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 .extracted]

	Function: advec_mom_kernel(int, int, int, int, clover::Buffer2D<double>&, clover::Buffer2D<double>&, ...	Module: exec	Source: advec_mom.cpp:218-222 [...]	Coverage (incl. loops): 3.32% \| (excl. loops): 0.00%

/home/eoseret/qaas_runs_GNR/173-814-2343/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp: 218 - 222

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

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

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

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

221:         vel1(i, j) = (vel1(i, j) * node_mass_pre(i, j) + mom_flux(i + 0, j - 1) - mom_flux(i, j)) / node_mass_post(i, j);

222:       }

/home/eoseret/qaas_runs_GNR/173-814-2343/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]; }

0x4222a0 PUSH	%RBP

0x4222a1 MOV	%RSP,%RBP

0x4222a4 PUSH	%R15

0x4222a6 PUSH	%R14

0x4222a8 PUSH	%R13

0x4222aa PUSH	%R12

0x4222ac PUSH	%RBX

0x4222ad SUB	$0xd8,%RSP

0x4222b4 MOV	0x30(%RBP),%RAX

0x4222b8 MOV	0x20(%RBP),%R14

0x4222bc MOV	0x18(%RBP),%R10

0x4222c0 MOV	0x10(%RBP),%EBX

0x4222c3 MOVL	$0,-0x38(%RBP)

0x4222ca TEST	%RAX,%RAX

0x4222cd JS	422729

0x4222d3 MOV	%R9,%R15

0x4222d6 MOV	%R8,%R12

0x4222d9 MOV	%RDX,%R13

0x4222dc MOV	%RCX,-0x30(%RBP)

0x4222e0 MOV	(%RDI),%ESI

0x4222e2 MOVQ	$0,-0x50(%RBP)

0x4222ea MOV	%RAX,-0x48(%RBP)

0x4222ee MOVQ	$0x1,-0x98(%RBP)

0x4222f9 SUB	$0x8,%RSP

0x4222fd LEA	-0x98(%RBP),%RAX

0x422304 LEA	-0x38(%RBP),%RCX

0x422308 LEA	-0x50(%RBP),%R8

0x42230c LEA	-0x48(%RBP),%R9

0x422310 MOV	$0x4b7730,%EDI

0x422315 MOV	%ESI,-0x34(%RBP)

0x422318 MOV	$0x22,%EDX

0x42231d PUSH	$0x1

0x42231f PUSH	$0x1

0x422321 PUSH	%RAX

0x422322 MOV	%R10,-0x40(%RBP)

0x422326 CALL	4031f0 <__kmpc_for_static_init_8@plt>

0x42232b MOV	-0x40(%RBP),%R10

0x42232f ADD	$0x20,%RSP

0x422333 MOV	-0x50(%RBP),%RCX

0x422337 MOV	-0x48(%RBP),%RDX

0x42233b CMP	%RDX,%RCX

0x42233e JA	42270b

0x422344 MOV	%R14,%R9

0x422347 MOV	%RDX,%RAX

0x42234a SUB	%R10D,%R9D

0x42234d MOV	(%R13),%RDX

0x422351 MOV	0x10(%R13),%R8

0x422355 MOV	(%R12),%RSI

0x422359 MOV	0x10(%R12),%RDI

0x42235e MOV	(%R15),%R11

0x422361 MOV	0x10(%R15),%R15

0x422365 MOV	-0x30(%RBP),%R14

0x422369 MOV	(%R14),%R12

0x42236c MOV	0x10(%R14),%R14

0x422370 INC	%RAX

0x422373 MOV	%RAX,-0x30(%RBP)

0x422377 SUB	%RCX,%RAX

0x42237a MOV	$-0x2,%R13D

0x422380 AND	%RAX,%R13

0x422383 MOV	%RSI,-0x90(%RBP)

0x42238a MOV	%RDI,-0x88(%RBP)

0x422391 MOV	%R12,-0x80(%RBP)

0x422395 JE	42273b

0x42239b MOV	%RAX,-0x58(%RBP)

0x42239f MOV	%R9,-0x68(%RBP)

0x4223a3 MOVD	%R9D,%XMM0

0x4223a8 PSHUFD	$0x44,%XMM0,%XMM1

0x4223ad MOVD	%EBX,%XMM0

0x4223b1 PSHUFD	$0x50,%XMM0,%XMM0

0x4223b6 MOVDQA	%XMM0,-0xf0(%RBP)

0x4223be MOVQ	%R10,%XMM0

0x4223c3 PSHUFD	$0x44,%XMM0,%XMM0

0x4223c8 MOVDQA	%XMM0,-0xe0(%RBP)

0x4223d0 MOV	%RDX,-0x60(%RBP)

0x4223d4 MOVQ	%RDX,%XMM0

0x4223d9 PSHUFD	$0x44,%XMM0,%XMM9

0x4223df MOVQ	%RSI,%XMM0

0x4223e4 PSHUFD	$0x44,%XMM0,%XMM11

0x4223ea MOVQ	%R11,%XMM0

0x4223ef PSHUFD	$0x44,%XMM0,%XMM14

0x4223f5 MOVQ	%R12,%XMM0

0x4223fa PSHUFD	$0x44,%XMM0,%XMM6

0x4223ff MOVQ	%RCX,%XMM0

0x422404 PSHUFD	$0x44,%XMM0,%XMM7

0x422409 PADDQ	0x7901f(%RIP),%XMM7

0x422411 XOR	%R10D,%R10D

0x422414 MOVQ	%XMM1,-0x78(%RBP)

0x422419 MOVDQA	%XMM1,-0x100(%RBP)

0x422421 PSHUFD	$-0x12,%XMM1,%XMM0

0x422426 MOVQ	%XMM0,-0x70(%RBP)

0x42242b MOVDQA	%XMM9,%XMM0

0x422430 PSRLQ	$0x20,%XMM0

0x422435 MOVDQA	%XMM0,-0xd0(%RBP)

0x42243d MOVQ	%R8,%XMM1

0x422442 PSHUFD	$0x44,%XMM1,%XMM0

0x422447 MOVDQA	%XMM0,-0xc0(%RBP)

0x42244f MOVDQA	%XMM11,%XMM10

0x422454 PSRLQ	$0x20,%XMM10

0x42245a MOVQ	%RDI,%XMM1

0x42245f PSHUFD	$0x44,%XMM1,%XMM0

0x422464 MOVDQA	%XMM0,-0xb0(%RBP)

0x42246c MOVDQA	%XMM14,%XMM1

0x422471 PSRLQ	$0x20,%XMM1

0x422476 NOPW	%CS:(%RAX,%RAX,1)

(162) 0x422480 MOVQ	%XMM7,%RAX

(162) 0x422485 XOR	%EDX,%EDX

(162) 0x422487 DIVQ	-0x78(%RBP)

(162) 0x42248b PSHUFD	$-0x12,%XMM7,%XMM2

(162) 0x422490 MOVQ	%XMM2,%RDX

(162) 0x422495 MOVQ	%RAX,%XMM2

(162) 0x42249a MOV	%RDX,%RAX

(162) 0x42249d XOR	%EDX,%EDX

(162) 0x42249f DIVQ	-0x70(%RBP)

(162) 0x4224a3 MOVQ	%RAX,%XMM3

(162) 0x4224a8 PUNPCKLQDQ	%XMM3,%XMM2

(162) 0x4224ac PSHUFD	$-0x18,%XMM2,%XMM15

(162) 0x4224b2 PADDD	-0xf0(%RBP),%XMM15

(162) 0x4224bb PMULUDQ	-0x100(%RBP),%XMM2

(162) 0x4224c3 MOVDQA	%XMM7,%XMM3

(162) 0x4224c7 PSUBQ	%XMM2,%XMM3

(162) 0x4224cb PADDQ	-0xe0(%RBP),%XMM3

(162) 0x4224d3 PSLLQ	$0x20,%XMM3

(162) 0x4224d8 PXOR	%XMM8,%XMM8

(162) 0x4224dd PCMPGTD	%XMM15,%XMM8

(162) 0x4224e2 PSHUFD	$-0x13,%XMM3,%XMM13

(162) 0x4224e8 MOVDQA	%XMM15,%XMM2

(162) 0x4224ed PUNPCKLDQ	%XMM8,%XMM2

(162) 0x4224f2 MOVDQA	%XMM9,%XMM5

(162) 0x4224f7 MOVDQA	%XMM13,%XMM4

(162) 0x4224fc MOVDQA	-0xd0(%RBP),%XMM0

(162) 0x422504 PMULUDQ	%XMM2,%XMM0

(162) 0x422508 PUNPCKLDQ	0x7828f(%RIP),%XMM8

(162) 0x422511 PSRAD	$0x1f,%XMM4

(162) 0x422516 MOVDQA	%XMM9,%XMM3

(162) 0x42251b PMULUDQ	%XMM8,%XMM3

(162) 0x422520 PADDQ	%XMM0,%XMM3

(162) 0x422524 PUNPCKLDQ	%XMM4,%XMM13

(162) 0x422529 MOVDQA	%XMM10,%XMM0

(162) 0x42252e PMULUDQ	%XMM2,%XMM0

(162) 0x422532 MOVDQA	%XMM11,%XMM4

(162) 0x422537 PMULUDQ	%XMM2,%XMM5

(162) 0x42253b PMULUDQ	%XMM8,%XMM4

(162) 0x422540 PADDQ	%XMM0,%XMM4

(162) 0x422544 MOVDQA	%XMM11,%XMM0

(162) 0x422549 PSLLQ	$0x20,%XMM3

(162) 0x42254e PMULUDQ	%XMM2,%XMM0

(162) 0x422552 PSLLQ	$0x20,%XMM4

(162) 0x422557 PADDD	0x79060(%RIP),%XMM15

(162) 0x422560 PADDQ	%XMM5,%XMM3

(162) 0x422564 PXOR	%XMM12,%XMM12

(162) 0x422569 PCMPGTD	%XMM15,%XMM12

(162) 0x42256e PUNPCKLDQ	%XMM12,%XMM15

(162) 0x422573 PADDQ	%XMM0,%XMM4

(162) 0x422577 MOVDQA	%XMM14,%XMM5

(162) 0x42257c PMULUDQ	%XMM15,%XMM5

(162) 0x422581 PSHUFD	$-0x2c,%XMM12,%XMM0

(162) 0x422587 PMULUDQ	%XMM1,%XMM15

(162) 0x42258c PMULUDQ	%XMM14,%XMM0

(162) 0x422591 PADDQ	%XMM15,%XMM0

(162) 0x422596 PADDQ	%XMM13,%XMM3

(162) 0x42259b PSLLQ	$0x3,%XMM3

(162) 0x4225a0 PADDQ	-0xc0(%RBP),%XMM3

(162) 0x4225a8 MOVQ	%XMM3,%RAX

(162) 0x4225ad PSHUFD	$-0x12,%XMM3,%XMM3

(162) 0x4225b2 MOVQ	%XMM3,%RDX

(162) 0x4225b7 PADDQ	%XMM13,%XMM4

(162) 0x4225bc PSLLQ	$0x3,%XMM4

(162) 0x4225c1 PADDQ	-0xb0(%RBP),%XMM4

(162) 0x4225c9 MOVQ	%XMM4,%R9

(162) 0x4225ce PSHUFD	$-0x12,%XMM4,%XMM3

(162) 0x4225d3 PSLLQ	$0x20,%XMM0

(162) 0x4225d8 PADDQ	%XMM13,%XMM5

(162) 0x4225dd MOVQ	%XMM3,%RSI

(162) 0x4225e2 PADDQ	%XMM0,%XMM5

(162) 0x4225e6 MOVDQA	%XMM1,%XMM0

(162) 0x4225ea MOVDQA	%XMM14,%XMM3

(162) 0x4225ef PMULUDQ	%XMM2,%XMM0

(162) 0x4225f3 PMULUDQ	%XMM8,%XMM3

(162) 0x4225f8 PADDQ	%XMM0,%XMM3

(162) 0x4225fc MOVDQA	%XMM14,%XMM0

(162) 0x422601 PMULUDQ	%XMM2,%XMM0

(162) 0x422605 PSLLQ	$0x20,%XMM3

(162) 0x42260a PADDQ	%XMM0,%XMM3

(162) 0x42260e MOVQ	%R15,%XMM0

(162) 0x422613 PSLLQ	$0x3,%XMM5

(162) 0x422618 PSHUFD	$0x44,%XMM0,%XMM0

(162) 0x42261d PADDQ	%XMM13,%XMM3

(162) 0x422622 PSLLQ	$0x3,%XMM3

(162) 0x422627 PADDQ	%XMM0,%XMM5

(162) 0x42262b PADDQ	%XMM0,%XMM3

(162) 0x42262f MOVQ	%XMM3,%RDI

(162) 0x422634 PSHUFD	$-0x12,%XMM3,%XMM0

(162) 0x422639 MOVQ	%XMM0,%R12

(162) 0x42263e MOVDQA	%XMM6,%XMM0

(162) 0x422642 MOVDQA	%XMM6,%XMM3

(162) 0x422646 PSRLQ	$0x20,%XMM3

(162) 0x42264b PMULUDQ	%XMM2,%XMM0

(162) 0x42264f PMULUDQ	%XMM2,%XMM3

(162) 0x422653 PMULUDQ	%XMM6,%XMM8

(162) 0x422658 PADDQ	%XMM3,%XMM8

(162) 0x42265d PSLLQ	$0x20,%XMM8

(162) 0x422663 PADDQ	%XMM0,%XMM8

(162) 0x422668 MOVSD	(%R9),%XMM0

(162) 0x42266d MOVHPD	(%RSI),%XMM0

(162) 0x422671 PADDQ	%XMM13,%XMM8

(162) 0x422676 MOVQ	%R14,%XMM2

(162) 0x42267b PSHUFD	$0x44,%XMM2,%XMM2

(162) 0x422680 PSLLQ	$0x3,%XMM8

(162) 0x422686 PADDQ	%XMM8,%XMM2

(162) 0x42268b MOVQ	%XMM5,%RSI

(162) 0x422690 PSHUFD	$-0x12,%XMM5,%XMM3

(162) 0x422695 MOVQ	%XMM3,%R9

(162) 0x42269a MOVSD	(%RAX),%XMM3

(162) 0x42269e MOVHPD	(%RDX),%XMM3

(162) 0x4226a2 MULPD	%XMM3,%XMM0

(162) 0x4226a6 MOVSD	(%RSI),%XMM3

(162) 0x4226aa MOVHPD	(%R9),%XMM3

(162) 0x4226af ADDPD	%XMM0,%XMM3

(162) 0x4226b3 MOVSD	(%RDI),%XMM0

(162) 0x4226b7 MOVHPD	(%R12),%XMM0

(162) 0x4226bd SUBPD	%XMM0,%XMM3

(162) 0x4226c1 MOVQ	%XMM2,%RSI

(162) 0x4226c6 PSHUFD	$-0x12,%XMM2,%XMM0

(162) 0x4226cb MOVQ	%XMM0,%RDI

(162) 0x4226d0 MOVSD	(%RSI),%XMM0

(162) 0x4226d4 MOVHPD	(%RDI),%XMM0

(162) 0x4226d8 DIVPD	%XMM0,%XMM3

(162) 0x4226dc MOVLPD	%XMM3,(%RAX)

(162) 0x4226e0 MOVHPD	%XMM3,(%RDX)

(162) 0x4226e4 PADDQ	0x78d64(%RIP),%XMM7

(162) 0x4226ec ADD	$0x2,%R10

(162) 0x4226f0 CMP	%R13,%R10

(162) 0x4226f3 JB	422480

0x4226f9 CMP	%R13,-0x58(%RBP)

0x4226fd MOV	-0x68(%RBP),%R9

0x422701 MOV	-0x40(%RBP),%R10

0x422705 MOV	-0x60(%RBP),%R12

0x422709 JNE	422743

0x42270b MOV	$0x4b7750,%EDI

0x422710 MOV	-0x34(%RBP),%ESI

0x422713 ADD	$0xd8,%RSP

0x42271a POP	%RBX

0x42271b POP	%R12

0x42271d POP	%R13

0x42271f POP	%R14

0x422721 POP	%R15

0x422723 POP	%RBP

0x422724 JMP	403050

0x422729 ADD	$0xd8,%RSP

0x422730 POP	%RBX

0x422731 POP	%R12

0x422733 POP	%R13

0x422735 POP	%R14

0x422737 POP	%R15

0x422739 POP	%RBP

0x42273a RET

0x42273b MOV	%RDX,%R12

0x42273e JMP	4227d4

0x422743 ADD	%R13,%RCX

0x422746 JMP	4227d4

0x42274b NOPL	(%RAX,%RAX,1)

(161) 0x422750 MOV	%RCX,%RAX

(161) 0x422753 XOR	%EDX,%EDX

(161) 0x422755 DIV	%R9

(161) 0x422758 LEA	(%RAX,%RBX,1),%EDI

(161) 0x42275b ADD	%R10D,%EDX

(161) 0x42275e MOVSXD	%EDX,%RSI

(161) 0x422761 MOVSXD	%EDI,%RDX

(161) 0x422764 MOV	-0x90(%RBP),%RDI

(161) 0x42276b IMUL	%RDX,%RDI

(161) 0x42276f ADD	%RSI,%RDI

(161) 0x422772 MOV	-0x88(%RBP),%R13

(161) 0x422779 MOVSD	(%R13,%RDI,8),%XMM0

(161) 0x422780 MOV	%R12,%RDI

(161) 0x422783 IMUL	%RDX,%RDI

(161) 0x422787 ADD	%RSI,%RDI

(161) 0x42278a MULSD	(%R8,%RDI,8),%XMM0

(161) 0x422790 ADD	%EBX,%EAX

(161) 0x422792 DEC	%EAX

(161) 0x422794 CLTQ

(161) 0x422796 IMUL	%R11,%RAX

(161) 0x42279a ADD	%RSI,%RAX

(161) 0x42279d ADDSD	(%R15,%RAX,8),%XMM0

(161) 0x4227a3 MOV	%R11,%RAX

(161) 0x4227a6 IMUL	%RDX,%RAX

(161) 0x4227aa ADD	%RSI,%RAX

(161) 0x4227ad SUBSD	(%R15,%RAX,8),%XMM0

(161) 0x4227b3 IMUL	-0x80(%RBP),%RDX

(161) 0x4227b8 ADD	%RSI,%RDX

(161) 0x4227bb DIVSD	(%R14,%RDX,8),%XMM0

(161) 0x4227c1 MOVSD	%XMM0,(%R8,%RDI,8)

(161) 0x4227c7 INC	%RCX

(161) 0x4227ca CMP	%RCX,-0x30(%RBP)

(161) 0x4227ce JE	42270b

(161) 0x4227d4 MOV	%RCX,%RAX

(161) 0x4227d7 SHR	$0x20,%RAX

(161) 0x4227db JNE	422750

(161) 0x4227e1 MOV	%ECX,%EAX

(161) 0x4227e3 XOR	%EDX,%EDX

(161) 0x4227e5 DIV	%R9D

(161) 0x4227e8 JMP	422758

0x4227ed NOPL	(%RAX)

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

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	advec_mom.cpp:218-222
Module	exec

The function is defined in:

/home/eoseret/qaas_runs_GNR/173-814-2343/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:218-222
/home/eoseret/qaas_runs_GNR/173-814-2343/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h:69

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 22.33 to 10.50 cycles (2.13x speedup).

Workaround

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

Vectorization

Your function is poorly vectorized. Only 15% of vector register length is used (average across all SSE/AVX instructions). By fully vectorizing your function, you can lower the cost of an iteration from 22.33 to 2.36 cycles (9.46x speedup).

Details

32% 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.
27% of SSE/AVX stores are used in vector version.
14% of SSE/AVX addition or subtraction instructions are used in vector version.
51% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

Since your execution units are vector units, only a fully 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.

__kmpc_for_static_init_8@plt: 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-814-2343/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:218

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 180 bytes. The binary function is storing 216 bytes.

General properties

nb instructions	133
nb uops	134
loop length	570
used x86 registers	16
used mmx registers	0
used xmm registers	8
used ymm registers	0
used zmm registers	0
nb stack references	25

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	22.33 cycles
front end	22.33 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	7.70	10.50	12.00	12.00	16.00	10.50	7.70	16.00	16.00	16.00	7.60	12.00
cycles	7.70	10.50	12.00	12.00	16.00	10.50	7.70	16.00	16.00	16.00	7.60	12.00

Execution ports to units layout:

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

Cycles executing div or sqrt instructions

Cycles summary

Front-end	22.33
Dispatch	16.00
Overall L1	22.33

Vectorization ratios

all	32%
load	7%
store	27%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	14%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	51%

Vector efficiency ratios

all	15%
load	12%
store	14%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	13%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	18%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 22.33 cycles. At this rate:

4% of peak load performance is reached (8.06 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
7% of peak store performance is reached (9.67 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 22.33 to 16.00 cycles (1.40x speedup).

ASM code

In the binary file, the address of the function is: 4222a0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
PUSH %R15	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %R14	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %R13	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %R12	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %RBX	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
SUB $0xd8,%RSP	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
MOV 0x30(%RBP),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x20(%RBP),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x18(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x10(%RBP),%EBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
MOVL $0,-0x38(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
TEST %RAX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (12.5%)
JS 422729 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x489>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %R9,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %R8,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RDX,%R13	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RCX,-0x30(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV (%RDI),%ESI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
MOVQ $0,-0x50(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOV %RAX,-0x48(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVQ $0x1,-0x98(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
SUB $0x8,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
LEA -0x98(%RBP),%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x38(%RBP),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x50(%RBP),%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x48(%RBP),%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x4b7730,%EDI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV %ESI,-0x34(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOV $0x22,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
PUSH $0x1	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH $0x1	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %RAX	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %R10,-0x40(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
CALL 4031f0 <__kmpc_for_static_init_8@plt>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV -0x40(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
ADD $0x20,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV -0x50(%RBP),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV -0x48(%RBP),%RDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
CMP %RDX,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JA 42270b <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x46b>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %R14,%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RDX,%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
SUB %R10D,%R9D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
MOV (%R13),%RDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x10(%R13),%R8	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV (%R12),%RSI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x10(%R12),%RDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV (%R15),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x10(%R15),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV -0x30(%RBP),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV (%R14),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x10(%R14),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
INC %RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RAX,-0x30(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
SUB %RCX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV $-0x2,%R13D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
AND %RAX,%R13	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
MOV %RSI,-0x90(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %RDI,-0x88(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %R12,-0x80(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
JE 42273b <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x49b>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %RAX,-0x58(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %R9,-0x68(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVD %R9D,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	scal (6.3%)
PSHUFD $0x44,%XMM0,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVD %EBX,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	scal (6.3%)
PSHUFD $0x50,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM0,-0xf0(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
MOVQ %R10,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM0,-0xe0(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
MOV %RDX,-0x60(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVQ %RDX,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM9	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %RSI,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM11	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %R11,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM14	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %R12,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM6	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %RCX,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM7	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
PADDQ 0x7901f(%RIP),%XMM7	1	0.33	0.33	0.33	0.33	0	0.33	0	0	0	0	0	0.33	1	0.33	vect (25.0%)
XOR %R10D,%R10D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
MOVQ %XMM1,-0x78(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVDQA %XMM1,-0x100(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
PSHUFD $-0x12,%XMM1,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %XMM0,-0x70(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVDQA %XMM9,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM0	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	0.50	vect (25.0%)
MOVDQA %XMM0,-0xd0(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
MOVQ %R8,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM0,-0xc0(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
MOVDQA %XMM11,%XMM10	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM10	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	0.50	vect (25.0%)
MOVQ %RDI,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM0,-0xb0(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
MOVDQA %XMM14,%XMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM1	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	0.50	vect (25.0%)
NOPW %CS:(%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMP %R13,-0x58(%RBP)	1	0.20	0.20	0.33	0.33	0	0.20	0.20	0	0	0	0.20	0.33	1	0.33	scal (12.5%)
MOV -0x68(%RBP),%R9	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x40(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x60(%RBP),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
JNE 422743 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x4a3>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV $0x4b7750,%EDI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV -0x34(%RBP),%ESI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
ADD $0xd8,%RSP	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
POP %RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R13	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %RBP	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
JMP 403050 <__kmpc_for_static_fini@plt>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
ADD $0xd8,%RSP	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
POP %RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R13	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %RBP	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
RET	1	0.50	0	0.33	0.33	0	0	0.50	0	0	0	0	0.33	0	2.13	N/A
MOV %RDX,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
JMP 4227d4 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x534>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
ADD %R13,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JMP 4227d4 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x534>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A

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

Source file and lines	advec_mom.cpp:218-222
Module	exec

The function is defined in:

/home/eoseret/qaas_runs_GNR/173-814-2343/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:218-222
/home/eoseret/qaas_runs_GNR/173-814-2343/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/context.h:69

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Code clean check

Workaround

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

Vectorization

Details

32% 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.
27% of SSE/AVX stores are used in vector version.
14% of SSE/AVX addition or subtraction instructions are used in vector version.
51% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

Since your execution units are vector units, only a fully 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

__kmpc_for_static_init_8@plt: 1 occurrences►
- /home/eoseret/qaas_runs_GNR/173-814-2343/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:218

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 180 bytes. The binary function is storing 216 bytes.

General properties

nb instructions	133
nb uops	134
loop length	570
used x86 registers	16
used mmx registers	0
used xmm registers	8
used ymm registers	0
used zmm registers	0
nb stack references	25

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	22.33 cycles
front end	22.33 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	7.70	10.50	12.00	12.00	16.00	10.50	7.70	16.00	16.00	16.00	7.60	12.00
cycles	7.70	10.50	12.00	12.00	16.00	10.50	7.70	16.00	16.00	16.00	7.60	12.00

Execution ports to units layout:

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

Cycles executing div or sqrt instructions

Cycles summary

Front-end	22.33
Dispatch	16.00
Overall L1	22.33

Vectorization ratios

all	32%
load	7%
store	27%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	14%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	51%

Vector efficiency ratios

all	15%
load	12%
store	14%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	13%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	18%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 22.33 cycles. At this rate:

4% of peak load performance is reached (8.06 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
7% of peak store performance is reached (9.67 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: 4222a0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
PUSH %R15	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %R14	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %R13	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %R12	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %RBX	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
SUB $0xd8,%RSP	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
MOV 0x30(%RBP),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x20(%RBP),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x18(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x10(%RBP),%EBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
MOVL $0,-0x38(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
TEST %RAX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (12.5%)
JS 422729 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x489>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %R9,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %R8,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RDX,%R13	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RCX,-0x30(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV (%RDI),%ESI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
MOVQ $0,-0x50(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOV %RAX,-0x48(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVQ $0x1,-0x98(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
SUB $0x8,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
LEA -0x98(%RBP),%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x38(%RBP),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x50(%RBP),%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x48(%RBP),%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x4b7730,%EDI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV %ESI,-0x34(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (6.3%)
MOV $0x22,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
PUSH $0x1	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH $0x1	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
PUSH %RAX	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %R10,-0x40(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
CALL 4031f0 <__kmpc_for_static_init_8@plt>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV -0x40(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
ADD $0x20,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV -0x50(%RBP),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV -0x48(%RBP),%RDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
CMP %RDX,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JA 42270b <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x46b>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %R14,%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RDX,%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
SUB %R10D,%R9D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
MOV (%R13),%RDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x10(%R13),%R8	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV (%R12),%RSI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x10(%R12),%RDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV (%R15),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x10(%R15),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV -0x30(%RBP),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV (%R14),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x10(%R14),%R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
INC %RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RAX,-0x30(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
SUB %RCX,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV $-0x2,%R13D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
AND %RAX,%R13	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
MOV %RSI,-0x90(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %RDI,-0x88(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %R12,-0x80(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
JE 42273b <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x49b>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %RAX,-0x58(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %R9,-0x68(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVD %R9D,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	scal (6.3%)
PSHUFD $0x44,%XMM0,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVD %EBX,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	scal (6.3%)
PSHUFD $0x50,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM0,-0xf0(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
MOVQ %R10,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM0,-0xe0(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
MOV %RDX,-0x60(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVQ %RDX,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM9	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %RSI,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM11	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %R11,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM14	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %R12,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM6	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %RCX,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM7	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
PADDQ 0x7901f(%RIP),%XMM7	1	0.33	0.33	0.33	0.33	0	0.33	0	0	0	0	0	0.33	1	0.33	vect (25.0%)
XOR %R10D,%R10D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
MOVQ %XMM1,-0x78(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVDQA %XMM1,-0x100(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
PSHUFD $-0x12,%XMM1,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %XMM0,-0x70(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOVDQA %XMM9,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM0	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	0.50	vect (25.0%)
MOVDQA %XMM0,-0xd0(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
MOVQ %R8,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM0,-0xc0(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
MOVDQA %XMM11,%XMM10	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM10	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	0.50	vect (25.0%)
MOVQ %RDI,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM0,-0xb0(%RBP)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0-1	0.50	vect (25.0%)
MOVDQA %XMM14,%XMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM1	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	0.50	vect (25.0%)
NOPW %CS:(%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMP %R13,-0x58(%RBP)	1	0.20	0.20	0.33	0.33	0	0.20	0.20	0	0	0	0.20	0.33	1	0.33	scal (12.5%)
MOV -0x68(%RBP),%R9	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x40(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x60(%RBP),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
JNE 422743 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x4a3>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV $0x4b7750,%EDI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV -0x34(%RBP),%ESI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
ADD $0xd8,%RSP	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
POP %RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R13	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %RBP	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
JMP 403050 <__kmpc_for_static_fini@plt>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
ADD $0xd8,%RSP	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
POP %RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R13	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R14	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
POP %RBP	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
RET	1	0.50	0	0.33	0.33	0	0	0.50	0	0	0	0	0.33	0	2.13	N/A
MOV %RDX,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
JMP 4227d4 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x534>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
ADD %R13,%RCX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JMP 4227d4 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.extracted+0x534>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A

Name	Coverage (%)	Time (s)
▼advec_mom_kernel(int, int, int, int, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clo...–	3.32	1.31
○Loop 162 - advec_mom.cpp:219-222 - exec	3.32	1.31
○Loop 161 - advec_mom.cpp:219-222 - exec	0.00	0.00

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

CALL instructions

Details

Type of elements and instruction set

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

CALL instructions

Details

Type of elements and instruction set

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code