OV - exec - Loop 150

	Loop Id: 150	Module: exec	Source: advec_mom_kernel.f90:115-118	Coverage: 2.65%

0x423d00 MOV	(%R15),%R11    [5]

0x423d03 IMUL	%RBX,%R11

0x423d07 ADD	%RDX,%R11

0x423d0a MOV	(%R13),%RAX    [3]

0x423d0e IMUL	%RBX,%RAX

0x423d12 ADD	%RSI,%RAX

0x423d15 MOVSD	(%R10,%RAX,1),%XMM0    [1]

0x423d1b MOVSD	%XMM0,(%R10,%R11,1)    [1]

0x423d21 MOV	(%R14),%RAX    [4]

0x423d24 IMUL	%RBX,%RAX

0x423d28 ADD	%R8,%RAX

0x423d2b MOVSD	0x8(%R10,%RAX,1),%XMM0    [1]

0x423d32 MOV	(%R15),%R11    [5]

0x423d35 IMUL	%RBX,%R11

0x423d39 ADD	%RDX,%R11

0x423d3c ADDSD	(%R10,%R11,1),%XMM0    [1]

0x423d42 SUBSD	(%R10,%RAX,1),%XMM0    [1]

0x423d48 MOV	(%R12),%RAX    [2]

0x423d4c IMUL	%RBX,%RAX

0x423d50 ADD	%RDI,%RAX

0x423d53 MOVSD	%XMM0,(%R10,%RAX,1)    [1]

0x423d59 ADD	$0x8,%R10

0x423d5d CMP	%R10,%RCX

0x423d60 JNE	423d00

/home/eoseret/qaas_runs_ZEN5/174-049-0948/intel/CloverLeaf1.3-FC/build/CloverLeaf1.3-FC/CloverLeaf_ref/kernels/advec_mom_kernel.f90: 115 - 118

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

115:         DO j=x_min-2,x_max+2

116:           post_vol(j,k)=volume(j,k)

117:           pre_vol(j,k)=post_vol(j,k)+vol_flux_x(j+1,k  )-vol_flux_x(j,k)

118:         ENDDO

Coverage (%)	Name	Source Location	Module
►53.82+	__kmp_invoke_microtask		libiomp5.so
○	__kmp_invoke_task_func		libiomp5.so
○	__kmp_launch_thread		libiomp5.so
○	_INTERNALea951ae6::__kmp_launc[...]		libiomp5.so
○	start_thread		libc.so.6
○	__GI___clone3		libc.so.6
►35.08+	__kmp_invoke_microtask		libiomp5.so
○	__kmp_invoke_task_func		libiomp5.so
○	__kmp_launch_thread		libiomp5.so
○	_INTERNALea951ae6::__kmp_launc[...]		libiomp5.so
►4.25+	__kmp_invoke_microtask		libiomp5.so
○	__kmp_invoke_task_func		libiomp5.so
○	__kmp_launch_thread		libiomp5.so
○	_INTERNALea951ae6::__kmp_launc[...]		libiomp5.so
○	start_thread		libc.so.6
►2.46+	__kmp_invoke_microtask		libiomp5.so
○	__kmp_invoke_task_func		libiomp5.so
○	__kmp_launch_thread		libiomp5.so
►2.09+	__kmp_invoke_microtask		libiomp5.so
○	__kmp_invoke_task_func		libiomp5.so
○	__kmp_fork_call		libiomp5.so
○	__kmpc_fork_call		libiomp5.so
○	advec_mom_kernel	advec_mom_kernel.f90:256	exec
○	advec_mom_driver	advec_mom_driver.f90:60	exec
○	advection	advection.f90:107	exec
○	hydro	hydro.f90:64	exec
○	main		exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start		exec
►2.05+	__kmp_invoke_microtask		libiomp5.so
○	__kmp_invoke_task_func		libiomp5.so
○	__kmp_fork_call		libiomp5.so
○	__kmpc_fork_call		libiomp5.so
○	advec_mom_kernel	advec_mom_kernel.f90:256	exec
○	advec_mom_driver	advec_mom_driver.f90:60	exec
○	advection	advection.f90:108	exec
○	hydro	hydro.f90:64	exec
○	main		exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start		exec

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	1.92
CQA speedup if FP arith vectorized	1.74
CQA speedup if fully vectorized	8.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.05
Bottlenecks	micro-operation queue,
Function	advec_mom_kernel_mod_mp_advec_mom_kernel_.DIR.OMP.PARALLEL.2
Source	advec_mom_kernel.f90:115-118
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	2.88
CQA cycles if no scalar integer	1.50
CQA cycles if FP arith vectorized	1.66
CQA cycles if fully vectorized	0.36
Front-end cycles	2.88
P0 cycles	2.00
P1 cycles	2.00
P2 cycles	2.00
P3 cycles	2.00
P4 cycles	2.00
P5 cycles	2.00
P6 cycles	2.75
P7 cycles	2.75
P8 cycles	2.75
P9 cycles	2.75
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	1.00
P13 cycles	1.00
P14 cycles	1.00
P15 cycles	1.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	24.00
Nb uops	23.00
Nb loads	9.00
Nb stores	2.00
Nb stack references	0.00
FLOP/cycle	0.70
Nb FLOP add-sub	2.00
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	30.61
Bytes prefetched	0.00
Bytes loaded	72.00
Bytes stored	16.00
Stride 0	4.00
Stride 1	0.00
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	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
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	12.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	NA

Metric	Value
CQA speedup if no scalar integer	1.92
CQA speedup if FP arith vectorized	1.74
CQA speedup if fully vectorized	8.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.05
Bottlenecks	micro-operation queue,
Function	advec_mom_kernel_mod_mp_advec_mom_kernel_.DIR.OMP.PARALLEL.2
Source	advec_mom_kernel.f90:115-118
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	2.88
CQA cycles if no scalar integer	1.50
CQA cycles if FP arith vectorized	1.66
CQA cycles if fully vectorized	0.36
Front-end cycles	2.88
P0 cycles	2.00
P1 cycles	2.00
P2 cycles	2.00
P3 cycles	2.00
P4 cycles	2.00
P5 cycles	2.00
P6 cycles	2.75
P7 cycles	2.75
P8 cycles	2.75
P9 cycles	2.75
P10 cycles	0.00
P11 cycles	0.00
P12 cycles	1.00
P13 cycles	1.00
P14 cycles	1.00
P15 cycles	1.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	24.00
Nb uops	23.00
Nb loads	9.00
Nb stores	2.00
Nb stack references	0.00
FLOP/cycle	0.70
Nb FLOP add-sub	2.00
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	30.61
Bytes prefetched	0.00
Bytes loaded	72.00
Bytes stored	16.00
Stride 0	4.00
Stride 1	0.00
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	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
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	12.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	NA

Path /

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

Function	advec_mom_kernel_mod_mp_advec_mom_kernel_.DIR.OMP.PARALLEL.2
Source file and lines	advec_mom_kernel.f90:115-118
Module	exec

The loop is defined in /home/eoseret/qaas_runs_ZEN5/174-049-0948/intel/CloverLeaf1.3-FC/build/CloverLeaf1.3-FC/CloverLeaf_ref/kernels/advec_mom_kernel.f90:115-118.

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

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 2.88 to 1.50 cycles (1.92x speedup).

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)
To reference allocatable arrays, use "allocatable" instead of "pointer" pointers or qualify them with the "contiguous" attribute (Fortran 2008)
For structures, limit to one indirection. For example, use a_b%c instead of a%b%c with a_b set to a%b before this loop

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 2.88 to 0.36 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:
- use the vec-report option to understand why your loop was not vectorized. If "existence of vector dependences", try the IVDEP directive. If, using IVDEP, "vectorization possible but seems inefficient", try the VECTOR ALWAYS directive.
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: Fortran storage order is column-major: do i do j a(i,j) = b(i,j) (slow, non stride 1) => do i do j a(j,i) = b(i,j) (fast, stride 1)
- If your loop streams arrays of structures (AoS), try to use structures of arrays instead (SoA): do i a(i)%x = b(i)%x (slow, non stride 1) => do 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

Type of elements and instruction set

2 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 2 FP arithmetical operations:

2: addition or subtraction

The binary loop is loading 72 bytes (9 double precision FP elements). The binary loop is storing 16 bytes (2 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.02 FP operations per loaded or stored byte.

Unroll opportunity

Loop is 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: !DIR$ UNROLL_AND_JAM(N), UNROLL_AND_JAM=N, UNROLL(N) or UNROLL=N

General properties

nb instructions	24
nb uops	23
loop length	98
used x86 registers	13
used mmx registers	0
used xmm registers	1
used ymm registers	0
used zmm registers	0
nb stack references	0

Front-end

ASSUMED MACRO FUSION FIT IN UOP CACHE

micro-operation queue	2.88 cycles
front end	2.88 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	2.00	2.00	2.00	2.00	2.00	2.00	2.75	2.75	2.75	2.75	0.00	0.00	1.00	1.00	1.00	1.00
cycles	2.00	2.00	2.00	2.00	2.00	2.00	2.75	2.75	2.75	2.75	0.00	0.00	1.00	1.00	1.00	1.00

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	2.88
Dispatch	2.75
Data deps.	1.00
Overall L1	2.88

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	NA (no other vectorizable/vectorized instructions)

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	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 2.88 cycles. At this rate:

9% of peak load performance is reached (25.04 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
4% of peak store performance is reached (5.57 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).

ASM code

In the binary file, the address of the loop is: 423d00

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P12	P13	P14	P15	Latency	Recip. throughput	Vectorization
MOV (%R15),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
IMUL %RBX,%R11	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	N/A
ADD %RDX,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV (%R13),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
IMUL %RBX,%RAX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	N/A
ADD %RSI,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOVSD (%R10,%RAX,1),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.50	scal (12.5%)
MOVSD %XMM0,(%R10,%R11,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0.50	0.50	1	0.50	scal (12.5%)
MOV (%R14),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
IMUL %RBX,%RAX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	N/A
ADD %R8,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOVSD 0x8(%R10,%RAX,1),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.50	scal (12.5%)
MOV (%R15),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
IMUL %RBX,%R11	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	N/A
ADD %RDX,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
ADDSD (%R10,%R11,1),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	2	0.50	scal (12.5%)
SUBSD (%R10,%RAX,1),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	2	0.50	scal (12.5%)
MOV (%R12),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
IMUL %RBX,%RAX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	N/A
ADD %RDI,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOVSD %XMM0,(%R10,%RAX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0.50	0.50	1	0.50	scal (12.5%)
ADD $0x8,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
CMP %R10,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
JNE 423d00 <advec_mom_kernel_mod_mp_advec_mom_kernel_.DIR.OMP.PARALLEL.2+0x930>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A

Function	advec_mom_kernel_mod_mp_advec_mom_kernel_.DIR.OMP.PARALLEL.2
Source file and lines	advec_mom_kernel.f90:115-118
Module	exec

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 2.88 to 1.50 cycles (1.92x speedup).

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)
To reference allocatable arrays, use "allocatable" instead of "pointer" pointers or qualify them with the "contiguous" attribute (Fortran 2008)
For structures, limit to one indirection. For example, use a_b%c instead of a%b%c with a_b set to a%b before this loop

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 2.88 to 0.36 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:
- use the vec-report option to understand why your loop was not vectorized. If "existence of vector dependences", try the IVDEP directive. If, using IVDEP, "vectorization possible but seems inefficient", try the VECTOR ALWAYS directive.
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: Fortran storage order is column-major: do i do j a(i,j) = b(i,j) (slow, non stride 1) => do i do j a(j,i) = b(i,j) (fast, stride 1)
- If your loop streams arrays of structures (AoS), try to use structures of arrays instead (SoA): do i a(i)%x = b(i)%x (slow, non stride 1) => do 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

Type of elements and instruction set

2 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 2 FP arithmetical operations:

2: addition or subtraction

The binary loop is loading 72 bytes (9 double precision FP elements). The binary loop is storing 16 bytes (2 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.02 FP operations per loaded or stored byte.

Unroll opportunity

Loop is data access bound.

Workaround

General properties

nb instructions	24
nb uops	23
loop length	98
used x86 registers	13
used mmx registers	0
used xmm registers	1
used ymm registers	0
used zmm registers	0
nb stack references	0

Front-end

ASSUMED MACRO FUSION FIT IN UOP CACHE

micro-operation queue	2.88 cycles
front end	2.88 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	2.00	2.00	2.00	2.00	2.00	2.00	2.75	2.75	2.75	2.75	0.00	0.00	1.00	1.00	1.00	1.00
cycles	2.00	2.00	2.00	2.00	2.00	2.00	2.75	2.75	2.75	2.75	0.00	0.00	1.00	1.00	1.00	1.00

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	2.88
Dispatch	2.75
Data deps.	1.00
Overall L1	2.88

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	NA (no other vectorizable/vectorized instructions)

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	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 2.88 cycles. At this rate:

9% of peak load performance is reached (25.04 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
4% of peak store performance is reached (5.57 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).

ASM code

In the binary file, the address of the loop is: 423d00

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P12	P13	P14	P15	Latency	Recip. throughput	Vectorization
MOV (%R15),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
IMUL %RBX,%R11	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	N/A
ADD %RDX,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV (%R13),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
IMUL %RBX,%RAX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	N/A
ADD %RSI,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOVSD (%R10,%RAX,1),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.50	scal (12.5%)
MOVSD %XMM0,(%R10,%R11,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0.50	0.50	1	0.50	scal (12.5%)
MOV (%R14),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
IMUL %RBX,%RAX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	N/A
ADD %R8,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOVSD 0x8(%R10,%RAX,1),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.50	scal (12.5%)
MOV (%R15),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
IMUL %RBX,%R11	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	N/A
ADD %RDX,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
ADDSD (%R10,%R11,1),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	2	0.50	scal (12.5%)
SUBSD (%R10,%RAX,1),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0	2	0.50	scal (12.5%)
MOV (%R12),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
IMUL %RBX,%RAX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	N/A
ADD %RDI,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOVSD %XMM0,(%R10,%RAX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0.50	0.50	1	0.50	scal (12.5%)
ADD $0x8,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
CMP %R10,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
JNE 423d00 <advec_mom_kernel_mod_mp_advec_mom_kernel_.DIR.OMP.PARALLEL.2+0x930>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

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

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

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