OV - exec - Function main.omp

	Function: main.omp_outlined.18	Module: exec	Source: stream.c:290-292	Coverage (incl. loops): 0.48% \| (excl. loops): 0.00%

/beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/build/stream-triad/src/stream.c: 290 - 292

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

290: #pragma omp parallel for

291:     for (j = 0; j < STREAM_ARRAY_SIZE; j++)

292: 		a[j] = 2.0E0 * a[j];

0x2033c0 PUSH	%RBP

0x2033c1 MOV	%RSP,%RBP

0x2033c4 PUSH	%R14

0x2033c6 PUSH	%RBX

0x2033c7 SUB	$0x20,%RSP

0x2033cb MOV	(%RDI),%R14D

0x2033ce LEA	0x31cb(%RIP),%RBX

0x2033d5 MOVQ	$0,-0x28(%RBP)

0x2033dd MOVQ	$0x3344ffff,-0x18(%RBP)

0x2033e5 MOVQ	$0x1,-0x30(%RBP)

0x2033ed MOVL	$0,-0x1c(%RBP)

0x2033f4 SUB	$0x8,%RSP

0x2033f8 MOV	$0x1100,%RDI

0x203402 LEA	-0x30(%RBP),%RAX

0x203406 LEA	-0x1c(%RBP),%RCX

0x20340a LEA	-0x28(%RBP),%R8

0x20340e LEA	-0x18(%RBP),%R9

0x203412 MOV	%R14D,%ESI

0x203415 MOV	$0x22,%EDX

0x20341a ADD	%RBX,%RDI

0x20341d PUSH	$0x1

0x20341f PUSH	$0x1

0x203421 PUSH	%RAX

0x203422 CALL	2043a0 <@plt_start@+0x80>

0x203427 ADD	$0x20,%RSP

0x20342b MOV	-0x18(%RBP),%RCX

0x20342f MOV	$0x3344ffff,%EAX

0x203434 CMP	$0x3344ffff,%RCX

0x20343b CMOVL	%RCX,%RAX

0x20343f MOV	-0x28(%RBP),%RCX

0x203443 MOV	%RAX,-0x18(%RBP)

0x203447 CMP	%RAX,%RCX

0x20344a JG	2034f8

0x203450 MOV	%RAX,%RSI

0x203453 SUB	%RCX,%RSI

0x203456 MOV	$0x1190,%RDX

0x203460 INC	%RSI

0x203463 CMP	$0x8,%RSI

0x203467 JB	2034d4

0x203469 LEA	(%RBX,%RDX,1),%R8

0x20346d MOV	%RSI,%RDI

0x203470 AND	$-0x8,%RDI

0x203474 XOR	%R9D,%R9D

0x203477 LEA	0x30(%R8,%RCX,8),%R8

0x20347c ADD	%RDI,%RCX

0x20347f NOP

(7) 0x203480 VMOVUPD	-0x30(%R8,%R9,8),%XMM0

(7) 0x203487 VMOVUPD	-0x20(%R8,%R9,8),%XMM1

(7) 0x20348e VMOVUPD	-0x10(%R8,%R9,8),%XMM2

(7) 0x203495 VMOVUPD	(%R8,%R9,8),%XMM3

(7) 0x20349b VADDPD	%XMM0,%XMM0,%XMM0

(7) 0x20349f VADDPD	%XMM1,%XMM1,%XMM1

(7) 0x2034a3 VADDPD	%XMM2,%XMM2,%XMM2

(7) 0x2034a7 VADDPD	%XMM3,%XMM3,%XMM3

(7) 0x2034ab VMOVUPD	%XMM0,-0x30(%R8,%R9,8)

(7) 0x2034b2 VMOVUPD	%XMM1,-0x20(%R8,%R9,8)

(7) 0x2034b9 VMOVUPD	%XMM2,-0x10(%R8,%R9,8)

(7) 0x2034c0 VMOVUPD	%XMM3,(%R8,%R9,8)

(7) 0x2034c6 ADD	$0x8,%R9

(7) 0x2034ca CMP	%R9,%RDI

(7) 0x2034cd JNE	203480

0x2034cf CMP	%RDI,%RSI

0x2034d2 JE	2034f8

0x2034d4 DEC	%RCX

0x2034d7 ADD	%RBX,%RDX

0x2034da NOPW	(%RAX,%RAX,1)

(6) 0x2034e0 VMOVSD	0x8(%RDX,%RCX,8),%XMM0

(6) 0x2034e6 VADDSD	%XMM0,%XMM0,%XMM0

(6) 0x2034ea VMOVSD	%XMM0,0x8(%RDX,%RCX,8)

(6) 0x2034f0 INC	%RCX

(6) 0x2034f3 CMP	%RAX,%RCX

(6) 0x2034f6 JL	2034e0

0x2034f8 MOV	$0x1118,%RAX

0x203502 MOV	%R14D,%ESI

0x203505 ADD	%RAX,%RBX

0x203508 MOV	%RBX,%RDI

0x20350b CALL	2043c0 <@plt_start@+0xa0>

0x203510 ADD	$0x20,%RSP

0x203514 POP	%RBX

0x203515 POP	%R14

0x203517 POP	%RBP

0x203518 RET

0x203519 NOPL	(%RAX)

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

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

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

Coverage (%)	Name	Source Location	Module
►99.41+	__kmp_invoke_microtask		libomp.so
○	__kmp_invoke_task_func		libomp.so

Coverage (%)	Name	Source Location	Module
►99.25+	__kmp_invoke_microtask		libomp.so
○	__kmp_invoke_task_func		libomp.so

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

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 /

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

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run 1x8

Source file and lines	stream.c:290-292
Module	exec

The function is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/build/stream-triad/src/stream.c:290-292.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Your function is not vectorized. Only 9% of vector register length is used (average across all SSE/AVX instructions). By vectorizing your function, you can lower the cost of an iteration from 10.33 to 0.72 cycles (14.38x 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 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)

Execution units bottlenecks

Found no such bottlenecks but see expert reports for more complex bottlenecks.

No data for this section

CALL instructions

Detected function call instructions.

Details

Calling (and then returning from) a function prevents many compiler optimizations (like vectorization), breaks control flow (which reduces pipeline performance) and executes extra instructions to save/restore the registers used inside it, which is very expensive (dozens of cycles). Consider to inline small functions.

unknown: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/build/stream-triad/src/stream.c:290
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/build/stream-triad/src/stream.c:290

Complex instructions

Detected COMPLEX INSTRUCTIONS.

Details

These instructions generate more than one micro-operation and only one of them can be decoded during a cycle and the extra micro-operations increase pressure on execution units.

LEA: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/build/stream-triad/src/stream.c:290

Type of elements and instruction set

No instructions are processing arithmetic or math operations on FP elements. This function is probably writing/copying data or processing integer elements.

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 20 bytes. The binary function is storing 36 bytes.

General properties

nb instructions	62
nb uops	62
loop length	249
used x86 registers	11
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	4

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	10.33 cycles
front end	10.33 cycles

Back-end

	ALU0/BRU0	ALU1	ALU2	ALU3	BRU1	AGU0	AGU1	AGU2	FP0	FP1	FP2	FP3	FP4	FP5
uops	8.00	8.00	8.00	8.00	3.00	3.67	3.67	3.67	0.00	0.00	0.00	0.00	0.00	0.00
cycles	8.00	8.00	8.00	8.00	3.00	3.67	3.67	3.67	0.00	0.00	0.00	0.00	0.00	0.00

Execution ports to units layout:

ALU0/BRU0: ALU, BRU
ALU1: ALU
ALU2: ALU
ALU3: ALU
BRU1: BRU
AGU0 (256 bits): store address, load
AGU1 (256 bits): store address, load
AGU2 (256 bits): store address, load
FP0 (256 bits): VPU, DIV/SQRT
FP1 (256 bits): VPU, DIV/SQRT
FP2 (256 bits): VPU
FP3 (256 bits): VPU
FP4 (256 bits): FP store data
FP5 (256 bits): FP store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	10.33
Dispatch	8.00
Overall L1	10.33

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	0%

Vector efficiency ratios

all	9%
load	6%
store	7%
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	10%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 10.33 cycles. At this rate:

2% of peak load performance is reached (1.94 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (3.48 out of 64.00 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). By removing all these bottlenecks, you can lower the cost of an iteration from 10.33 to 8.00 cycles (1.29x speedup).

ASM code

In the binary file, the address of the function is: 2033c0

Instruction	Nb FU	ALU0/BRU0	ALU1	ALU2	ALU3	BRU1	AGU0	AGU1	AGU2	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0	0	0	0	1	0.50	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0.17	N/A
PUSH %R14	1	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RBX	1	0	0	0	0	0	0	0	0	1	0.50	N/A
SUB $0x20,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV (%RDI),%R14D	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33	scal (6.3%)
LEA 0x31cb(%RIP),%RBX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOVQ $0,-0x28(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50	scal (6.3%)
MOVQ $0x3344ffff,-0x18(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50	scal (6.3%)
MOVQ $0x1,-0x30(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50	scal (6.3%)
MOVL $0,-0x1c(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50	scal (6.3%)
SUB $0x8,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV $0x1100,%RDI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.33	N/A
LEA -0x30(%RBP),%RAX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
LEA -0x1c(%RBP),%RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
LEA -0x28(%RBP),%R8	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
LEA -0x18(%RBP),%R9	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV %R14D,%ESI	1	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
MOV $0x22,%EDX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
ADD %RBX,%RDI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
PUSH $0x1	1	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH $0x1	1	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RAX	1	0	0	0	0	0	0	0	0	1	0.50	N/A
CALL 2043a0 <@plt_start@+0x80>	2	0.50	0	0	0	0.50	0.33	0.33	0.33	0	0.50	N/A
ADD $0x20,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV -0x18(%RBP),%RCX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33	N/A
MOV $0x3344ffff,%EAX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
CMP $0x3344ffff,%RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
CMOVL %RCX,%RAX	1	0.50	0	0	0.50	0	0	0	0	1	0.50	N/A
MOV -0x28(%RBP),%RCX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33	N/A
MOV %RAX,-0x18(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50	scal (12.5%)
CMP %RAX,%RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
JG 2034f8 <main.omp_outlined.18+0x138>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1	N/A
MOV %RAX,%RSI	1	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
SUB %RCX,%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
MOV $0x1190,%RDX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.33	N/A
INC %RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
CMP $0x8,%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
JB 2034d4 <main.omp_outlined.18+0x114>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1	N/A
LEA (%RBX,%RDX,1),%R8	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV %RSI,%RDI	1	0	0	0	0	0	0	0	0	0	0.17	N/A
AND $-0x8,%RDI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
XOR %R9D,%R9D	1	0	0	0	0	0	0	0	0	0	0.25	scal (6.3%)
LEA 0x30(%R8,%RCX,8),%R8	2	0	0	0.50	0.50	0	0.33	0.33	0.33	2	0.50	N/A
ADD %RDI,%RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
NOP	0	0	0	0	0	0	0	0	0	0	0.09	N/A
CMP %RDI,%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
JE 2034f8 <main.omp_outlined.18+0x138>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1	N/A
DEC %RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
ADD %RBX,%RDX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
NOPW (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0.09	N/A
MOV $0x1118,%RAX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.33	N/A
MOV %R14D,%ESI	1	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
ADD %RAX,%RBX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV %RBX,%RDI	1	0	0	0	0	0	0	0	0	0	0.17	N/A
CALL 2043c0 <@plt_start@+0xa0>	2	0.50	0	0	0	0.50	0.33	0.33	0.33	0	0.50	N/A
ADD $0x20,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R14	1	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	1	0.33	N/A
RET	1	0.50	0	0	0	0.50	0	0	0	1	0.50	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0.09	N/A

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run 1x8

Source file and lines	stream.c:290-292
Module	exec

gain
potential
hint
expert

Vectorization

Details

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)

Execution units bottlenecks

Found no such bottlenecks but see expert reports for more complex bottlenecks.

No data for this section

CALL instructions

Detected function call instructions.

Details

unknown: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/build/stream-triad/src/stream.c:290
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/build/stream-triad/src/stream.c:290

Complex instructions

Detected COMPLEX INSTRUCTIONS.

Details

These instructions generate more than one micro-operation and only one of them can be decoded during a cycle and the extra micro-operations increase pressure on execution units.

LEA: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/build/stream-triad/src/stream.c:290

Type of elements and instruction set

No instructions are processing arithmetic or math operations on FP elements. This function is probably writing/copying data or processing integer elements.

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 20 bytes. The binary function is storing 36 bytes.

General properties

nb instructions	62
nb uops	62
loop length	249
used x86 registers	11
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	4

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	10.33 cycles
front end	10.33 cycles

Back-end

	ALU0/BRU0	ALU1	ALU2	ALU3	BRU1	AGU0	AGU1	AGU2	FP0	FP1	FP2	FP3	FP4	FP5
uops	8.00	8.00	8.00	8.00	3.00	3.67	3.67	3.67	0.00	0.00	0.00	0.00	0.00	0.00
cycles	8.00	8.00	8.00	8.00	3.00	3.67	3.67	3.67	0.00	0.00	0.00	0.00	0.00	0.00

Execution ports to units layout:

ALU0/BRU0: ALU, BRU
ALU1: ALU
ALU2: ALU
ALU3: ALU
BRU1: BRU
AGU0 (256 bits): store address, load
AGU1 (256 bits): store address, load
AGU2 (256 bits): store address, load
FP0 (256 bits): VPU, DIV/SQRT
FP1 (256 bits): VPU, DIV/SQRT
FP2 (256 bits): VPU
FP3 (256 bits): VPU
FP4 (256 bits): FP store data
FP5 (256 bits): FP store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	10.33
Dispatch	8.00
Overall L1	10.33

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	0%

Vector efficiency ratios

all	9%
load	6%
store	7%
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	10%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 10.33 cycles. At this rate:

2% of peak load performance is reached (1.94 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (3.48 out of 64.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

ASM code

In the binary file, the address of the function is: 2033c0

Instruction	Nb FU	ALU0/BRU0	ALU1	ALU2	ALU3	BRU1	AGU0	AGU1	AGU2	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0	0	0	0	1	0.50	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0.17	N/A
PUSH %R14	1	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RBX	1	0	0	0	0	0	0	0	0	1	0.50	N/A
SUB $0x20,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV (%RDI),%R14D	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33	scal (6.3%)
LEA 0x31cb(%RIP),%RBX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOVQ $0,-0x28(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50	scal (6.3%)
MOVQ $0x3344ffff,-0x18(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50	scal (6.3%)
MOVQ $0x1,-0x30(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50	scal (6.3%)
MOVL $0,-0x1c(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50	scal (6.3%)
SUB $0x8,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV $0x1100,%RDI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.33	N/A
LEA -0x30(%RBP),%RAX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
LEA -0x1c(%RBP),%RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
LEA -0x28(%RBP),%R8	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
LEA -0x18(%RBP),%R9	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV %R14D,%ESI	1	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
MOV $0x22,%EDX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
ADD %RBX,%RDI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
PUSH $0x1	1	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH $0x1	1	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RAX	1	0	0	0	0	0	0	0	0	1	0.50	N/A
CALL 2043a0 <@plt_start@+0x80>	2	0.50	0	0	0	0.50	0.33	0.33	0.33	0	0.50	N/A
ADD $0x20,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV -0x18(%RBP),%RCX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33	N/A
MOV $0x3344ffff,%EAX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
CMP $0x3344ffff,%RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
CMOVL %RCX,%RAX	1	0.50	0	0	0.50	0	0	0	0	1	0.50	N/A
MOV -0x28(%RBP),%RCX	1	0	0	0	0	0	0.33	0.33	0.33	3	0.33	N/A
MOV %RAX,-0x18(%RBP)	1	0	0	0	0	0	0.33	0.33	0.33	4	0.50	scal (12.5%)
CMP %RAX,%RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
JG 2034f8 <main.omp_outlined.18+0x138>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1	N/A
MOV %RAX,%RSI	1	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
SUB %RCX,%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
MOV $0x1190,%RDX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.33	N/A
INC %RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
CMP $0x8,%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
JB 2034d4 <main.omp_outlined.18+0x114>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1	N/A
LEA (%RBX,%RDX,1),%R8	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV %RSI,%RDI	1	0	0	0	0	0	0	0	0	0	0.17	N/A
AND $-0x8,%RDI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
XOR %R9D,%R9D	1	0	0	0	0	0	0	0	0	0	0.25	scal (6.3%)
LEA 0x30(%R8,%RCX,8),%R8	2	0	0	0.50	0.50	0	0.33	0.33	0.33	2	0.50	N/A
ADD %RDI,%RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
NOP	0	0	0	0	0	0	0	0	0	0	0.09	N/A
CMP %RDI,%RSI	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
JE 2034f8 <main.omp_outlined.18+0x138>	1	0.50	0	0	0	0.50	0	0	0	1	0.50-1	N/A
DEC %RCX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
ADD %RBX,%RDX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
NOPW (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0.09	N/A
MOV $0x1118,%RAX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.33	N/A
MOV %R14D,%ESI	1	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
ADD %RAX,%RBX	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
MOV %RBX,%RDI	1	0	0	0	0	0	0	0	0	0	0.17	N/A
CALL 2043c0 <@plt_start@+0xa0>	2	0.50	0	0	0	0.50	0.33	0.33	0.33	0	0.50	N/A
ADD $0x20,%RSP	1	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R14	1	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	1	0.33	N/A
RET	1	0.50	0	0	0	0.50	0	0	0	1	0.50	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0.09	N/A

▶Scalability Runs Description

Run 1x8	Number processes: 1Number nodes: NANumber processes per node: 1Run Command: <executable>MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Dataset: Run Directory: /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/run/oneview_runs/multicore/aocc_4/oneview_run_1773741069OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threadsOMP_NUM_THREADS: 8
Run 1x64	Number processes: 1Number processes per node: 1Run Command: <executable>MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Dataset: Run Directory: /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/run/oneview_runs/multicore/aocc_4/oneview_run_1773741069OMP_NUM_THREADS: 64OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x96	Number processes: 1Number processes per node: 1Run Command: <executable>MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Dataset: Run Directory: /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/run/oneview_runs/multicore/aocc_4/oneview_run_1773741069OMP_NUM_THREADS: 96OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x128	Number processes: 1Number processes per node: 1Run Command: <executable>MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Dataset: Run Directory: /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/run/oneview_runs/multicore/aocc_4/oneview_run_1773741069OMP_NUM_THREADS: 128OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x160	Number processes: 1Number processes per node: 1Run Command: <executable>MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Dataset: Run Directory: /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/run/oneview_runs/multicore/aocc_4/oneview_run_1773741069OMP_NUM_THREADS: 160OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads
Run 1x192	Number processes: 1Number processes per node: 1Run Command: <executable>MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Dataset: Run Directory: /beegfs/hackathon/users/eoseret/qaas_runs_test/gpu04sas.benchmarkcenter.megware.com/177-374-0012/stream-triad/run/oneview_runs/multicore/aocc_4/oneview_run_1773741069OMP_NUM_THREADS: 192OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

(1x8) Efficiency	(1x8) Potential Speed-Up (%)	(1x64) Efficiency	(1x64) Potential Speed-Up (%)	(1x96) Efficiency	(1x96) Potential Speed-Up (%)	(1x128) Efficiency	(1x128) Potential Speed-Up (%)	(1x160) Efficiency	(1x160) Potential Speed-Up (%)	(1x192) Efficiency	(1x192) Potential Speed-Up (%)
1	0	0.29	0.28	0.19	0.33	0.11	0.46	0.1	0.39	0.07	0.5

Speed-Up and Efficiency

▶Detailed Speed-Up and Efficiency

Run	Number of threads	Efficiency (ideal is 1)	Speedup	Ideal Speedup	Time (s)	Coverage (%)
1x8	8	1	1	1	0.035000000149012	0.4823012650013
1x64	64	0.29	2.29	8	0.015000000596046	0.39148518443108
1x96	96	0.19	2.26	12	0.019999999552965	0.40777167677879
1x128	128	0.11	1.7	16	0.02000000141561	0.51430183649063
1x160	160	0.1	2.01	20	0.02000000141561	0.4315759241581
1x192	192	0.07	1.67	24	0.02000000141561	0.53748768568039

Name	Coverage (%)	Time (s)
▼main.omp_outlined.18–	0.48	0.04
○Loop 7 - stream.c:292-292 - exec	0.48	0.04
○Loop 6 - stream.c:291-292 - exec	0.00	0.00

Report Configuration

Vectorization

Details

Workaround

Execution units bottlenecks

CALL instructions

Details

Complex instructions

Details

Type of elements and instruction set

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

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

CALL instructions

Details

Complex instructions

Details

Type of elements and instruction set

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

▶Scalability Runs Description

▼Columns Filter

Speed-Up and Efficiency

▶Detailed Speed-Up and Efficiency