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: advec_mom.cpp:85-88 [...]	Coverage (incl. loops): 1.32% \| (excl. loops): 0.00%

/home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h: 69 - 69

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

69:   T &operator()(size_t i, size_t j) const { return data[i + j * sizeX]; }

/home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp: 85 - 88

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

85: #pragma omp parallel for simd collapse(2)

86:       for (int j = (y_min + 1); j < (y_max + 1 + 2); j++) {

87:         for (int i = (x_min - 2 + 1); i < (x_max + 2 + 2); i++) {

88:           node_flux(i, j) = 0.25 * (mass_flux_x(i + 0, j - 1) + mass_flux_x(i, j) + mass_flux_x(i + 1, j - 1) + mass_flux_x(i + 1, j + 0));

0x235340 PUSH	%RBP

0x235341 MOV	%RSP,%RBP

0x235344 PUSH	%R15

0x235346 PUSH	%R14

0x235348 PUSH	%R13

0x23534a PUSH	%R12

0x23534c PUSH	%RBX

0x23534d SUB	$0x58,%RSP

0x235351 MOV	(%RCX),%R12D

0x235354 MOV	(%RDX),%R14D

0x235357 LEA	0x1(%R14),%R15D

0x23535b ADD	$0x3,%R12D

0x23535f SUB	%R15D,%R12D

0x235362 JLE	235795

0x235368 MOV	(%R9),%EBX

0x23536b MOV	(%R8),%R13D

0x23536e ADD	$0x4,%EBX

0x235371 CMP	%EBX,%R13D

0x235374 JG	235795

0x23537a LEA	-0x1(%R13),%EAX

0x23537e MOV	(%RDI),%ESI

0x235380 MOVQ	$0,-0x48(%RBP)

0x235388 MOVQ	$0x1,-0x78(%RBP)

0x235390 MOVL	$0,-0x40(%RBP)

0x235397 SUB	%EAX,%EBX

0x235399 MOV	%RAX,-0x60(%RBP)

0x23539d IMUL	%RBX,%R12

0x2353a1 DEC	%R12

0x2353a4 MOV	%R12,-0x38(%RBP)

0x2353a8 SUB	$0x8,%RSP

0x2353ac LEA	-0x40(%RBP),%RCX

0x2353b0 LEA	-0x78(%RBP),%RAX

0x2353b4 LEA	0x258a5(%RIP),%RDI

0x2353bb LEA	-0x48(%RBP),%R8

0x2353bf LEA	-0x38(%RBP),%R9

0x2353c3 MOV	%ESI,-0x3c(%RBP)

0x2353c6 MOV	$0x22,%EDX

0x2353cb PUSH	$0x1

0x2353cd PUSH	$0x1

0x2353cf PUSH	%RAX

0x2353d0 CALL	259680 <@plt_start@+0x530>

0x2353d5 ADD	$0x20,%RSP

0x2353d9 MOV	-0x38(%RBP),%RAX

0x2353dd MOV	-0x48(%RBP),%RCX

0x2353e1 CMP	%R12,%RAX

0x2353e4 CMOVL	%RAX,%R12

0x2353e8 MOV	%R12,-0x38(%RBP)

0x2353ec CMP	%R12,%RCX

0x2353ef JG	235783

0x2353f5 MOV	0x10(%RBP),%RDX

0x2353f9 MOV	0x18(%RBP),%RAX

0x2353fd MOV	%RBX,-0x30(%RBP)

0x235401 MOV	-0x60(%RBP),%RSI

0x235405 MOV	%R12,%R11

0x235408 SUB	%RCX,%R11

0x23540b INC	%R11

0x23540e MOV	(%RDX),%R10

0x235411 MOV	(%RAX),%RBX

0x235414 MOV	0x10(%RAX),%RDI

0x235418 MOV	0x10(%RDX),%R9

0x23541c MOV	%R10,-0x70(%RBP)

0x235420 CMP	$0x4,%R11

0x235424 JAE	23542f

0x235426 MOV	-0x30(%RBP),%R8

0x23542a JMP	2356dc

0x23542f VBROADCASTSD	-0x242a8(%RIP),%YMM9

0x235438 VPBROADCASTQ	%RCX,%YMM0

0x23543e VPADDQ	-0x24166(%RIP),%YMM0,%YMM0

0x235446 VPBROADCASTQ	-0x2427f(%RIP),%YMM10

0x23544f MOV	-0x30(%RBP),%RAX

0x235453 MOV	%R11,-0x58(%RBP)

0x235457 AND	$-0x4,%R11

0x23545b VPBROADCASTD	%R15D,%XMM2

0x235461 VPBROADCASTD	%ESI,%XMM3

0x235467 VPBROADCASTD	%R14D,%XMM4

0x23546d VPBROADCASTQ	%RBX,%YMM5

0x235473 VPBROADCASTD	%R13D,%XMM6

0x235479 VPBROADCASTQ	%R10,%YMM7

0x23547f MOV	%RBX,-0x50(%RBP)

0x235483 ADD	%R11,%RCX

0x235486 MOV	%R11,%RBX

0x235489 VPBROADCASTQ	%RAX,%YMM1

0x23548f VEXTRACTI128	$0x1,%YMM1,%XMM8

0x235495 NOPW	%CS:(%RAX,%RAX,1)

(155) 0x2354a0 VEXTRACTI128	$0x1,%YMM0,%XMM11

(155) 0x2354a6 VPEXTRQ	$0x1,%XMM8,%R10

(155) 0x2354ac VPEXTRQ	$0x1,%XMM11,%RAX

(155) 0x2354b2 CQTO

(155) 0x2354b4 IDIV	%R10

(155) 0x2354b7 VMOVQ	%XMM8,%R10

(155) 0x2354bc VMOVQ	%RAX,%XMM12

(155) 0x2354c1 VMOVQ	%XMM11,%RAX

(155) 0x2354c6 CQTO

(155) 0x2354c8 IDIV	%R10

(155) 0x2354cb VPEXTRQ	$0x1,%XMM1,%R10

(155) 0x2354d1 VMOVQ	%RAX,%XMM11

(155) 0x2354d6 VPEXTRQ	$0x1,%XMM0,%RAX

(155) 0x2354dc CQTO

(155) 0x2354de VPUNPCKLQDQ	%XMM12,%XMM11,%XMM11

(155) 0x2354e3 IDIV	%R10

(155) 0x2354e6 VMOVQ	%XMM1,%R10

(155) 0x2354eb VMOVQ	%RAX,%XMM12

(155) 0x2354f0 VMOVQ	%XMM0,%RAX

(155) 0x2354f5 CQTO

(155) 0x2354f7 IDIV	%R10

(155) 0x2354fa ADD	$-0x4,%R11

(155) 0x2354fe VMOVQ	%RAX,%XMM13

(155) 0x235503 VPUNPCKLQDQ	%XMM12,%XMM13,%XMM12

(155) 0x235508 VINSERTI128	$0x1,%XMM11,%YMM12,%YMM11

(155) 0x23550e VPMOVQD	%YMM11,%XMM12

(155) 0x235514 VPMULLQ	%YMM1,%YMM11,%YMM11

(155) 0x23551a VPSUBQ	%YMM11,%YMM0,%YMM11

(155) 0x23551f VPADDQ	%YMM0,%YMM10,%YMM0

(155) 0x235523 VPMOVQD	%YMM11,%XMM14

(155) 0x235529 VPADDD	%XMM2,%XMM12,%XMM13

(155) 0x23552d VPADDD	%XMM4,%XMM12,%XMM12

(155) 0x235531 VPMOVSXDQ	%XMM12,%YMM12

(155) 0x235536 VPMOVSXDQ	%XMM13,%YMM13

(155) 0x23553b VPMULLQ	%YMM12,%YMM5,%YMM12

(155) 0x235541 VPADDD	%XMM3,%XMM14,%XMM11

(155) 0x235545 VPADDD	%XMM6,%XMM14,%XMM14

(155) 0x235549 VPMOVSXDQ	%XMM11,%YMM11

(155) 0x23554e VPMOVSXDQ	%XMM14,%YMM14

(155) 0x235553 VPADDQ	%YMM11,%YMM12,%YMM15

(155) 0x235558 VPADDQ	%YMM14,%YMM12,%YMM12

(155) 0x23555d VMOVQ	%XMM15,%RAX

(155) 0x235562 VPEXTRQ	$0x1,%XMM15,%RDX

(155) 0x235568 VEXTRACTI128	$0x1,%YMM15,%XMM15

(155) 0x23556e VMOVQ	%XMM15,%R10

(155) 0x235573 VMOVSD	(%RDI,%RAX,8),%XMM16

(155) 0x23557a VPEXTRQ	$0x1,%XMM15,%R8

(155) 0x235580 VMOVSD	(%RDI,%R10,8),%XMM15

(155) 0x235586 VMOVHPD	(%RDI,%RDX,8),%XMM16,%XMM16

(155) 0x23558d VMOVHPD	(%RDI,%R8,8),%XMM15,%XMM15

(155) 0x235593 VINSERTF32X4	$0x1,%XMM15,%YMM16,%YMM15

(155) 0x23559a VPMULLQ	%YMM13,%YMM5,%YMM16

(155) 0x2355a0 VPMULLQ	%YMM13,%YMM7,%YMM13

(155) 0x2355a6 VPADDQ	%YMM11,%YMM16,%YMM17

(155) 0x2355ac VPADDQ	%YMM14,%YMM16,%YMM14

(155) 0x2355b2 VPADDQ	%YMM11,%YMM13,%YMM11

(155) 0x2355b7 VMOVQ	%XMM17,%RAX

(155) 0x2355bd VPEXTRQ	$0x1,%XMM17,%RDX

(155) 0x2355c4 VEXTRACTI32X4	$0x1,%YMM17,%XMM17

(155) 0x2355cb VMOVQ	%XMM17,%R10

(155) 0x2355d1 VMOVSD	(%RDI,%RAX,8),%XMM18

(155) 0x2355d8 VPEXTRQ	$0x1,%XMM17,%R8

(155) 0x2355df VPEXTRQ	$0x1,%XMM12,%RAX

(155) 0x2355e5 VMOVSD	(%RDI,%R10,8),%XMM17

(155) 0x2355ec VMOVHPD	(%RDI,%RDX,8),%XMM18,%XMM18

(155) 0x2355f3 VMOVQ	%XMM12,%RDX

(155) 0x2355f8 VEXTRACTI128	$0x1,%YMM12,%XMM12

(155) 0x2355fe VMOVHPD	(%RDI,%R8,8),%XMM17,%XMM17

(155) 0x235605 VMOVQ	%XMM12,%R8

(155) 0x23560a VPEXTRQ	$0x1,%XMM12,%R10

(155) 0x235610 VMOVSD	(%RDI,%R8,8),%XMM12

(155) 0x235616 VMOVHPD	(%RDI,%R10,8),%XMM12,%XMM12

(155) 0x23561c VINSERTF32X4	$0x1,%XMM17,%YMM18,%YMM17

(155) 0x235623 VADDPD	%YMM15,%YMM17,%YMM15

(155) 0x235629 VMOVSD	(%RDI,%RDX,8),%XMM17

(155) 0x235630 VPEXTRQ	$0x1,%XMM14,%RDX

(155) 0x235636 VMOVHPD	(%RDI,%RAX,8),%XMM17,%XMM17

(155) 0x23563d VMOVQ	%XMM14,%RAX

(155) 0x235642 VEXTRACTI128	$0x1,%YMM14,%XMM14

(155) 0x235648 VMOVQ	%XMM14,%R8

(155) 0x23564d VMOVSD	(%RDI,%RAX,8),%XMM16

(155) 0x235654 VPEXTRQ	$0x1,%XMM14,%R10

(155) 0x23565a VPEXTRQ	$0x1,%XMM11,%RAX

(155) 0x235660 VMOVSD	(%RDI,%R8,8),%XMM14

(155) 0x235666 VMOVHPD	(%RDI,%RDX,8),%XMM16,%XMM16

(155) 0x23566d VMOVQ	%XMM11,%RDX

(155) 0x235672 VEXTRACTI128	$0x1,%YMM11,%XMM11

(155) 0x235678 VMOVHPD	(%RDI,%R10,8),%XMM14,%XMM14

(155) 0x23567e VMOVQ	%XMM11,%R8

(155) 0x235683 VPEXTRQ	$0x1,%XMM11,%R10

(155) 0x235689 VINSERTF32X4	$0x1,%XMM12,%YMM17,%YMM12

(155) 0x235690 VINSERTF32X4	$0x1,%XMM14,%YMM16,%YMM14

(155) 0x235697 VADDPD	%YMM14,%YMM12,%YMM12

(155) 0x23569c VADDPD	%YMM12,%YMM15,%YMM12

(155) 0x2356a1 VMULPD	%YMM9,%YMM12,%YMM12

(155) 0x2356a6 VEXTRACTF128	$0x1,%YMM12,%XMM11

(155) 0x2356ac VMOVLPD	%XMM12,(%R9,%RDX,8)

(155) 0x2356b2 VMOVHPD	%XMM12,(%R9,%RAX,8)

(155) 0x2356b8 VMOVLPD	%XMM11,(%R9,%R8,8)

(155) 0x2356be VMOVHPD	%XMM11,(%R9,%R10,8)

(155) 0x2356c4 JNE	2354a0

0x2356ca MOV	-0x30(%RBP),%R8

0x2356ce CMP	%RBX,-0x58(%RBP)

0x2356d2 MOV	-0x50(%RBP),%RBX

0x2356d6 JE	235783

0x2356dc VMOVSD	-0x24554(%RIP),%XMM0

0x2356e4 MOV	%R9,-0x68(%RBP)

0x2356e8 NOPL	(%RAX,%RAX,1)

(156) 0x2356f0 MOV	%RCX,%RAX

(156) 0x2356f3 CQTO

(156) 0x2356f5 MOV	%RBX,%R11

(156) 0x2356f8 MOV	%RSI,%R9

(156) 0x2356fb IDIV	%R8

(156) 0x2356fe LEA	(%RSI,%RDX,1),%R10D

(156) 0x235702 MOV	%R13,%RSI

(156) 0x235705 MOV	%R15,%R13

(156) 0x235708 MOVSXD	%R10D,%R10

(156) 0x23570b LEA	(%R15,%RAX,1),%R8D

(156) 0x23570f ADD	%R14D,%EAX

(156) 0x235712 MOV	%R14,%R15

(156) 0x235715 MOVSXD	%R8D,%R8

(156) 0x235718 CLTQ

(156) 0x23571a IMUL	%R8,%R11

(156) 0x23571e IMUL	%RBX,%RAX

(156) 0x235722 LEA	(%R11,%R10,1),%R14

(156) 0x235726 VMOVSD	(%RDI,%R14,8),%XMM1

(156) 0x23572c LEA	(%RAX,%R10,1),%R14

(156) 0x235730 VADDSD	(%RDI,%R14,8),%XMM1,%XMM1

(156) 0x235736 MOV	%R15,%R14

(156) 0x235739 MOV	%R13,%R15

(156) 0x23573c MOV	%RSI,%R13

(156) 0x23573f ADD	%R13D,%EDX

(156) 0x235742 MOV	%R9,%RSI

(156) 0x235745 MOV	-0x68(%RBP),%R9

(156) 0x235749 MOVSXD	%EDX,%RDX

(156) 0x23574c ADD	%RDX,%RAX

(156) 0x23574f ADD	%RDX,%R11

(156) 0x235752 VADDSD	(%RDI,%RAX,8),%XMM1,%XMM1

(156) 0x235757 MOV	-0x70(%RBP),%RAX

(156) 0x23575b VADDSD	(%RDI,%R11,8),%XMM1,%XMM1

(156) 0x235761 IMUL	%RAX,%R8

(156) 0x235765 ADD	%R10,%R8

(156) 0x235768 VMULSD	%XMM0,%XMM1,%XMM1

(156) 0x23576c VMOVSD	%XMM1,(%R9,%R8,8)

(156) 0x235772 MOV	-0x30(%RBP),%R8

(156) 0x235776 CMP	%R12,%RCX

(156) 0x235779 LEA	0x1(%RCX),%RCX

(156) 0x23577d JL	2356f0

0x235783 MOV	-0x3c(%RBP),%ESI

0x235786 LEA	0x254eb(%RIP),%RDI

0x23578d VZEROUPPER

0x235790 CALL	259690 <@plt_start@+0x540>

0x235795 ADD	$0x58,%RSP

0x235799 POP	%RBX

0x23579a POP	%R12

0x23579c POP	%R13

0x23579e POP	%R14

0x2357a0 POP	%R15

0x2357a2 POP	%RBP

0x2357a3 RET

0x2357a4 NOPW	%CS:(%RAX,%RAX,1)

Coverage (%)	Name	Source Location	Module

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 aocc_8

Source file and lines	advec_mom.cpp:85-88
Module	exec

The function is defined in:

/home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h:69
/home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:85-88

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 14.00 to 5.00 cycles (2.80x speedup).

Workaround

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

Vectorization

Your function is not vectorized. Only 11% of vector register length is used (average across all SSE/AVX instructions).

Details

5% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

7% of SSE/AVX loads are used in vector version.
0% of SSE/AVX stores are used in vector version.
12% of SSE/AVX addition or subtraction instructions are used in vector version.
0% of SSE/AVX multiply instructions are used in vector version.
9% 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►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:85
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:85

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►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
VPBROADCASTD: 4 occurrences►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
VPBROADCASTQ: 4 occurrences►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86

Type of elements and instruction set

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 192 bytes. The binary function is storing 88 bytes.

General properties

nb instructions	104
nb uops	112
loop length	435
used x86 registers	16
used mmx registers	0
used xmm registers	6
used ymm registers	6
used zmm registers	0
nb stack references	13

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	14.00 cycles
front end	14.00 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	9.50	9.50	9.50	9.50	0.50	5.00	5.00	0.50	0.00	0.00
cycles	6.50	6.50	6.50	6.50	6.50	6.50	9.50	9.50	9.50	9.50	0.50	5.00	5.00	0.50	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	14.00
Dispatch	9.50
Overall L1	14.00

Vectorization ratios

INT

all	6%
load	9%
store	0%
mul	0%
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
other	10%

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

INT+FP

all	5%
load	7%
store	0%
mul	0%
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	9%

Vector efficiency ratios

INT

all	11%
load	13%
store	10%
mul	12%
add-sub	14%
fma	NA (no fma vectorizable/vectorized instructions)
other	12%

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

INT+FP

all	11%
load	13%
store	10%
mul	12%
add-sub	14%
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 14.00 cycles. At this rate:

5% of peak load performance is reached (13.71 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
4% of peak store performance is reached (6.29 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 14.00 to 9.50 cycles (1.47x speedup).

ASM code

In the binary file, the address of the function is: 235340

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	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	0	0	0	0.17	N/A
PUSH %R15	1	0	0	0	0	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	0	0	0	0	1	0.50	N/A
PUSH %R13	1	0	0	0	0	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	0	0	0	0	1	0.50	N/A
PUSH %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SUB $0x58,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV (%RCX),%R12D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (6.3%)
MOV (%RDX),%R14D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
LEA 0x1(%R14),%R15D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
ADD $0x3,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
SUB %R15D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
JLE 235795 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0x455>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
MOV (%R9),%EBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (6.3%)
MOV (%R8),%R13D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
ADD $0x4,%EBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
CMP %EBX,%R13D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
JG 235795 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0x455>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
LEA -0x1(%R13),%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
MOV (%RDI),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (6.3%)
MOVQ $0,-0x48(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (6.3%)
MOVQ $0x1,-0x78(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (6.3%)
MOVL $0,-0x40(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (6.3%)
SUB %EAX,%EBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV %RAX,-0x60(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
IMUL %RBX,%R12	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	scal (12.5%)
DEC %R12	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %R12,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
SUB $0x8,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
LEA -0x40(%RBP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
LEA -0x78(%RBP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
LEA 0x258a5(%RIP),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
LEA -0x48(%RBP),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
LEA -0x38(%RBP),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
MOV %ESI,-0x3c(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (6.3%)
MOV $0x22,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
PUSH $0x1	1	0	0	0	0	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	0	0	0	0	1	0.50	N/A
PUSH %RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CALL 259680 <@plt_start@+0x530>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0.50	N/A
ADD $0x20,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV -0x38(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV -0x48(%RBP),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
CMP %R12,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
CMOVL %RAX,%R12	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.33	scal (12.5%)
MOV %R12,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
CMP %R12,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
JG 235783 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0x443>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
MOV 0x10(%RBP),%RDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV 0x18(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV %RBX,-0x30(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
MOV -0x60(%RBP),%RSI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (12.5%)
MOV %R12,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
SUB %RCX,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
INC %R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV (%RDX),%R10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (12.5%)
MOV (%RAX),%RBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (12.5%)
MOV 0x10(%RAX),%RDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV 0x10(%RDX),%R9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV %R10,-0x70(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
CMP $0x4,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
JAE 23542f <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0xef>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
MOV -0x30(%RBP),%R8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
JMP 2356dc <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0x39c>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	1	N/A
VBROADCASTSD -0x242a8(%RIP),%YMM9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
VPBROADCASTQ %RCX,%YMM0	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
VPADDQ -0x24166(%RIP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	1	0.50	vect (50.0%)
VPBROADCASTQ -0x2427f(%RIP),%YMM10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	1	0.50	scal (12.5%)
MOV -0x30(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV %R11,-0x58(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
AND $-0x4,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
VPBROADCASTD %R15D,%XMM2	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTD %ESI,%XMM3	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTD %R14D,%XMM4	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTQ %RBX,%YMM5	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
VPBROADCASTD %R13D,%XMM6	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTQ %R10,%YMM7	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
MOV %RBX,-0x50(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
ADD %R11,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %R11,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
VPBROADCASTQ %RAX,%YMM1	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
VEXTRACTI128 $0x1,%YMM1,%XMM8	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	3	0.50	vect (25.0%)
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
MOV -0x30(%RBP),%R8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
CMP %RBX,-0x58(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
MOV -0x50(%RBP),%RBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (12.5%)
JE 235783 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0x443>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
VMOVSD -0x24554(%RIP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.50	scal (12.5%)
MOV %R9,-0x68(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
NOPL (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
MOV -0x3c(%RBP),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (6.3%)
LEA 0x254eb(%RIP),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 259690 <@plt_start@+0x540>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0.50	N/A
ADD $0x58,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0	0	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	0	0	0	0	1	0.33	N/A
POP %R13	1	0	0	0	0	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	0	0	0	0	1	0.33	N/A
POP %R15	1	0	0	0	0	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	0	0	0	0	1	0.33	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	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	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 aocc_8

Source file and lines	advec_mom.cpp:85-88
Module	exec

The function is defined in:

/home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h:69
/home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:85-88

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 14.00 to 5.00 cycles (2.80x speedup).

Workaround

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

Vectorization

Your function is not vectorized. Only 11% of vector register length is used (average across all SSE/AVX instructions).

Details

5% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

7% of SSE/AVX loads are used in vector version.
0% of SSE/AVX stores are used in vector version.
12% of SSE/AVX addition or subtraction instructions are used in vector version.
0% of SSE/AVX multiply instructions are used in vector version.
9% 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►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:85
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:85

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►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
VPBROADCASTD: 4 occurrences►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
VPBROADCASTQ: 4 occurrences►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:86

Type of elements and instruction set

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 192 bytes. The binary function is storing 88 bytes.

General properties

nb instructions	104
nb uops	112
loop length	435
used x86 registers	16
used mmx registers	0
used xmm registers	6
used ymm registers	6
used zmm registers	0
nb stack references	13

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	14.00 cycles
front end	14.00 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	9.50	9.50	9.50	9.50	0.50	5.00	5.00	0.50	0.00	0.00
cycles	6.50	6.50	6.50	6.50	6.50	6.50	9.50	9.50	9.50	9.50	0.50	5.00	5.00	0.50	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	14.00
Dispatch	9.50
Overall L1	14.00

Vectorization ratios

INT

all	6%
load	9%
store	0%
mul	0%
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
other	10%

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

INT+FP

all	5%
load	7%
store	0%
mul	0%
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	9%

Vector efficiency ratios

INT

all	11%
load	13%
store	10%
mul	12%
add-sub	14%
fma	NA (no fma vectorizable/vectorized instructions)
other	12%

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

INT+FP

all	11%
load	13%
store	10%
mul	12%
add-sub	14%
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 14.00 cycles. At this rate:

5% of peak load performance is reached (13.71 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
4% of peak store performance is reached (6.29 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: 235340

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	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	0	0	0	0.17	N/A
PUSH %R15	1	0	0	0	0	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	0	0	0	0	1	0.50	N/A
PUSH %R13	1	0	0	0	0	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	0	0	0	0	1	0.50	N/A
PUSH %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
SUB $0x58,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV (%RCX),%R12D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (6.3%)
MOV (%RDX),%R14D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
LEA 0x1(%R14),%R15D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
ADD $0x3,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
SUB %R15D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
JLE 235795 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0x455>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
MOV (%R9),%EBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (6.3%)
MOV (%R8),%R13D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
ADD $0x4,%EBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
CMP %EBX,%R13D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
JG 235795 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0x455>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
LEA -0x1(%R13),%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
MOV (%RDI),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (6.3%)
MOVQ $0,-0x48(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (6.3%)
MOVQ $0x1,-0x78(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (6.3%)
MOVL $0,-0x40(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (6.3%)
SUB %EAX,%EBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV %RAX,-0x60(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
IMUL %RBX,%R12	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	3	0.33	scal (12.5%)
DEC %R12	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %R12,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
SUB $0x8,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
LEA -0x40(%RBP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
LEA -0x78(%RBP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
LEA 0x258a5(%RIP),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
LEA -0x48(%RBP),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
LEA -0x38(%RBP),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
MOV %ESI,-0x3c(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (6.3%)
MOV $0x22,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
PUSH $0x1	1	0	0	0	0	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	0	0	0	0	1	0.50	N/A
PUSH %RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CALL 259680 <@plt_start@+0x530>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0.50	N/A
ADD $0x20,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV -0x38(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV -0x48(%RBP),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
CMP %R12,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
CMOVL %RAX,%R12	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.33	scal (12.5%)
MOV %R12,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
CMP %R12,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
JG 235783 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0x443>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
MOV 0x10(%RBP),%RDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV 0x18(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV %RBX,-0x30(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
MOV -0x60(%RBP),%RSI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (12.5%)
MOV %R12,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
SUB %RCX,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
INC %R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV (%RDX),%R10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (12.5%)
MOV (%RAX),%RBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (12.5%)
MOV 0x10(%RAX),%RDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV 0x10(%RDX),%R9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV %R10,-0x70(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
CMP $0x4,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
JAE 23542f <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0xef>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
MOV -0x30(%RBP),%R8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
JMP 2356dc <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0x39c>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	1	N/A
VBROADCASTSD -0x242a8(%RIP),%YMM9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
VPBROADCASTQ %RCX,%YMM0	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
VPADDQ -0x24166(%RIP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	1	0.50	vect (50.0%)
VPBROADCASTQ -0x2427f(%RIP),%YMM10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	1	0.50	scal (12.5%)
MOV -0x30(%RBP),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
MOV %R11,-0x58(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
AND $-0x4,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
VPBROADCASTD %R15D,%XMM2	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTD %ESI,%XMM3	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTD %R14D,%XMM4	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTQ %RBX,%YMM5	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
VPBROADCASTD %R13D,%XMM6	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTQ %R10,%YMM7	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
MOV %RBX,-0x50(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
ADD %R11,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %R11,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
VPBROADCASTQ %RAX,%YMM1	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
VEXTRACTI128 $0x1,%YMM1,%XMM8	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	3	0.50	vect (25.0%)
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
MOV -0x30(%RBP),%R8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	N/A
CMP %RBX,-0x58(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	0	0	0	0	1	0.25	scal (12.5%)
MOV -0x50(%RBP),%RBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (12.5%)
JE 235783 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.8+0x443>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
VMOVSD -0x24554(%RIP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	0.50	scal (12.5%)
MOV %R9,-0x68(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	4	0.50	scal (12.5%)
NOPL (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
MOV -0x3c(%RBP),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	3	0.25	scal (6.3%)
LEA 0x254eb(%RIP),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.25	N/A
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 259690 <@plt_start@+0x540>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0.50	N/A
ADD $0x58,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0	0	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	0	0	0	0	1	0.33	N/A
POP %R13	1	0	0	0	0	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	0	0	0	0	1	0.33	N/A
POP %R15	1	0	0	0	0	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	0	0	0	0	1	0.33	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	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	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...–	1.32	0.49
○Loop 155 - advec_mom.cpp:86-88 - exec	1.32	0.49
○Loop 156 - advec_mom.cpp:86-88 - exec	0.00	0.00

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

512-bits vectorization

Workaround

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

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

512-bits vectorization

Workaround

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