Merge pull request #377 from Diggsey/apply-mask

Refactor `apply_mask` implementation, removing dead code paths and re…

src/ws/mask.rs

@@ -1,88 +1,106 @@
 //! This is code from [Tungstenite project](https://github.com/snapview/tungstenite-rs)
 #![cfg_attr(feature = "cargo-clippy", allow(cast_ptr_alignment))]
-use std::cmp::min;
+use std::slice;
-use std::mem::uninitialized;
 use std::ptr::copy_nonoverlapping;
 
+// Holds a slice guaranteed to be shorter than 8 bytes
+struct ShortSlice<'a>(&'a mut [u8]);
+
+impl<'a> ShortSlice<'a> {
+    unsafe fn new(slice: &'a mut [u8]) -> Self {
+        // Sanity check for debug builds
+        debug_assert!(slice.len() < 8);
+        ShortSlice(slice)
+    }
+    fn len(&self) -> usize {
+        self.0.len()
+    }
+}
+
 /// Faster version of `apply_mask()` which operates on 8-byte blocks.
-///
-/// unsafe because uses pointer math and bit operations for performance
 #[inline]
 #[cfg_attr(feature = "cargo-clippy", allow(cast_lossless))]
 pub(crate) fn apply_mask(buf: &mut [u8], mask_u32: u32) {
-    unsafe {
+    // Extend the mask to 64 bits
-        let mut ptr = buf.as_mut_ptr();
+    let mut mask_u64 = ((mask_u32 as u64) << 32) | (mask_u32 as u64);
-        let mut len = buf.len();
+    // Split the buffer into three segments
+    let (head, mid, tail) = align_buf(buf);
 
-        // Possible first unaligned block.
+    // Initial unaligned segment
-        let head = min(len, (8 - (ptr as usize & 0x7)) & 0x3);
+    let head_len = head.len();
-        let mask_u32 = if head > 0 {
+    if head_len > 0 {
-            let n = if head > 4 { head - 4 } else { head };
+        xor_short(head, mask_u64);
-
-            let mask_u32 = if n > 0 {
-                xor_mem(ptr, mask_u32, n);
-                ptr = ptr.offset(head as isize);
-                len -= n;
         if cfg!(target_endian = "big") {
-                    mask_u32.rotate_left(8 * n as u32)
+            mask_u64 = mask_u64.rotate_left(8 * head_len as u32);
         } else {
-                    mask_u32.rotate_right(8 * n as u32)
+            mask_u64 = mask_u64.rotate_right(8 * head_len as u32);
-                }
-            } else {
-                mask_u32
-            };
-
-            if head > 4 {
-                *(ptr as *mut u32) ^= mask_u32;
-                ptr = ptr.offset(4);
-                len -= 4;
-            }
-            mask_u32
-        } else {
-            mask_u32
-        };
-
-        if len > 0 {
-            debug_assert_eq!(ptr as usize % 4, 0);
-        }
-
-        // Properly aligned middle of the data.
-        if len >= 8 {
-            let mut mask_u64 = mask_u32 as u64;
-            mask_u64 = mask_u64 << 32 | mask_u32 as u64;
-
-            while len >= 8 {
-                *(ptr as *mut u64) ^= mask_u64;
-                ptr = ptr.offset(8);
-                len -= 8;
         }
     }
-
+    // Aligned segment
-        while len >= 4 {
+    for v in mid {
-            *(ptr as *mut u32) ^= mask_u32;
+        *v ^= mask_u64;
-            ptr = ptr.offset(4);
-            len -= 4;
-        }
-
-        // Possible last block.
-        if len > 0 {
-            xor_mem(ptr, mask_u32, len);
     }
+    // Final unaligned segment
+    if tail.len() > 0 {
+        xor_short(tail, mask_u64);
     }
 }
 
 #[inline]
 // TODO: copy_nonoverlapping here compiles to call memcpy. While it is not so
-// inefficient, it could be done better. The compiler does not see that len is
+// inefficient, it could be done better. The compiler does not understand that
-// limited to 3.
+// a `ShortSlice` must be smaller than a u64.
-unsafe fn xor_mem(ptr: *mut u8, mask: u32, len: usize) {
+fn xor_short(buf: ShortSlice, mask: u64) {
-    let mut b: u32 = uninitialized();
+    // Unsafe: we know that a `ShortSlice` fits in a u64
+    unsafe {
+        let (ptr, len) = (buf.0.as_mut_ptr(), buf.0.len());
+        let mut b: u64 = 0;
         #[allow(trivial_casts)]
         copy_nonoverlapping(ptr, &mut b as *mut _ as *mut u8, len);
         b ^= mask;
         #[allow(trivial_casts)]
         copy_nonoverlapping(&b as *const _ as *const u8, ptr, len);
     }
+}
+
+#[inline]
+// Unsafe: caller must ensure the buffer has the correct size and alignment
+unsafe fn cast_slice(buf: &mut [u8]) -> &mut [u64] {
+    // Assert correct size and alignment in debug builds
+    debug_assert!(buf.len() & 0x7 == 0);
+    debug_assert!(buf.as_ptr() as usize & 0x7 == 0);
+
+    slice::from_raw_parts_mut(buf.as_mut_ptr() as *mut u64, buf.len() >> 3)
+}
+
+#[inline]
+// Splits a slice into three parts: an unaligned short head and tail, plus an aligned
+// u64 mid section.
+fn align_buf(buf: &mut [u8]) -> (ShortSlice, &mut [u64], ShortSlice) {
+    let start_ptr = buf.as_ptr() as usize;
+    let end_ptr = start_ptr + buf.len();
+
+    // Round *up* to next aligned boundary for start
+    let start_aligned = (start_ptr+7) & !0x7;
+    // Round *down* to last aligned boundary for end
+    let end_aligned = end_ptr & !0x7;
+    
+    if end_aligned >= start_aligned {
+        // We have our three segments (head, mid, tail)
+        let (tmp, tail) = buf.split_at_mut(end_aligned - start_ptr);
+        let (head, mid) = tmp.split_at_mut(start_aligned - start_ptr);
+
+        // Unsafe: we know the middle section is correctly aligned, and the outer
+        // sections are smaller than 8 bytes
+        unsafe { (ShortSlice::new(head), cast_slice(mid), ShortSlice(tail)) }
+    } else {
+        // We didn't cross even one aligned boundary!
+
+        // Unsafe: The outer sections are smaller than 8 bytes
+        unsafe { (ShortSlice::new(buf), &mut [], ShortSlice::new(&mut [])) }
+    }
+}
+
 
 #[cfg(test)]
 mod tests {