fix(wc):GNU wc-cpu.sh (#9144)

* wc: align SIMD policy integration

- use SimdPolicy::detect with hardware feature labeling
- keep SIMD behavior respecting GLIBC_TUNABLES
- consolidate wc SIMD debug output and tests

* refactor(wc): simplify SIMD feature collection in debug output

- Changed multi-line SIMD feature vector creation to a single-line expression for improved readability and consistency with surrounding code.
- No functional changes; only stylistic refactoring in the wc debug logic.

* feat(wc): enhance debug output for SIMD hardware support limitations

Add new localization strings and logic to provide detailed debug information when SIMD support is limited by GLIBC_TUNABLES, including lists of disabled and enabled features. Refactor SIMD allowance check for better accuracy in detecting runtime support.

* refactor: consolidate SIMD feature handling in wc command

Refactor SIMD feature detection and reporting in the wc utility by introducing a WcSimdFeatures struct to group enabled, disabled, and runtime-disabled features. This replaces multiple separate functions with a single function, improving code organization and efficiency by reducing redundant iterations over feature lists. Also rename helper functions for clarity and update debug output logic accordingly.

* Update src/uu/wc/src/wc.rs

Co-authored-by: Dorian Péron <[email protected]>

* Update src/uu/wc/locales/en-US.ftl

Co-authored-by: Dorian Péron <[email protected]>

* Update src/uu/wc/locales/fr-FR.ftl

Co-authored-by: Dorian Péron <[email protected]>

* feat(wc): import show_error for enhanced error reporting

Add the show_error import from uucore to enable better error handling in the wc utility, allowing for consistent error messages in line with the project's style.

---------

Co-authored-by: Dorian Péron <[email protected]>

Author: mattsu2020

.vscode/cspell.dictionaries/jargon.wordlist.txt

@@ -1,4 +1,6 @@
 AFAICT
+asimd
+ASIMD
 alloc
 arity
 autogenerate
@@ -71,6 +73,7 @@ hardlink
 hardlinks
 hasher
 hashsums
+hwcaps
 infile
 iflag
 iflags
@@ -150,6 +153,8 @@ tokenize
 toolchain
 totalram
 truthy
+tunables
+TUNABLES
 ucase
 unbuffered
 udeps

src/uu/wc/Cargo.toml

@@ -18,16 +18,21 @@ workspace = true
 path = "src/wc.rs"
 
 [dependencies]
-clap = { workspace = true }
-uucore = { workspace = true, features = ["parser", "pipes", "quoting-style"] }
 bytecount = { workspace = true, features = ["runtime-dispatch-simd"] }
+clap = { workspace = true }
+fluent = { workspace = true }
 thiserror = { workspace = true }
+uucore = { workspace = true, features = [
+  "hardware",
+  "parser",
+  "pipes",
+  "quoting-style",
+] }
 unicode-width = { workspace = true }
-fluent = { workspace = true }
 
 [target.'cfg(unix)'.dependencies]
-nix = { workspace = true }
 libc = { workspace = true }
+nix = { workspace = true }
 
 [dev-dependencies]
 divan = { workspace = true }

src/uu/wc/locales/en-US.ftl

@@ -31,3 +31,10 @@ decoder-error-io = underlying bytestream error: { $error }
 # Other messages
 wc-standard-input = standard input
 wc-total = total
+
+# Debug messages
+wc-debug-hw-unavailable = debug: hardware support unavailable on this CPU
+wc-debug-hw-using = debug: using hardware support (features: { $features })
+wc-debug-hw-disabled-env = debug: hardware support disabled by environment
+wc-debug-hw-disabled-glibc = debug: hardware support disabled by GLIBC_TUNABLES ({ $features })
+wc-debug-hw-limited-glibc = debug: hardware support limited by GLIBC_TUNABLES (disabled: { $disabled }; enabled: { $enabled })

src/uu/wc/locales/fr-FR.ftl

@@ -31,3 +31,10 @@ decoder-error-io = erreur du flux d'octets sous-jacent : { $error }
 # Autres messages
 wc-standard-input = entrée standard
 wc-total = total
+
+# Messages de débogage
+wc-debug-hw-unavailable = debug : prise en charge matérielle indisponible sur ce CPU
+wc-debug-hw-using = debug : utilisation de l'accélération matérielle (fonctions : { $features })
+wc-debug-hw-disabled-env = debug : prise en charge matérielle désactivée par l'environnement
+wc-debug-hw-disabled-glibc = debug : prise en charge matérielle désactivée par GLIBC_TUNABLES ({ $features })
+wc-debug-hw-limited-glibc = debug : prise en charge matérielle limitée par GLIBC_TUNABLES (désactivé : { $disabled } ; activé : { $enabled })

src/uu/wc/src/count_fast.rs

@@ -4,7 +4,8 @@
 // file that was distributed with this source code.
 
 // cSpell:ignore sysconf
-use crate::word_count::WordCount;
+use crate::{wc_simd_allowed, word_count::WordCount};
+use uucore::hardware::SimdPolicy;
 
 use super::WordCountable;
 
@@ -232,6 +233,8 @@ pub(crate) fn count_bytes_chars_and_lines_fast<
 ) -> (WordCount, Option<io::Error>) {
     let mut total = WordCount::default();
     let buf: &mut [u8] = &mut AlignedBuffer::default().data;
+    let policy = SimdPolicy::detect();
+    let simd_allowed = wc_simd_allowed(policy);
     loop {
         match handle.read(buf) {
             Ok(0) => return (total, None),
@@ -240,10 +243,18 @@ pub(crate) fn count_bytes_chars_and_lines_fast<
                     total.bytes += n;
                 }
                 if COUNT_CHARS {
-                    total.chars += bytecount::num_chars(&buf[..n]);
+                    total.chars += if simd_allowed {
+                        bytecount::num_chars(&buf[..n])
+                    } else {
+                        bytecount::naive_num_chars(&buf[..n])
+                    };
                 }
                 if COUNT_LINES {
-                    total.lines += bytecount::count(&buf[..n], b'\n');
+                    total.lines += if simd_allowed {
+                        bytecount::count(&buf[..n], b'\n')
+                    } else {
+                        bytecount::naive_count(&buf[..n], b'\n')
+                    };
                 }
             }
             Err(ref e) if e.kind() == ErrorKind::Interrupted => (),

src/uu/wc/src/wc.rs

@@ -29,9 +29,10 @@ use uucore::translate;
 use uucore::{
     error::{FromIo, UError, UResult},
     format_usage,
+    hardware::{HardwareFeature, HasHardwareFeatures as _, SimdPolicy},
     parser::shortcut_value_parser::ShortcutValueParser,
     quoting_style::{self, QuotingStyle},
-    show,
+    show, show_error,
 };
 
 use crate::{
@@ -49,6 +50,7 @@ struct Settings<'a> {
     show_lines: bool,
     show_words: bool,
     show_max_line_length: bool,
+    debug: bool,
     files0_from: Option<Input<'a>>,
     total_when: TotalWhen,
 }
@@ -62,6 +64,7 @@ impl Default for Settings<'_> {
             show_lines: true,
             show_words: true,
             show_max_line_length: false,
+            debug: false,
             files0_from: None,
             total_when: TotalWhen::default(),
         }
@@ -85,6 +88,7 @@ impl<'a> Settings<'a> {
             show_lines: matches.get_flag(options::LINES),
             show_words: matches.get_flag(options::WORDS),
             show_max_line_length: matches.get_flag(options::MAX_LINE_LENGTH),
+            debug: matches.get_flag(options::DEBUG),
             files0_from,
             total_when,
         };
@@ -95,6 +99,7 @@ impl<'a> Settings<'a> {
             Self {
                 files0_from: settings.files0_from,
                 total_when,
+                debug: settings.debug,
                 ..Default::default()
             }
         }
@@ -122,6 +127,7 @@ mod options {
     pub static MAX_LINE_LENGTH: &str = "max-line-length";
     pub static TOTAL: &str = "total";
     pub static WORDS: &str = "words";
+    pub static DEBUG: &str = "debug";
 }
 static ARG_FILES: &str = "files";
 static STDIN_REPR: &str = "-";
@@ -445,6 +451,12 @@ pub fn uu_app() -> Command {
                 .help(translate!("wc-help-words"))
                 .action(ArgAction::SetTrue),
         )
+        .arg(
+            Arg::new(options::DEBUG)
+                .long(options::DEBUG)
+                .action(ArgAction::SetTrue)
+                .hide(true),
+        )
         .arg(
             Arg::new(ARG_FILES)
                 .action(ArgAction::Append)
@@ -805,6 +817,74 @@ fn escape_name_wrapper(name: &OsStr) -> String {
         .expect("All escaped names with the escaping option return valid strings.")
 }
 
+fn hardware_feature_label(feature: HardwareFeature) -> &'static str {
+    match feature {
+        HardwareFeature::Avx512 => "AVX512F",
+        HardwareFeature::Avx2 => "AVX2",
+        HardwareFeature::PclMul => "PCLMUL",
+        HardwareFeature::Vmull => "VMULL",
+        HardwareFeature::Sse2 => "SSE2",
+        HardwareFeature::Asimd => "ASIMD",
+    }
+}
+
+fn is_simd_runtime_feature(feature: &HardwareFeature) -> bool {
+    matches!(
+        feature,
+        HardwareFeature::Avx2 | HardwareFeature::Sse2 | HardwareFeature::Asimd
+    )
+}
+
+fn is_simd_debug_feature(feature: &HardwareFeature) -> bool {
+    matches!(
+        feature,
+        HardwareFeature::Avx512
+            | HardwareFeature::Avx2
+            | HardwareFeature::Sse2
+            | HardwareFeature::Asimd
+    )
+}
+
+struct WcSimdFeatures {
+    enabled: Vec<HardwareFeature>,
+    disabled: Vec<HardwareFeature>,
+    disabled_runtime: Vec<HardwareFeature>,
+}
+
+fn wc_simd_features(policy: &SimdPolicy) -> WcSimdFeatures {
+    let enabled = policy
+        .iter_features()
+        .filter(is_simd_runtime_feature)
+        .collect();
+
+    let mut disabled = Vec::new();
+    let mut disabled_runtime = Vec::new();
+    for feature in policy.disabled_features() {
+        if is_simd_debug_feature(&feature) {
+            disabled.push(feature);
+        }
+        if is_simd_runtime_feature(&feature) {
+            disabled_runtime.push(feature);
+        }
+    }
+
+    WcSimdFeatures {
+        enabled,
+        disabled,
+        disabled_runtime,
+    }
+}
+
+pub(crate) fn wc_simd_allowed(policy: &SimdPolicy) -> bool {
+    let disabled_features = policy.disabled_features();
+    if disabled_features.iter().any(is_simd_runtime_feature) {
+        return false;
+    }
+    policy
+        .iter_features()
+        .any(|feature| is_simd_runtime_feature(&feature))
+}
+
 fn wc(inputs: &Inputs, settings: &Settings) -> UResult<()> {
     let mut total_word_count = WordCount::default();
     let mut num_inputs: usize = 0;
@@ -814,6 +894,51 @@ fn wc(inputs: &Inputs, settings: &Settings) -> UResult<()> {
         _ => (compute_number_width(inputs, settings), true),
     };
 
+    if settings.debug {
+        let policy = SimdPolicy::detect();
+        let features = wc_simd_features(policy);
+
+        let enabled: Vec<&'static str> = features
+            .enabled
+            .iter()
+            .copied()
+            .map(hardware_feature_label)
+            .collect();
+        let disabled: Vec<&'static str> = features
+            .disabled
+            .iter()
+            .copied()
+            .map(hardware_feature_label)
+            .collect();
+
+        let enabled_empty = enabled.is_empty();
+        let disabled_empty = disabled.is_empty();
+        let runtime_disabled = !features.disabled_runtime.is_empty();
+
+        if enabled_empty && !runtime_disabled {
+            show_error!("{}", translate!("wc-debug-hw-unavailable"));
+        } else if runtime_disabled {
+            show_error!(
+                "{}",
+                translate!("wc-debug-hw-disabled-glibc", "features" => disabled.join(", "))
+            );
+        } else if !enabled_empty && disabled_empty {
+            show_error!(
+                "{}",
+                translate!("wc-debug-hw-using", "features" => enabled.join(", "))
+            );
+        } else {
+            show_error!(
+                "{}",
+                translate!(
+                    "wc-debug-hw-limited-glibc",
+                    "disabled" => disabled.join(", "),
+                    "enabled" => enabled.join(", ")
+                )
+            );
+        }
+    }
+
     for maybe_input in inputs.try_iter(settings)? {
         num_inputs += 1;
 

src/uucore/src/lib/features/hardware.rs

@@ -214,8 +214,9 @@ impl SimdPolicy {
         }
     }
 
+    /// Returns true if any SIMD feature remains enabled after applying GLIBC_TUNABLES.
     pub fn allows_simd(&self) -> bool {
-        self.disabled_by_env.is_empty()
+        self.iter_features().next().is_some()
     }
 
     pub fn disabled_features(&self) -> Vec<HardwareFeature> {

tests/by-util/test_wc.rs

@@ -808,3 +808,69 @@ fn wc_w_words_with_emoji_separator() {
         .succeeds()
         .stdout_contains("3");
 }
+
+#[cfg(unix)]
+#[test]
+fn test_simd_respects_glibc_tunables() {
+    // Ensure debug output reflects that SIMD paths are disabled via GLIBC_TUNABLES
+    let debug_output = new_ucmd!()
+        .args(&["-l", "--debug", "/dev/null"])
+        .env("GLIBC_TUNABLES", "glibc.cpu.hwcaps=-AVX2,-AVX512F")
+        .succeeds()
+        .stderr_str()
+        .to_string();
+    assert!(
+        !debug_output.contains("using hardware support"),
+        "SIMD should be reported as disabled when GLIBC_TUNABLES blocks AVX features: {debug_output}"
+    );
+    assert!(
+        debug_output.contains("hardware support disabled"),
+        "Debug output should acknowledge GLIBC_TUNABLES restrictions: {debug_output}"
+    );
+
+    // WC results should be identical with and without GLIBC_TUNABLES overrides
+    let sample_sizes = [0usize, 1, 7, 128, 513, 999];
+    use std::fmt::Write as _;
+    for &lines in &sample_sizes {
+        let content: String = (0..lines).fold(String::new(), |mut acc, i| {
+            // Build the input buffer efficiently without allocating per line.
+            let _ = writeln!(acc, "{i}");
+            acc
+        });
+
+        let base = new_ucmd!()
+            .arg("-l")
+            .pipe_in(content.clone())
+            .succeeds()
+            .stdout_str()
+            .trim()
+            .to_string();
+
+        let no_avx512 = new_ucmd!()
+            .arg("-l")
+            .env("GLIBC_TUNABLES", "glibc.cpu.hwcaps=-AVX512F")
+            .pipe_in(content.clone())
+            .succeeds()
+            .stdout_str()
+            .trim()
+            .to_string();
+
+        let no_avx2_avx512 = new_ucmd!()
+            .arg("-l")
+            .env("GLIBC_TUNABLES", "glibc.cpu.hwcaps=-AVX2,-AVX512F")
+            .pipe_in(content)
+            .succeeds()
+            .stdout_str()
+            .trim()
+            .to_string();
+
+        assert_eq!(
+            base, no_avx512,
+            "Line counts should not change when AVX512 is disabled (lines={lines})"
+        );
+        assert_eq!(
+            base, no_avx2_avx512,
+            "Line counts should not change when AVX2/AVX512 are disabled (lines={lines})"
+        );
+    }
+}