OV - exec - Loop 165

0x421e40 LD1D	{Z5.D}, P2/Z, [X14, X11,LSL #3]    [3]

0x421e44 LD1D	{Z6.D}, P2/Z, [X16, X11,LSL #3]    [5]

0x421e48 LD1D	{Z7.D}, P2/Z, [X24, X11,LSL #3]    [1]

0x421e4c LD1D	{Z17.D}, P2/Z, [X18, X11,LSL #3]    [4]

0x421e50 FADD	Z16.D, Z5.D, Z6.D

0x421e54 FADD	Z18.D, Z16.D, Z7.D

0x421e58 FADD	Z19.D, Z18.D, Z17.D

0x421e5c FMUL	Z20.D, Z19.D, Z4.D

0x421e60 ST1D	{Z20.D}, P2, [X1, X11,LSL #3]    [2]

0x421e64 ADD	X11, X11, X8

0x421e68 WHILELO	P2.D, X11, X15

0x421e6c B.NE	421e40

/home/hbollore/qaas/qaas-runs/169-814-4157/intel/CloverLeaf/build/CloverLeaf/CloverLeaf_ref/kernels/advec_mom_kernel.f90: 196 - 196

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

196:               +mass_flux_y(j-1,k+1)+mass_flux_y(j  ,k+1))

Coverage (%)	Name	Source Location	Module
►50.00+	GOMP_parallel		libgomp.so.1.0.0
○	advec_mom_kernel	advec_mom_kernel.f90:256	exec
○	advec_mom_driver	advec_mom_driver.f90:81	exec
○	advection	advection.f90:77	exec
○	hydro	hydro.f90:64	exec
○	main	clover_leaf.f90:41	exec
○	__libc_start_main		libc-2.31.so
○	_start	clover_leaf.f90:41	exec
►50.00+	GOMP_parallel		libgomp.so.1.0.0
○	advec_mom_kernel	advec_mom_kernel.f90:256	exec
○	advec_mom_driver	advec_mom_driver.f90:81	exec
○	advection	advection.f90:107	exec
○	hydro	hydro.f90:64	exec
○	main	clover_leaf.f90:41	exec
○	__libc_start_main		libc-2.31.so
○	_start	clover_leaf.f90:41	exec

Path /

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	1.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.25
Bottlenecks	P10, P11,
Function	advec_mom_kernel._omp_fn.0
Source	advec_mom_kernel.f90:196-196
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	2.50
CQA cycles if no scalar integer	2.50
CQA cycles if FP arith vectorized	2.50
CQA cycles if fully vectorized	2.50
Front-end cycles	1.50
DIV/SQRT cycles	0.50
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.25
P3 cycles	1.00
P4 cycles	0.25
P5 cycles	2.00
P6 cycles	2.00
P7 cycles	2.00
P8 cycles	2.00
P9 cycles	2.50
P10 cycles	2.50
P11 cycles	0.00
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	12.00
Nb uops	12.00
Nb loads	NA
Nb stores	1.00
Nb stack references	0.00
FLOP/cycle	6.40
Nb FLOP add-sub	12.00
Nb FLOP mul	4.00
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	64.00
Bytes prefetched	0.00
Bytes loaded	128.00
Bytes stored	32.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	100.00
Vectorization ratio load	100.00
Vectorization ratio store	100.00
Vectorization ratio mul	100.00
Vectorization ratio add_sub	100.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	NA
Vector-efficiency ratio all	100.00
Vector-efficiency ratio load	100.00
Vector-efficiency ratio store	100.00
Vector-efficiency ratio mul	100.00
Vector-efficiency ratio add_sub	100.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	NA

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	1.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.25
Bottlenecks	P10, P11,
Function	advec_mom_kernel._omp_fn.0
Source	advec_mom_kernel.f90:196-196
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	2.50
CQA cycles if no scalar integer	2.50
CQA cycles if FP arith vectorized	2.50
CQA cycles if fully vectorized	2.50
Front-end cycles	1.50
DIV/SQRT cycles	0.50
P0 cycles	0.50
P1 cycles	0.50
P2 cycles	0.25
P3 cycles	1.00
P4 cycles	0.25
P5 cycles	2.00
P6 cycles	2.00
P7 cycles	2.00
P8 cycles	2.00
P9 cycles	2.50
P10 cycles	2.50
P11 cycles	0.00
P12 cycles	0.00
P13 cycles	0.00
P14 cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	12.00
Nb uops	12.00
Nb loads	NA
Nb stores	1.00
Nb stack references	0.00
FLOP/cycle	6.40
Nb FLOP add-sub	12.00
Nb FLOP mul	4.00
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	64.00
Bytes prefetched	0.00
Bytes loaded	128.00
Bytes stored	32.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	100.00
Vectorization ratio load	100.00
Vectorization ratio store	100.00
Vectorization ratio mul	100.00
Vectorization ratio add_sub	100.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	NA
Vector-efficiency ratio all	100.00
Vector-efficiency ratio load	100.00
Vector-efficiency ratio store	100.00
Vector-efficiency ratio mul	100.00
Vector-efficiency ratio add_sub	100.00
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._omp_fn.0
Source file and lines	advec_mom_kernel.f90:196-196
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/169-814-4157/intel/CloverLeaf/build/CloverLeaf/CloverLeaf_ref/kernels/advec_mom_kernel.f90:196.

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

gain
potential
hint
expert

Vectorization

Your loop is fully vectorized, using full register length.

Details

All VPU instructions are used in vector version (process two or more data elements in vector registers).

FMA

Presence of both ADD/SUB and MUL operations.

Workaround

Try to change order in which elements are evaluated (using parentheses) in arithmetic expressions containing both ADD/SUB and MUL operations to enable your compiler to generate FMA instructions wherever possible. For instance a + b*c is a valid FMA (MUL then ADD). However (a+b)* c cannot be translated into an FMA (ADD then MUL).

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

The binary loop is composed of 16 FP arithmetical operations:

12: addition or subtraction
4: multiply

The binary loop is loading 128 bytes. The binary loop is storing 32 bytes.

Arithmetic intensity

Arithmetic intensity is 0.10 FP operations per loaded or stored byte.

General properties

nb instructions	12
loop length	48
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

1.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	0.50	0.50	0.50	0.25	1.00	0.25	2.00	2.00	0.50	0.50	2.50	2.50	0.00	0.00	0.00
cycles	0.50	0.50	0.50	0.25	1.00	0.25	2.00	2.00	2.00	2.00	2.50	2.50	0.00	0.00	0.00

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

Cycles summary

Front-end	1.50
Data deps.	1.00
Overall L1	2.50

Vectorization ratios

INT

all	100%
load	100%
store	100%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	NA (no other vectorizable/vectorized instructions)

FP

all	100%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	100%
add-sub	100%
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	100%
load	100%
store	100%
mul	100%
add-sub	100%
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 2.50 cycles. At this rate:

53% of peak load performance is reached (51.20 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
26% of peak store performance is reached (12.80 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 421e40

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput
LD1D {Z5.D}, P2/Z, [X14, X11,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z6.D}, P2/Z, [X16, X11,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z7.D}, P2/Z, [X24, X11,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z17.D}, P2/Z, [X18, X11,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
FADD Z16.D, Z5.D, Z6.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FADD Z18.D, Z16.D, Z7.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FADD Z19.D, Z18.D, Z17.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FMUL Z20.D, Z19.D, Z4.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	3	0.50
ST1D {Z20.D}, P2, [X1, X11,LSL #3]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	2	0.50
ADD X11, X11, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.25
WHILELO P2.D, X11, X15	1	0	0	0	0	1	0	0	0	0	0	0	0	3	2
B.NE 421e40 <__advec_mom_kernel_mod_MOD_advec_mom_kernel._omp_fn.0+0xd00>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50

Function	advec_mom_kernel._omp_fn.0
Source file and lines	advec_mom_kernel.f90:196-196
Module	exec

The loop is defined in /home/hbollore/qaas/qaas-runs/169-814-4157/intel/CloverLeaf/build/CloverLeaf/CloverLeaf_ref/kernels/advec_mom_kernel.f90:196.

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

gain
potential
hint
expert

Vectorization

Your loop is fully vectorized, using full register length.

Details

All VPU instructions are used in vector version (process two or more data elements in vector registers).

FMA

Presence of both ADD/SUB and MUL operations.

Workaround

Try to change order in which elements are evaluated (using parentheses) in arithmetic expressions containing both ADD/SUB and MUL operations to enable your compiler to generate FMA instructions wherever possible. For instance a + b*c is a valid FMA (MUL then ADD). However (a+b)* c cannot be translated into an FMA (ADD then MUL).

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

The binary loop is composed of 16 FP arithmetical operations:

12: addition or subtraction
4: multiply

The binary loop is loading 128 bytes. The binary loop is storing 32 bytes.

Arithmetic intensity

Arithmetic intensity is 0.10 FP operations per loaded or stored byte.

General properties

nb instructions	12
loop length	48
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

1.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	0.50	0.50	0.50	0.25	1.00	0.25	2.00	2.00	0.50	0.50	2.50	2.50	0.00	0.00	0.00
cycles	0.50	0.50	0.50	0.25	1.00	0.25	2.00	2.00	2.00	2.00	2.50	2.50	0.00	0.00	0.00

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

Cycles summary

Front-end	1.50
Data deps.	1.00
Overall L1	2.50

Vectorization ratios

INT

all	100%
load	100%
store	100%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	NA (no other vectorizable/vectorized instructions)

FP

all	100%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	100%
add-sub	100%
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	100%
load	100%
store	100%
mul	100%
add-sub	100%
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 2.50 cycles. At this rate:

53% of peak load performance is reached (51.20 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
26% of peak store performance is reached (12.80 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the loop is: 421e40

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput
LD1D {Z5.D}, P2/Z, [X14, X11,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z6.D}, P2/Z, [X16, X11,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z7.D}, P2/Z, [X24, X11,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
LD1D {Z17.D}, P2/Z, [X18, X11,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	6	0.50
FADD Z16.D, Z5.D, Z6.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FADD Z18.D, Z16.D, Z7.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FADD Z19.D, Z18.D, Z17.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	2	0.50
FMUL Z20.D, Z19.D, Z4.D	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	3	0.50
ST1D {Z20.D}, P2, [X1, X11,LSL #3]	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	2	0.50
ADD X11, X11, X8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.25
WHILELO P2.D, X11, X15	1	0	0	0	0	1	0	0	0	0	0	0	0	3	2
B.NE 421e40 <__advec_mom_kernel_mod_MOD_advec_mom_kernel._omp_fn.0+0xd00>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1	0.50

Report Configuration

Vectorization

Details

FMA

Workaround

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

Arithmetic intensity

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code

Vectorization

Details

FMA

Workaround

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

Arithmetic intensity

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code