OV - exec - Function main.omp

	Function: main.omp_outlined.18	Module: exec	Source: stream.c:290-292	Coverage (incl. loops): 0.99% \| (excl. loops): 0.00%

/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/177-365-3972/stream-copy/build/stream-copy/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];

0x203530 PUSH	%RBP

0x203531 MOV	%RSP,%RBP

0x203534 PUSH	%R14

0x203536 PUSH	%RBX

0x203537 SUB	$0x20,%RSP

0x20353b MOV	(%RDI),%EBX

0x20353d LEA	-0x7(%RIP),%RAX

0x203544 MOV	$0x3143,%R14

0x20354e MOVQ	$0,-0x28(%RBP)

0x203556 MOVQ	$0x3344ffff,-0x18(%RBP)

0x20355e MOVQ	$0x1,-0x30(%RBP)

0x203566 MOVL	$0,-0x1c(%RBP)

0x20356d ADD	%RAX,%R14

0x203570 SUB	$0x8,%RSP

0x203574 MOV	$-0x1260,%RDI

0x20357e LEA	-0x30(%RBP),%RAX

0x203582 LEA	-0x1c(%RBP),%RCX

0x203586 MOV	$-0x1028,%R10

0x203590 LEA	-0x28(%RBP),%R8

0x203594 LEA	-0x18(%RBP),%R9

0x203598 MOV	%EBX,%ESI

0x20359a MOV	$0x22,%EDX

0x20359f ADD	%R14,%RDI

0x2035a2 PUSH	$0x1

0x2035a4 PUSH	$0x1

0x2035a6 PUSH	%RAX

0x2035a7 CALLQ	(%R14,%R10,1)

0x2035ab ADD	$0x20,%RSP

0x2035af MOV	-0x18(%RBP),%RCX

0x2035b3 MOV	$0x3344ffff,%EAX

0x2035b8 CMP	$0x3344ffff,%RCX

0x2035bf CMOVL	%RCX,%RAX

0x2035c3 MOV	-0x28(%RBP),%RCX

0x2035c7 MOV	%RAX,-0x18(%RBP)

0x2035cb CMP	%RAX,%RCX

0x2035ce JG	2036e8

0x2035d4 MOV	%RAX,%RSI

0x2035d7 SUB	%RCX,%RSI

0x2035da MOV	$0x30,%RDX

0x2035e4 INC	%RSI

0x2035e7 CMP	$0x4,%RSI

0x2035eb JB	2036c3

0x2035f1 CMP	$0x20,%RSI

0x2035f5 JAE	2035fe

0x2035f7 XOR	%EDI,%EDI

0x2035f9 JMP	20368a

0x2035fe LEA	(%R14,%RDX,1),%R8

0x203602 MOV	%RSI,%RDI

0x203605 AND	$-0x20,%RDI

0x203609 XOR	%R9D,%R9D

0x20360c LEA	0xc0(%R8,%RCX,8),%R8

0x203614 NOPW	%CS:(%RAX,%RAX,1)

(8) 0x203620 VMOVUPD	-0xc0(%R8,%R9,8),%ZMM0

(8) 0x203628 VMOVUPD	-0x80(%R8,%R9,8),%ZMM1

(8) 0x203630 VMOVUPD	-0x40(%R8,%R9,8),%ZMM2

(8) 0x203638 VMOVUPD	(%R8,%R9,8),%ZMM3

(8) 0x20363f VADDPD	%ZMM0,%ZMM0,%ZMM0

(8) 0x203645 VADDPD	%ZMM1,%ZMM1,%ZMM1

(8) 0x20364b VADDPD	%ZMM2,%ZMM2,%ZMM2

(8) 0x203651 VADDPD	%ZMM3,%ZMM3,%ZMM3

(8) 0x203657 VMOVUPD	%ZMM0,-0xc0(%R8,%R9,8)

(8) 0x20365f VMOVUPD	%ZMM1,-0x80(%R8,%R9,8)

(8) 0x203667 VMOVUPD	%ZMM2,-0x40(%R8,%R9,8)

(8) 0x20366f VMOVUPD	%ZMM3,(%R8,%R9,8)

(8) 0x203676 ADD	$0x20,%R9

(8) 0x20367a CMP	%R9,%RDI

(8) 0x20367d JNE	203620

0x20367f CMP	%RDI,%RSI

0x203682 JE	2036e8

0x203684 TEST	$0x1c,%SIL

0x203688 JE	2036c0

0x20368a LEA	(%R14,%RDX,1),%R9

0x20368e MOV	%RSI,%R8

0x203691 AND	$-0x4,%R8

0x203695 LEA	(%R9,%RCX,8),%R9

0x203699 ADD	%R8,%RCX

0x20369c NOPL	(%RAX)

(9) 0x2036a0 VMOVUPD	(%R9,%RDI,8),%YMM0

(9) 0x2036a6 VADDPD	%YMM0,%YMM0,%YMM0

(9) 0x2036aa VMOVUPD	%YMM0,(%R9,%RDI,8)

(9) 0x2036b0 ADD	$0x4,%RDI

(9) 0x2036b4 CMP	%RDI,%R8

(9) 0x2036b7 JNE	2036a0

0x2036b9 CMP	%R8,%RSI

0x2036bc JNE	2036c3

0x2036be JMP	2036e8

0x2036c0 ADD	%RDI,%RCX

0x2036c3 DEC	%RCX

0x2036c6 ADD	%R14,%RDX

0x2036c9 NOPL	(%RAX)

(7) 0x2036d0 VMOVSD	0x8(%RDX,%RCX,8),%XMM0

(7) 0x2036d6 VADDSD	%XMM0,%XMM0,%XMM0

(7) 0x2036da VMOVSD	%XMM0,0x8(%RDX,%RCX,8)

(7) 0x2036e0 INC	%RCX

(7) 0x2036e3 CMP	%RAX,%RCX

(7) 0x2036e6 JL	2036d0

0x2036e8 MOV	$-0x1248,%RDI

0x2036f2 MOV	$-0x1018,%RAX

0x2036fc MOV	%EBX,%ESI

0x2036fe ADD	%R14,%RDI

0x203701 VZEROUPPER

0x203704 CALLQ	(%R14,%RAX,1)

0x203708 ADD	$0x20,%RSP

0x20370c POP	%RBX

0x20370d POP	%R14

0x20370f POP	%RBP

0x203710 RET

0x203711 NOPW	%CS:(%RAX,%RAX,1)

Coverage (%)	Name	Source Location	Module
►98.96+	__kmp_invoke_microtask		libomp.so
○	__kmp_invoke_task_func		libomp.so
►1.04+	__kmp_invoke_microtask		libomp.so
○	__kmp_invoke_task_func		libomp.so
○	unknown kernel region	unknown kernel region	kernel

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 orig_0

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

The function is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/177-365-3972/stream-copy/build/stream-copy/src/stream.c:290-292.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Your function is probably not vectorized. Only 10% 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 10.63 to 0.76 cycles (13.94x speedup).

Details

Store and arithmetical 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/gmz17.benchmarkcenter.megware.com/177-365-3972/stream-copy/build/stream-copy/src/stream.c:290
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/177-365-3972/stream-copy/build/stream-copy/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.

CMOVL: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/177-365-3972/stream-copy/build/stream-copy/src/stream.c:291
LEA: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/177-365-3972/stream-copy/build/stream-copy/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 36 bytes. The binary function is storing 36 bytes.

General properties

nb instructions	81
nb uops	85
loop length	352
used x86 registers	12
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	10.63 cycles
front end	10.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	8.67	8.67	8.67	8.67	8.67	8.67	3.25	3.25	3.25	3.25	0.00	0.00	0.00	0.00	0.00	0.00
cycles	8.67	8.67	8.67	8.67	8.67	8.67	3.25	3.25	3.25	3.25	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	10.63
Dispatch	8.67
Overall L1	10.63

Vectorization ratios

all	5%
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	10%

Vector efficiency ratios

all	10%
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	12%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 10.63 cycles. At this rate:

1% of peak load performance is reached (3.39 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (3.39 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 10.63 to 8.67 cycles (1.23x speedup).

ASM code

In the binary file, the address of the function is: 203530

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 %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 $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 (%RDI),%EBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (6.3%)
LEA -0x7(%RIP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x3143,%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOVQ $0,-0x28(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVQ $0x3344ffff,-0x18(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVQ $0x1,-0x30(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
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%)
ADD %RAX,%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
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
MOV $-0x1260,%RDI	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),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
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
MOV $-0x1028,%R10	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),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA -0x18(%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	N/A
ADD %R14,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
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
CALLQ (%R14,%R10,1)	4	0	0	0	0.33	0.33	0.33	0.50	0.50	0.50	0.50	0	2.20	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 -0x18(%RBP),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV $0x3344ffff,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP $0x3344ffff,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
CMOVL %RCX,%RAX	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	1	0.33	N/A
MOV -0x28(%RBP),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %RAX,-0x18(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
CMP %RAX,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JG 2036e8 <main.omp_outlined.18+0x1b8>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
MOV %RAX,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
SUB %RCX,%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 $0x30,%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
INC %RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
CMP $0x4,%RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JB 2036c3 <main.omp_outlined.18+0x193>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
CMP $0x20,%RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JAE 2035fe <main.omp_outlined.18+0xce>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
XOR %EDI,%EDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
JMP 20368a <main.omp_outlined.18+0x15a>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	N/A
LEA (%R14,%RDX,1),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %RSI,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
AND $-0x20,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
XOR %R9D,%R9D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA 0xc0(%R8,%RCX,8),%R8	2	0	0	0.50	0.50	0	0	0.25	0.25	0.25	0.25	2	0.50	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
CMP %RDI,%RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JE 2036e8 <main.omp_outlined.18+0x1b8>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
TEST $0x1c,%SIL	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
JE 2036c0 <main.omp_outlined.18+0x190>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
LEA (%R14,%RDX,1),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %RSI,%R8	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
AND $-0x4,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA (%R9,%RCX,8),%R9	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	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
CMP %R8,%RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JNE 2036c3 <main.omp_outlined.18+0x193>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
JMP 2036e8 <main.omp_outlined.18+0x1b8>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	N/A
ADD %RDI,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
DEC %RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ADD %R14,%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
MOV $-0x1248,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $-0x1018,%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 %R14,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALLQ (%R14,%RAX,1)	4	0	0	0	0.33	0.33	0.33	0.50	0.50	0.50	0.50	0	2.20	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
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 %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run orig_0

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/gmz17.benchmarkcenter.megware.com/177-365-3972/stream-copy/build/stream-copy/src/stream.c:290
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/177-365-3972/stream-copy/build/stream-copy/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.

CMOVL: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/177-365-3972/stream-copy/build/stream-copy/src/stream.c:291
LEA: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/177-365-3972/stream-copy/build/stream-copy/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 36 bytes. The binary function is storing 36 bytes.

General properties

nb instructions	81
nb uops	85
loop length	352
used x86 registers	12
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	10.63 cycles
front end	10.63 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	8.67	8.67	8.67	8.67	8.67	8.67	3.25	3.25	3.25	3.25	0.00	0.00	0.00	0.00	0.00	0.00
cycles	8.67	8.67	8.67	8.67	8.67	8.67	3.25	3.25	3.25	3.25	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	10.63
Dispatch	8.67
Overall L1	10.63

Vectorization ratios

all	5%
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	10%

Vector efficiency ratios

all	10%
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	12%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 10.63 cycles. At this rate:

1% of peak load performance is reached (3.39 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (3.39 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: 203530

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 %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 $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 (%RDI),%EBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (6.3%)
LEA -0x7(%RIP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x3143,%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOVQ $0,-0x28(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVQ $0x3344ffff,-0x18(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOVQ $0x1,-0x30(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
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%)
ADD %RAX,%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
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
MOV $-0x1260,%RDI	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),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
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
MOV $-0x1028,%R10	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),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA -0x18(%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	N/A
ADD %R14,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
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
CALLQ (%R14,%R10,1)	4	0	0	0	0.33	0.33	0.33	0.50	0.50	0.50	0.50	0	2.20	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 -0x18(%RBP),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV $0x3344ffff,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP $0x3344ffff,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
CMOVL %RCX,%RAX	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	1	0.33	N/A
MOV -0x28(%RBP),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %RAX,-0x18(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
CMP %RAX,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JG 2036e8 <main.omp_outlined.18+0x1b8>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
MOV %RAX,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
SUB %RCX,%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 $0x30,%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
INC %RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
CMP $0x4,%RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JB 2036c3 <main.omp_outlined.18+0x193>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
CMP $0x20,%RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JAE 2035fe <main.omp_outlined.18+0xce>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
XOR %EDI,%EDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
JMP 20368a <main.omp_outlined.18+0x15a>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	N/A
LEA (%R14,%RDX,1),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %RSI,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
AND $-0x20,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
XOR %R9D,%R9D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA 0xc0(%R8,%RCX,8),%R8	2	0	0	0.50	0.50	0	0	0.25	0.25	0.25	0.25	2	0.50	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
CMP %RDI,%RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JE 2036e8 <main.omp_outlined.18+0x1b8>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
TEST $0x1c,%SIL	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
JE 2036c0 <main.omp_outlined.18+0x190>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
LEA (%R14,%RDX,1),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %RSI,%R8	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
AND $-0x4,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA (%R9,%RCX,8),%R9	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	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
CMP %R8,%RSI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JNE 2036c3 <main.omp_outlined.18+0x193>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
JMP 2036e8 <main.omp_outlined.18+0x1b8>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	N/A
ADD %RDI,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
DEC %RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ADD %R14,%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
MOV $-0x1248,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $-0x1018,%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 %R14,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALLQ (%R14,%RAX,1)	4	0	0	0	0.33	0.33	0.33	0.50	0.50	0.50	0.50	0	2.20	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
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 %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A

Name	Coverage (%)	Time (s)
▼main.omp_outlined.18–	0.99	0.03
○Loop 8 - stream.c:292-292 - exec	0.99	0.03
○Loop 7 - stream.c:291-292 - exec	0.00	0.00
○Loop 9 - stream.c:292-292 - 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