OV - exec - Function qmcplusplus::ParticleSet::setActive(int)

	Function: qmcplusplus::ParticleSet::setActive(int)	Module: exec	Source: ParticleSet.cpp:256-261 [...]	Coverage: 0.01%

/usr/include/c++/11/bits/refwrap.h: 338 - 338

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

338:       { return *_M_data; }

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.cpp: 256 - 261

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

256: {

257:   ScopedTimer local_timer(timers[Timer_setActive]);

258:

259:   for (size_t i = 0; i < DistTables.size(); i++)

260:     DistTables[i]->evaluate(*this, iat);

261: }

/usr/include/c++/11/bits/stl_vector.h: 919 - 919

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

919:       { return size_type(this->_M_impl._M_finish - this->_M_impl._M_start); }

/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h: 242 - 249

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

242:   ScopeGuard(TIMER& t) : timer(t) { timer.start(); }

[...]

249:   ~ScopeGuard() { timer.stop(); }

0x462dc0 STP	X29, X30, [SP, #960]!

0x462dc4 ADD	X29, SP, #0

0x462dc8 STP	X19, X20, [SP, #16]

0x462dcc ADD	X20, X0, #2576

0x462dd0 STP	X21, X22, [SP, #32]

0x462dd4 ORR	X21, XZR, X0

0x462dd8 ORR	W22, WZR, W1

0x462ddc LDR	X0, [X0, #2728]

0x462de0 STR	X23, [SP, #48]

0x462de4 LDR	X23, [X0, #8]

0x462de8 ORR	X0, XZR, X23

0x462dec BL	479f20

0x462df0 LDR	X1, [X20, #8]

0x462df4 LDR	X2, [X21, #2576]

0x462df8 CMP	X1, X2

0x462dfc B.EQ	462e30

0x462e00 MOVZ	X19, #0

(1083) 0x462e04 LDR	X0, [X2, X19,LSL #3]

(1083) 0x462e08 ORR	X1, XZR, X21

(1083) 0x462e0c ORR	W2, WZR, W22

(1083) 0x462e10 LDR	X3, [X0]

(1083) 0x462e14 LDR	X4, [X3, #24]

(1083) 0x462e18 BLR	X4

(1083) 0x462e1c LDP	X2, X5, [X20]

(1083) 0x462e20 ADD	X19, X19, #1

(1083) 0x462e24 SUB	X6, X5, X2

(1083) 0x462e28 CMP	X19, X6,ASR #3

(1083) 0x462e2c B.CC	462e04

0x462e30 ORR	X0, XZR, X23

0x462e34 BL	47a180

0x462e38 LDP	X19, X20, [SP, #16]

0x462e3c LDP	X21, X22, [SP, #32]

0x462e40 LDR	X23, [SP, #48]

0x462e44 LDP	X29, X30, [SP], #64

0x462e48 RET

0x462e4c ORR	X21, XZR, X0

0x462e50 ORR	X0, XZR, X23

0x462e54 BL	47a180

0x462e58 ORR	X0, XZR, X21

0x462e5c BL	403aa0

Coverage (%)	Name	Source Location	Module
►100.00+	main._omp_fn.1	refwrap.h:337	exec
○	GOMP_parallel		libomp.so
○	main	miniqmc.cpp:404	exec
○	__libc_start_main		libc-2.31.so
○	_start	miniqmc.cpp:546	exec

Path /

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

Source file and lines	ParticleSet.cpp:256-261
Module	exec

The function is defined in:

/usr/include/c++/11/bits/refwrap.h:338
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.cpp:256-261
/usr/include/c++/11/bits/stl_vector.h:919
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:242,249

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Your function is not vectorized. Only 20% of vector register length is used (average across all VPU instructions). By vectorizing your function, you can lower the cost of an iteration from 4.00 to 0.53 cycles (7.53x 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)

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 4.00 to 2.00 cycles (2.00x speedup).

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 88 bytes. The binary function is storing 56 bytes.

General properties

nb instructions	29
loop length	116
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

3.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	3.00	3.00	4.00	4.00	4.00	4.00	0.00	0.00	0.00	0.00	4.00	4.00	4.00	2.00	2.00
cycles	3.00	3.00	4.00	4.00	4.00	4.00	0.00	0.00	0.00	0.00	4.00	4.00	4.00	2.00	2.00

Cycles executing div or sqrt instructions

Cycles summary

Front-end	3.63
Overall L1	4.00

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 4.00 cycles. At this rate:

22% of peak load performance is reached (22.00 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
29% of peak store performance is reached (14.00 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the function is: 462dc0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P10	P11	P12	P13	P14	Latency	Recip. throughput
STP X29, X30, [SP, #960]!	1	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0.50	0.50	1	0.50
ADD X29, SP, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
STP X19, X20, [SP, #16]	1	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
ADD X20, X0, #2576	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
STP X21, X22, [SP, #32]	1	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
ORR X21, XZR, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ORR W22, WZR, W1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR X0, [X0, #2728]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
STR X23, [SP, #48]	1	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
LDR X23, [X0, #8]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
ORR X0, XZR, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
BL 479f20 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE5startEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	1	0.50
LDR X1, [X20, #8]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X2, [X21, #2576]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X1, X2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 462e30 <_ZN11qmcplusplus11ParticleSet9setActiveEi+0x70>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
MOVZ X19, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ORR X0, XZR, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
BL 47a180 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	1	0.50
LDP X19, X20, [SP, #16]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
LDP X21, X22, [SP, #32]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
LDR X23, [SP, #48]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDP X29, X30, [SP], #64	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
RET	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
ORR X21, XZR, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ORR X0, XZR, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
BL 47a180 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	1	0.50
ORR X0, XZR, X21	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
BL 403aa0 <@plt_start@+0x420>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	1	0.50

Source file and lines	ParticleSet.cpp:256-261
Module	exec

The function is defined in:

/usr/include/c++/11/bits/refwrap.h:338
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Particle/ParticleSet.cpp:256-261
/usr/include/c++/11/bits/stl_vector.h:919
/home/hbollore/qaas/qaas-runs/169-816-8620/intel/miniqmc/build/miniqmc/src/Utilities/NewTimer.h:242,249

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)

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 4.00 to 2.00 cycles (2.00x speedup).

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 88 bytes. The binary function is storing 56 bytes.

General properties

nb instructions	29
loop length	116
nb stack references	0

Front-end

FIT IN UOP CACHE

front end

3.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	3.00	3.00	4.00	4.00	4.00	4.00	0.00	0.00	0.00	0.00	4.00	4.00	4.00	2.00	2.00
cycles	3.00	3.00	4.00	4.00	4.00	4.00	0.00	0.00	0.00	0.00	4.00	4.00	4.00	2.00	2.00

Cycles executing div or sqrt instructions

Cycles summary

Front-end	3.63
Overall L1	4.00

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 4.00 cycles. At this rate:

22% of peak load performance is reached (22.00 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
29% of peak store performance is reached (14.00 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

ASM code

In the binary file, the address of the function is: 462dc0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P10	P11	P12	P13	P14	Latency	Recip. throughput
STP X29, X30, [SP, #960]!	1	0	0	0.25	0.25	0.25	0.25	0.50	0.50	0	0.50	0.50	1	0.50
ADD X29, SP, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
STP X19, X20, [SP, #16]	1	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
ADD X20, X0, #2576	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
STP X21, X22, [SP, #32]	1	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
ORR X21, XZR, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ORR W22, WZR, W1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
LDR X0, [X0, #2728]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
STR X23, [SP, #48]	1	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50
LDR X23, [X0, #8]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
ORR X0, XZR, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
BL 479f20 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE5startEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	1	0.50
LDR X1, [X20, #8]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDR X2, [X21, #2576]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
CMP X1, X2	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.33
B.EQ 462e30 <_ZN11qmcplusplus11ParticleSet9setActiveEi+0x70>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
MOVZ X19, #0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ORR X0, XZR, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
BL 47a180 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	1	0.50
LDP X19, X20, [SP, #16]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
LDP X21, X22, [SP, #32]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
LDR X23, [SP, #48]	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33
LDP X29, X30, [SP], #64	1	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1
RET	1	0.50	0.50	0	0	0	0	0	0	0	0	0	1	0.50
ORR X21, XZR, X0	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
ORR X0, XZR, X23	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
BL 47a180 <_ZN11qmcplusplus9TimerTypeINS_8CPUClockEE4stopEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	1	0.50
ORR X0, XZR, X21	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	1	0.25
BL 403aa0 <@plt_start@+0x420>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	1	0.50

Name	Coverage (%)	Time (s)
▼qmcplusplus::ParticleSet::setActive(int)–	0.01	0
○Loop 1083 - ParticleSet.cpp:259-260 - exec	0.01	0

Report Configuration

Vectorization

Details

Workaround

Expensive FP math instructions/calls

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

Expensive FP math instructions/calls

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