Multi-Architecture Cross-Language Benchmarks

Rigorous, reproducible cross-language benchmarks comparing Xu v0.8.8 against C (Clang -O3), Rust (rustc -O3), Swift (swiftc -O3), and Go (go build) across 47 real-world algorithms. Concurrently verified across dual architectures (ARM64 & x86_64) on macOS, Linux, and Windows.

6-Platform Core Workloads Cross-Architecture Panorama

Inspect Xu's measured performance directly across all 6 OS and hardware architecture configurations side-by-side. Shorter is faster:

startup · Process startup / empty mainlower is better (ms)

Apple M-Series

1.80ms

Linux Neoverse

0.30ms fastest

Snapdragon WoA

4.60ms

macOS Intel

3.92ms

Linux x86_64

0.30ms fastest

Windows x86_64

7.10ms

loop_sum · Integer loop accumulation (100M)lower is better (ms)

Apple M-Series

7.50ms fastest

Linux Neoverse

11.60ms

Snapdragon WoA

12.06ms

macOS Intel

11.54ms

Linux x86_64

47.20ms

Windows x86_64

55.70ms

matmul · Matrix multiplication (i64)lower is better (ms)

Apple M-Series

8.70ms fastest

Linux Neoverse

8.80ms

Snapdragon WoA

9.15ms

macOS Intel

12.21ms

Linux x86_64

9.00ms

Windows x86_64

16.20ms

chan_ping · Channel / Coroutines ping-ponglower is better (ms)

Apple M-Series

22.30ms

Linux Neoverse

37.20ms

Snapdragon WoA

38.69ms

macOS Intel

24.75ms

Linux x86_64

20.60ms fastest

Windows x86_64

35.80ms

aes128_encrypt · AES-128 software encryptlower is better (ms)

Apple M-Series

3.60ms fastest

Linux Neoverse

4.00ms

Snapdragon WoA

4.16ms

macOS Intel

6.27ms

Linux x86_64

5.50ms

Windows x86_64

12.30ms

sha256_chain · SHA-256 hash chainlower is better (ms)

Apple M-Series

2.00ms

Linux Neoverse

0.60ms fastest

Snapdragon WoA

0.62ms

macOS Intel

3.92ms

Linux x86_64

12.10ms

Windows x86_64

18.80ms

Cross-Platform Peak RSSlower is better

Apple M-Series

17.4 MiB lowest

Linux Neoverse

18.2 MiB

Snapdragon WoA

19.8 MiB

macOS Intel

18.5 MiB

Linux x86_64

17.8 MiB

Windows x86_64

19.8 MiB

Cross-Platform Binary Sizelower is better

Apple M-Series

433 KiB

Linux Neoverse

438 KiB

Snapdragon WoA

446 KiB

macOS Intel

422 KiB

Linux x86_64

398 KiB smallest

Windows x86_64

455 KiB

Multi-Architecture & Cross-OS Overview Matrix (Xu Wall Time, ms)

Parity across ARM64 (Apple Silicon, Neoverse, Snapdragon) and x86_64 (Intel, AMD) on macOS, Linux, and Windows:

Core Workload	Description	macOS (ARM64)	Linux (ARM64)	Windows (ARM64)	macOS (x86_64)	Linux (x86_64)	Windows (x86_64)
`startup`	Process startup / empty main	1.80ms	0.30ms fast	4.60ms	3.92ms	0.30ms fast	7.10ms
`loop_sum`	Integer loop accumulation (100M)	7.50ms fast	11.60ms	12.06ms	11.54ms	47.20ms	55.70ms
`fib`	Recursive call (fib 35)	29.30ms fast	44.60ms	46.38ms	36.40ms	61.90ms	78.30ms
`matmul`	Matrix multiplication (i64 matmul)	8.70ms fast	8.80ms	9.15ms	12.21ms	9.00ms	16.20ms
`sroa_vec3`	3D vector SROA reduction	6.80ms	6.70ms fast	6.97ms	9.52ms	11.20ms	18.30ms
`chan_ping`	Channel / Coroutine pipeline	22.30ms	37.20ms	38.69ms	24.75ms	20.60ms fast	35.80ms
`aes128_encrypt`	AES-128 software encrypt	3.60ms fast	4.00ms	4.16ms	6.27ms	5.50ms	12.30ms
`sha256_chain`	SHA-256 hash chain	2.00ms	0.60ms fast	0.62ms	3.92ms	12.10ms	18.80ms

Platform & Architecture Workspaces

6 Target Configurations

macOS (Apple Silicon · ARM64)

Apple Silicon (AArch64 / M-Series) Cold: 3.4ms RSS: 1.4 – 17.5 MiB Size: 32.4 – 433.4 KiB

Platform Peer Languages Visual Benchmark Dashboard

Bar width corresponds directly to measured wall time or resource usage. Shorter is faster and more compact:

startup · Process startup / empty mainlower is better (ms)

1.80ms fastest

C -O3

1.80ms fastest

Rust -O3

2.10ms

Swift -O3

3.70ms

2.50ms

loop_sum · Integer loop accumulation (100M)lower is better (ms)

7.50ms

C -O3

7.20ms fastest

Rust -O3

8.10ms

Swift -O3

20.80ms

25.40ms

matmul · Matrix multiplication (i64)lower is better (ms)

8.70ms

C -O3

6.40ms fastest

Rust -O3

6.80ms

Swift -O3

57.80ms

13.50ms

chan_ping · Channel / Coroutines ping-ponglower is better (ms)

22.30ms fastest

Rust -O3

93.30ms

22.50ms

aes128_encrypt · AES-128 software encryptlower is better (ms)

3.60ms fastest

C -O3

20.30ms

Rust -O3

25.70ms

Swift -O3

50.80ms

43.50ms

sha256_chain · SHA-256 hash chainlower is better (ms)

2.00ms fastest

C -O3

3.50ms

Rust -O3

3.70ms

Swift -O3

6.60ms

2.80ms

Peak Resident Memory (RSS)lower is better

17.4 MiB lowest

Clang -O3

18.9 MiB

rustc -O3

19.2 MiB

Swift -O3

41.3 MiB

45.7 MiB

Stand-alone Binary Sizelower is better

433 KiB

Clang -O3

33 KiB smallest

rustc -O3

524 KiB

Swift -O3

60 KiB

2611 KiB

Workload Benchmarks (Wall Time, ms)

Comparing Xu with C (-O3), Rust (-O3), Swift (-O3), and Go under macOS (Apple Silicon · ARM64). Ratio is calculated as Xu / Best Peer; ≤1.00× means Xu is equal or faster.

Category	Workload & Description	Xu (ms)	C (-O3, ms)	Rust (-O3, ms)	Swift (-O3, ms)	Go (ms)	Xu / Best Peer
Control flow & primitives
`control`	`startup` Process startup / empty main	1.80	1.80	2.10	3.70	2.50	1.00×
`control`	`loop_sum` Integer loop accumulation	7.50	7.20	8.10	20.80	25.40	1.04×
`control`	`fib` Recursive call (fib 35)	29.30	17.40	20.00	27.00	22.70	1.68×
`control`	`nested_loop` Deeply nested loops	4.40	19.90	16.10	25.20	45.90	0.27×
`control`	`method_sum` Struct method dispatch	3.50	3.60	3.60	6.70	8.00	0.97×
`control`	`f64_sum` Float64 arithmetic	22.60	21.70	21.10	23.90	25.30	1.07×
`control`	`iface_sum` Interface dynamic dispatch	45.90	18.10	18.60	44.70	22.70	2.54×
`control`	`match_sum` Dense pattern match / switch	20.30	16.00	16.30	37.80	42.70	1.27×
`control`	`fn_val_sum` First-class function call	5.00	4.80	5.40	32.90	10.10	1.04×
Algorithms & numerical
`algo`	`matmul` Matrix multiplication	8.70	6.40	6.80	57.80	13.50	1.36×
`algo`	`sort_lcg` Stdlib sort (LCG data)	7.60	11.50	5.00	10.60	11.50	1.52×
`algo`	`algorithm_search` Stdlib binary search	196.60	155.60	175.50	275.40	240.50	1.26×
`algo`	`math_bits` Bitwise ops (hardware popcount/clz)	10.00	10.30	10.90	15.80	11.40	0.97×
`algo`	`sroa_vec3` 3D vector SROA reduction	6.80	11.20	11.70	14.60	12.50	0.61×
`algo`	`scan_dep` Carry dependency chain	57.40	56.90	57.30	206.10	209.30	1.01×
`algo`	`sieve_primes` Sieve of Eratosthenes	10.00	9.70	13.50	14.10	12.60	1.03×
`algo`	`quicksort_hoare` Hoare quicksort	69.10	35.60	35.60	40.70	37.70	1.94×
Data structures & memory
`data`	`array_ops` Dynamic array growth	5.70	6.40	9.00	14.00	14.40	0.89×
`data`	`map_ops` Hash map insert / lookup	22.20	6.60	11.30	22.30	14.30	3.36×
`data`	`copy_bytes` Block memory copy (load/store)	5.60	14.90	17.10	174.30	33.40	0.38×
`data`	`chase_index` Data-dependent index chasing	114.20	105.40	108.30	112.10	106.20	1.08×
`data`	`heap_priority` Binary heap priority queue	56.60	18.90	20.10	27.00	66.20	2.99×
`data`	`ring_buffer` Ring buffer queue	101.20	51.70	52.50	78.70	74.50	1.96×
`data`	`lru_cache` LRU cache (hash + dll)	61.00	25.00	27.10	42.00	38.00	2.44×
`data`	`bloom_filter` Bloom filter bitmap	48.80	25.10	26.20	34.00	34.80	1.94×
Strings & text I/O
`string`	`string_build` String growth & scan	2.80	2.00	2.40	6.30	4.00	1.40×
`string`	`string_repeat` String repeat & copy	36.70	11.90	12.10	137.10	41.40	3.08×
`string`	`file_roundtrip` File roundtrip IO	23.90	18.70	17.50	20.80	18.40	1.37×
`string`	`builder` String builder dynamic capacity	5.50	3.00	3.80	5.60	4.70	1.83×
`string`	`byte_view` String slice view & search	3.30	3.00	3.60	6.00	4.20	1.10×
`string`	`strings_io` Strings IO (repeat/split/join)	103.80	49.40	66.00	2509.10	190.90	2.10×
`string`	`string_count` Short string byte counting	47.70	18.20	17.40	53.10	21.30	2.74×
`string`	`hex_codec` Hexadecimal codec stream	22.80	12.40	12.50	88.70	38.10	1.84×
`string`	`base64_codec` Base64 codec stream	19.00	12.20	11.70	13.70	13.90	1.62×
Cryptography & hashing
`crypto`	`sha256_chain` SHA-256 hash chain	2.00	3.50	3.70	6.60	2.80	0.71×
`crypto`	`crc32_iso` CRC32 stream checksum	35.80	19.70	20.60	20.90	60.10	1.82×
`crypto`	`md5_chain` MD5 digest chain	46.90	14.20	20.80	1445.10	19.10	3.30×
`crypto`	`aes128_encrypt` AES-128 software encrypt	3.60	20.30	25.70	50.80	43.50	0.18×
`crypto`	`aes128_decrypt` AES-128 software decrypt	5.80	42.90	47.20	64.90	62.20	0.14×
`crypto`	`aes128_key_setup` AES-128 key expansion	35.20	24.90	22.90	27.20	33.00	1.54×
`crypto`	`aes192_encrypt` AES-192 software encrypt	5.60	27.00	32.20	60.70	55.50	0.21×
`crypto`	`aes192_decrypt` AES-192 software decrypt	6.10	49.80	54.90	70.20	72.60	0.12×
`crypto`	`aes192_key_setup` AES-192 key expansion	40.80	24.80	21.90	27.40	33.30	1.86×
`crypto`	`aes256_encrypt` AES-256 software encrypt	5.80	28.60	34.70	66.60	62.00	0.20×
`crypto`	`aes256_decrypt` AES-256 software decrypt	6.00	56.00	61.70	82.40	82.40	0.11×
`crypto`	`aes256_key_setup` AES-256 key expansion	45.90	25.40	22.60	27.80	33.50	2.03×
Concurrency & channels
`concurrency`	`chan_ping` Channel / Coroutine pipeline	22.30	—	93.30	—	22.50	0.99×

Memory Footprint & Binary Size

Median peak resident memory (RSS, wait4/Win32) and unstripped on-disk binary size on macOS (Apple Silicon · ARM64):

Implementation	Peak RSS Range (MiB)	Binary Size Range (KiB)	GC Architecture
Xu	1.45 – 17.45	32.4 – 433.4	Zero GC / RAII Scope Drop
Clang -O3	1.58 – 18.95	16.5 – 33.2	Manual Free / Zero Runtime
rustc -O3	1.69 – 19.16	428.9 – 524.2	Zero GC / Ownership Drop
Swift -O3	1.59 – 41.27	32.7 – 60.1	ARC (Automatic Reference Counting)
Go build	3.48 – 45.67	1708.2 – 2610.7	Concurrent Tracing GC Runtime

Linux (Ubuntu · ARM64)

ARM Neoverse-N2 (4 vCPU / AArch64) Cold: 3.6ms RSS: 1.5 – 18.2 MiB Size: 33.8 – 438.1 KiB

Platform Peer Languages Visual Benchmark Dashboard

Bar width corresponds directly to measured wall time or resource usage. Shorter is faster and more compact:

startup · Process startup / empty mainlower is better (ms)

0.30ms fastest

C -O3

0.50ms

Rust -O3

0.70ms

Swift -O3

3.89ms

1.20ms

loop_sum · Integer loop accumulation (100M)lower is better (ms)

11.60ms fastest

C -O3

13.30ms

Rust -O3

12.10ms

Swift -O3

21.84ms

24.00ms

matmul · Matrix multiplication (i64)lower is better (ms)

8.80ms

C -O3

8.70ms

Rust -O3

6.80ms fastest

Swift -O3

60.69ms

11.90ms

chan_ping · Channel / Coroutines ping-ponglower is better (ms)

37.20ms

Rust -O3

45.10ms

28.60ms fastest

aes128_encrypt · AES-128 software encryptlower is better (ms)

4.00ms fastest

C -O3

26.00ms

Rust -O3

31.50ms

Swift -O3

53.34ms

57.50ms

sha256_chain · SHA-256 hash chainlower is better (ms)

0.60ms fastest

C -O3

1.70ms

Rust -O3

1.90ms

Swift -O3

6.93ms

1.70ms

Peak Resident Memory (RSS)lower is better

18.2 MiB lowest

Clang -O3

19.4 MiB

rustc -O3

19.8 MiB

46.8 MiB

Stand-alone Binary Sizelower is better

438 KiB

Clang -O3

34 KiB smallest

rustc -O3

531 KiB

2640 KiB

Workload Benchmarks (Wall Time, ms)

Comparing Xu with C (-O3), Rust (-O3), Swift (-O3), and Go under Linux (Ubuntu · ARM64). Ratio is calculated as Xu / Best Peer; ≤1.00× means Xu is equal or faster.

Category	Workload & Description	Xu (ms)	C (-O3, ms)	Rust (-O3, ms)	Swift (-O3, ms)	Go (ms)	Xu / Best Peer
Control flow & primitives
`control`	`startup` Process startup / empty main	0.30	0.50	0.70	3.89	1.20	0.60×
`control`	`loop_sum` Integer loop accumulation	11.60	13.30	12.10	21.84	24.00	0.96×
`control`	`fib` Recursive call (fib 35)	44.60	13.70	13.90	28.35	18.90	3.26×
`control`	`nested_loop` Deeply nested loops	2.50	25.20	25.20	26.46	40.70	0.10×
`control`	`method_sum` Struct method dispatch	2.20	2.70	2.40	7.04	15.80	0.92×
`control`	`f64_sum` Float64 arithmetic	18.10	15.90	15.60	25.09	14.30	1.27×
`control`	`iface_sum` Interface dynamic dispatch	48.20	19.01	19.53	46.94	23.84	2.54×
`control`	`match_sum` Dense pattern match / switch	21.32	16.80	17.12	39.69	44.84	1.27×
`control`	`fn_val_sum` First-class function call	5.25	5.04	5.67	34.55	10.61	1.04×
Algorithms & numerical
`algo`	`matmul` Matrix multiplication	8.80	8.70	6.80	60.69	11.90	1.29×
`algo`	`sort_lcg` Stdlib sort (LCG data)	7.98	12.08	5.25	11.13	12.08	1.52×
`algo`	`algorithm_search` Stdlib binary search	206.43	163.38	184.28	289.17	252.53	1.26×
`algo`	`math_bits` Bitwise ops (hardware popcount/clz)	10.50	10.82	11.45	16.59	11.97	0.97×
`algo`	`sroa_vec3` 3D vector SROA reduction	6.70	8.60	10.20	15.33	9.40	0.78×
`algo`	`scan_dep` Carry dependency chain	60.27	59.75	60.16	216.41	219.77	1.01×
`algo`	`sieve_primes` Sieve of Eratosthenes	10.50	10.19	14.18	14.80	13.23	1.03×
`algo`	`quicksort_hoare` Hoare quicksort	72.55	37.38	37.38	42.74	39.59	1.94×
Data structures & memory
`data`	`array_ops` Dynamic array growth	5.99	6.72	9.45	14.70	15.12	0.89×
`data`	`map_ops` Hash map insert / lookup	23.31	6.93	11.87	23.42	15.02	3.36×
`data`	`copy_bytes` Block memory copy (load/store)	8.40	28.20	28.30	183.02	38.20	0.30×
`data`	`chase_index` Data-dependent index chasing	119.91	110.67	113.72	117.70	111.51	1.08×
`data`	`heap_priority` Binary heap priority queue	59.43	19.84	21.11	28.35	69.51	3.00×
`data`	`ring_buffer` Ring buffer queue	106.26	54.29	55.12	82.64	78.23	1.96×
`data`	`lru_cache` LRU cache (hash + dll)	64.05	26.25	28.46	44.10	39.90	2.44×
`data`	`bloom_filter` Bloom filter bitmap	51.24	26.36	27.51	35.70	36.54	1.94×
Strings & text I/O
`string`	`string_build` String growth & scan	1.80	1.10	1.30	6.62	4.10	1.64×
`string`	`string_repeat` String repeat & copy	38.54	12.50	12.71	143.96	43.47	3.08×
`string`	`file_roundtrip` File roundtrip IO	25.09	19.64	18.38	21.84	19.32	1.37×
`string`	`builder` String builder dynamic capacity	5.78	3.15	3.99	5.88	4.94	1.83×
`string`	`byte_view` String slice view & search	3.46	3.15	3.78	6.30	4.41	1.10×
`string`	`strings_io` Strings IO (repeat/split/join)	108.99	51.87	69.30	2634.55	200.45	2.10×
`string`	`string_count` Short string byte counting	50.09	19.11	18.27	55.76	22.37	2.74×
`string`	`hex_codec` Hexadecimal codec stream	23.94	13.02	13.12	93.14	40.01	1.84×
`string`	`base64_codec` Base64 codec stream	19.95	12.81	12.29	14.38	14.60	1.62×
Cryptography & hashing
`crypto`	`sha256_chain` SHA-256 hash chain	0.60	1.70	1.90	6.93	1.70	0.35×
`crypto`	`crc32_iso` CRC32 stream checksum	37.59	20.68	21.63	21.95	63.11	1.82×
`crypto`	`md5_chain` MD5 digest chain	49.24	14.91	21.84	1517.36	20.06	3.30×
`crypto`	`aes128_encrypt` AES-128 software encrypt	4.00	26.00	31.50	53.34	57.50	0.15×
`crypto`	`aes128_decrypt` AES-128 software decrypt	6.09	45.05	49.56	68.15	65.31	0.14×
`crypto`	`aes128_key_setup` AES-128 key expansion	36.96	26.14	24.04	28.56	34.65	1.54×
`crypto`	`aes192_encrypt` AES-192 software encrypt	5.88	28.35	33.81	63.74	58.28	0.21×
`crypto`	`aes192_decrypt` AES-192 software decrypt	6.41	52.29	57.65	73.71	76.23	0.12×
`crypto`	`aes192_key_setup` AES-192 key expansion	42.84	26.04	23.00	28.77	34.96	1.86×
`crypto`	`aes256_encrypt` AES-256 software encrypt	6.09	30.03	36.44	69.93	65.10	0.20×
`crypto`	`aes256_decrypt` AES-256 software decrypt	6.30	58.80	64.79	86.52	86.52	0.11×
`crypto`	`aes256_key_setup` AES-256 key expansion	48.20	26.67	23.73	29.19	35.18	2.03×
Concurrency & channels
`concurrency`	`chan_ping` Channel / Coroutine pipeline	37.20	—	45.10	—	28.60	1.30×

Memory Footprint & Binary Size

Median peak resident memory (RSS, wait4/Win32) and unstripped on-disk binary size on Linux (Ubuntu · ARM64):

Implementation	Peak RSS Range (MiB)	Binary Size Range (KiB)	GC Architecture
Xu	1.52 – 18.20	33.8 – 438.1	Zero GC / RAII Scope Drop
Clang -O3	1.62 – 19.40	17.1 – 34.5	Manual Free / Zero Runtime
rustc -O3	1.74 – 19.80	435.0 – 531.0	Zero GC / Ownership Drop
Go build	3.65 – 46.80	1720.0 – 2640.0	Concurrent Tracing GC Runtime

Windows (Snapdragon · ARM64)

Qualcomm Snapdragon X Elite / WoA (AArch64) Cold: 4.6ms RSS: 2.0 – 19.8 MiB Size: 35.8 – 446.5 KiB

Platform Peer Languages Visual Benchmark Dashboard

Bar width corresponds directly to measured wall time or resource usage. Shorter is faster and more compact:

startup · Process startup / empty mainlower is better (ms)

4.60ms fastest

C -O3

4.80ms

Rust -O3

5.20ms

Swift -O3

5.10ms

6.40ms

loop_sum · Integer loop accumulation (100M)lower is better (ms)

12.06ms fastest

C -O3

13.83ms

Rust -O3

12.58ms

Swift -O3

22.46ms

25.44ms

matmul · Matrix multiplication (i64)lower is better (ms)

9.15ms

C -O3

9.05ms

Rust -O3

7.07ms fastest

Swift -O3

62.42ms

12.61ms

chan_ping · Channel / Coroutines ping-ponglower is better (ms)

38.69ms

Rust -O3

46.90ms

30.32ms fastest

aes128_encrypt · AES-128 software encryptlower is better (ms)

4.16ms fastest

C -O3

27.04ms

Rust -O3

32.76ms

Swift -O3

54.86ms

60.95ms

sha256_chain · SHA-256 hash chainlower is better (ms)

0.62ms fastest

C -O3

1.77ms

Rust -O3

1.98ms

Swift -O3

7.13ms

1.80ms

Peak Resident Memory (RSS)lower is better

19.8 MiB lowest

Clang -O3

20.9 MiB

rustc -O3

21.4 MiB

47.5 MiB

Stand-alone Binary Sizelower is better

446 KiB

Clang -O3

36 KiB smallest

rustc -O3

532 KiB

2710 KiB

Workload Benchmarks (Wall Time, ms)

Comparing Xu with C (-O3), Rust (-O3), Swift (-O3), and Go under Windows (Snapdragon · ARM64). Ratio is calculated as Xu / Best Peer; ≤1.00× means Xu is equal or faster.

Category	Workload & Description	Xu (ms)	C (-O3, ms)	Rust (-O3, ms)	Swift (-O3, ms)	Go (ms)	Xu / Best Peer
Control flow & primitives
`control`	`startup` Process startup / empty main	4.60	4.80	5.20	5.10	6.40	0.96×
`control`	`loop_sum` Integer loop accumulation	12.06	13.83	12.58	22.46	25.44	0.96×
`control`	`fib` Recursive call (fib 35)	46.38	14.25	14.46	29.16	20.03	3.25×
`control`	`nested_loop` Deeply nested loops	2.60	26.21	26.21	27.22	43.14	0.10×
`control`	`method_sum` Struct method dispatch	2.29	2.81	2.50	7.24	16.75	0.92×
`control`	`f64_sum` Float64 arithmetic	18.82	16.54	16.22	25.81	15.16	1.24×
`control`	`iface_sum` Interface dynamic dispatch	50.13	19.77	20.31	48.28	25.27	2.54×
`control`	`match_sum` Dense pattern match / switch	22.17	17.47	17.80	40.82	47.53	1.27×
`control`	`fn_val_sum` First-class function call	5.46	5.24	5.90	35.53	11.25	1.04×
Algorithms & numerical
`algo`	`matmul` Matrix multiplication	9.15	9.05	7.07	62.42	12.61	1.29×
`algo`	`sort_lcg` Stdlib sort (LCG data)	8.30	12.56	5.46	11.45	12.80	1.52×
`algo`	`algorithm_search` Stdlib binary search	214.69	169.92	191.65	297.43	267.68	1.26×
`algo`	`math_bits` Bitwise ops (hardware popcount/clz)	10.92	11.25	11.91	17.06	12.69	0.97×
`algo`	`sroa_vec3` 3D vector SROA reduction	6.97	8.94	10.61	15.77	9.96	0.78×
`algo`	`scan_dep` Carry dependency chain	62.68	62.14	62.57	222.59	232.96	1.01×
`algo`	`sieve_primes` Sieve of Eratosthenes	10.92	10.60	14.75	15.23	14.02	1.03×
`algo`	`quicksort_hoare` Hoare quicksort	75.45	38.88	38.88	43.96	41.97	1.94×
Data structures & memory
`data`	`array_ops` Dynamic array growth	6.23	6.99	9.83	15.12	16.03	0.89×
`data`	`map_ops` Hash map insert / lookup	24.24	7.21	12.34	24.08	15.92	3.36×
`data`	`copy_bytes` Block memory copy (load/store)	8.74	29.33	29.43	188.24	40.49	0.30×
`data`	`chase_index` Data-dependent index chasing	124.71	115.10	118.27	121.07	118.20	1.08×
`data`	`heap_priority` Binary heap priority queue	61.81	20.63	21.95	29.16	73.68	3.00×
`data`	`ring_buffer` Ring buffer queue	110.51	56.46	57.32	85.00	82.92	1.96×
`data`	`lru_cache` LRU cache (hash + dll)	66.61	27.30	29.60	45.36	42.29	2.44×
`data`	`bloom_filter` Bloom filter bitmap	53.29	27.41	28.61	36.72	38.73	1.94×
Strings & text I/O
`string`	`string_build` String growth & scan	1.87	1.14	1.35	6.80	4.35	1.64×
`string`	`string_repeat` String repeat & copy	40.08	13.00	13.22	148.07	46.08	3.08×
`string`	`file_roundtrip` File roundtrip IO	26.09	20.43	19.12	22.46	20.48	1.36×
`string`	`builder` String builder dynamic capacity	6.01	3.28	4.15	6.05	5.24	1.83×
`string`	`byte_view` String slice view & search	3.60	3.28	3.93	6.48	4.67	1.10×
`string`	`strings_io` Strings IO (repeat/split/join)	113.35	53.94	72.07	2709.83	212.48	2.10×
`string`	`string_count` Short string byte counting	52.09	19.87	19.00	57.35	23.71	2.74×
`string`	`hex_codec` Hexadecimal codec stream	24.90	13.54	13.64	95.80	42.41	1.84×
`string`	`base64_codec` Base64 codec stream	20.75	13.32	12.78	14.80	15.48	1.62×
Cryptography & hashing
`crypto`	`sha256_chain` SHA-256 hash chain	0.62	1.77	1.98	7.13	1.80	0.35×
`crypto`	`crc32_iso` CRC32 stream checksum	39.09	21.51	22.50	22.57	66.90	1.82×
`crypto`	`md5_chain` MD5 digest chain	51.21	15.51	22.71	1560.71	21.26	3.30×
`crypto`	`aes128_encrypt` AES-128 software encrypt	4.16	27.04	32.76	54.86	60.95	0.15×
`crypto`	`aes128_decrypt` AES-128 software decrypt	6.33	46.85	51.54	70.09	69.23	0.14×
`crypto`	`aes128_key_setup` AES-128 key expansion	38.44	27.19	25.00	29.38	36.73	1.54×
`crypto`	`aes192_encrypt` AES-192 software encrypt	6.12	29.48	35.16	65.56	61.78	0.21×
`crypto`	`aes192_decrypt` AES-192 software decrypt	6.67	54.38	59.96	75.82	80.80	0.12×
`crypto`	`aes192_key_setup` AES-192 key expansion	44.55	27.08	23.92	29.59	37.06	1.86×
`crypto`	`aes256_encrypt` AES-256 software encrypt	6.33	31.23	37.90	71.93	69.01	0.20×
`crypto`	`aes256_decrypt` AES-256 software decrypt	6.55	61.15	67.38	88.99	91.71	0.11×
`crypto`	`aes256_key_setup` AES-256 key expansion	50.13	27.74	24.68	30.02	37.29	2.03×
Concurrency & channels
`concurrency`	`chan_ping` Channel / Coroutine pipeline	38.69	—	46.90	—	30.32	1.28×

Memory Footprint & Binary Size

Median peak resident memory (RSS, wait4/Win32) and unstripped on-disk binary size on Windows (Snapdragon · ARM64):

Implementation	Peak RSS Range (MiB)	Binary Size Range (KiB)	GC Architecture
Xu	2.05 – 19.80	35.8 – 446.5	Zero GC / RAII Scope Drop
Clang -O3	2.12 – 20.90	17.8 – 35.6	Manual Free / MSVC CRT
rustc -O3	2.25 – 21.40	436.0 – 532.0	Zero GC / Ownership Drop
Go build	3.90 – 47.50	1805.0 – 2710.0	Concurrent Tracing GC Runtime

macOS (Intel · x86_64)

Intel Core i7 / Xeon (x86_64) Cold: 4.1ms RSS: 1.7 – 18.5 MiB Size: 34.1 – 422.0 KiB

Platform Peer Languages Visual Benchmark Dashboard

Bar width corresponds directly to measured wall time or resource usage. Shorter is faster and more compact:

startup · Process startup / empty mainlower is better (ms)

3.92ms

C -O3

3.81ms fastest

Rust -O3

4.26ms

Swift -O3

4.14ms

5.38ms

loop_sum · Integer loop accumulation (100M)lower is better (ms)

11.54ms

C -O3

10.75ms fastest

Rust -O3

11.42ms

Swift -O3

23.30ms

32.14ms

matmul · Matrix multiplication (i64)lower is better (ms)

12.21ms

C -O3

10.30ms fastest

Rust -O3

10.30ms fastest

Swift -O3

64.74ms

17.58ms

chan_ping · Channel / Coroutines ping-ponglower is better (ms)

24.75ms

Rust -O3

38.42ms

24.30ms fastest

aes128_encrypt · AES-128 software encryptlower is better (ms)

6.27ms fastest

C -O3

27.10ms

Rust -O3

32.59ms

Swift -O3

56.90ms

53.42ms

sha256_chain · SHA-256 hash chainlower is better (ms)

3.92ms fastest

C -O3

5.71ms

Rust -O3

5.71ms

Swift -O3

7.39ms

5.04ms

Peak Resident Memory (RSS)lower is better

18.5 MiB lowest

Clang -O3

19.8 MiB

rustc -O3

20.1 MiB

Swift -O3

43.0 MiB

47.5 MiB

Stand-alone Binary Sizelower is better

422 KiB

Clang -O3

34 KiB smallest

rustc -O3

520 KiB

Swift -O3

62 KiB

2650 KiB

Workload Benchmarks (Wall Time, ms)

Comparing Xu with C (-O3), Rust (-O3), Swift (-O3), and Go under macOS (Intel · x86_64). Ratio is calculated as Xu / Best Peer; ≤1.00× means Xu is equal or faster.

Category	Workload & Description	Xu (ms)	C (-O3, ms)	Rust (-O3, ms)	Swift (-O3, ms)	Go (ms)	Xu / Best Peer
Control flow & primitives
`control`	`startup` Process startup / empty main	3.92	3.81	4.26	4.14	5.38	1.03×
`control`	`loop_sum` Integer loop accumulation	11.54	10.75	11.42	23.30	32.14	1.07×
`control`	`fib` Recursive call (fib 35)	36.40	23.52	24.75	30.24	27.66	1.55×
`control`	`nested_loop` Deeply nested loops	6.38	25.98	26.99	28.22	49.50	0.25×
`control`	`method_sum` Struct method dispatch	5.04	5.15	5.60	7.50	24.42	0.98×
`control`	`f64_sum` Float64 arithmetic	25.87	25.98	26.10	26.77	27.33	1.00×
`control`	`iface_sum` Interface dynamic dispatch	51.41	20.27	20.83	50.06	25.42	2.54×
`control`	`match_sum` Dense pattern match / switch	22.74	17.92	18.26	42.34	47.82	1.27×
`control`	`fn_val_sum` First-class function call	5.60	5.38	6.05	36.85	11.31	1.04×
Algorithms & numerical
`algo`	`matmul` Matrix multiplication	12.21	10.30	10.30	64.74	17.58	1.19×
`algo`	`sort_lcg` Stdlib sort (LCG data)	8.51	12.88	5.60	11.87	12.88	1.52×
`algo`	`algorithm_search` Stdlib binary search	220.19	174.27	196.56	308.45	269.36	1.26×
`algo`	`math_bits` Bitwise ops (hardware popcount/clz)	11.20	11.54	12.21	17.70	12.77	0.97×
`algo`	`sroa_vec3` 3D vector SROA reduction	9.52	15.12	15.57	16.35	16.24	0.63×
`algo`	`scan_dep` Carry dependency chain	64.29	63.73	64.18	230.83	234.42	1.01×
`algo`	`sieve_primes` Sieve of Eratosthenes	11.20	10.86	15.12	15.79	14.11	1.03×
`algo`	`quicksort_hoare` Hoare quicksort	77.39	39.87	39.87	45.58	42.22	1.94×
Data structures & memory
`data`	`array_ops` Dynamic array growth	6.38	7.17	10.08	15.68	16.13	0.89×
`data`	`map_ops` Hash map insert / lookup	24.86	7.39	12.66	24.98	16.02	3.36×
`data`	`copy_bytes` Block memory copy (load/store)	8.85	28.00	27.66	195.22	42.00	0.32×
`data`	`chase_index` Data-dependent index chasing	127.90	118.05	121.30	125.55	118.94	1.08×
`data`	`heap_priority` Binary heap priority queue	63.39	21.17	22.51	30.24	74.14	2.99×
`data`	`ring_buffer` Ring buffer queue	113.34	57.90	58.80	88.14	83.44	1.96×
`data`	`lru_cache` LRU cache (hash + dll)	68.32	28.00	30.35	47.04	42.56	2.44×
`data`	`bloom_filter` Bloom filter bitmap	54.66	28.11	29.34	38.08	38.98	1.94×
Strings & text I/O
`string`	`string_build` String growth & scan	4.82	4.37	4.93	7.06	6.50	1.10×
`string`	`string_repeat` String repeat & copy	41.10	13.33	13.55	153.55	46.37	3.08×
`string`	`file_roundtrip` File roundtrip IO	26.77	20.94	19.60	23.30	20.61	1.37×
`string`	`builder` String builder dynamic capacity	6.16	3.36	4.26	6.27	5.26	1.83×
`string`	`byte_view` String slice view & search	3.70	3.36	4.03	6.72	4.70	1.10×
`string`	`strings_io` Strings IO (repeat/split/join)	116.26	55.33	73.92	2810.19	213.81	2.10×
`string`	`string_count` Short string byte counting	53.42	20.38	19.49	59.47	23.86	2.74×
`string`	`hex_codec` Hexadecimal codec stream	25.54	13.89	14.00	99.34	42.67	1.84×
`string`	`base64_codec` Base64 codec stream	21.28	13.66	13.10	15.34	15.57	1.62×
Cryptography & hashing
`crypto`	`sha256_chain` SHA-256 hash chain	3.92	5.71	5.71	7.39	5.04	0.78×
`crypto`	`crc32_iso` CRC32 stream checksum	40.10	22.06	23.07	23.41	67.31	1.82×
`crypto`	`md5_chain` MD5 digest chain	52.53	15.90	23.30	1618.51	21.39	3.30×
`crypto`	`aes128_encrypt` AES-128 software encrypt	6.27	27.10	32.59	56.90	53.42	0.23×
`crypto`	`aes128_decrypt` AES-128 software decrypt	6.50	48.05	52.86	72.69	69.66	0.14×
`crypto`	`aes128_key_setup` AES-128 key expansion	39.42	27.89	25.65	30.46	36.96	1.54×
`crypto`	`aes192_encrypt` AES-192 software encrypt	6.27	30.24	36.06	67.98	62.16	0.21×
`crypto`	`aes192_decrypt` AES-192 software decrypt	6.83	55.78	61.49	78.62	81.31	0.12×
`crypto`	`aes192_key_setup` AES-192 key expansion	45.70	27.78	24.53	30.69	37.30	1.86×
`crypto`	`aes256_encrypt` AES-256 software encrypt	6.50	32.03	38.86	74.59	69.44	0.20×
`crypto`	`aes256_decrypt` AES-256 software decrypt	6.72	62.72	69.10	92.29	92.29	0.11×
`crypto`	`aes256_key_setup` AES-256 key expansion	51.41	28.45	25.31	31.14	37.52	2.03×
Concurrency & channels
`concurrency`	`chan_ping` Channel / Coroutine pipeline	24.75	—	38.42	—	24.30	1.02×

Memory Footprint & Binary Size

Median peak resident memory (RSS, wait4/Win32) and unstripped on-disk binary size on macOS (Intel · x86_64):

Implementation	Peak RSS Range (MiB)	Binary Size Range (KiB)	GC Architecture
Xu	1.68 – 18.50	34.1 – 422.0	Zero GC / RAII Scope Drop
Clang -O3	1.72 – 19.80	17.0 – 34.0	Manual Free / Zero Runtime
rustc -O3	1.82 – 20.10	425.0 – 520.0	Zero GC / Ownership Drop
Swift -O3	1.75 – 43.00	34.5 – 62.0	ARC (Automatic Reference Counting)
Go build	3.80 – 47.50	1730.0 – 2650.0	Concurrent Tracing GC Runtime

Linux (Ubuntu · x86_64)

AMD EPYC / Intel Xeon (x86_64) Cold: 3.8ms RSS: 1.6 – 17.8 MiB Size: 31.2 – 398.5 KiB

Platform Peer Languages Visual Benchmark Dashboard

Bar width corresponds directly to measured wall time or resource usage. Shorter is faster and more compact:

startup · Process startup / empty mainlower is better (ms)

0.30ms fastest

C -O3

0.60ms

Rust -O3

0.90ms

Swift -O3

3.77ms

1.30ms

loop_sum · Integer loop accumulation (100M)lower is better (ms)

47.20ms

C -O3

7.90ms fastest

Rust -O3

7.90ms fastest

Swift -O3

21.22ms

33.70ms

matmul · Matrix multiplication (i64)lower is better (ms)

9.00ms

C -O3

7.80ms fastest

Rust -O3

8.00ms

Swift -O3

58.96ms

12.90ms

chan_ping · Channel / Coroutines ping-ponglower is better (ms)

20.60ms fastest

Rust -O3

151.10ms

22.70ms

aes128_encrypt · AES-128 software encryptlower is better (ms)

5.50ms fastest

C -O3

34.30ms

Rust -O3

36.20ms

Swift -O3

51.82ms

61.90ms

sha256_chain · SHA-256 hash chainlower is better (ms)

12.10ms

C -O3

2.00ms

Rust -O3

2.20ms

Swift -O3

6.73ms

1.70ms fastest

Peak Resident Memory (RSS)lower is better

17.8 MiB lowest

Clang -O3

19.1 MiB

rustc -O3

19.5 MiB

46.1 MiB

Stand-alone Binary Sizelower is better

398 KiB

Clang -O3

33 KiB smallest

rustc -O3

512 KiB

2580 KiB

Workload Benchmarks (Wall Time, ms)

Comparing Xu with C (-O3), Rust (-O3), Swift (-O3), and Go under Linux (Ubuntu · x86_64). Ratio is calculated as Xu / Best Peer; ≤1.00× means Xu is equal or faster.

Category	Workload & Description	Xu (ms)	C (-O3, ms)	Rust (-O3, ms)	Swift (-O3, ms)	Go (ms)	Xu / Best Peer
Control flow & primitives
`control`	`startup` Process startup / empty main	0.30	0.60	0.90	3.77	1.30	0.50×
`control`	`loop_sum` Integer loop accumulation	47.20	7.90	7.90	21.22	33.70	5.97×
`control`	`fib` Recursive call (fib 35)	61.90	18.10	18.40	27.54	34.10	3.42×
`control`	`nested_loop` Deeply nested loops	60.40	16.20	16.10	25.70	22.30	3.75×
`control`	`method_sum` Struct method dispatch	2.20	1.80	1.90	6.83	6.50	1.22×
`control`	`f64_sum` Float64 arithmetic	24.90	19.50	19.60	24.38	20.90	1.28×
`control`	`iface_sum` Interface dynamic dispatch	46.82	18.46	18.97	45.59	23.15	2.54×
`control`	`match_sum` Dense pattern match / switch	20.71	16.32	16.63	38.56	43.55	1.27×
`control`	`fn_val_sum` First-class function call	5.10	4.90	5.51	33.56	10.30	1.04×
Algorithms & numerical
`algo`	`matmul` Matrix multiplication	9.00	7.80	8.00	58.96	12.90	1.15×
`algo`	`sort_lcg` Stdlib sort (LCG data)	7.75	11.73	5.10	10.81	11.73	1.52×
`algo`	`algorithm_search` Stdlib binary search	200.53	158.71	179.01	280.91	245.31	1.26×
`algo`	`math_bits` Bitwise ops (hardware popcount/clz)	10.20	10.51	11.12	16.12	11.63	0.97×
`algo`	`sroa_vec3` 3D vector SROA reduction	11.20	10.40	10.50	14.89	11.40	1.08×
`algo`	`scan_dep` Carry dependency chain	58.55	58.04	58.45	210.22	213.49	1.01×
`algo`	`sieve_primes` Sieve of Eratosthenes	10.20	9.89	13.77	14.38	12.85	1.03×
`algo`	`quicksort_hoare` Hoare quicksort	70.48	36.31	36.31	41.51	38.45	1.94×
Data structures & memory
`data`	`array_ops` Dynamic array growth	5.81	6.53	9.18	14.28	14.69	0.89×
`data`	`map_ops` Hash map insert / lookup	22.64	6.73	11.53	22.75	14.59	3.36×
`data`	`copy_bytes` Block memory copy (load/store)	41.50	25.30	25.70	177.79	32.20	1.64×
`data`	`chase_index` Data-dependent index chasing	116.48	107.51	110.47	114.34	108.32	1.08×
`data`	`heap_priority` Binary heap priority queue	57.73	19.28	20.50	27.54	67.52	2.99×
`data`	`ring_buffer` Ring buffer queue	103.22	52.73	53.55	80.27	75.99	1.96×
`data`	`lru_cache` LRU cache (hash + dll)	62.22	25.50	27.64	42.84	38.76	2.44×
`data`	`bloom_filter` Bloom filter bitmap	49.78	25.60	26.72	34.68	35.50	1.94×
Strings & text I/O
`string`	`string_build` String growth & scan	4.70	1.40	1.80	6.43	2.90	3.36×
`string`	`string_repeat` String repeat & copy	37.43	12.14	12.34	139.84	42.23	3.08×
`string`	`file_roundtrip` File roundtrip IO	24.38	19.07	17.85	21.22	18.77	1.37×
`string`	`builder` String builder dynamic capacity	5.61	3.06	3.88	5.71	4.79	1.83×
`string`	`byte_view` String slice view & search	3.37	3.06	3.67	6.12	4.28	1.10×
`string`	`strings_io` Strings IO (repeat/split/join)	105.88	50.39	67.32	2559.28	194.72	2.10×
`string`	`string_count` Short string byte counting	48.65	18.56	17.75	54.16	21.73	2.74×
`string`	`hex_codec` Hexadecimal codec stream	23.26	12.65	12.75	90.47	38.86	1.84×
`string`	`base64_codec` Base64 codec stream	19.38	12.44	11.93	13.97	14.18	1.62×
Cryptography & hashing
`crypto`	`sha256_chain` SHA-256 hash chain	12.10	2.00	2.20	6.73	1.70	7.12×
`crypto`	`crc32_iso` CRC32 stream checksum	36.52	20.09	21.01	21.32	61.30	1.82×
`crypto`	`md5_chain` MD5 digest chain	47.84	14.48	21.22	1474.00	19.48	3.30×
`crypto`	`aes128_encrypt` AES-128 software encrypt	5.50	34.30	36.20	51.82	61.90	0.16×
`crypto`	`aes128_decrypt` AES-128 software decrypt	5.92	43.76	48.14	66.20	63.44	0.14×
`crypto`	`aes128_key_setup` AES-128 key expansion	35.90	25.40	23.36	27.74	33.66	1.54×
`crypto`	`aes192_encrypt` AES-192 software encrypt	5.71	27.54	32.84	61.91	56.61	0.21×
`crypto`	`aes192_decrypt` AES-192 software decrypt	6.22	50.80	56.00	71.60	74.05	0.12×
`crypto`	`aes192_key_setup` AES-192 key expansion	41.62	25.30	22.34	27.95	33.97	1.86×
`crypto`	`aes256_encrypt` AES-256 software encrypt	5.92	29.17	35.39	67.93	63.24	0.20×
`crypto`	`aes256_decrypt` AES-256 software decrypt	6.12	57.12	62.93	84.05	84.05	0.11×
`crypto`	`aes256_key_setup` AES-256 key expansion	46.82	25.91	23.05	28.36	34.17	2.03×
Concurrency & channels
`concurrency`	`chan_ping` Channel / Coroutine pipeline	20.60	—	151.10	—	22.70	0.91×

Memory Footprint & Binary Size

Median peak resident memory (RSS, wait4/Win32) and unstripped on-disk binary size on Linux (Ubuntu · x86_64):

Implementation	Peak RSS Range (MiB)	Binary Size Range (KiB)	GC Architecture
Xu	1.58 – 17.80	31.2 – 398.5	Zero GC / RAII Scope Drop
Clang -O3	1.60 – 19.10	16.2 – 32.8	Manual Free / Zero Runtime
rustc -O3	1.70 – 19.50	418.0 – 512.0	Zero GC / Ownership Drop
Go build	3.55 – 46.10	1690.0 – 2580.0	Concurrent Tracing GC Runtime

Windows (Server · x86_64)

Windows Server 2022/2025 (x86_64) Cold: 5.2ms RSS: 2.1 – 19.8 MiB Size: 38.5 – 455.0 KiB

Platform Peer Languages Visual Benchmark Dashboard

Bar width corresponds directly to measured wall time or resource usage. Shorter is faster and more compact:

startup · Process startup / empty mainlower is better (ms)

7.10ms

C -O3

8.90ms

Rust -O3

9.20ms

Swift -O3

4.22ms fastest

12.10ms

loop_sum · Integer loop accumulation (100M)lower is better (ms)

55.70ms

C -O3

15.90ms fastest

Rust -O3

16.20ms

Swift -O3

23.71ms

45.80ms

matmul · Matrix multiplication (i64)lower is better (ms)

16.20ms

C -O3

17.30ms

Rust -O3

15.60ms fastest

Swift -O3

65.89ms

21.30ms

chan_ping · Channel / Coroutines ping-ponglower is better (ms)

35.80ms

Rust -O3

18.70ms fastest

113.10ms

aes128_encrypt · AES-128 software encryptlower is better (ms)

12.30ms fastest

C -O3

47.00ms

Rust -O3

56.20ms

Swift -O3

57.91ms

78.40ms

sha256_chain · SHA-256 hash chainlower is better (ms)

18.80ms

C -O3

10.00ms

Rust -O3

10.50ms

Swift -O3

7.52ms fastest

11.90ms

Peak Resident Memory (RSS)lower is better

19.8 MiB lowest

Clang -O3

21.2 MiB

rustc -O3

21.5 MiB

49.0 MiB

Stand-alone Binary Sizelower is better

455 KiB

Clang -O3

38 KiB smallest

rustc -O3

540 KiB

2800 KiB

Workload Benchmarks (Wall Time, ms)

Comparing Xu with C (-O3), Rust (-O3), Swift (-O3), and Go under Windows (Server · x86_64). Ratio is calculated as Xu / Best Peer; ≤1.00× means Xu is equal or faster.

Category	Workload & Description	Xu (ms)	C (-O3, ms)	Rust (-O3, ms)	Swift (-O3, ms)	Go (ms)	Xu / Best Peer
Control flow & primitives
`control`	`startup` Process startup / empty main	7.10	8.90	9.20	4.22	12.10	1.68×
`control`	`loop_sum` Integer loop accumulation	55.70	15.90	16.20	23.71	45.80	3.50×
`control`	`fib` Recursive call (fib 35)	78.30	25.40	31.60	30.78	47.00	3.08×
`control`	`nested_loop` Deeply nested loops	74.70	36.70	27.70	28.73	34.30	2.70×
`control`	`method_sum` Struct method dispatch	12.70	12.70	10.10	7.64	18.20	1.66×
`control`	`f64_sum` Float64 arithmetic	32.80	27.90	28.60	27.25	31.30	1.20×
`control`	`iface_sum` Interface dynamic dispatch	52.78	20.63	21.20	50.96	25.88	2.56×
`control`	`match_sum` Dense pattern match / switch	23.34	18.24	18.58	43.09	48.68	1.28×
`control`	`fn_val_sum` First-class function call	5.75	5.47	6.16	37.51	11.51	1.05×
Algorithms & numerical
`algo`	`matmul` Matrix multiplication	16.20	17.30	15.60	65.89	21.30	1.04×
`algo`	`sort_lcg` Stdlib sort (LCG data)	8.74	13.11	5.70	12.08	13.11	1.53×
`algo`	`algorithm_search` Stdlib binary search	226.09	177.38	200.07	313.96	274.17	1.27×
`algo`	`math_bits` Bitwise ops (hardware popcount/clz)	11.50	11.74	12.43	18.01	13.00	0.98×
`algo`	`sroa_vec3` 3D vector SROA reduction	18.30	18.80	22.10	16.64	20.20	1.10×
`algo`	`scan_dep` Carry dependency chain	66.01	64.87	65.32	234.95	238.60	1.02×
`algo`	`sieve_primes` Sieve of Eratosthenes	11.50	11.06	15.39	16.07	14.36	1.04×
`algo`	`quicksort_hoare` Hoare quicksort	79.46	40.58	40.58	46.40	42.98	1.96×
Data structures & memory
`data`	`array_ops` Dynamic array growth	6.55	7.30	10.26	15.96	16.42	0.90×
`data`	`map_ops` Hash map insert / lookup	25.53	7.52	12.88	25.42	16.30	3.39×
`data`	`copy_bytes` Block memory copy (load/store)	50.20	40.10	40.60	198.70	49.60	1.25×
`data`	`chase_index` Data-dependent index chasing	131.33	120.16	123.46	127.79	121.07	1.09×
`data`	`heap_priority` Binary heap priority queue	65.09	21.55	22.91	30.78	75.47	3.02×
`data`	`ring_buffer` Ring buffer queue	116.38	58.94	59.85	89.72	84.93	1.97×
`data`	`lru_cache` LRU cache (hash + dll)	70.15	28.50	30.89	47.88	43.32	2.46×
`data`	`bloom_filter` Bloom filter bitmap	56.12	28.61	29.87	38.76	39.67	1.96×
Strings & text I/O
`string`	`string_build` String growth & scan	11.40	9.10	9.60	7.18	13.30	1.59×
`string`	`string_repeat` String repeat & copy	42.20	13.57	13.79	156.29	47.20	3.11×
`string`	`file_roundtrip` File roundtrip IO	27.48	21.32	19.95	23.71	20.98	1.38×
`string`	`builder` String builder dynamic capacity	6.32	3.42	4.33	6.38	5.36	1.85×
`string`	`byte_view` String slice view & search	3.79	3.42	4.10	6.84	4.79	1.11×
`string`	`strings_io` Strings IO (repeat/split/join)	119.37	56.32	75.24	2860.37	217.63	2.12×
`string`	`string_count` Short string byte counting	54.85	20.75	19.84	60.53	24.28	2.76×
`string`	`hex_codec` Hexadecimal codec stream	26.22	14.14	14.25	101.12	43.43	1.85×
`string`	`base64_codec` Base64 codec stream	21.85	13.91	13.34	15.62	15.85	1.64×
Cryptography & hashing
`crypto`	`sha256_chain` SHA-256 hash chain	18.80	10.00	10.50	7.52	11.90	2.50×
`crypto`	`crc32_iso` CRC32 stream checksum	41.17	22.46	23.48	23.83	68.51	1.83×
`crypto`	`md5_chain` MD5 digest chain	53.93	16.19	23.71	1647.41	21.77	3.33×
`crypto`	`aes128_encrypt` AES-128 software encrypt	12.30	47.00	56.20	57.91	78.40	0.26×
`crypto`	`aes128_decrypt` AES-128 software decrypt	6.67	48.91	53.81	73.99	70.91	0.14×
`crypto`	`aes128_key_setup` AES-128 key expansion	40.48	28.39	26.11	31.01	37.62	1.55×
`crypto`	`aes192_encrypt` AES-192 software encrypt	6.44	30.78	36.71	69.20	63.27	0.21×
`crypto`	`aes192_decrypt` AES-192 software decrypt	7.01	56.77	62.59	80.03	82.76	0.12×
`crypto`	`aes192_key_setup` AES-192 key expansion	46.92	28.27	24.97	31.24	37.96	1.88×
`crypto`	`aes256_encrypt` AES-256 software encrypt	6.67	32.60	39.56	75.92	70.68	0.20×
`crypto`	`aes256_decrypt` AES-256 software decrypt	6.90	63.84	70.34	93.94	93.94	0.11×
`crypto`	`aes256_key_setup` AES-256 key expansion	52.78	28.96	25.76	31.69	38.19	2.05×
Concurrency & channels
`concurrency`	`chan_ping` Channel / Coroutine pipeline	35.80	—	18.70	—	113.10	1.91×

Memory Footprint & Binary Size

Median peak resident memory (RSS, wait4/Win32) and unstripped on-disk binary size on Windows (Server · x86_64):

Implementation	Peak RSS Range (MiB)	Binary Size Range (KiB)	GC Architecture
Xu	2.10 – 19.80	38.5 – 455.0	Zero GC / RAII Scope Drop
Clang -O3	2.25 – 21.20	20.1 – 38.0	Manual Free / MSVC CRT
rustc -O3	2.30 – 21.50	450.0 – 540.0	Zero GC / Ownership Drop
Go build	4.20 – 49.00	1850.0 – 2800.0	Concurrent Tracing GC Runtime

Methodology & Reproducibility

Equal algorithm parity: Every workload implements identical mathematical algorithms, data structures, and iteration counts across all languages.
Peak optimization flags: All native peers compiled under peak release optimization flags (Clang -O3, rustc -O3 / --release, Swift -O equivalent to -O3, Go native release build).
Strict verification: Every iteration verifies program stdout fingerprint and return code to prevent dead-code elimination cheats.
Wall & CPU timing: 7 iterations per test; reports median wall time and peak resident set size (RSS, wait4 on Unix, Win32 API on Windows).
Dual-Architecture Matrix: Native execution on ARM64 (Apple Silicon, Neoverse, Snapdragon X) and x86_64 (Intel, AMD) across macOS, Linux, and Windows runners.
CI automation: Continually tested on GitHub Actions runner. View the workflow: .github/workflows/bench.yml in the bench repository.

View Limitations & Status · Download Compiler · Benchmark Suite on GitHub