structures access

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

Details

Constant unknown stride: 7 occurrence(s)
Irregular (variable stride) or indirect: 5 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)
Try to remove indirect accesses. If applicable, precompute elements out of the innermost loop.

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 68 bytes. The binary loop is storing 16 bytes.

General properties

nb instructions	20
nb uops	19.50
loop length	89
used x86 registers	9.50
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	9

Front-end

MACRO FUSION NOT POSSIBLE

micro-operation queue	2.44 cycles
front end	2.44 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	1.83	1.83	1.83	1.83	1.83	1.83	2.63	2.63	2.63	2.63	0.00	0.00	0.00	0.00	0.00	0.00
cycles	1.83	1.83	1.83	1.83	1.83	1.83	2.63	2.63	2.63	2.63	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)	1.00

Cycles summary

Front-end	2.44
Dispatch	2.63
Data deps.	1.00
Overall L1	2.63

Vectorization ratios

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

Vector efficiency ratios

all	12%
load	12%
store	12%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	10%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 2.63 cycles. At this rate:

10% of peak load performance is reached (26.29 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
4% of peak store performance is reached (5.71 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

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::For<4l, 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>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#1}>&>(RAJA::internal::LoopData, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#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/pattern/kernel/For.hpp:137
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/src/Kripke/Kernel/SweepSubdomain.cpp:88-90,96-106

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. Only 11% of vector register length is used (average across all SSE/AVX instructions). By vectorizing your loop, you can lower the cost of an iteration from 3.50 to 0.34 cycles (10.18x 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
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 3.50 to 3.13 cycles (1.12x 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: 7 occurrence(s)
Irregular (variable stride) or indirect: 5 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)
Try to remove indirect accesses. If applicable, precompute elements out of the innermost loop.

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 88 bytes. The binary loop is storing 24 bytes.

General properties

nb instructions	26
nb uops	25
loop length	117
used x86 registers	11
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	12

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	3.13 cycles
front end	3.13 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	2.17	2.17	2.17	2.17	2.17	2.17	3.50	3.50	3.50	3.50	0.00	0.00	0.00	0.00	0.00	0.00
cycles	2.17	2.17	2.17	2.17	2.17	2.17	3.50	3.50	3.50	3.50	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)	1.00

Cycles summary

Front-end	3.13
Dispatch	3.50
Data deps.	1.00
Overall L1	3.50

Vectorization ratios

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

Vector efficiency ratios

all	11%
load	12%
store	12%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	9%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 3.50 cycles. At this rate:

9% of peak load performance is reached (25.14 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (6.86 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: 26f64f

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
MOV -0x50(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV -0x60(%RBP),%R10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (12.5%)
ADD -0x1d8(%RBP),%R13	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%)
INC %R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
ADD %RAX,%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
ADD %RAX,%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
ADD %RAX,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV %R10,%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R10,-0x60(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
CMP %R10,-0x1e0(%RBP)	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%)
JE 26f550 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl1EEEEJNS2_3ForILl2ENS_6policy10sequential8seq_execEJNS8_ILl3ESB_JNS8_ILl4ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEEEENS0_9LoopTypesINS5_4listIJvvvvvEEESK_EEE4execIRNS0_8LoopDataINS5_5tupleIJNS_4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSU_EElEENSQ_INSS_INST_5GroupElPSY_EElEENSQ_INSR_24strided_numeric_iteratorINST_5ZoneKElPS13_EElEENSQ_INS12_INST_5ZoneJElPS17_EElEENSQ_INS12_INST_5ZoneIElPS1B_EElEEEEENSP_IJEEENS5_9resources2v14HostEJZNK9SweepSdomclINST_11ArchLayoutTINST_12ArchT_OpenMPENST_11LayoutT_DGZEEEEEvT_RNST_4Core9DataStoreENST_6SdomIdEEUlSU_SY_S13_S17_S1B_E_EEEEEvOS1Q_.omp_outlined+0x5b0>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
CMPQ $0,-0x1e8(%RBP)	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 -0x58(%RBP),%RBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (12.5%)
JLE 26f620 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl1EEEEJNS2_3ForILl2ENS_6policy10sequential8seq_execEJNS8_ILl3ESB_JNS8_ILl4ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEEEENS0_9LoopTypesINS5_4listIJvvvvvEEESK_EEE4execIRNS0_8LoopDataINS5_5tupleIJNS_4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSU_EElEENSQ_INSS_INST_5GroupElPSY_EElEENSQ_INSR_24strided_numeric_iteratorINST_5ZoneKElPS13_EElEENSQ_INS12_INST_5ZoneJElPS17_EElEENSQ_INS12_INST_5ZoneIElPS1B_EElEEEEENSP_IJEEENS5_9resources2v14HostEJZNK9SweepSdomclINST_11ArchLayoutTINST_12ArchT_OpenMPENST_11LayoutT_DGZEEEEEvT_RNST_4Core9DataStoreENST_6SdomIdEEUlSU_SY_S13_S17_S1B_E_EEEEEvOS1Q_.omp_outlined+0x680>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV -0x60(%RBP),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV -0x1d0(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
XOR %R11D,%R11D	1	0	0	0	0	0	0	0	0	0	0	0	0.25	scal (6.3%)
XOR %R14D,%R14D	1	0	0	0	0	0	0	0	0	0	0	0	0.25	scal (6.3%)
IMUL -0x138(%RBP),%RCX	1	0.33	0.33	0.33	0	0	0	0.25	0.25	0.25	0.25	3	0.33	N/A
ADD -0x140(%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	N/A
ADD %RCX,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %RCX,-0x78(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV %RAX,-0x70(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV -0x1c8(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
NOP	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
JMP 26f620 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl1EEEEJNS2_3ForILl2ENS_6policy10sequential8seq_execEJNS8_ILl3ESB_JNS8_ILl4ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEEEENS0_9LoopTypesINS5_4listIJvvvvvEEESK_EEE4execIRNS0_8LoopDataINS5_5tupleIJNS_4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSU_EElEENSQ_INSS_INST_5GroupElPSY_EElEENSQ_INSR_24strided_numeric_iteratorINST_5ZoneKElPS13_EElEENSQ_INS12_INST_5ZoneJElPS17_EElEENSQ_INS12_INST_5ZoneIElPS1B_EElEEEEENSP_IJEEENS5_9resources2v14HostEJZNK9SweepSdomclINST_11ArchLayoutTINST_12ArchT_OpenMPENST_11LayoutT_DGZEEEEEvT_RNST_4Core9DataStoreENST_6SdomIdEEUlSU_SY_S13_S17_S1B_E_EEEEEvOS1Q_.omp_outlined+0x680>	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::For<4l, 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>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#1}>&>(RAJA::internal::LoopData, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long>, RAJA::Span, long> >, camp::tuple<>, camp::resources::v1::Host, SweepSdom::operator() >(Kripke::ArchLayoutT, Kripke::Core::DataStore&, Kripke::SdomId) const::{lambda(Kripke::Direction, Kripke::Group, Kripke::ZoneK, Kripke::ZoneJ, Kripke::ZoneI)#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/pattern/kernel/For.hpp:137
/home/eoseret/qaas_runs_ZEN5/173-940-2141/intel/Kripke/build/Kripke/src/Kripke/Kernel/SweepSubdomain.cpp:88-90,96-106

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
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 1.50 cycles (1.17x 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

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	14
nb uops	14
loop length	61
used x86 registers	8
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	1.75 cycles
front end	1.75 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	1.50	1.50	1.50	1.50	1.50	1.50	1.75	1.75	1.75	1.75	0.00	0.00	0.00	0.00	0.00	0.00
cycles	1.50	1.50	1.50	1.50	1.50	1.50	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)	1.00

Cycles summary

Front-end	1.75
Dispatch	1.75
Data deps.	1.00
Overall L1	1.75

Vectorization ratios

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

Vector efficiency ratios

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

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 1.75 to 1.50 cycles (1.17x speedup).

ASM code

In the binary file, the address of the loop is: 26f64f

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
MOV -0x50(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (12.5%)
MOV -0x60(%RBP),%R10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (12.5%)
ADD -0x1d8(%RBP),%R13	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%)
INC %R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
ADD %RAX,%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
ADD %RAX,%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
ADD %RAX,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV %R10,%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
MOV %R10,-0x60(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
CMP %R10,-0x1e0(%RBP)	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%)
JE 26f550 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl1EEEEJNS2_3ForILl2ENS_6policy10sequential8seq_execEJNS8_ILl3ESB_JNS8_ILl4ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEEEENS0_9LoopTypesINS5_4listIJvvvvvEEESK_EEE4execIRNS0_8LoopDataINS5_5tupleIJNS_4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSU_EElEENSQ_INSS_INST_5GroupElPSY_EElEENSQ_INSR_24strided_numeric_iteratorINST_5ZoneKElPS13_EElEENSQ_INS12_INST_5ZoneJElPS17_EElEENSQ_INS12_INST_5ZoneIElPS1B_EElEEEEENSP_IJEEENS5_9resources2v14HostEJZNK9SweepSdomclINST_11ArchLayoutTINST_12ArchT_OpenMPENST_11LayoutT_DGZEEEEEvT_RNST_4Core9DataStoreENST_6SdomIdEEUlSU_SY_S13_S17_S1B_E_EEEEEvOS1Q_.omp_outlined+0x5b0>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
CMPQ $0,-0x1e8(%RBP)	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
MOV -0x58(%RBP),%RBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (12.5%)
JLE 26f620 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl1EEEEJNS2_3ForILl2ENS_6policy10sequential8seq_execEJNS8_ILl3ESB_JNS8_ILl4ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEEEENS0_9LoopTypesINS5_4listIJvvvvvEEESK_EEE4execIRNS0_8LoopDataINS5_5tupleIJNS_4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSU_EElEENSQ_INSS_INST_5GroupElPSY_EElEENSQ_INSR_24strided_numeric_iteratorINST_5ZoneKElPS13_EElEENSQ_INS12_INST_5ZoneJElPS17_EElEENSQ_INS12_INST_5ZoneIElPS1B_EElEEEEENSP_IJEEENS5_9resources2v14HostEJZNK9SweepSdomclINST_11ArchLayoutTINST_12ArchT_OpenMPENST_11LayoutT_DGZEEEEEvT_RNST_4Core9DataStoreENST_6SdomIdEEUlSU_SY_S13_S17_S1B_E_EEEEEvOS1Q_.omp_outlined+0x680>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A

Report Configuration

Vectorization

Details

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

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

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