OV - exec - Loop 96

	Loop Id: 96	Module: exec	Source: timestep.c:88-94	Coverage: 0.68%

0x410810 MOVSXD	(%R15,%R9,4),%R14    [4]

0x410814 SAL	$0x4,%R14

0x410818 VMOVSD	-0x28(%R8,%R12,1),%XMM1    [1]

0x41081f VDIVSD	0x8(%R10,%R14,1),%XMM0,%XMM2    [3]

0x410826 VFMADD213SD	-0x28(%RDI,%R12,1),%XMM2,%XMM1    [2]

0x41082d VMOVSD	%XMM1,-0x28(%RDI,%R12,1)    [2]

0x410834 VMOVSD	-0x20(%R8,%R12,1),%XMM1    [1]

0x41083b VFMADD213SD	-0x20(%RDI,%R12,1),%XMM2,%XMM1    [2]

0x410842 VMOVSD	%XMM1,-0x20(%RDI,%R12,1)    [2]

0x410849 VMOVSD	-0x18(%R8,%R12,1),%XMM1    [1]

0x410850 VFMADD213SD	-0x18(%RDI,%R12,1),%XMM2,%XMM1    [2]

0x410857 VMOVSD	%XMM1,-0x18(%RDI,%R12,1)    [2]

0x41085e MOVSXD	0x4(%R15,%R9,4),%R14    [4]

0x410863 SAL	$0x4,%R14

0x410867 VMOVSD	-0x10(%R8,%R12,1),%XMM1    [1]

0x41086e VDIVSD	0x8(%R10,%R14,1),%XMM0,%XMM2    [3]

0x410875 VFMADD213SD	-0x10(%RDI,%R12,1),%XMM2,%XMM1    [2]

0x41087c VMOVSD	%XMM1,-0x10(%RDI,%R12,1)    [2]

0x410883 VMOVSD	-0x8(%R8,%R12,1),%XMM1    [1]

0x41088a VFMADD213SD	-0x8(%RDI,%R12,1),%XMM2,%XMM1    [2]

0x410891 VMOVSD	%XMM1,-0x8(%RDI,%R12,1)    [2]

0x410898 VMOVSD	(%R8,%R12,1),%XMM1    [1]

0x41089e VFMADD213SD	(%RDI,%R12,1),%XMM2,%XMM1    [2]

0x4108a4 VMOVSD	%XMM1,(%RDI,%R12,1)    [2]

0x4108aa ADD	$0x2,%R9

0x4108ae ADD	$0x30,%R12

0x4108b2 ADD	$-0x2,%R13D

0x4108b6 JNE	410810

/home/eoseret/qaas_runs_GNR/173-814-1858/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c: 88 - 94

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

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;

Coverage (%)	Name	Source Location	Module
►100.00+	__kmp_invoke_microtask		libiomp5.so
○	__kmp_invoke_task_func		libiomp5.so

Path /

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.68
CQA speedup if fully vectorized	2.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.37
Bottlenecks	P0,
Function	advancePosition.extracted
Source	timestep.c:88-94
Source loop unroll info	not unrolled or unrolled with no peel/tail loop
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	8.00
CQA cycles if no scalar integer	8.00
CQA cycles if FP arith vectorized	4.75
CQA cycles if fully vectorized	4.00
Front-end cycles	5.83
P0 cycles	4.00
P1 cycles	4.00
P2 cycles	5.33
P3 cycles	5.33
P4 cycles	3.00
P5 cycles	0.00
P6 cycles	3.00
P7 cycles	3.00
P8 cycles	3.00
P9 cycles	3.00
P10 cycles	0.00
P11 cycles	5.33
DIV/SQRT cycles	8.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	28.00
Nb uops	27.00
Nb loads	16.00
Nb stores	6.00
Nb stack references	0.00
FLOP/cycle	1.75
Nb FLOP add-sub	0.00
Nb FLOP mul	0.00
Nb FLOP fma	6.00
Nb FLOP div	2.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	21.00
Bytes prefetched	0.00
Bytes loaded	120.00
Bytes stored	48.00
Stride 0	1.00
Stride 1	3.00
Stride n	0.00
Stride unknown	0.00
Stride indirect	0.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	NA
Vectorization ratio fma	0.00
Vectorization ratio div_sqrt	0.00
Vectorization ratio other	NA
Vector-efficiency ratio all	12.50
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	12.50
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	NA
Vector-efficiency ratio fma	12.50
Vector-efficiency ratio div_sqrt	12.50
Vector-efficiency ratio other	NA

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.68
CQA speedup if fully vectorized	2.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.37
Bottlenecks	P0,
Function	advancePosition.extracted
Source	timestep.c:88-94
Source loop unroll info	not unrolled or unrolled with no peel/tail loop
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	8.00
CQA cycles if no scalar integer	8.00
CQA cycles if FP arith vectorized	4.75
CQA cycles if fully vectorized	4.00
Front-end cycles	5.83
P0 cycles	4.00
P1 cycles	4.00
P2 cycles	5.33
P3 cycles	5.33
P4 cycles	3.00
P5 cycles	0.00
P6 cycles	3.00
P7 cycles	3.00
P8 cycles	3.00
P9 cycles	3.00
P10 cycles	0.00
P11 cycles	5.33
DIV/SQRT cycles	8.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	28.00
Nb uops	27.00
Nb loads	16.00
Nb stores	6.00
Nb stack references	0.00
FLOP/cycle	1.75
Nb FLOP add-sub	0.00
Nb FLOP mul	0.00
Nb FLOP fma	6.00
Nb FLOP div	2.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	21.00
Bytes prefetched	0.00
Bytes loaded	120.00
Bytes stored	48.00
Stride 0	1.00
Stride 1	3.00
Stride n	0.00
Stride unknown	0.00
Stride indirect	0.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	NA
Vectorization ratio fma	0.00
Vectorization ratio div_sqrt	0.00
Vectorization ratio other	NA
Vector-efficiency ratio all	12.50
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	12.50
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	NA
Vector-efficiency ratio fma	12.50
Vector-efficiency ratio div_sqrt	12.50
Vector-efficiency ratio other	NA

Path /

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

Function	advancePosition.extracted
Source file and lines	timestep.c:88-94
Module	exec

The loop is defined in /home/eoseret/qaas_runs_GNR/173-814-1858/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c:88-94.

The related source loop is not unrolled or unrolled with no peel/tail loop.

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 8.00 to 4.00 cycles (2.00x speedup).

Details

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

Workaround

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

Execution units bottlenecks

Performance is limited by 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 8.00 to 5.83 cycles (1.37x 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
Check whether you really need double precision. If not, switch to single precision to speedup execution

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 8.00 to 5.50 cycles (1.45x speedup).

FMA

Detected 6 FMA (fused multiply-add) operations.

Type of elements and instruction set

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

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

The binary loop is composed of 14 FP arithmetical operations:

6: addition or subtraction (all inside FMA instructions)
6: multiply (all inside FMA instructions)
2: divide

The binary loop is loading 120 bytes (15 double precision FP elements). The binary loop is storing 48 bytes (6 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.08 FP operations per loaded or stored byte.

Unroll opportunity

Loop is potentially data access bound.

Workaround

Unroll your loop if trip count is significantly higher than target unroll factor and if some data references are common to consecutive iterations. This can be done manually. Or with the unroll (resp. unroll_and_jam) directive on top of the inner (resp. surrounding) loop. You can enforce an unroll factor: #pragma unroll_and_jam N, unroll_and_jam(N), unroll N or unroll(N)

General properties

nb instructions	28
nb uops	27
loop length	172
used x86 registers	8
used mmx registers	0
used xmm registers	3
used ymm registers	0
used zmm registers	0
nb stack references	0

Front-end

ASSUMED MACRO FUSION FIT IN UOP CACHE

micro-operation queue	5.83 cycles
front end	5.83 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	4.00	4.00	5.33	5.33	3.00	0.00	3.00	3.00	3.00	3.00	0.00	5.33
cycles	4.00	4.00	5.33	5.33	3.00	0.00	3.00	3.00	3.00	3.00	0.00	5.33

Execution ports to units layout:

P0 (256 bits): VPU, BRU, ALU, DIV/SQRT
P1 (256 bits): ALU, VPU
P2 (512 bits): load
P3 (512 bits): load
P4 (512 bits): store data
P5 (512 bits): ALU, VPU
P6: ALU, BRU
P7: store address
P8: store address
P9 (512 bits): store data
P10: ALU
P11 (512 bits): load

Cycles executing div or sqrt instructions	8.00
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	5.83
Dispatch	5.33
DIV/SQRT	8.00
Data deps.	1.00
Overall L1	8.00

Vectorization ratios

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

Vector efficiency ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 8.00 cycles. At this rate:

7% of peak load performance is reached (15.00 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
4% of peak store performance is reached (6.00 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: 410810

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
MOVSXD (%R15,%R9,4),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
SAL $0x4,%R14	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
VMOVSD -0x28(%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VDIVSD 0x8(%R10,%R14,1),%XMM0,%XMM2	1	1	0	0.33	0.33	0	0	0	0	0	0	0	0.33	13-15	4	scal (12.5%)
VFMADD213SD -0x28(%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,-0x28(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD -0x20(%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VFMADD213SD -0x20(%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,-0x20(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD -0x18(%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VFMADD213SD -0x18(%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,-0x18(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVSXD 0x4(%R15,%R9,4),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
SAL $0x4,%R14	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
VMOVSD -0x10(%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VDIVSD 0x8(%R10,%R14,1),%XMM0,%XMM2	1	1	0	0.33	0.33	0	0	0	0	0	0	0	0.33	13-15	4	scal (12.5%)
VFMADD213SD -0x10(%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,-0x10(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD -0x8(%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VFMADD213SD -0x8(%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,-0x8(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD (%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VFMADD213SD (%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
ADD $0x2,%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD $0x30,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD $-0x2,%R13D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JNE 410810 <advancePosition.extracted+0x120>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A

Function	advancePosition.extracted
Source file and lines	timestep.c:88-94
Module	exec

The loop is defined in /home/eoseret/qaas_runs_GNR/173-814-1858/intel/CoMD/build/CoMD/CoMD/src-openmp/timestep.c:88-94.

The related source loop is not unrolled or unrolled with no peel/tail loop.

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 8.00 to 4.00 cycles (2.00x speedup).

Details

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

Workaround

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

Execution units bottlenecks

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
Check whether you really need double precision. If not, switch to single precision to speedup execution

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 8.00 to 5.50 cycles (1.45x speedup).

FMA

Detected 6 FMA (fused multiply-add) operations.

Type of elements and instruction set

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

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

The binary loop is composed of 14 FP arithmetical operations:

6: addition or subtraction (all inside FMA instructions)
6: multiply (all inside FMA instructions)
2: divide

The binary loop is loading 120 bytes (15 double precision FP elements). The binary loop is storing 48 bytes (6 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.08 FP operations per loaded or stored byte.

Unroll opportunity

Loop is potentially data access bound.

Workaround

General properties

nb instructions	28
nb uops	27
loop length	172
used x86 registers	8
used mmx registers	0
used xmm registers	3
used ymm registers	0
used zmm registers	0
nb stack references	0

Front-end

ASSUMED MACRO FUSION FIT IN UOP CACHE

micro-operation queue	5.83 cycles
front end	5.83 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	4.00	4.00	5.33	5.33	3.00	0.00	3.00	3.00	3.00	3.00	0.00	5.33
cycles	4.00	4.00	5.33	5.33	3.00	0.00	3.00	3.00	3.00	3.00	0.00	5.33

Execution ports to units layout:

P0 (256 bits): VPU, BRU, ALU, DIV/SQRT
P1 (256 bits): ALU, VPU
P2 (512 bits): load
P3 (512 bits): load
P4 (512 bits): store data
P5 (512 bits): ALU, VPU
P6: ALU, BRU
P7: store address
P8: store address
P9 (512 bits): store data
P10: ALU
P11 (512 bits): load

Cycles executing div or sqrt instructions	8.00
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	5.83
Dispatch	5.33
DIV/SQRT	8.00
Data deps.	1.00
Overall L1	8.00

Vectorization ratios

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

Vector efficiency ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 8.00 cycles. At this rate:

7% of peak load performance is reached (15.00 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
4% of peak store performance is reached (6.00 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: 410810

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
MOVSXD (%R15,%R9,4),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
SAL $0x4,%R14	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
VMOVSD -0x28(%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VDIVSD 0x8(%R10,%R14,1),%XMM0,%XMM2	1	1	0	0.33	0.33	0	0	0	0	0	0	0	0.33	13-15	4	scal (12.5%)
VFMADD213SD -0x28(%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,-0x28(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD -0x20(%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VFMADD213SD -0x20(%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,-0x20(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD -0x18(%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VFMADD213SD -0x18(%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,-0x18(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVSXD 0x4(%R15,%R9,4),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
SAL $0x4,%R14	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
VMOVSD -0x10(%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VDIVSD 0x8(%R10,%R14,1),%XMM0,%XMM2	1	1	0	0.33	0.33	0	0	0	0	0	0	0	0.33	13-15	4	scal (12.5%)
VFMADD213SD -0x10(%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,-0x10(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD -0x8(%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VFMADD213SD -0x8(%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,-0x8(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD (%R8,%R12,1),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VFMADD213SD (%RDI,%R12,1),%XMM2,%XMM1	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
VMOVSD %XMM1,(%RDI,%R12,1)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
ADD $0x2,%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD $0x30,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD $-0x2,%R13D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JNE 410810 <advancePosition.extracted+0x120>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A

Report Configuration

Vectorization

Details

Workaround

Execution units bottlenecks

Workaround

Expensive FP math instructions/calls

FMA

Type of elements and instruction set

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

Arithmetic intensity

Unroll opportunity

Workaround

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

FMA

Type of elements and instruction set

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

Arithmetic intensity

Unroll opportunity

Workaround

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code