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:157-160 [...]	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: 157 - 160

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

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

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

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

160:           node_flux(i, j) = 0.25 * (mass_flux_y(i - 1, j + 0) + mass_flux_y(i, j) + mass_flux_y(i - 1, j + 1) + mass_flux_y(i + 0, j + 1));

0x236ac0 PUSH	%RBP

0x236ac1 MOV	%RSP,%RBP

0x236ac4 PUSH	%R15

0x236ac6 PUSH	%R14

0x236ac8 PUSH	%R13

0x236aca PUSH	%R12

0x236acc PUSH	%RBX

0x236acd SUB	$0x58,%RSP

0x236ad1 MOV	(%RCX),%R15D

0x236ad4 MOV	(%RDX),%R14D

0x236ad7 ADD	$0x4,%R15D

0x236adb CMP	%R15D,%R14D

0x236ade JG	236f08

0x236ae4 MOV	(%R9),%EBX

0x236ae7 MOV	(%R8),%R12D

0x236aea LEA	0x1(%R12),%R13D

0x236aef ADD	$0x3,%EBX

0x236af2 CMP	%R13D,%EBX

0x236af5 JLE	236f08

0x236afb LEA	-0x1(%R14),%EAX

0x236aff SUB	%R13D,%EBX

0x236b02 MOV	(%RDI),%ESI

0x236b04 MOVQ	$0,-0x48(%RBP)

0x236b0c MOVQ	$0x1,-0x78(%RBP)

0x236b14 MOVL	$0,-0x38(%RBP)

0x236b1b SUB	%EAX,%R15D

0x236b1e MOV	%RAX,-0x60(%RBP)

0x236b22 IMUL	%RBX,%R15

0x236b26 DEC	%R15

0x236b29 MOV	%R15,-0x30(%RBP)

0x236b2d SUB	$0x8,%RSP

0x236b31 LEA	-0x38(%RBP),%RCX

0x236b35 LEA	-0x78(%RBP),%RAX

0x236b39 LEA	0x24240(%RIP),%RDI

0x236b40 LEA	-0x48(%RBP),%R8

0x236b44 LEA	-0x30(%RBP),%R9

0x236b48 MOV	%ESI,-0x34(%RBP)

0x236b4b MOV	$0x22,%EDX

0x236b50 PUSH	$0x1

0x236b52 PUSH	$0x1

0x236b54 PUSH	%RAX

0x236b55 CALL	259680 <@plt_start@+0x530>

0x236b5a ADD	$0x20,%RSP

0x236b5e MOV	-0x30(%RBP),%RAX

0x236b62 MOV	-0x48(%RBP),%RCX

0x236b66 CMP	%R15,%RAX

0x236b69 CMOVL	%RAX,%R15

0x236b6d MOV	%R15,-0x30(%RBP)

0x236b71 CMP	%R15,%RCX

0x236b74 JG	236ef6

0x236b7a MOV	0x18(%RBP),%RAX

0x236b7e MOV	0x10(%RBP),%RDX

0x236b82 MOV	-0x60(%RBP),%RSI

0x236b86 MOV	%R15,%R11

0x236b89 SUB	%RCX,%R11

0x236b8c INC	%R11

0x236b8f MOV	(%RAX),%R8

0x236b92 MOV	0x10(%RAX),%RDI

0x236b96 MOV	(%RDX),%RAX

0x236b99 MOV	0x10(%RDX),%R9

0x236b9d MOV	%R8,-0x40(%RBP)

0x236ba1 MOV	%RAX,-0x70(%RBP)

0x236ba5 CMP	$0x4,%R11

0x236ba9 JB	236e47

0x236baf VBROADCASTSD	-0x25a28(%RIP),%YMM9

0x236bb8 VPBROADCASTQ	%RCX,%YMM0

0x236bbe VPADDQ	-0x258e6(%RIP),%YMM0,%YMM0

0x236bc6 VPBROADCASTQ	-0x259ff(%RIP),%YMM10

0x236bcf MOV	%R11,-0x58(%RBP)

0x236bd3 AND	$-0x4,%R11

0x236bd7 VPBROADCASTQ	%RBX,%YMM1

0x236bdd VPBROADCASTD	%ESI,%XMM2

0x236be3 VPBROADCASTD	%R13D,%XMM3

0x236be9 VPBROADCASTD	%R12D,%XMM4

0x236bef VPBROADCASTQ	%R8,%YMM5

0x236bf5 VPBROADCASTD	%R14D,%XMM6

0x236bfb VPBROADCASTQ	%RAX,%YMM7

0x236c01 ADD	%R11,%RCX

0x236c04 VEXTRACTI128	$0x1,%YMM1,%XMM8

0x236c0a MOV	%R11,-0x50(%RBP)

0x236c0e XCHG	%AX,%AX

(164) 0x236c10 VEXTRACTI128	$0x1,%YMM0,%XMM11

(164) 0x236c16 VPEXTRQ	$0x1,%XMM8,%R10

(164) 0x236c1c VPEXTRQ	$0x1,%XMM11,%RAX

(164) 0x236c22 CQTO

(164) 0x236c24 IDIV	%R10

(164) 0x236c27 VMOVQ	%XMM8,%R10

(164) 0x236c2c VMOVQ	%RAX,%XMM12

(164) 0x236c31 VMOVQ	%XMM11,%RAX

(164) 0x236c36 CQTO

(164) 0x236c38 IDIV	%R10

(164) 0x236c3b VPEXTRQ	$0x1,%XMM1,%R10

(164) 0x236c41 VMOVQ	%RAX,%XMM11

(164) 0x236c46 VPEXTRQ	$0x1,%XMM0,%RAX

(164) 0x236c4c CQTO

(164) 0x236c4e VPUNPCKLQDQ	%XMM12,%XMM11,%XMM11

(164) 0x236c53 IDIV	%R10

(164) 0x236c56 VMOVQ	%XMM1,%R10

(164) 0x236c5b VMOVQ	%RAX,%XMM12

(164) 0x236c60 VMOVQ	%XMM0,%RAX

(164) 0x236c65 CQTO

(164) 0x236c67 IDIV	%R10

(164) 0x236c6a ADD	$-0x4,%R11

(164) 0x236c6e VMOVQ	%RAX,%XMM13

(164) 0x236c73 VPUNPCKLQDQ	%XMM12,%XMM13,%XMM12

(164) 0x236c78 VINSERTI128	$0x1,%XMM11,%YMM12,%YMM11

(164) 0x236c7e VPMOVQD	%YMM11,%XMM12

(164) 0x236c84 VPMULLQ	%YMM1,%YMM11,%YMM11

(164) 0x236c8a VPSUBQ	%YMM11,%YMM0,%YMM11

(164) 0x236c8f VPADDQ	%YMM0,%YMM10,%YMM0

(164) 0x236c93 VPMOVQD	%YMM11,%XMM11

(164) 0x236c99 VPADDD	%XMM2,%XMM12,%XMM13

(164) 0x236c9d VPADDD	%XMM6,%XMM12,%XMM12

(164) 0x236ca1 VPMOVSXDQ	%XMM13,%YMM13

(164) 0x236ca6 VPMOVSXDQ	%XMM12,%YMM12

(164) 0x236cab VPMULLQ	%YMM13,%YMM5,%YMM15

(164) 0x236cb1 VPMULLQ	%YMM12,%YMM5,%YMM12

(164) 0x236cb7 VPADDD	%XMM3,%XMM11,%XMM14

(164) 0x236cbb VPADDD	%XMM4,%XMM11,%XMM11

(164) 0x236cbf VPMOVSXDQ	%XMM11,%YMM11

(164) 0x236cc4 VPMOVSXDQ	%XMM14,%YMM14

(164) 0x236cc9 VPADDQ	%YMM11,%YMM15,%YMM16

(164) 0x236ccf VPADDQ	%YMM14,%YMM15,%YMM15

(164) 0x236cd4 VPADDQ	%YMM11,%YMM12,%YMM11

(164) 0x236cd9 VPADDQ	%YMM14,%YMM12,%YMM12

(164) 0x236cde VMOVQ	%XMM16,%RAX

(164) 0x236ce4 VPEXTRQ	$0x1,%XMM16,%RDX

(164) 0x236ceb VEXTRACTI32X4	$0x1,%YMM16,%XMM16

(164) 0x236cf2 VMOVQ	%XMM16,%R10

(164) 0x236cf8 VMOVSD	(%RDI,%RAX,8),%XMM17

(164) 0x236cff VPEXTRQ	$0x1,%XMM16,%R8

(164) 0x236d06 VMOVQ	%XMM15,%RAX

(164) 0x236d0b VMOVSD	(%RDI,%R10,8),%XMM16

(164) 0x236d12 VMOVHPD	(%RDI,%RDX,8),%XMM17,%XMM17

(164) 0x236d19 VPEXTRQ	$0x1,%XMM15,%RDX

(164) 0x236d1f VEXTRACTI128	$0x1,%YMM15,%XMM15

(164) 0x236d25 VMOVHPD	(%RDI,%R8,8),%XMM16,%XMM16

(164) 0x236d2c VMOVQ	%XMM15,%R8

(164) 0x236d31 VPEXTRQ	$0x1,%XMM15,%R10

(164) 0x236d37 VMOVSD	(%RDI,%R8,8),%XMM15

(164) 0x236d3d VMOVHPD	(%RDI,%R10,8),%XMM15,%XMM15

(164) 0x236d43 VINSERTF32X4	$0x1,%XMM16,%YMM17,%YMM16

(164) 0x236d4a VMOVSD	(%RDI,%RAX,8),%XMM17

(164) 0x236d51 VPEXTRQ	$0x1,%XMM11,%RAX

(164) 0x236d57 VMOVHPD	(%RDI,%RDX,8),%XMM17,%XMM17

(164) 0x236d5e VMOVQ	%XMM11,%RDX

(164) 0x236d63 VEXTRACTI128	$0x1,%YMM11,%XMM11

(164) 0x236d69 VMOVQ	%XMM11,%R8

(164) 0x236d6e VPEXTRQ	$0x1,%XMM11,%R10

(164) 0x236d74 VMOVSD	(%RDI,%R8,8),%XMM11

(164) 0x236d7a VMOVHPD	(%RDI,%R10,8),%XMM11,%XMM11

(164) 0x236d80 VINSERTF32X4	$0x1,%XMM15,%YMM17,%YMM15

(164) 0x236d87 VADDPD	%YMM16,%YMM15,%YMM15

(164) 0x236d8d VMOVSD	(%RDI,%RDX,8),%XMM16

(164) 0x236d94 VPEXTRQ	$0x1,%XMM12,%RDX

(164) 0x236d9a VMOVHPD	(%RDI,%RAX,8),%XMM16,%XMM16

(164) 0x236da1 VMOVQ	%XMM12,%RAX

(164) 0x236da6 VEXTRACTI128	$0x1,%YMM12,%XMM12

(164) 0x236dac VMOVQ	%XMM12,%R8

(164) 0x236db1 VPEXTRQ	$0x1,%XMM12,%R10

(164) 0x236db7 VMOVSD	(%RDI,%R8,8),%XMM12

(164) 0x236dbd VMOVHPD	(%RDI,%R10,8),%XMM12,%XMM12

(164) 0x236dc3 VINSERTF32X4	$0x1,%XMM11,%YMM16,%YMM11

(164) 0x236dca VMOVSD	(%RDI,%RAX,8),%XMM16

(164) 0x236dd1 VMOVHPD	(%RDI,%RDX,8),%XMM16,%XMM16

(164) 0x236dd8 VINSERTF32X4	$0x1,%XMM12,%YMM16,%YMM12

(164) 0x236ddf VADDPD	%YMM12,%YMM11,%YMM11

(164) 0x236de4 VPMULLQ	%YMM13,%YMM7,%YMM12

(164) 0x236dea VADDPD	%YMM11,%YMM15,%YMM11

(164) 0x236def VPADDQ	%YMM14,%YMM12,%YMM12

(164) 0x236df4 VPEXTRQ	$0x1,%XMM12,%RAX

(164) 0x236dfa VMOVQ	%XMM12,%RDX

(164) 0x236dff VEXTRACTI128	$0x1,%YMM12,%XMM12

(164) 0x236e05 VMOVQ	%XMM12,%R8

(164) 0x236e0a VPEXTRQ	$0x1,%XMM12,%R10

(164) 0x236e10 VMULPD	%YMM9,%YMM11,%YMM11

(164) 0x236e15 VMOVLPD	%XMM11,(%R9,%RDX,8)

(164) 0x236e1b VMOVHPD	%XMM11,(%R9,%RAX,8)

(164) 0x236e21 VEXTRACTF128	$0x1,%YMM11,%XMM11

(164) 0x236e27 VMOVLPD	%XMM11,(%R9,%R8,8)

(164) 0x236e2d VMOVHPD	%XMM11,(%R9,%R10,8)

(164) 0x236e33 JNE	236c10

0x236e39 MOV	-0x50(%RBP),%RAX

0x236e3d CMP	%RAX,-0x58(%RBP)

0x236e41 JE	236ef6

0x236e47 VMOVSD	-0x25cbf(%RIP),%XMM0

0x236e4f MOV	%R9,-0x68(%RBP)

0x236e53 NOPW	%CS:(%RAX,%RAX,1)

(163) 0x236e60 MOV	%RCX,%RAX

(163) 0x236e63 CQTO

(163) 0x236e65 MOV	%RBX,%R9

(163) 0x236e68 IDIV	%RBX

(163) 0x236e6b MOV	-0x40(%RBP),%R11

(163) 0x236e6f MOV	%RSI,%RBX

(163) 0x236e72 LEA	(%R13,%RDX,1),%R10D

(163) 0x236e77 ADD	%R12D,%EDX

(163) 0x236e7a MOVSXD	%EDX,%RDX

(163) 0x236e7d MOVSXD	%R10D,%R10

(163) 0x236e80 LEA	(%RSI,%RAX,1),%R8D

(163) 0x236e84 MOV	%R14,%RSI

(163) 0x236e87 MOV	%R13,%R14

(163) 0x236e8a MOV	%R12,%R13

(163) 0x236e8d MOVSXD	%R8D,%R8

(163) 0x236e90 IMUL	%R8,%R11

(163) 0x236e94 LEA	(%R11,%RDX,1),%R12

(163) 0x236e98 ADD	%R10,%R11

(163) 0x236e9b VMOVSD	(%RDI,%R11,8),%XMM1

(163) 0x236ea1 VADDSD	(%RDI,%R12,8),%XMM1,%XMM1

(163) 0x236ea7 MOV	%R13,%R12

(163) 0x236eaa MOV	%R14,%R13

(163) 0x236ead MOV	%RSI,%R14

(163) 0x236eb0 ADD	%R14D,%EAX

(163) 0x236eb3 MOV	%RBX,%RSI

(163) 0x236eb6 MOV	%R9,%RBX

(163) 0x236eb9 MOV	-0x68(%RBP),%R9

(163) 0x236ebd CLTQ

(163) 0x236ebf IMUL	-0x40(%RBP),%RAX

(163) 0x236ec4 ADD	%RAX,%RDX

(163) 0x236ec7 ADD	%R10,%RAX

(163) 0x236eca VADDSD	(%RDI,%RDX,8),%XMM1,%XMM1

(163) 0x236ecf VADDSD	(%RDI,%RAX,8),%XMM1,%XMM1

(163) 0x236ed4 MOV	-0x70(%RBP),%RAX

(163) 0x236ed8 IMUL	%RAX,%R8

(163) 0x236edc ADD	%R10,%R8

(163) 0x236edf VMULSD	%XMM0,%XMM1,%XMM1

(163) 0x236ee3 VMOVSD	%XMM1,(%R9,%R8,8)

(163) 0x236ee9 CMP	%R15,%RCX

(163) 0x236eec LEA	0x1(%RCX),%RCX

(163) 0x236ef0 JL	236e60

0x236ef6 MOV	-0x34(%RBP),%ESI

0x236ef9 LEA	0x23e98(%RIP),%RDI

0x236f00 VZEROUPPER

0x236f03 CALL	259690 <@plt_start@+0x540>

0x236f08 ADD	$0x58,%RSP

0x236f0c POP	%RBX

0x236f0d POP	%R12

0x236f0f POP	%R13

0x236f11 POP	%R14

0x236f13 POP	%R15

0x236f15 POP	%RBP

0x236f16 RET

0x236f17 NOPW	(%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:157-160
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:157-160

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 13.50 to 5.00 cycles (2.70x 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

6% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

10% 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:157
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:157

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:158
VPBROADCASTD: 4 occurrences►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
VPBROADCASTQ: 4 occurrences►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158

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 168 bytes. The binary function is storing 88 bytes.

General properties

nb instructions	100
nb uops	108
loop length	417
used x86 registers	15
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	13.50 cycles
front end	13.50 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	8.75	8.75	8.75	8.75	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	8.75	8.75	8.75	8.75	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	13.50
Dispatch	8.75
Overall L1	13.50

Vectorization ratios

INT

all	6%
load	12%
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	6%
load	10%
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	14%
store	10%
mul	12%
add-sub	14%
fma	NA (no fma vectorizable/vectorized instructions)
other	11%

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	11%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 13.50 cycles. At this rate:

4% of peak load performance is reached (12.44 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (6.52 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 13.50 to 8.75 cycles (1.54x speedup).

ASM code

In the binary file, the address of the function is: 236ac0

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),%R15D	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
ADD $0x4,%R15D	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 %R15D,%R14D	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 236f08 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.16+0x448>	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),%R12D	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(%R12),%R13D	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,%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 %R13D,%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%)
JLE 236f08 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.16+0x448>	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(%R14),%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
SUB %R13D,%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 (%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,-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 (6.3%)
SUB %EAX,%R15D	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,%R15	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 %R15	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 %R15,-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%)
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 -0x38(%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 0x24240(%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 -0x30(%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,-0x34(%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 -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 -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 %R15,%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,%R15	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 %R15,-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%)
CMP %R15,%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 236ef6 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.16+0x436>	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 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 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 -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 %R15,%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 (%RAX),%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
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 (%RDX),%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 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 %R8,-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 (12.5%)
MOV %RAX,-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%)
JB 236e47 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.16+0x387>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
VBROADCASTSD -0x25a28(%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 -0x258e6(%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 -0x259ff(%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 %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%)
VPBROADCASTQ %RBX,%YMM1	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
VPBROADCASTD %ESI,%XMM2	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTD %R13D,%XMM3	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTD %R12D,%XMM4	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTQ %R8,%YMM5	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
VPBROADCASTD %R14D,%XMM6	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTQ %RAX,%YMM7	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	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
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%)
MOV %R11,-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%)
XCHG %AX,%AX	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
MOV -0x50(%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
CMP %RAX,-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%)
JE 236ef6 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.16+0x436>	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 -0x25cbf(%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%)
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 -0x34(%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 0x23e98(%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 (%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:157-160
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:157-160

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 13.50 to 5.00 cycles (2.70x 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

6% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

10% 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:157
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:157

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:158
VPBROADCASTD: 4 occurrences►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
VPBROADCASTQ: 4 occurrences►
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158
- /home/eoseret/qaas_runs_ZEN5/173-937-4090/intel/CloverLeaf2.0-CXX/build/CloverLeaf2.0-CXX/src/omp/advec_mom.cpp:158

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 168 bytes. The binary function is storing 88 bytes.

General properties

nb instructions	100
nb uops	108
loop length	417
used x86 registers	15
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	13.50 cycles
front end	13.50 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	8.75	8.75	8.75	8.75	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	8.75	8.75	8.75	8.75	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	13.50
Dispatch	8.75
Overall L1	13.50

Vectorization ratios

INT

all	6%
load	12%
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	6%
load	10%
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	14%
store	10%
mul	12%
add-sub	14%
fma	NA (no fma vectorizable/vectorized instructions)
other	11%

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	11%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 13.50 cycles. At this rate:

4% of peak load performance is reached (12.44 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (6.52 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: 236ac0

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),%R15D	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
ADD $0x4,%R15D	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 %R15D,%R14D	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 236f08 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.16+0x448>	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),%R12D	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(%R12),%R13D	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,%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 %R13D,%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%)
JLE 236f08 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.16+0x448>	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(%R14),%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
SUB %R13D,%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 (%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,-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 (6.3%)
SUB %EAX,%R15D	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,%R15	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 %R15	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 %R15,-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%)
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 -0x38(%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 0x24240(%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 -0x30(%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,-0x34(%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 -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 -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 %R15,%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,%R15	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 %R15,-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%)
CMP %R15,%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 236ef6 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.16+0x436>	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 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 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 -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 %R15,%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 (%RAX),%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
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 (%RDX),%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 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 %R8,-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 (12.5%)
MOV %RAX,-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%)
JB 236e47 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.16+0x387>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	1	0.50-1	N/A
VBROADCASTSD -0x25a28(%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 -0x258e6(%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 -0x259ff(%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 %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%)
VPBROADCASTQ %RBX,%YMM1	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
VPBROADCASTD %ESI,%XMM2	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTD %R13D,%XMM3	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTD %R12D,%XMM4	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTQ %R8,%YMM5	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (12.5%)
VPBROADCASTD %R14D,%XMM6	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	scal (6.3%)
VPBROADCASTQ %RAX,%YMM7	2	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	1	1	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
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%)
MOV %R11,-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%)
XCHG %AX,%AX	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
MOV -0x50(%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
CMP %RAX,-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%)
JE 236ef6 <_Z16advec_mom_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_S2_RNS_8Buffer1DIdEES5_iii.omp_outlined.16+0x436>	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 -0x25cbf(%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%)
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 -0x34(%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 0x23e98(%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 (%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 164 - advec_mom.cpp:158-160 - exec	1.32	0.49
○Loop 163 - advec_mom.cpp:158-160 - 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