Sort: bench ahash

Cargo.toml

@@ -1,7 +1,7 @@
 # coreutils (uutils)
 # * see the repository LICENSE, README, and CONTRIBUTING files for more information
 
-# spell-checker:ignore (libs) bigdecimal datetime serde wincode gethostid kqueue libselinux mangen memmap uuhelp startswith constness expl unnested logind cfgs interner
+# spell-checker:ignore (libs) ahash bigdecimal datetime serde wincode gethostid kqueue libselinux mangen memmap uuhelp startswith constness expl unnested logind cfgs interner
 
 [package]
 name = "coreutils"
@@ -311,6 +311,7 @@ readme = "README.package.md"
 version = "0.6.0"
 
 [workspace.dependencies]
+ahash = "0.8.12"
 ansi-width = "0.1.0"
 bigdecimal = "0.4"
 binary-heap-plus = "0.5.0"
@@ -329,7 +330,6 @@ dns-lookup = { version = "3.0.0" }
 exacl = "0.12.0"
 file_diff = "1.0.0"
 filetime = "0.2.23"
-fnv = "1.0.7"
 fs_extra = "1.3.0"
 fts-sys = "0.2.16"
 gcd = "2.3"

src/uu/sort/Cargo.toml

@@ -1,4 +1,4 @@
-# spell-checker:ignore bigdecimal
+# spell-checker:ignore ahash bigdecimal
 
 [package]
 name = "uu_sort"
@@ -29,7 +29,6 @@ bigdecimal = { workspace = true }
 binary-heap-plus = { workspace = true }
 clap = { workspace = true }
 compare = { workspace = true }
-fnv = { workspace = true }
 itertools = { workspace = true }
 memchr = { workspace = true }
 rand = { workspace = true }
@@ -45,6 +44,7 @@ uucore = { workspace = true, features = [
   "i18n-collator",
 ] }
 fluent = { workspace = true }
+ahash = { workspace = true }
 
 [target.'cfg(not(target_os = "redox"))'.dependencies]
 ctrlc = { workspace = true }

src/uu/sort/src/sort.rs

@@ -18,22 +18,21 @@ mod merge;
 mod numeric_str_cmp;
 mod tmp_dir;
 
+use ahash::AHashMap;
 use bigdecimal::BigDecimal;
 use chunks::LineData;
 use clap::builder::ValueParser;
 use clap::{Arg, ArgAction, ArgMatches, Command};
 use custom_str_cmp::custom_str_cmp;
-
 use ext_sort::ext_sort;
-use fnv::FnvHasher;
 use numeric_str_cmp::{NumInfo, NumInfoParseSettings, human_numeric_str_cmp, numeric_str_cmp};
 use rand::{Rng, rng};
 use rayon::prelude::*;
 use std::cmp::Ordering;
 use std::env;
 use std::ffi::{OsStr, OsString};
 use std::fs::{File, OpenOptions};
-use std::hash::{Hash, Hasher};
+use std::hash::{BuildHasher, Hash, Hasher};
 use std::io::{BufRead, BufReader, BufWriter, Read, Write, stdin, stdout};
 use std::num::{IntErrorKind, NonZero};
 use std::ops::Range;
@@ -1681,7 +1680,7 @@ fn index_legacy_warnings(processed_args: &[OsString], legacy_warnings: &mut [Leg
         return;
     }
 
-    let mut index_by_arg = std::collections::HashMap::new();
+    let mut index_by_arg = AHashMap::default();
     for (warning_idx, warning) in legacy_warnings.iter().enumerate() {
         index_by_arg.insert(warning.arg_index, warning_idx);
     }
@@ -2909,7 +2908,8 @@ fn salt_from_random_source(path: &Path) -> UResult<[u8; SALT_LEN]> {
     let mut reader = open_with_open_failed_error(path)?;
     let mut buf = [0u8; BUF_LEN];
     let mut total = 0usize;
-    let mut hasher = FnvHasher::default();
+    // freeze seed for --random-source
+    let mut hasher = ahash::RandomState::with_seeds(1, 1, 1, 1).build_hasher();
 
     loop {
         let n = reader
@@ -2934,7 +2934,8 @@ fn salt_from_random_source(path: &Path) -> UResult<[u8; SALT_LEN]> {
     }
 
     let first = hasher.finish();
-    let mut second_hasher = FnvHasher::default();
+    // freeze seed for --random-source
+    let mut second_hasher = ahash::RandomState::with_seeds(2, 2, 2, 2).build_hasher();
     second_hasher.write(RANDOM_SOURCE_TAG);
     second_hasher.write_u64(first);
     let second = second_hasher.finish();
@@ -2946,9 +2947,8 @@ fn salt_from_random_source(path: &Path) -> UResult<[u8; SALT_LEN]> {
 }
 
 fn get_hash<T: Hash>(t: &T) -> u64 {
-    let mut s = FnvHasher::default();
-    t.hash(&mut s);
-    s.finish()
+    // Is reproducibility of get_hash itself needed for --random-source ?
+    ahash::RandomState::with_seeds(0, 0, 0, 0).hash_one(t)
 }
 
 fn random_shuffle(a: &[u8], b: &[u8], salt: &[u8]) -> Ordering {
@@ -3086,13 +3086,6 @@ mod tests {
         buffer
     }
 
-    #[test]
-    fn test_get_hash() {
-        let a = "Ted".to_string();
-
-        assert_eq!(2_646_829_031_758_483_623, get_hash(&a));
-    }
-
     #[test]
     fn test_random_shuffle() {
         let a = b"Ted";