Motivation:
The writeSpinCount currently loops over the same buffer, gathering
write, file write, or other write operation multiple times but will
continue writing until there is nothing left or the OS doesn't accept
any data for that specific write. However if the OS keeps accepting
writes there is no way to limit how much time we spend on a specific
socket. This can lead to unfair consumption of resources dedicated to a
single socket.
We currently don't limit the amount of bytes we attempt to write per
gathering write. If there are many more bytes pending relative to the
SO_SNDBUF size we will end up building iov arrays with more elements
than can be written, which results in extra iteration, conditionals,
and book keeping.
Modifications:
- writeSpinCount should limit the number of system calls we make to
write data, instead of applying to individual write operations
- IovArray should support a maximum number of bytes
- IovArray should support composite buffers of greater than size 1024
- We should auto-scale the amount of data that we attempt to write per
gathering write operation relative to SO_SNDBUF and how much data is
successfully written
- The non-unsafe path should also support a maximum number of bytes,
and respect the IOV_MAX limit
Result:
Write resource consumption can be bounded and gathering writes have
a limit relative to the amount of data which can actually be accepted
by the socket.
Motivation:
If large amounts of data is being transferred it is difficult to correlate the amount we attempt to read vs the maximum amount that the OS will actually buffer and deliver to the application. For exmaple some OSes may dynicamlly update the SO_RCVBUF size or otherwise dynamically adjust how much data is delieved to the application. In these circumstances it can reduce latency to just call read() on the socket another time to see if there is really any data remaining instead of giving up the maxMessagesPerRead quantum and going back to the selector to read later.
Motifications:
- Add DefaultMaxMessagesRecvByteBufAllocator#respectMaybeMoreData which provides a way to ignore the maybeMoreData function which may not account for the current data pending, and if it does this maybe racy.
Result:
Option to always use the full maxMessagesPerRead quantum before going back to the selector.
Motivation:
SslHandler will do aggregation of writes by default in an attempt to improve goodput and reduce the number of discrete buffers which must be accumulated. However if aggregation is not possible then a CompositeByteBuf is used to accumulate multiple buffers. Using a CompositeByteBuf doesn't provide any of the benefits of better goodput and in the case of small + large writes (e.g. http/2 frame header + data) this can reduce the amount of data that can be passed to writev by about half. This has the impact of increasing latency as well as reducing goodput.
Modifications:
- SslHandler should prefer copying instead of using a CompositeByteBuf
Result:
Better goodput (and potentially improved latency) at the cost of copy operations.
Motivation:
AdaptiveRecvByteBufAllocator currently adjusts the ByteBuf allocation size guess when readComplete is called. However the default configuration for number of reads before readComplete is called is 16. This means that there will be 16 reads done before any adjustment is done. If there is a large amount of data pending AdaptiveRecvByteBufAllocator will be slow to adjust the allocation size guess. In addition to being slow the result of only updating the guess in readComplete means that we must go back to the selector and wait to be woken up again when data is ready to read. Going back to the selector is an expensive operations and can add significant latency if there is large amount of data pending to read.
Modifications:
- AdaptiveRecvByteBufAllocator should check on each read if a step up is necessary. The step down process is left unchanged and can be more gradual at the cost of potentially over allocating.
Result:
AdaptiveRecvByteBufAllocator increases the guess size during the read loop to reduce latency when large amounts of data is being read.
Automatic-Module-Name entry provides a stable JDK9 module name, when Netty is used in a modular JDK9 applications. More info: http://blog.joda.org/2017/05/java-se-9-jpms-automatic-modules.html
When Netty migrates to JDK9 in the future, the entry can be replaced by actual module-info descriptor.
Modification:
The POM-s are configured to put the correct module names to the manifest.
Result:
Fixes#7218.
Motivation:
`FixedChannelPool` allows users to configure `acquireTimeoutMillis`
and expects given value to be greater or equal to zero when timeout
action is supplied. However, validation error message said that
value is expected to be greater or equal to one. Code performs
check against zero.
Modifications:
Changed error message to say that value greater or equal to
zero is expected. Added test to check that zero is an acceptable
value.
Result:
Exception with right error message is thrown.
Motivation:
AbstractCoalescingBufferQueue#add accounts for void promises, but AbstractCoalescingBufferQueue#addFirst does not. These methods should be consistent.
Modifications:
- AbstractCoalescingBufferQueue#addFirst should account for void promises and share code with AbstractCoalescingBufferQueue#add
Result:
More correct void promise handling in AbstractCoalescingBufferQueue.
complete
Motivation:
SslHandler removes a Buffer/Promise pair from
AbstractCoalescingBufferQueue when wrapping data. However it is possible
the SSLEngine will not consume the entire buffer. In this case
SslHandler adds the Buffer back to the queue, but doesn't add the
Promise back to the queue. This may result in the promise completing
immediately in finishFlush, and generally not correlating to the
completion of writing the corresponding Buffer
Modifications:
- AbstractCoalescingBufferQueue#addFirst should also support adding the
ChannelPromise
- In the event of a handshake timeout we should immediately fail pending
writes immediately to get a more accurate exception
Result:
Fixes https://github.com/netty/netty/issues/7378.
Motivation:
We need to set readPending to false when we detect a EOF while issue a read as otherwise we may not unregister from the Selector / Epoll / KQueue and so keep on receving wakeups.
The important bit is that we may even get a wakeup for a read event but will still will only be able to read 0 bytes from the socket, so we need to be very careful when we clear the readPending. This can happen because we generally using edge-triggered mode for our native transports and because of the nature of edge-triggered we may schedule an read event just to find out there is nothing left to read atm (because we completely drained the socket on the previous read).
Modifications:
Set readPending to false when EOF is detected.
Result:
Fixes [#7255].
Motivation:
HTTP/2 allows writes of 0 length data frames. However in some cases EMPTY_BUFFER is used instead of the actual buffer that was written. This may mask writes of released buffers or otherwise invalid buffer objects. It is also possible that if the buffer is invalid AbstractCoalescingBufferQueue will not release the aggregated buffer nor fail the associated promise.
Modifications:
- DefaultHttp2FrameCodec should take care to fail the promise, even if releasing the data throws
- AbstractCoalescingBufferQueue should release any aggregated data and fail the associated promise if something goes wrong during aggregation
Result:
More correct handling of invalid buffers in HTTP/2 code.
This reverts commit 413c7c2cd8 as it introduced an regression when edge-triggered mode is used which is true for our native transports by default. With 413c7c2cd8 included it was possible that we set readPending to false by mistake even if we would be interested in read more.
Motivation:
readPending is currently only set to false if data is delivered to the application, however this may result in duplicate events being received from the selector in the event that the socket was closed.
Modifications:
- We should set readPending to false before each read attempt for all
transports besides NIO.
- Based upon the Javadocs it is possible that NIO may have spurious
wakeups [1]. In this case we should be more cautious and only set
readPending to false if data was actually read.
[1] https://docs.oracle.com/javase/7/docs/api/java/nio/channels/SelectionKey.html
That a selection key's ready set indicates that its channel is ready for some operation category is a hint, but not a guarantee, that an operation in such a category may be performed by a thread without causing the thread to block.
Result:
Notification from the selector (or simulated events from kqueue/epoll ET) in the event of socket closure.
Fixes https://github.com/netty/netty/issues/7255
Motivation:
A regression was introduced in 86e653e which had the effect that the writability was not updated for a Channel while queueing data in the SslHandler.
Modifications:
- Factor out code that will increment / decrement pending bytes and use it in AbstractCoalescingBufferQueue and PendingWriteQueue
- Add test-case
Result:
Channel writability changes are triggered again.
Motivation:
Without a 'serialVersionUID' field, any change to a class will make
previously serialized versions unreadable.
Modifications:
Add missed 'serialVersionUID' field for all Serializable
classes.
Result:
Proper deserialization of previously serialized objects.
Motivation:
There are many @SuppressWarnings("unchecked") in the code for the same purpose that we want to do this return:
@SuppressWarnings("unchecked")
public B someMethod() {
......
return (B) this;
}
Modification:
Add a method self() and reuse in all these return lines:
@SuppressWarnings("unchecked")
private B self() {
return (B) this;
}
Result:
Then only one @SuppressWarnings("unchecked") left in the code.
Motivation:
When SO_LINGER is used we run doClose() on the GlobalEventExecutor by default so we need to ensure we schedule all code that needs to be run on the EventLoop on the EventLoop in doClose. Beside this there are also threading issues when calling shutdownOutput(...)
Modifications:
- Schedule removal from EventLoop to the EventLoop
- Correctly handle shutdownOutput and shutdown in respect with threading-model
- Add unit tests
Result:
Fixes [#7159].
Motivation:
A `DefaultChannelId` has final `hashCode` field calculated in the constructor. We can use it in `equals` to the fast return for different objects.
Modifications:
Use `hashCode` field in `DefaultChannelId.equals()`.
Result:
Fast `equals` on negative scenarios.
Motivation:
We should not log by default if the promise is a VoidChannelPromise as its try* methods will always return false.
Modifications:
Do an instanceof check to determine if we should log or not by default
Result:
No more noise in the logs when using a VoidChannelPromise.
Motivation:
If AutoClose is false and there is a IoException then AbstractChannel will not close the channel but instead just fail flushed element in the ChannelOutboundBuffer. AbstractChannel also notifies of writability changes, which may lead to an infinite loop if the peer has closed its read side of the socket because we will keep accepting more data but continuously fail because the peer isn't accepting writes.
Modifications:
- If the transport throws on a write we should acknowledge that the output side of the channel has been shutdown and cleanup. If the channel can't accept more data because it is full, and still healthy it is not expected to throw. However if the channel is not healthy it will throw and is not expected to accept any more writes. In this case we should shutdown the output for Channels that support this feature and otherwise just close.
- Connection-less protocols like UDP can remain the same because the channel may disconnected temporarily.
- Make sure AbstractUnsafe#shutdownOutput is called because the shutdown on the socket may throw an exception.
Result:
More correct handling of write failure when AutoClose is false.
Motivation:
ShutdownOutput now fails all pending writes in the ChannelOutboundBuffer and sets it to null. However the Close code path uses the ChannelOutboundBuffer as an indication that the close operation is in progress and exits early and will not call doClose. This will lead to the Channel not actually being fully closed.
Bug introduced by 237a4da1b7
Modifications:
- AbstractChannel#close shouldn't exit early just because outboundBuffer is null, and instead should use additional state closeInitiated to avoid duplicate close operations
Result:
AbstractChannel#close(..) after AbstractChannel#shutdownOutbound() will still invoke doClose and cleanup Channel state.
Motivation:
Continuing to make netty happy when compiling through errorprone.
Modification:
Mostly comments, some minor switch statement changes.
Result:
No more compiler errors!
Motivation:
Calling `newInstance()` on a Class object can bypass compile time
checked Exception propagation. This is noted in Java Puzzlers,
as well as in ErrorProne:
http://errorprone.info/bugpattern/ClassNewInstance
Modifications:
Use the niladic constructor to create a new instance.
Result:
Compile time safety for checked exceptions
Motivation:
Our http2 child channel implementation was not 100 % complete and had a few bugs. Beside this the performance overhead was non-trivial.
Modifications:
There are a lot of modifications, the most important....
* Http2FrameCodec extends Http2ConnectionHandler and Http2MultiplexCodec extends Http2FrameCodec to reduce performance heads and inter-dependencies on handlers in the pipeline
* Correctly handle outbound flow control for child channels
* Support unknow frame types in Http2FrameCodec and Http2MultiplexCodec
* Use a consistent way how to create Http2ConnectionHandler, Http2FrameCodec and Http2MultiplexCodec (via a builder)
* Remove Http2Codec and Http2CodecBuilder as the user should just use Http2MultipleCodec and Http2MultiplexCodecBuilder now
* Smart handling of flushes from child channels to reduce overhead
* Reduce object allocations
* child channels always use the same EventLoop as the parent Channel to reduce overhead and simplify implementation.
* Not extend AbstractChannel for the child channel implementation to reduce overhead in terms of performance and memory usage
* Remove Http2FrameStream.managedState(...) as the user of the child channel api should just use Channel.attr(...)
Result:
Http2MultiplexCodec (and so child channels) and Http2FrameCodec are more correct, faster and more feature complete.
Motivation:
This PR (unfortunately) does 4 things:
1) Add outbound flow control to the Http2MultiplexCodec:
The HTTP/2 child channel API should interact with HTTP/2 outbound/remote flow control. That is,
if a H2 stream used up all its flow control window, the corresponding child channel should be
marked unwritable and a writability-changed event should be fired. Similarly, a unwritable
child channel should be marked writable and a writability-event should be fired, once a
WINDOW_UPDATE frame has been received. The changes are (mostly) contained in ChannelOutboundBuffer,
AbstractHttp2StreamChannel and Http2MultiplexCodec.
2) Introduce a Http2Stream2 object, that is used instead of stream identifiers on stream frames. A
Http2Stream2 object allows an application to attach state to it, and so a application handler
no longer needs to maintain stream state (i.e. in a map(id -> state)) himself.
3) Remove stream state events, which are no longer necessary due to the introduction of Http2Stream2.
Also those stream state events have been found hard and complex to work with, when porting gRPC
to the Http2FrameCodec.
4) Add support for HTTP/2 frames that have not yet been implemented, like PING and SETTINGS. Also add
a Http2FrameCodecBuilder that exposes options from the Http2ConnectionHandler API that couldn't else
be used with the frame codec, like buffering outbound streams, window update ratio, frame logger, etc.
Modifications:
1) A child channel's writability and a H2 stream's outbound flow control window interact, as described
in the motivation. A channel handler is free to ignore the channel's writability, in which case the
parent channel is reponsible for buffering writes until a WINDOW_UPDATE is received.
The connection-level flow control window is ignored for now. That is, a child channel's writability
is only affected by the stream-level flow control window. So a child channel could be marked writable,
even though the connection-level flow control window is zero.
2) Modify Http2StreamFrame and the Http2FrameCodec to take a Http2Stream2 object intstead of a primitive
integer. Introduce a special Http2ChannelDuplexHandler that has newStream() and forEachActiveStream()
methods. It's recommended for a user to extend from this handler, to use those advanced features.
3) As explained in the documentation, a new inbound stream active can be detected by checking if the
Http2Stream2.managedState() of a Http2HeadersFrame is null. An outbound stream active can be detected
by adding a listener to the ChannelPromise of the write of the first Http2HeadersFrame. A stream
closed event can be listened to by adding a listener to the Http2Stream2.closeFuture().
4) Add a simple Http2FrameCodecBuilder and implement the missing frame types.
Result:
1) The Http2MultiplexCodec supports outbound flow control.
2) The Http2FrameCodec API makes it easy for a user to manage custom stream specific state and to create
new outbound streams.
3) The Http2FrameCodec API is much cleaner and easier to work with. Hacks like the ChannelCarryingHeadersFrame
are no longer necessary.
4) The Http2FrameCodec now also supports PING and SETTINGS frames. The Http2FrameCodecBuilder allows the Http2FrameCodec
to use some of the rich features of the Http2ConnectionHandler API.
Motivation:
When using the OIO transport we need to act on byte[] when writing and reading from / to the underyling Socket. So we should ensure we use heap buffers by default to reduce memory copies.
Modifications:
Ensure we prefer heap buffers by default for the OIO transport.
Result:
Possible less memory copies.
Motivation:
We need to ensure we always null out (or set) the address on the java.net.DatagramPacket when doing read or write operation as the same instance is used across different calls.
Modifications:
Null out the address if needed.
Result:
Ensure the correct remote address is used when connect / disconnect between calls and also mix these with calls that directly specify the remote address for adatagram packets.
Motivation:
We need to support SO_TIMEOUT for the OioDatagramChannel but we miss this atm as we not have special handling for it in the DatagramChannelConfig impl that we use. Because of this the following log lines showed up when running the testsuite:
20:31:26.299 [main] WARN io.netty.bootstrap.Bootstrap - Unknown channel option 'SO_TIMEOUT' for channel '[id: 0x7cb9183c]'
Modifications:
- Add OioDatagramChannelConfig and impl
- Correctly set SO_TIMEOUT in testsuite
Result:
Support SO_TIMEOUT for OioDatagramChannel and so faster execution of datagram related tests in the testsuite
Motivation:
Implementations of DuplexChannel delegate the shutdownOutput to the underlying transport, but do not take any action on the ChannelOutboundBuffer. In the event of a write failure due to the underlying transport failing and application may attempt to shutdown the output and allow the read side the transport to finish and detect the close. However this may result in an issue where writes are failed, this generates a writability change, we continue to write more data, and this may lead to another writability change, and this loop may continue. Shutting down the output should fail all pending writes and not allow any future writes to avoid this scenario.
Modifications:
- Implementations of DuplexChannel should null out the ChannelOutboundBuffer and fail all pending writes
Result:
More controlled sequencing for shutting down the output side of a channel.
Motivation:
When a user called ctx.close() and used the EmbeddedChannel we did not correctly run all pending tasks which means channelInactive was never called.
Modifications:
Ensure we run all pending tasks after all operations that may change the Channel state and are part of the Channel.Unsafe impl.
Result:
Fixes [#6894].
Motivation:
ErrorProne complains that the array override doesn't match the
vararg super call. See http://errorprone.info/bugpattern/Overrides
Additionally, almost every other Future uses the vararg form, so
it would be stylistically consistent to keep it that way.
Modifications:
Use vararg override.
Result:
Cleaner, less naggy code.
Motivation:
DefaultChannelPipeline.estimatorHandle needs to be volatile as its accessed from different threads.
Modifications:
Make DefaultChannelPipeline.estimatorHandle volatile and correctly init it via CAS
Result:
No more race.
Motivation:
We previously used pollLast() to retrieve a Channel from the queue that backs SimpleChannelPool. This could lead to the problem that some Channels are very unfrequently used and so when these are used the connection was already be closed and so could not be reused.
Modifications:
Allow to configure if the last recent used Channel should be used or the "oldest".
Result:
More flexible usage of ChannelPools
Motivation:
Some ChannelOptions must be set before the Channel is really registered to have the desired effect.
Modifications:
Add another constructor argument which allows to not register the EmbeddedChannel to its EventLoop until the user calls register().
Result:
More flexible usage of EmbeddedChannel. Also Fixes [#6968].
Motivation:
We had recently a report that the issue [#6607] is still not fixed.
Modifications:
Add a testcase to prove the issue is fixed.
Result:
More tests.
Motivation:
JCTools 2.0.2 provides an unbounded MPSC linked queue. Before we shaded JCTools we had our own unbounded MPSC linked queue and used it in various places but gave this up because there was no public equivalent available in JCTools at the time.
Modifications:
- Use JCTool's MPSC linked queue when no upper bound is specified
Result:
Fixes https://github.com/netty/netty/issues/5951
Motivation:
Each call to SSL_write may introduce about ~100 bytes of overhead. The OpenSslEngine (based upon OpenSSL) is not able to do gathering writes so this means each wrap operation will incur the ~100 byte overhead. This commit attempts to increase goodput by aggregating the plaintext in chunks of <a href="https://tools.ietf.org/html/rfc5246#section-6.2">2^14</a>. If many small chunks are written this can increase goodput, decrease the amount of calls to SSL_write, and decrease overall encryption operations.
Modifications:
- Introduce SslHandlerCoalescingBufferQueue in SslHandler which will aggregate up to 2^14 chunks of plaintext by default
- Introduce SslHandler#setWrapDataSize to control how much data should be aggregated for each write. Aggregation can be disabled by setting this value to <= 0.
Result:
Better goodput when using SslHandler and the OpenSslEngine.
Motivation:
The behaviour of the FixedChannelPool.release was inconsistent with the
SimpleChannelPool implementation, in that given promise is returned.
In the FixedChannelPool implementation a new promise was return and
this meant that the completion of that promise can be different.
Specifically on releasing a channel to a closed pool, the parameter
promise is failed with an IllegalStateException but the returned one
will have been successful (as it was completed by call to super
.release)
Modification:
Return the given promise as the result of FixedChannelPool.release
Result:
Returned promise will reflect the result of the release operation.
Motivation:
Channels returned to a FixedChannelPool after closing it remain active.
Since channels that where acquired from the pool are not closed during the close operation, they remain open even after releasing the channel back to the pool where they are then in accessible and become in-effect a connection leak.
Modification:
Close the released channel on releasing back to a closed pool.
Result:
Much harder to create a connection leak by closing an active
FixedChannelPool instance.
