Motivation:
The usage of Invocation level for JMH fixture methods (setup/teardown) inccurs in a significant overhead
in the benchmark time (see org.openjdk.jmh.annotations.Level documentation).
In the case of CodecInputListBenchmark, benchmarks are far too small (less than 50ns) and the Invocation
level setup offsets the measurement considerably.
On such cases, the recommended fix patch is to include the setup/teardown code in the benchmark method.
Modifications:
Include the setup/teardown code in the relevant benchmark methods.
Remove the setup/teardown methods from the benchmark class.
Result:
We run the entire benchmark 10 times with default parameters we observed:
- ArrayList benchmark affected directly by JMH overhead is now from 15-80% faster.
- CodecList benchmark is now 50% faster than original (even with the setup code being measured).
- Recyclable ArrayList is ~30% slower.
- All benchmarks have significant different means (ANOVA) and medians (Moore)
Mode: Throughput (Higher the better)
Method Full params Factor Modified (Median) Original (Median)
recyclableArrayList (elements = 1) 0.615520967 21719082.75 35285691.2
recyclableArrayList (elements = 4) 0.699553431 17149442.76 24514843.31
arrayList (elements = 4) 1.152666631 27120407.18 23528404.88
codecOutList (elements = 1) 1.527275908 67251089.04 44033359.47
codecOutList (elements = 4) 1.596917095 59174088.78 37055204.03
arrayList (elements = 1) 1.878616889 62188238.24 33103204.06
Environment:
Tests run on a Computational server with CPU: E5-1660-3.3GHZ (6 cores + HT), 64 GB RAM.
Motivation:
The usage of Invocation level for JMH fixture methods (setup/teardown) inccurs in a significant impact in
in the benchmark time (see org.openjdk.jmh.annotations.Level documentation).
When the benchmark and the setup/teardown is too small (less than a milisecond) the Invocation level might saturate the system with
timestamp requests and iteration synchronizations which introduce artificial latency, throughput, and scalability bottlenecks.
In the HeadersBenchmark, all benchmarks take less than 100ns and the Invocation level setup offsets the measurement considerably.
As fixture methods is defined for the entire class, this overhead also impacts every single benchmark in this class, not only
the ones that use the emptyHttpHeaders object (cleaned in the setup).
The recommended fix patch here is to include the setup/teardown code in the benchmark where the object is used.
Modifications:
Include the setup/teardown code in the relevant benchmark methods.
Remove the setup/teardown method of Invocation level from the benchmark class.
Result:
We run all benchmarks from HeadersBenchmark 10 times with default parameter, we observe:
- Benchmarks that were not directly affected by the fix patch, improved execution time.
For instance, http2Remove with (exampleHeader = THREE) had its median reported as 2x faster than the original version.
- Benchmarks that had the setup code inserted (eg. http2AddAllFastest) did not suffer a significant punch in the execution time,
as the benchmarks are not dominated by the clear().
Environment:
Tests run on a Computational server with CPU: E5-1660-3.3GHZ (6 cores + HT), 64 GB RAM.
Motivation:
We deviate from the AbstractChannel implementation on deregistration by
failing the provided promise if the channel is already deregistered. In
contrast, AbstractChannel will always set the promise to successfully
done.
Modification:
Change the
Http2MultiplexCodec.DefaultHttp2StreamChannel.Http2ChannelUnsafe to
always set the promise provided to deregister as done as is the
case in AbstractChannel.
Motivation:
If a wrapped cookie value with an invalid charcater is passed to the strict
encoder, an exception is thrown on validation but the error message contains
a character at the wrong position.
Modifications:
Print `unwrappedValue.charAt(pos)` instead of `value.charAt(pos)`.
Result:
The exception indicates the correct invalid character in the unwrapped cookie.
Motivation:
Eliminate avoidable backing array reallocations when constructing
composite ByteBufs from existing buffer arrays/Iterables. This also
applies to the Unpooled.wrappedBuffer(...) methods.
Modifications:
Ensure the initial components ComponentList is sized at least as large
as the provided buffer array/Iterable in the CompositeByteBuffer
constructors.
In single-arg Unpooled.wrappedBuffer(...) methods, set maxNumComponents
to the count of provided buffers, rather than a fixed default of 16. It
seems likely that most usage of these involves wrapping a list without
subsequent modification, particularly since they return a ByteBuf rather
than CompositeByteBuf. If a different/larger max is required there are
already the wrappedBuffer(int, ...) variants.
In fact the current behaviour could be considered inconsistent - if you
call Unpooled.wrappedBuffer(int, ByteBuf) with a single buffer, you
might expect to subsequently be able to add buffers to it (since you
specified a max related to consolidation), but it will in fact return
just a slice of the provided ByteBuf.
Result:
Fewer and smaller allocations in some cases when using CompositeByteBufs
or Unpooled.wrappedBuffer(...).
Motivation:
Some of the cipher protocol combos that were used are no longer included in more recent OpenSSL releases.
Modifications:
Remove some combos that were used for testing.
Result:
Tests also pass in more recent OpenSSL versions (1.1.0+).
Motivation:
This reverts commit 4b728cd5bc as it was fixes in Java 11 ea+17.
Modification:
Revert previous added workaround as this is fixed in Java 11 now.
Result:
No more workaround for test included.
Motivation:
netty-tcnative 2.0.9 did not contain all native code for boringssl due a release mistake.
Modifications:
Update to 2.0.10
Result:
Use latest netty-tcnative release.
Motivation:
netty-tcnative 2.0.9.Final was released which fixes a memory leak that can happen if client auth is used via client side.
Modifications:
Update to latest netty-tcnative.
Result:
No more memory leak.
Motivation:
Epoll and Kqueue channels have internal state which forces
a single read operation after channel construction. This
violates the Channel#read() interface which indicates that
data shouldn't be delivered until this method is called.
The behavior is also inconsistent with the NIO transport.
Modifications:
- Epoll and Kqueue shouldn't unconditionally read upon
initialization, and instead should rely upon Channel#read()
or auto_read.
Result:
Epoll and Kqueue are more consistent with NIO.
Motivation:
There is an inconsistency between the order of events in the
StreamChannel implementation in Http2MultiplexCodec and other Channel
implementations that extend AbstractChannel where channelInactive and
channelUnregistered events are not performed 'later'. This can cause an
unexected order of events for ChannelHandler implementations that call
Channel.close() in response to some event.
Modification:
The Http2MultiplexCodec.DefaultHttp2StreamChannel.Http2ChannelUnsafe was
modified to bounce the deregistration and channelInactive events through
the parent channels EventLoop.
Result:
Stream events are now in the proper order.
Fixes#8018.
Motivation
There is a cost to concatenating strings and calling methods that will be wasted if the Logger's level is not enabled.
Modifications
Check if Log level is enabled before producing log statement. These are just a few cases found by RegEx'ing in the code.
Result
Tiny bit more efficient code.
Motivation:
Currently there is not a clear way to provide a byte array to a netty
ByteBuf and be informed when it is released. This is a would be a
valuable addition for projects that integrate with netty but also pool
their own byte arrays.
Modification:
Modified the UnpooledHeapByteBuf class so that the freeArray method is
protected visibility instead of default. This will allow a user to
subclass the UnpooledHeapByteBuf, provide a byte array, and override
freeArray to return the byte array to a pool when it is called.
Additionally this makes this implementation equivalent to
UnpooledDirectByteBuf (freeDirect is protected).
Additionally allocateArray is also made protect to provide another override
option for subclasses.
Result:
Users can override UnpooledHeapByteBuf#freeArray and
UnpooledHeapByteBuf#allocateArray.
Motivation:
Http2MultiplexCodec doesn't currently have an API for using the response
of a h2c upgrade request.
Modifications:
Add a new API to the Http2MultiplexCodecBuilder which allows for setting
an upgrade handler and wire it into the Http2MultiplexCodec
implementation.
Result:
When using the Http2MultiplexCodec with h2c upgrades the upgrade handler
will get added to the Http2StreamChannel which represents the
half-closed (local) response of stream 1. It is then up to the user to
manage the transition from the IO channel pipeline configuration
necessary for making the h2c upgrade request to a form where it can read
the response from the new stream channel.
Fixes#7947.
Motivation:
The `ByteBuffer emptyPingBuf()` method of Http2CodecUtils is has been dead
code since DefaultHttp2PingFrame switched from using a ByteBuf to represent
the 8 octets to a long.
Modifications:
Remove the method and the unused static ByteBuf.
Result:
Less dead code.
Fixes#8002
Motivation:
When using conscrypt some NPEs were logged, these were fixed in the latest release.
Modifications:
Update to conscrypt 1.1.3.
Result:
Fixes https://github.com/netty/netty/issues/7988.
Motivation
The HttpObjectEncoder raises an IllegalStateException due to an illegal state but doesn't mention what the state was. It could be useful for debugging purposes to figure out what happened.
Modifications
Mention the HttpObjectEncoder's state in the message of the IllegalStateException.
Result
An exception with more information what caused it.
Motivation:
The implementation of CharSequenceValueConverter.convertToByte did not correctly handle AsciiString if the length != 1.
Modifications:
- Only use fast-path for AsciiString with length of 1.
- Add unit tests.
Result:
Fixes https://github.com/netty/netty/issues/7990
Motivation:
SslHandlerTest tried to get access to the SslHandler in the pipeline via pipeline.get(...) which may return null if the channel was already closed and so the pipeline was teared down.
This showed up in a test run as:
```
-------------------------------------------------------------------------------
Test set: io.netty.handler.ssl.SslHandlerTest
-------------------------------------------------------------------------------
Tests run: 17, Failures: 0, Errors: 1, Skipped: 1, Time elapsed: 0.802 sec <<< FAILURE! - in io.netty.handler.ssl.SslHandlerTest
testCloseOnHandshakeFailure(io.netty.handler.ssl.SslHandlerTest) Time elapsed: 0.188 sec <<< ERROR!
java.lang.NullPointerException
at io.netty.handler.ssl.SslHandlerTest.testCloseOnHandshakeFailure(SslHandlerTest.java:640)
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.base/java.lang.reflect.Method.invoke(Method.java:564)
at org.junit.runners.model.FrameworkMethod$1.runReflectiveCall(FrameworkMethod.java:50)
at org.junit.internal.runners.model.ReflectiveCallable.run(ReflectiveCallable.java:12)
at org.junit.runners.model.FrameworkMethod.invokeExplosively(FrameworkMethod.java:47)
at org.junit.internal.runners.statements.InvokeMethod.evaluate(InvokeMethod.java:17)
at org.junit.internal.runners.statements.FailOnTimeout$CallableStatement.call(FailOnTimeout.java:298)
at org.junit.internal.runners.statements.FailOnTimeout$CallableStatement.call(FailOnTimeout.java:292)
at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264)
at java.base/java.lang.Thread.run(Thread.java:844)
```
Modifications:
Use an AtomicReference to propagate the SslHandler instance to the outer scope.
Result:
No more NPE.
Motivation:
Currently, when passing custom headers to a WebSocketClientHandshaker,
if values are added for headers that are reserved for use in the
websocket handshake performed with the server, these custom values can
be used by the server to compute the websocket handshake challenge. If
the server computes the response to the challenge with the custom header
values, rather than the values computed by the client handshaker, the
handshake may fail.
Modifications:
Update the client handshaker implementations to add the custom header
values first, and then set the reserved websocket header values.
Result:
Reserved websocket handshake headers, if present in the custom headers
passed to the client handshaker, will not be propagated to the server.
Instead the client handshaker will propagate the values it generates.
Fixes#7973.
Motivation:
If we can not replace the internal used Set of the Selector there is no need to create an SelectedSelectionKeySet instance.
Modification:
Only create SelectedSelectionKeySet if we will replace the internal set.
Result:
Less object creation in some cases and cleaner code.
Motivation:
We should allow to schedule tasks with a delay up to Long.MAX_VALUE as we did pre 4.1.25.Final.
Modifications:
Just ensure we not overflow and put the correct max limits in place when schedule a timer. At worse we will get a wakeup to early and then schedule a new timeout.
Result:
Fixes https://github.com/netty/netty/issues/7970.
Motivation:
Java11 disallow draining any remaining bytes from the socket if a write causes a connection reset. This should be completely safe to do. At the moment if a write is causing a connection-reset you basically loose all the pending bytes that are sitting on the socket and are waiting to be read.
This happens because SocketOutputStream.write(…) may call AbstractPlainSocketImpl.setConnectionReset(…). Once this method is called any read(…) call will just throw a SocketException without even attempt to read any remaining data.
This is related:
- https://bugs.openjdk.java.net/browse/JDK-8199329
- http://hg.openjdk.java.net/jdk/jdk/rev/92cca24c8807
- http://mail.openjdk.java.net/pipermail/net-dev/2018-May/011511.html
Modifications:
Tolarate if remaining bytes could not be read when using OIO.
Result:
Be able to build Netty and run testsuite while using Java11
Motivation:
We also need to run our tests while using boringssl-static to ensure everything works when using it. Beside this its sometimes useful to be able to just get a shell and so interactive work in the docker instance.
Modifications:
- Add configs for shell
- Add configs for testing with boringssl-static
- Ensure we not share .m2 when running tests
Result:
More complete docker setup.
Motivation:
The maven surefire plugin will trim stacktraces by default which makes these kind of use-less when trying to understand why an test failed because one was thrown.
Modifications:
Configure the plugin to not trim the stacktrace.
Result:
Easier to debug test-failures.
Motivation:
This is a followup for #7860. In the fix for #7860 we only partly fixed the problem as Http2UnknownFrame did not correctly extend HttpStreamFrame and so only worked when using the Http2FrameCodec. We need to have it extend HttpStreamFrame as otherwise Http2MultiplexCodec will reject to handle it correctly.
Modifications:
- Let Http2UnknownFrame extend HttpStreamFrame
- Add unit tests for writing and reading Http2UnkownFrame instances when the Http2MultiplexCodec is used.
Result:
Fixes https://github.com/netty/netty/issues/7969.
motivation: setup for testing across different permutations of linux and java versions
changes:
* refactor docker file to allow dynamic versions of centos and java
* add docker compose driver files for centos 6, 7 and java 1.8, 1.9, 1.10, 1.11
* update instructions
Motivation:
The websockets abstract test suite does not run against the 08
implementation in the 08 version of the test suite.
Modifications:
Update the WebSocketClientHandshaker08Test to instantiate a new
WebSocketClientHandshaker08 rather than an 07 handshaker.
Result:
The WebSocketClientHandshaker08Test now tests the 08 implementation.
Motivation:
On J9 / OpenJ9 netty initializes this value with 64M, even the direct accessible memory is actually unbounded.
Modifications:
Skip usage of VM.maxDirectMemory() on J9 / OpenJ9
Result:
More correct direct memory limit detection. Fixes#7654.
Motivation:
A long time ago we deprecated AUTO_CLOSE but it turned out this feature is still useful because if a write error is detected there still maybe data to read, and if we close the channel automatically we will lose data
Modifications:
- Remove `@Deprecated` tag for AUTO_CLOSE, setAutoClose(...) and isAutoClose(...)
- Fix javadocs on ChannelConfig to correctly tell the default value of AUTO_CLOSE.
Result:
Less warnings.
* Read until all data is consumed when EOF is detected even if readPending is false and auto-read is disabled.
Motivation:
We should better always notify the user of EOF even if the user did not request any data as otherwise we may never be notified when the remote peer closes the connection. This should be ok as the amount of extra data we may read and so fire through the pipeline is limited by SO_RECVBUF.
Modifications:
- Always drain the socket when EOF is detected.
- Add testcase
Result:
No risk for the user to be not notified of EOF.
Motivation:
Currently, on recipt of a PongWebSocketFrame, the
WebSocketProtocolHandler will drop the frame, rather than passing it
along so it can be referenced by other handlers.
Modifications:
Add boolean field to WebSocketProtocolHandler to indicate whether Pong
frames should be dropped or propagated, defaulting to "true" to preserve
existing functionality.
Add new constructors to the client and server implementations of
WebSocketProtocolHandler that allow for overriding the behavior for the
handling of Pong frames.
Result:
PongWebSocketFrames are passed along the channel, if specified.
Motivation:
DefaultHttpResponse did not respect its status when compute the hashCode and check for equality.
Modifications:
Correctly implement hashCode and equals
Result:
Fixes https://github.com/netty/netty/issues/7964.
Motivation:
We need to ensure we only return from close() after all work is done as otherwise we may close the EventExecutor before we dispatched everything.
Modifications:
Correctly wait on operations to complete before return.
Result:
Fixes https://github.com/netty/netty/issues/7901.
Motivation:
When a sender sends too large of headers it should not unnecessarily
kill the connection, as killing the connection is a heavy-handed
solution while SETTINGS_MAX_HEADER_LIST_SIZE is advisory and may be
ignored.
The maxHeaderListSizeGoAway limit in HpackDecoder is unnecessary because
any headers causing the list to exceeding the max size can simply be
thrown away. In addition, DefaultHttp2FrameReader.HeadersBlockBuilder
limits the entire block to maxHeaderListSizeGoAway. Thus individual
literals are limited to maxHeaderListSizeGoAway.
(Technically, literals are limited to 1.6x maxHeaderListSizeGoAway,
since the canonical Huffman code has a maximum compression ratio of
.625. However, the "unnecessary" limit in HpackDecoder was also being
applied to compressed sizes.)
Modifications:
Remove maxHeaderListSizeGoAway checking in HpackDecoder and instead
eagerly throw away any headers causing the list to exceed
maxHeaderListSize.
Result:
Fewer large header cases will trigger connection-killing.
DefaultHttp2FrameReader.HeadersBlockBuilder will still kill the
connection when maxHeaderListSizeGoAway is exceeded, however.
Fixes#7887
Motivation:
When we build the key-material we should use the ByteBufAllocator used by the ReferenceCountedOpenSslEngine when possible.
Modifications:
Whenever we have access to the ReferenceCountedOpenSslEngine we use its allocator.
Result:
Use correct allocator
Motivation:
We currently have only interopt tests for Conscrypt, we should also have non-interopt tests.
Modifications:
Add ConscryptSslEngineTest
Result:
More tests
Motivation:
We added a workaround for Java 11 as it not produced a connect-reset when SO_LINGER with 0 was set and NIO was used. This was fixed in the latest ea release of Java 11:
- http://hg.openjdk.java.net/jdk/jdk/rev/ea54197f4fe4
- https://bugs.openjdk.java.net/browse/JDK-8203059
Modifications:
Revert workaround.
Result:
Test that Java 11 behave the same way as earlier Java versions again.
Motivation:
We did not guard against the case of calling malloc(0) when creating a ByteBuffer without a Cleaner. The problem is that malloc(0) can have different behaviour, it either return a null-pointer or a valid pointer that you can pass to free.
The real problem arise if Unsafe.allocateMemory(0) returns 0 and we use it as the memoryAddress of the ByteBuffer. The problem here is that native libraries test for 0 and handle it as a null-ptr. This is for example true in SSL.bioSetByteBuffer(...) which would throw a NPE when 0 is used as memoryAddress and so produced errors during SSL usage.
Modifications:
- Always allocate 1 byte as minimum (even if we ask for an empty buffer).
- Add unit test.
Result:
No more errors possible because of malloc(0).
Motivation:
We previously did not correctly take into account when we could not wrap (and so produce) the full SSL record with an alert when the SSLEngine was closed.
There are two problems here:
- If we call wrap(...) with an empty dst buffer after closeOutbound() was called we will not notify the user if we could not store the whole SSLRecord into the dst buffer and so we may produce incomplete SSLRecords
Modifications:
Add unit test which failed before.
Result:
Correctly handle the case when the dst buffer is not big enough and and alert needs to be produced.
Motivation:
https://github.com/netty/netty/pull/7943 had a bug which caused to not have the argument passed to the delegating method.
Modifications:
Add argument to release call.
Result:
Correctly delegate method.
Motivation:
Some of the tests failed when using BoringSSL as some protocol / cipher combinations are not supported and it uses a different alert when the cert is not valid yet.
Modification:
- Remove protocol / cipher combos that are not supported by BoringSSL
- Test for different alert when using BoringSSL
Result:
Not test failures when using BoringSSL.
Motivation:
https://github.com/netty/netty/pull/7941 proved that its easy to not correctly clear the error stack sometimes. We should do carefully test this.
Modifications:
Add a new SSLEngine wrapper that is used during tests, which verifies that the error stack is empty after each method call.
Result:
Better testing.
Motivation:
We sometimes did not correctly detect when a protocol is not enabled when using netty-tcnative as we did not take into account when the option flag is 0 (as for example in BoringSSL for SSLv2).
Modifications:
- Correctly take an option flag with 0 into account.
- Add unit tests.
Result:
Fixes https://github.com/netty/netty/issues/7935.
