OV - exec - Function revert_kernel(int, int, int, int, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&, clover::Buffer2D<double>&) [clone .extracted]

	Function: revert_kernel(int, int, int, int, clover::Buffer2D<double>&, clover::Buffer2D<double>&, cl ...	Module: exec	Source: revert.cpp:34-39 [...]	Coverage (incl. loops): 1.97% \| (excl. loops): 0.00%

/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]; }

/home/eoseret/qaas_runs_GNR/173-814-2343/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/revert.cpp: 34 - 39

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

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

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

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

37:       density1(i, j) = density0(i, j);

38:       energy1(i, j) = energy0(i, j);

39:     }

0x444250 PUSH	%RBP

0x444251 MOV	%RSP,%RBP

0x444254 PUSH	%R15

0x444256 PUSH	%R14

0x444258 PUSH	%R13

0x44425a PUSH	%R12

0x44425c PUSH	%RBX

0x44425d SUB	$0xb8,%RSP

0x444264 MOV	%RCX,%R12

0x444267 MOV	%RDX,%R13

0x44426a MOV	0x30(%RBP),%RAX

0x44426e MOV	0x20(%RBP),%RBX

0x444272 MOV	0x18(%RBP),%RCX

0x444276 MOV	0x10(%RBP),%RDX

0x44427a MOVL	$0,-0x38(%RBP)

0x444281 TEST	%RAX,%RAX

0x444284 JS	444657

0x44428a MOV	%R9,%R14

0x44428d MOV	%R8,%R15

0x444290 MOV	%RCX,-0x50(%RBP)

0x444294 MOV	(%RDI),%ESI

0x444296 MOVQ	$0,-0x60(%RBP)

0x44429e MOV	%RAX,-0x58(%RBP)

0x4442a2 MOVQ	$0x1,-0x88(%RBP)

0x4442ad SUB	$0x8,%RSP

0x4442b1 LEA	-0x88(%RBP),%RAX

0x4442b8 LEA	-0x38(%RBP),%RCX

0x4442bc LEA	-0x60(%RBP),%R8

0x4442c0 MOV	%RDX,-0x40(%RBP)

0x4442c4 LEA	-0x58(%RBP),%R9

0x4442c8 MOV	$0x4b8890,%EDI

0x4442cd MOV	%ESI,-0x34(%RBP)

0x4442d0 MOV	$0x22,%EDX

0x4442d5 PUSH	$0x1

0x4442d7 PUSH	$0x1

0x4442d9 PUSH	%RAX

0x4442da CALL	4031f0 <__kmpc_for_static_init_8@plt>

0x4442df MOV	-0x40(%RBP),%R9

0x4442e3 ADD	$0x20,%RSP

0x4442e7 MOV	-0x60(%RBP),%RCX

0x4442eb MOV	-0x58(%RBP),%RSI

0x4442ef CMP	%RSI,%RCX

0x4442f2 JA	444639

0x4442f8 MOV	-0x50(%RBP),%RDI

0x4442fc SUB	%EDI,%EBX

0x4442fe MOV	(%R13),%R10

0x444302 MOV	0x10(%R13),%RAX

0x444306 MOV	(%R12),%RDX

0x44430a MOV	0x10(%R12),%R12

0x44430f MOV	%R14,%R8

0x444312 MOV	(%R15),%R14

0x444315 MOV	0x10(%R15),%R11

0x444319 MOV	%R11,-0x30(%RBP)

0x44431d MOV	(%R8),%R15

0x444320 MOV	0x10(%R8),%R13

0x444324 INC	%RSI

0x444327 MOV	%RSI,%R11

0x44432a SUB	%RCX,%R11

0x44432d MOV	$-0x2,%R8D

0x444333 AND	%R11,%R8

0x444336 MOV	%RBX,-0x48(%RBP)

0x44433a JE	444669

0x444340 MOV	%R11,-0x68(%RBP)

0x444344 MOVD	%EBX,%XMM0

0x444348 PSHUFD	$0x44,%XMM0,%XMM2

0x44434d MOVQ	%R9,%XMM0

0x444352 PSHUFD	$0x44,%XMM0,%XMM0

0x444357 MOVDQA	%XMM0,-0xd0(%RBP)

0x44435f MOVQ	%RDI,%XMM0

0x444364 PSHUFD	$0x44,%XMM0,%XMM0

0x444369 MOVDQA	%XMM0,-0xc0(%RBP)

0x444371 MOV	%R10,-0x80(%RBP)

0x444375 MOVQ	%R10,%XMM0

0x44437a PSHUFD	$0x44,%XMM0,%XMM8

0x444380 MOV	%RDX,-0x70(%RBP)

0x444384 MOVQ	%RDX,%XMM0

0x444389 PSHUFD	$0x44,%XMM0,%XMM4

0x44438e MOV	%R14,%R11

0x444391 MOVQ	%R14,%XMM0

0x444396 PSHUFD	$0x44,%XMM0,%XMM5

0x44439b MOV	%R15,%RBX

0x44439e MOVQ	%R15,%XMM0

0x4443a3 MOVQ	%RCX,%XMM1

0x4443a8 PSHUFD	$0x44,%XMM1,%XMM6

0x4443ad PADDQ	0x5707b(%RIP),%XMM6

0x4443b5 PSHUFD	$0x44,%XMM0,%XMM7

0x4443ba XOR	%R10D,%R10D

0x4443bd MOVQ	%XMM2,%RDI

0x4443c2 MOVDQA	%XMM2,-0xe0(%RBP)

0x4443ca PSHUFD	$-0x12,%XMM2,%XMM0

0x4443cf MOVQ	%XMM0,%R15

0x4443d4 MOVDQA	%XMM8,%XMM11

0x4443d9 PSRLQ	$0x20,%XMM11

0x4443df MOV	%RAX,-0x78(%RBP)

0x4443e3 MOVQ	%RAX,%XMM1

0x4443e8 PSHUFD	$0x44,%XMM1,%XMM0

0x4443ed MOVDQA	%XMM0,-0xb0(%RBP)

0x4443f5 MOVDQA	%XMM4,%XMM10

0x4443fa PSRLQ	$0x20,%XMM10

0x444400 MOV	%R12,%R14

0x444403 MOVQ	%R12,%XMM1

0x444408 PSHUFD	$0x44,%XMM1,%XMM0

0x44440d MOVDQA	%XMM0,-0xa0(%RBP)

0x444415 MOVDQA	%XMM5,%XMM12

0x44441a PSRLQ	$0x20,%XMM12

0x444420 MOVQ	-0x30(%RBP),%XMM1

0x444425 PSHUFD	$0x44,%XMM1,%XMM13

0x44442b MOVDQA	%XMM7,%XMM14

0x444430 PSRLQ	$0x20,%XMM14

0x444436 NOPW	%CS:(%RAX,%RAX,1)

(304) 0x444440 MOVQ	%XMM6,%RAX

(304) 0x444445 XOR	%EDX,%EDX

(304) 0x444447 DIV	%RDI

(304) 0x44444a MOVQ	%RAX,%XMM1

(304) 0x44444f PSHUFD	$-0x12,%XMM6,%XMM2

(304) 0x444454 MOVQ	%XMM2,%RAX

(304) 0x444459 XOR	%EDX,%EDX

(304) 0x44445b DIV	%R15

(304) 0x44445e MOVQ	%RAX,%XMM2

(304) 0x444463 PUNPCKLQDQ	%XMM2,%XMM1

(304) 0x444467 MOVDQA	%XMM1,%XMM15

(304) 0x44446c PADDQ	-0xd0(%RBP),%XMM15

(304) 0x444475 PMULUDQ	-0xe0(%RBP),%XMM1

(304) 0x44447d MOVDQA	%XMM6,%XMM2

(304) 0x444481 PSUBQ	%XMM1,%XMM2

(304) 0x444485 PADDQ	-0xc0(%RBP),%XMM2

(304) 0x44448d PSLLQ	$0x20,%XMM2

(304) 0x444492 PSHUFD	$-0x13,%XMM2,%XMM9

(304) 0x444498 PSHUFD	$-0x18,%XMM15,%XMM2

(304) 0x44449e PSLLQ	$0x20,%XMM15

(304) 0x4444a4 PSRAD	$0x1f,%XMM15

(304) 0x4444aa MOVDQA	%XMM9,%XMM1

(304) 0x4444af PSHUFD	$-0x13,%XMM15,%XMM3

(304) 0x4444b5 PUNPCKLDQ	%XMM3,%XMM2

(304) 0x4444b9 MOVDQA	%XMM8,%XMM3

(304) 0x4444be PSRAD	$0x1f,%XMM1

(304) 0x4444c3 PMULUDQ	%XMM2,%XMM3

(304) 0x4444c7 MOVDQA	%XMM11,%XMM0

(304) 0x4444cc PMULUDQ	%XMM2,%XMM0

(304) 0x4444d0 PUNPCKLDQ	%XMM1,%XMM9

(304) 0x4444d5 PSRLQ	$0x20,%XMM15

(304) 0x4444db MOVDQA	%XMM8,%XMM1

(304) 0x4444e0 PMULUDQ	%XMM15,%XMM1

(304) 0x4444e5 PADDQ	%XMM0,%XMM1

(304) 0x4444e9 PSLLQ	$0x20,%XMM1

(304) 0x4444ee PADDQ	%XMM9,%XMM3

(304) 0x4444f3 PADDQ	%XMM1,%XMM3

(304) 0x4444f7 PSLLQ	$0x3,%XMM3

(304) 0x4444fc MOVDQA	%XMM10,%XMM0

(304) 0x444501 PMULUDQ	%XMM2,%XMM0

(304) 0x444505 MOVDQA	%XMM4,%XMM1

(304) 0x444509 PADDQ	-0xb0(%RBP),%XMM3

(304) 0x444511 PMULUDQ	%XMM15,%XMM1

(304) 0x444516 PADDQ	%XMM0,%XMM1

(304) 0x44451a MOVDQA	%XMM4,%XMM0

(304) 0x44451e MOVQ	%XMM3,%RDX

(304) 0x444523 PMULUDQ	%XMM2,%XMM0

(304) 0x444527 PSLLQ	$0x20,%XMM1

(304) 0x44452c PADDQ	%XMM9,%XMM0

(304) 0x444531 PSHUFD	$-0x12,%XMM3,%XMM3

(304) 0x444536 PADDQ	%XMM1,%XMM0

(304) 0x44453a PSLLQ	$0x3,%XMM0

(304) 0x44453f PADDQ	-0xa0(%RBP),%XMM0

(304) 0x444547 MOVQ	%XMM3,%R9

(304) 0x44454c MOVQ	%XMM0,%R12

(304) 0x444551 PSHUFD	$-0x12,%XMM0,%XMM0

(304) 0x444556 MOVDQA	%XMM5,%XMM1

(304) 0x44455a MOVQ	%XMM0,%RAX

(304) 0x44455f MOVDQA	%XMM12,%XMM0

(304) 0x444564 PMULUDQ	%XMM2,%XMM0

(304) 0x444568 MOVDQA	%XMM5,%XMM3

(304) 0x44456c PMULUDQ	%XMM2,%XMM1

(304) 0x444570 PMULUDQ	%XMM15,%XMM3

(304) 0x444575 PADDQ	%XMM0,%XMM3

(304) 0x444579 PSLLQ	$0x20,%XMM3

(304) 0x44457e MOVSD	(%RDX),%XMM0

(304) 0x444582 PADDQ	%XMM9,%XMM1

(304) 0x444587 PADDQ	%XMM3,%XMM1

(304) 0x44458b MOVSD	(%R9),%XMM3

(304) 0x444590 MOVSD	%XMM0,(%R12)

(304) 0x444596 PSLLQ	$0x3,%XMM1

(304) 0x44459b PADDQ	%XMM13,%XMM1

(304) 0x4445a0 MOVQ	%XMM1,%R9

(304) 0x4445a5 PSHUFD	$-0x12,%XMM1,%XMM0

(304) 0x4445aa MOVQ	%XMM0,%RDX

(304) 0x4445af MOVDQA	%XMM7,%XMM0

(304) 0x4445b3 PMULUDQ	%XMM2,%XMM0

(304) 0x4445b7 PMULUDQ	%XMM14,%XMM2

(304) 0x4445bc PMULUDQ	%XMM7,%XMM15

(304) 0x4445c1 PADDQ	%XMM2,%XMM15

(304) 0x4445c6 PADDQ	%XMM9,%XMM0

(304) 0x4445cb MOVSD	%XMM3,(%RAX)

(304) 0x4445cf PSLLQ	$0x20,%XMM15

(304) 0x4445d5 PADDQ	%XMM15,%XMM0

(304) 0x4445da MOVQ	%R13,%XMM1

(304) 0x4445df PSLLQ	$0x3,%XMM0

(304) 0x4445e4 PSHUFD	$0x44,%XMM1,%XMM1

(304) 0x4445e9 PADDQ	%XMM0,%XMM1

(304) 0x4445ed MOVQ	(%R9),%XMM0

(304) 0x4445f2 MOVQ	%XMM1,%RAX

(304) 0x4445f7 PSHUFD	$-0x12,%XMM1,%XMM1

(304) 0x4445fc MOVQ	%XMM1,%R9

(304) 0x444601 MOVQ	(%RDX),%XMM1

(304) 0x444605 MOVQ	%XMM0,(%RAX)

(304) 0x444609 MOVQ	%XMM1,(%R9)

(304) 0x44460e PADDQ	0x56e3a(%RIP),%XMM6

(304) 0x444616 ADD	$0x2,%R10

(304) 0x44461a CMP	%R8,%R10

(304) 0x44461d JB	444440

0x444623 CMP	%R8,-0x68(%RBP)

0x444627 MOV	-0x40(%RBP),%R9

0x44462b MOV	-0x80(%RBP),%R10

0x44462f MOV	-0x78(%RBP),%R15

0x444633 MOV	-0x70(%RBP),%R12

0x444637 JNE	44467a

0x444639 MOV	$0x4b88b0,%EDI

0x44463e MOV	-0x34(%RBP),%ESI

0x444641 ADD	$0xb8,%RSP

0x444648 POP	%RBX

0x444649 POP	%R12

0x44464b POP	%R13

0x44464d POP	%R14

0x44464f POP	%R15

0x444651 POP	%RBP

0x444652 JMP	403050

0x444657 ADD	$0xb8,%RSP

0x44465e POP	%RBX

0x44465f POP	%R12

0x444661 POP	%R13

0x444663 POP	%R14

0x444665 POP	%R15

0x444667 POP	%RBP

0x444668 RET

0x444669 MOV	%R15,%RBX

0x44466c MOV	%R14,%R11

0x44466f MOV	%R12,%R14

0x444672 MOV	%RAX,%R15

0x444675 MOV	%RDX,%R12

0x444678 JMP	4446e2

0x44467a ADD	%R8,%RCX

0x44467d JMP	4446e2

0x44467f NOP

(303) 0x444680 MOV	%RCX,%RAX

(303) 0x444683 XOR	%EDX,%EDX

(303) 0x444685 DIVQ	-0x48(%RBP)

(303) 0x444689 MOV	-0x30(%RBP),%R8

(303) 0x44468d ADD	%R9D,%EAX

(303) 0x444690 ADD	-0x50(%RBP),%EDX

(303) 0x444693 MOVSXD	%EDX,%RDX

(303) 0x444696 CLTQ

(303) 0x444698 MOV	%R10,%RDI

(303) 0x44469b IMUL	%RAX,%RDI

(303) 0x44469f ADD	%RDX,%RDI

(303) 0x4446a2 MOVSD	(%R15,%RDI,8),%XMM0

(303) 0x4446a8 MOV	%R12,%RDI

(303) 0x4446ab IMUL	%RAX,%RDI

(303) 0x4446af ADD	%RDX,%RDI

(303) 0x4446b2 MOVSD	%XMM0,(%R14,%RDI,8)

(303) 0x4446b8 MOV	%R11,%RDI

(303) 0x4446bb IMUL	%RAX,%RDI

(303) 0x4446bf ADD	%RDX,%RDI

(303) 0x4446c2 MOVQ	(%R8,%RDI,8),%XMM0

(303) 0x4446c8 IMUL	%RBX,%RAX

(303) 0x4446cc ADD	%RDX,%RAX

(303) 0x4446cf MOVQ	%XMM0,(%R13,%RAX,8)

(303) 0x4446d6 INC	%RCX

(303) 0x4446d9 CMP	%RCX,%RSI

(303) 0x4446dc JE	444639

(303) 0x4446e2 MOV	%RCX,%RAX

(303) 0x4446e5 SHR	$0x20,%RAX

(303) 0x4446e9 JNE	444680

(303) 0x4446eb MOV	%ECX,%EAX

(303) 0x4446ed XOR	%EDX,%EDX

(303) 0x4446ef DIVL	-0x48(%RBP)

(303) 0x4446f2 JMP	444689

0x4446f4 NOPW	%CS:(%RAX,%RAX,1)

0x4446fe XCHG	%AX,%AX

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	revert.cpp:34-39
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/context.h:69
/home/eoseret/qaas_runs_GNR/173-814-2343/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/revert.cpp:34-39

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 24.33 to 11.00 cycles (2.21x 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 24.33 to 2.53 cycles (9.63x speedup).

Details

30% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

6% of SSE/AVX loads are used in vector version.
27% of SSE/AVX stores are used in vector version.
11% of SSE/AVX addition or subtraction instructions are used in vector version.
44% 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/revert.cpp:34

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 200 bytes. The binary function is storing 176 bytes.

General properties

nb instructions	145
nb uops	146
loop length	601
used x86 registers	16
used mmx registers	0
used xmm registers	13
used ymm registers	0
used zmm registers	0
nb stack references	22

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	24.33 cycles
front end	24.33 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	8.70	11.00	12.67	12.67	14.00	11.00	8.70	14.00	14.00	14.00	8.60	12.67
cycles	8.70	11.00	12.67	12.67	14.00	11.00	8.70	14.00	14.00	14.00	8.60	12.67

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	24.33
Dispatch	14.00
Overall L1	24.33

Vectorization ratios

all	30%
load	6%
store	27%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	11%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	44%

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	17%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 24.33 cycles. At this rate:

4% of peak load performance is reached (8.22 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (7.23 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 24.33 to 14.00 cycles (1.74x speedup).

ASM code

In the binary file, the address of the function is: 444250

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 $0xb8,%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 %RCX,%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 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),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x18(%RBP),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x10(%RBP),%RDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
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 444657 <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x407>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %R9,%R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %R8,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RCX,-0x50(%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,-0x60(%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,-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%)
MOVQ $0x1,-0x88(%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 -0x88(%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 -0x60(%RBP),%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RDX,-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%)
LEA -0x58(%RBP),%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x4b8890,%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
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),%R9	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 -0x60(%RBP),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV -0x58(%RBP),%RSI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
CMP %RSI,%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 444639 <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x3e9>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV -0x50(%RBP),%RDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
SUB %EDI,%EBX	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),%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(%R13),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV (%R12),%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(%R12),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %R14,%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV (%R15),%R14	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),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV %R11,-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 (%R8),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x10(%R8),%R13	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
INC %RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RSI,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
SUB %RCX,%R11	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
MOV $-0x2,%R8D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
AND %R11,%R8	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 %RBX,-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%)
JE 444669 <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x419>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %R11,-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 %EBX,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	scal (6.3%)
PSHUFD $0x44,%XMM0,%XMM2	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %R9,%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,-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 %RDI,%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,-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%)
MOV %R10,-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%)
MOVQ %R10,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM8	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOV %RDX,-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%)
MOVQ %RDX,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM4	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOV %R14,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOVQ %R14,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM5	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOV %R15,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOVQ %R15,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
MOVQ %RCX,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM6	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
PADDQ 0x5707b(%RIP),%XMM6	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%)
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%)
XOR %R10D,%R10D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
MOVQ %XMM2,%RDI	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
MOVDQA %XMM2,-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%)
PSHUFD $-0x12,%XMM2,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %XMM0,%R15	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
MOVDQA %XMM8,%XMM11	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM11	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	0.50	vect (25.0%)
MOV %RAX,-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%)
MOVQ %RAX,%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 %XMM4,%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%)
MOV %R12,%R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOVQ %R12,%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,-0xa0(%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 %XMM5,%XMM12	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM12	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	0.50	vect (25.0%)
MOVQ -0x30(%RBP),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM13	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM7,%XMM14	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM14	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 %R8,-0x68(%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 -0x40(%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 -0x80(%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 -0x78(%RBP),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV -0x70(%RBP),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
JNE 44467a <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x42a>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV $0x4b88b0,%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 $0xb8,%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 $0xb8,%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 %R15,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %R14,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %R12,%R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RAX,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RDX,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
JMP 4446e2 <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x492>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
ADD %R8,%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 4446e2 <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x492>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
NOP	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
NOPW %CS:(%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
XCHG %AX,%AX	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	revert.cpp:34-39
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/context.h:69
/home/eoseret/qaas_runs_GNR/173-814-2343/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/revert.cpp:34-39

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

30% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

6% of SSE/AVX loads are used in vector version.
27% of SSE/AVX stores are used in vector version.
11% of SSE/AVX addition or subtraction instructions are used in vector version.
44% 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/revert.cpp:34

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 200 bytes. The binary function is storing 176 bytes.

General properties

nb instructions	145
nb uops	146
loop length	601
used x86 registers	16
used mmx registers	0
used xmm registers	13
used ymm registers	0
used zmm registers	0
nb stack references	22

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	24.33 cycles
front end	24.33 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	8.70	11.00	12.67	12.67	14.00	11.00	8.70	14.00	14.00	14.00	8.60	12.67
cycles	8.70	11.00	12.67	12.67	14.00	11.00	8.70	14.00	14.00	14.00	8.60	12.67

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	24.33
Dispatch	14.00
Overall L1	24.33

Vectorization ratios

all	30%
load	6%
store	27%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	11%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	44%

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	17%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 24.33 cycles. At this rate:

4% of peak load performance is reached (8.22 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (7.23 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: 444250

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 $0xb8,%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 %RCX,%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 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),%RBX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV 0x18(%RBP),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x10(%RBP),%RDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
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 444657 <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x407>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %R9,%R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %R8,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RCX,-0x50(%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,-0x60(%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,-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%)
MOVQ $0x1,-0x88(%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 -0x88(%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 -0x60(%RBP),%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RDX,-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%)
LEA -0x58(%RBP),%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x4b8890,%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
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),%R9	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 -0x60(%RBP),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV -0x58(%RBP),%RSI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
CMP %RSI,%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 444639 <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x3e9>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV -0x50(%RBP),%RDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
SUB %EDI,%EBX	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),%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(%R13),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV (%R12),%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(%R12),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %R14,%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV (%R15),%R14	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),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV %R11,-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 (%R8),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV 0x10(%R8),%R13	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
INC %RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RSI,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
SUB %RCX,%R11	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
MOV $-0x2,%R8D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
AND %R11,%R8	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 %RBX,-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%)
JE 444669 <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x419>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV %R11,-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 %EBX,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	3	1	scal (6.3%)
PSHUFD $0x44,%XMM0,%XMM2	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %R9,%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,-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 %RDI,%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,-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%)
MOV %R10,-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%)
MOVQ %R10,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM8	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOV %RDX,-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%)
MOVQ %RDX,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM4	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOV %R14,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOVQ %R14,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM0,%XMM5	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOV %R15,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOVQ %R15,%XMM0	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
MOVQ %RCX,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM6	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
PADDQ 0x5707b(%RIP),%XMM6	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%)
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%)
XOR %R10D,%R10D	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (6.3%)
MOVQ %XMM2,%RDI	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
MOVDQA %XMM2,-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%)
PSHUFD $-0x12,%XMM2,%XMM0	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %XMM0,%R15	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
MOVDQA %XMM8,%XMM11	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM11	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	0.50	vect (25.0%)
MOV %RAX,-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%)
MOVQ %RAX,%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 %XMM4,%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%)
MOV %R12,%R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOVQ %R12,%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,-0xa0(%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 %XMM5,%XMM12	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM12	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	0.50	vect (25.0%)
MOVQ -0x30(%RBP),%XMM1	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM13	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM7,%XMM14	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM14	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 %R8,-0x68(%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 -0x40(%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 -0x80(%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 -0x78(%RBP),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV -0x70(%RBP),%R12	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
JNE 44467a <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x42a>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV $0x4b88b0,%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 $0xb8,%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 $0xb8,%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 %R15,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %R14,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %R12,%R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RAX,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %RDX,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
JMP 4446e2 <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x492>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
ADD %R8,%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 4446e2 <_Z13revert_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x492>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
NOP	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
NOPW %CS:(%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
XCHG %AX,%AX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A

Name	Coverage (%)	Time (s)
▼revert_kernel(int, int, int, int, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&) [clone .extracted]–	1.97	0.78
○Loop 304 - revert.cpp:35-39 - exec	1.97	0.78
○Loop 303 - revert.cpp:35-39 - 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