Commit Graph

1365 Commits

Author SHA1 Message Date
Trustin Lee
ba92cd4c3e Work around the system configuration issue that causes NioSocketChannelTest to fail
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.
2014-05-28 09:40:59 +09:00
Norman Maurer
dd0782990b [#2485] Use RecvByteBufAllocator for all allocations related to read from Channel
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.
2014-05-10 15:28:02 +02:00
Norman Maurer
d4f074186d [#2454] Correctly return null when DefaultChannelPipeline.firstContext() is called on empty pipeline
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.
2014-05-04 21:09:58 +02:00
Norman Maurer
bf5c928796 Not cause busy loop when interrupt Thread of NioEventLoop
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()
2014-04-28 08:05:46 +02:00
Trustin Lee
bdab831ba5 Undeprecate deregister() and chanelUnregistered()
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
2014-04-25 16:53:59 +09:00
Norman Maurer
c9faaa6e3e Make NioSocketChannelTest more bullet-proof
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.
2014-04-23 09:26:09 +02:00
Norman Maurer
579ef0c4c0 [#2400] Not close LocalChannel during deregister() to allow register to other EventLoop
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
2014-04-21 10:07:14 +02:00
Norman Maurer
c97c8b7b8e [#2144] Fix NPE in Local transport caused by a race
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
2014-04-17 14:24:36 +02:00
Norman Maurer
c2dc993c42 [#2375] [#2404] Fix bug in respecting ChannelConfig.setAutoRead(false) and also fix Channel.read() for OIO
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.
2014-04-17 07:37:34 +02:00
Norman Maurer
e8b5c9ccae [#2390] Minimize memory usage of NioDatagramChannel
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
2014-04-15 06:55:51 +02:00
Norman Maurer
012166803a [#2353] Use a privileged block to get ClassLoader and System property if needed
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
2014-04-08 13:59:03 +02:00
Norman Maurer
2aa35922b4 [#2363] SelectedSelectionKeySet may hold strong reference to SelectionKey after Channel is closed
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.
2014-04-05 19:29:58 +02:00
Norman Maurer
5cd939f634 [#2362] AbstractChannel.AbstractUnsafe.write(...) is slow
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
2014-04-04 09:45:03 +02:00
Norman Maurer
aa74f00006 [#2349] Correctly handle cancelled ChannelPromise in DefaultChannelHandlerContext
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.
2014-03-31 07:29:31 +02:00
CoNDoRip
68670ba195 Allow specifying SelectorProvider when constructing an NIO channel #2311
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.
2014-03-26 21:08:15 +01:00
Norman Maurer
a17d28605a [#2326] Add constructor to NioServerSocketChannel which accepts a ServerSocketChannel
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.
2014-03-16 07:00:44 -07:00
Norman Maurer
535b024913 [#2323] Make it clear a Channel must be closed to release all resources
Motivation:
Ensure the user know the Channel must be closed to release resources like filehandles.

Modifications:
Add some extra javadoc.

Result:
More clear documentation
2014-03-16 06:56:49 -07:00
Norman Maurer
e0b2f34a37 [#2308] Use SelectorProvider.open*() to open NIO channels and so remove condition when create new NIO channels.
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.
2014-03-13 06:46:05 +01:00
Norman Maurer
ef76907422 Remove condition in ChannelHandlerAdapter.isSharable() by caching the result of the annotation lookup.
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
2014-03-12 12:31:22 +01:00
Norman Maurer
c838790400 Corrected inconsistencies in the Javadoc.
Port of 80030493b9f7824c02dc88260c2579afc0aed8bc
2014-03-04 06:34:14 +01:00
Jakob Buchgraber
15d28f0c59 Added asserts to make sure ChannelHandlers are removed from the pipeline 2014-03-03 06:43:32 +01:00
Norman Maurer
bdedde1294 [#1259] Add optimized queue for SCMP pattern and use it in NIO and native transport
This queue also produces less GC then CLQ when make use of OneTimeTask
2014-02-27 11:44:06 +01:00
Norman Maurer
c538fd5098 Fix check to clear READ_OP and EPOLLIN. Part of [#2254] 2014-02-22 20:06:35 +01:00
Norman Maurer
b00e2a19d3 [#2254] Correctly handle Channel.read() and ChannelHandlerContext.read()
This includes also when it is called from channelRead(...) and channelReadComplete(...) methods.
2014-02-22 18:45:36 +01:00
Norman Maurer
7fe10fe635 [#2254] Fix regression in handling autoRead and Channel.read()
This regression was introduced by e0b39159657c9eb711047bc32367537c4870d467
2014-02-21 08:39:41 +01:00
Norman Maurer
32e28950e1 Move marking ChannelPromise for writes uncancellable to addFlush for keep things simple 2014-02-17 16:14:25 +01:00
Trustin Lee
f4074a8811 Fix compilation errors introduced by sloppy merge 2014-02-15 11:31:01 -08:00
Trustin Lee
11a235ffe9 Use thread local direct buffer for I/O when the current allocator is unpooled
- 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
2014-02-15 11:27:04 -08:00
Trustin Lee
b18c8fe688 Determine the default allocator from system property
- Add ByteBufAllocator.DEFAULT
- The default allocator is 'unpooled'
2014-02-14 13:05:57 -08:00
Norman Maurer
69a36b8bea Prettify exception message 2014-02-13 06:46:51 +01:00
Norman Maurer
62ebe88981 Allow to set IoRatio to 100% 2014-02-12 15:22:39 +01:00
Norman Maurer
80e6f9adf4 Correctly respect isAutoRead() and make it consistent across OIO/NIO 2014-02-11 20:46:46 +01:00
Norman Maurer
7041a9238e Allow to cancel non-flushed writes 2014-02-11 19:42:49 +01:00
Trustin Lee
d52dc3b740 Do not warn if failed to mark a void promise as success
- it's always supposed to fail.
2014-02-10 15:03:46 -08:00
Trustin Lee
96b0a949e1 Make most outbound operations cancellable / More robust promise update
- Inspired by #2214 by @normanmaurer
- Call setUncancellable() before performing an outbound operation
- Add safeSetSuccess/Failure() and use them wherever
2014-02-10 14:57:18 -08:00
Trustin Lee
041d37e0c8 Make most outbound operations are cancellable
- Inspired by #2214
- It actually reduces the chance of trying to marking a cancelled promise as success again, which raises an IllegalStateException.
2014-02-10 14:06:54 -08:00
Norman Maurer
87602fde25 [#2215] DefaultChannelHandlerContext tasks needs to be volatile to ensure every thread only see full initialized instances 2014-02-08 10:33:44 +01:00
Norman Maurer
9bee78f91c Provide an optimized AtomicIntegerFieldUpdater, AtomicLongFieldUpdater and AtomicReferenceFieldUpdater 2014-02-06 20:08:45 +01:00
Norman Maurer
64c3f58279 Not wakeup the EventLoop for writes as they will not cause a flush anyway 2014-02-01 13:45:27 +01:00
Norman Maurer
f94b563bcd Better exception message to tell the user why it is not supported 2014-01-30 07:02:22 +01:00
Norman Maurer
80ed147652 [#2164] Only reregister SelectionKeys that are still valid 2014-01-29 07:18:29 +01:00
Trustin Lee
9b0e5b9148 Fix API documentation on the usage of AttributeKey 2014-01-28 13:59:11 +09:00
Norman Maurer
f0c7c901d0 Add testcase to try to reproduce #2144 2014-01-23 07:04:42 +01:00
William Kemper
39b8cc348a fix grouping for isActive - socket.isBound is almost always true and should not override 'isOpen' 2014-01-17 07:14:00 +01:00
Trustin Lee
237ba27499 Better exception message 2014-01-13 23:32:49 +09:00
Trustin Lee
3373f83cbb Fix a bug where DefaultChannelPipelineTest.testFireChannelRegistered() triggers channelRegistered() twice 2014-01-13 23:25:32 +09:00
Norman Maurer
13d65d7ccf [#2104] Make sure we only act on the SelectionKey if it is valid 2014-01-09 18:27:56 +01:00
Trustin Lee
751943ed00 Fix a potential NPE due to the race between a connection attempt and its cancellation
- should fix #2086
2014-01-09 19:24:44 +09:00
Norman Maurer
0bbc3facec [#2086] Fix race which could produce NPE in AbstractNioUnsafe.finishConnect 2014-01-09 08:22:34 +01:00
milenkovicm
393f7b2306 ChannelOutboundBuffer returns total pending write size
total pending write size may be used to optimize write batching
2014-01-07 06:52:07 +01:00
Veebs
dd8f4bc0c4 Fix typo 2014-01-03 11:15:14 +01:00
Norman Maurer
fc2e6916d9 Fix regression which was introduced by 89a7cb8e71
Related to [#2060]
2013-12-21 21:39:51 +01:00
Trustin Lee
bab227213e Increase the default maxMessagesPerRead of AbstractNioByteChannel to 16
- Related: #2079
2013-12-21 20:08:42 +09:00
Trustin Lee
1fcd19a28f Fix a bug where adaptive recvbuf size prediction doesn't work correctly when maxMessagesPerRead is > 1 2013-12-21 19:57:11 +09:00
Norman Maurer
b889d3f559 [#2079] Stop reading once the NIO byte Channel was complete drained 2013-12-21 10:38:07 +01:00
Trustin Lee
89a7cb8e71 More graceful registration failure
- Fixes #2060
- Ensure to return a future/promise implementation that does not fail with 'not registered to an event loop' error for registration operations
- If there is no usable event loop available, GlobalEventExecutor.INSTANCE is used as a fallback.
2013-12-21 18:08:58 +09:00
Trustin Lee
ceaebd37ed Rename flushAndWrite() to writeAndFlush()
- Fixes: #2066
- Fixed inspection warnings
2013-12-16 15:02:13 +09:00
Trustin Lee
82b6e75b96 Rename flushAndWrite() to writeAndFlush()
- Related: #2066
2013-12-16 14:58:40 +09:00
Norman Maurer
e136227ee6 [#2065] Fix NPE in AbstractOioByteChannel during write to the socket 2013-12-15 11:41:04 +01:00
Trustin Lee
419b5a8de7 Disable logging temporarily when running testRegistrationAfterShutdown 2013-12-08 14:18:39 +09:00
Trustin Lee
0097b904b7 Disable logging temporarily when running testRegistrationAfterShutdown2 2013-12-08 14:13:43 +09:00
Trustin Lee
4116de8360 Fix NoSuchElementException raised by ChannelInitializer
.. again.
2013-12-07 11:04:40 +09:00
Trustin Lee
02703e8fc1 Revert "Fix NoSuchElementException raised by ChannelInitializer"
This reverts commit 3c453f5dba.
2013-12-07 11:04:33 +09:00
Norman Maurer
b3d8c81557 Fix all leaks reported during tests
- One notable leak is from WebSocketFrameAggregator
- All other leaks are from tests
2013-12-07 00:44:56 +09:00
Trustin Lee
51428004b3 Fix NoSuchElementException raised by ChannelInitializer 2013-12-07 00:39:21 +09:00
Norman Maurer
7ec70d0210 Merge package private interfaces into public ones. Related to [#1989] and [#1991] 2013-11-25 10:37:34 +01:00
Trustin Lee
fd776274c9 Additional fix for potential race condition which occurs when a user cancels a connection attempt
- Fixes #1986
2013-11-18 17:00:43 +09:00
Trustin Lee
54d3c99469 Fix an unexpected IllegalStateException from a selector loop when a user cancels a connection attempt
- Fixes #1986
2013-11-18 16:33:21 +09:00
Trustin Lee
ec10409314 Saner toString() implementation for EmbeddedSocketAddress 2013-11-08 18:04:40 +09:00
Norman Maurer
7dddbbb2bd Fix possible leak when a write is rejected 2013-11-07 10:24:15 +01:00
Norman Maurer
77054d7ecb Fix regression which lead to leak buffers when nothing could be read from the Channel.
This was introduced as part of #1812, but fortunatualy was not part of any release.
2013-11-07 07:12:19 +01:00
Trustin Lee
d2032254b7 Enable AUTO_CLOSE opton by default for backward compatibility / Deprecare AUTO_CLOSE option because it's gone in 5.0.
- Related #1952
2013-11-05 17:40:23 +09:00
Trustin Lee
c7b66545b4 Add AUTO_CLOSE option
- Fixes #1952
- If AUTO_CLOSE is turned on, Netty will close the channel immediately and automatically on write failure.  The default is false.
2013-11-05 17:24:36 +09:00
Trustin Lee
54db9ec725 Use StringUtil.simpleClassName(..) instead of Class.getSimpleName() where necessary
- Class.getSimpleName() doesn't render anonymous classes very well
- + some minor cleanup
2013-11-04 19:46:15 +09:00
Norman Maurer
e9967e55ad [#1940] Add javadoc to explain how the FileChannel is closed when using DefaultFileRegion 2013-11-02 15:59:33 +01:00
Trustin Lee
e1baa3f9b4 Simplify 2013-11-02 20:03:41 +09:00
Trustin Lee
72c78ebe75 Add an important notice about future migration issue in SimpleChannelInboundHandler.
- Related: #1590
2013-11-02 20:03:15 +09:00
Norman Maurer
b53fa5dd70 [#1812] All to have NioMessageUnsafe.read() inlined 2013-10-26 17:44:25 +02:00
Trustin Lee
8986245b47 Deprecate UniqueName and its subtypes' constructors / Add valueOf() for easier future migration. 2013-10-25 20:53:47 +09:00
Norman Maurer
544d68b396 [#1812] Allow for inline for most common cases when use NioByteUnsafe.read() 2013-10-25 13:32:21 +02:00
Norman Maurer
e4358ae6b8 [#1947] Handle RejectExecutionException graceful for outbound operations 2013-10-25 07:51:17 +02:00
Norman Maurer
9fd8bf329d More efficient handling of incomplete writes.
The problem with the old way was that we always set the OP_WRITE when the buffer could not be written
until the write-spin-count was reached. This means that in some cases the channel was still be writable
but we just was not able to write out the data quick enough. For this cases we should better break out the
write loop and schedule a write to be picked up later in the EventLoop, when other tasks was executed.
The OP_WRITE will only be set if a write actual returned 0 which means there is no more room for writing data
and this we need to wait for the os to notify us.
2013-10-17 20:23:47 +09:00
Norman Maurer
53fcff2eab [#1920] Fix IndexOutOfBoundsException when using PooledByteBufAllocator with SCTP and NIO Datagram channels 2013-10-16 09:53:26 +02:00
Norman Maurer
df442b9b6a #1924] Only log RejectExecutorException when call invokeLater(...) 2013-10-15 15:22:46 +02:00
Norman Maurer
068c75a025 [#1924] Correctly fail promise when EventExecutor was shutdown in between deregister 2013-10-15 14:01:59 +02:00
Trustin Lee
c9b7f1f1b5 Ensure the selector implementation can be instrumented before attempting instrumentation
- Fixes #1908
2013-10-11 12:08:51 +09:00
Trustin Lee
d7f9b1ee76 Fixes the problem where the promise of the outbound operation that causes a channel closure is notified after channelInactive()
- Fixes #1897
2013-10-08 12:24:16 +09:00
Norman Maurer
20a16ae8dc Use direct ByteBuf for the test to make sure it is not copied 2013-10-07 08:08:10 +02:00
Norman Maurer
c0936fc8e7 [#1890] Correctly expand ByteBuffer array in all cases
The problem was that we did not handle the case correctly when doubling the array was not enough. We need to keep doubling until everything fits in.
2013-10-06 15:29:35 +02:00
Bill Gallagher
013ac44d3a [#1832] - Channel writability change notifications sometimes fail to fire 2013-09-30 20:01:39 +02:00
Norman Maurer
b4fa8af079 Cache underlying ByteBuffers and count in ChannelOutboundBuffer.Entry to reduce object creation and so GC pressure
Beside this it also helps to reduce CPU usage as nioBufferCount() is quite expensive when used on CompositeByteBuf which are
nested and contains a lot of components
2013-09-26 20:37:39 +02:00
Norman Maurer
ae904a1f56 Introduce a new ChannelOption called DATAGRAM_CHANNEL_ACTIVE_ON_REGISTRATION. Related to [#1830]
This ChannelOption allows to tell the DatagramChannel implementation to be active as soon as they are registrated to their EventLoop. This can be used to make it possible to write to a not bound DatagramChannel.
The ChannelOption is marked as @deprecated as I'm looking for a better solution in master which breaks default behaviour with 4.0 branch.
2013-09-24 11:46:49 +02:00
Sasha Zverev
a2624a833e Slip in DefaultOioSocketChannelConfig (setAllowHalfClosure used to ignore argument) 2013-09-21 20:13:52 +02:00
Norman Maurer
f034f90fb3 [#1836] Add comment to explain why read is triggered 2013-09-17 06:58:47 +02:00
Norman Maurer
357677d8fa [#1830] Add testcase for write to not bound DatagramChannel impls and revert change in OIO as it breaks things as the udnerlying socket lazy binds 2013-09-12 09:27:43 +02:00
Norman Maurer
f4f04a08c2 Make DatagramChannel impls .isActive() return true when the underlying Channel is open. This also fixes [#1830] 2013-09-11 20:37:21 +02:00
Norman Maurer
7d9388a3cc Allow to only register a Channel via AbstractBootstrap and bind/connect it later. Related to [#1829] 2013-09-11 20:14:53 +02:00
Norman Maurer
af499b5fb4 Mark ChannelHandler.exceptionCaught(...) as deprecated in preparation to move to ChannelInboundHandler. Related to [#1808] 2013-09-08 20:11:51 +02:00
Trustin Lee
ad0c456209 Ensure operationProgressed is invoked even on completion
- Fixes #1809
2013-09-05 18:37:14 +09:00
Norman Maurer
92cec8d2ac Split up the nioBuffers() method to allow for inline. Related to #1812
This move less common method patterns to extra methods and so make the nioBuffers() method with most common pattern (backed by one ByteBuffer) small enough for inlining.
2013-09-05 09:22:13 +02:00
Norman Maurer
c97971b53a [#1805] Fix example in javadocs of SimpleChannelInboundHandler 2013-09-03 20:30:56 +02:00
Norman Maurer
25c226a835 Make sure only direct ByteBuffer are passed to the underlying jdk Channel.
This is needed because of otherwise the JDK itself will do an extra ByteBuffer copy with it's own pool implementation. Even worth it will be done
multiple times if the ByteBuffer is always only partial written. With this change the copy is done inside of netty using it's own allocator and
only be done one time in all cases.
2013-09-02 20:17:53 +02:00
Norman Maurer
4a5c840271 @Deprecated ChannelOutboundHandler.deregister(...) 2013-09-01 12:54:29 +02:00
Norman Maurer
b8866aef33 No need to check instanceof everytime to find the next inbound / outbound context 2013-08-30 06:31:58 +02:00
Norman Maurer
32deb2c22b [#1785] Fix incorrect javadocs 2013-08-27 06:55:14 +02:00
Norman Maurer
5e9b199296 @deprecated all methods which are related to deregister as it will be removed in 4.1.0.Final 2013-08-26 11:36:56 +02:00
Norman Maurer
80d30c3dd8 Small code improvements 2013-08-26 07:52:47 +02:00
Norman Maurer
77d4222db5 [#1777] Use correct Thread when close per channel 2013-08-25 21:22:53 +02:00
Norman Maurer
f76c01c3aa [#1782] Fix IndexOutOfBoundException with direct buffer and gathering writes 2013-08-24 18:16:05 +02:00
Trustin Lee
7aefd0cbdb Make AbstractBootstrap public
.. to work around the issue with JDK reflection described here:

  http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4283544

Fixes #1780
2013-08-24 17:20:53 +09:00
Norman Maurer
9ca20b73d3 Add testcase to show channelRegistered is called 2013-08-23 17:17:28 +02:00
bgallagher
51fa795819 fix race condition in test 2013-08-23 16:20:02 +02:00
Norman Maurer
206dc2a391 [#1772] Make sure ChannelOutboundBuffer.recycle() does also reset unflushed, flushed and tail.
This fix a IndexOutOfBoundsException which as triggered if recycle() did cut down the buffer[] to the initial size.
2013-08-23 09:01:16 +02:00
Norman Maurer
9fd35d09a2 [#1770] Fix NPE which was thrown if connection timeout was disabled and the connect did not finish directly when using NIO 2013-08-22 12:38:56 +02:00
bgallagher
6a2f340ec0 trim buffers before recycling 2013-08-21 20:28:54 +02:00
bgallagher
fb619f2394 fix writability callback 2013-08-21 16:39:50 +02:00
Norman Maurer
217b8e255c [#1763] Fill ChannelOutboundBuffer.nioBuffers with null on close to allow the content to be GC'ed 2013-08-20 21:15:05 +02:00
Trustin Lee
15cfa47ad9 Fix checkstyle 2013-08-20 14:40:28 +09:00
Trustin Lee
caf91b9c06 Fix IllegalStateException triggered while shutting down ThreadPerChannelEventLoopGroup
- Fix #1718
- Add the test case contributed by @mkw
2013-08-20 14:37:57 +09:00
bgallagher
a383988cdb add struct to replace parallel arrays consolidate flushed & unflushed buffers 2013-08-18 19:19:57 +02:00
bgallagher
06e250e493 remove unused initialCapacity 2013-08-17 14:10:33 +02:00
bgallagher
9f88552f12 remove some dead code 2013-08-10 20:46:48 +02:00
Norman Maurer
280afda013 [#1715] Remove extra call socket.configureBlocking(false) 2013-08-10 20:33:19 +02:00
bgallagher
59c2fd8813 [#1723] invoke close in eventloop 2013-08-10 08:19:16 +02:00
Norman Maurer
9d8c4ded50 Move method in correct scope 2013-08-09 13:30:04 +02:00
Norman Maurer
4d0621855a Remove unused package private method and so remove some complexity 2013-08-09 11:44:41 +02:00
Norman Maurer
b54937ab50 [#1711] Allow to use ChannelOption for set / get MessageSizeEstimator 2013-08-08 20:37:35 +02:00
Norman Maurer
f40a3f34b1 [#1692] Allow to set WRITE_BUFFER_HIGH_WATER_MARK and WRITE_BUFFER_LOW_WATER_MARK via ChannelOption 2013-08-08 20:30:01 +02:00
Norman Maurer
fc6213604d Make sure ChannelOutboundBuffer not throw NPE during increment/decrement pending writes 2013-08-08 13:42:25 +02:00
Norman Maurer
b934b6009c Share some code, related to [#1707]] 2013-08-07 20:28:33 +02:00
Norman Maurer
5ef30b6d8b [#1707] More efficient implementation of ChannelHandlerContext.writeAndFlush(..) 2013-08-07 20:15:10 +02:00
Norman Maurer
5b6e762bf4 Allow to have the Channel GC'ed ASAP 2013-08-07 10:32:03 +02:00
Norman Maurer
48a7a21541 Correctly update Channel.isWritable() when the write happens from outside the EventLoop in a fast fashion. Fixes [#1697]
Introduce a new interface called MessageSizeEstimator. This can be specific per Channel (via ChannelConfig). The MessageSizeEstimator will be used to estimate for a message that should be written. The default implementation handles ByteBuf, ByteBufHolder and FileRegion. A user is free to plug-in his/her own implementation for different behaviour.
2013-08-07 10:10:43 +02:00
Norman Maurer
de44917e62 Use recyclable write task to reduce GC pressure 2013-08-05 14:59:58 +02:00
Trustin Lee
17b91f416d Remove an irrelevant comment
.. because firedChannelActive is gone long ago
2013-08-05 16:50:25 +09:00
Norman Maurer
a06295fe0a Correctly fix problem in ByteToMessageDecoder and ReplayingDecoder which could let to have a released buffer passed to the decode methods.
This fixes #1664 and revert also the original commit which was meant to fix it 3b1881b523 . The problem with the original commit was that it could delay handlerRemove(..) calls and so mess up the order or forward bytes to late.
2013-08-01 09:54:07 +02:00
Trustin Lee
81de2275e4 Remove unnecessary traversal of pipeline which occurs when a Channel is closed
- Remove unnecessary ascending traversal of pipeline in DefaultChannelHandlerContext.freeInbound()
- Move DefaultChannelHandlerContext.teardownAll() to DefaultChannelPipeline
2013-07-31 20:45:37 +09:00
Trustin Lee
3b1881b523 Make sure ChannelHandler.handlerRemoved() is invoked while handler methods are not running
- Fixes #1664
2013-07-31 20:35:51 +09:00
Norman Maurer
3049f5ddc4 Fix checkstyle 2013-07-31 12:21:50 +02:00
Trustin Lee
c79a3cdefe Fix a bug in NioSocketChannel.doWrite() where flush() triggered from a ChannelFutureListener is ignored
- Fixes #1679
2013-07-31 19:13:45 +09:00
Norman Maurer
ca29f1a37d [#1683] Mark unused ChannelOption @deprecated, to prepare for removal in 4.1.0.Final 2013-07-31 10:50:39 +02:00
Trustin Lee
e81368af06 Simplify handler invocations in DefaultChannelHandlerContext
- Sometimes we delegated executor.inEventLoop() to the next context and sometimes we did not.  Now we always check inEventLoop() before delegating.
2013-07-31 16:34:32 +09:00
Norman Maurer
e09aea5902 Rename methods of EmbeddedChannel and mark old ones as @deprecated 2013-07-31 16:04:08 +09:00
Jeff Pinner
3922c518cd Clean up AbstractUnsafe#bind and AbstractUnsafe#disconnect. 2013-07-29 06:54:42 +02:00
Jeff Pinner
bf2430d255 Change AbstractChannel#doRegister return type from Runnable to void. 2013-07-27 20:22:02 +02:00
Jeff Pinner
ca59c1201e Change AbstractChannel#doDeregister return type from Runnable to void. 2013-07-26 22:21:41 +02:00
Norman Maurer
15279b2525 [#1654] Notify close future after the close was complete 2013-07-26 14:21:11 +02:00
Norman Maurer
cffbba94ee [#1654] Remove AbstractChannel.doPreClose() and notify close future on register failure. Based on patch of @jpinner 2013-07-26 14:10:31 +02:00
Norman Maurer
5f363e8ade Revert "remove AbstractChannel#doPreClose"
This reverts commit 241f856cc7.
2013-07-26 11:23:11 +02:00
Jeff Pinner
241f856cc7 remove AbstractChannel#doPreClose 2013-07-26 10:12:01 +02:00
Norman Maurer
fabefba791 [#1650] Make it possible to override isSharable() to simplify things for clojure users 2013-07-25 20:24:27 +02:00
Norman Maurer
6ff87cc20d Small optimitation to limit volatile reads 2013-07-25 09:28:54 +02:00
Trustin Lee
5e4169802f Fix potential reentrance issue in ChannelOutboundBuffer 2013-07-24 11:26:03 +09:00
Norman Maurer
0b17586731 [#1623] Fix javadocs for ChannelGroup 2013-07-23 09:03:42 +02:00
Trustin Lee
7903697c36 Remove Arrays.fill(..., null)
.. because we can just set each element to null while looping
2013-07-23 15:09:14 +09:00
Trustin Lee
a89b17fa94 Merge ChannelOutboundBuffer.failUnflushed() and recycle() into a single method and make sure it is run later on reentrance
- Previously, failUnflushed() did not run when inFail is true, which made unflushed writes are not released on reentrance.   This has been fixed by this commit.
- Also, AbstractUnsafe.outboundBuffer is set to null as early as possible to remove the chance of any write attempts made after the closure.
2013-07-23 14:34:10 +09:00
Trustin Lee
d4aa5b53d6 Simplify the validation logic proposed in 25f96b1644 2013-07-23 13:48:22 +09:00
kerr
ada07cb9e0 Fix types in javadocs 2013-07-22 19:14:36 +02:00
Norman Maurer
25f96b1644 [#1606] Reduce overhead during writes because of ChannelPromise validation 2013-07-22 11:28:02 +02:00
Norman Maurer
81612f8e9b [#1624] Fix resource leak when writing to a closed / not-open channel 2013-07-22 10:59:06 +02:00
Norman Maurer
0f6cc0cc7b [#1630] Fix re-entrance bug in ChannelOutboundBuffer.remove(...) 2013-07-22 10:44:33 +02:00
tsingxu
125070958d Fix up ChannelPipeline javadocs by add lost "don't " 2013-07-22 07:14:42 +02:00
Norman Maurer
7f86550ef8 [#1619] Fix bug in ChannelOutboundBuffer which could lead to invalid write order when fireChannelWritabilityChanged() write another message 2013-07-20 10:03:53 +02:00
Norman Maurer
60529ac3ee Fix checkstyle 2013-07-20 08:09:08 +02:00
Jeff Pinner
6c9c151d66 minor documentation cleanup 2013-07-20 08:01:14 +02:00
Norman Maurer
4d69848aec Fix broken javadocs 2013-07-19 07:40:12 +02:00
Trustin Lee
19c92ceb59 Separate 'progress total' and 'pending number of bytes'
- Reverted the recent changes in AbstractChannel.calculateMessageSize()
2013-07-19 14:09:08 +09:00
Trustin Lee
762adfcb69 Update HttpStaticFileServer example / Fix bugs related with progress notification
- Fix a bug in DefaultProgressivePromise.tryProgress() where the notification is dropped
 - Fix a bug in AbstractChannel.calculateMessageSize() where FileRegion is not counted
 - HttpStaticFileServer example now uses zero copy file transfer if possible.
2013-07-19 13:21:32 +09:00
Trustin Lee
f96a8e5951 Implement ProgressivePromise notification in NIO byte channels and ChunkedWriteHandler
- Refine the contract of GenericProgressiveFutureListener.
- Negative 'total' now means 'unknown', which is useful for ChunkedWriteHandler.
2013-07-19 12:53:23 +09:00
Trustin Lee
0653efcd75 Even better promise validation messages 2013-07-19 09:38:25 +09:00
Trustin Lee
d5e202d755 Better promise validation messages 2013-07-19 09:37:56 +09:00
Trustin Lee
faf8b76b5a Fix a potential bug where AbstractUnsafe.flush0() fails to clear the inFlush0 flag
- Fixes #1609
2013-07-19 08:56:17 +09:00
Trustin Lee
8178322c5c A little bit of Javadoc for ChannelOutboundBuffer 2013-07-19 01:17:00 +09:00
Trustin Lee
fef838f197 Make sure channelInactive() and channelDeregistered() are triggered even if outboundBuffer.recycle() fails 2013-07-19 01:09:19 +09:00
Trustin Lee
8828d5327a Move inFlush0 and outboundBuffer in AbstractChannel to AbstractUnsafe
.. because they are accessed by AbstractUnsafe much more often.
2013-07-18 23:33:39 +09:00
Trustin Lee
5e5df6bd09 Fix NPE 2013-07-18 23:29:13 +09:00
Trustin Lee
4cd7e62555 Make ChannelOutboundBuffer recycled 2013-07-18 23:26:45 +09:00
Trustin Lee
46ea0d4e7b Implement gathering writes in NioSocketChannel
- Add some support methods in ChannelOutputBuffer
2013-07-18 23:14:39 +09:00
Trustin Lee
4f0a952241 Checkstyle 2013-07-18 21:01:01 +09:00
Trustin Lee
bcef796dc7 Rewrite ChannelOutboundBuffer
- Merge MessageList into ChannelOutboundBuffer
- Make ChannelOutboundBuffer a queue-like data structure so that it is nearly impossible to leak a message
- Make ChannelOutboundBuffer public so that AbstractChannel can expose it to its subclasses.
- TODO: Re-enable gathering write in NioSocketChannel
2013-07-18 20:59:14 +09:00
Trustin Lee
568166d554 Cleanup 2013-07-18 10:29:34 +09:00
Trustin Lee
473af5c98e Fix checkstyle 2013-07-18 10:25:16 +09:00
Trustin Lee
e28594952b Make the cleanup logic in ChannelOutboundBuffer more robust
- Fixes #1601
2013-07-18 10:23:26 +09:00
Trustin Lee
9c8d980a74 Fix a bug where a ChannelFutureListener can allow Unsafe.close() and Unsafe.flush() to overlap and reenter from flush() to close().
- Fixes #1600
2013-07-18 10:14:55 +09:00
Trustin Lee
88cbdb50d2 Fail unflushed writes with ClosedChannelException
... instead of cryptic exception message.
2013-07-17 21:25:51 +09:00
Trustin Lee
5f235eafc3 Fix checkstyle 2013-07-17 21:17:52 +09:00
Trustin Lee
31ecd17212 Fix test failure 2013-07-17 21:16:52 +09:00
Trustin Lee
cc9b9567b9 Make sure all write attempts made after a channel is closed are marked as failure 2013-07-17 21:14:07 +09:00
Trustin Lee
a8d67b0282 Fix data structure corruption and resource leak in ChannelOutboundBuffer 2013-07-17 21:02:56 +09:00
Trustin Lee
dcf7896172 Do not close ServerChannel on IOException while accepting an incoming connection
.. so that we can recover from 'too many open files'

- Fixes #1578
2013-07-16 18:39:29 +09:00
Norman Maurer
2de491aa7f Use UnpooledByteBufAllocator as default 2013-07-15 10:44:33 +02:00
Norman Maurer
d5f052f39c Rename ChannelGroupMatcher to ChannelMatcher 2013-07-15 10:42:36 +02:00
Norman Maurer
9dfad35423 Fix checkstyle 2013-07-14 18:32:10 +02:00
Norman Maurer
75997545ed Fix checkstyle 2013-07-14 18:13:47 +02:00
Norman Maurer
c20c852602 More javadoc fixes 2013-07-14 18:04:33 +02:00
Norman Maurer
43d22f6d23 Remove unused class which was package private anyway 2013-07-14 17:54:07 +02:00
Norman Maurer
a07abee55f Add and correct javadocs 2013-07-13 19:42:02 +02:00
Norman Maurer
2af7db361b [#1573] Fix NPE which could be triggered one failed promises that trigger a close89 2013-07-12 22:57:03 +02:00
Norman Maurer
2871079c4a [#1557] Make the contract of Channel.Unsafe.flush() more clear
* Remove boolean parameter from Channel.Unsafe.flush() method
* Move NIO related things to AbstractNioChannel.NioUnsafe
2013-07-12 19:45:31 +02:00
Trustin Lee
660729bd56 Kill the cruft 2013-07-12 20:41:29 +09:00