OV - exec - Function advec_mom_kernel(int, int, int, int, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer1D<double>&, clover::Buffer1D<double>&, int, int, int) [clone ._omp

	Function: advec_mom_kernel(int, int, int, int, clover::Buffer2D<double>&, clover::Buffer2D<double>&, ...	Module: exec	Source: :0-0	Coverage (incl. loops): 3.06% \| (excl. loops): 0.00%

  ***  This Panel is Intentionally Left Blank.  ***

It is due to a lack of debug symbols in the given object
(loop or function).

0x426440 PUSH	%R15

0x426442 PUSH	%R14

0x426444 PUSH	%R13

0x426446 PUSH	%R12

0x426448 PUSH	%RBP

0x426449 PUSH	%RBX

0x42644a MOV	%RDI,%RBX

0x42644d SUB	$0x88,%RSP

0x426454 MOV	0x30(%RDI),%EAX

0x426457 MOV	0x2c(%RBX),%ESI

0x42645a LEA	-0x1(%RAX),%EBP

0x42645d MOV	0x34(%RDI),%EAX

0x426460 MOV	0x28(%RDI),%EDI

0x426463 ADD	$0x4,%EAX

0x426466 SUB	$0x1,%EDI

0x426469 MOV	%EAX,0x48(%RSP)

0x42646d MOV	%EDI,0x4c(%RSP)

0x426471 CMP	%EAX,%EBP

0x426473 JGE	42673a

0x426479 MOV	%EAX,%R13D

0x42647c LEA	0x4(%RSI),%R14D

0x426480 SUB	%EBP,%R13D

0x426483 CMP	%R14D,%EDI

0x426486 JGE	42673a

0x42648c MOV	%R14D,%R15D

0x42648f SUB	%EDI,%R15D

0x426492 MOV	%R15D,0x50(%RSP)

0x426497 CALL	404660 <omp_get_num_threads@plt>

0x42649c MOV	%EAX,%R12D

0x42649f CALL	404550 <omp_get_thread_num@plt>

0x4264a4 XOR	%EDX,%EDX

0x4264a6 MOV	%EAX,%ECX

0x4264a8 MOV	%R15D,%EAX

0x4264ab IMUL	%R13D,%EAX

0x4264af DIV	%R12D

0x4264b2 MOV	%EAX,%EDI

0x4264b4 CMP	%EDX,%ECX

0x4264b6 JB	42674c

0x4264bc IMUL	%EDI,%ECX

0x4264bf LEA	(%RCX,%RDX,1),%ESI

0x4264c2 LEA	(%RDI,%RSI,1),%EAX

0x4264c5 MOV	%EAX,0x54(%RSP)

0x4264c9 CMP	%EAX,%ESI

0x4264cb JAE	42673a

0x4264d1 XOR	%EDX,%EDX

0x4264d3 MOV	%ESI,%EAX

0x4264d5 MOV	0x4c(%RSP),%ECX

0x4264d9 DIVL	0x50(%RSP)

0x4264dd LEA	(%RDX,%RCX,1),%R15D

0x4264e1 MOV	0x10(%RBX),%RCX

0x4264e5 MOV	%R14D,%EDX

0x4264e8 ADD	%EBP,%EAX

0x4264ea SUB	%R15D,%EDX

0x4264ed MOV	%RCX,0x58(%RSP)

0x4264f2 MOV	0x8(%RBX),%RCX

0x4264f6 CMP	%EDX,%EDI

0x4264f8 CMOVA	%EDX,%EDI

0x4264fb MOV	%RCX,0x60(%RSP)

0x426500 MOV	0x20(%RBX),%RCX

0x426504 MOV	%RCX,0x68(%RSP)

0x426509 MOV	(%RBX),%RCX

0x42650c MOV	0x18(%RBX),%RBX

0x426510 MOV	%RCX,0x70(%RSP)

0x426515 MOVSXD	%EAX,%RCX

0x426518 LEA	(%RSI,%RDI,1),%EAX

0x42651b MOV	%RBX,0x78(%RSP)

0x426520 MOV	%EAX,0x1c(%RSP)

0x426524 CMP	%EAX,%ESI

0x426526 JAE	426726

0x42652c NOPL	(%RAX)

  (215) 0x426530 MOV	0x58(%RSP),%RAX

  (215) 0x426535 LEA	0x1(%RCX),%RBX

  (215) 0x426539 MOV	%RBX,%RBP

  (215) 0x42653c MOV	%RBX,0x10(%RSP)

  (215) 0x426541 MOV	0x10(%RAX),%R10

  (215) 0x426545 MOV	(%RAX),%RSI

  (215) 0x426548 MOV	0x60(%RSP),%RAX

  (215) 0x42654d IMUL	%RCX,%RSI

  (215) 0x426551 MOV	%R10,0x38(%RSP)

  (215) 0x426556 MOV	0x10(%RAX),%R13

  (215) 0x42655a MOV	(%RAX),%RAX

  (215) 0x42655d IMUL	%RAX,%RBP

  (215) 0x426561 MOV	%RSI,0x40(%RSP)

  (215) 0x426566 MOV	%RBP,%R8

  (215) 0x426569 SUB	%RAX,%R8

  (215) 0x42656c MOV	0x68(%RSP),%RAX

  (215) 0x426571 MOV	%R8,0x30(%RSP)

  (215) 0x426576 MOV	0x10(%RAX),%RBX

  (215) 0x42657a MOV	(%RAX),%R9

  (215) 0x42657d MOV	0x70(%RSP),%RAX

  (215) 0x426582 IMUL	%RCX,%R9

  (215) 0x426586 MOV	%RBX,0x8(%RSP)

  (215) 0x42658b MOV	(%RAX),%R14

  (215) 0x42658e MOV	0x10(%RAX),%R12

  (215) 0x426592 MOV	0x78(%RSP),%RAX

  (215) 0x426597 IMUL	%RCX,%R14

  (215) 0x42659b MOV	%R9,0x28(%RSP)

  (215) 0x4265a0 IMUL	(%RAX),%RCX

  (215) 0x4265a4 MOV	0x10(%RAX),%RDX

  (215) 0x4265a8 MOV	%RDX,(%RSP)

  (215) 0x4265ac MOV	%RCX,0x20(%RSP)

  (215) 0x4265b1 CMP	$0x1,%EDI

  (215) 0x4265b4 JE	42668a

  (215) 0x4265ba MOVSXD	%R15D,%RAX

  (215) 0x4265bd LEA	(%RSI,%RAX,1),%RDX

  (215) 0x4265c1 MOV	0x8(%RSP),%RSI

  (215) 0x4265c6 LEA	(%R10,%RDX,8),%RBX

  (215) 0x4265ca LEA	(%RBP,%RAX,1),%RDX

  (215) 0x4265cf LEA	(%R13,%RDX,8),%R11

  (215) 0x4265d4 LEA	(%R8,%RAX,1),%RDX

  (215) 0x4265d8 LEA	(%R13,%RDX,8),%R10

  (215) 0x4265dd LEA	(%R9,%RAX,1),%RDX

  (215) 0x4265e1 LEA	(%RSI,%RDX,8),%R9

  (215) 0x4265e5 LEA	0x1(%R14,%RAX,1),%RDX

  (215) 0x4265ea ADD	%RCX,%RAX

  (215) 0x4265ed SAL	$0x3,%RDX

  (215) 0x4265f1 LEA	(%R12,%RDX,1),%R8

  (215) 0x4265f5 LEA	-0x8(%R12,%RDX,1),%RSI

  (215) 0x4265fa MOV	(%RSP),%RDX

  (215) 0x4265fe LEA	(%RDX,%RAX,8),%RCX

  (215) 0x426602 MOV	%EDI,%EDX

  (215) 0x426604 XOR	%EAX,%EAX

  (215) 0x426606 SHR	$0x1,%EDX

  (215) 0x426608 SAL	$0x4,%RDX

  (215) 0x42660c NOPW	%CS:(%RAX,%RAX,1)

  (215) 0x426617 NOPW	%CS:(%RAX,%RAX,1)

  (215) 0x426622 NOPW	%CS:(%RAX,%RAX,1)

  (215) 0x42662d NOPW	%CS:(%RAX,%RAX,1)

  (215) 0x426638 NOPL	(%RAX,%RAX,1)

    (216) 0x426640 MOVUPD	(%R11,%RAX,1),%XMM0

    (216) 0x426646 MOVUPD	(%RBX,%RAX,1),%XMM1

    (216) 0x42664b MOVUPD	(%R10,%RAX,1),%XMM2

    (216) 0x426651 ADDPD	%XMM1,%XMM0

    (216) 0x426655 SUBPD	%XMM2,%XMM0

    (216) 0x426659 MOVUPS	%XMM0,(%R9,%RAX,1)

    (216) 0x42665e MOVUPD	(%R8,%RAX,1),%XMM3

    (216) 0x426664 MOVUPD	(%RSI,%RAX,1),%XMM4

    (216) 0x426669 ADDPD	%XMM3,%XMM0

    (216) 0x42666d SUBPD	%XMM4,%XMM0

    (216) 0x426671 MOVUPS	%XMM0,(%RCX,%RAX,1)

    (216) 0x426675 ADD	$0x10,%RAX

    (216) 0x426679 CMP	%RAX,%RDX

    (216) 0x42667c JNE	426640

  (215) 0x42667e TEST	$0x1,%DIL

  (215) 0x426682 JE	4266f4

  (215) 0x426684 AND	$-0x2,%EDI

  (215) 0x426687 ADD	%EDI,%R15D

  (215) 0x42668a MOV	0x40(%RSP),%RDX

  (215) 0x42668f MOVSXD	%R15D,%RAX

  (215) 0x426692 MOV	0x38(%RSP),%RDI

  (215) 0x426697 ADD	$0x1,%R15D

  (215) 0x42669b ADD	%RAX,%RBP

  (215) 0x42669e MOV	0x8(%RSP),%RBX

  (215) 0x4266a3 MOVSXD	%R15D,%R15

  (215) 0x4266a6 ADD	%RAX,%RDX

  (215) 0x4266a9 MOVSD	(%R13,%RBP,8),%XMM0

  (215) 0x4266b0 ADD	%R14,%R15

  (215) 0x4266b3 ADDSD	(%RDI,%RDX,8),%XMM0

  (215) 0x4266b8 MOV	0x30(%RSP),%RDX

  (215) 0x4266bd ADD	%RAX,%RDX

  (215) 0x4266c0 SUBSD	(%R13,%RDX,8),%XMM0

  (215) 0x4266c7 MOV	0x28(%RSP),%RDX

  (215) 0x4266cc ADD	%RAX,%RDX

  (215) 0x4266cf MOVSD	%XMM0,(%RBX,%RDX,8)

  (215) 0x4266d4 MOV	0x20(%RSP),%RDX

  (215) 0x4266d9 ADDSD	(%R12,%R15,8),%XMM0

  (215) 0x4266df ADD	%RAX,%RDX

  (215) 0x4266e2 ADD	%R14,%RAX

  (215) 0x4266e5 SUBSD	(%R12,%RAX,8),%XMM0

  (215) 0x4266eb MOV	(%RSP),%RAX

  (215) 0x4266ef MOVSD	%XMM0,(%RAX,%RDX,8)

  (215) 0x4266f4 MOV	0x1c(%RSP),%ESI

  (215) 0x4266f8 MOV	0x10(%RSP),%RCX

  (215) 0x4266fd CMP	%ECX,0x48(%RSP)

  (215) 0x426701 JLE	42673a

  (215) 0x426703 MOV	0x54(%RSP),%EDI

  (215) 0x426707 MOV	0x50(%RSP),%EDX

  (215) 0x42670b MOV	0x4c(%RSP),%R15D

  (215) 0x426710 SUB	%ESI,%EDI

  (215) 0x426712 CMP	%EDX,%EDI

  (215) 0x426714 CMOVA	%EDX,%EDI

  (215) 0x426717 LEA	(%RSI,%RDI,1),%EAX

  (215) 0x42671a MOV	%EAX,0x1c(%RSP)

  (215) 0x42671e CMP	%EAX,%ESI

  (215) 0x426720 JB	426530

(217) 0x426726 LEA	0x1(%RCX),%RAX

(217) 0x42672a MOV	%RAX,0x10(%RSP)

(217) 0x42672f MOV	0x10(%RSP),%RCX

(217) 0x426734 CMP	%ECX,0x48(%RSP)

(217) 0x426738 JG	426703

0x42673a ADD	$0x88,%RSP

0x426741 POP	%RBX

0x426742 POP	%RBP

0x426743 POP	%R12

0x426745 POP	%R13

0x426747 POP	%R14

0x426749 POP	%R15

0x42674b RET

0x42674c ADD	$0x1,%EDI

0x42674f XOR	%EDX,%EDX

0x426751 JMP	4264bc

0x426756 NOPW	%CS:(%RAX,%RAX,1)

Coverage (%)	Name	Source Location	Module
►96.79+	gomp_thread_start	team.c:130	libgomp.so.1.0.0
○	start_thread		libc.so.6
○	__GI___clone3		libc.so.6
►1.57+	GOMP_parallel	libgomp.h:982	libgomp.so.1.0.0
○	advec_mom_kernel(int, int, int[...]		exec
○	advec_mom_driver(global_variab[...]		exec
○	advection(global_variables&)		exec
○	hydro(global_variables&, paral[...]		exec
○	main		exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start		exec
►1.53+	GOMP_parallel	libgomp.h:982	libgomp.so.1.0.0
○	advec_mom_kernel(int, int, int[...]		exec
○	advec_mom_driver(global_variab[...]		exec
○	advection(global_variables&)		exec
○	hydro(global_variables&, paral[...]		exec
○	main		exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start		exec

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
Module	exec

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Your function is not vectorized. Only 7% 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 12.00 to 0.75 cycles (16.00x 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)

Execution units bottlenecks

Performance is limited by execution of divide and square root operations (the divide/square root unit is a bottleneck). By removing all these bottlenecks, you can lower the cost of an iteration from 12.00 to 10.63 cycles (1.13x speedup).

Workaround

Reduce the number of division or square root instructions:

If denominator is constant over iterations, use reciprocal (replace x/y with x*(1/y)). Check precision impact.

Expensive FP math instructions/calls

Detected performance impact from expensive FP math instructions/calls. By removing/reexpressing them, you can lower the cost of an iteration from 12.00 to 10.38 cycles (1.16x speedup).

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.

omp_get_num_threads@plt: 1 occurrences►
omp_get_thread_num@plt: 1 occurrences►

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.

CMOVA: 1 occurrences►
DIV: 2 occurrences►

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 64 bytes. The binary function is storing 60 bytes.

General properties

nb instructions	82
nb uops	85
loop length	278
used x86 registers	12
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	10

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	6.50	6.50	6.50	6.50	6.50	6.50	5.75	5.75	5.75	5.75	0.00	0.00	0.00	0.00	0.00	0.00
cycles	6.50	6.50	6.50	6.50	6.50	6.50	5.75	5.75	5.75	5.75	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

12.00

Cycles summary

Front-end	10.63
Dispatch	6.50
DIV/SQRT	12.00
Overall L1	12.00

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	0%
other	0%

Vector efficiency ratios

all	7%
load	6%
store	9%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	6%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	6%
other	6%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 12.00 cycles. At this rate:

2% of peak load performance is reached (5.33 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
3% of peak store performance is reached (5.00 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 426440

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
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 %R13	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R12	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RBP	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
MOV %RDI,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SUB $0x88,%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(%RDI),%EAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x2c(%RBX),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
LEA -0x1(%RAX),%EBP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV 0x34(%RDI),%EAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x28(%RDI),%EDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
ADD $0x4,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SUB $0x1,%EDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %EAX,0x48(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOV %EDI,0x4c(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
CMP %EAX,%EBP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JGE 42673a <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x2fa>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV %EAX,%R13D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
LEA 0x4(%RSI),%R14D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SUB %EBP,%R13D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
CMP %R14D,%EDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JGE 42673a <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x2fa>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV %R14D,%R15D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
SUB %EDI,%R15D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
MOV %R15D,0x50(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
CALL 404660 <omp_get_num_threads@plt>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
MOV %EAX,%R12D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
CALL 404550 <omp_get_thread_num@plt>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
MOV %EAX,%ECX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R15D,%EAX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %R13D,%EAX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
DIV %R12D	2	0	0	0	2	0	0	0	0	0	0	10-17	6	scal (6.3%)
MOV %EAX,%EDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMP %EDX,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JB 42674c <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x30c>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
IMUL %EDI,%ECX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
LEA (%RCX,%RDX,1),%ESI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA (%RDI,%RSI,1),%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %EAX,0x54(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
CMP %EAX,%ESI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JAE 42673a <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x2fa>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
MOV %ESI,%EAX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV 0x4c(%RSP),%ECX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
DIVL 0x50(%RSP)	2	0	0	0	2	0	0	0.25	0.25	0.25	0.25	10-17	6	scal (6.3%)
LEA (%RDX,%RCX,1),%R15D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV 0x10(%RBX),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %R14D,%EDX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
ADD %EBP,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SUB %R15D,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %RCX,0x58(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV 0x8(%RBX),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
CMP %EDX,%EDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
CMOVA %EDX,%EDI	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	1	0.33	N/A
MOV %RCX,0x60(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV 0x20(%RBX),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %RCX,0x68(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV (%RBX),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x18(%RBX),%RBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %RCX,0x70(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOVSXD %EAX,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA (%RSI,%RDI,1),%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %RBX,0x78(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV %EAX,0x1c(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
CMP %EAX,%ESI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JAE 426726 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x2e6>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
ADD $0x88,%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	0.33	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R12	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R13	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50	N/A
ADD $0x1,%EDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
JMP 4264bc <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x7c>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	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 gcc_0

Source file and lines
Module	exec

Warnings:
Ignoring paths for analysis

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)

Execution units bottlenecks

Workaround

Reduce the number of division or square root instructions:

If denominator is constant over iterations, use reciprocal (replace x/y with x*(1/y)). Check precision impact.

Expensive FP math instructions/calls

Detected performance impact from expensive FP math instructions/calls. By removing/reexpressing them, you can lower the cost of an iteration from 12.00 to 10.38 cycles (1.16x speedup).

CALL instructions

Detected function call instructions.

Details

omp_get_num_threads@plt: 1 occurrences►
omp_get_thread_num@plt: 1 occurrences►

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.

CMOVA: 1 occurrences►
DIV: 2 occurrences►

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 64 bytes. The binary function is storing 60 bytes.

General properties

nb instructions	82
nb uops	85
loop length	278
used x86 registers	12
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	10

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	6.50	6.50	6.50	6.50	6.50	6.50	5.75	5.75	5.75	5.75	0.00	0.00	0.00	0.00	0.00	0.00
cycles	6.50	6.50	6.50	6.50	6.50	6.50	5.75	5.75	5.75	5.75	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

12.00

Cycles summary

Front-end	10.63
Dispatch	6.50
DIV/SQRT	12.00
Overall L1	12.00

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	0%
other	0%

Vector efficiency ratios

all	7%
load	6%
store	9%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	6%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	6%
other	6%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 12.00 cycles. At this rate:

2% of peak load performance is reached (5.33 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
3% of peak store performance is reached (5.00 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 426440

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
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 %R13	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R12	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RBP	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
MOV %RDI,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SUB $0x88,%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(%RDI),%EAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x2c(%RBX),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
LEA -0x1(%RAX),%EBP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV 0x34(%RDI),%EAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x28(%RDI),%EDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
ADD $0x4,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SUB $0x1,%EDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %EAX,0x48(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
MOV %EDI,0x4c(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
CMP %EAX,%EBP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JGE 42673a <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x2fa>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV %EAX,%R13D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
LEA 0x4(%RSI),%R14D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SUB %EBP,%R13D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
CMP %R14D,%EDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JGE 42673a <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x2fa>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV %R14D,%R15D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
SUB %EDI,%R15D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
MOV %R15D,0x50(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
CALL 404660 <omp_get_num_threads@plt>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
MOV %EAX,%R12D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
CALL 404550 <omp_get_thread_num@plt>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
MOV %EAX,%ECX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R15D,%EAX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %R13D,%EAX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
DIV %R12D	2	0	0	0	2	0	0	0	0	0	0	10-17	6	scal (6.3%)
MOV %EAX,%EDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMP %EDX,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JB 42674c <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x30c>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
IMUL %EDI,%ECX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
LEA (%RCX,%RDX,1),%ESI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA (%RDI,%RSI,1),%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %EAX,0x54(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
CMP %EAX,%ESI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JAE 42673a <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x2fa>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
MOV %ESI,%EAX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV 0x4c(%RSP),%ECX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
DIVL 0x50(%RSP)	2	0	0	0	2	0	0	0.25	0.25	0.25	0.25	10-17	6	scal (6.3%)
LEA (%RDX,%RCX,1),%R15D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV 0x10(%RBX),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %R14D,%EDX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
ADD %EBP,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SUB %R15D,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %RCX,0x58(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV 0x8(%RBX),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
CMP %EDX,%EDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
CMOVA %EDX,%EDI	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	1	0.33	N/A
MOV %RCX,0x60(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV 0x20(%RBX),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %RCX,0x68(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV (%RBX),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x18(%RBX),%RBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %RCX,0x70(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOVSXD %EAX,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA (%RSI,%RDI,1),%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %RBX,0x78(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV %EAX,0x1c(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (6.3%)
CMP %EAX,%ESI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
JAE 426726 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x2e6>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
ADD $0x88,%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	0.33	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R12	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R13	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50	N/A
ADD $0x1,%EDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
XOR %EDX,%EDX	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
JMP 4264bc <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii._omp_fn.0+0x7c>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A

Name	Coverage (%)	Time (s)
▼advec_mom_kernel(int, int, int, int, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clo...–	3.06	1.17
▼Loop 217 - - exec–	0.00	0.00
▼Loop 215 - - exec–	0.00	0.01
○Loop 216 - - exec	3.06	1.17

Report Configuration

Vectorization

Details

Workaround

Execution units bottlenecks

Workaround

Expensive FP math instructions/calls

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

Workaround

Expensive FP math instructions/calls

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