OV - exec - Loop 1019

	Loop Id: 1019	Module: exec	Source: par_multi_interp.c:272-276	Coverage: 0.01%

0x4539c0 INC	%RCX

0x4539c3 ADD	$0x8,%RSI

0x4539c7 ADD	$-0x8,%RDI

0x4539cb JE	453817

0x4539d1 MOV	(%R12,%RSI,8),%R8    [1]

0x4539d5 CMP	$-0x3,%R8

0x4539d9 JE	453ac0

0x4539df CMP	$0x1,%R8

0x4539e3 JNE	4539e8

0x4539e5 INC	%RCX

0x4539e8 MOV	0x8(%R12,%RSI,8),%R8    [1]

0x4539ed CMP	$0x1,%R8

0x4539f1 JE	453ad2

0x4539f7 CMP	$-0x3,%R8

0x4539fb JNE	453a00

0x4539fd INC	%RDX

0x453a00 MOV	0x10(%R12,%RSI,8),%R8    [1]

0x453a05 CMP	$0x1,%R8

0x453a09 JE	453ae4

0x453a0f CMP	$-0x3,%R8

0x453a13 JNE	453a18

0x453a15 INC	%RDX

0x453a18 MOV	0x18(%R12,%RSI,8),%R8    [1]

0x453a1d CMP	$0x1,%R8

0x453a21 JE	453af6

0x453a27 CMP	$-0x3,%R8

0x453a2b JNE	453a30

0x453a2d INC	%RDX

0x453a30 MOV	0x20(%R12,%RSI,8),%R8    [1]

0x453a35 CMP	$0x1,%R8

0x453a39 JE	453b08

0x453a3f CMP	$-0x3,%R8

0x453a43 JNE	453a48

0x453a45 INC	%RDX

0x453a48 MOV	0x28(%R12,%RSI,8),%R8    [1]

0x453a4d CMP	$0x1,%R8

0x453a51 JE	453b1a

0x453a57 CMP	$-0x3,%R8

0x453a5b JNE	453a60

0x453a5d INC	%RDX

0x453a60 MOV	0x30(%R12,%RSI,8),%R8    [1]

0x453a65 CMP	$0x1,%R8

0x453a69 JE	453b2c

0x453a6f CMP	$-0x3,%R8

0x453a73 JNE	453a78

0x453a75 INC	%RDX

0x453a78 MOV	0x38(%R12,%RSI,8),%R8    [1]

0x453a7d CMP	$0x1,%R8

0x453a81 JNE	453b3e

0x453a87 JMP	4539c0

0x453ac0 INC	%RDX

0x453ac3 MOV	0x8(%R12,%RSI,8),%R8    [1]

0x453ac8 CMP	$0x1,%R8

0x453acc JNE	4539f7

0x453ad2 INC	%RCX

0x453ad5 MOV	0x10(%R12,%RSI,8),%R8    [1]

0x453ada CMP	$0x1,%R8

0x453ade JNE	453a0f

0x453ae4 INC	%RCX

0x453ae7 MOV	0x18(%R12,%RSI,8),%R8    [1]

0x453aec CMP	$0x1,%R8

0x453af0 JNE	453a27

0x453af6 INC	%RCX

0x453af9 MOV	0x20(%R12,%RSI,8),%R8    [1]

0x453afe CMP	$0x1,%R8

0x453b02 JNE	453a3f

0x453b08 INC	%RCX

0x453b0b MOV	0x28(%R12,%RSI,8),%R8    [1]

0x453b10 CMP	$0x1,%R8

0x453b14 JNE	453a57

0x453b1a INC	%RCX

0x453b1d MOV	0x30(%R12,%RSI,8),%R8    [1]

0x453b22 CMP	$0x1,%R8

0x453b26 JNE	453a6f

0x453b2c INC	%RCX

0x453b2f MOV	0x38(%R12,%RSI,8),%R8    [1]

0x453b34 CMP	$0x1,%R8

0x453b38 JE	4539c0

0x453b3e CMP	$-0x3,%R8

0x453b42 JNE	4539c3

0x453b48 INC	%RDX

0x453b4b JMP	4539c3

/home/eoseret/qaas_runs_GNR/173-927-0874/intel/AMG/build/AMG/AMG/parcsr_ls/par_multi_interp.c: 272 - 276

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

272: #pragma omp parallel for private(i) reduction(+:n_coarse,n_SF ) HYPRE_SMP_SCHEDULE

273: #endif

274:    for (i=0; i < n_fine; i++)

275:       if (CF_marker[i] == 1) n_coarse++;

276:       else if (CF_marker[i] == -3) n_SF++;

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

Path /

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	8.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.13
Bottlenecks	micro-operation queue,
Function	hypre_BoomerAMGBuildMultipass.extracted.85
Source	par_multi_interp.c:272-276
Source loop unroll info	unrolled by 15
Source loop unroll confidence level	max
Unroll/vectorization loop type	main
Unroll factor	15
CQA cycles	13.50
CQA cycles if no scalar integer	13.50
CQA cycles if FP arith vectorized	13.50
CQA cycles if fully vectorized	1.69
Front-end cycles	13.50
P0 cycles	12.00
P1 cycles	7.80
P2 cycles	5.00
P3 cycles	5.00
P4 cycles	0.00
P5 cycles	7.60
P6 cycles	12.00
P7 cycles	0.00
P8 cycles	0.00
P9 cycles	0.00
P10 cycles	7.60
P11 cycles	5.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	82.00
Nb uops	81.00
Nb loads	15.00
Nb stores	0.00
Nb stack references	0.00
FLOP/cycle	0.00
Nb FLOP add-sub	0.00
Nb FLOP mul	0.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	8.89
Bytes prefetched	0.00
Bytes loaded	120.00
Bytes stored	0.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	NA
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	12.50
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	NA
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	12.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	12.50

Metric	Value
CQA speedup if no scalar integer	1.00
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	8.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.13
Bottlenecks	micro-operation queue,
Function	hypre_BoomerAMGBuildMultipass.extracted.85
Source	par_multi_interp.c:272-276
Source loop unroll info	unrolled by 15
Source loop unroll confidence level	max
Unroll/vectorization loop type	main
Unroll factor	15
CQA cycles	13.50
CQA cycles if no scalar integer	13.50
CQA cycles if FP arith vectorized	13.50
CQA cycles if fully vectorized	1.69
Front-end cycles	13.50
P0 cycles	12.00
P1 cycles	7.80
P2 cycles	5.00
P3 cycles	5.00
P4 cycles	0.00
P5 cycles	7.60
P6 cycles	12.00
P7 cycles	0.00
P8 cycles	0.00
P9 cycles	0.00
P10 cycles	7.60
P11 cycles	5.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	82.00
Nb uops	81.00
Nb loads	15.00
Nb stores	0.00
Nb stack references	0.00
FLOP/cycle	0.00
Nb FLOP add-sub	0.00
Nb FLOP mul	0.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	8.89
Bytes prefetched	0.00
Bytes loaded	120.00
Bytes stored	0.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	NA
Vectorization ratio mul	NA
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	12.50
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	NA
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	12.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	12.50

Path /

Average path: Display a virtual path defined by average values of all real paths

Function	hypre_BoomerAMGBuildMultipass.extracted.85
Source file and lines	par_multi_interp.c:272-276
Module	exec

The loop is defined in /home/eoseret/qaas_runs_GNR/173-927-0874/intel/AMG/build/AMG/AMG/parcsr_ls/par_multi_interp.c:272-276.

It is main loop of related source loop which is unrolled by 15 (including vectorization).
Warnings:

Ignoring paths for analysis
Too many paths. If you really need to analyze all of the 6561 paths individually, rerun with max-paths=6561
RecMII not computed since number of paths is unknown or > max_paths
Streams not analyzed since number of paths is unknown or > max_paths

Try to simplify control and/or increase the maximum number of paths per function/loop through the 'max-paths-nb' option.

This loop has 6561 execution paths.

The presence of multiple execution paths is typically the main/first bottleneck.
Try to simplify control inside loop: ideally, try to remove all conditional expressions, for example by (if applicable):

hoisting them (moving them outside the loop)
turning them into conditional moves, MIN or MAX

Ex: if (x<0) x=0 => x = (x<0 ? 0 : x) (or MAX(0,x) after defining the corresponding macro)

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. 8 data elements could be processed at once in vector registers. By vectorizing your loop, you can lower the cost of an iteration from 13.50 to 1.69 cycles (8.00x speedup).

Details

All SSE/AVX instructions are used in scalar version (process only one data element in vector registers). Since your execution units are vector units, only a vectorized loop can use their full power.

Workaround

Try another compiler or update/tune your current one:
- recompile with ffast-math (included in Ofast) to extend loop vectorization to FP reductions.
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop 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

Type of elements and instruction set

No instructions are processing arithmetic or math operations on FP elements. This loop is probably writing/copying data or processing integer elements.

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

The binary loop does not contain any FP arithmetical operations. The binary loop is loading 120 bytes.

General properties

nb instructions	82
nb uops	81
loop length	348
used x86 registers	6
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	0

Front-end

ASSUMED MACRO FUSION 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
uops	12.00	7.80	5.00	5.00	0.00	7.60	12.00	0.00	0.00	0.00	7.60	5.00
cycles	12.00	7.80	5.00	5.00	0.00	7.60	12.00	0.00	0.00	0.00	7.60	5.00

Execution ports to units layout:

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

Cycles executing div or sqrt instructions

Cycles summary

Front-end	13.50
Dispatch	12.00
Overall L1	13.50

Vectorization ratios

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

Vector efficiency ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 13.50 cycles. At this rate:

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

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 12.00 cycles (1.13x speedup).

ASM code

In the binary file, the address of the loop is: 4539d1

Instruction	Nb FU	P0	P1	P2	P3	P5	P6	P10	P11	Latency	Recip. throughput	Vectorization
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
ADD $0x8,%RSI	1	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD $-0x8,%RDI	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
JE 453817 <hypre_BoomerAMGBuildMultipass.extracted.85+0xb7>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
MOV (%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453ac0 <hypre_BoomerAMGBuildMultipass.extracted.85+0x360>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 4539e8 <hypre_BoomerAMGBuildMultipass.extracted.85+0x288>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x8(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453ad2 <hypre_BoomerAMGBuildMultipass.extracted.85+0x372>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a00 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2a0>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x10(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453ae4 <hypre_BoomerAMGBuildMultipass.extracted.85+0x384>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a18 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2b8>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x18(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453af6 <hypre_BoomerAMGBuildMultipass.extracted.85+0x396>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a30 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2d0>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x20(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453b08 <hypre_BoomerAMGBuildMultipass.extracted.85+0x3a8>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a48 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2e8>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x28(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453b1a <hypre_BoomerAMGBuildMultipass.extracted.85+0x3ba>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a60 <hypre_BoomerAMGBuildMultipass.extracted.85+0x300>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x30(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453b2c <hypre_BoomerAMGBuildMultipass.extracted.85+0x3cc>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a78 <hypre_BoomerAMGBuildMultipass.extracted.85+0x318>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x38(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453b3e <hypre_BoomerAMGBuildMultipass.extracted.85+0x3de>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
JMP 4539c0 <hypre_BoomerAMGBuildMultipass.extracted.85+0x260>	1	0	0	0	0	0	0	0	0	0	2.08	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x8(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 4539f7 <hypre_BoomerAMGBuildMultipass.extracted.85+0x297>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x10(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a0f <hypre_BoomerAMGBuildMultipass.extracted.85+0x2af>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x18(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a27 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2c7>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x20(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a3f <hypre_BoomerAMGBuildMultipass.extracted.85+0x2df>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x28(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a57 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2f7>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x30(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a6f <hypre_BoomerAMGBuildMultipass.extracted.85+0x30f>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x38(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 4539c0 <hypre_BoomerAMGBuildMultipass.extracted.85+0x260>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 4539c3 <hypre_BoomerAMGBuildMultipass.extracted.85+0x263>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
JMP 4539c3 <hypre_BoomerAMGBuildMultipass.extracted.85+0x263>	1	0	0	0	0	0	0	0	0	0	2.08	N/A

Function	hypre_BoomerAMGBuildMultipass.extracted.85
Source file and lines	par_multi_interp.c:272-276
Module	exec

Ignoring paths for analysis
Too many paths. If you really need to analyze all of the 6561 paths individually, rerun with max-paths=6561
RecMII not computed since number of paths is unknown or > max_paths
Streams not analyzed since number of paths is unknown or > max_paths

hoisting them (moving them outside the loop)
turning them into conditional moves, MIN or MAX

Ex: if (x<0) x=0 => x = (x<0 ? 0 : x) (or MAX(0,x) after defining the corresponding macro)

gain
potential
hint
expert

Vectorization

Details

All SSE/AVX instructions are used in scalar version (process only one data element in vector registers). Since your execution units are vector units, only a vectorized loop can use their full power.

Workaround

Try another compiler or update/tune your current one:
- recompile with ffast-math (included in Ofast) to extend loop vectorization to FP reductions.
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop 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

Type of elements and instruction set

No instructions are processing arithmetic or math operations on FP elements. This loop is probably writing/copying data or processing integer elements.

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

The binary loop does not contain any FP arithmetical operations. The binary loop is loading 120 bytes.

General properties

nb instructions	82
nb uops	81
loop length	348
used x86 registers	6
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	0

Front-end

ASSUMED MACRO FUSION 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
uops	12.00	7.80	5.00	5.00	0.00	7.60	12.00	0.00	0.00	0.00	7.60	5.00
cycles	12.00	7.80	5.00	5.00	0.00	7.60	12.00	0.00	0.00	0.00	7.60	5.00

Execution ports to units layout:

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

Cycles executing div or sqrt instructions

Cycles summary

Front-end	13.50
Dispatch	12.00
Overall L1	13.50

Vectorization ratios

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

Vector efficiency ratios

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

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 13.50 cycles. At this rate:

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

Front-end bottlenecks

ASM code

In the binary file, the address of the loop is: 4539d1

Instruction	Nb FU	P0	P1	P2	P3	P5	P6	P10	P11	Latency	Recip. throughput	Vectorization
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
ADD $0x8,%RSI	1	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD $-0x8,%RDI	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
JE 453817 <hypre_BoomerAMGBuildMultipass.extracted.85+0xb7>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
MOV (%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453ac0 <hypre_BoomerAMGBuildMultipass.extracted.85+0x360>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 4539e8 <hypre_BoomerAMGBuildMultipass.extracted.85+0x288>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x8(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453ad2 <hypre_BoomerAMGBuildMultipass.extracted.85+0x372>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a00 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2a0>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x10(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453ae4 <hypre_BoomerAMGBuildMultipass.extracted.85+0x384>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a18 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2b8>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x18(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453af6 <hypre_BoomerAMGBuildMultipass.extracted.85+0x396>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a30 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2d0>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x20(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453b08 <hypre_BoomerAMGBuildMultipass.extracted.85+0x3a8>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a48 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2e8>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x28(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453b1a <hypre_BoomerAMGBuildMultipass.extracted.85+0x3ba>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a60 <hypre_BoomerAMGBuildMultipass.extracted.85+0x300>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x30(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 453b2c <hypre_BoomerAMGBuildMultipass.extracted.85+0x3cc>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a78 <hypre_BoomerAMGBuildMultipass.extracted.85+0x318>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x38(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453b3e <hypre_BoomerAMGBuildMultipass.extracted.85+0x3de>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
JMP 4539c0 <hypre_BoomerAMGBuildMultipass.extracted.85+0x260>	1	0	0	0	0	0	0	0	0	0	2.08	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x8(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 4539f7 <hypre_BoomerAMGBuildMultipass.extracted.85+0x297>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x10(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a0f <hypre_BoomerAMGBuildMultipass.extracted.85+0x2af>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x18(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a27 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2c7>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x20(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a3f <hypre_BoomerAMGBuildMultipass.extracted.85+0x2df>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x28(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a57 <hypre_BoomerAMGBuildMultipass.extracted.85+0x2f7>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x30(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 453a6f <hypre_BoomerAMGBuildMultipass.extracted.85+0x30f>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RCX	1	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x38(%R12,%RSI,8),%R8	1	0	0	0.33	0.33	0	0	0	0.33	1	0.33	scal (12.5%)
CMP $0x1,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JE 4539c0 <hypre_BoomerAMGBuildMultipass.extracted.85+0x260>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
CMP $-0x3,%R8	1	0.20	0.20	0	0	0.20	0.20	0.20	0	1	0.20	scal (12.5%)
JNE 4539c3 <hypre_BoomerAMGBuildMultipass.extracted.85+0x263>	1	0.50	0	0	0	0	0.50	0	0	0	0.50	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	1	0.17	N/A
JMP 4539c3 <hypre_BoomerAMGBuildMultipass.extracted.85+0x263>	1	0	0	0	0	0	0	0	0	0	2.08	N/A

Report Configuration

Vectorization

Details

Workaround

Execution units bottlenecks

Type of elements and instruction set

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

Vectorization

Details

Workaround

Execution units bottlenecks

Type of elements and instruction set

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code