Commit Graph

380 Commits

Author SHA1 Message Date
Norman Maurer
b967805f32 [maven-release-plugin] prepare for next development iteration 2017-08-24 15:38:22 +02:00
Norman Maurer
da8e010a42 [maven-release-plugin] prepare release netty-4.1.15.Final 2017-08-24 15:37:59 +02:00
Nikolay Fedorovskikh
ba27456653 Use the index-based AsciiString constructor instead of substring()
Motivation:
The construction `new AsciiString(string.substring(...))` can be replaced with the `new AsciiString(string, start, length)` to avoid extra allocation.

Modifications:
Apply the described replacement in `HttpConversionUtil#setHttp2Authority`.

Result:
Less allocations.
2017-08-18 09:48:05 +02:00
Nikolay Fedorovskikh
4875a2aad4 Immediate caching the strings wrapped to AsciiString
Motivation:
The `AsciiString#toString` method calculate string value and cache it into field. If an `AsciiString` created from the `String` value, we can avoid rebuilding strings if we cache them immediately when creating `AsciiString`. It would be useful for constants strings, which already stored in the JVMs string table, or in cases where an unavoidable `#toString `method call is assumed.

Modifications:
- Add new static method `AsciiString#cache(String)` which save string value into cache field.
- Apply a "benign" data race in the `#hashCode` and `#toString` methods.

Result:
Less memory usage in some `AsciiString` use cases.
2017-08-15 06:22:14 +02:00
Norman Maurer
08284dbbcd Ensure we call promise.setUncancellable() before trying to process in DefaultHttp2StreamChannel.
Motivation:

We should call promise.setUncancellable() in DefaultHttp2StreamChannel.Unsafe impl to detect if the operation was cancelled.

Modifications:

Add promise.setUncancellable() calls

Result:

More correct handling of cancelled promises
2017-08-12 09:19:36 +02:00
Norman Maurer
74f24a5c19 Finish work on http2 child channel implementation and http2 frame api.
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.
2017-08-11 12:41:28 +02:00
buchgr
3a2b462a67 Remove the concept of pending streams. The close future can only be accessed once a stream is active. 2017-08-11 12:41:28 +02:00
buchgr
5380c7c3e3 HTTP/2 Child Channel and FrameCodec Feature Parity.
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.
2017-08-11 12:41:28 +02:00
Norman Maurer
348745608f Fix incorrect javadocs in Http2RemoteFlowController
Motivation:

The javadocs of Http2RemoteFlowController.isWritable(...) are incorrect.

Modifications:

Update javadocs to reflect reality.

Result:

Correct javadocs.
2017-08-08 07:47:18 +02:00
Scott Mitchell
0f8ecdcad5 Http2FrameLogger frame labels incorrect
Motivation:
The labels identifying the frame types in Http2FrameLogger are not always correct.

Modification:
- Correct the string labels to indicate the right frame type in Http2FrameLogger

Result:
Logs are more correct.
2017-08-07 10:24:17 -07:00
Norman Maurer
d56f403c69 First call channelReadComplete(...) before flush(...) for better performance
Motivation:

In Http2ConnectionHandler we call flush(...) in channelReadComplete(...) to ensure we update the flow-controller and write stuff to the remote peer. We should better flip the order and so may be able to pick up more bytes.

Modifications:

Change order of calls.

Result:

Better performance
2017-08-04 11:55:35 +02:00
Norman Maurer
52f384b37f [maven-release-plugin] prepare for next development iteration 2017-08-02 12:55:10 +00:00
Norman Maurer
8cc1071881 [maven-release-plugin] prepare release netty-4.1.14.Final 2017-08-02 12:54:51 +00:00
chhsiao90
8320a45c15 Configures HTTP2 pipeline with more proper way
Motivation:

When we use pipeline.replace and we still had ongoing inbound, then
there will be some problem that inbound message would go to wrong
handlers. So we add handler first, and remove self after add, so that
the next handler will be the correct one.

Modifications:

Uses remove after addAfter instead of replace.

Result:

Fixed #6881
2017-08-02 06:58:55 +02:00
Norman Maurer
2988fb8eeb Ensure Http2FrameCodec uses Http2Settings.defaultSettings()
Motivation:

Http2FrameCodec should use Http2Settings.defaultSettings() when no Http2Settings were specified by the user.

Modifications:

Replace new Http2Settings() with Http2Settings.defaultSettings()

Result:

Use correct Http2Settings by default when using Http2FrameCodec in all cases.
2017-08-01 07:07:05 +02:00
Vladimir Gordiychuk
fe8ecea366 Http2FrameLogger avoid hex dump of the ByteBufs when log disabled
Motivation:

Currentry logger create hex dump even if log write will not apply.
It's unecessary GC overhead.

Modifications:

Restore optimization from #3492

Result:

Fixes #7025
2017-07-26 21:21:04 +02:00
Spencer Fang
732b145842 Http2ConnectionHandler: allow graceful shutdown to wait forever
Motivation:

There should be a way to allow graceful shutdown to wait for all open streams to close without a timeout. Using gracefulShutdownTimeoutMillis with a large value is a bit of a hack, and has a gotcha that sufficiently large values will overflow the long, resulting in a ClosingChannelFutureListener that executes immediately.

Modification:

Allow to use gracefulShutdownTimeoutMillis(-1) to express waiting until all streams are closed.

Result:

We can now shutdown the connection without a forced timeout.
2017-07-26 20:40:24 +02:00
Scott Mitchell
a91df58ca1 HTTP/2 enforce HTTP message flow
Motivation:
codec-http2 currently does not strictly enforce the HTTP/1.x semantics with respect to the number of headers defined in RFC 7540 Section 8.1 [1]. We currently don't validate the number of headers nor do we validate that the trailing headers should indicate EOS.

[1] https://tools.ietf.org/html/rfc7540#section-8.1

Modifications:
- DefaultHttp2ConnectionDecoder should only allow decoding of a single headers and a single trailers
- DefaultHttp2ConnectionEncoder should only allow encoding of a single headers and optionally a single trailers

Result:
Constraints of RFC 7540 restricting the number of headers/trailers is enforced.
2017-07-19 13:37:23 -07:00
Norman Maurer
dbd82e07b1 Let Http2ServerUpgradeCodec support Http2FrameCodec
Motivation:

Http2ServerUpgradeCodec should support Http2FrameCodec.

Modifications:

- Add support for Http2FrameCodec
- Add example that uses Http2FrameCodec

Result:

More flexible use of Http2ServerUpgradeCodec
2017-07-19 11:12:10 +02:00
Norman Maurer
2a376eeb1b [maven-release-plugin] prepare for next development iteration 2017-07-06 13:24:06 +02:00
Norman Maurer
c7f8168324 [maven-release-plugin] prepare release netty-4.1.13.Final 2017-07-06 13:23:51 +02:00
Scott Mitchell
449befa003 Workaround IBM's J9 JVM getSupportedCipherSuites() returning SSL_ prefix cipher names
Motivation:
IBM's J9 JVM utilizes a custom cipher naming scheme with SSL_ prefix [1] instead of the TLS_ prefix defined by TLS RFCs and the JSSE cihper suite names [2]. IBM's documentation says that the SSL_ prefix are "interchangeable" with cipher names with the TLS_ prefix [1]. To work around this issue we parse the supported cipher list and see an SSL_ prefix we can also add the same cipher with the TLS_ prefix. For more details see a discussion on IBM's forums [3] and IBM's issue tracker [4].

[1] https://www.ibm.com/support/knowledgecenter/en/SSYKE2_8.0.0/com.ibm.java.security.component.80.doc/security-component/jsse2Docs/ciphersuites.html
[2] http://docs.oracle.com/javase/8/docs/technotes/guides/security/StandardNames.html#ciphersuites
[3] https://www.ibm.com/developerworks/community/forums/html/topic?id=9b5a56a9-fa46-4031-b33b-df91e28d77c2
[4] https://www.ibm.com/developerworks/rfe/execute?use_case=viewRfe&CR_ID=71770

Modifications:
- When parsing the supported cipher list to get the supported ciphers and we encounter a SSL_ prefix we should also add a TLS_ prefix cipher.
- Remove SSL_ prefix ciphers from Http2SecurityUtil.

Result:
Work around for IBM JVM's custom naming scheme covers more cases for supported cipher suites.
2017-07-05 09:05:42 -04:00
Scott Mitchell
bc46a99eaa DefaultHttp2ConnectionEncoder#writeHeaders shouldn't send GO_AWAY if stream is closed
Motivation:
DefaultHttp2ConnectionEncoder#writeHeaders attempts to find a stream object, and if one doesn't exist it tries to create one. However in the event that the local endpoint has received a RST_STREAM frame before writing the response headers we attempt to create a stream. Since this stream ID is for the incorrect endpoint we then generate a GO_AWAY for what appears to be a protocol error, but can instead be failed locally.

Modifications:
- Just fail the local promise in the above situation instead of sending a GO_AWAY

Result:
Less severe consequences if the server asynchronously sends headers after a RST_STREAM has been received.
Fixes https://github.com/netty/netty/issues/6906.
2017-06-28 12:06:00 -04:00
Scott Mitchell
d3c44ef985 Update Http2SecurityUtil cipher suites
Motivation:
Mozilla's Server Side cipher suite recommendations have been updated [1].

[1] https://wiki.mozilla.org/Security/Server_Side_TLS#Modern_compatibility

Modifications:
- Update Http2SecurityUtil to exclude older ciphers.
- Remove support for DHE ciphersuites because they are now Intermediate and BoringSSL dropped support for these ciphers [2]

[2] https://boringssl.googlesource.com/boringssl/+/7e06de5d2d1b53c57c0c81e8d6ba4122b64cf626

Result:
Updated default ciphers for HTTP/2.
2017-06-28 11:40:28 -04:00
Nikolay Fedorovskikh
01eb428b39 Move methods for decode hex dump into StringUtil
Motivation:

PR #6811 introduced a public utility methods to decode hex dump and its parts, but they are not visible from netty-common.

Modifications:

1. Move the `decodeHexByte`, `decodeHexDump` and `decodeHexNibble` methods into `StringUtils`.
2. Apply these methods where applicable.
3. Remove similar methods from other locations (e.g. `HpackHex` test class).

Result:

Less code duplication.
2017-06-23 18:52:42 +02:00
Scott Mitchell
5934ae8fd2 Http2FrameLogger Updates
Motivation:
The Http2FrameLogger uses a custom format when logging events. We should use the more familiar format of 'channel event type: details' and single line logging for more consistent debugging.

Modifications:
- Http2FrameLogger should not use a StringBuilder and instead should directly use the Logger
- Http2FrameLogger should use the more consistent format defined above

Result:
Http2FrameLogger's logging formate is more consistent with other log events.
2017-06-21 17:12:38 -07:00
Nikolay Fedorovskikh
8f8be31226 Remove unnecessary conversions
Motivation:

In a `HttpConversionUtil#toHttp2Headers` a status code conversion can be replaced with using `HttpResponseStatus#codeAsText` method.

Modifications:

Apply `HttpResponseStatus#codeAsText` method.

Result:

Less allocations.
2017-06-21 21:53:48 +02:00
Norman Maurer
575baf5050 Use more aggressive expanding strategy in HpackHuffmanDecoder
Motivation:

Before we always expanded the buffer by the initialCapacity which by default is 32 bytes. This may lead to many expansions of the buffer until we finally reached the point that the buffer can fin everything.

Modifications:

Double the buffer size until the threshold of >= 1024 is hit. After this will grow it by the initialCapacity

Result:

Less expansion of the buffer (and so allocations / copies) when the intialCapacity is not big enough. Fixes [#6864].
2017-06-15 06:56:44 +02:00
Norman Maurer
3f0085c267 Do proper bounds-checking in HpackHuffmanDecoder to reduce overhead of IndexOutOfBoundsException creation
Motivation:

HpackHuffmanDecoder.Decoder did not do any bound-checking but just catched IndexOutOfBoundsException to detect if the array needs to grow. This can be very expensive because of fillInStackTrace()

Modifications:

Add proper bounds checking and grow the array if needed without catching IndexOutOfBoundsException.

Result:

Less overhead if the array needs to grow.
2017-06-13 07:43:29 +02:00
Scott Mitchell
f00638af52 AbstractHttp2ConnectionHandlerBuilder support for HPACK huffman decoder initial size
Motivation:
Depending on the use case it may make sense to increase or decrease the initial size of the buffer used during the HPACK huffman decode process. This is currently not exposed through the AbstractHttp2ConnectionHandlerBuilder.

Modifications:
- Add a method to AbstractHttp2ConnectionHandlerBuilder which allows the initial size of the buffer used during the HPACK huffman decode prcoess to be configured.

Result:
AbstractHttp2ConnectionHandlerBuilder provides more control of codec-http2 knobs.
2017-06-12 16:36:43 -07:00
Norman Maurer
fd67a2354d [maven-release-plugin] prepare for next development iteration 2017-06-08 21:06:24 +02:00
Norman Maurer
3acd5c68ea [maven-release-plugin] prepare release netty-4.1.12.Final 2017-06-08 21:06:01 +02:00
Norman Maurer
0db2901f4d [maven-release-plugin] prepare for next development iteration 2017-05-11 16:00:55 +02:00
Norman Maurer
f7a19d330c [maven-release-plugin] prepare release netty-4.1.11.Final 2017-05-11 16:00:16 +02:00
Nikolay Fedorovskikh
94e9448ae3 Simplify JUnit assertions
Motivation:

Some JUnit assert calls can be replaced by simpler.

Modifications:

Replacement with a more suitable methods.

Result:

More informative JUnit reports.
2017-05-09 20:19:10 +02:00
Moses Nakamura
cf26227c6c Supply a builder for Http2Codec
Motivation:

DefaultHttp2FrameWriter has constructors that it would be a hassle to
expose as configuration parameters on Http2Codec. We should instead
make a builder for Http2Codec.

Modifications:

Get rid of the public constructors on Http2Codec and instead make sure
you can always use the builder where you would have used the constructor
before.

Result:

Http2Codec can be configured more flexibly, and the SensitivityDetector
can be configured.
2017-05-05 09:32:46 -07:00
Norman Maurer
6915ec3bb9 [maven-release-plugin] prepare for next development iteration 2017-04-29 14:10:00 +02:00
Norman Maurer
f30f242fee [maven-release-plugin] prepare release netty-4.1.10.Final 2017-04-29 14:09:32 +02:00
Scott Mitchell
d21f2adb98 HTTP/2 StreamByteDistributor improve parameter validation
Motivation:
Each StreamByteDistributor may allow for priority in different ways, but there are certain characteristics which are invalid regardless of the distribution algorithm. We should validate these invalid characteristics at the flow controller level.

Modifications:
- Disallow negative stream IDs from being used. These streams may be accepted by the WeightedFairQueueByteDistributor and cause state for other valid streams to be evicted.
- Improve unit tests to verify limits are enforced.

Result:
Boundary conditions related to the priority parameters are validated more strictly.
2017-04-24 17:17:18 -07:00
Eric Anderson
799350c369 Fix HTTP/2 dependency tree corruption
Motivation:

Chrome was randomly getting stuck loading the tiles examples.
Investigation showed that the Netty flow controller thought it had
nothing to send for the connection even though some streams has queued
data and window available.

Modifications:

Fixed an accounting error where an implicitly created parent was not
being added to the dependency tree, thus it and all of its children were
orphaned from the connection's tree and would never have data written.

Result:

Fixes #6621
2017-04-22 08:26:47 -07:00
Eric Anderson
c6ad9338b3 Avoid infinite loop in HTTP/2 distributor toString()
Motivation:

Although effectively unused, the toString() of
WeightedFairQueueByteDistributor.State is useful for debugging. It
accidentally had an infinite loop, as it would recurse infinitely
between a parent and its child, which makes it less useful for
debugging.

Modifications:

Prune the infinite loop by using the parent's streamId instead of the
parent's toString().

Result:

Faster, less stack-overflowing toString()
2017-04-22 08:16:41 -07:00
Nikolay Fedorovskikh
0692bf1b6a fix the typos 2017-04-20 04:56:09 +02:00
Scott Mitchell
1bc5bc69e3 HTTP/2 Allow more time for EventLoopGroup to shutdown in test that use LocalChannel
Motivation:
The CI servers have reported leaks while building the HTTP/2 unit tests. The unit tests attempt to wait for the channels to be closed before exiting the test, but we should wait in case there are any tasks pending on the EventLoopGroup's task queues.

Modifications:
- Change the Future.sync() operations to Future.syncUninterruptibly()
- HTTP/2 unit tests which use local channel should wait for 5 seconds before shutting down the EventLoopGroups

Result:
More likely that any cleanup related tasks will execute before the unit tests are shutdown.
2017-04-06 17:56:21 -07:00
Scott Mitchell
225d10e1ad HTTP/2 Make DefaultHttp2HeadersDecoder's Http2Headers object creation extensible
Motivation:
It is generally useful to override DefaultHttp2HeadersDecoder's creation of a new Http2Headers object so more optimized versions can be substituted if the use case allows for it.

Modifications:
- DefaultHttp2HeadersDecoder should support an overridable method to generate the new Http2Headers object for each decode operation

Result:
DefaultHttp2HeadersDecoder is more extensible.
Fixes https://github.com/netty/netty/issues/6591.
2017-04-03 11:19:14 -07:00
Scott Mitchell
e8da5e5162 Revert "Expose HTTP/2 HpackDecoder (#6589)"
This reverts commit f4c635d30b.
2017-04-03 11:19:09 -07:00
Scott Mitchell
36c6a61d33 HTTP/2 remove unnecessary buffer operations
Motivation:
codec-http2 has some helper methods to write to ByteBuf in a big endian fashion. This is the default memory structure for ByteBuf so these helper methods are not necessary.

Modifications:
- remove writeUnsignedInt and writeUnsignedShort

Result:
codec-http2 has less ByteBuf helper methods which are not necessary.
2017-03-31 15:23:39 -07:00
chhsiao90
0ee36fef00 Accept two ways to start HTTP/2 over clear text
Motivation:

HTTP/2 support two ways to start on a no-tls tcp connection,
http/1.1 upgrade and prior knowlege methodology to start HTTP/2.
Currently, the http2-server from example only support
starting by upgrade. I think we can do a simple dispatch by peek first
bytes from inbound that match to prior knowledge preface or not and
determine which handlers to set into pipeline.

Modifications:

Add ClearTextHttp2ServerUpgradeHandler to support start HTTP/2 via clear
text with two approach. And update example/http2-server to support
this functionality.

Result:

netty HTTP/2 and the example http2-server accept for two ways to start
HTTP/2 over clear text.

Fixed memory leak problem

Update fields to final

Rename ClearText to cleartext

Addressed comments for code improvement

- Always prefer static, final, and private if possible
- Add UnstableApi annotation
- Used EmbeddedChannel.readInbound instead of unhandled inbound handler
- More assertion

Update javadoc for CleartextHttp2ServerUpgradeHandler

Rename ClearTextHttp2ServerUpgradeHandler to CleartextHttp2ServerUpgradeHandler

Removed redundant code about configure pipeline

nit: PriorKnowledgeHandler

Removed Mockito.spy, investigate conn state instead

Add Http2UpgradeEvent

Check null of the constructor arguments

Rename Http2UpgradeEvent to PriorKnowledgeUpgradeEvent

Update unit test
2017-03-31 15:21:48 -07:00
Nathan Mittler
f4c635d30b Expose HTTP/2 HpackDecoder (#6589)
Motivation:

gRPC (and potentially other libraries) has an optimized header processor that requires direct access to the HpackDecoder.

Modifications:

Make the HpackDecoder and its constructors public.

Result:

Fixes #6579
2017-03-31 10:51:01 -07:00
Norman Maurer
40bead56c4 Revert "http2: Http2StreamChannel now shares options of its parent channel"
This reverts commit 7467106630.
2017-03-31 07:45:56 +02:00
Vladimir Kostyukov
7467106630 http2: Http2StreamChannel now shares options of its parent channel
Motivation

Http2StreamChannel ignores options of its parent channel when being created. That leads to surprising results when, for example, unpooled allocator could be silently replaced with pooled allocator (default setting).

Modification

Copy parent channel's options over to the Http2StreamChannel.

Result

Channel options are now consistent between Http2StreamChannel and its parent channel. Newly added test passes on this branch and fails on master. Fixes #6551.
2017-03-23 21:25:54 -07:00