Motivation:
Recycler is used in many places to reduce GC-pressure but is still not as fast as possible because of the internal datastructures used.
Modification:
- Rewrite Recycler to use a WeakOrderQueue which makes minimal guaranteer about order and visibility for max performance.
- Recycling of the same object multiple times without acquire it will fail.
- Introduce a RecyclableMpscLinkedQueueNode which can be used for MpscLinkedQueueNodes that use Recycler
These changes are based on @belliottsmith 's work that was part of #2504.
Result:
Huge increase in performance.
4.0 branch without this commit:
Benchmark (size) Mode Samples Score Score error Units
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 00000 thrpt 20 116026994.130 2763381.305 ops/s
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 00256 thrpt 20 110823170.627 3007221.464 ops/s
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 01024 thrpt 20 118290272.413 7143962.304 ops/s
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 04096 thrpt 20 120560396.523 6483323.228 ops/s
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 16384 thrpt 20 114726607.428 2960013.108 ops/s
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 65536 thrpt 20 119385917.899 3172913.684 ops/s
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 297.617 sec - in io.netty.microbench.internal.RecyclableArrayListBenchmark
4.0 branch with this commit:
Benchmark (size) Mode Samples Score Score error Units
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 00000 thrpt 20 204158855.315 5031432.145 ops/s
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 00256 thrpt 20 205179685.861 1934137.841 ops/s
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 01024 thrpt 20 209906801.437 8007811.254 ops/s
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 04096 thrpt 20 214288320.053 6413126.689 ops/s
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 16384 thrpt 20 215940902.649 7837706.133 ops/s
i.n.m.i.RecyclableArrayListBenchmark.recycleSameThread 65536 thrpt 20 211141994.206 5017868.542 ops/s
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 297.648 sec - in io.netty.microbench.internal.RecyclableArrayListBenchmark
Motivation:
LocalServerChannel.doClose() calls LocalChannelRegistry.unregister(localAddress); without check if localAddress is null and so produce a NPE when pass null the used ConcurrentHashMapV8
Modification:
Check for localAddress != null before try to remove it from Map. Also added a unit test which showed the stacktrace of the error.
Result:
No more NPE during doClose().
Motivation:
At the moment AbstractBoostrap.bind(...) will always use the GlobalEventExecutor to notify the returned ChannelFuture if the registration is not done yet. This should only be done if the registration fails later. If it completes successful we should just notify with the EventLoop of the Channel.
Modification:
Use EventLoop of the Channel if possible to use the correct Thread to notify and so guaranteer the right order of events.
Result:
Use the correct EventLoop for notification
Motivation:
When Netty runs in a managed environment such as web application server,
Netty needs to provide an explicit way to remove the thread-local
variables it created to prevent class loader leaks.
FastThreadLocal uses different execution paths for storing a
thread-local variable depending on the type of the current thread.
It increases the complexity of thread-local removal.
Modifications:
- Moved FastThreadLocal and FastThreadLocalThread out of the internal
package so that a user can use it.
- FastThreadLocal now keeps track of all thread local variables it has
initialized, and calling FastThreadLocal.removeAll() will remove all
thread-local variables of the caller thread.
- Added FastThreadLocal.size() for diagnostics and tests
- Introduce InternalThreadLocalMap which is a mixture of hard-wired
thread local variable fields and extensible indexed variables
- FastThreadLocal now uses InternalThreadLocalMap to implement a
thread-local variable.
- Added ThreadDeathWatcher.unwatch() so that PooledByteBufAllocator
tells it to stop watching when its thread-local cache has been freed
by FastThreadLocal.removeAll().
- Added FastThreadLocalTest to ensure that removeAll() works
- Added microbenchmark for FastThreadLocal and JDK ThreadLocal
- Upgraded to JMH 0.9
Result:
- A user can remove all thread-local variables Netty created, as long as
he or she did not exit from the current thread. (Note that there's no
way to remove a thread-local variable from outside of the thread.)
- FastThreadLocal exposes more useful operations such as isSet() because
we always implement a thread local variable via InternalThreadLocalMap
instead of falling back to JDK ThreadLocal.
- FastThreadLocalBenchmark shows that this change improves the
performance of FastThreadLocal even more.
Motivation:
The code in ChannelOutboundBuffer can be simplified by using AtomicLongFieldUpdater.addAndGet(...)
Modification:
Replace our manual looping with AtomicLongFieldUpdater.addAndGet(...)
Result:
Cleaner code
Motivation:
If ChannelOutboundBuffer.addFlush() is called multiple times and flushed != unflushed it will still loop through all entries that are not flushed yet even if it is not needed anymore as these were marked uncancellable before.
Modifications:
Check if new messages were added since addFlush() was called and only if this was the case loop through all entries and try to mark the uncancellable.
Result:
Less overhead when ChannelOuboundBuffer.addFlush() is called multiple times without new messages been added.
Motivation:
Allow to make use of our new FastThreadLocal whereever possible
Modification:
Make use of an array to store FastThreadLocals and so allow to also use it in PooledByteBufAllocator that is instanced by users.
The maximal size of the array is configurable per system property to allow to tune it if needed. As default we use 64 entries which should be good enough.
Result:
More flexible usage of FastThreadLocal
Motivation:
Provide a faster ThreadLocal implementation
Modification:
Add a "FastThreadLocal" which uses an EnumMap and a predefined fixed set of possible thread locals (all of the static instances created by netty) that is around 10-20% faster than standard ThreadLocal in my benchmarks (and can be seen having an effect in the direct PooledByteBufAllocator benchmark that uses the DEFAULT ByteBufAllocator which uses this FastThreadLocal, as opposed to normal instantiations that do not, and in the new RecyclableArrayList benchmark);
Result:
Improved performance
Motivation:
Each of DefaultChannelPipeline instance creates an head and tail that wraps a handler. These are used to chain together other DefaultChannelHandlerContext that are created once a new ChannelHandler is added. There are a few things here that can be improved in terms of memory usage and initialization time.
Modification:
- Only generate the name for the tail and head one time as it will never change anyway
- Rename DefaultChannelHandlerContext to AbstractChannelHandlerContext and make it abstract
- Create a new DefaultChannelHandlerContext that is used when a ChannelHandler is added to the DefaultChannelPipeline
- Rename TailHandler to TailContext and HeadHandler to HeadContext and let them extend AbstractChannelHandlerContext. This way we can save 2 object creations per DefaultChannelPipeline
Result:
- Less memory usage because we have 2 less objects per DefaultChannelPipeline
- Faster creation of DefaultChannelPipeline as we not need to generate the name for the head and tail
Motivation:
At the moment ChannelFlushPromiseNotifier.add(....) takes an int value for pendingDataSize, which may be too small as a user may need to use a long. This can for example be useful when a user writes a FileRegion etc. Beside this the notify* method names are kind of missleading as these should not contain *Future* because it is about ChannelPromises.
Modification:
Add a new add(...) method that takes a long for pendingDataSize and @deprecated the old method. Beside this also @deprecated all *Future* methods and add methods that have *Promise* in the method name to better reflect usage.
Result:
ChannelFlushPromiseNotifier can be used with bigger data.
Motivation:
On some ill-configured systems, InetAddress.getLocalHost() fails. NioSocketChannelTest calls java.net.Socket.connect() and it internally invoked InetAddress.getLocalHost(), which causes the test failures in NioSocketChannelTes on such an ill-configured system.
Modifications:
Use NetUtil.LOCALHOST explicitly.
Result:
NioSocketChannelTest should not fail anymore.
Motivation:
At the moment we sometimes use only RecvByteBufAllocator.guess() to guess the next size and the use the ByteBufAllocator.* directly to allocate the buffer. We should always use RecvByteBufAllocator.allocate(...) all the time as this makes the behavior easier to adjust.
Modifications:
Change the read() implementations to make use of RecvByteBufAllocator.
Result:
Behavior is more consistent.
Motivation:
DefaultChannelPipeline.firstContext() should return null when the ipeline is empty. This is not the case atm.
Modification:
Fix incorrect check in DefaultChannelPipeline.firstContext() and add unit tests.
Result:
Correctly return null when DefaultChannelPipeline.firstContext() is called on empty pipeline.
Motivation:
Because Thread.currentThread().interrupt() will unblock Selector.select() we need to take special care when check if we need to rebuild the Selector. If the unblock was caused by the interrupt() we will clear it and move on as this is most likely a bug in a custom ChannelHandler or a library the user makes use of.
Modification:
Clear the interrupt state of the Thread if the Selector was unblock because of an interrupt and the returned keys was 0.
Result:
No more busy loop caused by Thread.currentThread().interrupt()
Motivation:
As discussed in #2250, it will become much less complicated to implement
deregistration and reregistration of a channel once #2250 is resolved.
Therefore, there's no need to deprecate deregister() and
channelUnregistered().
Modification:
- Undeprecate deregister() and channelUnregistered()
- Remove SuppressWarnings annotations where applicable
Result:
We (including @jakobbuchgraber) are now ready to play with #2250 at
master
Motivation:
I had the NioSocketChannelTest.testFlushCloseReentrance() fail sometimes on one of my linux installation. This change let it pass all the time.
Modification:
Set the SO_SNDBUF to a small value to force split writes
Result:
Test is passing all the time where it was sometimes fail before.
Motivation:
At the moment it is not possible to deregister a LocalChannel from its EventLoop and register it to another one as the LocalChannel is closed during the deregister.
Modification:
Not close the LocalChannel during dergister
Result:
It is now possible to deregister a LocalChannel and register it to another EventLoop
Motivation:
At the moment it is possible to see a NPE when the LocalSocketChannels doRegister() method is called and the LocalSocketChannels doClose() method is called before the registration was completed.
Modifications:
Make sure we delay the actual close until the registration task was executed.
Result:
No more NPE
Motivation:
At the moment ChanneConfig.setAutoRead(false) only is guaranteer to not have an extra channelRead(...) triggered when used from within the channelRead(...) or channelReadComplete(...) method. This is not the correct behaviour as it should also work from other methods that are triggered from within the EventLoop. For example a valid use case is to have it called from within a ChannelFutureListener, which currently not work as expected.
Beside this there is another bug which is kind of related. Currently Channel.read() will not work as expected for OIO as we will stop try to read even if nothing could be read there after one read operation on the socket (when the SO_TIMEOUT kicks in).
Modifications:
Implement the logic the right way for the NIO/OIO/SCTP and native transport, specific to the transport implementation. Also correctly handle Channel.read() for OIO transport by trigger a new read if SO_TIMEOUT was catched.
Result:
It is now also possible to use ChannelConfig.setAutoRead(false) from other methods that are called from within the EventLoop and have direct effect.
Motivation:
At the moment we create a HashMap that holds the MembershipKeys for multicast with every NioDatagramChannel even when most people not need it at al
Modifications:
Lazy create the HashMap when needed.
Result:
Less memory usage and less object creation
Motivation:
When using System.getProperty(...) and various methods to get a ClassLoader it will fail when a SecurityManager is in place.
Modifications:
Use a priveled block if needed. This work is based in the PR #2353 done by @anilsaldhana .
Result:
Code works also when SecurityManager is present
Motivation:
Because we not null out the array entry in the SelectionKey[] which is produced by SelectedSelectionKeySet.flip() we may end up with a few SelectionKeyreferences still hanging around here even after the Channel was closed. As these entries may be present at the end of the SelectionKey[] which is never updated for a long time as not enough SelectionKeys are ready.
Modifications:
Once we access the SelectionKey out of the SelectionKey[] we directly null it out.
Result:
Reference can be GC'ed right away once the Channel was closed.
Motivation:
At the moment we do a Channel.isActive() check in every AbstractChannel.AbstractUnsafe.write(...) call which gives quite some overhead as shown in the profiler when you write fast enough. We can eliminate the check and do something more smart here.
Modifications:
Remove the isActive() check and just check if the ChannelOutboundBuffer was set to null before, which means the Channel was closed. The rest will be handled in flush0() anyway.
Result:
Less overhead when doing many write calls
Motivation:
At the moment an IllegalArgumentException will be thrown if a ChannelPromise is cancelled while propagate through the ChannelPipeline. This is not correct, we should just stop to propagate it as it is valid to cancel at any time.
Modifications:
Stop propagate the operation through the ChannelPipeline once a ChannelPromise is cancelled.
Result:
No more IllegalArgumentException when cancel a ChannelPromise while moving through the ChannelPipeline.
Motivation:
At the moment we use the system-wide default selector provider for this invocation of the Java virtual machine when constructing a new NIO channel, which makes using an alternative SelectorProvider practically useless.
This change allows user specify his/her preferred SelectorProvider.
Modifications:
Add SelectorProvider as a param for current `private static *Channel newSocket` method of NioSocketChannel, NioServerSocketChannel and NioDatagramChannel.
Change default constructors of NioSocketChannel, NioServerSocketChannel and NioDatagramChannel to use DEFAULT_SELECTOR_PROVIDER when calling newSocket(SelectorProvider).
Add new constructors for NioSocketChannel, NioServerSocketChannel and NioDatagramChannel which allow user specify his/her preferred SelectorProvider.
Result:
Now users can specify his/her preferred SelectorProvider when constructing an NIO channel.
Motivation:
Allow the user to create a NioServerSocketChannel from an existing ServerSocketChannel.
Modifications:
Add an extra constructor
Result:
Now the user is be able to create a NioServerSocketChannel from an existing ServerSocketChannel, like he can do with all the other Nio*Channel implemntations.
Motivation:
Ensure the user know the Channel must be closed to release resources like filehandles.
Modifications:
Add some extra javadoc.
Result:
More clear documentation
Motivation:
At the moment we use SocketChannel.open(), ServerSocketChannel.open() and DatagramSocketChannel.open(...) within the constructor of our
NIO channels. This introduces a bottleneck if you create a lot of connections as these calls delegate to SelectorProvider.provider() which
uses synchronized internal. This change removed the bottleneck.
Modifications:
Obtain a static instance of the SelectorProvider and use SelectorProvider.openSocketChannel(), SelectorProvider.openServerSocketChannel() and
SelectorProvider.openDatagramChannel(). This eliminates the bottleneck as SelectorProvider.provider() is not called on every channel creation.
Result:
Less conditions when create new channels.
Motivation:
Remove the synchronization bottleneck and so speed up things
Modifications:
Introduce a ThreadLocal cache that holds mappings between classes of ChannelHandlerAdapater implementations and the result of checking if the @Sharable annotation is present.
This way we only will need to do the real check one time and server the other calls via the cache. A ThreadLocal and WeakHashMap combo is used to implement the cache
as this way we can minimize the conditions while still be sure we not leak class instances in containers.
Result:
Less conditions during adding ChannelHandlerAdapter to the ChannelPipeline
- Allocating and deallocating a direct buffer for I/O is an expensive
operation, so we have to at least have a pool of direct buffers if the
current allocator is not pooled
