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

300 lines
9.7 KiB
Java
Raw Normal View History

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
/*
* Copyright 2013 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:
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* 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
* License for the specific language governing permissions and limitations
* under the License.
*/
/*
* Written by Josh Bloch of Google Inc. and released to the public domain,
* as explained at http://creativecommons.org/publicdomain/zero/1.0/.
*/
package io.netty.channel;
import io.netty.util.ReferenceCountUtil;
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.util.internal.logging.InternalLogger;
import io.netty.util.internal.logging.InternalLoggerFactory;
import java.util.concurrent.atomic.AtomicIntegerFieldUpdater;
final class ChannelOutboundBuffer {
private static final InternalLogger logger = InternalLoggerFactory.getInstance(ChannelOutboundBuffer.class);
private static final int MIN_INITIAL_CAPACITY = 8;
MessageList currentMessageList;
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 currentMessageIndex;
private long currentMessageListSize;
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 MessageList[] messageLists;
private long[] messageListSizes;
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 MessageList unflushedMessageList;
private long unflushedMessageListSize;
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 int head;
private int tail;
private boolean inFail;
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 AbstractChannel channel;
private long pendingOutboundBytes;
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 static final AtomicIntegerFieldUpdater<ChannelOutboundBuffer> WRITABLE_UPDATER =
AtomicIntegerFieldUpdater.newUpdater(ChannelOutboundBuffer.class, "writable");
@SuppressWarnings("unused")
private volatile int writable = 1;
ChannelOutboundBuffer(AbstractChannel channel) {
this(channel, MIN_INITIAL_CAPACITY << 1);
}
@SuppressWarnings("unchecked")
ChannelOutboundBuffer(AbstractChannel channel, int initialCapacity) {
if (initialCapacity < 0) {
throw new IllegalArgumentException("initialCapacity: " + initialCapacity + " (expected: >= 0)");
}
// Find the best power of two to hold elements.
// Tests "<=" because arrays aren't kept full.
if (initialCapacity >= MIN_INITIAL_CAPACITY) {
initialCapacity |= initialCapacity >>> 1;
initialCapacity |= initialCapacity >>> 2;
initialCapacity |= initialCapacity >>> 4;
initialCapacity |= initialCapacity >>> 8;
initialCapacity |= initialCapacity >>> 16;
initialCapacity ++;
if (initialCapacity < 0) { // Too many elements, must back off
initialCapacity >>>= 1; // Good luck allocating 2 ^ 30 elements
}
} else {
initialCapacity = MIN_INITIAL_CAPACITY;
}
this.channel = channel;
messageLists = new MessageList[initialCapacity];
messageListSizes = new long[initialCapacity];
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
}
void addMessage(Object msg, ChannelPromise promise) {
MessageList unflushedMessageList = this.unflushedMessageList;
if (unflushedMessageList == null) {
this.unflushedMessageList = unflushedMessageList = MessageList.newInstance();
}
unflushedMessageList.add(msg, promise);
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 size = channel.calculateMessageSize(msg);
unflushedMessageListSize += size;
incrementPendingOutboundBytes(size);
}
void addFlush() {
MessageList unflushedMessageList = this.unflushedMessageList;
if (unflushedMessageList == null) {
return;
}
int tail = this.tail;
messageLists[tail] = unflushedMessageList;
messageListSizes[tail] = unflushedMessageListSize;
this.unflushedMessageList = null;
unflushedMessageListSize = 0;
if ((this.tail = (tail + 1) & (messageLists.length - 1)) == head) {
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
doubleCapacity();
}
}
private void incrementPendingOutboundBytes(int size) {
if (size == 0) {
return;
}
long newWriteBufferSize = pendingOutboundBytes += size;
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 highWaterMark = channel.config().getWriteBufferHighWaterMark();
if (newWriteBufferSize > highWaterMark) {
if (WRITABLE_UPDATER.compareAndSet(this, 1, 0)) {
channel.pipeline().fireChannelWritabilityChanged();
}
}
}
private void decrementPendingOutboundBytes(long size) {
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 (size == 0) {
return;
}
long newWriteBufferSize = pendingOutboundBytes -= size;
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 lowWaterMark = channel.config().getWriteBufferLowWaterMark();
if (newWriteBufferSize == 0 || newWriteBufferSize < lowWaterMark) {
if (WRITABLE_UPDATER.compareAndSet(this, 0, 1)) {
channel.pipeline().fireChannelWritabilityChanged();
}
}
}
private void doubleCapacity() {
assert head == tail;
int p = head;
int n = messageLists.length;
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 r = n - p; // number of elements to the right of p
int newCapacity = n << 1;
if (newCapacity < 0) {
throw new IllegalStateException("Sorry, deque too big");
}
@SuppressWarnings("unchecked")
MessageList[] a1 = new MessageList[newCapacity];
System.arraycopy(messageLists, p, a1, 0, r);
System.arraycopy(messageLists, 0, a1, r, p);
messageLists = a1;
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
long[] a2 = new long[newCapacity];
System.arraycopy(messageListSizes, p, a2, 0, r);
System.arraycopy(messageListSizes, 0, a2, r, p);
messageListSizes = a2;
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
head = 0;
tail = n;
}
boolean next() {
// FIXME: pendingOutboundBytes should be decreased when the messages are flushed.
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
decrementPendingOutboundBytes(currentMessageListSize);
int h = head;
MessageList e = messageLists[h]; // Element is null if deque empty
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 (e == null) {
currentMessageListSize = 0;
currentMessageList = null;
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 false;
}
currentMessageList = messageLists[h];
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
currentMessageIndex = 0;
currentMessageListSize = messageListSizes[h];
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
messageLists[h] = null;
messageListSizes[h] = 0;
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
head = h + 1 & messageLists.length - 1;
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 true;
}
boolean getWritable() {
return WRITABLE_UPDATER.get(this) == 1;
}
int size() {
return tail - head & messageLists.length - 1;
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
}
boolean isEmpty() {
return head == tail;
}
void clearUnflushed(Throwable cause) {
if (inFail) {
return;
}
MessageList unflushed = unflushedMessageList;
if (unflushed == null) {
return;
}
inFail = true;
// Release all unflushed messages.
Object[] messages = unflushed.messages();
ChannelPromise[] promises = unflushed.promises();
final int size = unflushed.size();
try {
for (int i = 0; i < size; i++) {
2013-07-18 03:29:34 +02:00
safeRelease(messages[i], promises[i], cause);
}
} finally {
unflushed.recycle();
decrementPendingOutboundBytes(unflushedMessageListSize);
unflushedMessageListSize = 0;
inFail = 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
}
}
void fail(Throwable cause) {
// Make sure that this method does not reenter. A listener added to the current promise can be notified by the
// current thread in the tryFailure() call of the loop below, and the listener can trigger another fail() call
// indirectly (usually by closing the channel.)
//
// See https://github.com/netty/netty/issues/1501
if (inFail) {
return;
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
}
try {
inFail = true;
if (currentMessageList == null) {
if (!next()) {
return;
}
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
}
do {
if (currentMessageList != null) {
// Store a local reference of current messages
// This is needed as a promise may have a listener attached that will close the channel
// The close will call next() which will set currentMessages to null and so
// trigger a NPE in the finally block if no local reference is used.
//
// See https://github.com/netty/netty/issues/1573
MessageList current = currentMessageList;
// Release all failed messages.
Object[] messages = current.messages();
ChannelPromise[] promises = current.promises();
final int size = current.size();
try {
for (int i = currentMessageIndex; i < size; i++) {
2013-07-18 03:29:34 +02:00
safeRelease(messages[i], promises[i], cause);
}
} finally {
current.recycle();
}
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
}
} while(next());
} finally {
inFail = 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
}
2013-07-18 03:29:34 +02:00
private static void safeRelease(Object message, ChannelPromise promise, Throwable cause) {
try {
ReferenceCountUtil.release(message);
} catch (Throwable t) {
logger.warn("Failed to release a message.", t);
}
2013-07-18 03:29:34 +02:00
ChannelPromise p = promise;
if (!(p instanceof VoidChannelPromise) && !p.tryFailure(cause)) {
logger.warn("Promise done already: {} - new exception is:", p, cause);
}
}
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
}