Commit Graph

285 Commits

Author SHA1 Message Date
Moses Nakamura
da8734a6f9 codec-http2: Mark requests as chunked in Http2ServerDowngrader
Motivation:

Http2ServerDowngrader doesn't mark chunked requests as chunked, even
though the natural conversion from http/2 requests to http/1.1 requests
is to chunked ones.

Modifications:

Mark requests that aren't already complete as chunked.

Result:

Requests will be chunked, and can later be aggregated if necessary.
2016-09-15 15:58:40 -07:00
buchgr
908464f161 make the http2 codec void promise ready.
Motivation:

Void promises need special treatment, as they don't behave like normal promises. One
cannot add a listener to a void promise for example.

Modifications:

Inspected the code for addListener() calls, and added extra logic for void promises
where necessary. Essentially, when writing a frame with a void promise, any errors
are reported via the channel pipeline's exceptionCaught event.

Result:

The HTTP/2 codec has better support for void promises.
2016-09-15 09:19:57 -07:00
Norman Maurer
4beb075dd3 Ensure we only call debugData.release() if we called debugData.retain()
Motivation:

We need to ensure we only call debugData.release() if we called debugData.retain(), otherwise we my see an IllegalReferenceCountException.

Modifications:

Only call release() if we called retain().

Result:

No more IllegalReferenceCountException possible.
2016-09-12 06:47:38 -07:00
Norman Maurer
51629245d0 Call debugData.retain() before we forward the frame to the pipeline
Motivation:

We need to call debugData.retain() before we forward the frame to the pipeline as ByteToMessageDecoder will call release() on the buffer.

Modifications:

Correctly retain debugData and fix the unit test to test for it.

Result:

No more IllegalReferenceCountException when using the Http2FrameCodec.
2016-09-12 06:45:13 -07:00
Norman Maurer
d2389a9339 [] HTTP/2: SETTINGS_HEADER_TABLE_SIZE should be an unsigned int
Motivation:

he HTTP/2 spec demands that the max value for SETTINGS_HEADER_TABLE_SIZE should be an unsigned 32-bit integer.

Modifications:

Change the limit to unsigned 32-bit integer and add tests.

Result:

Complient to rfc.
2016-09-09 13:20:32 +02:00
buchgr
515f8964b4 HTTP/2: Fix some errors reported by h2spec.
Motivation:

h2spec [1] reported 32 issues [2] with Netty's HTTP/2 implementation.

Modifications:

Fixed 11 of those issues. Mostly wrong error codes or added missing validation
code in the frame reader.

Result:

11 fewer errors. h2spec now reports: 70 tests, 48 passed, 1 skipped, 21 failed

[1] https://github.com/summerwind/h2spec
[2] https://github.com/netty/netty/issues/5761
2016-09-01 08:28:16 +02:00
Trustin Lee
5b46cf25c1 Fulfill the write promise of empty HTTP/2 DATA frames sooner
Motivation:

DefaultHttp2ConnectionEncoder.FlowControlledData.write() does not
complete the write promise of empty HTTP/2 DATA frames until either a
non-DATA frame, a non-empty DATA frame or a DATA frame with endOfStream
set. This makes the write promise of the empty DATA frame is notified
much later than it could be.

Modifications:

- Notify the write promise of the empty DATA frames immediately is the
  queue contains empty DATA frames only

Result:

The write promise of an empty DATA frame is notified sooner.
2016-08-26 08:45:09 +09:00
Scott Mitchell
208893aac9 HTTP/2 Hpack Encoder Cleanup
Motivation:
The HTTP/2 HPACK Encoder class has some code which is only used for test purposes. This code can be removed to reduce complexity and member variable count.

Modifications:
- Remove test code and update unit tests
- Other minor cleanup

Result:
Test code is removed from operational code.
2016-08-25 09:08:46 -07:00
Norman Maurer
5fd239c29c Ensure we not log missleading errors if the promise was already failed due errors
Motivation:

In DefaultHttp2ConnectionEncoder we fail the promise in in the FlowControlledData.error(...) method but also add it the CoalescingBufferQueue. Which can lead to have the promise failed by error(...) before it can be failed in CoalescingBufferQueue.

This can lead to confusing and missleading errors in the log like:
    2016-08-12 09:47:43,716       nettyIoExecutorGroup-1-9 [WARN ] PromiseNotifier                - Failed to mark a promise as failure because it's done already: DefaultChannelPromise@374225e0(failure: javax.net.ssl.SSLException: SSLEngine closed already)
    javax.net.ssl.SSLException: SSLEngine closed already
        at io.netty.handler.ssl.SslHandler.wrap(...)(Unknown Source) ~[netty-all-4.1.5.Final-SNAPSHOT.jar:?]

Modifications:

Ensure we only fail the queue (which will also fail the promise).

Result:

No more missleading logs.
2016-08-18 07:17:45 +02:00
buchgr
d568cfc14a HTTP/2: Treat MAX_CONCURRENT_STREAMS exceeded as a stream error.
Motivation:

As per the HTTP/2 spec, exceeding the MAX_CONCURRENT_STREAMS should be treated as a stream error as opposed to a connection error.

"An endpoint that receives a HEADERS frame that causes its advertised concurrent stream limit to be exceeded MUST treat this as a stream error (Section 5.4.2) of type PROTOCOL_ERROR or REFUSED_STREAM." http://httpwg.org/specs/rfc7540.html#rfc.section.5.1.2

Modifications:

Make the error a stream error.

Result:

It's a stream error.
2016-08-17 14:47:15 +02:00
Scott Mitchell
f73c4f24ee HTTP/2 HPACK Bounds Check Fix
Motivation:
21e8d84b79 changed the way bounds checking was done, but however a bounds check in the case of READ_LITERAL_HEADER_NAME_LENGTH_PREFIX was using an old value. This would delay when the bounds check would actually be done and potentially allow more allocation than necessary.

Modifications:
- Use the new length (index) in the bounds check instead of an old length (nameLength) which had not yet been assigned to the new value.

Result:
More correct bounds checking.
2016-08-13 11:04:26 -07:00
Scott Mitchell
21e8d84b79 HTTP/2 Simplify Headers Decode Bounds Checking
Motivation:
The HPACK decoder keeps state so that the decode method can be called multiple times with successive header fragments. This decoder also requires that a method is called to signify the decoding is complete. At this point status is returned to indicate if the max header size has been violated. Netty always accumulates the header fragments into a single buffer before attempting to HPACK decode process and so keeping state and delaying notification that bounds have been exceeded is not necessary.

Modifications:
- HPACK Decoder#decode(..) now must be called with a complete header block
- HPACK will terminate immediately if the maximum header length, or maximum number of headers is exceeded
- Reduce member variables in the HPACK Decoder class because they can now live in the decode(..) method

Result:
HPACK bounds checking is done earlier and less class state is needed.
2016-08-12 17:12:53 -07:00
Scott Mitchell
4d74bf3984 HTTP/2 MaxStreams cleanup
Motivation:
765e944d4d imposed a limit on the maximum number of stream in all states. However the default limit did not allow room for streams in addition to SETTINGS_MAX_CONCURRENT_STREAMS. This can mean streams in states outside the states which SETTINGS_MAX_CONCURRENT_STREAMS applies to may not be reliably created.

Modifications:
- The default limit should be larger than SETTINGS_MAX_CONCURRENT_STREAMS

Result:
More lenient limit is applied to maxStreams by default.
2016-08-11 09:38:58 -07:00
Scott Mitchell
765e944d4d HTTP/2 limit streams in all states
Motivation:
SETTINGS_MAX_CONCURRENT_STREAMS does not apply to idle streams and thus we do not apply any explicit limitations on how many idle streams can be created. This may allow a peer to consume an undesirable amount of resources.

Modifications:
- Each Endpoint should enforce a limit for streams in a any state. By default this limit will be the same as SETTINGS_MAX_CONCURRENT_STREAMS but can be overridden if necessary.

Result:
There is now a limit to how many IDLE streams can be created.
2016-08-11 09:01:37 -07:00
Scott Mitchell
a4ad68239e Http2ConnectionDecoder remove localSettings setter method
Motivation:
Http2ConnectionDecoder#localSettings(Http2Settings) is not used in codec-http2 and currently results in duplicated code.

Modifications:
- Remove Http2ConnectionDecoder#localSettings(Http2Settings)

Result:
Smaller interface and less duplicated code.
2016-08-10 12:23:03 -07:00
buchgr
328510468c Complete ChannelPromise for Http2WindowUpdateFrames in Http2FrameCodec. Fixes
Motivation:

The channel promise of a window update frame is not completed correctly,
depending on the failure or success of the operation.

Modification:

Succeed / Fail the promise if the window update succeeds / fails.

Result:

Correctly succeed / fail the promise.
2016-07-15 17:29:41 +02:00
Oliver Gould
f2ce28bf18 Satisfy write promise when writing an Http2WindowUpdateFrame to Http2FrameCodec.
Motivation:

When writing an Http2WindowUpdateFrame to an Http2FrameCodec, the
ChannelPromise is never satisfied, so callers cannot generically rely on the
write future being satisfied on success.

Modifications:

When writing Http2WindowUpdateFrame, Http2FrameCodec now satisfies the
ChannelPromise immediately.

Result:

The write future is satisfied on successful writes.

Fixes .
2016-07-13 21:31:34 +02:00
Scott Mitchell
9de90d07a9 DefaultHttp2RemoteFlowController reentry infinite loop
Motivation:
DefaultHttp2RemoteFlowController.writePendingBytes does not support reentry but does not enforce this constraint. Reentry is possible if the channel transitions from Writable -> Not Writable -> Writable during the distribution phase. This can happen if the user flushes when notified of the channel transitioning to the not writable state, and may be done if the user wants to fill the channel outbound buffer before flushing.

Modifications:
- DefaultHttp2RemoteFlowController.writePendingBytes should protect against reentry

Result:
DefaultHttp2RemoteFlowController will not allocate unexpected amounts or enter an infinite loop.
2016-07-13 09:12:33 -07:00
Scott Mitchell
4baff691b4 DefaultPromise make listeners not volatile
Motivation:
DefaultPromise has a listeners member variable which is volatile to allow for an optimization which makes notification of listeners less expensive when there are no listeners to notify. However this change makes all other operations involving the listeners member variable more costly. This optimization which requires listeners to be volatile can be removed to avoid volatile writes/reads for every access on the listeners member variable.

Modifications:
- DefaultPromise listeners is made non-volatile and the null check optimization is removed

Result:
DefaultPromise.listeners is no longer volatile.
2016-07-07 12:53:03 -07:00
Norman Maurer
6492cb98b2 Revert "DefaultPromise make listeners not volatile"
This reverts commit 4d8132ff24 as I missed something I want to discuss first.
2016-07-07 08:37:41 +02:00
Scott Mitchell
4d8132ff24 DefaultPromise make listeners not volatile
Motivation:
DefaultPromise has a listeners member variable which is volatile to allow for an optimization which makes notification of listeners less expensive when there are no listeners to notify. However this change makes all other operations involving the listeners member variable more costly. This optimization which requires listeners to be volatile can be removed to avoid volatile writes/reads for every access on the listeners member variable.

Modifications:
- DefaultPromise listeners is made non-volatile and the null check optimization is removed

Result:
DefaultPromise.listeners is no longer volatile.
2016-07-07 08:01:25 +02:00
Scott Mitchell
6af56ffe76 HPACK Encoder headerFields improvements
Motivation:
HPACK Encoder has a data structure which is similar to a previous version of DefaultHeaders. Some of the same improvements can be made.

Motivation:
- Enforce the restriction that the Encoder's headerFields length must be a power of two so we can use masking instead of modulo
- Use AsciiString.hashCode which already has optimizations instead of having yet another hash code algorithm in Encoder

Result:
Fixes https://github.com/netty/netty/issues/5357
2016-06-30 09:00:12 -07:00
Scott Mitchell
a7f7d9c8e0 Remove unsafe char[] access in PlatformDependent
Motivation:
PlatformDependent attempts to use reflection to get the underlying char[] (or byte[]) from String objects. This is fragile as if the String implementation does not utilize the full array, and instead uses a subset of the array, this optimization is invalid. OpenJDK6 and some earlier versions of OpenJDK7 String have the capability to use a subsection of the underlying char[].

Modifications:
- PlatformDependent should not attempt to use the underlying array from String (or other data types) via reflection

Result:
PlatformDependent hash code generation for CharSequence does not depend upon specific JDK implementation details.
2016-06-30 08:58:28 -07:00
Scott Mitchell
df41be6fc8 HTTP/2 Decoder validate that GOAWAY lastStreamId doesn't increase
Motivation:
The HTTP/2 RFC states in https://tools.ietf.org/html/rfc7540#section-6.8 that Endpoints MUST NOT increase the value they send in the last stream identifier however we don't enforce this when decoding GOAWAY frames.

Modifications:
- Throw a connection error if the peer attempts to increase the lastStreamId in a GOAWAY frame

Result:
RFC is more strictly enforced.
2016-06-29 07:53:13 -07:00
buchgr
3613d15bca Split Http2MultiplexCodec into Frame- and MultiplexCodec + Tests. Fixes
Motivation:

Quote from issue 4914:
"Http2MultiplexCodec currently does two things: mapping the existing h2 API to frames and managing the child channels.

It would be better if the two parts were separated. This would allow less-coupled development of the HTTP/2 handlers (flow control could be its own handler, for instance) and allow applications to insert themselves between all streams and the codec, which permits custom logic and could be used, in part, to implement custom frame types.

It would also greatly ease testing, as the child channel could be tested by itself without dealing with how frames are encoded on the wire."

Modifications:

- Split the Http2MultiplexCodec into Http2FrameCodec and Http2MultiplexCodec. The Http2FrameCodec interacts with the existing HTTP/2 callback-based API, while the Http2MulitplexCodec is completely independent of it and simply multiplexes Http2StreamFrames to the child channels. Additionally, the Http2Codec handler is introduced, which is a convenience class that simply sets up the Http2FrameCodec and Http2MultiplexCodec in the channel pipeline and removes itself.

- Improved test coverage quite a bit.

Result:

- The original Http2MultiplexCodec is split into Http2FrameCodec and Http2MultiplexCodec.
- More tests for higher confidence in the code.
2016-06-29 07:22:09 +02:00
Scott Mitchell
70651cc58d HpackUtil.equals performance improvement
Motivation:
PR  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
2016-06-27 14:37:39 -07:00
buchgr
73c4ad5f0a http2: count pad length field toward flow control. Fixes
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 ).

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.
2016-06-25 09:51:36 -07:00
Norman Maurer
b4d4c0034d Optimize HPACK usage to align more with Netty types and remove heavy object creations. Related to []
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 [] 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.
2016-06-22 14:26:05 +02:00
Tim Brooks
d964bf6f18 Remove usages of deprecated methods group() and childGroup().
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.
2016-06-21 14:06:57 +02:00
Scott Mitchell
6aa5f76d42 HTTP/2 DelegatingDecompressorFrameListener return bytes to flow control
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
2016-06-20 14:24:09 -07:00
Norman Maurer
e845670043 Set some StackTraceElement on pre-instantiated static exceptions
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.
2016-06-20 11:33:05 +02:00
Guido Medina
c3abb9146e Use shaded dependency on JCTools instead of copy and paste
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
2016-06-10 13:19:45 +02:00
Scott Mitchell
79f2e3604e HTTP/2 close only send GO_AWAY if one has not already been sent
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
2016-06-06 11:18:30 -07:00
Norman Maurer
844976a0a2 Ensure the same ByteBufAllocator is used in the EmbeddedChannel when compress / decompress. Related to []
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.
2016-05-31 09:08:33 +02:00
nmittler
79a06eaede Fix NPE in Http2ConnectionHandler.onHttpServerUpgrade
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 
2016-05-25 09:13:06 -07:00
Norman Maurer
7b25402e80 Add CompositeByteBuf.addComponent(boolean ...) method to simplify usage
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.
2016-05-21 19:52:16 +02:00
Scott Mitchell
1cb706ac93 HTTP/2 HPACK Header Name Validation and Trailing Padding
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
2016-05-17 13:42:16 -07:00
Trustin Lee
3a9f472161 Make retained derived buffers recyclable
Related:   

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
2016-05-17 11:16:13 +02:00
Norman Maurer
68cd670eb9 Remove ChannelHandlerInvoker
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.
2016-05-17 11:14:00 +02:00
Norman Maurer
979dc3e3e4 Fix one more possible leak as a follow up of 341a235fea 2016-05-13 18:38:37 +02:00
Norman Maurer
341a235fea Fix possible leaks in Http2ServerDowngraderTest
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.
2016-05-13 17:38:26 +02:00
Norman Maurer
4b24b0aac8 Annotate Http2ServerDowngrader with @UnstableApi
Motivation:

Everything in the http2 package should be considered unstable for now

Modifications:

Add missing annotation on Http2ServerDowngrader

Result:

Clearly mark class as unstable.
2016-05-13 08:41:46 +02:00
Moses Nakamura
4d2e91a10d codec-http2: Stop leaking in header downgrader test
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.
2016-05-10 09:04:52 +02:00
Scott Mitchell
d580245afc DefaultHttp2Connection.close Reentrant Modification
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
2016-05-09 14:16:30 -07:00
Moses Nakamura
9ce84dcb21 Http2 to Http1.1 converter on new Http2 server API
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 , and it's now simple to make your Http2MultiplexCodec speak
Http1.1
2016-05-09 13:24:46 -07:00
Norman Maurer
9229ed98e2 [] Add annotation which marks packages/interfaces/classes as unstable
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.
2016-05-09 15:16:35 +02:00
Scott Mitchell
b17ee6066f HTTP/2 Log Cleanup
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.
2016-05-06 10:52:00 -07:00
Scott Mitchell
467e5a1019 DefaultHttp2FrameReader stops reading if stream error
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
2016-04-30 10:41:06 -07:00
Trustin Lee
38d05abd84 Reorganize imports 2016-04-14 18:58:47 +09:00
Norman Maurer
d698746609 Add ByteBuf.asReadOnly()
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.
2016-04-14 10:51:20 +02:00
Norman Maurer
7c734fcf73 Fix resource leak in tests introduced by 69070c37ba. 2016-04-14 10:13:55 +02:00
Trustin Lee
0b078314b2 Add ByteBuf.isReadOnly()
Motivation:

It is sometimes useful to determins if a buffer is read-only.

Modifications:

Add ByteBuf.isReadOnly()

Result:

One more feature
2016-04-13 21:41:27 +09:00
Scott Mitchell
61cfdd7671 e24a5d8 compile error
Motivation:
e24a5d8 was cherry-picked but had a compile error.

Modifications:
- Fix the compile error in e24a5d8

Result:
Build now compiles.
2016-03-25 12:51:13 -07:00
Eric Anderson
e24a5d8839 Map HTTP/2 Streams to Channels
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).
2016-03-25 12:14:44 -07:00
Carsten Varming
d6bf388343 Prevent nepotism with generational GCs.
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.
2016-03-23 17:27:00 +01:00
Scott Mitchell
fc099292fd HTTP/2 DefaultHttp2ConnectionEncoder data frame size incorrect if error
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.
2016-03-18 11:38:01 -07:00
Norman Maurer
8ec594c6eb Change HttpServerUpgradeHandler.UpgradeCodec to allow aborting upgrade
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.
2016-03-18 17:01:59 +01:00
Norman Maurer
ed9d6c79bc [] Remove misleading argument from HttpServerUpgradeHandler.UpgradeCodec.upgradeTo
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.
2016-03-17 10:50:07 +01:00
Scott Mitchell
6a2425b846 HTTP/2 SimpleChannelPromiseAggregator don't fail fast
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.
2016-03-14 11:00:21 -07:00
Scott Mitchell
45849b2fa8 Deprecate PromiseAggregator
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.
2016-03-14 10:53:30 -07:00
Scott Mitchell
404666d247 HTTP/2 ByteBufUtil.writeUtf8 cleanup
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.
2016-03-11 14:24:45 -08:00
Scott Mitchell
bd6040a36e HTTP/2 DefaultHttp2Connection NPE
Motivation:
If while iterating the active streams a close operation occurs this will be queued and process after the iteration has completed to avoid a concurrent modification exception. However it is possible that during the iteration the stream which was closed could have been removed from the priority tree and its parent would be set to null. Then after the iteration completes the close operation will attempt to dereference the parent and results in a NPE.

Modifications:
- pending close operations should verify the stream's parent is not null before processing the event

Result:
No More NPE.
2016-03-11 07:41:04 -08:00
Scott Mitchell
900353af52 HTTP/2 Reduce Log Level
Motivation:
83c4aa6ad8 changed the log level to warn, but should have changed to debug.

Modifications:
- Change the log level to debug in Http2ConnectionHandler if the GO_AWAY fails to send. The write failure could be the result of the channel already being closed.

Result:
Fixes https://github.com/netty/netty/issues/4930.
2016-03-04 17:25:04 -08:00
Scott Mitchell
4b5b230802 HTTP/2 DefaultHttp2HeadersDecoder weighted average error
Motiviation:
cfcee5798d introduced code to resize the headers based upon a weighted average. The weight used for new entries was initialized using integer arithmetic when it should have been floating point arithmetic and so new values contribute 0 weight.

Modifications:
- Cast to float when initializing

Result:
Weighted average does not give 0 weight to new headers in DefaultHttp2HeadersDecoder.
2016-03-02 15:07:38 -08:00
Scott Mitchell
c4fbc0642d HTTP/2 stream removed from map before onStreamClosed called
Motivation:
The interface contract of Http2Connection.Listener.onStreamClosed says that the stream will be removed from the active stream map, and not necessarily the stream map. If the channel becomes inactive we may remove from the stream map before calling onStreamClosed.

Modifications:
- Don't remove from the stream map during iteration until after onStreamClosed is called

Result:
Expectations of onStreamClosed interface are not violated
2016-02-29 08:46:22 -08:00
Scott Mitchell
83c4aa6ad8 HTTP/2 Writes GO_AWAY on channelInactive
Motivation:
Http2ConnectionHandler inherits from ByteToMessageDecoder. ByteToMessageDecoder.channelInactive will attempt to decode any remaining data by calling the abstract decode method. If the Http2ConnectionHandler is in server mode, and no data has been exchanged yet, it will try to treat this data as an invalid connection preface and write a GO_AWAY. This is noisy in the logs and creates an illusion that there is a protocol violation when there has not been.

Modifications:
- If the channel is inactive the connection preface decode should not be executed.

Result:
Log files don't include misleading error messages related to connection preface errors.
2016-02-18 19:47:42 -08:00
Brendt Lucas
41d0a81691 Use ByteBufAllocator to allocate ByteBuf for FullHttpMessage Motivation: When converting SPDY or HTTP/2 frames to HTTP/1.x, netty always used an unpooled heap ByteBuf.
Modifications:
When constructing the FullHttpMessage pass in the ByteBuf to use via the ByteBufAllocator assigned via the context.

Result:
The ByteBuf assigned to the FullHttpMessage can now be configured as a pooled/unpooled, direct/heap based ByteBuf via the ByteBufAllocator used.
2016-02-17 19:55:52 -08:00
Moses Nakamura
f0f0b69d90 fixed "sensative" typo to read "sensitive" 2016-02-17 08:18:11 -08:00
Scott Mitchell
06e29e0d1b HTTP/2 codec may not always call Http2Connection.onStreamRemoved
Motivation:
Http2Connection.onStreamRemoved is not always called if Http2Connection.onStreamAdded is called. This is problematic as users may rely on the onStreamRemoved method to be called to release ByteBuf objects and do other cleanup.

Modifications:
- Http2Connection.close will remove all streams existing streams and prevent new ones from being created
- Http2ConnectionHandler will call the new close method in channelInactive

Result:
Http2Connection.onStreamRemoved is always called when Http2Connection.onStreamRemoved is called to preserve the Http2Connection guarantees.
Fixes https://github.com/netty/netty/issues/4838
2016-02-12 16:01:37 -08:00
Scott Mitchell
56e6e07b25 HTTP/2 RST_STREAM Regression f990f99
Motivation:
Commit f990f99 introduced a bug into the RST_STREAM processing that would prevent a RST_STREAM from being sent when it should have been. The promise would be marked as successful even though the RST_STREAM frame would never be sent.

Modifications:
- Fix conditional in Http2ConnectionHandler.resetStream to allow reset streams to be sent in all stream states besides IDLE.

Result:
RST_STREAM frames are now sent when they are supposed to be sent
Fixes https://github.com/netty/netty/issues/4856
2016-02-10 13:47:53 -08:00
Scott Mitchell
7a7160f176 HTTP/2 Buffer Leak if UTF8 Conversion Fails
Motivation:
Http2CodecUtil uses ByteBufUtil.writeUtf8 but does not account for it
throwing an exception. If an exception is thrown because the format is
not valid UTF16 encoded UTF8 then the buffer will leak.

Modifications:
- Make sure the buffer is released if an exception is thrown
- Ensure call sites of the Http2CodecUtil.toByteBuf can tolerate and
  exception being thrown

Result:
No leak if exception data can not be converted to UTF8.
2016-02-02 11:22:17 -08:00
Scott Mitchell
f990f9983d HTTP/2 Don't Flow Control Iniital Headers
Motivation:
Currently the initial headers for every stream is queued in the flow controller. Since the initial header frame may create streams the peer must receive these frames in the order in which they were created, or else this will be a protocol error and the connection will be closed. Tolerating the initial headers being queued would increase the complexity of the WeightedFairQueueByteDistributor and there is benefit of doing so is not clear.

Modifications:
- The initial headers will no longer be queued in the flow controllers

Result:
Fixes https://github.com/netty/netty/issues/4758
2016-02-01 13:37:43 -08:00
Scott Mitchell
5fb18e3415 InboundHttp2ToHttpAdapter leak and logic improvements
Motivation:
In HttpConversionUtil's toHttpRequest and toHttpResponse methods can
allocate FullHttpMessage objects, and if an exeception is thrown during
the header conversion then this object will not be released. If a
FullHttpMessage is not fired up the pipeline, and the stream is closed
then we remove from the map, but do not release the object. This leads
to a ByteBuf leak. Some of the logic related to stream lifetime management
and FullHttpMessage also predates the RFC being finalized and is not correct.

Modifications:
- Fix leaks in HttpConversionUtil
- Ensure the objects are released when they are removed from the map.
- Correct logic and unit tests where they are found to be incorrect.

Result:
Fixes https://github.com/netty/netty/issues/4780
Fixes https://github.com/netty/netty/issues/3619
2016-02-01 12:38:40 -08:00
Trustin Lee
4d6ab1d30d Fix missing trailing data on HTTP client upgrade
Motivation:

When HttpClientUpgradeHandler upgrades from HTTP/1 to another protocol,
it performs a two-step opertion:

1. Remove the SourceCodec (HttpClientCodec)
2. Add the UpgradeCodec

When HttpClientCodec is removed from the pipeline, the decoder being
removed triggers channelRead() event with the data left in its
cumulation buffer. However, this is not received by the UpgradeCodec
becuase it's not added yet. e.g. HTTP/2 SETTINGS frame sent by the
server can be missed out.

To fix the problem, we need to reverse the steps:

1. Add the UpgradeCodec
2. Remove the SourceCodec

However, this does not work as expected either, because UpgradeCodec can
send a greeting message such as HTTP/2 Preface. Such a greeting message
will be handled by the SourceCodec and will trigger an 'unsupported
message type' exception.

To fix the problem really, we need to make the upgrade process 3-step:

1. Remove/disable the encoder of SourceCodec
2. Add the UpgradeCodec
3. Remove the SourceCodec

Modifications:

- Add SourceCodec.prepareUpgradeFrom() so that SourceCodec can remove or
  disable its encoder
- Implement HttpClientCodec.prepareUpgradeFrom() properly
- Miscellaneous:
  - Log the related channel as well When logging the failure to send a
    GOAWAY

Result:

Cleartext HTTP/1-to-HTTP/2 upgrade works again.
2016-02-01 15:52:37 +01:00
Scott Mitchell
11bcb8790c Http2Connection stream id generation to support queueing
Motivation:
StreamBufferingEncoder provides queueing so that MAX_CONCURRENT_STREAMS is not violated. However the stream id generation provided by Http2Connection.nextStreamId() only returns the next stream id that is expected on the connection and does not account for queueing. The codec should provide a way to generate the next stream id for a given endpoint that functions with or without queueing.

Modifications:
- Change Http2Connection.nextStreamId to Http2Connection.incrementAndGetNextStreamId

Result:
Http2Connection can generate the next stream id in queued and non-queued scenarios.
Fixes https://github.com/netty/netty/issues/4704
2016-01-29 11:37:17 -08:00
Scott Mitchell
78b508a7eb AbstractHttp2ConnectionHandlerBuilder validateHeaders cannot be set with encoder/decoder
Motivation:
If validateHeaders is set in combination with the encoder/decoder it will be silently ignored. We should enforce the constraint that validateHeaders and encoder/decoder are mutually exclusive.

Modifications:
- Make sure either validateHeaders can be set or encoder/decoder.

Result:
AbstractHttp2ConnectionHandlerBuilder does not allow conflicting options to be set.
2016-01-29 00:33:45 -08:00
Scott Mitchell
e8850072e2 HTTP/2 DefaultHttp2RemoteFlowController frame merging with padding bug
Motivation:
DefaultHttp2RemoteFlowController does not correctly account for the padding in the event frames are merged. This causes the internal stat of DefaultHttp2RemoteFlowController to become corrupt and can result in attempting to write frames when there are none.

Modifications:
- Update DefaultHttp2RemoteFlowController to account for frame sizes not necessarily adding together.

Result:
DefaultHttp2RemoteFlowController internal state does not become corrupt when padding is present.
Fixes https://github.com/netty/netty/issues/4573
2016-01-27 14:52:00 -08:00
Alex Petrov
7bcae8919d Log the current channel in Http2FrameLogger
Motivation:
Currently it's impossible to distinguish which
connection the corresponding logged message	is
related to.

Modifications:
Http2FrameLogger is extended to support channel
id logging, usages in Inbound/Outbound Frame
Loggers are adjusted accordingly.

Result:
Logger outputs the channel id.
2016-01-18 10:52:08 +01:00
Scott Mitchell
7dba13f276 HttpConversionUtil remove throws from method signature
Motivation:
HttpConversionUtil.toHttp2Headers currently has a throws Exception as part of the signature. This comes from the signature of ByteProcessor.process, but is not necessary because the ByteProcessor used does not throw.

Modifications:
- Remove throws Exception from the signature of HttpConversionUtil.toHttp2Headers.

Result:
HttpConversionUtil.toHttp2Headers interface does not propagate a throws Exception when it is used.
2016-01-15 10:53:34 +01:00
Norman Maurer
8716b9d4bd Revert "Fix unnecessary boxing and incorrect Serializable"
This reverts commit 0ae6f17285.
2015-12-31 14:48:10 +01:00
Xiaoyan Lin
0ae6f17285 Fix unnecessary boxing and incorrect Serializable
Motivation:

- AbstractHttp2ConnectionHandlerBuilder.encoderEnforceMaxConcurrentStreams can be the primitive boolean
- SpdySession.StreamComparator should not be Serializable since SpdySession is not Serializable

Modifications:

Use boolean instead and remove Serializable

Result:

- Minor improvement for AbstractHttp2ConnectionHandlerBuilder
- StreamComparator is not Serializable any more
2015-12-31 10:45:24 +01:00
Xiaoyan Lin
475d901131 Fix errors reported by javadoc
Motivation:

Javadoc reports errors about invalid docs.

Modifications:

Fix some errors reported by javadoc.

Result:

A lot of javadoc errors are fixed by this patch.
2015-12-27 08:36:45 +01:00
Xiaoyan Lin
a96d52fe66 Fix javadoc links and tags
Motivation:

There are some wrong links and tags in javadoc.

Modifications:

Fix the wrong links and tags in javadoc.

Result:

These links will work correctly in javadoc.
2015-12-26 08:34:31 +01:00
Scott Mitchell
80cff236e4 HTTP/2 UniformStreamByteDistributor negative window shouldn't write
Motivation:
If the stream window is negative UniformStreamByteDistributor may write data. This is prohibited by the RFC https://tools.ietf.org/html/rfc7540#section-6.9.2.

Modifications:
- UniformStreamByteDistributor should use StreamState.isWriteAllowed()

Result:
UniformStreamByteDistributor is more complaint with HTTP/2 RFC.
Fixes https://github.com/netty/netty/issues/4545
2015-12-23 10:14:24 -08:00
Scott Mitchell
7b2f55ec2f HTTP/2 Remove RemoteFlowController.streamWritten
Motivation:
RemoteFlowController.streamWritten is not currently required. We should remove it to keep interfaces minimal.

Modifications:
- Remove RemoteFlowController.streamWritten

Result:
1 Less method in RemoteFlowController interface.
Fixes https://github.com/netty/netty/issues/4600
2015-12-22 16:59:40 -08:00
Scott Mitchell
72accceeac HTTP/2 remove PriorityStreamByteDistributor
Motivation:
PriorityStreamByteDistributor is now obsolete and can be replaced by WeightedFairQueueByteDistributor.

Modifications:
- Remove PriorityStreamByteDistributor and use WeightedFairQueueByteDistributor by default.

Result:
PriorityStreamByteDistributor no longer has to be maintained and is replaced by a better algorithm.
2015-12-21 08:56:50 -08:00
Scott Mitchell
9ac430f16f HTTP/2 DefaultHttp2RemoteFlowController Stream writability notification broken
Motivation:
DefaultHttp2RemoteFlowController.ListenerWritabilityMonitor no longer reliably detects when a stream's writability change occurs.

Modifications:
- Ensure writiability is reliabily reported by DefaultHttp2RemoteFlowController.ListenerWritabilityMonitor
- Fix infinite loop issue (https://github.com/netty/netty/issues/4588) detected when consolidating unit tests

Result:
Reliable stream writability change notification, and 1 less infinite loop in UniformStreamByteDistributor.
Fixes https://github.com/netty/netty/issues/4587
2015-12-21 10:01:33 +01:00
Trustin Lee
b39380ad83 Revamp InboundHttp2ToHttpAdapter builder API
Related:  

Motivation:

Consistency in our builder API design

Modifications:

- Add AbstractInboundHttp2ToHttpAdapterBuilder
- Replace the old 'Builder's with InboundHttp2ToHttpAdapterBuilder and
  InboundHttp2ToHttpPriorityAdapterBuilder

Result:

Builder API consistency
2015-12-18 12:44:46 +09:00
Scott Mitchell
904e70a4d4 HTTP/2 Weighted Fair Queue Byte Distributor
Motivation:
PriorityStreamByteDistributor uses a homegrown algorithm which distributes bytes to nodes in the priority tree. PriorityStreamByteDistributor has no concept of goodput which may result in poor utilization of network resources. PriorityStreamByteDistributor also has performance issues related to the tree traversal approach and number of nodes that must be visited. There also exists some more proven algorithms from the resource scheduling domain which PriorityStreamByteDistributor does not employ.

Modifications:
- Introduce a new ByteDistributor which uses elements from weighted fair queue schedulers

Result:
StreamByteDistributor which is sensitive to priority and uses a more familiar distribution concept.
Fixes https://github.com/netty/netty/issues/4462
2015-12-17 11:17:02 -08:00
Trustin Lee
2202e8f967 Revamp the Http2ConnectionHandler builder API
Related: 

Motivation:

- A user might want to extend Http2ConnectionHandler and define his/her
  own static inner Builder class that extends
  Http2ConnectionHandler.BuilderBase. This introduces potential
  confusion because there's already Http2ConnectionHandler.Builder. Your
  IDE will warn about this name duplication as well.
- BuilderBase exposes all setters with public modifier. A user's Builder
  might not want to expose them to enforce it to certain configuration.
  There's no way to hide them because it's public already and they are
  final.
- BuilderBase.build(Http2ConnectionDecoder, Http2ConnectionEncoder)
  ignores most properties exposed by BuilderBase, such as
  validateHeaders, frameLogger and encoderEnforceMaxConcurrentStreams.
  If any build() method ignores the properties exposed by the builder,
  there's something wrong.
- A user's Builder that extends BuilderBase might want to require more
  parameters in build(). There's no way to do that cleanly because
  build() is public and final already.

Modifications:

- Make BuilderBase and Builder top-level so that there's no duplicate
  name issue anymore.
  - Add AbstractHttp2ConnectionHandlerBuilder
  - Add Http2ConnectionHandlerBuilder
  - Add HttpToHttp2ConnectionHandlerBuilder
- Make all builder methods in AbstractHttp2ConnectionHandlerBuilder
  protected so that a subclass can choose which methods to expose
- Provide only a single build() method
  - Add connection() and codec() so that a user can still specify
    Http2Connection or Http2Connection(En|De)coder explicitly
  - Implement proper state validation mechanism so that it is prevented
    to invoke conflicting setters

Result:

Less confusing yet flexible builder API
2015-12-17 14:08:13 +09:00
Norman Maurer
f4386fb8e9 Allow to set Http2HeaderEncoder.SensitivityDetector in the Http2ConnectionHandler
Motivation:

Some times the user wants to set a Http2HeaderEncoder.SensitivityDetector when building a Http2ConnectionHandler.

Modifications:

Allow to set Http2HeaderEncoder.SensitivityDetector via builder.

Result:

More flexible building of Http2ConnectionHandler possible.
2015-12-10 14:34:04 +01:00
Norman Maurer
c6d43667d4 Add Http2HeadersEncoder.ALWAYS_SENSITIVE instance
Motivation:

We already provide a NEVER_SENSITIVE instance,we should add ALWAYS_SENSITIVE as well.

Modifications:

Add ALWAYS_SENSITIVE instance which will always return true when check for sesitive.

Result:

User can reuse code.
2015-12-10 12:59:04 +01:00
Scott Mitchell
6257091d12 HttpConversionUtil does not account for COOKIE compression
Motivation:
The HTTP/2 RFC allows for COOKIE values to be split into individual header elements to get more benefit from compression (https://tools.ietf.org/html/rfc7540#section-8.1.2.5). HttpConversionUtil was not accounting for this behavior.

Modifications:
- Modify HttpConversionUtil to support compressing and decompressing the COOKIE values

Result:
HttpConversionUtil is compatible with https://tools.ietf.org/html/rfc7540#section-8.1.2.5)
Fixes https://github.com/netty/netty/issues/4457
2015-12-08 20:00:31 -08:00
nmittler
8cd259896e No HTTP/2 RST_STREAM if no prior HEADERS were sent
Motivation:

Because we flow control HEADERS frames, it's possible that an intermediate error can result in a RST_STREAM frame being sent for a frame that the other endpoint is not yet aware of. This is a violation of the spec and will either result in spammy logs at the other endpoint or broken connections.

Modifications:

Modified the HTTP/2 handler so that it only sends RST_STREAM if it has sent at least one HEADERS frame to the remote endpoint for the stream.

Result:

Fixes 
2015-11-25 13:46:32 -08:00
nmittler
dbaeb3314e Allow HTTP2 frame writer to accept arbitrarily large frames
Motivation:

The encoder is currently responsible for chunking frames when writing in order to conform to max frame size. The frame writer would be a better place for this since it could perform a reuse the same promise aggregator for all the write and could also perform a single allocation for all of the frame headers.

Modifications:

Modified `DefaultHttp2FrameWriter` to perform the chunking and modified the contract in the `Http2FrameWriter` interface. Modified `DefaultHttp2ConnectionEncoder` to send give all allocated bytes to the writer.

Result:

Fixes 
2015-11-24 11:44:06 -08:00
nmittler
79ab756fa3 Use a single queue in UniformStreamByteDistributor
Motivation:

The UniformStreamByteDistributor currently processes all zero-length frames, regardless of add order. This means that we would always send HEADERS for all streams, possibly taking away bandwidth for streams that actually have data.

Modifications:

Empty frames are now treated the same as any other frame except that the algorithm will pop off the any empty frames at the head of the queue.

Result:

Empty frames require no extra processing.
2015-11-24 08:11:23 -08:00
Scott Mitchell
cfcee5798d Adjustable size of DefaultHeaders array
Motivation:
DefaultHeaders creates an array of size 16 for all headers. This may waste a good deal of memory if applications only have a small number of headers. This memory may be critical when the number of connections grows large.

Modifications:
- Make the size of the array for DefaultHeaders configurable

Result:
Applications can control the size of the DefaultHeaders array and save memory.
2015-11-23 15:38:08 -08:00
nmittler
2a2059d976 Adding UniformStreamByteDistributor
Motivation:

The current priority algorithm can yield poor per-stream goodput when either the number of streams is high or the connection window is small. When all priorities are the same (i.e. priority is disabled), we should be able to do better.

Modifications:

Added a new UniformStreamByteDistributor that ignores priority entirely and manages a queue of streams.  Each stream is allocated a minimum of 1KiB on each iteration.

Result:

Improved goodput when priority is not used.
2015-11-19 16:49:12 -08:00
nmittler
96f9b0b91b Remote flow controller incorrectly updates stream state
Motivation:

The `DefaultHttp2RemoteFlowController` does not correctly determine `hasFrame` when updating the stream state for the distributor. Adding a check to enforce `hasFrame` when `streamableBytes > 0` causes several test failures.

Modifications:

Modified `DefaultHttp2RemoteFlowController` to simplify the writing logic and to correct the bookkeeping for `hasFrame`.

Result:

The distributors are always called with valid arguments.
2015-11-18 11:32:18 -08:00
nmittler
8accc52b03 Forking Twitter's hpack
Motivation:

The twitter hpack project does not have the support that it used to have.  See discussion here: https://github.com/netty/netty/issues/4403.

Modifications:

Created a new module in Netty and copied the latest from twitter hpack master.

Result:

Netty no longer depends on twitter hpack.
2015-11-14 10:13:32 -08:00