OV - exec - Function qmcplusplus::DistanceTableBA<double, 3u, 39>::evaluate(qmcplusplus::ParticleSet&)

	Function: qmcplusplus::DistanceTableBA<double, 3u, 39>::evaluate(qmcplusplus::ParticleSet&)	Module: exec	Source: DistanceTableBA.h:66-105 [...]	Coverage: 0.01%

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Particle/DistanceTableBA.h: 66 - 105

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

66:   {

[...]

73:     const size_t ntgt_local  = Ntargets;

[...]

88:     for (size_t iat = 0; iat < ntgt_local; ++iat)

89:       for (size_t team_id = 0; team_id < num_teams; ++team_id)

90:       {

91:         T pos[D];

92:         for (int idim = 0; idim < D; idim++)

93:           pos[idim] = *(tgt_ptr + ntgt_padded * idim + iat);

94:

95:         const size_t first = ChunkSizePerTeam * team_id;

96:         const size_t last  = std::min(first + ChunkSizePerTeam, nsrc_padded);

97:

98:         PRAGMA_OFFLOAD("omp parallel for")

99:         for (size_t jel = first; jel < last; ++jel)

100:         {

101:           DTD_BConds<T, D, SC>::computeDistancesOffload(pos, src_ptr, nsrc_padded, dist_ptr + nsrc_padded * iat,

102:                                                         displ_ptr + nsrc_padded * D * iat, nsrc_padded, jel);

103:         }

104:       }

105:   }

0x44aa00 SUB	SP, SP, #192

0x44aa04 STP	X29, X30, [SP, #96]

0x44aa08 STP	X28, X27, [SP, #112]

0x44aa0c STP	X26, X25, [SP, #128]

0x44aa10 STP	X24, X23, [SP, #144]

0x44aa14 STP	X22, X21, [SP, #160]

0x44aa18 STP	X20, X19, [SP, #176]

0x44aa1c ADD	X29, SP, #96

0x44aa20 LDRSW	X8, [X0, #200]

0x44aa24 CBZ	W8, 44aa40

0x44aa28 LDRSW	X9, [X0, #196]

0x44aa2c ADD	X9, X9, #3

0x44aa30 AND	X19, X9, #8062

0x44aa34 ADD	X9, X19, #127

0x44aa38 CMP	X9, #128

0x44aa3c B.CS	44aa60

(1296) 0x44aa40 LDP	X20, X19, [SP, #176]

(1296) 0x44aa44 LDP	X22, X21, [SP, #160]

(1296) 0x44aa48 LDP	X24, X23, [SP, #144]

(1296) 0x44aa4c LDP	X26, X25, [SP, #128]

(1296) 0x44aa50 LDP	X28, X27, [SP, #112]

(1296) 0x44aa54 LDP	X29, X30, [SP, #96]

(1296) 0x44aa58 ADD	SP, SP, #192

(1296) 0x44aa5c RET

(1296) 0x44aa60 ADD	X10, X8, #3

(1296) 0x44aa64 UBFM	X9, X9, #7, #63

(1296) 0x44aa68 ORR	X11, XZR, XZR

(1296) 0x44aa6c ADD	X21, X0, #240

(1296) 0x44aa70 AND	X12, X10, #8062

(1296) 0x44aa74 LDR	X10, [X0, #64]

(1296) 0x44aa78 CMP	X9, #1

(1296) 0x44aa7c CSINC	X9, X9, XZR, #8

(1296) 0x44aa80 CMP	X8, #1

(1296) 0x44aa84 CSINC	X8, X8, XZR, #8

(1296) 0x44aa88 STP	X9, X12, [X29, #984]

(1296) 0x44aa8c STR	X8, [SP, #8]

(1296) 0x44aa90 UBFM	X8, X12, #63, #62

(1296) 0x44aa94 STR	X10, [SP, #32]

(1296) 0x44aa98 LDR	X10, [X0, #168]

(1296) 0x44aa9c STR	X8, [SP, #48]

(1296) 0x44aaa0 STR	X10, [SP, #24]

(1296) 0x44aaa4 LDR	X10, [X0, #8]

(1296) 0x44aaa8 LDR	X20, [X10, #1552]

(1296) 0x44aaac LDR	X10, [X1, #1552]

(1296) 0x44aab0 STR	X10, [SP, #16]

(1296) 0x44aab4 B	44aad4

0x44aab8 HINT	#0

0x44aabc HINT	#0

(1296) 0x44aac0 LDR	X11, [SP, #40]

(1296) 0x44aac4 LDR	X8, [SP, #8]

(1296) 0x44aac8 ADD	X11, X11, #1

(1296) 0x44aacc CMP	X11, X8

(1296) 0x44aad0 B.EQ	44aa40

(1296) 0x44aad4 LDR	X9, [SP, #32]

(1296) 0x44aad8 MADD	X8, X11, X19, XZR

(1296) 0x44aadc MOVZ	W10, #24

(1296) 0x44aae0 ORR	X25, XZR, XZR

(1296) 0x44aae4 MOVZ	W28, #128

(1296) 0x44aae8 STR	X11, [SP, #40]

(1296) 0x44aaec ADD	X22, X9, X8,LSL #3

(1296) 0x44aaf0 LDR	X9, [SP, #24]

(1296) 0x44aaf4 MADD	X23, X8, X10, X9

(1296) 0x44aaf8 LDR	X8, [SP, #16]

(1296) 0x44aafc ADD	X26, X8, X11,LSL #3

(1296) 0x44ab00 B	44ab18

  (1297) 0x44ab04 LDUR	X8, [X29, #472]

  (1297) 0x44ab08 ADD	X25, X25, #1

  (1297) 0x44ab0c ADD	X28, X28, #128

  (1297) 0x44ab10 CMP	X25, X8

  (1297) 0x44ab14 B.EQ	44aac0

  (1297) 0x44ab18 LDR	D0, [X26]

  (1297) 0x44ab1c LDUR	X8, [X29, #480]

  (1297) 0x44ab20 UBFM	X24, X25, #57, #56

  (1297) 0x44ab24 CMP	X28, X19

  (1297) 0x44ab28 CSEL	X27, X28, X19, #3

  (1297) 0x44ab2c STUR	D0, [X29, #488]

  (1297) 0x44ab30 LDR	D0, [X26, X8,LSL #3]

  (1297) 0x44ab34 LDR	X8, [SP, #48]

  (1297) 0x44ab38 STUR	D0, [X29, #496]

  (1297) 0x44ab3c LDR	D0, [X26, X8,LSL #3]

  (1297) 0x44ab40 ADD	X8, X24, #128

  (1297) 0x44ab44 CMP	X19, X8

  (1297) 0x44ab48 CSEL	X8, X19, X8, #3

  (1297) 0x44ab4c CMP	X24, X8

  (1297) 0x44ab50 STUR	D0, [X29, #504]

  (1297) 0x44ab54 B.CS	44ab04

  (1297) 0x44ab58 HINT	#0

  (1297) 0x44ab5c HINT	#0

    (1298) 0x44ab60 ORR	X0, XZR, X21

    (1298) 0x44ab64 ORR	X2, XZR, X20

    (1298) 0x44ab68 SUB	X1, X29, #24

    (1298) 0x44ab6c STR	WZR, [SP]

    (1298) 0x44ab70 ORR	W3, WZR, W19

    (1298) 0x44ab74 ORR	X4, XZR, X22

    (1298) 0x44ab78 ORR	X5, XZR, X23

    (1298) 0x44ab7c ORR	W6, WZR, W19

    (1298) 0x44ab80 ORR	W7, WZR, W24

    (1298) 0x44ab84 BL	44ad10

    (1298) 0x44ab88 ADD	X24, X24, #1

    (1298) 0x44ab8c CMP	X27, X24

    (1298) 0x44ab90 B.NE	44ab60

  (1297) 0x44ab94 B	44ab04

0x44ab98 HINT	#0

0x44ab9c HINT	#0

Coverage (%)	Name	Source Location	Module
►100.00+	qmcplusplus::ParticleSet::upda[...]	stl_vector.h:919	exec
○	.omp_outlined.	miniqmc.cpp:390	exec
○	__kmp_invoke_microtask		libomp.so

Path /

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

Source file and lines	DistanceTableBA.h:66-105
Module	exec

The function is defined in /home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Particle/DistanceTableBA.h:66,73,88-105.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Your function is not vectorized. 4 data elements could be processed at once in vector registers. By vectorizing your function, you can lower the cost of an iteration from 3.00 to 0.75 cycles (4.00x speedup).

Details

All VPU 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)

No data for this section

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 8 bytes. The binary function is storing 96 bytes.

General properties

nb instructions	20
loop length	80
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

2.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	1.00	1.00	1.50	1.50	1.50	1.50	0.00	0.00	0.00	0.00	3.00	3.00	2.00	3.00	3.00
cycles	1.00	1.00	1.50	1.50	1.50	1.50	0.00	0.00	0.00	0.00	3.00	3.00	2.00	3.00	3.00

Cycles executing div or sqrt instructions

Cycles summary

Front-end	2.00
Overall L1	3.00

Vectorization ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 3.00 cycles. At this rate:

2% of peak load performance is reached (2.67 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
66% of peak store performance is reached (32.00 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the function is: 44aa00

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput
SUB SP, SP, #192	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STP X29, X30, [SP, #96]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X28, X27, [SP, #112]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X26, X25, [SP, #128]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X24, X23, [SP, #144]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X22, X21, [SP, #160]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X20, X19, [SP, #176]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
ADD X29, SP, #96	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
LDRSW X8, [X0, #200]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CBZ W8, 44aa40 <_ZN11qmcplusplus15DistanceTableBAIdLj3ELi39EE8evaluateERNS_11ParticleSetE+0x40>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDRSW X9, [X0, #196]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
ADD X9, X9, #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
AND X19, X9, #8062	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X9, X19, #127	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X9, #128	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.CS 44aa60 <_ZN11qmcplusplus15DistanceTableBAIdLj3ELi39EE8evaluateERNS_11ParticleSetE+0x60>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
HINT #0
HINT #0
HINT #0
HINT #0

Source file and lines	DistanceTableBA.h:66-105
Module	exec

The function is defined in /home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Particle/DistanceTableBA.h:66,73,88-105.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Details

All VPU 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)

No data for this section

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 8 bytes. The binary function is storing 96 bytes.

General properties

nb instructions	20
loop length	80
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

2.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	1.00	1.00	1.50	1.50	1.50	1.50	0.00	0.00	0.00	0.00	3.00	3.00	2.00	3.00	3.00
cycles	1.00	1.00	1.50	1.50	1.50	1.50	0.00	0.00	0.00	0.00	3.00	3.00	2.00	3.00	3.00

Cycles executing div or sqrt instructions

Cycles summary

Front-end	2.00
Overall L1	3.00

Vectorization ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 3.00 cycles. At this rate:

2% of peak load performance is reached (2.67 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
66% of peak store performance is reached (32.00 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the function is: 44aa00

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput
SUB SP, SP, #192	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
STP X29, X30, [SP, #96]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X28, X27, [SP, #112]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X26, X25, [SP, #128]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X24, X23, [SP, #144]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X22, X21, [SP, #160]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
STP X20, X19, [SP, #176]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
ADD X29, SP, #96	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
LDRSW X8, [X0, #200]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CBZ W8, 44aa40 <_ZN11qmcplusplus15DistanceTableBAIdLj3ELi39EE8evaluateERNS_11ParticleSetE+0x40>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
LDRSW X9, [X0, #196]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
ADD X9, X9, #3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
AND X19, X9, #8062	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
ADD X9, X19, #127	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25
CMP X9, #128	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33
B.CS 44aa60 <_ZN11qmcplusplus15DistanceTableBAIdLj3ELi39EE8evaluateERNS_11ParticleSetE+0x60>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50
HINT #0
HINT #0
HINT #0
HINT #0

Name	Coverage (%)	Time (s)
▼qmcplusplus::DistanceTableBA::evaluate(qmcplusplus::ParticleSet&)–	0.01	0
▼Loop 1296 - DistanceTableBA.h:88-105 - exec–	0	0
▼Loop 1297 - DistanceTableBA.h:88-101 - exec–	0.01	0
○Loop 1298 - DistanceTableBA.h:99-101 - exec	0	0

Report Configuration

Vectorization

Details

Workaround

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code

Vectorization

Details

Workaround

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Cycles and memory resources usage

ASM code