Module hash 

Structs

CheckOptions

Options for check mode.

CheckResult

Result of check mode verification.

Enums

HashAlgorithm

Supported hash algorithms.

Functions

blake2b_hash_data

Hash raw data with BLAKE2b variable output length. output_bytes is the output size in bytes (e.g., 32 for 256-bit).

blake2b_hash_file

Hash a file with BLAKE2b variable output length. Uses mmap for large files (zero-copy), single-read for small files, and streaming read as fallback.

blake2b_hash_files_many

Batch-hash multiple files with BLAKE2b using multi-buffer SIMD.

blake2b_hash_reader

Hash a reader with BLAKE2b variable output length. Uses thread-local buffer for cache-friendly streaming.

blake2b_hash_stdin

Hash stdin with BLAKE2b variable output length. Tries fadvise if stdin is a regular file (shell redirect), then streams.

check_file

Verify checksums from a check file. Each line should be “hash filename” or “hash *filename” or “ALGO (filename) = hash”.

hash_bytes

Compute hash of a byte slice directly (zero-copy fast path).

hash_file

Hash a file by path. Uses mmap for large files (zero-copy, no read() syscalls), single-read + single-shot hash for small files, and streaming read as fallback.

hash_file_nostat

Hash a file without fstat — just open, read until EOF, hash. For many-file workloads (100+ tiny files), skipping fstat saves ~5µs/file. Uses a two-tier buffer strategy: small stack buffer (4KB) for the initial read, then falls back to a larger stack buffer (64KB) or streaming hash for bigger files. For benchmark’s 55-byte files: one read() fills the 4KB buffer, hash immediately.

hash_files_parallel

Batch-hash multiple files with SHA-256/MD5 using std::thread::scope. Uses lightweight OS threads instead of rayon’s work-stealing pool. For single-invocation tools, this saves ~300µs of rayon thread pool initialization (spawning N-1 threads + setting up work-stealing deques). Returns results in input order.

hash_reader

Compute hash of data from a reader, returning hex string.

hash_stdin

Hash stdin. Uses fadvise for file redirects, streaming for pipes.

parse_check_line

Parse a checksum line in any supported format.

parse_check_line_tag

Parse a BSD-style tag line: “ALGO (filename) = hash” Returns (expected_hash, filename, optional_bits). bits is the hash length parsed from the algo name (e.g., BLAKE2b-256 -> Some(256)).

print_hash

Print hash result in GNU format: “hash filename\n” Uses raw byte writes to avoid std::fmt overhead.

print_hash_tag

Print hash result in BSD tag format: “ALGO (filename) = hash\n”

print_hash_tag_b2sum

Print hash in BSD tag format with BLAKE2b length info: “BLAKE2b (filename) = hash” for 512-bit, or “BLAKE2b-256 (filename) = hash” for other lengths.

print_hash_tag_b2sum_zero

Print hash in BSD tag format with BLAKE2b length info and NUL terminator.

print_hash_tag_zero

Print hash in BSD tag format with NUL terminator.

print_hash_zero

Print hash in GNU format with NUL terminator instead of newline.

readahead_files

Issue readahead hints for a list of file paths to warm the page cache. Uses POSIX_FADV_WILLNEED which is non-blocking and batches efficiently. Only issues hints for files >= 1MB; small files are read fast enough that the fadvise syscall overhead isn’t worth it.

readahead_files_all

Issue readahead hints for ALL file paths (no size threshold). For multi-file benchmarks, even small files benefit from batched readahead.

should_use_parallel

Check if parallel hashing is worthwhile for the given file paths. Always parallelize with 2+ files — rayon’s thread pool is lazily initialized once and reused, so per-file work-stealing overhead is negligible (~1µs). Removing the stat()-based size check eliminates N extra syscalls for N files.

write_hash_line

Build and write the standard GNU hash output line in a single write() call. Format: “hash filename\n” or “hash *filename\n” (binary mode). For escaped filenames: “\hash escaped_filename\n”.

write_hash_tag_line

Build and write BSD tag format output in a single write() call. Format: “ALGO (filename) = hash\n”