netty5/transport/src/main/java/io/netty/channel/AbstractChannel.java

893 lines
28 KiB
Java
Raw Normal View History

/*
2012-06-04 22:31:44 +02:00
* Copyright 2012 The Netty Project
*
* The Netty Project licenses this file to you under the Apache License,
* version 2.0 (the "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at:
*
2012-06-04 22:31:44 +02:00
* http://www.apache.org/licenses/LICENSE-2.0
*
2009-08-28 09:15:49 +02:00
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
2009-08-28 09:15:49 +02:00
* License for the specific language governing permissions and limitations
* under the License.
*/
2011-12-09 04:38:59 +01:00
package io.netty.channel;
import io.netty.buffer.ByteBuf;
ByteBufAllocator API w/ ByteBuf perf improvements This commit introduces a new API for ByteBuf allocation which fixes issue #643 along with refactoring of ByteBuf for simplicity and better performance. (see #62) A user can configure the ByteBufAllocator of a Channel via ChannelOption.ALLOCATOR or ChannelConfig.get/setAllocator(). The default allocator is currently UnpooledByteBufAllocator.HEAP_BY_DEFAULT. To allocate a buffer, do not use Unpooled anymore. do the following: ctx.alloc().buffer(...); // allocator chooses the buffer type. ctx.alloc().heapBuffer(...); ctx.alloc().directBuffer(...); To deallocate a buffer, use the unsafe free() operation: ((UnsafeByteBuf) buf).free(); The following is the list of the relevant changes: - Add ChannelInboundHandler.freeInboundBuffer() and ChannelOutboundHandler.freeOutboundBuffer() to let a user free the buffer he or she allocated. ChannelHandler adapter classes implement is already, so most users won't need to call free() by themselves. freeIn/OutboundBuffer() methods are invoked when a Channel is closed and deregistered. - All ByteBuf by contract must implement UnsafeByteBuf. To access an unsafe operation: ((UnsafeByteBuf) buf).internalNioBuffer() - Replace WrappedByteBuf and ByteBuf.Unsafe with UnsafeByteBuf to simplify overall class hierarchy and to avoid unnecesary instantiation of Unsafe instances on an unsafe operation. - Remove buffer reference counting which is confusing - Instantiate SwappedByteBuf lazily to avoid instantiation cost - Rename ChannelFutureFactory to ChannelPropertyAccess and move common methods between Channel and ChannelHandlerContext there. Also made it package-private to hide it from a user. - Remove unused unsafe operations such as newBuffer() - Add DetectionUtil.canFreeDirectBuffer() so that an allocator decides which buffer type to use safely
2012-11-15 22:04:37 +01:00
import io.netty.buffer.ByteBufAllocator;
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
import io.netty.buffer.ByteBufHolder;
import io.netty.util.DefaultAttributeMap;
import io.netty.util.internal.PlatformDependent;
import io.netty.util.internal.logging.InternalLogger;
import io.netty.util.internal.logging.InternalLoggerFactory;
import java.io.EOFException;
import java.io.IOException;
import java.net.InetSocketAddress;
import java.net.SocketAddress;
import java.nio.channels.ClosedChannelException;
import java.util.Random;
import java.util.concurrent.ConcurrentMap;
/**
2008-09-02 09:13:20 +02:00
* A skeletal {@link Channel} implementation.
*/
public abstract class AbstractChannel extends DefaultAttributeMap implements Channel {
private static final InternalLogger logger = InternalLoggerFactory.getInstance(AbstractChannel.class);
static final ConcurrentMap<Integer, Channel> allChannels = PlatformDependent.newConcurrentHashMap();
private static final Random random = new Random();
/**
* Generates a negative unique integer ID. This method generates only
* negative integers to avoid conflicts with user-specified IDs where only
* non-negative integers are allowed.
*/
private static Integer allocateId(Channel channel) {
int idVal = random.nextInt();
if (idVal > 0) {
idVal = -idVal;
} else if (idVal == 0) {
idVal = -1;
}
Integer id;
for (;;) {
id = Integer.valueOf(idVal);
// Loop until a unique ID is acquired.
// It should be found in one loop practically.
if (allChannels.putIfAbsent(id, channel) == null) {
// Successfully acquired.
return id;
} else {
// Taken by other channel at almost the same moment.
idVal --;
if (idVal >= 0) {
idVal = -1;
}
}
}
}
private final Channel parent;
private final Integer id;
private final Unsafe unsafe;
private final DefaultChannelPipeline pipeline;
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
private final ChannelOutboundBuffer outboundBuffer = new ChannelOutboundBuffer(this);
private final ChannelFuture succeededFuture = new SucceededChannelFuture(this, null);
private final VoidChannelPromise voidPromise = new VoidChannelPromise(this, true);
private final VoidChannelPromise unsafeVoidPromise = new VoidChannelPromise(this, false);
private final CloseFuture closeFuture = new CloseFuture(this);
private volatile SocketAddress localAddress;
private volatile SocketAddress remoteAddress;
private volatile EventLoop eventLoop;
private volatile boolean registered;
private ClosedChannelException closedChannelException;
private boolean inFlushNow;
private boolean flushNowPending;
/** Cache for the string representation of this channel */
private boolean strValActive;
private String strVal;
2008-09-02 09:13:20 +02:00
/**
* Creates a new instance.
*
* @param id
* the unique non-negative integer ID of this channel.
* Specify {@code null} to auto-generate a unique negative integer
* ID.
2008-09-02 09:13:20 +02:00
* @param parent
* the parent of this channel. {@code null} if there's no parent.
*/
protected AbstractChannel(Channel parent, Integer id) {
if (id == null) {
id = allocateId(this);
} else {
if (id.intValue() < 0) {
throw new IllegalArgumentException("id: " + id + " (expected: >= 0)");
}
if (allChannels.putIfAbsent(id, this) != null) {
throw new IllegalArgumentException("duplicate ID: " + id);
}
}
this.parent = parent;
this.id = id;
unsafe = newUnsafe();
pipeline = new DefaultChannelPipeline(this);
closeFuture().addListener(new ChannelFutureListener() {
@Override
public void operationComplete(ChannelFuture future) {
allChannels.remove(id());
}
});
}
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
@Override
public boolean isWritable() {
return outboundBuffer.getWritable();
}
@Override
public final Integer id() {
return id;
}
@Override
public Channel parent() {
return parent;
}
@Override
public ChannelPipeline pipeline() {
return pipeline;
}
ByteBufAllocator API w/ ByteBuf perf improvements This commit introduces a new API for ByteBuf allocation which fixes issue #643 along with refactoring of ByteBuf for simplicity and better performance. (see #62) A user can configure the ByteBufAllocator of a Channel via ChannelOption.ALLOCATOR or ChannelConfig.get/setAllocator(). The default allocator is currently UnpooledByteBufAllocator.HEAP_BY_DEFAULT. To allocate a buffer, do not use Unpooled anymore. do the following: ctx.alloc().buffer(...); // allocator chooses the buffer type. ctx.alloc().heapBuffer(...); ctx.alloc().directBuffer(...); To deallocate a buffer, use the unsafe free() operation: ((UnsafeByteBuf) buf).free(); The following is the list of the relevant changes: - Add ChannelInboundHandler.freeInboundBuffer() and ChannelOutboundHandler.freeOutboundBuffer() to let a user free the buffer he or she allocated. ChannelHandler adapter classes implement is already, so most users won't need to call free() by themselves. freeIn/OutboundBuffer() methods are invoked when a Channel is closed and deregistered. - All ByteBuf by contract must implement UnsafeByteBuf. To access an unsafe operation: ((UnsafeByteBuf) buf).internalNioBuffer() - Replace WrappedByteBuf and ByteBuf.Unsafe with UnsafeByteBuf to simplify overall class hierarchy and to avoid unnecesary instantiation of Unsafe instances on an unsafe operation. - Remove buffer reference counting which is confusing - Instantiate SwappedByteBuf lazily to avoid instantiation cost - Rename ChannelFutureFactory to ChannelPropertyAccess and move common methods between Channel and ChannelHandlerContext there. Also made it package-private to hide it from a user. - Remove unused unsafe operations such as newBuffer() - Add DetectionUtil.canFreeDirectBuffer() so that an allocator decides which buffer type to use safely
2012-11-15 22:04:37 +01:00
@Override
public ByteBufAllocator alloc() {
return config().getAllocator();
}
@Override
public EventLoop eventLoop() {
2012-06-02 03:34:19 +02:00
EventLoop eventLoop = this.eventLoop;
if (eventLoop == null) {
throw new IllegalStateException("channel not registered to an event loop");
}
return eventLoop;
}
@Override
public SocketAddress localAddress() {
SocketAddress localAddress = this.localAddress;
if (localAddress == null) {
try {
this.localAddress = localAddress = unsafe().localAddress();
} catch (Throwable t) {
// Sometimes fails on a closed socket in Windows.
return null;
}
}
return localAddress;
}
protected void invalidateLocalAddress() {
localAddress = null;
}
@Override
public SocketAddress remoteAddress() {
SocketAddress remoteAddress = this.remoteAddress;
if (remoteAddress == null) {
try {
this.remoteAddress = remoteAddress = unsafe().remoteAddress();
} catch (Throwable t) {
// Sometimes fails on a closed socket in Windows.
return null;
}
}
return remoteAddress;
}
/**
* Reset the stored remoteAddress
*/
protected void invalidateRemoteAddress() {
remoteAddress = null;
}
@Override
public boolean isRegistered() {
return registered;
}
@Override
public ChannelFuture bind(SocketAddress localAddress) {
return pipeline.bind(localAddress);
}
@Override
public ChannelFuture connect(SocketAddress remoteAddress) {
return pipeline.connect(remoteAddress);
}
@Override
public ChannelFuture connect(SocketAddress remoteAddress, SocketAddress localAddress) {
return pipeline.connect(remoteAddress, localAddress);
}
@Override
public ChannelFuture disconnect() {
return pipeline.disconnect();
}
@Override
public ChannelFuture close() {
return pipeline.close();
}
@Override
public ChannelFuture deregister() {
return pipeline.deregister();
}
@Override
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
public ChannelFuture write(Object msg) {
return pipeline.write(msg);
}
@Override
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
public ChannelFuture write(MessageList<?> msgs) {
return pipeline.write(msgs);
}
@Override
public ChannelFuture bind(SocketAddress localAddress, ChannelPromise promise) {
return pipeline.bind(localAddress, promise);
}
@Override
public ChannelFuture connect(SocketAddress remoteAddress, ChannelPromise promise) {
return pipeline.connect(remoteAddress, promise);
}
@Override
public ChannelFuture connect(SocketAddress remoteAddress, SocketAddress localAddress, ChannelPromise promise) {
return pipeline.connect(remoteAddress, localAddress, promise);
}
@Override
public ChannelFuture disconnect(ChannelPromise promise) {
return pipeline.disconnect(promise);
}
@Override
public ChannelFuture close(ChannelPromise promise) {
return pipeline.close(promise);
}
@Override
public ChannelFuture deregister(ChannelPromise promise) {
return pipeline.deregister(promise);
}
Read only when requested (read-on-demand) This pull request introduces a new operation called read() that replaces the existing inbound traffic control method. EventLoop now performs socket reads only when the read() operation has been issued. Once the requested read() operation is actually performed, EventLoop triggers an inboundBufferSuspended event that tells the handlers that the requested read() operation has been performed and the inbound traffic has been suspended again. A handler can decide to continue reading or not. Unlike other outbound operations, read() does not use ChannelFuture at all to avoid GC cost. If there's a good reason to create a new future per read at the GC cost, I'll change this. This pull request consequently removes the readable property in ChannelHandlerContext, which means how the traffic control works changed significantly. This pull request also adds a new configuration property ChannelOption.AUTO_READ whose default value is true. If true, Netty will call ctx.read() for you. If you need a close control over when read() is called, you can set it to false. Another interesting fact is that non-terminal handlers do not really need to call read() at all. Only the last inbound handler will have to call it, and that's just enough. Actually, you don't even need to call it at the last handler in most cases because of the ChannelOption.AUTO_READ mentioned above. There's no serious backward compatibility issue. If the compiler complains your handler does not implement the read() method, add the following: public void read(ChannelHandlerContext ctx) throws Exception { ctx.read(); } Note that this pull request certainly makes bounded inbound buffer support very easy, but itself does not add the bounded inbound buffer support.
2012-12-30 13:53:59 +01:00
@Override
public void read() {
pipeline.read();
}
@Override
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
public ChannelFuture write(Object msg, ChannelPromise promise) {
return pipeline.write(msg, promise);
}
@Override
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
public ChannelFuture write(MessageList<?> msgs, ChannelPromise promise) {
return pipeline.write(msgs, promise);
}
@Override
public ChannelPromise newPromise() {
return new DefaultChannelPromise(this);
}
@Override
public ChannelProgressivePromise newProgressivePromise() {
return new DefaultChannelProgressivePromise(this);
}
@Override
public ChannelFuture newSucceededFuture() {
return succeededFuture;
}
@Override
public ChannelFuture newFailedFuture(Throwable cause) {
return new FailedChannelFuture(this, null, cause);
}
@Override
public ChannelFuture closeFuture() {
return closeFuture;
}
@Override
public Unsafe unsafe() {
return unsafe;
}
/**
* Create a new {@link AbstractUnsafe} instance which will be used for the life-time of the {@link Channel}
*/
protected abstract AbstractUnsafe newUnsafe();
/**
* Returns the ID of this channel.
*/
@Override
public final int hashCode() {
return id;
}
/**
* Returns {@code true} if and only if the specified object is identical
* with this channel (i.e: {@code this == o}).
*/
@Override
public final boolean equals(Object o) {
return this == o;
}
/**
* Compares the {@linkplain #id() ID} of the two channels.
*/
@Override
public final int compareTo(Channel o) {
return id().compareTo(o.id());
}
2008-09-02 09:13:20 +02:00
/**
* Returns the {@link String} representation of this channel. The returned
* string contains the {@linkplain #id() ID}, {@linkplain #localAddress() local address},
* and {@linkplain #remoteAddress() remote address} of this channel for
2008-09-02 09:13:20 +02:00
* easier identification.
*/
@Override
public String toString() {
boolean active = isActive();
if (strValActive == active && strVal != null) {
return strVal;
}
2012-05-11 03:47:45 +02:00
SocketAddress remoteAddr = remoteAddress();
SocketAddress localAddr = localAddress();
if (remoteAddr != null) {
SocketAddress srcAddr;
SocketAddress dstAddr;
if (parent == null) {
srcAddr = localAddr;
dstAddr = remoteAddr;
} else {
2012-05-11 03:47:45 +02:00
srcAddr = remoteAddr;
dstAddr = localAddr;
}
2012-05-11 03:47:45 +02:00
strVal = String.format("[id: 0x%08x, %s %s %s]", id, srcAddr, active? "=>" : ":>", dstAddr);
} else if (localAddr != null) {
strVal = String.format("[id: 0x%08x, %s]", id, localAddr);
} else {
strVal = String.format("[id: 0x%08x]", id);
}
strValActive = active;
return strVal;
}
2009-02-16 15:32:40 +01:00
@Override
public final ChannelPromise voidPromise() {
return voidPromise;
}
/**
* {@link Unsafe} implementation which sub-classes must extend and use.
*/
protected abstract class AbstractUnsafe implements Unsafe {
private final Runnable flushLaterTask = new Runnable() {
@Override
public void run() {
flushNowPending = false;
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
flush();
}
};
@Override
public final SocketAddress localAddress() {
return localAddress0();
}
@Override
public final SocketAddress remoteAddress() {
return remoteAddress0();
}
@Override
public final void register(EventLoop eventLoop, final ChannelPromise promise) {
if (eventLoop == null) {
throw new NullPointerException("eventLoop");
}
if (isRegistered()) {
promise.setFailure(new IllegalStateException("registered to an event loop already"));
return;
}
if (!isCompatible(eventLoop)) {
2013-03-21 11:02:34 +01:00
promise.setFailure(
new IllegalStateException("incompatible event loop type: " + eventLoop.getClass().getName()));
return;
}
AbstractChannel.this.eventLoop = eventLoop;
if (eventLoop.inEventLoop()) {
register0(promise);
} else {
try {
eventLoop.execute(new Runnable() {
@Override
public void run() {
register0(promise);
}
});
} catch (Throwable t) {
logger.warn(
"Force-closing a channel whose registration task was unaccepted by an event loop: {}",
AbstractChannel.this, t);
closeForcibly();
promise.setFailure(t);
}
}
}
private void register0(ChannelPromise promise) {
try {
// check if the channel is still open as it could be closed in the mean time when the register
// call was outside of the eventLoop
if (!ensureOpen(promise)) {
return;
}
Runnable postRegisterTask = doRegister();
registered = true;
promise.setSuccess();
pipeline.fireChannelRegistered();
if (postRegisterTask != null) {
postRegisterTask.run();
}
if (isActive()) {
pipeline.fireChannelActive();
}
} catch (Throwable t) {
// Close the channel directly to avoid FD leak.
closeForcibly();
if (!promise.tryFailure(t)) {
logger.warn(
"Tried to fail the registration promise, but it is complete already. " +
"Swallowing the cause of the registration failure:", t);
}
closeFuture.setClosed();
}
}
@Override
public final void bind(final SocketAddress localAddress, final ChannelPromise promise) {
if (!ensureOpen(promise)) {
return;
}
try {
boolean wasActive = isActive();
// See: https://github.com/netty/netty/issues/576
if (!PlatformDependent.isWindows() && !PlatformDependent.isRoot() &&
Boolean.TRUE.equals(config().getOption(ChannelOption.SO_BROADCAST)) &&
localAddress instanceof InetSocketAddress &&
!((InetSocketAddress) localAddress).getAddress().isAnyLocalAddress()) {
// Warn a user about the fact that a non-root user can't receive a
// broadcast packet on *nix if the socket is bound on non-wildcard address.
logger.warn(
"A non-root user can't receive a broadcast packet if the socket " +
"is not bound to a wildcard address; binding to a non-wildcard " +
"address (" + localAddress + ") anyway as requested.");
}
doBind(localAddress);
promise.setSuccess();
if (!wasActive && isActive()) {
pipeline.fireChannelActive();
}
} catch (Throwable t) {
promise.setFailure(t);
closeIfClosed();
}
}
@Override
public final void disconnect(final ChannelPromise promise) {
try {
boolean wasActive = isActive();
doDisconnect();
promise.setSuccess();
if (wasActive && !isActive()) {
invokeLater(new Runnable() {
@Override
public void run() {
pipeline.fireChannelInactive();
}
});
}
} catch (Throwable t) {
promise.setFailure(t);
closeIfClosed();
}
}
@Override
public final void close(final ChannelPromise promise) {
boolean wasActive = isActive();
if (closeFuture.setClosed()) {
try {
doClose();
promise.setSuccess();
} catch (Throwable t) {
promise.setFailure(t);
}
if (closedChannelException == null) {
closedChannelException = new ClosedChannelException();
}
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
// fail all queued messages
if (outboundBuffer.next()) {
outboundBuffer.fail(closedChannelException);
}
if (wasActive && !isActive()) {
invokeLater(new Runnable() {
@Override
public void run() {
pipeline.fireChannelInactive();
}
});
}
deregister(voidPromise());
} else {
// Closed already.
promise.setSuccess();
}
}
@Override
public final void closeForcibly() {
try {
doClose();
} catch (Exception e) {
logger.warn("Failed to close a channel.", e);
}
}
@Override
public final void deregister(final ChannelPromise promise) {
if (!registered) {
promise.setSuccess();
return;
}
Runnable postTask = null;
try {
postTask = doDeregister();
} catch (Throwable t) {
logger.warn("Unexpected exception occurred while deregistering a channel.", t);
} finally {
if (registered) {
registered = false;
promise.setSuccess();
invokeLater(new Runnable() {
@Override
public void run() {
pipeline.fireChannelUnregistered();
}
});
} else {
// Some transports like local and AIO does not allow the deregistration of
// an open channel. Their doDeregister() calls close(). Consequently,
// close() calls deregister() again - no need to fire channelUnregistered.
promise.setSuccess();
}
if (postTask != null) {
postTask.run();
}
}
}
Read only when requested (read-on-demand) This pull request introduces a new operation called read() that replaces the existing inbound traffic control method. EventLoop now performs socket reads only when the read() operation has been issued. Once the requested read() operation is actually performed, EventLoop triggers an inboundBufferSuspended event that tells the handlers that the requested read() operation has been performed and the inbound traffic has been suspended again. A handler can decide to continue reading or not. Unlike other outbound operations, read() does not use ChannelFuture at all to avoid GC cost. If there's a good reason to create a new future per read at the GC cost, I'll change this. This pull request consequently removes the readable property in ChannelHandlerContext, which means how the traffic control works changed significantly. This pull request also adds a new configuration property ChannelOption.AUTO_READ whose default value is true. If true, Netty will call ctx.read() for you. If you need a close control over when read() is called, you can set it to false. Another interesting fact is that non-terminal handlers do not really need to call read() at all. Only the last inbound handler will have to call it, and that's just enough. Actually, you don't even need to call it at the last handler in most cases because of the ChannelOption.AUTO_READ mentioned above. There's no serious backward compatibility issue. If the compiler complains your handler does not implement the read() method, add the following: public void read(ChannelHandlerContext ctx) throws Exception { ctx.read(); } Note that this pull request certainly makes bounded inbound buffer support very easy, but itself does not add the bounded inbound buffer support.
2012-12-30 13:53:59 +01:00
@Override
public void beginRead() {
2013-01-09 13:28:31 +01:00
if (!isActive()) {
return;
}
try {
doBeginRead();
} catch (final Exception e) {
invokeLater(new Runnable() {
@Override
public void run() {
pipeline.fireExceptionCaught(e);
}
});
close(voidPromise());
Read only when requested (read-on-demand) This pull request introduces a new operation called read() that replaces the existing inbound traffic control method. EventLoop now performs socket reads only when the read() operation has been issued. Once the requested read() operation is actually performed, EventLoop triggers an inboundBufferSuspended event that tells the handlers that the requested read() operation has been performed and the inbound traffic has been suspended again. A handler can decide to continue reading or not. Unlike other outbound operations, read() does not use ChannelFuture at all to avoid GC cost. If there's a good reason to create a new future per read at the GC cost, I'll change this. This pull request consequently removes the readable property in ChannelHandlerContext, which means how the traffic control works changed significantly. This pull request also adds a new configuration property ChannelOption.AUTO_READ whose default value is true. If true, Netty will call ctx.read() for you. If you need a close control over when read() is called, you can set it to false. Another interesting fact is that non-terminal handlers do not really need to call read() at all. Only the last inbound handler will have to call it, and that's just enough. Actually, you don't even need to call it at the last handler in most cases because of the ChannelOption.AUTO_READ mentioned above. There's no serious backward compatibility issue. If the compiler complains your handler does not implement the read() method, add the following: public void read(ChannelHandlerContext ctx) throws Exception { ctx.read(); } Note that this pull request certainly makes bounded inbound buffer support very easy, but itself does not add the bounded inbound buffer support.
2012-12-30 13:53:59 +01:00
}
}
@Override
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
public void write(MessageList<?> msgs, ChannelPromise promise) {
outboundBuffer.add(msgs, promise);
flush();
}
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
private void flush() {
if (!inFlushNow) { // Avoid re-entrance
try {
// Flush immediately only when there's no pending flush.
// If there's a pending flush operation, event loop will call flushNow() later,
// and thus there's no need to call it now.
if (!isFlushPending()) {
flushNow();
}
} catch (Throwable t) {
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
outboundBuffer.fail(t);
close(voidPromise());
}
} else {
if (!flushNowPending) {
flushNowPending = true;
eventLoop().execute(flushLaterTask);
}
}
}
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
@Override
public final void flushNow() {
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
if (inFlushNow) {
return;
}
inFlushNow = true;
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
final ChannelOutboundBuffer outboundBuffer = AbstractChannel.this.outboundBuffer;
try {
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
for (;;) {
ChannelPromise promise = outboundBuffer.currentPromise;
if (promise == null) {
if (!outboundBuffer.next()) {
break;
}
promise = outboundBuffer.currentPromise;
}
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
MessageList<Object> messages = outboundBuffer.currentMessages;
int messageIndex = outboundBuffer.currentMessageIndex;
int messageCount = messages.size();
if (messageCount == 0) {
messages.recycle();
promise.trySuccess();
if (!outboundBuffer.next()) {
break;
} else {
continue;
}
}
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
int writtenMessages = doWrite(messages, messageIndex);
outboundBuffer.currentMessageIndex = messageIndex += writtenMessages;
if (messageIndex >= messageCount) {
messages.recycle();
promise.trySuccess();
if (!outboundBuffer.next()) {
break;
}
} else {
// Could not flush the current write request completely. Try again later.
break;
}
}
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
} catch (Throwable t) {
outboundBuffer.fail(t);
if (t instanceof IOException) {
close(voidPromise());
}
} finally {
inFlushNow = false;
}
}
@Override
public ChannelPromise voidPromise() {
return unsafeVoidPromise;
}
protected final boolean ensureOpen(ChannelPromise promise) {
if (isOpen()) {
return true;
}
Exception e = new ClosedChannelException();
promise.setFailure(e);
return false;
}
protected final void closeIfClosed() {
if (isOpen()) {
return;
}
close(voidPromise());
}
private void invokeLater(Runnable task) {
// This method is used by outbound operation implementations to trigger an inbound event later.
// They do not trigger an inbound event immediately because an outbound operation might have been
// triggered by another inbound event handler method. If fired immediately, the call stack
// will look like this for example:
//
// handlerA.inboundBufferUpdated() - (1) an inbound handler method closes a connection.
// -> handlerA.ctx.close()
// -> channel.unsafe.close()
// -> handlerA.channelInactive() - (2) another inbound handler method called while in (1) yet
//
// which means the execution of two inbound handler methods of the same handler overlap undesirably.
eventLoop().execute(task);
}
}
/**
* Return {@code true} if the given {@link EventLoop} is compatible with this instance.
*/
protected abstract boolean isCompatible(EventLoop loop);
/**
* Returns the {@link SocketAddress} which is bound locally.
*/
protected abstract SocketAddress localAddress0();
/**
* Return the {@link SocketAddress} which the {@link Channel} is connected to.
*/
protected abstract SocketAddress remoteAddress0();
/**
* Is called after the {@link Channel} is registered with its {@link EventLoop} as part of the register process.
* You can return a {@link Runnable} which will be run as post-task of the registration process.
*
* Sub-classes may override this method as it will just return {@code null}
*/
protected Runnable doRegister() throws Exception {
return null;
}
/**
* Bind the {@link Channel} to the {@link SocketAddress}
*/
protected abstract void doBind(SocketAddress localAddress) throws Exception;
/**
* Disconnect this {@link Channel} from its remote peer
*/
protected abstract void doDisconnect() throws Exception;
/**
* Will be called before the actual close operation will be performed. Sub-classes may override this as the default
* is to do nothing.
*/
protected void doPreClose() throws Exception {
// NOOP by default
}
/**
* Close the {@link Channel}
*/
protected abstract void doClose() throws Exception;
/**
* Deregister the {@link Channel} from its {@link EventLoop}.
* You can return a {@link Runnable} which will be run as post-task of the registration process.
*
* Sub-classes may override this method
*/
protected Runnable doDeregister() throws Exception {
return null;
}
Read only when requested (read-on-demand) This pull request introduces a new operation called read() that replaces the existing inbound traffic control method. EventLoop now performs socket reads only when the read() operation has been issued. Once the requested read() operation is actually performed, EventLoop triggers an inboundBufferSuspended event that tells the handlers that the requested read() operation has been performed and the inbound traffic has been suspended again. A handler can decide to continue reading or not. Unlike other outbound operations, read() does not use ChannelFuture at all to avoid GC cost. If there's a good reason to create a new future per read at the GC cost, I'll change this. This pull request consequently removes the readable property in ChannelHandlerContext, which means how the traffic control works changed significantly. This pull request also adds a new configuration property ChannelOption.AUTO_READ whose default value is true. If true, Netty will call ctx.read() for you. If you need a close control over when read() is called, you can set it to false. Another interesting fact is that non-terminal handlers do not really need to call read() at all. Only the last inbound handler will have to call it, and that's just enough. Actually, you don't even need to call it at the last handler in most cases because of the ChannelOption.AUTO_READ mentioned above. There's no serious backward compatibility issue. If the compiler complains your handler does not implement the read() method, add the following: public void read(ChannelHandlerContext ctx) throws Exception { ctx.read(); } Note that this pull request certainly makes bounded inbound buffer support very easy, but itself does not add the bounded inbound buffer support.
2012-12-30 13:53:59 +01:00
/**
* Schedule a read operation.
*/
protected abstract void doBeginRead() throws Exception;
/**
* Flush the content of the given {@link ByteBuf} to the remote peer.
*
* Sub-classes may override this as this implementation will just thrown an {@link UnsupportedOperationException}
*
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
* @return the number of written messages
*/
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
protected abstract int doWrite(MessageList<Object> msgs, int index) throws Exception;
protected static void checkEOF(FileRegion region) throws IOException {
if (region.transfered() < region.count()) {
throw new EOFException("Expected to be able to write "
+ region.count() + " bytes, but only wrote "
+ region.transfered());
}
}
/**
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
* Calculate the number of bytes a message takes up in memory. Sub-classes may override this if they use different
* messages then {@link ByteBuf} or {@link ByteBufHolder}. If the size can not be calculated 0 should be returned.
*/
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
protected int calculateMessageSize(Object message) {
if (message instanceof ByteBuf) {
return ((ByteBuf) message).readableBytes();
}
if (message instanceof ByteBufHolder) {
return ((ByteBufHolder) message).content().readableBytes();
}
Revamp the core API to reduce memory footprint and consumption The API changes made so far turned out to increase the memory footprint and consumption while our intention was actually decreasing them. Memory consumption issue: When there are many connections which does not exchange data frequently, the old Netty 4 API spent a lot more memory than 3 because it always allocates per-handler buffer for each connection unless otherwise explicitly stated by a user. In a usual real world load, a client doesn't always send requests without pausing, so the idea of having a buffer whose life cycle if bound to the life cycle of a connection didn't work as expected. Memory footprint issue: The old Netty 4 API decreased overall memory footprint by a great deal in many cases. It was mainly because the old Netty 4 API did not allocate a new buffer and event object for each read. Instead, it created a new buffer for each handler in a pipeline. This works pretty well as long as the number of handlers in a pipeline is only a few. However, for a highly modular application with many handlers which handles connections which lasts for relatively short period, it actually makes the memory footprint issue much worse. Changes: All in all, this is about retaining all the good changes we made in 4 so far such as better thread model and going back to the way how we dealt with message events in 3. To fix the memory consumption/footprint issue mentioned above, we made a hard decision to break the backward compatibility again with the following changes: - Remove MessageBuf - Merge Buf into ByteBuf - Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler - Similar changes were made to the adapter classes - Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler - Similar changes were made to the adapter classes - Introduce MessageList which is similar to `MessageEvent` in Netty 3 - Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList) - Replace flush(ctx, promise) with write(ctx, MessageList, promise) - Remove ByteToByteEncoder/Decoder/Codec - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf> - Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel - Add SimpleChannelInboundHandler which is sometimes more useful than ChannelInboundHandlerAdapter - Bring back Channel.isWritable() from Netty 3 - Add ChannelInboundHandler.channelWritabilityChanges() event - Add RecvByteBufAllocator configuration property - Similar to ReceiveBufferSizePredictor in Netty 3 - Some existing configuration properties such as DatagramChannelConfig.receivePacketSize is gone now. - Remove suspend/resumeIntermediaryDeallocation() in ByteBuf This change would have been impossible without @normanmaurer's help. He fixed, ported, and improved many parts of the changes.
2013-05-28 13:40:19 +02:00
return 0;
}
/**
* Return {@code true} if a flush to the {@link Channel} is currently pending.
*/
protected abstract boolean isFlushPending();
final class CloseFuture extends DefaultChannelPromise {
CloseFuture(AbstractChannel ch) {
super(ch);
}
@Override
public ChannelPromise setSuccess() {
throw new IllegalStateException();
}
@Override
public ChannelPromise setFailure(Throwable cause) {
throw new IllegalStateException();
}
@Override
public boolean trySuccess() {
throw new IllegalStateException();
}
@Override
public boolean tryFailure(Throwable cause) {
throw new IllegalStateException();
}
boolean setClosed() {
try {
doPreClose();
} catch (Exception e) {
logger.warn("doPreClose() raised an exception.", e);
}
return super.trySuccess();
}
}
}