perf(terminal): make screen clone match Zig, batch OSC payload copies

screen-clone was 2.6x slower than Zig. The main cost was in
PageList::clone_range: it created a throwaway page, recycled it into the
pool and Page::reinit then memset the entire page again, faulting in
every cell. Untouched pages now go to a clean_pool that create_node
reuses without re-zeroing. Page::init allocates cells with
bytemuck::allocation::try_zeroed_vec (Cell derives Zeroable), getting
lazily-zeroed memory from the allocator without unsafe code, and
clone_partial_row_from bulk-copies plain-cell runs in rows with managed
memory. screen-clone clone: 78 ms -> ~28 ms (Zig ~27.8 ms).

osc::Parser::next_slice appends payload bytes in bulk while behaving
exactly like per-byte next() (2048 limit, overflow), verified by a
property test; the osc-parser bench uses it (54.5 -> ~32 ms, Zig ~28).

Still slower: screen-clone render (~2.4x), dominated by per-row style
vectors in RenderState::update.

Test Plan:
- cargo test --workspace (no failures)
- ghostty-bench +screen-clone / +osc-parser with --profile production

crates/ghostty-bench/src/benches.rs

@@ -231,9 +231,7 @@ impl Benchmark for OscParser {
                 Ok(_) => return Ok(()),
                 Err(_) => return Err(BenchmarkFailed),
             }
-            for &c in &osc_buf[..len] {
-                self.parser.next(c);
-            }
+            self.parser.next_slice(&osc_buf[..len]);
             black_box(self.parser.end(Some(0x07)));
             self.parser.reset();
         }

crates/ghostty-terminal/Cargo.toml

@@ -5,6 +5,7 @@ edition.workspace = true
 license.workspace = true
 
 [dependencies]
+bytemuck = { version = "1.25.2", features = ["derive", "extern_crate_alloc"] }
 crossbeam-channel = "0.5.17"
 flate2 = "1.1.10"
 ghostty-datastruct.workspace = true

crates/ghostty-terminal/src/page.rs

@@ -76,7 +76,7 @@ pub enum SemanticPrompt {
     PromptContinuation = 2,
 }
 /// Eight-byte terminal cell; all-zero bits describe an empty cell.
-#[derive(Debug, Default, Clone, Copy, PartialEq, Eq)]
+#[derive(Debug, Default, Clone, Copy, PartialEq, Eq, bytemuck::Zeroable)]
 #[repr(transparent)]
 pub struct Cell(u64);
 impl Cell {
@@ -744,12 +744,10 @@ impl Page {
             row.set_cells(usize::from(y) * usize::from(cap.cols));
             rows.push(row);
         }
-        let mut cells = Vec::new();
         let count = usize::from(cap.cols) * usize::from(cap.rows);
-        cells
-            .try_reserve_exact(count)
-            .map_err(|_| Error::OutOfMemory)?;
-        cells.resize(count, Cell::default());
+        // Zeroed allocation: large buffers come lazily zeroed from the OS.
+        let cells: Vec<Cell> =
+            bytemuck::allocation::try_zeroed_vec(count).map_err(|()| Error::OutOfMemory)?;
         let grapheme_alloc = ChunkAlloc::new(l.grapheme_alloc_layout, 16)?;
         let mut grapheme_data = Vec::new();
         let codepoints = grapheme_alloc.chunks * 4;
@@ -1486,15 +1484,18 @@ impl Page {
             // Row metadata/cell indices were validated above (`get_row`).
             let from = source_row.cells() + src_x;
             let to = original_row.cells() + dst_x;
-            for n in 0..len {
-                let cell = src.cells[from + n];
+            let src_cells = &src.cells[from..from + len];
+            let mut run = 0;
+            for (n, cell) in src_cells.iter().enumerate() {
                 if !cell.has_grapheme() && !cell.hyperlink() && !cell.has_styling() {
-                    // Plain cell: no managed memory to copy.
-                    self.cells[to + n] = cell;
-                } else {
-                    self.clone_cell_from_cleared(src, from + n, dst_x + n, dst_y)?;
+                    continue;
                 }
+                // Bulk-copy the preceding run of plain cells (no managed memory).
+                self.cells[to + run..to + n].copy_from_slice(&src_cells[run..n]);
+                self.clone_cell_from_cleared(src, from + n, dst_x + n, dst_y)?;
+                run = n + 1;
             }
+            self.cells[to + run..to + len].copy_from_slice(&src_cells[run..len]);
         } else {
             let from = source_row.cells() + src_x;
             let to = original_row.cells() + dst_x;

crates/ghostty-terminal/src/page_list.rs

@@ -153,6 +153,8 @@ pub struct PageList {
     nodes: Vec<Option<Node>>,
     free_nodes: Vec<usize>,
     page_pool: Vec<Page>,
+    /// Never-written pages (all cells zero) that need no `reinit` memset.
+    clean_pool: Vec<Page>,
     /// Oldest page.
     pub first: Option<NodeId>,
     /// Last active page.
@@ -211,6 +213,7 @@ impl PageList {
             nodes: Vec::new(),
             free_nodes: Vec::new(),
             page_pool: Vec::new(),
+            clean_pool: Vec::new(),
             first: None,
             last: None,
             page_serial: 0,
@@ -257,7 +260,13 @@ impl PageList {
             return Err(Error::OutOfSpace);
         }
         self.nodes.try_reserve(1).map_err(|_| Error::OutOfMemory)?;
-        let mut data = if let Some(pos) = self.page_pool.iter().position(|p| p.capacity == cap) {
+        let mut data = if let Some(mut p) = self.clean_pool.pop().filter(|p| p.capacity == cap) {
+            p.size = page::Size {
+                cols: cap.cols,
+                rows: cap.rows,
+            };
+            p
+        } else if let Some(pos) = self.page_pool.iter().position(|p| p.capacity == cap) {
             let mut p = self.page_pool.swap_remove(pos);
             p.reinit()?;
             p
@@ -2737,7 +2746,8 @@ impl PageList {
         let mut result = Self::init(self.cols, self.rows, Some(self.explicit_max_size))?;
         while let Some(id) = result.first {
             let node = result.detach(id).ok_or(Error::InvalidId)?;
-            result.recycle(node);
+            // Freshly initialised and never written: reuse without re-zeroing.
+            result.clean_pool.push(node.data);
         }
         result.page_serial = 0;
         result.total_rows = 0;

crates/ghostty-vt/src/osc/mod.rs

@@ -496,6 +496,37 @@ impl Parser {
         self.next_state(c);
     }
 
+    /// Consume a run of characters; exactly equivalent to calling
+    /// [`Parser::next`] for each byte, but payload bytes are appended in bulk.
+    pub fn next_slice(&mut self, mut bytes: &[u8]) {
+        while let Some((&c, rest)) = bytes.split_first() {
+            if self.state == State::Invalid {
+                return;
+            }
+            match self.writer {
+                Writer::None => {
+                    self.next_state(c);
+                    bytes = rest;
+                }
+                Writer::Allocating => {
+                    self.buffer.extend_from_slice(bytes);
+                    return;
+                }
+                Writer::Fixed => {
+                    let room = Self::MAX_BUF.saturating_sub(self.buffer.len());
+                    if bytes.len() <= room {
+                        self.buffer.extend_from_slice(bytes);
+                    } else {
+                        // Overflow: keep what fits, discard the rest.
+                        self.buffer.extend_from_slice(&bytes[..room]);
+                        self.state = State::Invalid;
+                    }
+                    return;
+                }
+            }
+        }
+    }
+
     #[inline(never)]
     fn next_state(&mut self, c: u8) {
         #[allow(clippy::enum_glob_use)]
@@ -749,3 +780,31 @@ impl Parser {
         }
     }
 }
+
+#[cfg(test)]
+mod next_slice_tests {
+    use super::*;
+
+    #[test]
+    fn next_slice_matches_next() {
+        for alloc in [false, true] {
+            for prefix in ["0;", "52;c;", "4;", "7;", "8;;", "99;", "x"] {
+                for n in [0usize, 1, 5, 2040, 2046, 2047, 2048, 2049, 2050, 3000] {
+                    let mut input = prefix.as_bytes().to_vec();
+                    input.extend((0..n).map(|i| b'a' + (i % 26) as u8));
+                    let mut a = Parser::init(alloc);
+                    let mut b = Parser::init(alloc);
+                    for &c in &input {
+                        a.next(c);
+                    }
+                    let mid = input.len() / 2;
+                    b.next_slice(&input[..mid]);
+                    b.next_slice(&input[mid..]);
+                    assert_eq!(a.state, b.state);
+                    assert_eq!(a.buffer, b.buffer);
+                    assert_eq!(a.writer, b.writer);
+                }
+            }
+        }
+    }
+}