OV - exec - Function main.extracted.65

	Function: main.extracted.65	Module: exec	Source: stream.c:354-356	Coverage (incl. loops): 27.54% \| (excl. loops): 0.00%

/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-365-2841/stream/build/stream/src/stream.c: 354 - 356

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

354: #pragma omp parallel for

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

356: 	    a[j] = b[j]+scalar*c[j];

0x403610 PUSH	%RBP

0x403611 MOV	%RSP,%RBP

0x403614 PUSH	%R15

0x403616 PUSH	%R14

0x403618 PUSH	%RBX

0x403619 SUB	$0x28,%RSP

0x40361d MOVL	$0,-0x1c(%RBP)

0x403624 MOV	(%RDI),%EBX

0x403626 MOVQ	$0,-0x30(%RBP)

0x40362e MOVQ	$0x3344ffff,-0x28(%RBP)

0x403636 MOVQ	$0x1,-0x38(%RBP)

0x40363e SUB	$0x8,%RSP

0x403642 LEA	-0x38(%RBP),%RAX

0x403646 LEA	-0x30(%RBP),%R8

0x40364a MOV	$0x4ceb89240,%RDI

0x403654 MOV	$0x401050,%R10

0x40365e LEA	-0x1c(%RBP),%RCX

0x403662 LEA	-0x28(%RBP),%R9

0x403666 MOV	%EBX,%ESI

0x403668 MOV	$0x22,%EDX

0x40366d PUSH	$0x1

0x40366f PUSH	$0x1

0x403671 PUSH	%RAX

0x403672 CALL	%R10

0x403675 ADD	$0x20,%RSP

0x403679 MOV	-0x30(%RBP),%R8

0x40367d MOV	-0x28(%RBP),%RAX

0x403681 CMP	%RAX,%R8

0x403684 JA	40373e

0x40368a MOV	%RAX,%R10

0x40368d SUB	%R8,%R10

0x403690 MOV	$0x334906060,%RDX

0x40369a MOV	$0x19a686060,%RSI

0x4036a4 MOV	$0x4040d0,%R9

0x4036ae MOV	$0x406060,%RDI

0x4036b8 INC	%R10

0x4036bb MOV	%R10,%RCX

0x4036be AND	$0x7ffffffc,%RCX

0x4036c5 JE	403706

0x4036c7 VBROADCASTSD	(%R9),%YMM0

0x4036cc LEA	(,%R8,8),%R11

0x4036d4 XOR	%R14D,%R14D

0x4036d7 NOPW	(%RAX,%RAX,1)

(22) 0x4036e0 LEA	(%R11,%R14,8),%R15

(22) 0x4036e4 ADD	$0x4,%R14

(22) 0x4036e8 VMOVUPD	(%RDX,%R15,1),%YMM1

(22) 0x4036ee VFMADD213PD	(%RSI,%R15,1),%YMM0,%YMM1

(22) 0x4036f4 VMOVUPD	%YMM1,(%RDI,%R15,1)

(22) 0x4036fa CMP	%RCX,%R14

(22) 0x4036fd JB	4036e0

0x4036ff CMP	%RCX,%R10

0x403702 JNE	403708

0x403704 JMP	40373e

0x403706 XOR	%ECX,%ECX

0x403708 VMOVSD	(%R9),%XMM0

0x40370d SUB	%RCX,%RAX

0x403710 ADD	%R8,%RCX

0x403713 SUB	%R8,%RAX

0x403716 SAL	$0x3,%RCX

0x40371a XOR	%R8D,%R8D

0x40371d INC	%RAX

(21) 0x403720 LEA	(%RCX,%R8,8),%R9

(21) 0x403724 INC	%R8

(21) 0x403727 VMOVSD	(%RDX,%R9,1),%XMM1

(21) 0x40372d VFMADD213SD	(%RSI,%R9,1),%XMM0,%XMM1

(21) 0x403733 VMOVSD	%XMM1,(%RDI,%R9,1)

(21) 0x403739 CMP	%R8,%RAX

(21) 0x40373c JNE	403720

0x40373e MOV	$0x4ceb89260,%RDI

0x403748 MOV	$0x401030,%RAX

0x403752 MOV	%EBX,%ESI

0x403754 ADD	$0x28,%RSP

0x403758 POP	%RBX

0x403759 POP	%R14

0x40375b POP	%R15

0x40375d POP	%RBP

0x40375e VZEROUPPER

0x403761 JMP	%RAX

Coverage (%)	Name	Source Location	Module
►100.00+	__kmp_invoke_microtask		libiomp5.so
○	__kmp_invoke_task_func		libiomp5.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 icx_4

Source file and lines	stream.c:354-356
Module	exec

The function is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-365-2841/stream/build/stream/src/stream.c:354-356.

Warnings:
Ignoring paths for analysis

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 8.00 to 1.25 cycles (6.40x speedup).

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)

Vectorization

Your function is probably not vectorized. Only 10% of vector register length is used (average across all SSE/AVX instructions).

Details

Store and arithmetical SSE/AVX instructions are used in scalar version (process only one data element in vector registers).

Workaround

Read the "512-bits vectorization" report at "Potential" confidence level.

Execution units bottlenecks

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

512-bits vectorization

On some x86 processors supporting 512-bits vectorization, compilers are often too conservative and limit vectorization to 256 bits. Performance can then be improved by enforcing 512-bits vectorization, especially with many vectorized and high trip count loops. 512-bits vectorization performance overhead (compared to 256-bits) is generally lower on newer processors.

Workaround

Recompile with -mprefer-vector-width=512

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: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-365-2841/stream/build/stream/src/stream.c:354

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 36 bytes. The binary function is storing 28 bytes.

General properties

nb instructions	64
nb uops	64
loop length	278
used x86 registers	14
used mmx registers	0
used xmm registers	1
used ymm registers	1
used zmm registers	0
nb stack references	4

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	8.00 cycles
front end	8.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	5.67	5.67	5.67	5.67	5.67	5.67	2.50	2.50	2.50	2.50	0.00	0.00	0.00	0.00	0.00	0.00
cycles	5.67	5.67	5.67	5.67	5.67	5.67	2.50	2.50	2.50	2.50	0.00	0.00	0.00	0.00	0.00	0.00

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

Cycles summary

Front-end	8.00
Dispatch	5.67
Overall L1	8.00

Vectorization ratios

INT

all	4%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
other	6%

all	0%
load	0%
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

INT+FP

all	4%
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	6%

Vector efficiency ratios

INT

all	10%
load	6%
store	6%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
other	11%

all	12%
load	12%
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	12%

INT+FP

all	10%
load	10%
store	6%
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	11%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 8.00 cycles. At this rate:

1% of peak load performance is reached (4.50 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (3.50 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). By removing all these bottlenecks, you can lower the cost of an iteration from 8.00 to 5.67 cycles (1.41x speedup).

ASM code

In the binary file, the address of the function is: 403610

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	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	0	0.17	N/A
PUSH %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RBX	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SUB $0x28,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOVL $0,-0x1c(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOV (%RDI),%EBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (6.3%)
MOVQ $0,-0x30(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVQ $0x3344ffff,-0x28(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVQ $0x1,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
SUB $0x8,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA -0x38(%RBP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA -0x30(%RBP),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x4ceb89240,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x401050,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
LEA -0x1c(%RBP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA -0x28(%RBP),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %EBX,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
MOV $0x22,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
PUSH $0x1	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH $0x1	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RAX	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CALL %R10	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	2	N/A
ADD $0x20,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV -0x30(%RBP),%R8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV -0x28(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
CMP %RAX,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JA 40373e <main.extracted.65+0x12e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
MOV %RAX,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
SUB %R8,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV $0x334906060,%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV $0x19a686060,%RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV $0x4040d0,%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x406060,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
INC %R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV %R10,%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
AND $0x7ffffffc,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JE 403706 <main.extracted.65+0xf6>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
VBROADCASTSD (%R9),%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
LEA (,%R8,8),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
XOR %R14D,%R14D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
NOPW (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
CMP %RCX,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JNE 403708 <main.extracted.65+0xf8>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
JMP 40373e <main.extracted.65+0x12e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	N/A
XOR %ECX,%ECX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
VMOVSD (%R9),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	1	0.50	scal (12.5%)
SUB %RCX,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ADD %R8,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
SUB %R8,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SAL $0x3,%RCX	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.33	scal (12.5%)
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
INC %RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x4ceb89260,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x401030,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %EBX,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
ADD $0x28,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
JMP %RAX	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run icx_4

Source file and lines	stream.c:354-356
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 8.00 to 1.25 cycles (6.40x speedup).

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)

Vectorization

Your function is probably not vectorized. Only 10% of vector register length is used (average across all SSE/AVX instructions).

Details

Store and arithmetical SSE/AVX instructions are used in scalar version (process only one data element in vector registers).

Workaround

Read the "512-bits vectorization" report at "Potential" confidence level.

Execution units bottlenecks

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

512-bits vectorization

Workaround

Recompile with -mprefer-vector-width=512

CALL instructions

Detected function call instructions.

Details

unknown: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-365-2841/stream/build/stream/src/stream.c:354

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 36 bytes. The binary function is storing 28 bytes.

General properties

nb instructions	64
nb uops	64
loop length	278
used x86 registers	14
used mmx registers	0
used xmm registers	1
used ymm registers	1
used zmm registers	0
nb stack references	4

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	8.00 cycles
front end	8.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	5.67	5.67	5.67	5.67	5.67	5.67	2.50	2.50	2.50	2.50	0.00	0.00	0.00	0.00	0.00	0.00
cycles	5.67	5.67	5.67	5.67	5.67	5.67	2.50	2.50	2.50	2.50	0.00	0.00	0.00	0.00	0.00	0.00

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

Cycles summary

Front-end	8.00
Dispatch	5.67
Overall L1	8.00

Vectorization ratios

INT

all	4%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
other	6%

all	0%
load	0%
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

INT+FP

all	4%
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	6%

Vector efficiency ratios

INT

all	10%
load	6%
store	6%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
other	11%

all	12%
load	12%
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	12%

INT+FP

all	10%
load	10%
store	6%
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	11%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 8.00 cycles. At this rate:

1% of peak load performance is reached (4.50 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (3.50 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

ASM code

In the binary file, the address of the function is: 403610

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	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	0	0.17	N/A
PUSH %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RBX	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SUB $0x28,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOVL $0,-0x1c(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOV (%RDI),%EBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (6.3%)
MOVQ $0,-0x30(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVQ $0x3344ffff,-0x28(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVQ $0x1,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
SUB $0x8,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA -0x38(%RBP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA -0x30(%RBP),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x4ceb89240,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x401050,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
LEA -0x1c(%RBP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA -0x28(%RBP),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %EBX,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
MOV $0x22,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
PUSH $0x1	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH $0x1	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RAX	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CALL %R10	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	2	N/A
ADD $0x20,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV -0x30(%RBP),%R8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV -0x28(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
CMP %RAX,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JA 40373e <main.extracted.65+0x12e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
MOV %RAX,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
SUB %R8,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV $0x334906060,%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV $0x19a686060,%RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV $0x4040d0,%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x406060,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
INC %R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV %R10,%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
AND $0x7ffffffc,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JE 403706 <main.extracted.65+0xf6>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
VBROADCASTSD (%R9),%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
LEA (,%R8,8),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
XOR %R14D,%R14D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
NOPW (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
CMP %RCX,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JNE 403708 <main.extracted.65+0xf8>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
JMP 40373e <main.extracted.65+0x12e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	N/A
XOR %ECX,%ECX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
VMOVSD (%R9),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	1	0.50	scal (12.5%)
SUB %RCX,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ADD %R8,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
SUB %R8,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SAL $0x3,%RCX	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.33	scal (12.5%)
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
INC %RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x4ceb89260,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x401030,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %EBX,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
ADD $0x28,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
JMP %RAX	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A

Name	Coverage (%)	Time (s)
▼main.extracted.65–	27.54	5.50
○Loop 22 - stream.c:355-356 - exec	27.54	5.41
○Loop 21 - stream.c:355-356 - exec	0.00	0.00

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

512-bits vectorization

Workaround

CALL 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

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

512-bits vectorization

Workaround

CALL 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