perf(web): optimize encodeInto() (#15922)

cli/bench/encode_into.js

@@ -0,0 +1,23 @@
+// Copyright 2018-2022 the Deno authors. All rights reserved. MIT license.
+const queueMicrotask = globalThis.queueMicrotask || process.nextTick;
+let [total, count] = typeof Deno !== "undefined"
+  ? Deno.args
+  : [process.argv[2], process.argv[3]];
+
+total = total ? parseInt(total, 0) : 50;
+count = count ? parseInt(count, 10) : 1000000;
+
+function bench(fun) {
+  const start = Date.now();
+  for (let i = 0; i < count; i++) fun();
+  const elapsed = Date.now() - start;
+  const rate = Math.floor(count / (elapsed / 1000));
+  console.log(`time ${elapsed} ms rate ${rate}`);
+  if (--total) queueMicrotask(() => bench(fun));
+}
+
+const encoder = new TextEncoder();
+const data = "hello world";
+const out = new Uint8Array(100);
+
+bench(() => encoder.encodeInto(data, out));

ext/web/08_text_encoding.js

@@ -25,6 +25,7 @@
     TypedArrayPrototypeSubarray,
     TypedArrayPrototypeSlice,
     Uint8Array,
+    Uint32Array,
   } = window.__bootstrap.primordials;
 
   class TextDecoder {
@@ -199,10 +200,16 @@
         context: "Argument 2",
         allowShared: true,
       });
-      return ops.op_encoding_encode_into(source, destination);
+      ops.op_encoding_encode_into(source, destination, encodeIntoBuf);
+      return {
+        read: encodeIntoBuf[0],
+        written: encodeIntoBuf[1],
+      };
     }
   }
 
+  const encodeIntoBuf = new Uint32Array(2);
+
   webidl.configurePrototype(TextEncoder);
   const TextEncoderPrototype = TextEncoder.prototype;
 

ext/web/lib.rs

@@ -10,7 +10,9 @@ use deno_core::error::type_error;
 use deno_core::error::AnyError;
 use deno_core::include_js_files;
 use deno_core::op;
+use deno_core::serde_v8;
 use deno_core::url::Url;
+use deno_core::v8;
 use deno_core::ByteString;
 use deno_core::CancelHandle;
 use deno_core::Extension;
@@ -19,11 +21,11 @@ use deno_core::Resource;
 use deno_core::ResourceId;
 use deno_core::U16String;
 use deno_core::ZeroCopyBuf;
+
 use encoding_rs::CoderResult;
 use encoding_rs::Decoder;
 use encoding_rs::DecoderResult;
 use encoding_rs::Encoding;
-use serde::Serialize;
 use std::borrow::Cow;
 use std::cell::RefCell;
 use std::fmt;
@@ -314,46 +316,25 @@ impl Resource for TextDecoderResource {
   }
 }
 
-#[derive(Serialize)]
+#[op(v8)]
-#[serde(rename_all = "camelCase")]
-struct EncodeIntoResult {
-  read: usize,
-  written: usize,
-}
-
-#[op]
 fn op_encoding_encode_into(
-  input: String,
+  scope: &mut v8::HandleScope,
+  input: serde_v8::Value,
   buffer: &mut [u8],
-) -> EncodeIntoResult {
+  out_buf: &mut [u32],
-  // Since `input` is already UTF-8, we can simply find the last UTF-8 code
+) -> Result<(), AnyError> {
-  // point boundary from input that fits in `buffer`, and copy the bytes up to
+  let s = v8::Local::<v8::String>::try_from(input.v8_value)?;
-  // that point.
-  let boundary = if buffer.len() >= input.len() {
-    input.len()
-  } else {
-    let mut boundary = buffer.len();
 
-    // The maximum length of a UTF-8 code point is 4 bytes.
+  let mut nchars = 0;
-    for _ in 0..4 {
+  out_buf[1] = s.write_utf8(
-      if input.is_char_boundary(boundary) {
+    scope,
-        break;
+    buffer,
-      }
+    Some(&mut nchars),
-      debug_assert!(boundary > 0);
+    v8::WriteOptions::NO_NULL_TERMINATION
-      boundary -= 1;
+      | v8::WriteOptions::REPLACE_INVALID_UTF8,
-    }
+  ) as u32;
-
+  out_buf[0] = nchars as u32;
-    debug_assert!(input.is_char_boundary(boundary));
+  Ok(())
-    boundary
-  };
-
-  buffer[..boundary].copy_from_slice(input[..boundary].as_bytes());
-
-  EncodeIntoResult {
-    // The `read` output parameter is measured in UTF-16 code units.
-    read: input[..boundary].encode_utf16().count(),
-    written: boundary,
-  }
 }
 
 /// Creates a [`CancelHandle`] resource that can be used to cancel invocations of certain ops.

ops/lib.rs

@@ -721,42 +721,20 @@ fn codegen_arg(
     };
   }
   // Fast path for &/&mut [u8] and &/&mut [u32]
-  if let Some(ty) = is_ref_slice(&**ty) {
+  match is_ref_slice(&**ty) {
-    let (ptr_ty, mutability) = match ty {
+    None => {}
-      SliceType::U8 => (quote!([u8]), quote!(&)),
+    Some(SliceType::U32Mut) => {
-      SliceType::U8Mut => (quote!([u8]), quote!(&mut)),
+      let blck = codegen_u32_mut_slice(core, idx);
-    };
       return quote! {
-      let #ident = {
+        let #ident = #blck;
-        let value = args.get(#idx as i32);
-        if let Ok(view) = #core::v8::Local::<#core::v8::ArrayBufferView>::try_from(value) {
-          let (offset, len) = (view.byte_offset(), view.byte_length());
-          let buffer = match view.buffer(scope) {
-              Some(v) => v,
-              None => {
-                return #core::_ops::throw_type_error(scope, format!("Expected ArrayBufferView at position {}", #idx));
-              }
       };
-          let store = buffer.get_backing_store();
-          if store.is_shared() {
-            return #core::_ops::throw_type_error(scope, format!("Expected non-shared ArrayBufferView at position {}", #idx));
-          }
-          unsafe { #mutability *(&store[offset..offset + len] as *const _ as *mut #ptr_ty) }
-        } else {
-          let b: #core::v8::Local<#core::v8::ArrayBuffer> = match value.try_into() {
-            Ok(v) => v,
-            Err(_) => {
-              return #core::_ops::throw_type_error(scope, format!("Expected ArrayBuffer at position {}", #idx));
     }
+    Some(_) => {
+      let blck = codegen_u8_slice(core, idx);
+      return quote! {
+        let #ident = #blck;
       };
-          let store = b.get_backing_store();
-          if store.is_shared() {
-            return #core::_ops::throw_type_error(scope, format!("Expected non-shared ArrayBufferView at position {}", #idx));
     }
-          unsafe { #mutability *(&store[0..b.byte_length()] as *const _ as *mut #ptr_ty) }
-        }
-      };
-    };
   }
   // Otherwise deserialize it via serde_v8
   quote! {
@@ -771,6 +749,54 @@ fn codegen_arg(
   }
 }
 
+fn codegen_u8_slice(core: &TokenStream2, idx: usize) -> TokenStream2 {
+  quote! {{
+    let value = args.get(#idx as i32);
+    match #core::v8::Local::<#core::v8::ArrayBuffer>::try_from(value) {
+      Ok(b) => {
+        let store = b.data() as *mut u8;
+        // SAFETY: rust guarantees that lifetime of slice is no longer than the call.
+        unsafe { ::std::slice::from_raw_parts_mut(store, b.byte_length()) }
+      },
+      Err(_) => {
+        if let Ok(view) = #core::v8::Local::<#core::v8::ArrayBufferView>::try_from(value) {
+          let (offset, len) = (view.byte_offset(), view.byte_length());
+          let buffer = match view.buffer(scope) {
+              Some(v) => v,
+              None => {
+                return #core::_ops::throw_type_error(scope, format!("Expected ArrayBufferView at position {}", #idx));
+              }
+          };
+          let store = buffer.data() as *mut u8;
+          // SAFETY: rust guarantees that lifetime of slice is no longer than the call.
+          unsafe { ::std::slice::from_raw_parts_mut(store.add(offset), len) }
+        } else {
+          return #core::_ops::throw_type_error(scope, format!("Expected ArrayBufferView at position {}", #idx));
+        }
+      }
+    }}
+  }
+}
+
+fn codegen_u32_mut_slice(core: &TokenStream2, idx: usize) -> TokenStream2 {
+  quote! {
+    if let Ok(view) = #core::v8::Local::<#core::v8::Uint32Array>::try_from(args.get(#idx as i32)) {
+      let (offset, len) = (view.byte_offset(), view.byte_length());
+      let buffer = match view.buffer(scope) {
+          Some(v) => v,
+          None => {
+            return #core::_ops::throw_type_error(scope, format!("Expected Uint32Array at position {}", #idx));
+          }
+      };
+      let store = buffer.data() as *mut u8;
+      // SAFETY: buffer from Uint32Array. Rust guarantees that lifetime of slice is no longer than the call.
+      unsafe { ::std::slice::from_raw_parts_mut(store.add(offset) as *mut u32, len / 4) }
+    } else {
+      return #core::_ops::throw_type_error(scope, format!("Expected Uint32Array at position {}", #idx));
+    }
+  }
+}
+
 fn codegen_sync_ret(
   core: &TokenStream2,
   output: &syn::ReturnType,
@@ -849,6 +875,7 @@ fn is_option_string(ty: impl ToTokens) -> bool {
 enum SliceType {
   U8,
   U8Mut,
+  U32Mut,
 }
 
 fn is_ref_slice(ty: impl ToTokens) -> Option<SliceType> {
@@ -858,6 +885,9 @@ fn is_ref_slice(ty: impl ToTokens) -> Option<SliceType> {
   if is_u8_slice_mut(&ty) {
     return Some(SliceType::U8Mut);
   }
+  if is_u32_slice_mut(&ty) {
+    return Some(SliceType::U32Mut);
+  }
   None
 }
 
@@ -869,6 +899,10 @@ fn is_u8_slice_mut(ty: impl ToTokens) -> bool {
   tokens(ty) == "& mut [u8]"
 }
 
+fn is_u32_slice_mut(ty: impl ToTokens) -> bool {
+  tokens(ty) == "& mut [u32]"
+}
+
 fn is_optional_fast_callback_option(ty: impl ToTokens) -> bool {
   tokens(&ty).contains("Option < & mut FastApiCallbackOptions")
 }

tools/wpt/expectation.json

@@ -991,94 +991,8 @@
     "api-replacement-encodings.any.worker.html": true,
     "api-surrogates-utf8.any.html": true,
     "api-surrogates-utf8.any.worker.html": true,
-    "encodeInto.any.html": [
+    "encodeInto.any.html": true,
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 0, filler 0",
+    "encodeInto.any.worker.html": true,
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 4, filler random"
-    ],
-    "encodeInto.any.worker.html": [
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with Hi and destination length 0, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with A and destination length 10, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with 𝌆 and destination length 4, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with 𝌆A and destination length 3, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with \ud834A\udf06A¥Hi and destination length 10, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with A\udf06 and destination length 4, offset 4, filler random",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 0, filler 0",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 4, filler 0",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 0, filler 128",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 4, filler 128",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 0, filler random",
-      "encodeInto() into SharedArrayBuffer with ¥¥ and destination length 4, offset 4, filler random"
-    ],
     "idlharness.any.html": true,
     "idlharness.any.worker.html": true,
     "iso-2022-jp-decoder.any.html": true,
@@ -1117,7 +1031,8 @@
       "encode-utf8.any.html": true,
       "encode-utf8.any.worker.html": true,
       "readable-writable-properties.any.html": true,
-      "readable-writable-properties.any.worker.html": true
+      "readable-writable-properties.any.worker.html": true,
+      "realms.window.html": false
     },
     "textdecoder-arguments.any.html": true,
     "textdecoder-arguments.any.worker.html": true,
@@ -1162,7 +1077,9 @@
     "single-byte-decoder.window.html?TextDecoder": true,
     "textdecoder-eof.any.html": true,
     "textdecoder-eof.any.worker.html": true,
-    "idlharness-shadowrealm.window.html": false
+    "idlharness-shadowrealm.window.html": false,
+    "single-byte-decoder.window.html?XMLHttpRequest": false,
+    "single-byte-decoder.window.html?document": false
   },
   "hr-time": {
     "monotonic-clock.any.html": true,