Skip to content

src: use stack allocation for small string encoding #62431

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
thisalihassan wants to merge 1 commit into
base: main
Choose a base branch
from
+91 −89
Open

src: use stack allocation for small string encoding #62431

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
180 changes: 91 additions & 89 deletions src/string_bytes.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -145,12 +145,49 @@ class ExternString: public ResourceType {
size_t length_;
};

typedef ExternString<String::ExternalOneByteStringResource, char>
ExternOneByteString;
typedef ExternString<String::ExternalStringResource, uint16_t>
ExternTwoByteString;

template <typename EncodeFn>
static MaybeLocal<Value> EncodeOneByteString(Isolate* isolate,
size_t length,
EncodeFn encode) {
// 512B stack threshold: covers common small outputs (hex SHA-256/512, UUIDs).
// Larger thresholds were benchmarked
MaybeStackBuffer<char, 512> buf(length);
encode(buf.out());
// Copy stack-backed data, but release heap-backed storage to V8.
if (buf.IsAllocated()) {
char* data = buf.out();
buf.Release();
return ExternOneByteString::New(isolate, data, length);
}
return String::NewFromOneByte(isolate,
reinterpret_cast<const uint8_t*>(buf.out()),
v8::NewStringType::kNormal,
static_cast<int>(length));
}

typedef ExternString<String::ExternalOneByteStringResource,
char> ExternOneByteString;
typedef ExternString<String::ExternalStringResource,
uint16_t> ExternTwoByteString;

template <typename EncodeFn>
static MaybeLocal<Value> EncodeTwoByteString(Isolate* isolate,
size_t char_length,
EncodeFn encode) {
// 256 uint16_t = 512 bytes on the stack, matching the one-byte
MaybeStackBuffer<uint16_t, 256> buf(char_length);
encode(buf.out());
// Copy stack-backed data, but release heap-backed storage to V8.
if (buf.IsAllocated()) {
uint16_t* data = buf.out();
buf.Release();
return ExternTwoByteString::New(isolate, data, char_length);
}
return String::NewFromTwoByte(isolate,
buf.out(),
v8::NewStringType::kNormal,
static_cast<int>(char_length));
}

template <>
MaybeLocal<Value> ExternOneByteString::NewExternal(
Expand Down Expand Up @@ -513,27 +550,23 @@ MaybeLocal<Value> StringBytes::Encode(Isolate* isolate,
MaybeLocal<String> val;

switch (encoding) {
case BUFFER:
{
auto maybe_buf = Buffer::Copy(isolate, buf, buflen);
Local<v8::Object> buf;
if (!maybe_buf.ToLocal(&buf)) {
isolate->ThrowException(node::ERR_MEMORY_ALLOCATION_FAILED(isolate));
}
return buf;
case BUFFER: {
auto maybe_buf = Buffer::Copy(isolate, buf, buflen);
Local<v8::Object> buf;
if (!maybe_buf.ToLocal(&buf)) {
isolate->ThrowException(node::ERR_MEMORY_ALLOCATION_FAILED(isolate));
}
return buf;
}

case ASCII:
buflen = keep_buflen_in_range(buflen);
if (simdutf::validate_ascii_with_errors(buf, buflen).error) {
// The input contains non-ASCII bytes.
char* out = node::UncheckedMalloc(buflen);
if (out == nullptr) {
isolate->ThrowException(node::ERR_MEMORY_ALLOCATION_FAILED(isolate));
return MaybeLocal<Value>();
}
nbytes::ForceAscii(buf, out, buflen);
return ExternOneByteString::New(isolate, out, buflen);

return EncodeOneByteString(isolate, buflen, [buf, buflen](char* dst) {
nbytes::ForceAscii(buf, dst, buflen);
});
} else {
return ExternOneByteString::NewFromCopy(isolate, buf, buflen);
}
Expand All @@ -557,14 +590,12 @@ MaybeLocal<Value> StringBytes::Encode(Isolate* isolate,
isolate->ThrowException(ERR_STRING_TOO_LONG(isolate));
return MaybeLocal<Value>();
}
uint16_t* dst = node::UncheckedMalloc<uint16_t>(u16size);
if (u16size != 0 && dst == nullptr) {
THROW_ERR_MEMORY_ALLOCATION_FAILED(isolate);
return MaybeLocal<Value>();
}
size_t utf16len = simdutf::convert_valid_utf8_to_utf16(
buf, buflen, reinterpret_cast<char16_t*>(dst));
return ExternTwoByteString::New(isolate, dst, utf16len);
return EncodeTwoByteString(
isolate, u16size, [buf, buflen, u16size](uint16_t* dst) {
size_t written = simdutf::convert_valid_utf8_to_utf16(
buf, buflen, reinterpret_cast<char16_t*>(dst));
CHECK_EQ(written, u16size);
});
}

val =
Expand All @@ -583,77 +614,52 @@ MaybeLocal<Value> StringBytes::Encode(Isolate* isolate,
case BASE64: {
buflen = keep_buflen_in_range(buflen);
size_t dlen = simdutf::base64_length_from_binary(buflen);
char* dst = node::UncheckedMalloc(dlen);
if (dst == nullptr) {
isolate->ThrowException(node::ERR_MEMORY_ALLOCATION_FAILED(isolate));
return MaybeLocal<Value>();
}

size_t written = simdutf::binary_to_base64(buf, buflen, dst);
CHECK_EQ(written, dlen);

return ExternOneByteString::New(isolate, dst, dlen);
return EncodeOneByteString(isolate, dlen, [buf, buflen, dlen](char* dst) {
size_t written = simdutf::binary_to_base64(buf, buflen, dst);
CHECK_EQ(written, dlen);
});
}

case BASE64URL: {
buflen = keep_buflen_in_range(buflen);
size_t dlen =
simdutf::base64_length_from_binary(buflen, simdutf::base64_url);
char* dst = node::UncheckedMalloc(dlen);
if (dst == nullptr) {
isolate->ThrowException(node::ERR_MEMORY_ALLOCATION_FAILED(isolate));
return MaybeLocal<Value>();
}

size_t written =
simdutf::binary_to_base64(buf, buflen, dst, simdutf::base64_url);
CHECK_EQ(written, dlen);

return ExternOneByteString::New(isolate, dst, dlen);
return EncodeOneByteString(isolate, dlen, [buf, buflen, dlen](char* dst) {
size_t written =
simdutf::binary_to_base64(buf, buflen, dst, simdutf::base64_url);
CHECK_EQ(written, dlen);
});
}

case HEX: {
buflen = keep_buflen_in_range(buflen);
size_t dlen = buflen * 2;
char* dst = node::UncheckedMalloc(dlen);
if (dst == nullptr) {
isolate->ThrowException(node::ERR_MEMORY_ALLOCATION_FAILED(isolate));
return MaybeLocal<Value>();
}
size_t written = nbytes::HexEncode(buf, buflen, dst, dlen);
CHECK_EQ(written, dlen);

return ExternOneByteString::New(isolate, dst, dlen);
return EncodeOneByteString(isolate, dlen, [buf, buflen, dlen](char* dst) {
size_t written = nbytes::HexEncode(buf, buflen, dst, dlen);
CHECK_EQ(written, dlen);
});
}

case UCS2: {
buflen = keep_buflen_in_range(buflen);
size_t str_len = buflen / 2;
if constexpr (IsBigEndian()) {
uint16_t* dst = node::UncheckedMalloc<uint16_t>(str_len);
if (str_len != 0 && dst == nullptr) {
isolate->ThrowException(node::ERR_MEMORY_ALLOCATION_FAILED(isolate));
return MaybeLocal<Value>();
}
for (size_t i = 0, k = 0; k < str_len; i += 2, k += 1) {
// The input is in *little endian*, because that's what Node.js
// expects, so the high byte comes after the low byte.
const uint8_t hi = static_cast<uint8_t>(buf[i + 1]);
const uint8_t lo = static_cast<uint8_t>(buf[i + 0]);
dst[k] = static_cast<uint16_t>(hi) << 8 | lo;
}
return ExternTwoByteString::New(isolate, dst, str_len);
return EncodeTwoByteString(
isolate, str_len, [buf, str_len](uint16_t* dst) {
for (size_t i = 0, k = 0; k < str_len; i += 2, k += 1) {
// The input is in *little endian*, because that's what Node.js
// expects, so the high byte comes after the low byte.
const uint8_t hi = static_cast<uint8_t>(buf[i + 1]);
const uint8_t lo = static_cast<uint8_t>(buf[i + 0]);
dst[k] = static_cast<uint16_t>(hi) << 8 | lo;
}
});
}
if (reinterpret_cast<uintptr_t>(buf) % 2 != 0) {
// Unaligned data still means we can't directly pass it to V8.
char* dst = node::UncheckedMalloc(buflen);
if (dst == nullptr) {
isolate->ThrowException(node::ERR_MEMORY_ALLOCATION_FAILED(isolate));
return MaybeLocal<Value>();
}
memcpy(dst, buf, buflen);
return ExternTwoByteString::New(
isolate, reinterpret_cast<uint16_t*>(dst), str_len);
return EncodeTwoByteString(
isolate, str_len, [buf, buflen](uint16_t* dst) {
memcpy(dst, buf, buflen);
});
}
return ExternTwoByteString::NewFromCopy(
isolate, reinterpret_cast<const uint16_t*>(buf), str_len);
Expand All @@ -675,15 +681,11 @@ MaybeLocal<Value> StringBytes::Encode(Isolate* isolate,
// https://nodejs.org/api/buffer.html regarding Node's "ucs2"
// encoding specification
if constexpr (IsBigEndian()) {
uint16_t* dst = node::UncheckedMalloc<uint16_t>(buflen);
if (dst == nullptr) {
isolate->ThrowException(node::ERR_MEMORY_ALLOCATION_FAILED(isolate));
return MaybeLocal<Value>();
}
size_t nbytes = buflen * sizeof(uint16_t);
memcpy(dst, buf, nbytes);
CHECK(nbytes::SwapBytes16(reinterpret_cast<char*>(dst), nbytes));
return ExternTwoByteString::New(isolate, dst, buflen);
return EncodeTwoByteString(isolate, buflen, [buf, buflen](uint16_t* dst) {
size_t nbytes = buflen * sizeof(uint16_t);
memcpy(dst, buf, nbytes);
CHECK(nbytes::SwapBytes16(reinterpret_cast<char*>(dst), nbytes));
});
} else {
return ExternTwoByteString::NewFromCopy(isolate, buf, buflen);
}
Expand Down
Loading