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

	Function: ideal_gas_kernel(int, int, int, int, clover::Buffer2D<double>&, clover::Buffer2D<double>&, ...	Module: exec	Source: ideal_gas.cpp:37-46 [...]	Coverage (incl. loops): 4.23% \| (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/ideal_gas.cpp: 37 - 46

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

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

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

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

40:       double v = 1.0 / density(i, j);

41:       pressure(i, j) = (1.4 - 1.0) * density(i, j) * energy(i, j);

42:       double pressurebyenergy = (1.4 - 1.0) * density(i, j);

43:       double pressurebyvolume = -density(i, j) * pressure(i, j);

44:       double sound_speed_squared = v * v * (pressure(i, j) * pressurebyenergy - pressurebyvolume);

45:       soundspeed(i, j) = std::sqrt(sound_speed_squared);

46:     }

0x438700 PUSH	%RBP

0x438701 MOV	%RSP,%RBP

0x438704 PUSH	%R15

0x438706 PUSH	%R14

0x438708 PUSH	%R13

0x43870a PUSH	%R12

0x43870c PUSH	%RBX

0x43870d SUB	$0xa8,%RSP

0x438714 MOV	0x30(%RBP),%RAX

0x438718 MOV	0x20(%RBP),%RBX

0x43871c MOV	0x18(%RBP),%R10

0x438720 MOV	0x10(%RBP),%R11

0x438724 MOVL	$0,-0x30(%RBP)

0x43872b TEST	%RAX,%RAX

0x43872e JS	438b41

0x438734 MOV	%R9,%R14

0x438737 MOV	%R8,%R15

0x43873a MOV	%RCX,%R12

0x43873d MOV	%RDX,%R13

0x438740 MOV	(%RDI),%ESI

0x438742 MOVQ	$0,-0x68(%RBP)

0x43874a MOV	%RAX,-0x60(%RBP)

0x43874e MOVQ	$0x1,-0x90(%RBP)

0x438759 SUB	$0x8,%RSP

0x43875d LEA	-0x90(%RBP),%RAX

0x438764 LEA	-0x30(%RBP),%RCX

0x438768 LEA	-0x68(%RBP),%R8

0x43876c LEA	-0x60(%RBP),%R9

0x438770 MOV	$0x4b81d0,%EDI

0x438775 MOV	%ESI,-0x2c(%RBP)

0x438778 MOV	$0x22,%EDX

0x43877d PUSH	$0x1

0x43877f PUSH	$0x1

0x438781 PUSH	%RAX

0x438782 MOV	%R10,-0x48(%RBP)

0x438786 MOV	%R11,-0x40(%RBP)

0x43878a CALL	4031f0 <__kmpc_for_static_init_8@plt>

0x43878f MOV	-0x40(%RBP),%R11

0x438793 MOV	-0x48(%RBP),%R10

0x438797 ADD	$0x20,%RSP

0x43879b MOV	-0x68(%RBP),%RCX

0x43879f MOV	-0x60(%RBP),%RSI

0x4387a3 CMP	%RSI,%RCX

0x4387a6 JA	438b23

0x4387ac SUB	%R10D,%EBX

0x4387af MOV	(%R13),%RAX

0x4387b3 MOV	0x10(%R13),%R8

0x4387b7 MOV	(%R12),%RDX

0x4387bb MOV	0x10(%R12),%R12

0x4387c0 MOV	(%R15),%RDI

0x4387c3 MOV	0x10(%R15),%R15

0x4387c7 MOV	(%R14),%R9

0x4387ca MOV	0x10(%R14),%R14

0x4387ce INC	%RSI

0x4387d1 MOV	%RSI,-0x38(%RBP)

0x4387d5 SUB	%RCX,%RSI

0x4387d8 MOV	$-0x2,%R13D

0x4387de MOV	%RSI,-0x78(%RBP)

0x4387e2 AND	%RSI,%R13

0x4387e5 MOV	%RDX,-0x58(%RBP)

0x4387e9 MOV	%R9,-0x80(%RBP)

0x4387ed MOV	%RBX,-0x50(%RBP)

0x4387f1 MOV	%RDI,-0x88(%RBP)

0x4387f8 JE	438b53

0x4387fe MOVD	%EBX,%XMM0

0x438802 PSHUFD	$0x44,%XMM0,%XMM2

0x438807 MOVQ	%R11,%XMM1

0x43880c PSHUFD	$0x44,%XMM1,%XMM0

0x438811 MOVDQA	%XMM0,-0xc0(%RBP)

0x438819 MOVQ	%R10,%XMM1

0x43881e PSHUFD	$0x44,%XMM1,%XMM0

0x438823 MOVDQA	%XMM0,-0xb0(%RBP)

0x43882b MOV	%RAX,-0x70(%RBP)

0x43882f MOVQ	%RAX,%XMM1

0x438834 PSHUFD	$0x44,%XMM1,%XMM0

0x438839 MOVQ	%RDX,%XMM1

0x43883e PSHUFD	$0x44,%XMM1,%XMM4

0x438843 MOVQ	%RDI,%XMM1

0x438848 PSHUFD	$0x44,%XMM1,%XMM5

0x43884d MOVQ	%R9,%XMM1

0x438852 PSHUFD	$0x44,%XMM1,%XMM6

0x438857 MOVQ	%RCX,%XMM1

0x43885c PSHUFD	$0x44,%XMM1,%XMM7

0x438861 PADDQ	0x62bc7(%RIP),%XMM7

0x438869 XOR	%R10D,%R10D

0x43886c MOVQ	%XMM2,%RDI

0x438871 MOVDQA	%XMM2,-0xd0(%RBP)

0x438879 PSHUFD	$-0x12,%XMM2,%XMM1

0x43887e MOVQ	%XMM1,%R9

0x438883 MOVDQA	%XMM0,%XMM9

0x438888 PSRLQ	$0x20,%XMM9

0x43888e MOVQ	%R8,%XMM1

0x438893 PSHUFD	$0x44,%XMM1,%XMM1

0x438898 MOVDQA	%XMM1,-0xa0(%RBP)

0x4388a0 MOVDQA	%XMM4,%XMM12

0x4388a5 PSRLQ	$0x20,%XMM12

0x4388ab MOV	%R12,%RBX

0x4388ae MOVQ	%R12,%XMM1

0x4388b3 PSHUFD	$0x44,%XMM1,%XMM13

0x4388b9 MOVDQA	%XMM5,%XMM14

0x4388be PSRLQ	$0x20,%XMM14

0x4388c4 NOPW	%CS:(%RAX,%RAX,1)

(234) 0x4388d0 MOVQ	%XMM7,%RAX

(234) 0x4388d5 XOR	%EDX,%EDX

(234) 0x4388d7 DIV	%RDI

(234) 0x4388da MOVQ	%RAX,%XMM1

(234) 0x4388df PSHUFD	$-0x12,%XMM7,%XMM2

(234) 0x4388e4 MOVQ	%XMM2,%RAX

(234) 0x4388e9 XOR	%EDX,%EDX

(234) 0x4388eb DIV	%R9

(234) 0x4388ee MOVQ	%RAX,%XMM2

(234) 0x4388f3 PUNPCKLQDQ	%XMM2,%XMM1

(234) 0x4388f7 MOVDQA	%XMM1,%XMM10

(234) 0x4388fc PMULUDQ	-0xd0(%RBP),%XMM1

(234) 0x438904 PADDQ	-0xc0(%RBP),%XMM10

(234) 0x43890d MOVDQA	%XMM7,%XMM2

(234) 0x438911 PSUBQ	%XMM1,%XMM2

(234) 0x438915 PADDQ	-0xb0(%RBP),%XMM2

(234) 0x43891d PSLLQ	$0x20,%XMM2

(234) 0x438922 PSHUFD	$-0x18,%XMM10,%XMM11

(234) 0x438928 PSLLQ	$0x20,%XMM10

(234) 0x43892e PSRAD	$0x1f,%XMM10

(234) 0x438934 PSHUFD	$-0x13,%XMM2,%XMM15

(234) 0x43893a PSHUFD	$-0x13,%XMM10,%XMM1

(234) 0x438940 PUNPCKLDQ	%XMM1,%XMM11

(234) 0x438945 MOVDQA	%XMM0,%XMM1

(234) 0x438949 MOVDQA	%XMM15,%XMM2

(234) 0x43894e PMULUDQ	%XMM11,%XMM1

(234) 0x438953 MOVDQA	%XMM9,%XMM8

(234) 0x438958 PMULUDQ	%XMM11,%XMM8

(234) 0x43895d PSRAD	$0x1f,%XMM2

(234) 0x438962 PSRLQ	$0x20,%XMM10

(234) 0x438968 MOVDQA	%XMM0,%XMM3

(234) 0x43896c PMULUDQ	%XMM10,%XMM3

(234) 0x438971 PUNPCKLDQ	%XMM2,%XMM15

(234) 0x438976 PADDQ	%XMM8,%XMM3

(234) 0x43897b PSLLQ	$0x20,%XMM3

(234) 0x438980 PADDQ	%XMM15,%XMM1

(234) 0x438985 PADDQ	%XMM3,%XMM1

(234) 0x438989 PSLLQ	$0x3,%XMM1

(234) 0x43898e PADDQ	-0xa0(%RBP),%XMM1

(234) 0x438996 PSHUFD	$-0x12,%XMM1,%XMM2

(234) 0x43899b MOVQ	%XMM1,%RDX

(234) 0x4389a0 MOVQ	%XMM2,%RAX

(234) 0x4389a5 MOVSD	(%RDX),%XMM2

(234) 0x4389a9 MOVHPD	(%RAX),%XMM2

(234) 0x4389ad MOVDQA	%XMM12,%XMM3

(234) 0x4389b2 PMULUDQ	%XMM11,%XMM3

(234) 0x4389b7 MOVDQA	%XMM4,%XMM8

(234) 0x4389bc MOVAPD	0x62c1c(%RIP),%XMM1

(234) 0x4389c4 PMULUDQ	%XMM10,%XMM8

(234) 0x4389c9 PADDQ	%XMM3,%XMM8

(234) 0x4389ce MOVDQA	%XMM4,%XMM3

(234) 0x4389d2 DIVPD	%XMM2,%XMM1

(234) 0x4389d6 PMULUDQ	%XMM11,%XMM3

(234) 0x4389db PSLLQ	$0x20,%XMM8

(234) 0x4389e1 PADDQ	%XMM15,%XMM3

(234) 0x4389e6 PADDQ	%XMM8,%XMM3

(234) 0x4389eb PSLLQ	$0x3,%XMM3

(234) 0x4389f0 PADDQ	%XMM13,%XMM3

(234) 0x4389f5 PSHUFD	$-0x12,%XMM3,%XMM8

(234) 0x4389fb MOVQ	%XMM3,%R12

(234) 0x438a00 MOVQ	%XMM8,%R11

(234) 0x438a05 MOVDQA	%XMM14,%XMM3

(234) 0x438a0a MOVDQA	%XMM5,%XMM8

(234) 0x438a0f PMULUDQ	%XMM11,%XMM3

(234) 0x438a14 PMULUDQ	%XMM10,%XMM8

(234) 0x438a19 PADDQ	%XMM3,%XMM8

(234) 0x438a1e MOVDQA	%XMM5,%XMM3

(234) 0x438a22 PMULUDQ	%XMM11,%XMM3

(234) 0x438a27 PSLLQ	$0x20,%XMM8

(234) 0x438a2d PADDQ	%XMM15,%XMM3

(234) 0x438a32 PADDQ	%XMM8,%XMM3

(234) 0x438a37 MULPD	0x65db1(%RIP),%XMM2

(234) 0x438a3f MOVSD	(%R12),%XMM8

(234) 0x438a45 MOVHPD	(%R11),%XMM8

(234) 0x438a4a MULPD	%XMM2,%XMM8

(234) 0x438a4f PSLLQ	$0x3,%XMM3

(234) 0x438a54 MOVQ	%R15,%XMM2

(234) 0x438a59 PSHUFD	$0x44,%XMM2,%XMM2

(234) 0x438a5e PADDQ	%XMM3,%XMM2

(234) 0x438a62 MOVQ	%XMM2,%R11

(234) 0x438a67 PSHUFD	$-0x12,%XMM2,%XMM2

(234) 0x438a6c MOVQ	%XMM2,%R12

(234) 0x438a71 MOVLPD	%XMM8,(%R11)

(234) 0x438a76 MOVHPD	%XMM8,(%R12)

(234) 0x438a7c MOVSD	(%RDX),%XMM2

(234) 0x438a80 MOVHPD	(%RAX),%XMM2

(234) 0x438a84 MULPD	%XMM1,%XMM1

(234) 0x438a88 MULPD	%XMM8,%XMM2

(234) 0x438a8d MULPD	0x65d6b(%RIP),%XMM1

(234) 0x438a95 MULPD	%XMM1,%XMM2

(234) 0x438a99 MOVDQA	%XMM6,%XMM1

(234) 0x438a9d MOVDQA	%XMM6,%XMM3

(234) 0x438aa1 PMULUDQ	%XMM11,%XMM1

(234) 0x438aa6 PSRLQ	$0x20,%XMM3

(234) 0x438aab PMULUDQ	%XMM11,%XMM3

(234) 0x438ab0 PMULUDQ	%XMM6,%XMM10

(234) 0x438ab5 PADDQ	%XMM3,%XMM10

(234) 0x438aba PADDQ	%XMM15,%XMM1

(234) 0x438abf PSLLQ	$0x20,%XMM10

(234) 0x438ac5 PADDQ	%XMM10,%XMM1

(234) 0x438aca PSLLQ	$0x3,%XMM1

(234) 0x438acf MOVQ	%R14,%XMM3

(234) 0x438ad4 PSHUFD	$0x44,%XMM3,%XMM3

(234) 0x438ad9 PADDQ	%XMM1,%XMM3

(234) 0x438add MOVQ	%XMM3,%RAX

(234) 0x438ae2 PSHUFD	$-0x12,%XMM3,%XMM1

(234) 0x438ae7 MOVQ	%XMM1,%RDX

(234) 0x438aec SQRTPD	%XMM2,%XMM1

(234) 0x438af0 MOVLPD	%XMM1,(%RAX)

(234) 0x438af4 MOVHPD	%XMM1,(%RDX)

(234) 0x438af8 PADDQ	0x62950(%RIP),%XMM7

(234) 0x438b00 ADD	$0x2,%R10

(234) 0x438b04 CMP	%R13,%R10

(234) 0x438b07 JB	4388d0

0x438b0d CMP	%R13,-0x78(%RBP)

0x438b11 MOV	-0x48(%RBP),%R10

0x438b15 MOV	-0x40(%RBP),%R11

0x438b19 MOV	-0x70(%RBP),%R12

0x438b1d MOV	-0x38(%RBP),%RSI

0x438b21 JNE	438b62

0x438b23 MOV	$0x4b81f0,%EDI

0x438b28 MOV	-0x2c(%RBP),%ESI

0x438b2b ADD	$0xa8,%RSP

0x438b32 POP	%RBX

0x438b33 POP	%R12

0x438b35 POP	%R13

0x438b37 POP	%R14

0x438b39 POP	%R15

0x438b3b POP	%RBP

0x438b3c JMP	403050

0x438b41 ADD	$0xa8,%RSP

0x438b48 POP	%RBX

0x438b49 POP	%R12

0x438b4b POP	%R13

0x438b4d POP	%R14

0x438b4f POP	%R15

0x438b51 POP	%RBP

0x438b52 RET

0x438b53 MOV	%R12,%RBX

0x438b56 MOV	%RAX,%R12

0x438b59 MOV	-0x38(%RBP),%RSI

0x438b5d JMP	438c04

0x438b62 ADD	%R13,%RCX

0x438b65 JMP	438c04

0x438b6a NOPW	(%RAX,%RAX,1)

(233) 0x438b70 MOV	%RCX,%RAX

(233) 0x438b73 XOR	%EDX,%EDX

(233) 0x438b75 DIVQ	-0x50(%RBP)

(233) 0x438b79 MOV	-0x58(%RBP),%R9

(233) 0x438b7d ADD	%R11D,%EAX

(233) 0x438b80 ADD	%R10D,%EDX

(233) 0x438b83 MOVSXD	%EDX,%RDX

(233) 0x438b86 CLTQ

(233) 0x438b88 MOV	%R12,%RDI

(233) 0x438b8b IMUL	%RAX,%RDI

(233) 0x438b8f ADD	%RDX,%RDI

(233) 0x438b92 MOVSD	(%R8,%RDI,8),%XMM0

(233) 0x438b98 MOVSD	0x61ef8(%RIP),%XMM1

(233) 0x438ba0 DIVSD	%XMM0,%XMM1

(233) 0x438ba4 MULSD	0x65c64(%RIP),%XMM0

(233) 0x438bac IMUL	%RAX,%R9

(233) 0x438bb0 ADD	%RDX,%R9

(233) 0x438bb3 MULSD	(%RBX,%R9,8),%XMM0

(233) 0x438bb9 MOV	-0x88(%RBP),%R9

(233) 0x438bc0 IMUL	%RAX,%R9

(233) 0x438bc4 ADD	%RDX,%R9

(233) 0x438bc7 MOVSD	%XMM0,(%R15,%R9,8)

(233) 0x438bcd MULSD	%XMM1,%XMM1

(233) 0x438bd1 MULSD	0x65c3f(%RIP),%XMM1

(233) 0x438bd9 MULSD	%XMM0,%XMM1

(233) 0x438bdd MULSD	(%R8,%RDI,8),%XMM1

(233) 0x438be3 IMUL	-0x80(%RBP),%RAX

(233) 0x438be8 ADD	%RDX,%RAX

(233) 0x438beb XORPS	%XMM0,%XMM0

(233) 0x438bee SQRTSD	%XMM1,%XMM0

(233) 0x438bf2 MOVSD	%XMM0,(%R14,%RAX,8)

(233) 0x438bf8 INC	%RCX

(233) 0x438bfb CMP	%RCX,%RSI

(233) 0x438bfe JE	438b23

(233) 0x438c04 MOV	%RCX,%RAX

(233) 0x438c07 SHR	$0x20,%RAX

(233) 0x438c0b JNE	438b70

(233) 0x438c11 MOV	%ECX,%EAX

(233) 0x438c13 XOR	%EDX,%EDX

(233) 0x438c15 DIVL	-0x50(%RBP)

(233) 0x438c18 JMP	438b79

0x438c1d 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	ideal_gas.cpp:37-46
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/ideal_gas.cpp:37-46

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 22.50 to 10.50 cycles (2.14x 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 22.50 to 2.35 cycles (9.57x speedup).

Details

28% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

5% of SSE/AVX loads are used in vector version.
22% of SSE/AVX stores are used in vector version.
11% of SSE/AVX addition or subtraction instructions are used in vector version.
47% 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/ideal_gas.cpp:37

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 168 bytes.

General properties

nb instructions	134
nb uops	135
loop length	566
used x86 registers	16
used mmx registers	0
used xmm registers	11
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	22.50 cycles
front end	22.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	8.30	10.50	12.67	12.67	14.00	10.50	8.30	14.00	14.00	14.00	8.40	12.67
cycles	8.30	10.50	12.67	12.67	14.00	10.50	8.30	14.00	14.00	14.00	8.40	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	22.50
Dispatch	14.00
Overall L1	22.50

Vectorization ratios

all	28%
load	5%
store	22%
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	47%

Vector efficiency ratios

all	15%
load	12%
store	13%
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 22.50 cycles. At this rate:

4% of peak load performance is reached (8.89 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (7.47 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 22.50 to 14.00 cycles (1.61x speedup).

ASM code

In the binary file, the address of the function is: 438700

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 $0xa8,%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),%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),%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),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOVL $0,-0x30(%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 438b41 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x441>	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	N/A
MOV %R8,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
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 (%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,-0x68(%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,-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 $0x1,-0x90(%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 -0x90(%RBP),%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x30(%RBP),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x68(%RBP),%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x60(%RBP),%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x4b81d0,%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,-0x2c(%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,-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%)
MOV %R11,-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),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x48(%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 -0x68(%RBP),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV -0x60(%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 438b23 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x423>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
SUB %R10D,%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),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
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),%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 (%R15),%RDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
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 (%R14),%R9	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
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 %RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RSI,-0x38(%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,%RSI	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,%R13D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV %RSI,-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%)
AND %RSI,%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 %RDX,-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,-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%)
MOV %RBX,-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,-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%)
JE 438b53 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x453>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
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 %R11,%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%)
MOVQ %R10,%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%)
MOV %RAX,-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 %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%)
MOVQ %RDX,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM4	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	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,%XMM5	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %R9,%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%)
MOVQ %RCX,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM7	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
PADDQ 0x62bc7(%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 %XMM2,%RDI	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
MOVDQA %XMM2,-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%)
PSHUFD $-0x12,%XMM2,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %XMM1,%R9	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
MOVDQA %XMM0,%XMM9	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM9	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	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,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM1,-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 %XMM4,%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%)
MOV %R12,%RBX	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,%XMM13	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM5,%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 %R13,-0x78(%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 -0x48(%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 -0x40(%RBP),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
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
MOV -0x38(%RBP),%RSI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
JNE 438b62 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x462>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV $0x4b81f0,%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 -0x2c(%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 $0xa8,%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 $0xa8,%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 %R12,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RAX,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV -0x38(%RBP),%RSI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
JMP 438c04 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x504>	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 438c04 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x504>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPW (%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	ideal_gas.cpp:37-46
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/ideal_gas.cpp:37-46

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

28% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

5% of SSE/AVX loads are used in vector version.
22% of SSE/AVX stores are used in vector version.
11% of SSE/AVX addition or subtraction instructions are used in vector version.
47% 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/ideal_gas.cpp:37

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 168 bytes.

General properties

nb instructions	134
nb uops	135
loop length	566
used x86 registers	16
used mmx registers	0
used xmm registers	11
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	22.50 cycles
front end	22.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	8.30	10.50	12.67	12.67	14.00	10.50	8.30	14.00	14.00	14.00	8.40	12.67
cycles	8.30	10.50	12.67	12.67	14.00	10.50	8.30	14.00	14.00	14.00	8.40	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	22.50
Dispatch	14.00
Overall L1	22.50

Vectorization ratios

all	28%
load	5%
store	22%
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	47%

Vector efficiency ratios

all	15%
load	12%
store	13%
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 22.50 cycles. At this rate:

4% of peak load performance is reached (8.89 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (7.47 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: 438700

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 $0xa8,%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),%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),%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),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOVL $0,-0x30(%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 438b41 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x441>	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	N/A
MOV %R8,%R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
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 (%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,-0x68(%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,-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 $0x1,-0x90(%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 -0x90(%RBP),%RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x30(%RBP),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x68(%RBP),%R8	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA -0x60(%RBP),%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x4b81d0,%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,-0x2c(%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,-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%)
MOV %R11,-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),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
MOV -0x48(%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 -0x68(%RBP),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV -0x60(%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 438b23 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x423>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
SUB %R10D,%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),%RAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
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),%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 (%R15),%RDI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
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 (%R14),%R9	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
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 %RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RSI,-0x38(%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,%RSI	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,%R13D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV %RSI,-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%)
AND %RSI,%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 %RDX,-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,-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%)
MOV %RBX,-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,-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%)
JE 438b53 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x453>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
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 %R11,%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%)
MOVQ %R10,%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%)
MOV %RAX,-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 %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%)
MOVQ %RDX,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM4	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	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,%XMM5	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %R9,%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%)
MOVQ %RCX,%XMM1	1	0	0	0	0	0	1	0	0	0	0	0	0	1	1	scal (12.5%)
PSHUFD $0x44,%XMM1,%XMM7	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
PADDQ 0x62bc7(%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 %XMM2,%RDI	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
MOVDQA %XMM2,-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%)
PSHUFD $-0x12,%XMM2,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVQ %XMM1,%R9	1	1	0	0	0	0	0	0	0	0	0	0	0	1	1	scal (12.5%)
MOVDQA %XMM0,%XMM9	1	0	0	0	0	0	0	0	0	0	0	0	0	0-1	0.17	vect (25.0%)
PSRLQ $0x20,%XMM9	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	1-2	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,%XMM1	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM1,-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 %XMM4,%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%)
MOV %R12,%RBX	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,%XMM13	1	0	0.50	0	0	0	0.50	0	0	0	0	0	0	1	0.50	vect (25.0%)
MOVDQA %XMM5,%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 %R13,-0x78(%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 -0x48(%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 -0x40(%RBP),%R11	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
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
MOV -0x38(%RBP),%RSI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
JNE 438b62 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x462>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV $0x4b81f0,%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 -0x2c(%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 $0xa8,%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 $0xa8,%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 %R12,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV %RAX,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV -0x38(%RBP),%RSI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
JMP 438c04 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x504>	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 438c04 <_Z16ideal_gas_kerneliiiiRN6clover8Buffer2DIdEES2_S2_S2_.extracted+0x504>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPW (%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)
▼ideal_gas_kernel(int, int, int, int, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&, clover::Buffer2D&) [clone .extracted]–	4.23	1.67
○Loop 234 - ideal_gas.cpp:38-46 - exec	4.23	1.67
○Loop 233 - ideal_gas.cpp:38-46 - 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