diff --git a/Cargo.lock b/Cargo.lock
index f47ac348..eeab52a7 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -1522,6 +1522,7 @@ dependencies = [
  "tempfile",
  "textwrap",
  "uu_free",
+ "uu_hugetop",
  "uu_pgrep",
  "uu_pidof",
  "uu_pidwait",
@@ -2344,6 +2345,15 @@ dependencies = [
  "windows",
 ]
 
+[[package]]
+name = "uu_hugetop"
+version = "0.0.1"
+dependencies = [
+ "clap",
+ "tempfile",
+ "uucore",
+]
+
 [[package]]
 name = "uu_pgrep"
 version = "0.0.1"
diff --git a/Cargo.toml b/Cargo.toml
index 4a02eea0..33948491 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,3 +1,5 @@
+[workspace]
+members = ["src/uu/hugetop"]
 # procps (uutils)
 # * see the repository LICENSE, README, and CONTRIBUTING files for more information
 
@@ -33,6 +35,7 @@ uudoc = []
 
 feat_common_core = [
     "free",
+    "hugetop",
     "pgrep",
     "pidof",
     "pidwait",
@@ -94,6 +97,7 @@ uucore = { workspace = true }
 
 #
 free = { optional = true, version = "0.0.1", package = "uu_free", path = "src/uu/free" }
+hugetop = { optional = true, version = "0.0.1", package = "uu_hugetop", path = "src/uu/hugetop" }
 pgrep = { optional = true, version = "0.0.1", package = "uu_pgrep", path = "src/uu/pgrep" }
 pidof = { optional = true, version = "0.0.1", package = "uu_pidof", path = "src/uu/pidof" }
 pidwait = { optional = true, version = "0.0.1", package = "uu_pidwait", path = "src/uu/pidwait" }
diff --git a/README.md b/README.md
index 2a1d0bf9..37580ee4 100644
--- a/README.md
+++ b/README.md
@@ -13,6 +13,7 @@ Provides command line and full screen utilities for browsing procfs, a "pseudo"
 
 Ongoing:
 * `free`: Shows the amount of free and used memory in the system.
+* `hugetop`: Report hugepage usage of processes and the system as a whole.
 * `pgrep`: Searches for processes based on name and other attributes.
 * `pidof`: Find the process ID of a running program.
 * `pidwait`: Waits for a specific process to terminate.
@@ -30,9 +31,6 @@ Ongoing:
 * `w`: Shows who is logged on and what they are doing.
 * `watch`: Executes a program periodically, showing output fullscreen.
 
-TODO:
-* `hugetop`: Report hugepage usage of processes and the system as a whole.
-
 Elsewhere:
 
  * `kill` is already implemented in https://github.com/uutils/coreutils
diff --git a/src/uu/hugetop/Cargo.toml b/src/uu/hugetop/Cargo.toml
new file mode 100644
index 00000000..aa0eea47
--- /dev/null
+++ b/src/uu/hugetop/Cargo.toml
@@ -0,0 +1,28 @@
+[package]
+name = "uu_hugetop"
+description = "hugetop ~ (uutils) Report hugepage usage for the system and processes"
+repository = "https://github.com/uutils/procps/tree/main/src/uu/hugetop"
+authors.workspace = true
+categories.workspace = true
+edition.workspace = true
+homepage.workspace = true
+keywords.workspace = true
+license.workspace = true
+version.workspace = true
+
+[lints]
+workspace = true
+
+[dependencies]
+clap = { workspace = true }
+uucore = { workspace = true }
+
+[dev-dependencies]
+tempfile = { workspace = true }
+
+[lib]
+path = "src/hugetop.rs"
+
+[[bin]]
+name = "hugetop"
+path = "src/main.rs"
diff --git a/src/uu/hugetop/src/hugetop.rs b/src/uu/hugetop/src/hugetop.rs
new file mode 100644
index 00000000..0edde723
--- /dev/null
+++ b/src/uu/hugetop/src/hugetop.rs
@@ -0,0 +1,456 @@
+// This file is part of the uutils procps package.
+//
+// For the full copyright and license information, please view the LICENSE
+// file that was distributed with this source code.
+
+use clap::{value_parser, Arg, Command};
+use std::collections::BTreeMap;
+use std::fs;
+use std::path::Path;
+use std::thread::sleep;
+use std::time::{Duration, SystemTime, UNIX_EPOCH};
+use uucore::error::UResult;
+
+const DEFAULT_HUGEPAGES_ROOT: &str = "/sys/kernel/mm/hugepages";
+const SYS_NODES_ROOT: &str = "/sys/devices/system/node";
+const DEFAULT_PROC_ROOT: &str = "/proc";
+
+#[derive(Debug, Clone, PartialEq, Eq)]
+struct HugePagePool {
+    size_kb: u64,
+    total_pages: u64,
+    free_pages: u64,
+    reserved_pages: u64,
+    surplus_pages: u64,
+}
+
+#[derive(Debug, Clone, PartialEq, Eq)]
+struct ProcessHugeUsage {
+    pid: u32,
+    command: String,
+    anon_huge_kb: u64,
+    shared_hugetlb_kb: u64,
+    private_hugetlb_kb: u64,
+}
+
+impl ProcessHugeUsage {
+    fn total_kb(&self) -> u64 {
+        self.anon_huge_kb + self.shared_hugetlb_kb + self.private_hugetlb_kb
+    }
+}
+
+#[uucore::main]
+pub fn uumain(args: impl uucore::Args) -> UResult<()> {
+    let matches = uu_app().try_get_matches_from(args)?;
+
+    let limit = matches.get_one::<usize>("lines").copied();
+    let numa = matches.get_flag("numa");
+    let human = matches.get_flag("human");
+    let once = matches.get_flag("once");
+    let delay = *matches.get_one::<u64>("delay").unwrap_or(&0);
+
+    if once || delay == 0 {
+        run(numa, human, limit)
+    } else {
+        loop {
+            // Clear the terminal to roughly approximate hugetop's screen refresh behavior.
+            print!("\x1B[2J\x1B[H");
+            run(numa, human, limit)?;
+            sleep(Duration::from_secs(delay));
+        }
+    }
+}
+
+fn run(numa: bool, human: bool, limit: Option<usize>) -> UResult<()> {
+    print_summary(numa, human);
+    print_headings();
+    print_procs(human, limit);
+    Ok(())
+}
+
+pub fn uu_app() -> Command {
+    Command::new(uucore::util_name())
+        .about("Report hugepage usage of processes and the system as a whole")
+        .arg(
+            Arg::new("delay")
+                .short('d')
+                .long("delay")
+                .value_name("SECONDS")
+                .help("Delay between updates (0 = run once)")
+                .value_parser(value_parser!(u64)),
+        )
+        .arg(
+            Arg::new("numa")
+                .short('n')
+                .long("numa")
+                .help("Display per NUMA node huge page information")
+                .action(clap::ArgAction::SetTrue),
+        )
+        .arg(
+            Arg::new("once")
+                .short('o')
+                .long("once")
+                .help("Only display once, then exit")
+                .action(clap::ArgAction::SetTrue),
+        )
+        .arg(
+            Arg::new("human")
+                .short('H')
+                .long("human")
+                .help("Display human-readable output")
+                .action(clap::ArgAction::SetTrue),
+        )
+        .arg(
+            Arg::new("lines")
+                .short('l')
+                .long("lines")
+                .value_name("N")
+                .help("Show the top N processes")
+                .value_parser(value_parser!(usize)),
+        )
+}
+
+fn print_summary(numa: bool, human: bool) {
+    let now = SystemTime::now()
+        .duration_since(UNIX_EPOCH)
+        .map(|d| d.as_secs())
+        .unwrap_or_default();
+
+    println!("hugetop - {}", now);
+
+    let pools = match read_node_hugepage_pools() {
+        Ok(nodes) if numa => {
+            for (node, pools) in &nodes {
+                print_node(node, pools, human);
+            }
+            return;
+        }
+        Ok(nodes) => merge_node_pools(&nodes),
+        Err(_) => Vec::new(),
+    };
+
+    if pools.is_empty() {
+        let pools = read_hugepage_pools(Path::new(DEFAULT_HUGEPAGES_ROOT)).unwrap_or_default();
+        if pools.is_empty() {
+            println!("(no hugepage pools found)");
+            return;
+        }
+        print_node("node(s)", &pools, human);
+    } else {
+        print_node("node(s)", &pools, human);
+    }
+}
+
+fn print_headings() {
+    println!("{:>8} {:>10} {:>10} COMMAND", "PID", "SHARED", "PRIVATE");
+}
+
+fn print_procs(human: bool, limit: Option<usize>) {
+    let mut processes = match read_process_hugepage_usage(Path::new(DEFAULT_PROC_ROOT)) {
+        Ok(p) => p,
+        Err(_) => Vec::new(),
+    };
+
+    processes.sort_by_key(|usage| std::cmp::Reverse(usage.total_kb()));
+
+    let mut shown = 0usize;
+    let limit = limit.unwrap_or(processes.len());
+
+    for usage in processes {
+        if shown >= limit {
+            break;
+        }
+
+        let shared = format_kb(usage.shared_hugetlb_kb, human);
+        let private = format_kb(usage.private_hugetlb_kb, human);
+
+        println!(
+            "{:>8} {:>10} {:>10} {}",
+            usage.pid, shared, private, usage.command
+        );
+
+        shown += 1;
+    }
+}
+
+fn print_node(node: &str, pools: &[HugePagePool], human: bool) {
+    let mut line = String::new();
+    line.push_str(node);
+    line.push_str(":");
+
+    for (i, pool) in pools.iter().enumerate() {
+        if i > 0 {
+            line.push_str(",");
+        }
+
+        let size = if human {
+            humanized(pool.size_kb, false)
+        } else {
+            format!("{}kB", pool.size_kb)
+        };
+
+        line.push_str(&format!(" {} - {}/{}", size, pool.free_pages, pool.total_pages));
+    }
+
+    println!("{}", line);
+}
+
+fn format_kb(kb: u64, human: bool) -> String {
+    if human {
+        humanized(kb, false)
+    } else {
+        format!("{}", kb)
+    }
+}
+
+fn humanized(kib: u64, si: bool) -> String {
+    let b = kib * 1024;
+    let units = ['B', 'K', 'M', 'G', 'T', 'P'];
+    let mut level = 0;
+    let mut divisor = 1u64;
+
+    while level < units.len() - 1 && divisor * 100 <= b {
+        divisor *= if si { 1000 } else { 1024 };
+        level += 1;
+    }
+
+    if level == 0 {
+        return format!("{}{}", b, units[level]);
+    }
+
+    let value = (b as f64) / (divisor as f64);
+    let formatted_value = if (value * 10.0).round() < 100.0 {
+        format!("{:.1}", (value * 10.0).round() / 10.0)
+    } else {
+        (value as u64).to_string()
+    };
+
+    format!(
+        "{}{}{}",
+        formatted_value,
+        units[level].to_owned(),
+        if si { "" } else { "i" }
+    )
+}
+
+fn read_node_hugepage_pools() -> UResult<Vec<(String, Vec<HugePagePool>)>> {
+    let mut nodes = Vec::new();
+    let Ok(entries) = fs::read_dir(SYS_NODES_ROOT) else {
+        return Ok(nodes);
+    };
+
+    for entry in entries.flatten() {
+        let file_name = entry.file_name();
+        let name = match file_name.to_str() {
+            Some(n) if n.starts_with("node") => n.to_string(),
+            _ => continue,
+        };
+
+        let path = entry.path().join("hugepages");
+        if !path.is_dir() {
+            continue;
+        }
+
+        let pools = read_hugepage_pools(&path)?;
+        if pools.is_empty() {
+            continue;
+        }
+
+        nodes.push((name, pools));
+    }
+
+    Ok(nodes)
+}
+
+fn merge_node_pools(nodes: &[(String, Vec<HugePagePool>)]) -> Vec<HugePagePool> {
+    let mut map: BTreeMap<u64, HugePagePool> = BTreeMap::new();
+
+    for (_, pools) in nodes {
+        for pool in pools {
+            let entry = map.entry(pool.size_kb).or_insert_with(|| HugePagePool {
+                size_kb: pool.size_kb,
+                total_pages: 0,
+                free_pages: 0,
+                reserved_pages: 0,
+                surplus_pages: 0,
+            });
+            entry.total_pages += pool.total_pages;
+            entry.free_pages += pool.free_pages;
+            entry.reserved_pages += pool.reserved_pages;
+            entry.surplus_pages += pool.surplus_pages;
+        }
+    }
+
+    map.into_values().collect()
+}
+
+fn read_hugepage_pools(root: &Path) -> UResult<Vec<HugePagePool>> {
+    let mut pools = Vec::new();
+
+    let Ok(entries) = fs::read_dir(root) else {
+        return Ok(pools);
+    };
+
+    for entry in entries.flatten() {
+        let path = entry.path();
+        let Some(name) = path.file_name().and_then(|s| s.to_str()) else {
+            continue;
+        };
+
+        let Some(size_kb) = parse_hugepage_dir_name(name) else {
+            continue;
+        };
+
+        let total_pages = read_u64(path.join("nr_hugepages"));
+        let free_pages = read_u64(path.join("free_hugepages"));
+        let reserved_pages = read_u64(path.join("resv_hugepages"));
+        let surplus_pages = read_u64(path.join("surplus_hugepages"));
+
+        pools.push(HugePagePool {
+            size_kb,
+            total_pages,
+            free_pages,
+            reserved_pages,
+            surplus_pages,
+        });
+    }
+
+    pools.sort_by_key(|pool| pool.size_kb);
+    Ok(pools)
+}
+
+fn read_process_hugepage_usage(root: &Path) -> UResult<Vec<ProcessHugeUsage>> {
+    let mut usages = Vec::new();
+
+    let Ok(entries) = fs::read_dir(root) else {
+        return Ok(usages);
+    };
+
+    for entry in entries.flatten() {
+        let path = entry.path();
+        let Some(file_name) = path.file_name().and_then(|name| name.to_str()) else {
+            continue;
+        };
+
+        let Ok(pid) = file_name.parse::<u32>() else {
+            continue;
+        };
+
+        let Some((_, shared_hugetlb_kb, private_hugetlb_kb)) =
+            parse_smaps_rollup(&path.join("smaps_rollup"))
+        else {
+            continue;
+        };
+
+        let total_kb = shared_hugetlb_kb + private_hugetlb_kb;
+        if total_kb == 0 {
+            continue;
+        }
+
+        let command = fs::read_to_string(path.join("comm"))
+            .unwrap_or_else(|_| String::from("?"))
+            .trim()
+            .to_string();
+
+        usages.push(ProcessHugeUsage {
+            pid,
+            command,
+            anon_huge_kb: 0,
+            shared_hugetlb_kb,
+            private_hugetlb_kb,
+        });
+    }
+
+    Ok(usages)
+}
+
+fn parse_hugepage_dir_name(name: &str) -> Option<u64> {
+    let prefix = "hugepages-";
+    let suffix = "kB";
+
+    if !name.starts_with(prefix) || !name.ends_with(suffix) {
+        return None;
+    }
+
+    name[prefix.len()..name.len() - suffix.len()]
+        .parse::<u64>()
+        .ok()
+}
+
+fn parse_smaps_rollup(path: &Path) -> Option<(u64, u64, u64)> {
+    let content = fs::read_to_string(path).ok()?;
+
+    let mut anon_huge_kb = 0;
+    let mut shared_hugetlb_kb = 0;
+    let mut private_hugetlb_kb = 0;
+
+    for line in content.lines() {
+        if let Some(value) = parse_kb_field(line, "AnonHugePages:") {
+            anon_huge_kb = value;
+        } else if let Some(value) = parse_kb_field(line, "Shared_Hugetlb:") {
+            shared_hugetlb_kb = value;
+        } else if let Some(value) = parse_kb_field(line, "Private_Hugetlb:") {
+            private_hugetlb_kb = value;
+        }
+    }
+
+    Some((anon_huge_kb, shared_hugetlb_kb, private_hugetlb_kb))
+}
+
+fn parse_kb_field(line: &str, field: &str) -> Option<u64> {
+    let value = line.strip_prefix(field)?.trim();
+    let number = value.split_whitespace().next()?;
+    number.parse::<u64>().ok()
+}
+
+fn read_u64(path: impl AsRef<Path>) -> u64 {
+    fs::read_to_string(path)
+        .ok()
+        .and_then(|s| s.trim().parse::<u64>().ok())
+        .unwrap_or(0)
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use std::io::Write;
+
+    #[test]
+    fn parse_hugepage_name_works() {
+        assert_eq!(parse_hugepage_dir_name("hugepages-2048kB"), Some(2048));
+        assert_eq!(
+            parse_hugepage_dir_name("hugepages-1048576kB"),
+            Some(1_048_576)
+        );
+        assert_eq!(parse_hugepage_dir_name("hugepages-foo"), None);
+    }
+
+    #[test]
+    fn parse_smaps_rollup_works() {
+        let dir = tempfile::tempdir().unwrap();
+        let file = dir.path().join("smaps_rollup");
+        let mut f = fs::File::create(&file).unwrap();
+        writeln!(f, "AnonHugePages:      512 kB").unwrap();
+        writeln!(f, "Shared_Hugetlb:      64 kB").unwrap();
+        writeln!(f, "Private_Hugetlb:     32 kB").unwrap();
+
+        assert_eq!(parse_smaps_rollup(&file), Some((512, 64, 32)));
+    }
+
+    #[test]
+    fn reads_pools_from_tree() {
+        let dir = tempfile::tempdir().unwrap();
+        let pool = dir.path().join("hugepages-2048kB");
+        fs::create_dir(&pool).unwrap();
+        fs::write(pool.join("nr_hugepages"), "10\n").unwrap();
+        fs::write(pool.join("free_hugepages"), "3\n").unwrap();
+        fs::write(pool.join("resv_hugepages"), "2\n").unwrap();
+        fs::write(pool.join("surplus_hugepages"), "1\n").unwrap();
+
+        let pools = read_hugepage_pools(dir.path()).unwrap();
+        assert_eq!(pools.len(), 1);
+        assert_eq!(pools[0].size_kb, 2048);
+        assert_eq!(pools[0].total_pages, 10);
+        assert_eq!(pools[0].free_pages, 3);
+    }
+}
diff --git a/src/uu/hugetop/src/main.rs b/src/uu/hugetop/src/main.rs
new file mode 100644
index 00000000..1e6f1470
--- /dev/null
+++ b/src/uu/hugetop/src/main.rs
@@ -0,0 +1 @@
+uucore::bin!(uu_hugetop);
diff --git a/tests/by-util/test_hugetop.rs b/tests/by-util/test_hugetop.rs
new file mode 100644
index 00000000..e78957e1
--- /dev/null
+++ b/tests/by-util/test_hugetop.rs
@@ -0,0 +1,21 @@
+// This file is part of the uutils procps package.
+//
+// For the full copyright and license information, please view the LICENSE
+// file that was distributed with this source code.
+
+use uutests::new_ucmd;
+
+#[test]
+fn runs_successfully() {
+    new_ucmd!().succeeds();
+}
+
+#[test]
+fn supports_lines_option() {
+    new_ucmd!().arg("-l").arg("1").succeeds();
+}
+
+#[test]
+fn supports_numa_option() {
+    new_ucmd!().arg("-n").succeeds();
+}
diff --git a/tests/tests.rs b/tests/tests.rs
index 8df70274..a27a4294 100644
--- a/tests/tests.rs
+++ b/tests/tests.rs
@@ -23,6 +23,10 @@ mod test_pwdx;
 #[path = "by-util/test_free.rs"]
 mod test_free;
 
+#[cfg(feature = "hugetop")]
+#[path = "by-util/test_hugetop.rs"]
+mod test_hugetop;
+
 #[cfg(feature = "w")]
 #[path = "by-util/test_w.rs"]
 mod test_w;