Motivation:
PR #5355 modified interfaces to reduce GC related to the HPACK code. However this came with an anticipated performance regression related to HpackUtil.equals due to AsciiString's increase cost of charAt(..). We should mitigate this performance regression.
Modifications:
- Introduce an equals method in PlatformDependent which doesn't leak timing information and use this in HpcakUtil.equals
Result:
Fixes https://github.com/netty/netty/issues/5436
Motivation:
The HTTP/2 specification requires the pad length field of DATA, HEADERS and PUSH_PROMISE frames to be counted towards the flow control window. The current implementation doesn't do so (See #5434).
Furthermore, it's currently not possible to add one byte padding, as this would add the one byte pad length field as well as append one padding byte to the end of the frame.
Modifications:
Include the one byte pad length field in the padding parameter of the API. Thereby extending the allowed value range by one byte to 256 (inclusive). On the wire, a one byte padding is encoded with a pad length field with value zero and a 256 byte padding is encoded with a pad length field with value 255 and 255 bytes append to the end of the frame.
Result:
More correct padding.
Motivations:
The HPACK code was not really optimized and written with Netty types in mind. Because of this a lot of garbage was created due heavy object creation.
This was first reported in [#3597] and https://github.com/grpc/grpc-java/issues/1872 .
Modifications:
- Directly use ByteBuf as input and output
- Make use of ByteProcessor where possible
- Use AsciiString as this is the only thing we need for our http2 usage
Result:
Less garbage and better usage of Netty apis.
Motivation:
These methods were recently deprecated. However, they remained in use in several locations in Netty's codebase.
Modifications:
Netty's code will now access the bootstrap config to get the group or child group.
Result:
No impact on functionality.
Motivation:
If a single DATA frame ends up being decompressed into multiple frames by DelegatingDecompressorFrameListener the flow control accounting is delayed until all frames have been decompressed. However it is possible the user may want to return bytes to the flow controller which were not included in the onDataRead return value. In this case the amount of processed bytes has not been incremented and will lead to negative value for processed bytes.
Modifications:
- Http2Decompressor.incrementProcessedBytes should be called each time onDataRead is called to ensure all bytes are accounted for at the correct time
Result:
Fixes https://github.com/netty/netty/issues/5375
Motivation:
We use pre-instantiated exceptions in various places for performance reasons. These exceptions don't include a stacktrace which makes it hard to know where the exception was thrown. This is especially true as we use the same exception type (for example ChannelClosedException) in different places. Setting some StackTraceElements will provide more context as to where these exceptions original and make debugging easier.
Modifications:
Set a generated StackTraceElement on these pre-instantiated exceptions which at least contains the origin class and method name. The filename and linenumber are specified as unkown (as stated in the javadocs of StackTraceElement).
Result:
Easier to find the origin of a pre-instantiated exception.
Motivation:
JCTools supports both non-unsafe, unsafe versions of queues and JDK6 which allows us to shade the library in netty-common allowing it to stay "zero dependency".
Modifications:
- Remove copy paste JCTools code and shade the library (dependencies that are shaded should be removed from the <dependencies> section of the generated POM).
- Remove usage of OneTimeTask and remove it all together.
Result:
Less code to maintain and easier to update JCTools and less GC pressure as the queue implementation nt creates so much garbage
Motivation:
Http2ConnectionHandler will always send a GO_AWAY when the channel is closed. This may cause problems if the user is attempting to control when GO_AWAY is sent and the content of the GO_AWAY.
Modifications:
- When the channel is closed Http2ConnectionHandler should only send a GO_AWAY if one has not already been sent
Result:
The user has more control over when GO_AWAY is sent
Fixes https://github.com/netty/netty/issues/5307
Motivation:
The user may specify to use a different allocator then the default. In this case we need to ensure it is shared when creating the EmbeddedChannel inside of a ChannelHandler
Modifications:
Use the config of the "original" Channel in the EmbeddedChannel and so share the same allocator etc.
Result:
Same type of buffers are used.
Motivation:
Performing a server upgrade with a new initial flow control window will cause an NPE in the DefaultHttp2RemoteFlowController. This is due to the fact that the monitor does not check whether or not the channel is writable.
Modifications:
Added a check for channel writability before calling `writePendingBytes`. Also fixed a unit test that was supposed to be testing this :).
Result:
Fixes#5301
Motivation:
At the moment the user is responsible to increase the writer index of the composite buffer when a new component is added. We should add some methods that handle this for the user as this is the most popular usage of the composite buffer.
Modifications:
Add new methods that autoamtically increase the writerIndex when buffers are added.
Result:
Easier usage of CompositeByteBuf.
Motivation:
The HPACK code currently disallows empty header names. This is not explicitly forbidden by the HPACK RFC https://tools.ietf.org/html/rfc7541. However the HTTP/1.x RFC https://tools.ietf.org/html/rfc7230#section-3.2 and thus HTTP/2 both disallow empty header names, and so this precondition check should be moved from the HPACK code to the protocol level.
HPACK also requires that string literals which are huffman encoded must be treated as an encoding error if the string has more than 7 trailing padding bits https://tools.ietf.org/html/rfc7541#section-5.2, but this is currently not enforced.
Result:
- HPACK to allow empty header names
- HTTP/1.x and HTTP/2 header validation should not allow empty header names
- Enforce max of 7 trailing padding bits
Result:
Code is more compliant with the above mentioned RFCs
Fixes https://github.com/netty/netty/issues/5228
Related: #4333#4421#5128
Motivation:
slice(), duplicate() and readSlice() currently create a non-recyclable
derived buffer instance. Under heavy load, an application that creates a
lot of derived buffers can put the garbage collector under pressure.
Modifications:
- Add the following methods which creates a non-recyclable derived buffer
- retainedSlice()
- retainedDuplicate()
- readRetainedSlice()
- Add the new recyclable derived buffer implementations, which has its
own reference count value
- Add ByteBufHolder.retainedDuplicate()
- Add ByteBufHolder.replace(ByteBuf) so that..
- a user can replace the content of the holder in a consistent way
- copy/duplicate/retainedDuplicate() can delegate the holder
construction to replace(ByteBuf)
- Use retainedDuplicate() and retainedSlice() wherever possible
- Miscellaneous:
- Rename DuplicateByteBufTest to DuplicatedByteBufTest (missing 'D')
- Make ReplayingDecoderByteBuf.reject() return an exception instead of
throwing it so that its callers don't need to add dummy return
statement
Result:
Derived buffers are now recycled when created via retainedSlice() and
retainedDuplicate() and derived from a pooled buffer
Motivation:
We tried to provide the ability for the user to change the semantics of the threading-model by delegate the invoking of the ChannelHandler to the ChannelHandlerInvoker. Unfortunually this not really worked out quite well and resulted in just more complexity and splitting of code that belongs together. We should remove the ChannelHandlerInvoker again and just do the same as in 4.0
Modifications:
Remove ChannelHandlerInvoker again and replace its usage in Http2MultiplexCodec
Result:
Easier code and less bad abstractions.
Motivation:
We need to ensure we release all ReferenceCounted objects during tests to not leak.
Modifications:
Fix possible leaks by calling release()
Result:
No more leaks in tests.
Motivation:
Everything in the http2 package should be considered unstable for now
Modifications:
Add missing annotation on Http2ServerDowngrader
Result:
Clearly mark class as unstable.
Motivation:
We're leaking requests in our Http2ServerDowngrader tests when we
allocate a buffer using the local allocator.
Modification:
Release the request later when the request is constructed with the local
allocator.
Result:
Less leaky tests.
Motivation:
The DefaultHttp2Conneciton.close method accounts for active streams being iterated and attempts to avoid reentrant modifications of the underlying stream map by using iterators to remove from the stream map. However there are a few issues:
- While iterating over the stream map we don't prevent iterations over the active stream collection
- Removing a single stream may actually remove > 1 streams due to closed non-leaf streams being preserved in the priority tree which may result in NPE
Preserving closed non-leaf streams in the priority tree is no longer necessary with our current allocation algorithms, and so this feature (and related complexity) can be removed.
Modifications:
- DefaultHttp2Connection.close should prevent others from iterating over the active streams and reentrant modification scenarios which may result from this
- DefaultHttp2Connection should not keep closed stream in the priority tree
- Remove all associated code in DefaultHttp2RemoteFlowController which accounts for this case including the ReducedState object
- This includes fixing writability changes which depended on ReducedState
- Update unit tests
Result:
Fixes https://github.com/netty/netty/issues/5198
Motivation:
http/2 and http/1.1 have similar protocols, and it's useful to be able
to implement a single server against a single interface. There's an
injection from http/1.1 messages to http/2 ones, so it makes sense to
make folks program against http/1.1 and upgrade them under the hood.
Modifications:
added a MessageToMessageCodec<Http2StreamFrame, HttpObject> which turns
every kind of Http2StreamFrame domain object into an HttpObject domain
object, and then back again on the way out. This one is specialized for
servers, but it should be straightforward to make a symmetric one for
clients, or else extend this one.
Result:
fixes#5199, and it's now simple to make your Http2MultiplexCodec speak
Http1.1
Motivation:
Some codecs should be considered unstable as these are relative new. For this purpose we should introduce an annotation which these codecs should us to be marked as unstable in terms of API.
Modifications:
- Add UnstableApi annotation and use it on codecs that are not stable
- Move http2.hpack to http2.internal.hpack as it is internal.
Result:
Better document unstable APIs.
Motivation:
The format of some log lines used the printf style formatting instead of the logger {} formatting for variables. This would lead to printing out the literal printf tokens instead of substituting the value.
Modifications:
- Fix the format string for log statements which use printf style formatting
Result:
Logs actually capture the value of the variables instead of fixed string tokens.
Motivation:
DefaultHttp2FrameReader will stop reading data if any exception is thrown. However some exceptions are recoverable and we will lose data if we don't continue reading. For example some stream errors are recoverable.
Modifications:
- DefaultHttp2FrameReader should attempt to continue reading if a stream error is encountered.
Result:
Fixes https://github.com/netty/netty/issues/5186
Motivation:
We lately added ByteBuf.isReadOnly() which allows to detect if a buffer is read-only or not. We should add ByteBuf.asReadOnly() to allow easily access a read-only version of a buffer.
Modifications:
- Add ByteBuf.asReadOnly()
- Deprecate Unpooled.unmodifiableBuffer(Bytebuf)
Result:
More consistent api.
Motivation:
This allows using handlers for Streams in normal Netty-style. Frames are
read/written to the channel as messages, not directly as a
callback/method call. Handlers allow mixing and can ease HTTP/1 and
HTTP/2 interoperability by eventually supporting HTTP/1 handlers in
HTTP/2 and vise versa.
Modifications:
New handler Http2MultiplexCodec that converts from the current HTTP/2
API to a message-based API and child channels for streams.
Result:
The basics are done for server-side: new streams trigger creation of new
channels in much the same appearance to how new connections trigger new
channel creation. The basic frames HEADERS and DATA are handled, but
also GOAWAY and RST_STREAM.
Inbound flow control is implemented, but outbound is not. That will be
done later, along with not completing write promises on the child
channel until the write actually completes on the parent.
There is not yet support for outbound priority/weight, push promises,
and many other features.
There is a generic Object that may be set on stream frames. This also
paves the way for client-side support which needs a way to refer to
yet-to-be-created streams (due to how HEADERS allocates a stream id, and
the allocation order must be the same as transmission order).
Motivation:
If a single Encoder object is promoted to the old generation then every object
reachable from the promoted object will eventually be promoted as well. A queue
illustrates the problem very well. Say a sequence of inserts and deletions
generate an object graph:
A -> B -> C -> D -> E -> F -> G -> H,
the head of the queue is E, the tail of the queue is H, and A, B, C, D are
dead. If all queue nodes are in the young generation, then a young gc will
clean up the object graph and leave us with:
E -> F -> G -> H
on the other hand, if B and C were previously promoted to the old generation,
then a young collection assumes the refernece from C to D is from a live object
(this is a key result of generational gc, no need to mark the old generation).
Hence the young collection assumes the refence to D is a gc root and leave us
with the object graph:
B-> C -> D -> E -> F -> G -> H.
Eventually D, E, F, G, H, and all queue nodes ever seen from this point on will
be promoted, regardless of their global live or dead status. It is generally
trivial to fix nepotism issues by simply breaking the reference chain after
dequeuing a node.
Currently Encoder objects do not null their references when removed from the
hash map. We have observed a 20X increase in promoted Encoder objects due to
nepotism.
Modifications:
Null before, after, and next fields when removing Encoder objects from maps.
Result:
Fewer promoted Encoder objects, fewer Encoder objects in the old generation,
shorter young collection times, old collections spaced further apart (nepotism
is just really bad). Enjoy.
Motivation:
If an error occurs during a write operation then DefaultHttp2ConnectionEncoder.FlowControlledData will clear the CoalescingBufferQueue which will reset the queue's readable bytes to 0. To recover from an error the DefaultHttp2RemoteFlowController will attempt to return bytes to the flow control window, but since the frame has reset its own size this will lead to invalid flow control accounting.
Modifications:
- DefaultHttp2ConnectionEncoder.FlowControlledData should not reset its size if an error occurs
Result:
No more flow controller errors due to DefaultHttp2ConnectionEncoder.FlowControlledData setting its size to 0 if an error occurs.
Motivation:
HttpServerUpgradeHandler.UpgradeCodec.prepareUpgradeResponse should allow to abort the upgrade and so just continue with using HTTP. Beside this we should only pass in the response HttpHeaders as this is inline with the docs.
Modifications:
- UpgradeCodec.prepareUpgradeResponse now allows to return a boolean and so allows to specifiy if the upgrade should take place.
- Change the param from FullHttpResponse to HttpHeaders to be inline with the javadocs.
Result:
More flexible and correct handling of upgrades.
Motivation:
upgradeTo(...) takes the response as paramater, but the respone itself was already written to the Channel. This gives the user the impression the response can be changed or even act on it which may not be safe anymore once it was written and has been released.
Modifications:
Remove the response param from the method.
Result:
Less confusion and safer usage.
Motivation:
Http2Codec.SimpleChannelPromiseAggregator currently fails fast if as soon as a tryFailure or setFailure method is called. This can lead to write operations which pass the result of SimpleChannelPromiseAggregator.newPromise to multiple channel.write calls throwing exceptions due to the promise being already done. This behavior is not expected by most of the Netty codecs (SslHandler) and can also create unexpected leaks in the http2 codec (DefaultHttp2FrameWriter).
Modifications:
- Http2Codec.SimpleChannelPromiseAggregator shouldn't complete the promise until doneAllocatingPromises is called
- Usages of Http2Codec.SimpleChannelPromiseAggregator should be adjusted to handle the change in behavior
- What were leaks in DefaultHttp2FrameWriter should be fixed to catch any other cases where ctx.write may throw
Result:
SimpleChannelPromiseAggregator won't generate promises which are done when newPromise is called.
Motivation:
PromiseAggregator's API allows for the aggregate promise to complete before the user is done adding promises. In order to support this use case the API structure would need to change in a breaking manner.
Modifications:
- Deprecate PromiseAggregator and subclasses
- Introduce PromiseCombiner which corrects these issues
Result:
PromiseCombiner corrects the deficiencies in PromiseAggregator.
Motiviation:
691bc1690e made writeUtf8 consistent with String.getBytes() so that it never throws.
94f27be59b provided a writeUtf8 method which takes a ByteBufAllocator to do an appropriately sized buffer allocation.
Result:
- Assume writeUtf8 will not throw in HTTP/2 codec
- Use the new writeUtf8 method
Result:
Cleaner code in codec-http2.