OV - exec - Function main._omp

	Function: main._omp_fn.3	Module: exec	Source: stream.c:290-292	Coverage (incl. loops): 0.66% \| (excl. loops): 0.00%

/beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/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];

0x401f20 PUSH	%RBP

0x401f21 MOV	$0x401060,%RAX

0x401f2b MOV	%RSP,%RBP

0x401f2e PUSH	%RBX

0x401f2f AND	$-0x20,%RSP

0x401f33 CALL	%RAX

0x401f35 MOVSXD	%EAX,%RBX

0x401f38 MOV	$0x401030,%RAX

0x401f42 CALL	%RAX

0x401f44 MOVSXD	%EAX,%RCX

0x401f47 MOV	$0x33450000,%EAX

0x401f4c CQTO

0x401f4e IDIV	%RBX

0x401f51 CMP	%RDX,%RCX

0x401f54 JL	402168

0x401f5a IMUL	%RAX,%RCX

0x401f5e ADD	%RCX,%RDX

0x401f61 LEA	(%RAX,%RDX,1),%RSI

0x401f65 CMP	%RSI,%RDX

0x401f68 JGE	40215d

0x401f6e LEA	-0x1(%RAX),%RDI

0x401f72 MOV	%RDX,%R9

0x401f75 CMP	$0x2,%RDI

0x401f79 JBE	402178

0x401f7f MOV	%RAX,%RBX

0x401f82 MOV	$0x3349060e0,%R10

0x401f8c SHR	$0x2,%RBX

0x401f90 SAL	$0x5,%RBX

0x401f94 LEA	(%R10,%RDX,8),%R11

0x401f98 LEA	(%RBX,%R11,1),%R8

0x401f9c SUB	$0x20,%RBX

0x401fa0 SHR	$0x5,%RBX

0x401fa4 INC	%RBX

0x401fa7 AND	$0x7,%EBX

0x401faa JE	402069

0x401fb0 CMP	$0x1,%RBX

0x401fb4 JE	40204c

0x401fba CMP	$0x2,%RBX

0x401fbe JE	402038

0x401fc0 CMP	$0x3,%RBX

0x401fc4 JE	402024

0x401fc6 CMP	$0x4,%RBX

0x401fca JE	402011

0x401fcc CMP	$0x5,%RBX

0x401fd0 JE	401ffe

0x401fd2 CMP	$0x6,%RBX

0x401fd6 JE	401feb

0x401fd8 VMOVUPD	(%R11),%YMM0

0x401fdd ADD	$0x20,%R11

0x401fe1 VADDPD	%YMM0,%YMM0,%YMM1

0x401fe5 VMOVUPD	%YMM1,-0x20(%R11)

0x401feb VMOVUPD	(%R11),%YMM2

0x401ff0 ADD	$0x20,%R11

0x401ff4 VADDPD	%YMM2,%YMM2,%YMM3

0x401ff8 VMOVUPD	%YMM3,-0x20(%R11)

0x401ffe VMOVUPD	(%R11),%YMM4

0x402003 ADD	$0x20,%R11

0x402007 VADDPD	%YMM4,%YMM4,%YMM5

0x40200b VMOVUPD	%YMM5,-0x20(%R11)

0x402011 VMOVUPD	(%R11),%YMM6

0x402016 ADD	$0x20,%R11

0x40201a VADDPD	%YMM6,%YMM6,%YMM7

0x40201e VMOVUPD	%YMM7,-0x20(%R11)

0x402024 VMOVUPD	(%R11),%YMM8

0x402029 ADD	$0x20,%R11

0x40202d VADDPD	%YMM8,%YMM8,%YMM9

0x402032 VMOVUPD	%YMM9,-0x20(%R11)

0x402038 VMOVUPD	(%R11),%YMM10

0x40203d ADD	$0x20,%R11

0x402041 VADDPD	%YMM10,%YMM10,%YMM11

0x402046 VMOVUPD	%YMM11,-0x20(%R11)

0x40204c VMOVUPD	(%R11),%YMM12

0x402051 ADD	$0x20,%R11

0x402055 VADDPD	%YMM12,%YMM12,%YMM13

0x40205a VMOVUPD	%YMM13,-0x20(%R11)

0x402060 CMP	%R8,%R11

0x402063 JE	402113

(4) 0x402069 VMOVUPD	(%R11),%YMM14

(4) 0x40206e VMOVUPD	0x20(%R11),%YMM0

(4) 0x402074 VMOVUPD	0x40(%R11),%YMM2

(4) 0x40207a VMOVUPD	0x60(%R11),%YMM4

(4) 0x402080 VMOVUPD	0x80(%R11),%YMM6

(4) 0x402089 VMOVUPD	0xa0(%R11),%YMM8

(4) 0x402092 VMOVUPD	0xc0(%R11),%YMM10

(4) 0x40209b VMOVUPD	0xe0(%R11),%YMM12

(4) 0x4020a4 VADDPD	%YMM14,%YMM14,%YMM15

(4) 0x4020a9 VADDPD	%YMM0,%YMM0,%YMM1

(4) 0x4020ad VADDPD	%YMM2,%YMM2,%YMM3

(4) 0x4020b1 VADDPD	%YMM4,%YMM4,%YMM5

(4) 0x4020b5 VADDPD	%YMM6,%YMM6,%YMM7

(4) 0x4020b9 VADDPD	%YMM8,%YMM8,%YMM9

(4) 0x4020be VADDPD	%YMM10,%YMM10,%YMM11

(4) 0x4020c3 VADDPD	%YMM12,%YMM12,%YMM13

(4) 0x4020c8 VMOVUPD	%YMM15,(%R11)

(4) 0x4020cd VMOVUPD	%YMM1,0x20(%R11)

(4) 0x4020d3 VMOVUPD	%YMM3,0x40(%R11)

(4) 0x4020d9 VMOVUPD	%YMM5,0x60(%R11)

(4) 0x4020df VMOVUPD	%YMM7,0x80(%R11)

(4) 0x4020e8 VMOVUPD	%YMM9,0xa0(%R11)

(4) 0x4020f1 VMOVUPD	%YMM11,0xc0(%R11)

(4) 0x4020fa VMOVUPD	%YMM13,0xe0(%R11)

(4) 0x402103 ADD	$0x100,%R11

(4) 0x40210a CMP	%R8,%R11

(4) 0x40210d JNE	402069

0x402113 MOV	%RAX,%RCX

0x402116 AND	$-0x4,%RCX

0x40211a ADD	%RCX,%RDX

0x40211d CMP	%RCX,%RAX

0x402120 JE	402190

0x402122 VZEROUPPER

0x402125 SUB	%RCX,%RAX

0x402128 CMP	$0x1,%RAX

0x40212c JE	40214d

0x40212e ADD	%R9,%RCX

0x402131 LEA	(%R10,%RCX,8),%RSI

0x402135 VMOVUPD	(%RSI),%XMM14

0x402139 VADDPD	%XMM14,%XMM14,%XMM15

0x40213e VMOVUPD	%XMM15,(%RSI)

0x402142 TEST	$0x1,%AL

0x402144 JE	40215d

0x402146 AND	$-0x2,%RAX

0x40214a ADD	%RAX,%RDX

0x40214d VMOVSD	(%R10,%RDX,8),%XMM0

0x402153 VADDSD	%XMM0,%XMM0,%XMM1

0x402157 VMOVSD	%XMM1,(%R10,%RDX,8)

0x40215d MOV	-0x8(%RBP),%RBX

0x402161 LEAVE

0x402162 RET

0x402163 NOPL	(%RAX,%RAX,1)

0x402168 INC	%RAX

0x40216b XOR	%EDX,%EDX

0x40216d JMP	401f5a

0x402172 NOPW	(%RAX,%RAX,1)

0x402178 XOR	%ECX,%ECX

0x40217a MOV	$0x3349060e0,%R10

0x402184 JMP	402125

0x402186 NOPW	%CS:(%RAX,%RAX,1)

0x402190 VZEROUPPER

0x402193 MOV	-0x8(%RBP),%RBX

0x402197 LEAVE

0x402198 RET

0x402199 NOPL	(%RAX)

Coverage (%)	Name	Source Location	Module
○99.15	gomp_thread_start	team.c:130	libgomp.so.1.0.0

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 gcc_0

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

The function is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290-292.

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 21.17 to 8.67 cycles (2.44x speedup).

Workaround

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

Vectorization

Your function is partially vectorized. Only 31% of vector register length is used (average across all SSE/AVX instructions).

Details

57% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

88% of SSE/AVX loads are used in vector version.
88% of SSE/AVX stores are used in vector version.
88% of SSE/AVX addition or subtraction instructions are used in vector version.
0% of SSE/AVX divide and square root instructions are used in vector version.
11% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

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: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/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.

IDIV: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290
LEAVE: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290
VZEROUPPER: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290

Vector unaligned load/store instructions

Detected 14 optimal vector unaligned load/store instructions.

Details

VMOVUPD: 14 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292

Workaround

Use vector aligned instructions:

align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.

Conversion instructions

Detected expensive conversion instructions.

Details

CQTO: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290

Workaround

Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.

Type of elements and instruction set

1 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in scalar mode (one at a time). 1 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (two at a time). 7 AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (four at a time).

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

The binary function is composed of 31 FP arithmetical operations:

31: addition or subtraction

The binary function is loading 264 bytes (33 double precision FP elements). The binary function is storing 248 bytes (31 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.06 FP operations per loaded or stored byte.

General properties

nb instructions	115
nb uops	127
loop length	470
used x86 registers	12
used mmx registers	0
used xmm registers	4
used ymm registers	14
used zmm registers	0
nb stack references	1

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	21.17 cycles
front end	21.17 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	14.00	14.10	5.00	5.00	6.50	13.90	14.00	6.50	6.50	6.50	14.00	5.00
cycles	14.00	14.10	5.00	5.00	6.50	13.90	14.00	6.50	6.50	6.50	14.00	5.00

Execution ports to units layout:

P0 (256 bits): VPU, BRU, ALU, DIV/SQRT
P1 (256 bits): ALU, VPU
P2 (512 bits): load
P3 (512 bits): load
P4 (512 bits): store data
P5 (512 bits): ALU, VPU
P6: ALU, BRU
P7: store address
P8: store address
P9 (512 bits): store data
P10: ALU
P11 (512 bits): load

Cycles executing div or sqrt instructions

10.00

Cycles summary

Front-end	21.17
Dispatch	14.10
DIV/SQRT	10.00
Overall L1	21.17

Vectorization ratios

INT

all	11%
load	NA (no load vectorizable/vectorized instructions)
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)
other	11%

all	88%
load	88%
store	88%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	88%
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	57%
load	88%
store	88%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	88%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	0%
other	11%

Vector efficiency ratios

INT

all	13%
load	NA (no load vectorizable/vectorized instructions)
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)
other	13%

all	43%
load	43%
store	43%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	43%
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	31%
load	43%
store	43%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	43%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	12%
other	13%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 21.17 cycles. At this rate:

6% of peak load performance is reached (12.47 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
9% of peak store performance is reached (11.72 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 21.17 to 14.10 cycles (1.50x speedup).

ASM code

In the binary file, the address of the function is: 401f20

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV $0x401060,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
PUSH %RBX	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
AND $-0x20,%RSP	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	scal (12.5%)
CALL %RAX	2	0.50	0	0	0	0.50	0	0.50	0.50	0.50	0.50	0	0	0	2.14	N/A
MOVSXD %EAX,%RBX	1	0	0.33	0	0	0	0.33	0	0	0	0	0.33	0	1	0.33	N/A
MOV $0x401030,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
CALL %RAX	2	0.50	0	0	0	0.50	0	0.50	0.50	0.50	0.50	0	0	0	2.14	N/A
MOVSXD %EAX,%RCX	1	0	0.33	0	0	0	0.33	0	0	0	0	0.33	0	1	0.33	N/A
MOV $0x33450000,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
CQTO	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	1	0.50	scal (12.5%)
IDIV %RBX	5	0	3	0	0	0	0	0	0	0	0	0	0	11-16	10	scal (12.5%)
CMP %RDX,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JL 402168 <main._omp_fn.3+0x248>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
IMUL %RAX,%RCX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD %RCX,%RDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA (%RAX,%RDX,1),%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMP %RSI,%RDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JGE 40215d <main._omp_fn.3+0x23d>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
LEA -0x1(%RAX),%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RDX,%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
CMP $0x2,%RDI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JBE 402178 <main._omp_fn.3+0x258>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %RAX,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV $0x3349060e0,%R10	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
SHR $0x2,%RBX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
SAL $0x5,%RBX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
LEA (%R10,%RDX,8),%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA (%RBX,%R11,1),%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SUB $0x20,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
SHR $0x5,%RBX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
INC %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $0x7,%EBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 402069 <main._omp_fn.3+0x149>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x1,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 40204c <main._omp_fn.3+0x12c>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x2,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 402038 <main._omp_fn.3+0x118>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x3,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 402024 <main._omp_fn.3+0x104>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x4,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 402011 <main._omp_fn.3+0xf1>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x5,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 401ffe <main._omp_fn.3+0xde>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x6,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 401feb <main._omp_fn.3+0xcb>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
VMOVUPD (%R11),%YMM0	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM0,%YMM0,%YMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM1,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM2	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM2,%YMM2,%YMM3	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM3,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM4	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM4,%YMM4,%YMM5	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM5,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM6	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM6,%YMM6,%YMM7	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM7,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM8	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM8,%YMM8,%YMM9	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM9,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM10,%YMM10,%YMM11	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM11,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM12,%YMM12,%YMM13	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM13,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
CMP %R8,%R11	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 402113 <main._omp_fn.3+0x1f3>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %RAX,%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $-0x4,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %RCX,%RDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
CMP %RCX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 402190 <main._omp_fn.3+0x270>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
VZEROUPPER	2	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
SUB %RCX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
CMP $0x1,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 40214d <main._omp_fn.3+0x22d>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
ADD %R9,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA (%R10,%RCX,8),%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VMOVUPD (%RSI),%XMM14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (25.0%)
VADDPD %XMM14,%XMM14,%XMM15	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
VMOVUPD %XMM15,(%RSI)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
TEST $0x1,%AL	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	N/A
JE 40215d <main._omp_fn.3+0x23d>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
AND $-0x2,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %RAX,%RDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
VMOVSD (%R10,%RDX,8),%XMM0	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VADDSD %XMM0,%XMM0,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
VMOVSD %XMM1,(%R10,%RDX,8)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV -0x8(%RBP),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
LEAVE	3	0.60	0.60	0.33	0.33	0	0.60	0.60	0	0	0	0.60	0.33	2-6	2	N/A
RET	1	0.50	0	0.33	0.33	0	0	0.50	0	0	0	0	0.33	0	2.13	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
INC %RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
JMP 401f5a <main._omp_fn.3+0x3a>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPW (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
XOR %ECX,%ECX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x3349060e0,%R10	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
JMP 402125 <main._omp_fn.3+0x205>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
NOPW %CS:(%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VZEROUPPER	2	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
MOV -0x8(%RBP),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
LEAVE	3	0.60	0.60	0.33	0.33	0	0.60	0.60	0	0	0	0.60	0.33	2-6	2	N/A
RET	1	0.50	0	0.33	0.33	0	0	0.50	0	0	0	0	0.33	0	2.13	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run gcc_0

Source file and lines	stream.c:290-292
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 21.17 to 8.67 cycles (2.44x speedup).

Workaround

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

Vectorization

Your function is partially vectorized. Only 31% of vector register length is used (average across all SSE/AVX instructions).

Details

57% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

88% of SSE/AVX loads are used in vector version.
88% of SSE/AVX stores are used in vector version.
88% of SSE/AVX addition or subtraction instructions are used in vector version.
0% of SSE/AVX divide and square root instructions are used in vector version.
11% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

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: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/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.

IDIV: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290
LEAVE: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290
VZEROUPPER: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290

Vector unaligned load/store instructions

Detected 14 optimal vector unaligned load/store instructions.

Details

VMOVUPD: 14 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:292

Workaround

Use vector aligned instructions:

align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.

Conversion instructions

Detected expensive conversion instructions.

Details

CQTO: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/isix03.benchmarkcenter.megware.com/177-373-7190/stream-add/build/stream-add/src/stream.c:290

Workaround

Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.

Type of elements and instruction set

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

The binary function is composed of 31 FP arithmetical operations:

31: addition or subtraction

The binary function is loading 264 bytes (33 double precision FP elements). The binary function is storing 248 bytes (31 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.06 FP operations per loaded or stored byte.

General properties

nb instructions	115
nb uops	127
loop length	470
used x86 registers	12
used mmx registers	0
used xmm registers	4
used ymm registers	14
used zmm registers	0
nb stack references	1

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	21.17 cycles
front end	21.17 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	14.00	14.10	5.00	5.00	6.50	13.90	14.00	6.50	6.50	6.50	14.00	5.00
cycles	14.00	14.10	5.00	5.00	6.50	13.90	14.00	6.50	6.50	6.50	14.00	5.00

Execution ports to units layout:

P0 (256 bits): VPU, BRU, ALU, DIV/SQRT
P1 (256 bits): ALU, VPU
P2 (512 bits): load
P3 (512 bits): load
P4 (512 bits): store data
P5 (512 bits): ALU, VPU
P6: ALU, BRU
P7: store address
P8: store address
P9 (512 bits): store data
P10: ALU
P11 (512 bits): load

Cycles executing div or sqrt instructions

10.00

Cycles summary

Front-end	21.17
Dispatch	14.10
DIV/SQRT	10.00
Overall L1	21.17

Vectorization ratios

INT

all	11%
load	NA (no load vectorizable/vectorized instructions)
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)
other	11%

all	88%
load	88%
store	88%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	88%
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	57%
load	88%
store	88%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	88%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	0%
other	11%

Vector efficiency ratios

INT

all	13%
load	NA (no load vectorizable/vectorized instructions)
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)
other	13%

all	43%
load	43%
store	43%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	43%
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	31%
load	43%
store	43%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	43%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	12%
other	13%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 21.17 cycles. At this rate:

6% of peak load performance is reached (12.47 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
9% of peak store performance is reached (11.72 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: 401f20

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV $0x401060,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
PUSH %RBX	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
AND $-0x20,%RSP	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	scal (12.5%)
CALL %RAX	2	0.50	0	0	0	0.50	0	0.50	0.50	0.50	0.50	0	0	0	2.14	N/A
MOVSXD %EAX,%RBX	1	0	0.33	0	0	0	0.33	0	0	0	0	0.33	0	1	0.33	N/A
MOV $0x401030,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
CALL %RAX	2	0.50	0	0	0	0.50	0	0.50	0.50	0.50	0.50	0	0	0	2.14	N/A
MOVSXD %EAX,%RCX	1	0	0.33	0	0	0	0.33	0	0	0	0	0.33	0	1	0.33	N/A
MOV $0x33450000,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
CQTO	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	1	0.50	scal (12.5%)
IDIV %RBX	5	0	3	0	0	0	0	0	0	0	0	0	0	11-16	10	scal (12.5%)
CMP %RDX,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JL 402168 <main._omp_fn.3+0x248>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
IMUL %RAX,%RCX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD %RCX,%RDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA (%RAX,%RDX,1),%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMP %RSI,%RDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JGE 40215d <main._omp_fn.3+0x23d>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
LEA -0x1(%RAX),%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RDX,%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
CMP $0x2,%RDI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JBE 402178 <main._omp_fn.3+0x258>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %RAX,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV $0x3349060e0,%R10	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
SHR $0x2,%RBX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
SAL $0x5,%RBX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
LEA (%R10,%RDX,8),%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA (%RBX,%R11,1),%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SUB $0x20,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
SHR $0x5,%RBX	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0-2	0.50	N/A
INC %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $0x7,%EBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
JE 402069 <main._omp_fn.3+0x149>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x1,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 40204c <main._omp_fn.3+0x12c>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x2,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 402038 <main._omp_fn.3+0x118>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x3,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 402024 <main._omp_fn.3+0x104>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x4,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 402011 <main._omp_fn.3+0xf1>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x5,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 401ffe <main._omp_fn.3+0xde>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $0x6,%RBX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 401feb <main._omp_fn.3+0xcb>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
VMOVUPD (%R11),%YMM0	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM0,%YMM0,%YMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM1,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM2	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM2,%YMM2,%YMM3	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM3,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM4	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM4,%YMM4,%YMM5	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM5,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM6	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM6,%YMM6,%YMM7	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM7,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM8	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM8,%YMM8,%YMM9	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM9,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM10,%YMM10,%YMM11	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM11,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
VMOVUPD (%R11),%YMM12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (50.0%)
ADD $0x20,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VADDPD %YMM12,%YMM12,%YMM13	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (50.0%)
VMOVUPD %YMM13,-0x20(%R11)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (50.0%)
CMP %R8,%R11	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 402113 <main._omp_fn.3+0x1f3>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %RAX,%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND $-0x4,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %RCX,%RDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
CMP %RCX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 402190 <main._omp_fn.3+0x270>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
VZEROUPPER	2	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
SUB %RCX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
CMP $0x1,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JE 40214d <main._omp_fn.3+0x22d>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
ADD %R9,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA (%R10,%RCX,8),%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VMOVUPD (%RSI),%XMM14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	0-1	0.33	vect (25.0%)
VADDPD %XMM14,%XMM14,%XMM15	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	vect (25.0%)
VMOVUPD %XMM15,(%RSI)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
TEST $0x1,%AL	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	N/A
JE 40215d <main._omp_fn.3+0x23d>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
AND $-0x2,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %RAX,%RDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
VMOVSD (%R10,%RDX,8),%XMM0	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
VADDSD %XMM0,%XMM0,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	3	0.50	scal (12.5%)
VMOVSD %XMM1,(%R10,%RDX,8)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV -0x8(%RBP),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
LEAVE	3	0.60	0.60	0.33	0.33	0	0.60	0.60	0	0	0	0.60	0.33	2-6	2	N/A
RET	1	0.50	0	0.33	0.33	0	0	0.50	0	0	0	0	0.33	0	2.13	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
INC %RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
JMP 401f5a <main._omp_fn.3+0x3a>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPW (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
XOR %ECX,%ECX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x3349060e0,%R10	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.28	N/A
JMP 402125 <main._omp_fn.3+0x205>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
NOPW %CS:(%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VZEROUPPER	2	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
MOV -0x8(%RBP),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
LEAVE	3	0.60	0.60	0.33	0.33	0	0.60	0.60	0	0	0	0.60	0.33	2-6	2	N/A
RET	1	0.50	0	0.33	0.33	0	0	0.50	0	0	0	0	0.33	0	2.13	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A

Name	Coverage (%)	Time (s)
▼main._omp_fn.3–	0.66	0.02
○Loop 4 - stream.c:292-292 - exec	0.66	0.02

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

512-bits vectorization

Workaround

CALL instructions

Details

Complex instructions

Details

Vector unaligned load/store instructions

Details

Workaround

Conversion instructions

Details

Workaround

Type of elements and instruction set

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

Arithmetic intensity

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

Complex instructions

Details

Vector unaligned load/store instructions

Details

Workaround

Conversion instructions

Details

Workaround

Type of elements and instruction set

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

Arithmetic intensity

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code