OV - exec - Function advancePosition.omp

	Function: advancePosition.omp_outlined	Module: exec	Source: timestep.c:85-94	Coverage (incl. loops): 1.37% \| (excl. loops): 0.00%

/home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c: 85 - 94

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

85:    #pragma omp parallel for

86:    for (int iBox=0; iBox<nBoxes; iBox++)

87:    {

88:       for (int iOff=MAXATOMS*iBox,ii=0; ii<s->boxes->nAtoms[iBox]; ii++,iOff++)

89:       {

90:          int iSpecies = s->atoms->iSpecies[iOff];

91:          real_t invMass = 1.0/s->species[iSpecies].mass;

92:          s->atoms->r[iOff][0] += dt*s->atoms->p[iOff][0]*invMass;

93:          s->atoms->r[iOff][1] += dt*s->atoms->p[iOff][1]*invMass;

94:          s->atoms->r[iOff][2] += dt*s->atoms->p[iOff][2]*invMass;

0x211c20 PUSH	%RBP

0x211c21 MOV	%RSP,%RBP

0x211c24 PUSH	%R15

0x211c26 PUSH	%R14

0x211c28 PUSH	%R13

0x211c2a PUSH	%R12

0x211c2c PUSH	%RBX

0x211c2d SUB	$0x18,%RSP

0x211c31 MOV	(%RDX),%R12D

0x211c34 TEST	%R12D,%R12D

0x211c37 JLE	211cb4

0x211c39 MOV	(%RDI),%ESI

0x211c3b DEC	%R12D

0x211c3e MOV	%R8,%RBX

0x211c41 MOV	%RCX,%R15

0x211c44 MOVL	$0,-0x30(%RBP)

0x211c4b MOVL	$0x1,-0x3c(%RBP)

0x211c52 MOVL	$0,-0x38(%RBP)

0x211c59 MOV	%R12D,-0x2c(%RBP)

0x211c5d SUB	$0x8,%RSP

0x211c61 LEA	-0x3c(%RBP),%RAX

0x211c65 LEA	0x1e54(%RIP),%RDI

0x211c6c LEA	-0x38(%RBP),%RCX

0x211c70 LEA	-0x30(%RBP),%R8

0x211c74 LEA	-0x2c(%RBP),%R9

0x211c78 MOV	%ESI,-0x34(%RBP)

0x211c7b MOV	$0x22,%EDX

0x211c80 PUSH	$0x1

0x211c82 PUSH	$0x1

0x211c84 PUSH	%RAX

0x211c85 CALL	212580 <@plt_start@+0x180>

0x211c8a ADD	$0x20,%RSP

0x211c8e MOV	-0x2c(%RBP),%EAX

0x211c91 CMP	%R12D,%EAX

0x211c94 CMOVL	%EAX,%R12D

0x211c98 MOVSXD	-0x30(%RBP),%RAX

0x211c9c MOV	%R12D,-0x2c(%RBP)

0x211ca0 CMP	%R12D,%EAX

0x211ca3 JLE	211cc3

0x211ca5 MOV	-0x34(%RBP),%ESI

0x211ca8 LEA	0x1e29(%RIP),%RDI

0x211caf CALL	212590 <@plt_start@+0x190>

0x211cb4 ADD	$0x18,%RSP

0x211cb8 POP	%RBX

0x211cb9 POP	%R12

0x211cbb POP	%R13

0x211cbd POP	%R14

0x211cbf POP	%R15

0x211cc1 POP	%RBP

0x211cc2 RET

0x211cc3 MOV	(%R15),%RCX

0x211cc6 VMOVSD	-0xd49e(%RIP),%XMM0

0x211cce MOV	%EAX,%ESI

0x211cd0 SAL	$0x6,%ESI

0x211cd3 MOVSXD	%R12D,%RDI

0x211cd6 MOV	0x18(%RCX),%RDX

0x211cda MOV	0x78(%RDX),%RDX

0x211cde JMP	211cef

(110) 0x211ce0 ADD	$0x40,%ESI

(110) 0x211ce3 LEA	0x1(%RAX),%R8

(110) 0x211ce7 CMP	%RDI,%RAX

(110) 0x211cea MOV	%R8,%RAX

(110) 0x211ced JGE	211ca5

(110) 0x211cef MOV	(%RDX,%RAX,4),%R8D

(110) 0x211cf3 TEST	%R8D,%R8D

(110) 0x211cf6 JLE	211ce0

(110) 0x211cf8 MOV	0x20(%RCX),%R13

(110) 0x211cfc MOVSXD	%ESI,%R9

(110) 0x211cff MOV	0x28(%RCX),%R11

(110) 0x211d03 LEA	(%R9,%R9,2),%R10

(110) 0x211d07 SAL	$0x2,%R9

(110) 0x211d0b LEA	0x10(,%R10,8),%R10

(110) 0x211d13 MOV	0x18(%R13),%R15

(110) 0x211d17 MOV	0x20(%R13),%R12

(110) 0x211d1b ADD	0x10(%R13),%R9

(110) 0x211d1f XOR	%R13D,%R13D

(110) 0x211d22 NOPW	%CS:(%RAX,%RAX,1)

  (111) 0x211d30 MOVSXD	(%R9,%R13,4),%R14

  (111) 0x211d34 VMOVSD	(%RBX),%XMM2

  (111) 0x211d38 INC	%R13

  (111) 0x211d3b VMULSD	-0x10(%R12,%R10,1),%XMM2,%XMM2

  (111) 0x211d42 SAL	$0x4,%R14

  (111) 0x211d46 VDIVSD	0x8(%R11,%R14,1),%XMM0,%XMM1

  (111) 0x211d4d VFMADD213SD	-0x10(%R15,%R10,1),%XMM1,%XMM2

  (111) 0x211d54 VMOVSD	%XMM2,-0x10(%R15,%R10,1)

  (111) 0x211d5b VMOVSD	(%RBX),%XMM2

  (111) 0x211d5f VMULSD	-0x8(%R12,%R10,1),%XMM2,%XMM2

  (111) 0x211d66 VFMADD213SD	-0x8(%R15,%R10,1),%XMM1,%XMM2

  (111) 0x211d6d VMOVSD	%XMM2,-0x8(%R15,%R10,1)

  (111) 0x211d74 VMOVSD	(%RBX),%XMM2

  (111) 0x211d78 VMULSD	(%R12,%R10,1),%XMM2,%XMM2

  (111) 0x211d7e VFMADD213SD	(%R15,%R10,1),%XMM1,%XMM2

  (111) 0x211d84 VMOVSD	%XMM2,(%R15,%R10,1)

  (111) 0x211d8a ADD	$0x18,%R10

  (111) 0x211d8e CMP	%R13D,%R8D

  (111) 0x211d91 JNE	211d30

(110) 0x211d93 JMP	211ce0

0x211d98 NOPL	(%RAX,%RAX,1)

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 /

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

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run orig_0

Source file and lines	timestep.c:85-94
Module	exec

The function is defined in /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c:85-94.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Code clean check

Detected a slowdown caused by scalar integer instructions (typically used for address computation). By removing them, you can lower the cost of an iteration from 7.63 to 1.33 cycles (5.72x speedup).

Workaround

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

Vectorization

Your function is not vectorized. Only 6% 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 7.63 to 0.48 cycles (15.73x 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

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

No data for this section

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.

unknown: 2 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c:85
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c:85

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.

CMOVL: 1 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c:86

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 52 bytes. The binary function is storing 24 bytes.

General properties

nb instructions	59
nb uops	61
loop length	200
used x86 registers	14
used mmx registers	0
used xmm registers	1
used ymm registers	0
used zmm registers	0
nb stack references	5

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	7.63 cycles
front end	7.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	4.33	4.33	4.33	4.33	4.33	4.33	4.25	4.25	4.25	4.25	0.00	0.00	0.00	0.00	0.00	0.00
cycles	4.33	4.33	4.33	4.33	4.33	4.33	4.25	4.25	4.25	4.25	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

Cycles summary

Front-end	7.63
Dispatch	4.33
Overall L1	7.63

Vectorization ratios

INT

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

all	0%
load	0%
store	NA (no store vectorizable/vectorized instructions)
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	NA (no other vectorizable/vectorized instructions)

INT+FP

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

INT

all	6%
load	6%
store	6%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	6%
fma	NA (no fma vectorizable/vectorized instructions)
other	7%

all	12%
load	12%
store	NA (no store vectorizable/vectorized instructions)
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	NA (no other vectorizable/vectorized instructions)

INT+FP

all	6%
load	7%
store	6%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	6%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	7%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 7.63 cycles. At this rate:

2% of peak load performance is reached (6.82 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (3.15 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 7.63 to 4.33 cycles (1.76x speedup).

ASM code

In the binary file, the address of the function is: 211c20

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
PUSH %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R13	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R12	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RBX	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SUB $0x18,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV (%RDX),%R12D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (6.3%)
TEST %R12D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JLE 211cb4 <advancePosition.omp_outlined+0x94>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV (%RDI),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (6.3%)
DEC %R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
MOV %R8,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
MOV %RCX,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOVL $0,-0x30(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVL $0x1,-0x3c(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVL $0,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOV %R12D,-0x2c(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
SUB $0x8,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA -0x3c(%RBP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA 0x1e54(%RIP),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA -0x38(%RBP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA -0x30(%RBP),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA -0x2c(%RBP),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %ESI,-0x34(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOV $0x22,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
PUSH $0x1	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH $0x1	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RAX	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CALL 212580 <@plt_start@+0x180>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
ADD $0x20,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV -0x2c(%RBP),%EAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
CMP %R12D,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
CMOVL %EAX,%R12D	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	1	0.33	scal (6.3%)
MOVSXD -0x30(%RBP),%RAX	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 %R12D,-0x2c(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
CMP %R12D,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JLE 211cc3 <advancePosition.omp_outlined+0xa3>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV -0x34(%RBP),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (6.3%)
LEA 0x1e29(%RIP),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
CALL 212590 <@plt_start@+0x190>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
ADD $0x18,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R12	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R13	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50	N/A
MOV (%R15),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
VMOVSD -0xd49e(%RIP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	1	0.50	scal (12.5%)
MOV %EAX,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
SAL $0x6,%ESI	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.33	scal (6.3%)
MOVSXD %R12D,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV 0x18(%RCX),%RDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x78(%RDX),%RDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
JMP 211cef <advancePosition.omp_outlined+0xcf>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
NOPL (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run orig_0

Source file and lines	timestep.c:85-94
Module	exec

The function is defined in /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c:85-94.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Code clean check

Detected a slowdown caused by scalar integer instructions (typically used for address computation). By removing them, you can lower the cost of an iteration from 7.63 to 1.33 cycles (5.72x speedup).

Workaround

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

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

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

No data for this section

CALL instructions

Detected function call instructions.

Details

unknown: 2 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c:85
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c:85

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.

CMOVL: 1 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c:86

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 52 bytes. The binary function is storing 24 bytes.

General properties

nb instructions	59
nb uops	61
loop length	200
used x86 registers	14
used mmx registers	0
used xmm registers	1
used ymm registers	0
used zmm registers	0
nb stack references	5

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	7.63 cycles
front end	7.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	4.33	4.33	4.33	4.33	4.33	4.33	4.25	4.25	4.25	4.25	0.00	0.00	0.00	0.00	0.00	0.00
cycles	4.33	4.33	4.33	4.33	4.33	4.33	4.25	4.25	4.25	4.25	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

Cycles summary

Front-end	7.63
Dispatch	4.33
Overall L1	7.63

Vectorization ratios

INT

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

all	0%
load	0%
store	NA (no store vectorizable/vectorized instructions)
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	NA (no other vectorizable/vectorized instructions)

INT+FP

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

INT

all	6%
load	6%
store	6%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	6%
fma	NA (no fma vectorizable/vectorized instructions)
other	7%

all	12%
load	12%
store	NA (no store vectorizable/vectorized instructions)
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	NA (no other vectorizable/vectorized instructions)

INT+FP

all	6%
load	7%
store	6%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	6%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	7%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 7.63 cycles. At this rate:

2% of peak load performance is reached (6.82 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (3.15 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

ASM code

In the binary file, the address of the function is: 211c20

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
PUSH %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R13	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R12	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RBX	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SUB $0x18,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV (%RDX),%R12D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (6.3%)
TEST %R12D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JLE 211cb4 <advancePosition.omp_outlined+0x94>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV (%RDI),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (6.3%)
DEC %R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
MOV %R8,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
MOV %RCX,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOVL $0,-0x30(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVL $0x1,-0x3c(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVL $0,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOV %R12D,-0x2c(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
SUB $0x8,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA -0x3c(%RBP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA 0x1e54(%RIP),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA -0x38(%RBP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA -0x30(%RBP),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA -0x2c(%RBP),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %ESI,-0x34(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOV $0x22,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
PUSH $0x1	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH $0x1	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RAX	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CALL 212580 <@plt_start@+0x180>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
ADD $0x20,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV -0x2c(%RBP),%EAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
CMP %R12D,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
CMOVL %EAX,%R12D	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	1	0.33	scal (6.3%)
MOVSXD -0x30(%RBP),%RAX	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 %R12D,-0x2c(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
CMP %R12D,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JLE 211cc3 <advancePosition.omp_outlined+0xa3>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV -0x34(%RBP),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (6.3%)
LEA 0x1e29(%RIP),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
CALL 212590 <@plt_start@+0x190>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
ADD $0x18,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R12	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R13	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50	N/A
MOV (%R15),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
VMOVSD -0xd49e(%RIP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	1	0.50	scal (12.5%)
MOV %EAX,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
SAL $0x6,%ESI	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.33	scal (6.3%)
MOVSXD %R12D,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV 0x18(%RCX),%RDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x78(%RDX),%RDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
JMP 211cef <advancePosition.omp_outlined+0xcf>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
NOPL (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A

Name	Coverage (%)	Time (s)
▼advancePosition.omp_outlined–	1.37	0.39
▼Loop 110 - timestep.c:86-94 - exec–	0.01	0.01
○Loop 111 - timestep.c:88-94 - exec	1.36	0.38

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

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

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

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