OV - attention-gcc-native - Loop 24

	Loop Id: 24	Module: attention-gcc-native	Source: random.tcc:458-3374 [...]	Coverage: 0.29%

0x4108b8 MOVZ	X5, #5400

0x4108bc ADD	X8, X3, #1

0x4108c0 ADD	X0, SP, X5,UXTX

0x4108c4 MOVZ	X1, #22144

0x4108c8 LDR	X0, [X0, X3,LSL #3]    [3]

0x4108cc MOVK	X1, #40236

0x4108d0 STR	X8, [SP, #10392]    [1]

0x4108d4 UBFM	X3, X0, #11, #42

0x4108d8 EOR	X0, X0, X3

0x4108dc AND	X3, X1, X0,LSL #7

0x4108e0 EOR	X0, X0, X3

0x4108e4 AND	X3, X2, X0,LSL #15

0x4108e8 EOR	X0, X0, X3

0x4108ec EOR	X0, X0, X0,LSR #18

0x4108f0 UCVTF	S31, X0

0x4108f4 FMUL	S31, S31, S15

0x4108f8 FCMPE	S31, S13

0x4108fc FCSEL	S31, S31, S12, #4

0x410900 STR	S31, [X19, X22,LSL #2]    [2]

0x410904 CMP	X8, #623

0x410908 B.HI	4109f4

0x41090c MOVZ	X3, #5400

0x410910 MOVZ	X1, #22144

0x410914 ADD	X0, SP, X3,UXTX

0x410918 MOVK	X1, #40236

0x41091c LDR	X0, [X0, X8,LSL #3]    [7]

0x410920 ADD	X7, X8, #1

0x410924 STR	X7, [SP, #10392]    [1]

0x410928 UBFM	X3, X0, #11, #42

0x41092c EOR	X0, X0, X3

0x410930 AND	X3, X1, X0,LSL #7

0x410934 EOR	X0, X0, X3

0x410938 AND	X3, X2, X0,LSL #15

0x41093c EOR	X0, X0, X3

0x410940 EOR	X0, X0, X0,LSR #18

0x410944 UCVTF	S31, X0

0x410948 FMUL	S31, S31, S15

0x41094c FCMPE	S31, S13

0x410950 FCSEL	S31, S31, S12, #4

0x410954 STR	S31, [X21, X22,LSL #2]    [5]

0x410958 CMP	X7, #623

0x41095c B.HI	4109dc

0x410960 MOVZ	X0, #5400

0x410964 ADD	X3, X7, #1

0x410968 ADD	X0, SP, X0,UXTX

0x41096c MOVZ	X1, #22144

0x410970 LDR	X0, [X0, X7,LSL #3]    [6]

0x410974 MOVK	X1, #40236

0x410978 STR	X3, [SP, #10392]    [1]

0x41097c UBFM	X7, X0, #11, #42

0x410980 EOR	X0, X0, X7

0x410984 AND	X7, X1, X0,LSL #7

0x410988 EOR	X0, X0, X7

0x41098c AND	X7, X2, X0,LSL #15

0x410990 EOR	X0, X0, X7

0x410994 EOR	X0, X0, X0,LSR #18

0x410998 UCVTF	S31, X0

0x41099c FMUL	S31, S31, S15

0x4109a0 FCMPE	S31, S13

0x4109a4 B.GE	4111e4

0x4109a8 LDR	X0, [SP, #136]    [1]

0x4109ac STR	S31, [X20, X22,LSL #2]    [4]

0x4109b0 ADD	X22, X22, #1

0x4109b4 CMP	X22, X0

0x4109b8 B.CS	410a0c

0x4109bc CMP	X3, #623

0x4109c0 B.LS	4108b8

0x4109c4 MOVZ	X6, #5400

0x4109c8 ADD	X0, SP, X6,UXTX

0x4109cc BL	411ee0

0x4109d0 LDR	X3, [SP, #10392]    [1]

0x4109d4 MOVZ	X2, #61382

0x4109d8 B	4108b8

0x4109dc MOVZ	X1, #5400

0x4109e0 ADD	X0, SP, X1,UXTX

0x4109e4 BL	411ee0

0x4109e8 LDR	X7, [SP, #10392]    [1]

0x4109ec MOVZ	X2, #61382

0x4109f0 B	410960

0x4109f4 MOVZ	X4, #5400

0x4109f8 ADD	X0, SP, X4,UXTX

0x4109fc BL	411ee0

0x410a00 LDR	X8, [SP, #10392]    [1]

0x410a04 MOVZ	X2, #61382

0x410a08 B	41090c

0x4111e4 MOVN	W0, #49280

0x4111e8 FMOV	S31, W0

0x4111ec B	4109a8

/home/eoseret/llm-attention/attention_v2.cpp: 164 - 167

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

164:       for (size_t i = 0; i < elemsW; ++i) {

165:          h_WQ[i] = dist(rng);

166:          h_WK[i] = dist(rng);

167:          h_WV[i] = dist(rng);

/usr/include/c++/14/bits/random.tcc: 458 - 3374

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

458:       if (_M_p >= state_size)

459: 	_M_gen_rand();

460:

461:       // Calculate o(x(i)).

462:       result_type __z = _M_x[_M_p++];

463:       __z ^= (__z >> __u) & __d;

464:       __z ^= (__z << __s) & __b;

465:       __z ^= (__z << __t) & __c;

466:       __z ^= (__z >> __l);

[...]

3367: 	  __sum += _RealType(__urng() - __urng.min()) * __tmp;

3368: 	  __tmp *= __r;

3369: 	}

3370:       __ret = __sum / __tmp;

3371:       if (__builtin_expect(__ret >= _RealType(1), 0))

3372: 	{

3373: #if _GLIBCXX_USE_C99_MATH_FUNCS

3374: 	  __ret = std::nextafter(_RealType(1), _RealType(0));

Coverage (%)	Name	Source Location	Module
►100.00+	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start	stl_vector.h:1920	attention-gcc-native

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 /

Metric	Value
CQA speedup if no scalar integer	3.26
CQA speedup if FP arith vectorized	2.48
CQA speedup if fully vectorized	3.59
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.16
Bottlenecks	micro-operation queue,
Function	main
Source	attention_v2.cpp:164-167,random.tcc:458-466,random.tcc:3367-3374
Source loop unroll info	not unrolled or unrolled with no peel/tail loop
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	11.00
CQA cycles if no scalar integer	3.38
CQA cycles if FP arith vectorized	4.44
CQA cycles if fully vectorized	3.06
Front-end cycles	11.00
P0 cycles	6.00
P1 cycles	6.00
P2 cycles	9.50
P3 cycles	9.50
P4 cycles	9.50
P5 cycles	9.50
P6 cycles	9.50
P7 cycles	9.50
P8 cycles	4.00
P9 cycles	4.00
P10 cycles	2.00
P11 cycles	2.00
P12 cycles	4.33
P13 cycles	4.33
P14 cycles	4.33
P15 cycles	1.50
P16 cycles	1.50
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	88.00
Nb uops	88.00
Nb loads	NA
Nb stores	6.00
Nb stack references	7.00
FLOP/cycle	0.27
Nb FLOP add-sub	0.00
Nb FLOP mul	3.00
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	0.82
Bytes prefetched	0.00
Bytes loaded	0.00
Bytes stored	9.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	NA
Vectorization ratio store	0.00
Vectorization ratio mul	0.00
Vectorization ratio add_sub	NA
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	41.18
Vector-efficiency ratio load	NA
Vector-efficiency ratio store	37.50
Vector-efficiency ratio mul	25.00
Vector-efficiency ratio add_sub	NA
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	44.00

Metric	Value
CQA speedup if no scalar integer	3.26
CQA speedup if FP arith vectorized	2.48
CQA speedup if fully vectorized	3.59
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.16
Bottlenecks	micro-operation queue,
Function	main
Source	attention_v2.cpp:164-167,random.tcc:458-466,random.tcc:3367-3374
Source loop unroll info	not unrolled or unrolled with no peel/tail loop
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	11.00
CQA cycles if no scalar integer	3.38
CQA cycles if FP arith vectorized	4.44
CQA cycles if fully vectorized	3.06
Front-end cycles	11.00
P0 cycles	6.00
P1 cycles	6.00
P2 cycles	9.50
P3 cycles	9.50
P4 cycles	9.50
P5 cycles	9.50
P6 cycles	9.50
P7 cycles	9.50
P8 cycles	4.00
P9 cycles	4.00
P10 cycles	2.00
P11 cycles	2.00
P12 cycles	4.33
P13 cycles	4.33
P14 cycles	4.33
P15 cycles	1.50
P16 cycles	1.50
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	NA
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	88.00
Nb uops	88.00
Nb loads	NA
Nb stores	6.00
Nb stack references	7.00
FLOP/cycle	0.27
Nb FLOP add-sub	0.00
Nb FLOP mul	3.00
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	0.82
Bytes prefetched	0.00
Bytes loaded	0.00
Bytes stored	9.00
Stride 0	NA
Stride 1	NA
Stride n	NA
Stride unknown	NA
Stride indirect	NA
Vectorization ratio all	0.00
Vectorization ratio load	NA
Vectorization ratio store	0.00
Vectorization ratio mul	0.00
Vectorization ratio add_sub	NA
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	41.18
Vector-efficiency ratio load	NA
Vector-efficiency ratio store	37.50
Vector-efficiency ratio mul	25.00
Vector-efficiency ratio add_sub	NA
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	44.00

Path /

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

Function	main
Source file and lines	random.tcc:458-3374
Module	attention-gcc-native

The loop is defined in:

/home/eoseret/llm-attention/attention_v2.cpp:164-167
/usr/include/c++/14/bits/random.tcc:458-466,3367-3374

The related source loop is not unrolled or unrolled with no peel/tail loop.
Warnings:

Ignoring paths for analysis
Too many paths. Rerun with max-paths=16
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 16 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

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 11.00 to 3.38 cycles (3.26x speedup).

Workaround

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

Vectorization

Your loop is not vectorized. Only 41% of vector register length is used (average across all VPU instructions). By vectorizing your loop, you can lower the cost of an iteration from 11.00 to 3.06 cycles (3.59x speedup).

Details

All VPU 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
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)

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.

std::mersenne_twister_engine
::_M_gen_rand(): 3 occurrences►
- /usr/include/c++/14/bits/random.tcc:459
- /usr/include/c++/14/bits/random.tcc:459
- /usr/include/c++/14/bits/random.tcc:459

Conversion instructions

Detected expensive conversion instructions.

Details

UCVTF: 3 occurrences►
- /usr/include/c++/14/bits/random.tcc:3367
- /usr/include/c++/14/bits/random.tcc:3367
- /usr/include/c++/14/bits/random.tcc:3367

Workaround

Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.

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

The binary loop is composed of 3 FP arithmetical operations:

3: multiply

The binary loop is storing 9 bytes.

Arithmetic intensity

Arithmetic intensity is 0.33 FP operations per loaded or stored byte.

General properties

nb instructions	88
nb uops	88
loop length	352
used w registers	1
used x registers	13
used b registers	0
used h registers	0
used s registers	5
used d registers	0
used q registers	0
used v registers	0
used z registers	0
nb stack references	7

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	11.00 cycles
front end	11.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16
uops	6.00	6.00	9.50	9.50	9.50	9.50	9.50	9.50	4.00	4.00	2.00	2.00	4.33	4.33	4.33	1.50	1.50
cycles	6.00	6.00	9.50	9.50	9.50	9.50	9.50	9.50	4.00	4.00	2.00	2.00	4.33	4.33	4.33	1.50	1.50

Execution ports to units layout:

P0: BRU
P1: BRU
P2: ALU
P3: ALU
P4: ALU
P5: ALU
P6: ALU
P7: ALU
P8 (128 bits): VPU, FP store data, ALU, DIV/SQRT
P9 (128 bits): VPU, ALU, FP store data
P10 (128 bits): VPU, ALU, DIV/SQRT
P11 (128 bits): ALU, VPU
P12 (256 bits): store address, load
P13 (256 bits): store address, load
P14 (256 bits): load
P15 (64 bits): store data
P16 (64 bits): store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	11.00
Dispatch	9.50
Overall L1	11.00

Vectorization ratios

INT

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

all	0%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	0%
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	0%
load	NA (no load vectorizable/vectorized instructions)
store	0%
mul	0%
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%

Vector efficiency ratios

INT

all	47%
load	NA (no load vectorizable/vectorized instructions)
store	37%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	50%

all	25%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	25%
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	25%

INT+FP

all	41%
load	NA (no load vectorizable/vectorized instructions)
store	37%
mul	25%
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	44%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 11.00 cycles. At this rate:

1% of peak store performance is reached (0.82 out of 48.00 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 11.00 to 9.50 cycles (1.16x speedup).

ASM code

In the binary file, the address of the loop is: 4109bc

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16	Latency	Recip. throughput	Vectorization
MOVZ X5, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
ADD X8, X3, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD X0, SP, X5,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
MOVZ X1, #22144	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
LDR X0, [X0, X3,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
MOVK X1, #40236	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
STR X8, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
UBFM X3, X0, #11, #42	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X3, X1, X0,LSL #7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X3, X2, X0,LSL #15	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X0,LSR #18	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
UCVTF S31, X0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (50.0%)
FMUL S31, S31, S15	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	scal (25.0%)
FCMPE S31, S13	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
FCSEL S31, S31, S12, #4	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.25	scal (25.0%)
STR S31, [X19, X22,LSL #2]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	scal (25.0%)
CMP X8, #623	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (50.0%)
B.HI 4109f4 <main+0x834>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVZ X3, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOVZ X1, #22144	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
ADD X0, SP, X3,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
MOVK X1, #40236	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
LDR X0, [X0, X8,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X7, X8, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
STR X7, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
UBFM X3, X0, #11, #42	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X3, X1, X0,LSL #7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X3, X2, X0,LSL #15	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X0,LSR #18	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
UCVTF S31, X0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (50.0%)
FMUL S31, S31, S15	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	scal (25.0%)
FCMPE S31, S13	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
FCSEL S31, S31, S12, #4	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.25	scal (25.0%)
STR S31, [X21, X22,LSL #2]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	scal (25.0%)
CMP X7, #623	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (50.0%)
B.HI 4109dc <main+0x81c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVZ X0, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD X3, X7, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD X0, SP, X0,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
MOVZ X1, #22144	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
LDR X0, [X0, X7,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
MOVK X1, #40236	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
STR X3, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
UBFM X7, X0, #11, #42	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X7, X1, X0,LSL #7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X7, X2, X0,LSL #15	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X0,LSR #18	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
UCVTF S31, X0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (50.0%)
FMUL S31, S31, S15	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	scal (25.0%)
FCMPE S31, S13	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
B.GE 4111e4 <main+0x1024>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X0, [SP, #136]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
STR S31, [X20, X22,LSL #2]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	scal (25.0%)
ADD X22, X22, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
CMP X22, X0	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (50.0%)
B.CS 410a0c <main+0x84c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X3, #623	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (50.0%)
B.LS 4108b8 <main+0x6f8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVZ X6, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
ADD X0, SP, X6,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
BL 411ee0 <_ZNSt23mersenne_twister_engineImLm32ELm624ELm397ELm31ELm2567483615ELm11ELm4294967295ELm7ELm2636928640ELm15ELm4022730752ELm18ELm1812433253EE11_M_gen_randEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X3, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
MOVZ X2, #61382	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
B 4108b8 <main+0x6f8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVZ X1, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
ADD X0, SP, X1,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
BL 411ee0 <_ZNSt23mersenne_twister_engineImLm32ELm624ELm397ELm31ELm2567483615ELm11ELm4294967295ELm7ELm2636928640ELm15ELm4022730752ELm18ELm1812433253EE11_M_gen_randEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X7, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
MOVZ X2, #61382	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
B 410960 <main+0x7a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVZ X4, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD X0, SP, X4,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
BL 411ee0 <_ZNSt23mersenne_twister_engineImLm32ELm624ELm397ELm31ELm2567483615ELm11ELm4294967295ELm7ELm2636928640ELm15ELm4022730752ELm18ELm1812433253EE11_M_gen_randEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X8, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
MOVZ X2, #61382	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
B 41090c <main+0x74c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVN W0, #49280	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
FMOV S31, W0	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
B 4109a8 <main+0x7e8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Function	main
Source file and lines	random.tcc:458-3374
Module	attention-gcc-native

The loop is defined in:

/home/eoseret/llm-attention/attention_v2.cpp:164-167
/usr/include/c++/14/bits/random.tcc:458-466,3367-3374

The related source loop is not unrolled or unrolled with no peel/tail loop.
Warnings:

Ignoring paths for analysis
Too many paths. Rerun with max-paths=16
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

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 11.00 to 3.38 cycles (3.26x speedup).

Workaround

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

Vectorization

Details

All VPU 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
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)

No data for this section

CALL instructions

Detected function call instructions.

Details

std::mersenne_twister_engine
::_M_gen_rand(): 3 occurrences►
- /usr/include/c++/14/bits/random.tcc:459
- /usr/include/c++/14/bits/random.tcc:459
- /usr/include/c++/14/bits/random.tcc:459

Conversion instructions

Detected expensive conversion instructions.

Details

UCVTF: 3 occurrences►
- /usr/include/c++/14/bits/random.tcc:3367
- /usr/include/c++/14/bits/random.tcc:3367
- /usr/include/c++/14/bits/random.tcc:3367

Workaround

Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.

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

The binary loop is composed of 3 FP arithmetical operations:

3: multiply

The binary loop is storing 9 bytes.

Arithmetic intensity

Arithmetic intensity is 0.33 FP operations per loaded or stored byte.

General properties

nb instructions	88
nb uops	88
loop length	352
used w registers	1
used x registers	13
used b registers	0
used h registers	0
used s registers	5
used d registers	0
used q registers	0
used v registers	0
used z registers	0
nb stack references	7

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	11.00 cycles
front end	11.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16
uops	6.00	6.00	9.50	9.50	9.50	9.50	9.50	9.50	4.00	4.00	2.00	2.00	4.33	4.33	4.33	1.50	1.50
cycles	6.00	6.00	9.50	9.50	9.50	9.50	9.50	9.50	4.00	4.00	2.00	2.00	4.33	4.33	4.33	1.50	1.50

Execution ports to units layout:

P0: BRU
P1: BRU
P2: ALU
P3: ALU
P4: ALU
P5: ALU
P6: ALU
P7: ALU
P8 (128 bits): VPU, FP store data, ALU, DIV/SQRT
P9 (128 bits): VPU, ALU, FP store data
P10 (128 bits): VPU, ALU, DIV/SQRT
P11 (128 bits): ALU, VPU
P12 (256 bits): store address, load
P13 (256 bits): store address, load
P14 (256 bits): load
P15 (64 bits): store data
P16 (64 bits): store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	11.00
Dispatch	9.50
Overall L1	11.00

Vectorization ratios

INT

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

all	0%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	0%
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	0%
load	NA (no load vectorizable/vectorized instructions)
store	0%
mul	0%
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%

Vector efficiency ratios

INT

all	47%
load	NA (no load vectorizable/vectorized instructions)
store	37%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	50%

all	25%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	25%
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	25%

INT+FP

all	41%
load	NA (no load vectorizable/vectorized instructions)
store	37%
mul	25%
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	44%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 11.00 cycles. At this rate:

1% of peak store performance is reached (0.82 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

ASM code

In the binary file, the address of the loop is: 4109bc

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	P16	Latency	Recip. throughput	Vectorization
MOVZ X5, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
ADD X8, X3, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD X0, SP, X5,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
MOVZ X1, #22144	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
LDR X0, [X0, X3,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
MOVK X1, #40236	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
STR X8, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
UBFM X3, X0, #11, #42	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X3, X1, X0,LSL #7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X3, X2, X0,LSL #15	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X0,LSR #18	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
UCVTF S31, X0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (50.0%)
FMUL S31, S31, S15	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	scal (25.0%)
FCMPE S31, S13	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
FCSEL S31, S31, S12, #4	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.25	scal (25.0%)
STR S31, [X19, X22,LSL #2]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	scal (25.0%)
CMP X8, #623	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (50.0%)
B.HI 4109f4 <main+0x834>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVZ X3, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOVZ X1, #22144	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
ADD X0, SP, X3,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
MOVK X1, #40236	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
LDR X0, [X0, X8,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X7, X8, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
STR X7, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
UBFM X3, X0, #11, #42	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X3, X1, X0,LSL #7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X3, X2, X0,LSL #15	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X3	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X0,LSR #18	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
UCVTF S31, X0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (50.0%)
FMUL S31, S31, S15	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	scal (25.0%)
FCMPE S31, S13	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
FCSEL S31, S31, S12, #4	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	2	0.25	scal (25.0%)
STR S31, [X21, X22,LSL #2]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	scal (25.0%)
CMP X7, #623	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (50.0%)
B.HI 4109dc <main+0x81c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVZ X0, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD X3, X7, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD X0, SP, X0,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
MOVZ X1, #22144	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
LDR X0, [X0, X7,LSL #3]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
MOVK X1, #40236	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
STR X3, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
UBFM X7, X0, #11, #42	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X7, X1, X0,LSL #7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
AND X7, X2, X0,LSL #15	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X7	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
EOR X0, X0, X0,LSR #18	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
UCVTF S31, X0	1	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (50.0%)
FMUL S31, S31, S15	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	3	0.25	scal (25.0%)
FCMPE S31, S13	1	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	2	1	scal (25.0%)
B.GE 4111e4 <main+0x1024>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X0, [SP, #136]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
STR S31, [X20, X22,LSL #2]	1	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0.50	0.50	0	0	0	2	0.50	scal (25.0%)
ADD X22, X22, #1	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
CMP X22, X0	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (50.0%)
B.CS 410a0c <main+0x84c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
CMP X3, #623	1	0	0	0.25	0.25	0	0	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (50.0%)
B.LS 4108b8 <main+0x6f8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVZ X6, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
ADD X0, SP, X6,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
BL 411ee0 <_ZNSt23mersenne_twister_engineImLm32ELm624ELm397ELm31ELm2567483615ELm11ELm4294967295ELm7ELm2636928640ELm15ELm4022730752ELm18ELm1812433253EE11_M_gen_randEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X3, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
MOVZ X2, #61382	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
B 4108b8 <main+0x6f8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVZ X1, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
ADD X0, SP, X1,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
BL 411ee0 <_ZNSt23mersenne_twister_engineImLm32ELm624ELm397ELm31ELm2567483615ELm11ELm4294967295ELm7ELm2636928640ELm15ELm4022730752ELm18ELm1812433253EE11_M_gen_randEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X7, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
MOVZ X2, #61382	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
B 410960 <main+0x7a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVZ X4, #5400	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD X0, SP, X4,UXTX	1	0	0	0	0	0	0	0.50	0.50	0	0	0	0	0	0	0	0	0	2	0.50	N/A
BL 411ee0 <_ZNSt23mersenne_twister_engineImLm32ELm624ELm397ELm31ELm2567483615ELm11ELm4294967295ELm7ELm2636928640ELm15ELm4022730752ELm18ELm1812433253EE11_M_gen_randEv>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR X8, [SP, #10392]	1	0	0	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
MOVZ X2, #61382	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	scal (50.0%)
B 41090c <main+0x74c>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVN W0, #49280	1	0	0	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	1	0.17	N/A
FMOV S31, W0	1	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	2	0.25	scal (25.0%)
B 4109a8 <main+0x7e8>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

CALL instructions

Details

Conversion instructions

Details

Workaround

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

Arithmetic intensity

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

CALL instructions

Details

Conversion instructions

Details

Workaround

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

Arithmetic intensity

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code