serenity/AK/BitStream.h
Timothy Flynn eed956b473 AK: Increase LittleEndianOutputBitStream's buffer size and remove loops
This is very similar to the LittleEndianInputBitStream bit buffer change
from 8e834d4bb2.

We currently buffer one byte of data for the underlying stream. And when
we put bits onto that buffer, we do so 1 bit at a time.

This replaces the u8 buffer with a u64. And instead of looping at all,
we perform bitwise operations to write the desired number of bits.

Using the "enwik8" file as a test (100MB uncompressed, commonly used in
benchmarks: https://www.mattmahoney.net/dc/enwik8.zip), compression time
decreases from:

    13.62s to 10.9s on Serenity (cold)
    13.62s to 9.22s on Serenity (warm)
    2.93s to 2.32s on Linux

One caveat is that this requires explicitly flushing any leftover bits
when the caller is done with the stream. The byte buffer implementation
implicitly flushed its data every time the buffer was byte-aligned, as
doing so would always fill the byte. This is no longer the case. But for
now, this should be fine as the one user of this class, DEFLATE, already
has a "flush everything now that we're done" finalizer.
2023-04-02 10:54:37 +02:00

437 lines
12 KiB
C++

/*
* Copyright (c) 2021, kleines Filmröllchen <filmroellchen@serenityos.org>.
*
* SPDX-License-Identifier: BSD-2-Clause
*/
#pragma once
#include <AK/ByteBuffer.h>
#include <AK/MaybeOwned.h>
#include <AK/NumericLimits.h>
#include <AK/OwnPtr.h>
#include <AK/Stream.h>
namespace AK {
/// A stream wrapper class that allows you to read arbitrary amounts of bits
/// in big-endian order from another stream.
class BigEndianInputBitStream : public Stream {
public:
explicit BigEndianInputBitStream(MaybeOwned<Stream> stream)
: m_stream(move(stream))
{
}
// ^Stream
virtual ErrorOr<Bytes> read_some(Bytes bytes) override
{
if (m_current_byte.has_value() && is_aligned_to_byte_boundary()) {
bytes[0] = m_current_byte.release_value();
// FIXME: This accidentally slices off the first byte of the returned span.
return m_stream->read_some(bytes.slice(1));
}
align_to_byte_boundary();
return m_stream->read_some(bytes);
}
virtual ErrorOr<size_t> write_some(ReadonlyBytes bytes) override { return m_stream->write_some(bytes); }
virtual bool is_eof() const override { return m_stream->is_eof() && !m_current_byte.has_value(); }
virtual bool is_open() const override { return m_stream->is_open(); }
virtual void close() override
{
m_stream->close();
align_to_byte_boundary();
}
ErrorOr<bool> read_bit()
{
return read_bits<bool>(1);
}
/// Depending on the number of bits to read, the return type can be chosen appropriately.
/// This avoids a bunch of static_cast<>'s for the user.
// TODO: Support u128, u256 etc. as well: The concepts would be quite complex.
template<Unsigned T = u64>
ErrorOr<T> read_bits(size_t count)
{
if constexpr (IsSame<bool, T>) {
VERIFY(count == 1);
}
T result = 0;
size_t nread = 0;
while (nread < count) {
if (m_current_byte.has_value()) {
if constexpr (!IsSame<bool, T> && !IsSame<u8, T>) {
// read as many bytes as possible directly
if (((count - nread) >= 8) && is_aligned_to_byte_boundary()) {
// shift existing data over
result <<= 8;
result |= m_current_byte.value();
nread += 8;
m_current_byte.clear();
} else {
auto const bit = (m_current_byte.value() >> (7 - m_bit_offset)) & 1;
result <<= 1;
result |= bit;
++nread;
if (m_bit_offset++ == 7)
m_current_byte.clear();
}
} else {
// Always take this branch for booleans or u8: there's no purpose in reading more than a single bit
auto const bit = (m_current_byte.value() >> (7 - m_bit_offset)) & 1;
if constexpr (IsSame<bool, T>)
result = bit;
else {
result <<= 1;
result |= bit;
}
++nread;
if (m_bit_offset++ == 7)
m_current_byte.clear();
}
} else {
m_current_byte = TRY(m_stream->read_value<u8>());
m_bit_offset = 0;
}
}
return result;
}
/// Discards any sub-byte stream positioning the input stream may be keeping track of.
/// Non-bitwise reads will implicitly call this.
void align_to_byte_boundary()
{
m_current_byte.clear();
m_bit_offset = 0;
}
/// Whether we are (accidentally or intentionally) at a byte boundary right now.
ALWAYS_INLINE bool is_aligned_to_byte_boundary() const { return m_bit_offset == 0; }
private:
Optional<u8> m_current_byte;
size_t m_bit_offset { 0 };
MaybeOwned<Stream> m_stream;
};
class LittleEndianBitStream : public Stream {
protected:
using BufferType = u64;
static constexpr size_t bits_per_byte = 8u;
static constexpr size_t bit_buffer_size = sizeof(BufferType) * bits_per_byte;
explicit LittleEndianBitStream(MaybeOwned<Stream> stream)
: m_stream(move(stream))
{
}
template<Unsigned T>
static constexpr T lsb_mask(T bits)
{
constexpr auto max = NumericLimits<T>::max();
constexpr auto digits = NumericLimits<T>::digits();
return bits == 0 ? 0 : max >> (digits - bits);
}
ALWAYS_INLINE BufferType lsb_aligned_buffer() const
{
return m_bit_offset == bit_buffer_size ? 0 : m_bit_buffer >> m_bit_offset;
}
ALWAYS_INLINE bool is_aligned_to_byte_boundary() const { return m_bit_count % bits_per_byte == 0; }
MaybeOwned<Stream> m_stream;
BufferType m_bit_buffer { 0 };
u8 m_bit_offset { 0 };
u8 m_bit_count { 0 };
};
/// A stream wrapper class that allows you to read arbitrary amounts of bits
/// in little-endian order from another stream.
class LittleEndianInputBitStream : public LittleEndianBitStream {
public:
explicit LittleEndianInputBitStream(MaybeOwned<Stream> stream)
: LittleEndianBitStream(move(stream))
{
}
// ^Stream
virtual ErrorOr<Bytes> read_some(Bytes bytes) override
{
align_to_byte_boundary();
size_t bytes_read = 0;
auto buffer = bytes;
if (m_bit_count > 0) {
auto bits_to_read = min(buffer.size() * bits_per_byte, m_bit_count);
auto result = TRY(read_bits(bits_to_read));
bytes_read = bits_to_read / bits_per_byte;
buffer.overwrite(0, &result, bytes_read);
buffer = buffer.slice(bytes_read);
}
buffer = TRY(m_stream->read_some(buffer));
bytes_read += buffer.size();
return bytes.trim(bytes_read);
}
virtual ErrorOr<size_t> write_some(ReadonlyBytes bytes) override { return m_stream->write_some(bytes); }
virtual bool is_eof() const override { return m_stream->is_eof() && m_bit_count == 0; }
virtual bool is_open() const override { return m_stream->is_open(); }
virtual void close() override
{
m_stream->close();
align_to_byte_boundary();
}
ErrorOr<bool> read_bit()
{
return read_bits<bool>(1);
}
/// Depending on the number of bits to read, the return type can be chosen appropriately.
/// This avoids a bunch of static_cast<>'s for the user.
// TODO: Support u128, u256 etc. as well: The concepts would be quite complex.
template<Unsigned T = u64>
ErrorOr<T> read_bits(size_t count)
{
auto result = TRY(peek_bits<T>(count));
discard_previously_peeked_bits(count);
return result;
}
template<Unsigned T = u64>
ErrorOr<T> peek_bits(size_t count)
{
if (count > m_bit_count)
TRY(refill_buffer_from_stream());
return lsb_aligned_buffer() & lsb_mask<T>(min(count, m_bit_count));
}
ALWAYS_INLINE void discard_previously_peeked_bits(u8 count)
{
m_bit_offset += count;
m_bit_count -= count;
}
/// Discards any sub-byte stream positioning the input stream may be keeping track of.
/// Non-bitwise reads will implicitly call this.
u8 align_to_byte_boundary()
{
u8 remaining_bits = 0;
m_bit_buffer = lsb_aligned_buffer();
m_bit_offset = 0;
if (auto offset = m_bit_count % bits_per_byte; offset != 0) {
remaining_bits = m_bit_buffer & lsb_mask<u8>(offset);
discard_previously_peeked_bits(offset);
}
return remaining_bits;
}
private:
ErrorOr<void> refill_buffer_from_stream()
{
size_t bits_to_read = bit_buffer_size - m_bit_count;
size_t bytes_to_read = bits_to_read / bits_per_byte;
BufferType buffer = 0;
auto bytes = TRY(m_stream->read_some({ &buffer, bytes_to_read }));
m_bit_buffer = (buffer << m_bit_count) | lsb_aligned_buffer();
m_bit_count += bytes.size() * bits_per_byte;
m_bit_offset = 0;
return {};
}
};
/// A stream wrapper class that allows you to write arbitrary amounts of bits
/// in big-endian order to another stream.
class BigEndianOutputBitStream : public Stream {
public:
explicit BigEndianOutputBitStream(MaybeOwned<Stream> stream)
: m_stream(move(stream))
{
}
virtual ErrorOr<Bytes> read_some(Bytes) override
{
return Error::from_errno(EBADF);
}
virtual ErrorOr<size_t> write_some(ReadonlyBytes bytes) override
{
VERIFY(m_bit_offset == 0);
return m_stream->write_some(bytes);
}
template<Unsigned T>
ErrorOr<void> write_bits(T value, size_t bit_count)
{
VERIFY(m_bit_offset <= 7);
while (bit_count > 0) {
u8 next_bit = (value >> (bit_count - 1)) & 1;
bit_count--;
m_current_byte <<= 1;
m_current_byte |= next_bit;
m_bit_offset++;
if (m_bit_offset > 7) {
TRY(m_stream->write_value(m_current_byte));
m_bit_offset = 0;
m_current_byte = 0;
}
}
return {};
}
virtual bool is_eof() const override
{
return true;
}
virtual bool is_open() const override
{
return m_stream->is_open();
}
virtual void close() override
{
}
size_t bit_offset() const
{
return m_bit_offset;
}
ErrorOr<void> align_to_byte_boundary()
{
if (m_bit_offset == 0)
return {};
TRY(write_bits(0u, 8 - m_bit_offset));
VERIFY(m_bit_offset == 0);
return {};
}
private:
MaybeOwned<Stream> m_stream;
u8 m_current_byte { 0 };
size_t m_bit_offset { 0 };
};
/// A stream wrapper class that allows you to write arbitrary amounts of bits
/// in little-endian order to another stream.
class LittleEndianOutputBitStream : public LittleEndianBitStream {
public:
explicit LittleEndianOutputBitStream(MaybeOwned<Stream> stream)
: LittleEndianBitStream(move(stream))
{
}
virtual ErrorOr<Bytes> read_some(Bytes) override
{
return Error::from_errno(EBADF);
}
virtual ErrorOr<size_t> write_some(ReadonlyBytes bytes) override
{
VERIFY(is_aligned_to_byte_boundary());
if (m_bit_count > 0)
TRY(flush_buffer_to_stream());
return m_stream->write_some(bytes);
}
template<Unsigned T>
ErrorOr<void> write_bits(T value, size_t count)
{
if (m_bit_count == bit_buffer_size) {
TRY(flush_buffer_to_stream());
} else if (auto remaining = bit_buffer_size - m_bit_count; count >= remaining) {
m_bit_buffer |= (static_cast<BufferType>(value) & lsb_mask<BufferType>(remaining)) << m_bit_count;
m_bit_count = bit_buffer_size;
if (remaining != sizeof(value) * bits_per_byte)
value >>= remaining;
count -= remaining;
TRY(flush_buffer_to_stream());
}
if (count == 0)
return {};
m_bit_buffer |= static_cast<BufferType>(value) << m_bit_count;
m_bit_count += count;
return {};
}
ALWAYS_INLINE ErrorOr<void> flush_buffer_to_stream()
{
auto bytes_to_write = m_bit_count / bits_per_byte;
TRY(m_stream->write_until_depleted({ &m_bit_buffer, bytes_to_write }));
if (m_bit_count == bit_buffer_size) {
m_bit_buffer = 0;
m_bit_count = 0;
} else {
auto bits_written = bytes_to_write * bits_per_byte;
m_bit_buffer >>= bits_written;
m_bit_count -= bits_written;
}
return {};
}
virtual bool is_eof() const override
{
return true;
}
virtual bool is_open() const override
{
return m_stream->is_open();
}
virtual void close() override
{
}
size_t bit_offset() const
{
return m_bit_count;
}
ErrorOr<void> align_to_byte_boundary()
{
if (auto offset = m_bit_count % bits_per_byte; offset != 0)
TRY(write_bits<u8>(0u, bits_per_byte - offset));
return {};
}
};
}