OV - - Function _ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp

	Function: _ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_exe ...	Module: libkripke.so	Source: Collapse.hpp:81-81 [...]	Coverage: 2.63%

/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Operators.hpp: 307 - 307

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

307:     return Ret{lhs} + rhs;

/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Layout.hpp: 55 - 55

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

55:     return a * b;

/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/index/RangeSegment.hpp: 120 - 120

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

120:   RAJA_HOST_DEVICE RAJA_INLINE ~TypedRangeSegment() {}

/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/policy/openmp/kernel/Collapse.hpp: 81 - 81

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

81: #pragma omp parallel for private(i0, i1) firstprivate(privatizer) \

/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/policy/loop/forall.hpp: 59 - 59

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

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

/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/internal/Iterators.hpp: 55 - 177

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

55:       : val(rhs.val)

[...]

142:     return val - rhs.val;

[...]

177:     return value_type(val + rhs);

/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/View.hpp: 79 - 79

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

79:       : layout(V.layout), data(V.data)

/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/src/Kripke/Kernel/LPlusTimes.cpp: 57 - 57

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

57:             rhs(d,g,z) += ell_plus(d, nm) * phi_out(nm, g, z);

0x5b510 PUSH	%RBP

0x5b511 MOV	%RSP,%RBP

0x5b514 PUSH	%R15

0x5b516 PUSH	%R14

0x5b518 PUSH	%R13

0x5b51a PUSH	%R12

0x5b51c PUSH	%RBX

0x5b51d SUB	$0x68,%RSP

0x5b521 MOV	0x10(%RDI),%RAX

0x5b525 MOV	(%RAX),%RDX

0x5b528 MOV	0x20(%RAX),%RBX

0x5b52c MOV	0x18(%RAX),%RCX

0x5b530 MOV	0x38(%RAX),%RSI

0x5b534 MOV	0x68(%RAX),%R8

0x5b538 MOV	0xa8(%RAX),%R9

0x5b53f MOV	0xc0(%RAX),%R10

0x5b546 MOV	0xf0(%RAX),%R11

0x5b54d MOV	0x110(%RAX),%R15

0x5b554 MOV	0x10(%RAX),%R12

0x5b558 MOV	0x30(%RAX),%R13

0x5b55c MOV	0x60(%RAX),%R14

0x5b560 MOV	%RDX,-0x58(%RBP)

0x5b564 MOV	0x118(%RAX),%RDX

0x5b56b MOV	0x158(%RAX),%RAX

0x5b572 MOV	%RBX,-0x60(%RBP)

0x5b576 MOV	%RCX,-0x80(%RBP)

0x5b57a MOV	%RSI,-0x90(%RBP)

0x5b581 MOV	%R8,-0x50(%RBP)

0x5b585 MOV	%R9,-0x68(%RBP)

0x5b589 MOV	%R10,-0x40(%RBP)

0x5b58d MOV	%R11,-0x70(%RBP)

0x5b591 MOV	%R15,-0x88(%RBP)

0x5b598 MOV	%RDX,-0x48(%RBP)

0x5b59c MOV	%RAX,-0x78(%RBP)

0x5b5a0 MOV	0x8(%RDI),%RBX

0x5b5a4 MOV	(%RDI),%RDI

0x5b5a7 TEST	%RDI,%RDI

0x5b5aa MOV	%RDI,-0x38(%RBP)

0x5b5ae JLE	5b8a3

0x5b5b4 TEST	%RBX,%RBX

0x5b5b7 JLE	5b8a3

0x5b5bd CALL	9760 <omp_get_num_threads@plt>

0x5b5c2 MOVSXD	%EAX,%R15

0x5b5c5 CALL	9650 <omp_get_thread_num@plt>

0x5b5ca XOR	%EDX,%EDX

0x5b5cc MOVSXD	%EAX,%RSI

0x5b5cf MOV	-0x38(%RBP),%RAX

0x5b5d3 IMUL	%RBX,%RAX

0x5b5d7 DIV	%R15

0x5b5da MOV	%RAX,%RCX

0x5b5dd CMP	%RDX,%RSI

0x5b5e0 JB	5b8b9

0x5b5e6 IMUL	%RCX,%RSI

0x5b5ea LEA	(%RSI,%RDX,1),%RAX

0x5b5ee LEA	(%RCX,%RAX,1),%R8

0x5b5f2 CMP	%R8,%RAX

0x5b5f5 JAE	5b8a3

0x5b5fb XOR	%EDX,%EDX

0x5b5fd MOV	-0x88(%RBP),%R11

0x5b604 MOV	-0x80(%RBP),%R10

0x5b608 MOV	-0x90(%RBP),%RDI

0x5b60f DIV	%RBX

0x5b612 LEA	-0x1(%RCX),%R9

0x5b616 MOV	%RBX,-0x90(%RBP)

0x5b61d MOV	%R9,-0x80(%RBP)

0x5b621 XOR	%R8D,%R8D

0x5b624 MOV	%R11,%R15

0x5b627 SUB	%R12,%R10

0x5b62a SUB	%R13,%RDI

0x5b62d IMUL	%R12,%R15

0x5b631 SAL	$0x3,%R11

0x5b635 LEA	(%R15,%R13,1),%RSI

0x5b639 MOV	%RSI,-0x88(%RBP)

0x5b640 MOV	%RAX,-0x38(%RBP)

0x5b644 MOV	-0x38(%RBP),%R9

0x5b648 NOPL	(%RAX,%RAX,1)

(900) 0x5b650 TEST	%R10,%R10

(900) 0x5b653 JLE	5b889

(900) 0x5b659 MOV	-0x58(%RBP),%RAX

(900) 0x5b65d MOV	-0x40(%RBP),%RCX

(900) 0x5b661 MOV	-0x70(%RBP),%RSI

(900) 0x5b665 MOV	%R8,-0x38(%RBP)

(900) 0x5b669 MOV	-0x60(%RBP),%RBX

(900) 0x5b66d MOV	%RDX,%R8

(900) 0x5b670 ADD	%R9,%RAX

(900) 0x5b673 IMUL	%RAX,%RCX

(900) 0x5b677 LEA	(%RDX,%RBX,1),%R15

(900) 0x5b67b MOV	-0x88(%RBP),%RBX

(900) 0x5b682 IMUL	%R14,%RAX

(900) 0x5b686 ADD	%R12,%RCX

(900) 0x5b689 LEA	(%RSI,%RCX,8),%RSI

(900) 0x5b68d MOV	-0x48(%RBP),%RCX

(900) 0x5b691 IMUL	%R15,%RCX

(900) 0x5b695 ADD	%RBX,%RCX

(900) 0x5b698 MOV	-0x78(%RBP),%RBX

(900) 0x5b69c LEA	(%RBX,%RCX,8),%RCX

(900) 0x5b6a0 MOV	-0x50(%RBP),%RBX

(900) 0x5b6a4 IMUL	%RBX,%R15

(900) 0x5b6a8 ADD	%R15,%RAX

(900) 0x5b6ab MOV	-0x68(%RBP),%R15

(900) 0x5b6af ADD	%R13,%RAX

(900) 0x5b6b2 LEA	(%R15,%RAX,8),%RBX

(900) 0x5b6b6 XOR	%EAX,%EAX

(900) 0x5b6b8 MOV	%RAX,%RDX

(900) 0x5b6bb NOPL	(%RAX,%RAX,1)

  (901) 0x5b6c0 XOR	%EAX,%EAX

  (901) 0x5b6c2 TEST	%RDI,%RDI

  (901) 0x5b6c5 JLE	5b86f

  (901) 0x5b6cb MOV	%RDI,%R15

  (901) 0x5b6ce AND	$0x7,%R15D

  (901) 0x5b6d2 JE	5b7ad

  (901) 0x5b6d8 CMP	$0x1,%R15

  (901) 0x5b6dc JE	5b78d

  (901) 0x5b6e2 CMP	$0x2,%R15

  (901) 0x5b6e6 JE	5b776

  (901) 0x5b6ec CMP	$0x3,%R15

  (901) 0x5b6f0 JE	5b75f

  (901) 0x5b6f2 CMP	$0x4,%R15

  (901) 0x5b6f6 JE	5b748

  (901) 0x5b6f8 CMP	$0x5,%R15

  (901) 0x5b6fc JE	5b731

  (901) 0x5b6fe CMP	$0x6,%R15

  (901) 0x5b702 JE	5b71a

  (901) 0x5b704 MOV	$0x1,%EAX

  (901) 0x5b709 VMOVSD	(%RSI),%XMM0

  (901) 0x5b70d VMOVSD	(%RBX),%XMM7

  (901) 0x5b711 VFMADD132SD	(%RCX),%XMM7,%XMM0

  (901) 0x5b716 VMOVSD	%XMM0,(%RBX)

  (901) 0x5b71a VMOVSD	(%RBX,%RAX,8),%XMM2

  (901) 0x5b71f VMOVSD	(%RSI),%XMM1

  (901) 0x5b723 VFMADD132SD	(%RCX,%RAX,8),%XMM2,%XMM1

  (901) 0x5b729 VMOVSD	%XMM1,(%RBX,%RAX,8)

  (901) 0x5b72e INC	%RAX

  (901) 0x5b731 VMOVSD	(%RBX,%RAX,8),%XMM6

  (901) 0x5b736 VMOVSD	(%RSI),%XMM3

  (901) 0x5b73a VFMADD132SD	(%RCX,%RAX,8),%XMM6,%XMM3

  (901) 0x5b740 VMOVSD	%XMM3,(%RBX,%RAX,8)

  (901) 0x5b745 INC	%RAX

  (901) 0x5b748 VMOVSD	(%RBX,%RAX,8),%XMM5

  (901) 0x5b74d VMOVSD	(%RSI),%XMM4

  (901) 0x5b751 VFMADD132SD	(%RCX,%RAX,8),%XMM5,%XMM4

  (901) 0x5b757 VMOVSD	%XMM4,(%RBX,%RAX,8)

  (901) 0x5b75c INC	%RAX

  (901) 0x5b75f VMOVSD	(%RBX,%RAX,8),%XMM9

  (901) 0x5b764 VMOVSD	(%RSI),%XMM8

  (901) 0x5b768 VFMADD132SD	(%RCX,%RAX,8),%XMM9,%XMM8

  (901) 0x5b76e VMOVSD	%XMM8,(%RBX,%RAX,8)

  (901) 0x5b773 INC	%RAX

  (901) 0x5b776 VMOVSD	(%RBX,%RAX,8),%XMM11

  (901) 0x5b77b VMOVSD	(%RSI),%XMM10

  (901) 0x5b77f VFMADD132SD	(%RCX,%RAX,8),%XMM11,%XMM10

  (901) 0x5b785 VMOVSD	%XMM10,(%RBX,%RAX,8)

  (901) 0x5b78a INC	%RAX

  (901) 0x5b78d VMOVSD	(%RBX,%RAX,8),%XMM13

  (901) 0x5b792 VMOVSD	(%RSI),%XMM12

  (901) 0x5b796 VFMADD132SD	(%RCX,%RAX,8),%XMM13,%XMM12

  (901) 0x5b79c VMOVSD	%XMM12,(%RBX,%RAX,8)

  (901) 0x5b7a1 INC	%RAX

  (901) 0x5b7a4 CMP	%RAX,%RDI

  (901) 0x5b7a7 JE	5b86f

    (902) 0x5b7ad VMOVSD	(%RSI),%XMM14

    (902) 0x5b7b1 VMOVSD	(%RBX,%RAX,8),%XMM15

    (902) 0x5b7b6 VMOVSD	0x8(%RBX,%RAX,8),%XMM7

    (902) 0x5b7bc VFMADD132SD	(%RCX,%RAX,8),%XMM15,%XMM14

    (902) 0x5b7c2 VMOVSD	0x10(%RBX,%RAX,8),%XMM2

    (902) 0x5b7c8 VMOVSD	%XMM14,(%RBX,%RAX,8)

    (902) 0x5b7cd VMOVSD	0x18(%RBX,%RAX,8),%XMM6

    (902) 0x5b7d3 VMOVSD	(%RSI),%XMM0

    (902) 0x5b7d7 VMOVSD	0x20(%RBX,%RAX,8),%XMM5

    (902) 0x5b7dd VFMADD132SD	0x8(%RCX,%RAX,8),%XMM7,%XMM0

    (902) 0x5b7e4 VMOVSD	0x28(%RBX,%RAX,8),%XMM9

    (902) 0x5b7ea VMOVSD	%XMM0,0x8(%RBX,%RAX,8)

    (902) 0x5b7f0 VMOVSD	0x30(%RBX,%RAX,8),%XMM11

    (902) 0x5b7f6 VMOVSD	(%RSI),%XMM1

    (902) 0x5b7fa VMOVSD	0x38(%RBX,%RAX,8),%XMM13

    (902) 0x5b800 VFMADD132SD	0x10(%RCX,%RAX,8),%XMM2,%XMM1

    (902) 0x5b807 VMOVSD	%XMM1,0x10(%RBX,%RAX,8)

    (902) 0x5b80d VMOVSD	(%RSI),%XMM3

    (902) 0x5b811 VFMADD132SD	0x18(%RCX,%RAX,8),%XMM6,%XMM3

    (902) 0x5b818 VMOVSD	%XMM3,0x18(%RBX,%RAX,8)

    (902) 0x5b81e VMOVSD	(%RSI),%XMM4

    (902) 0x5b822 VFMADD132SD	0x20(%RCX,%RAX,8),%XMM5,%XMM4

    (902) 0x5b829 VMOVSD	%XMM4,0x20(%RBX,%RAX,8)

    (902) 0x5b82f VMOVSD	(%RSI),%XMM8

    (902) 0x5b833 VFMADD132SD	0x28(%RCX,%RAX,8),%XMM9,%XMM8

    (902) 0x5b83a VMOVSD	%XMM8,0x28(%RBX,%RAX,8)

    (902) 0x5b840 VMOVSD	(%RSI),%XMM10

    (902) 0x5b844 VFMADD132SD	0x30(%RCX,%RAX,8),%XMM11,%XMM10

    (902) 0x5b84b VMOVSD	%XMM10,0x30(%RBX,%RAX,8)

    (902) 0x5b851 VMOVSD	(%RSI),%XMM12

    (902) 0x5b855 VFMADD132SD	0x38(%RCX,%RAX,8),%XMM13,%XMM12

    (902) 0x5b85c VMOVSD	%XMM12,0x38(%RBX,%RAX,8)

    (902) 0x5b862 ADD	$0x8,%RAX

    (902) 0x5b866 CMP	%RAX,%RDI

    (902) 0x5b869 JNE	5b7ad

  (901) 0x5b86f INC	%RDX

  (901) 0x5b872 ADD	$0x8,%RSI

  (901) 0x5b876 ADD	%R11,%RCX

  (901) 0x5b879 CMP	%R10,%RDX

  (901) 0x5b87c JNE	5b6c0

(900) 0x5b882 MOV	%R8,%RDX

(900) 0x5b885 MOV	-0x38(%RBP),%R8

(900) 0x5b889 CMP	%R8,-0x80(%RBP)

(900) 0x5b88d JE	5b8a3

(900) 0x5b88f INC	%RDX

(900) 0x5b892 CMP	%RDX,-0x90(%RBP)

(900) 0x5b899 JLE	5b8b2

(900) 0x5b89b INC	%R8

(900) 0x5b89e JMP	5b650

0x5b8a3 ADD	$0x68,%RSP

0x5b8a7 POP	%RBX

0x5b8a8 POP	%R12

0x5b8aa POP	%R13

0x5b8ac POP	%R14

0x5b8ae POP	%R15

0x5b8b0 POP	%RBP

0x5b8b1 RET

(900) 0x5b8b2 INC	%R9

(900) 0x5b8b5 XOR	%EDX,%EDX

(900) 0x5b8b7 JMP	5b89b

0x5b8b9 INC	%RCX

0x5b8bc XOR	%EDX,%EDX

0x5b8be JMP	5b5e6

0x5b8c3 NOPW	%CS:(%RAX,%RAX,1)

0x5b8ce XCHG	%AX,%AX

Path /

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

Source file and lines	Collapse.hpp:81-81
Module	libkripke.so

The function is defined in:

/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Operators.hpp:307
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Layout.hpp:55
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/index/RangeSegment.hpp:120
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/policy/openmp/kernel/Collapse.hpp:81
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/policy/loop/forall.hpp:59
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/internal/Iterators.hpp:55,142,177
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/View.hpp:79
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/src/Kripke/Kernel/LPlusTimes.cpp:57

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Your function is not vectorized. Only 12% of vector register length is used (average across all SSE/AVX instructions). By vectorizing your function, you can lower the cost of an iteration from 24.00 to 6.00 cycles (4.00x 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 function can use their full power.

Workaround

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

Execution units bottlenecks

Performance is limited by execution of divide and square root operations (the divide/square root unit is a bottleneck). By removing all these bottlenecks, you can lower the cost of an iteration from 24.00 to 15.00 cycles (1.60x speedup).

Workaround

Reduce the number of division or square root instructions:

If denominator is constant over iterations, use reciprocal (replace x/y with x*(1/y)). Check precision impact. This will be done by your compiler with ffast-math or Ofast

Expensive FP math instructions/calls

Detected performance impact from expensive FP math instructions/calls. By removing/reexpressing them, you can lower the cost of an iteration from 24.00 to 14.67 cycles (1.64x speedup).

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.

omp_get_num_threads@plt: 1 occurrences
omp_get_thread_num@plt: 1 occurrences

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.

DIV: 2 occurrences

Type of elements and instruction set

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

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

The binary function does not contain any FP arithmetical operations. The binary function is loading 176 bytes. The binary function is storing 128 bytes.

General properties

nb instructions	89
nb uops	90
loop length	358
used x86 registers	16
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	15.00 cycles
front end	15.00 cycles

Back-end

	ALU0/BRU0	ALU1	ALU2	ALU3	BRU1	AGU0	AGU1	AGU2	FP0	FP1	FP2	FP3	FP4	FP5
uops	6.25	6.25	6.25	6.25	6.00	13.33	13.33	13.33	0.00	0.00	0.00	0.00	0.00	0.00
cycles	6.25	6.25	6.25	6.25	6.00	13.33	13.33	13.33	0.00	0.00	0.00	0.00	0.00	0.00

Execution ports to units layout:

ALU0/BRU0: ALU
ALU1: ALU
ALU2: ALU
ALU3: ALU
BRU1:
AGU0 (256 bits): store address, load
AGU1 (256 bits): store address, load
AGU2 (256 bits): store address, load
FP0 (256 bits): VPU, DIV/SQRT
FP1 (256 bits): VPU, DIV/SQRT
FP2 (256 bits): VPU
FP3 (256 bits): VPU
FP4 (256 bits): FP store data
FP5 (256 bits): FP store data

Cycles executing div or sqrt instructions

14.00-24.00

Cycles summary

Front-end	15.00
Dispatch	13.33
DIV/SQRT	14.00-24.00
Overall L1	15.00-24.00

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	0%
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	12%
other	11%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 24.00 cycles. At this rate:

7% of peak load performance is reached (7.33 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
8% of peak store performance is reached (5.33 out of 64.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 5b510

Instruction	Nb FU	ALU0/BRU0	ALU1	ALU2	ALU3	BRU1	AGU0	AGU1	AGU2	Latency	Recip. throughput
PUSH %RBP	1	0	0	0	0	0	0	0	0	1	0.50
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0.17
PUSH %R15	1	0	0	0	0	0	0	0	0	1	0.50
PUSH %R14	1	0	0	0	0	0	0	0	0	1	0.50
PUSH %R13	1	0	0	0	0	0	0	0	0	1	0.50
PUSH %R12	1	0	0	0	0	0	0	0	0	1	0.50
PUSH %RBX	1	0	0	0	0	0	0	0	0	1	0.50
SUB $0x68,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOV 0x10(%RDI),%RAX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV (%RAX),%RDX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x20(%RAX),%RBX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x18(%RAX),%RCX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x38(%RAX),%RSI	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x68(%RAX),%R8	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0xa8(%RAX),%R9	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0xc0(%RAX),%R10	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0xf0(%RAX),%R11	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x110(%RAX),%R15	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x10(%RAX),%R12	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x30(%RAX),%R13	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x60(%RAX),%R14	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV %RDX,-0x58(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV 0x118(%RAX),%RDX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x158(%RAX),%RAX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV %RBX,-0x60(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %RCX,-0x80(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %RSI,-0x90(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R8,-0x50(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R9,-0x68(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R10,-0x40(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R11,-0x70(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R15,-0x88(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %RDX,-0x48(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %RAX,-0x78(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV 0x8(%RDI),%RBX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV (%RDI),%RDI	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
TEST %RDI,%RDI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOV %RDI,-0x38(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
JLE 5b8a3 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp_fn.0+0x393>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1
TEST %RBX,%RBX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
JLE 5b8a3 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp_fn.0+0x393>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1
CALL 9760 <omp_get_num_threads@plt>	2	0.50	0	0	0	0.50	0.33	0.33	0.33	0	0.50
MOVSXD %EAX,%R15	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
CALL 9650 <omp_get_thread_num@plt>	2	0.50	0	0	0	0.50	0.33	0.33	0.33	0	0.50
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0.25
MOVSXD %EAX,%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOV -0x38(%RBP),%RAX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
IMUL %RBX,%RAX	1	0	1	0	0	0	0	0	0	3	1
DIV %R15	2	2	0	0	0	0	0	0	0	10-22	7-12
MOV %RAX,%RCX	1	0	0	0	0	0	0	0	0	0	0.17
CMP %RDX,%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
JB 5b8b9 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp_fn.0+0x3a9>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1
IMUL %RCX,%RSI	1	0	1	0	0	0	0	0	0	3	1
LEA (%RSI,%RDX,1),%RAX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LEA (%RCX,%RAX,1),%R8	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
CMP %R8,%RAX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
JAE 5b8a3 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp_fn.0+0x393>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0.25
MOV -0x88(%RBP),%R11	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV -0x80(%RBP),%R10	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV -0x90(%RBP),%RDI	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
DIV %RBX	2	2	0	0	0	0	0	0	0	10-22	7-12
LEA -0x1(%RCX),%R9	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOV %RBX,-0x90(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R9,-0x80(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0.25
MOV %R11,%R15	1	0	0	0	0	0	0	0	0	0	0.17
SUB %R12,%R10	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
SUB %R13,%RDI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
IMUL %R12,%R15	1	0	1	0	0	0	0	0	0	3	1
SAL $0x3,%R11	1	0	0.50	0.50	0	0	0	0	0	1	0.50
LEA (%R15,%R13,1),%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOV %RSI,-0x88(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %RAX,-0x38(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV -0x38(%RBP),%R9	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
NOPL (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0.09
ADD $0x68,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
POP %RBX	1	0	0	0	0	0	0	0	0	1	0.33
POP %R12	1	0	0	0	0	0	0	0	0	1	0.33
POP %R13	1	0	0	0	0	0	0	0	0	1	0.33
POP %R14	1	0	0	0	0	0	0	0	0	1	0.33
POP %R15	1	0	0	0	0	0	0	0	0	1	0.33
POP %RBP	1	0	0	0	0	0	0	0	0	1	0.33
RET	1	0.50	0	0	0	0.50	0	0	0	1	0.50
INC %RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0.25
JMP 5b5e6 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp_fn.0+0xd6>	1	0.50	0	0	0	0.50	0	0	0	0	1
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0.09
XCHG %AX,%AX	0	0	0	0	0	0	0	0	0	0	0.09

Source file and lines	Collapse.hpp:81-81
Module	libkripke.so

The function is defined in:

/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Operators.hpp:307
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/Layout.hpp:55
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/index/RangeSegment.hpp:120
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/policy/openmp/kernel/Collapse.hpp:81
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/policy/loop/forall.hpp:59
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/internal/Iterators.hpp:55,142,177
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/tpl/raja/include/RAJA/util/View.hpp:79
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/build/Kripke/src/Kripke/Kernel/LPlusTimes.cpp:57

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Details

Workaround

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

Execution units bottlenecks

Workaround

Reduce the number of division or square root instructions:

If denominator is constant over iterations, use reciprocal (replace x/y with x*(1/y)). Check precision impact. This will be done by your compiler with ffast-math or Ofast

Expensive FP math instructions/calls

Detected performance impact from expensive FP math instructions/calls. By removing/reexpressing them, you can lower the cost of an iteration from 24.00 to 14.67 cycles (1.64x speedup).

CALL instructions

Detected function call instructions.

Details

omp_get_num_threads@plt: 1 occurrences
omp_get_thread_num@plt: 1 occurrences

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.

DIV: 2 occurrences

Type of elements and instruction set

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

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

The binary function does not contain any FP arithmetical operations. The binary function is loading 176 bytes. The binary function is storing 128 bytes.

General properties

nb instructions	89
nb uops	90
loop length	358
used x86 registers	16
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	15.00 cycles
front end	15.00 cycles

Back-end

	ALU0/BRU0	ALU1	ALU2	ALU3	BRU1	AGU0	AGU1	AGU2	FP0	FP1	FP2	FP3	FP4	FP5
uops	6.25	6.25	6.25	6.25	6.00	13.33	13.33	13.33	0.00	0.00	0.00	0.00	0.00	0.00
cycles	6.25	6.25	6.25	6.25	6.00	13.33	13.33	13.33	0.00	0.00	0.00	0.00	0.00	0.00

Execution ports to units layout:

ALU0/BRU0: ALU
ALU1: ALU
ALU2: ALU
ALU3: ALU
BRU1:
AGU0 (256 bits): store address, load
AGU1 (256 bits): store address, load
AGU2 (256 bits): store address, load
FP0 (256 bits): VPU, DIV/SQRT
FP1 (256 bits): VPU, DIV/SQRT
FP2 (256 bits): VPU
FP3 (256 bits): VPU
FP4 (256 bits): FP store data
FP5 (256 bits): FP store data

Cycles executing div or sqrt instructions

14.00-24.00

Cycles summary

Front-end	15.00
Dispatch	13.33
DIV/SQRT	14.00-24.00
Overall L1	15.00-24.00

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	0%
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	12%
other	11%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 24.00 cycles. At this rate:

7% of peak load performance is reached (7.33 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
8% of peak store performance is reached (5.33 out of 64.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 5b510

Instruction	Nb FU	ALU0/BRU0	ALU1	ALU2	ALU3	BRU1	AGU0	AGU1	AGU2	Latency	Recip. throughput
PUSH %RBP	1	0	0	0	0	0	0	0	0	1	0.50
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0.17
PUSH %R15	1	0	0	0	0	0	0	0	0	1	0.50
PUSH %R14	1	0	0	0	0	0	0	0	0	1	0.50
PUSH %R13	1	0	0	0	0	0	0	0	0	1	0.50
PUSH %R12	1	0	0	0	0	0	0	0	0	1	0.50
PUSH %RBX	1	0	0	0	0	0	0	0	0	1	0.50
SUB $0x68,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOV 0x10(%RDI),%RAX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV (%RAX),%RDX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x20(%RAX),%RBX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x18(%RAX),%RCX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x38(%RAX),%RSI	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x68(%RAX),%R8	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0xa8(%RAX),%R9	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0xc0(%RAX),%R10	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0xf0(%RAX),%R11	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x110(%RAX),%R15	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x10(%RAX),%R12	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x30(%RAX),%R13	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x60(%RAX),%R14	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV %RDX,-0x58(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV 0x118(%RAX),%RDX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV 0x158(%RAX),%RAX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV %RBX,-0x60(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %RCX,-0x80(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %RSI,-0x90(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R8,-0x50(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R9,-0x68(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R10,-0x40(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R11,-0x70(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R15,-0x88(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %RDX,-0x48(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %RAX,-0x78(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV 0x8(%RDI),%RBX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV (%RDI),%RDI	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
TEST %RDI,%RDI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOV %RDI,-0x38(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
JLE 5b8a3 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp_fn.0+0x393>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1
TEST %RBX,%RBX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
JLE 5b8a3 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp_fn.0+0x393>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1
CALL 9760 <omp_get_num_threads@plt>	2	0.50	0	0	0	0.50	0.33	0.33	0.33	0	0.50
MOVSXD %EAX,%R15	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
CALL 9650 <omp_get_thread_num@plt>	2	0.50	0	0	0	0.50	0.33	0.33	0.33	0	0.50
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0.25
MOVSXD %EAX,%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOV -0x38(%RBP),%RAX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
IMUL %RBX,%RAX	1	0	1	0	0	0	0	0	0	3	1
DIV %R15	2	2	0	0	0	0	0	0	0	10-22	7-12
MOV %RAX,%RCX	1	0	0	0	0	0	0	0	0	0	0.17
CMP %RDX,%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
JB 5b8b9 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp_fn.0+0x3a9>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1
IMUL %RCX,%RSI	1	0	1	0	0	0	0	0	0	3	1
LEA (%RSI,%RDX,1),%RAX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
LEA (%RCX,%RAX,1),%R8	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
CMP %R8,%RAX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
JAE 5b8a3 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp_fn.0+0x393>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0.25
MOV -0x88(%RBP),%R11	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV -0x80(%RBP),%R10	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
MOV -0x90(%RBP),%RDI	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
DIV %RBX	2	2	0	0	0	0	0	0	0	10-22	7-12
LEA -0x1(%RCX),%R9	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOV %RBX,-0x90(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %R9,-0x80(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0.25
MOV %R11,%R15	1	0	0	0	0	0	0	0	0	0	0.17
SUB %R12,%R10	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
SUB %R13,%RDI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
IMUL %R12,%R15	1	0	1	0	0	0	0	0	0	3	1
SAL $0x3,%R11	1	0	0.50	0.50	0	0	0	0	0	1	0.50
LEA (%R15,%R13,1),%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
MOV %RSI,-0x88(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV %RAX,-0x38(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50
MOV -0x38(%RBP),%R9	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33
NOPL (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0.09
ADD $0x68,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
POP %RBX	1	0	0	0	0	0	0	0	0	1	0.33
POP %R12	1	0	0	0	0	0	0	0	0	1	0.33
POP %R13	1	0	0	0	0	0	0	0	0	1	0.33
POP %R14	1	0	0	0	0	0	0	0	0	1	0.33
POP %R15	1	0	0	0	0	0	0	0	0	1	0.33
POP %RBP	1	0	0	0	0	0	0	0	0	1	0.33
RET	1	0.50	0	0	0	0.50	0	0	0	1	0.50
INC %RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0.25
JMP 5b5e6 <_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS_4impl4SpanINS_9Iterators16numeric_iteratorIN6Kripke9DirectionElPSS_EESS_EENSO_INSQ_INSR_6MomentElPSW_EESW_EENSO_INSQ_INSR_5GroupElPS10_EES10_EENSO_INSQ_INSR_4ZoneElPS14_EES14_EEEEENSM_IJEEEJZNK14LPlusTimesSdomclINSR_11ArchLayoutTINSR_12ArchT_OpenMPENSR_11LayoutT_DGZEEEEEvT_NSR_6SdomIdERKNSR_4Core3SetES1L_S1L_S1L_RNS1I_5FieldIdJSW_S10_S14_EEERNS1M_IdJSS_S10_S14_EEERNS1M_IdJSS_SW_EEEEUlSS_SW_S10_S14_E_EEEEEvOS1G_._omp_fn.0+0xd6>	1	0.50	0	0	0	0.50	0	0	0	0	1
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0.09
XCHG %AX,%AX	0	0	0	0	0	0	0	0	0	0	0.09

▶Scalability Runs Description

Run 2x1	Number processes: 2Number nodes: 1Number processes per node: 2Run Command: <executable> --groups 1024 --procs 2,1,1MPI Command: mpirun -np <number_processes>Dataset: Run Directory: /beegfs/hackathon/users/eoseret/qaas_runs/170-850-6313/intel/Kripke/run/oneview_runs/compilers/gcc_13/oneview_run_1708510989OMP_NUM_THREADS: 1
Run 2x2	OMP_NUM_THREADS: 2
Run 2x4	OMP_NUM_THREADS: 4
Run 2x8	OMP_NUM_THREADS: 8
Run 2x16	OMP_NUM_THREADS: 16
Run 2x32	OMP_NUM_THREADS: 32
Run 2x64	OMP_NUM_THREADS: 64
Run 2x96	OMP_NUM_THREADS: 96

(2x1) Efficiency	(2x1) Potential Speed-Up (%)	(2x2) Efficiency	(2x2) Potential Speed-Up (%)	(2x4) Efficiency	(2x4) Potential Speed-Up (%)	(2x8) Efficiency	(2x8) Potential Speed-Up (%)	(2x16) Efficiency	(2x16) Potential Speed-Up (%)	(2x32) Efficiency	(2x32) Potential Speed-Up (%)	(2x64) Efficiency	(2x64) Potential Speed-Up (%)	(2x96) Efficiency	(2x96) Potential Speed-Up (%)
1	0	0.94	0.18	0.85	0.46	0.76	0.8	0.6	1.69	0.39	3.57	0.19	5.91	0.18	3.74

Speed-Up and Efficiency

▶Detailed Speed-Up and Efficiency

Run	Number of threads	Efficiency (ideal is 1)	Speedup	Ideal Speedup	Time (s)	Coverage (%)
2x1	2	1	1	1	27.08	2.63
2x2	4	0.94	1.87	2	14.26	2.83
2x4	8	0.85	3.4	4	8.14	3.08
2x8	16	0.76	6.08	8	5.11	3.34
2x16	32	0.6	9.57	16	3.23	4.2
2x32	64	0.39	12.38	32	2.47	5.83
2x64	128	0.19	12.05	64	2.19	7.28
2x96	192	0.18	17.55	96	1.7	4.58

Name	Coverage (%)	Time (s)
▼_ZN4RAJA8internal17StatementExecutorINS_9statement8CollapseINS_26omp_parallel_collapse_execEN4camp7int_seqIlJLl0ELl2EEEEJNS2_3ForILl1ENS_6policy4loop9loop_execEJNS8_ILl3ESB_JNS2_6LambdaILl0EJEEEEEEEEEEEEE4execIRNS0_8LoopDataINS5_4listIJSG_EEENS5_5tupleIJNS...–	2.63	26.5
▼Loop 900 - RangeSegment.hpp:120-120 - libkripke.so–	0	0
▼Loop 901 - forall.hpp:59-59 - libkripke.so–	0	0.01
○Loop 902 - forall.hpp:59-59 - libkripke.so	2.63	26.48

Report Configuration

Vectorization

Details

Workaround

Execution units bottlenecks

Workaround

Expensive FP math instructions/calls

CALL instructions

Details

Complex instructions

Details

Type of elements and instruction set

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

Vectorization

Details

Workaround

Execution units bottlenecks

Workaround

Expensive FP math instructions/calls

CALL instructions

Details

Complex instructions

Details

Type of elements and instruction set

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

▶Scalability Runs Description

▼Colums Filter

Speed-Up and Efficiency

▶Detailed Speed-Up and Efficiency