Commit Graph

5857 Commits

Author SHA1 Message Date
Norman Maurer
ac348956e3 [#2615] Correctly update SelectionKey after selector rebuild
Motivation:

When a select rebuild was triggered the reference to the SelectionKey is not updated in AbstractNioChannel. This will cause a CancelledKeyException later.

Modification:

Correctly update SelectionKey reference after rebuild

Result:

Fix exception
2014-06-27 17:09:27 +02:00
Trustin Lee
ff5b15acbd Upgrade JACOCO to the latest version
.. to fix the build errors triggered by the old JACOCO release
2014-06-27 17:23:03 +09:00
Trustin Lee
4f82b6e393 Enable verbose:gc only when 'leak' profile is disabled
The leak profile generates a lot of garbage which makes the verbose:gc
output too large.
2014-06-27 17:10:45 +09:00
Trustin Lee
7b2d2e22ba Remove the deprecated methods in SPDY
SPDY is an experimental package. No need to keep ABI compatibility.
2014-06-27 16:46:45 +09:00
Trustin Lee
2bdcedeffd Partially revert 4a13f66e13
Revert the removal of 'get' prefix from HTTP classes to ensure ABI
compatibility.  Note that this commit does not revert the changes in
SPDY, which is considered experimental.
2014-06-27 16:40:43 +09:00
Norman Maurer
522b3e1b92 [#2604] Not try to use sun.misc.Cleaner when on android
Motivation:

When a user tries to use netty on android it currently fails with "Could not find class 'sun.misc.Cleaner'"

Modification:

Encapsulate sun.misc.Cleaner usage in extra class to workaround this isssue.

Result:
Netty can be used on android again
2014-06-27 08:18:21 +02:00
Zhihui Jiao
d062500ca0 Fix inconsistent code in the doc 2014-06-27 06:48:47 +02:00
Norman Maurer
8869f6d5c1 [#2598] Add Epoll.isAvailable() which allows to check if epoll can be used.
Motivation:

At the moment there is no simple way for a user to check if the native epoll transport can be used on the running platform. Thus the user can only try to instance it and catch any exception and fallback to nio transport.

Modification:

Add Epoll.isAvailable() which allows to check if epoll can be used.

Result:

User can easily check if epoll transport can be used or not
2014-06-26 12:27:40 +02:00
Trustin Lee
a5097b937e Fix incorrect bytesBefore/indexOf() in ReplayingDecoderBuffer
Motivation:

bytesBefore(length, ...), bytesBefore(index, length, ...), and
indexOf(fromIndex, toIndex,...) in ReplayingDecoderBuffer are buggy.
They trigger 'REPLAY even when they don't need to.

Modification:

Implement the buggy methods properly so that REPLAYs are not triggered
unnecessarily.

Result:

Correct behvaior
2014-06-26 18:57:20 +09:00
Norman Maurer
d1c8bcb40f [#2605] Use SO_REUSEADDR on EpollServerSocketChannel to match defaults of java.nio.ServerSocketChannel impl
Motivation:

When using openjdk and oracle jdk's nio (while using the nio transport) the ServerSocketChannel uses SO_REUSEADDR by default. Our native transport should do the same to make it easier to switch between the different implementations and get the expected result.

Modification:

Change EpollServerSocketChannelConfig to set SO_REUSEADDR on the created socket.

Result:

SO_REUSEADDR is used by default on servers.
2014-06-26 11:53:23 +02:00
Norman Maurer
658fdffbad Reduce the memory copies in JdkZlibEncoder
Motivation:

At the moment we use a lot of unnecessary memory copies in JdkZlibEncoder. This is caused by either allocate a to small ByteBuf and expand it later or using a temporary byte array.
Beside this the memory footprint of JdkZlibEncoder is pretty high because of the byte[] used for compressing.

Modification:

- Override allocateBuffer(...) and calculate the estimatedsize in there, this reduce expanding of the ByteBuf later
- Not use byte[] in the instance itself but allocate a heap ByteBuf and write directly into the byte array

Result:

Less memory copies and smaller memory footprint
2014-06-26 11:12:06 +02:00
Trustin Lee
d8d0bbfc26 Optimize PoolChunk
- Using short[] for memoryMap did not improve performance. Reverting
  back to the original dual-byte[] structure in favor of simplicity.
- Optimize allocateRun() which yields small performence improvement
- Use local variable when member fields are accessed more than once
2014-06-26 17:06:29 +09:00
Trustin Lee
c41538050c Fix inspector warnings 2014-06-26 17:06:29 +09:00
Pavan Kumar
d2e36a49c7 Improve the allocation algorithm in PoolChunk
Motivation:

Depth-first search is not always efficient for buddy allocation.

Modification:

Employ a new faster search algorithm with different memoryMap layout.

Result:

With thread-local cache disabled, we see a lot of performance
improvment, especially when the size of the allocation is as small as
the page size, which had the largest search space previously.
2014-06-26 17:06:29 +09:00
Norman Maurer
c3f24444ef Use IntObjectMap to replace Map in EpollEventLoop.
Motivation:

We need to map from ints to AbstractEpollChannel in EpollEventLoop but there is no need for box to Integer.

Modification:

Replace Map with IntObjectMap.

Result:

No more auto-boxing needed.
2014-06-25 20:23:16 +02:00
Norman Maurer
b4b61c1f41 [#2599] Not use sun.nio.ch.DirectBuffer as it not exists on android
Motivation:

During some refactoring we changed PlatformDependend0 to use sun.nio.ch.DirectBuffer for release direct buffers. This broke support for android as the class does not exist there and so an exception is thrown.

Modification:

Use again the fieldoffset to get access to Cleaner for release direct buffers.

Result:
Netty can be used on android again
2014-06-25 15:01:57 +02:00
Alexey Parfenov
063ca10d87 Fix integer overflow in HttpObjectEncoder when handling chunked encoding and FileRegion > Integer.MAX_VALUE
Motivation:

Due to integer overflow bug, writes of FileRegions to http server pipeline (eg like one from HttpStaticFileServer example) with length greater than Integer.MAX_VALUE are ignored in 1/2 of cases (ie no data gets sent to client)

Modification:

Correctly handle chunk sized > Integer.MAX_VALUE

Result:

Be able to use FileRegion > Integer.MAX_VALUE when using chunked encoding.
2014-06-24 12:14:46 +02:00
Trustin Lee
4a13f66e13 Remove 'get' prefix from all HTTP/SPDY messages
Motivation:

Persuit for the consistency in method naming

Modifications:

- Remove the 'get' prefix from all HTTP/SPDY message classes
- Fix some inspector warnings

Result:

Consistency
Fixes #2594
2014-06-24 18:33:30 +09:00
onlychoice
30e22f5da3 Fix a typo in comment 2014-06-24 11:02:12 +02:00
Trustin Lee
7457801020 Deprecate SocksMessage.encodeAsByteBuf()
It was an internal use only method which became public by a mistake
during the review process.
2014-06-24 16:41:47 +09:00
Trustin Lee
71dce0193f Rename fromByte() to valueOf()
Motivation:

Persuit the consistency in method naming

Modifications:

Rename fromByte(byte) to valueOf(byte)

Result:

Consistency
2014-06-24 16:36:10 +09:00
Norman Maurer
790c63e8d2 Improve performance of Recycler
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
2014-06-24 08:09:19 +02:00
Norman Maurer
22f16e52bf MessageToByteEncoder always starts with ByteBuf that use initalCapacity == 0
Motivation:

MessageToByteEncoder always starts with ByteBuf that use initalCapacity == 0 when preferDirect is used. This is really wasteful in terms of performance as every first write into the buffer will cause an expand of the buffer itself.

Modifications:

 - Change ByteBufAllocator.ioBuffer() use the same default initialCapacity as heapBuffer() and directBuffer()
 - Add new allocateBuffer method to MessageToByteEncoder that allow the user to do some smarter allocation based on the message that will be encoded.

Result:

Less expanding of buffer and more flexibilty when allocate the buffer for encoding.
2014-06-24 13:55:02 +09:00
Norman Maurer
f0e2d0b77c Make use of HttpChunkedInput as this will also work when compression is used 2014-06-23 09:38:31 +02:00
Trustin Lee
7162d96ed5 Revert "Improve the allocation algorithm in PoolChunk"
This reverts commit 36305d7dce, which
seems to cause an assertion failure on our CI machine.
2014-06-21 19:19:49 +09:00
Trustin Lee
a1b87411fb Make sure OpenSslEngine is tested against transport-native-epoll 2014-06-21 18:29:00 +09:00
Trustin Lee
3775a6124e Remove padding utility classes
- It's not used anywhere
2014-06-21 17:59:20 +09:00
Trustin Lee
6834c2da51 Add missing last padding / Comment 2014-06-21 17:57:29 +09:00
Trustin Lee
b1a5ced729 Checkstyle / Overall clean-up / Fix serialization 2014-06-21 17:57:29 +09:00
nitsanw
a5d585a8ef Fix false sharing between head and tail reference in MpscLinkedQueue
Motivation:

The tail node reference writes (by producer threads) are very likely to
invalidate the cache line holding the headRef which is read by the
consumer threads in order to access the padded reference to the head
node. This is because the resulting layout for the object is:

- header
- Object AtomicReference.value -> Tail node
- Object MpscLinkedQueue.headRef -> PaddedRef -> Head node

This is 'passive' false sharing where one thread reads and the other
writes.  The current implementation suffers from further passive false
sharing potential from any and all neighbours to the queue object as no
pre/post padding is provided for the class fields.

Modifications:

Fix the memory layout by adding pre-post padding for the head node and
putting the tail node reference in the same object.

Result:

Fixed false sharing
2014-06-21 17:57:29 +09:00
nmittler
fd895e53f4 Adding int-to-object map implementation.
Motivation:

Maps with integer keys are used in several places (HTTP/2 code, for
example). To reduce the memory footprint of these structures, we need a
specialized map class that uses ints as keys.

Modifications:

Added IntObjectHashMap, which is uses open addressing and double hashing
for collision resolution.

Result:

A new int-based map class that can be shared across Netty.
2014-06-21 08:36:06 +02:00
Trustin Lee
cb994dd926 Fix the inconsistencies between performance tests in ByteBufAllocatorBenchmark
Motivation:

default*() tests are performing a test in a different way, and they must be same with other tests.

Modification:

Make sure default*() tests are same with the others

Result:

Easier to compare default and non-default allocators
2014-06-21 13:28:11 +09:00
Pavan Kumar
004ffbad90 Improve the allocation algorithm in PoolChunk
Motivation:

Depth-first search is not always efficient for buddy allocation.

Modification:

Employ a new faster search algorithm with different memoryMap layout.

Result:

With thread-local cache disabled, we see a lot of performance
improvment, especially when the size of the allocation is as small as
the page size, which had the largest search space previously:

-- master head --
Benchmark                (size) Mode    Score  Error Units
pooledDirectAllocAndFree  8192 thrpt  215.392  1.565 ops/ms
pooledDirectAllocAndFree 16384 thrpt  594.625  2.154 ops/ms
pooledDirectAllocAndFree 65536 thrpt 1221.520 18.965 ops/ms
pooledHeapAllocAndFree    8192 thrpt  217.175  1.653 ops/ms
pooledHeapAllocAndFree   16384 thrpt  587.250 14.827 ops/ms
pooledHeapAllocAndFree   65536 thrpt 1217.023 44.963 ops/ms

-- changes --
Benchmark                (size) Mode    Score  Error Units
pooledDirectAllocAndFree  8192 thrpt 3656.744 94.093 ops/ms
pooledDirectAllocAndFree 16384 thrpt 4087.152 22.921 ops/ms
pooledDirectAllocAndFree 65536 thrpt 4058.814 29.276 ops/ms
pooledHeapAllocAndFree    8192 thrpt 3640.355 44.418 ops/ms
pooledHeapAllocAndFree   16384 thrpt 4030.206 24.365 ops/ms
pooledHeapAllocAndFree   65536 thrpt 4103.991 70.991 ops/ms
2014-06-21 13:20:56 +09:00
Norman Maurer
58f4b4b7d9 [#2589] LocalServerChannel.doClose() throws NPE when localAddress == null
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().
2014-06-20 20:07:00 +02:00
Norman Maurer
19a1b603d0 Remove System.out.println(...) debug messages 2014-06-20 19:42:08 +02:00
Norman Maurer
d2b8560a76 [#2580] [#2587] Fix buffer corruption regression when ByteBuf.order(LITTLE_ENDIAN) is used
Motivation:

To improve the speed of ByteBuf with order LITTLE_ENDIAN and where the native order is also LITTLE_ENDIAN (intel) we introduces a new special SwappedByteBuf before in commit 4ad3984c8b. Unfortunally the commit has a flaw which does not handle correctly the case when a ByteBuf expands. This was caused because the memoryAddress was cached and never changed again even if the underlying buffer expanded. This can lead to corrupt data or even to SEGFAULT the JVM if you are lucky enough.

Modification:

Always lookup the actual memoryAddress of the wrapped ByteBuf.

Result:

No more data-corruption for ByteBuf with order LITTLE_ENDIAN and no JVM crashes.
2014-06-20 18:25:54 +02:00
Norman Maurer
1278467fec [#2586] Use correct EventLoop to notify delayed successful registration
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
2014-06-20 16:51:28 +02:00
Trustin Lee
fb538ea532 Refactor FastThreadLocal to simplify TLV management
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.
2014-06-19 21:08:16 +09:00
Norman Maurer
7279e48bef Small improvement in SimpleChannelInboundHandlerAdapter javadoc 2014-06-18 14:49:11 +02:00
Norman Maurer
917132e28d Make use of AtomicLongFieldUpdater.addAndGet(...) for cleaner code
Motivation:

The code in ChannelOutboundBuffer can be simplified by using AtomicLongFieldUpdater.addAndGet(...)

Modification:

Replace our manual looping with AtomicLongFieldUpdater.addAndGet(...)

Result:

Cleaner code
2014-06-17 19:50:14 +02:00
Norman Maurer
b627824b18 [#2577] ChannelOutboundBuffer.addFlush() unnecessary loop through all entries on multiple calls
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.
2014-06-17 09:29:16 +02:00
Trustin Lee
b8a7881588 Fix incorrect method signature of awaitInactivity()
- Related: #2084
2014-06-17 16:01:25 +09:00
Norman Maurer
6fdf1138ca [#2573] UnpooledUnsafeDirectByteBuf.setBytes(int,ByteBuf,int,int) fails to use fast-path when src has array
Motivation:

UnpooledUnsafeDirectByteBuf.setBytes(int,ByteBuf,int,int) fails to use fast-path when src uses an array as backing storage. This is because the if else uses the wrong ByteBuf for its check.

Modifications:

- Use correct ByteBuf when check for array as backing storage
- Also eliminate unecessary check in UnpooledDirectByteBuf which always fails anyway

Result:

Faster setBytes(...) when src ByteBuf is backed by an array.

No more IndexOutOfBoundsException or data-corruption.
2014-06-16 11:11:54 +02:00
Norman Maurer
a4bd566cef [#2572] Correctly calculate length of output buffer before inflate to fix IndexOutOfBoundException
Motivation:

JdkZlibDecoder fails to decode because the length of the output buffer is not calculated correctly.
This can cause an IndexOutOfBoundsException or data-corruption when the PooledByteBuffAllocator is used.

Modifications:

Correctly calculate the length

Result:

No more IndexOutOfBoundsException or data-corruption.
2014-06-16 10:22:02 +02:00
Phil.Baxter
101b9ded33 export sun security packages as optional 2014-06-15 20:59:55 +02:00
Trustin Lee
e88262861a Use FastThreadLocal in more places 2014-06-14 17:46:36 +09:00
Norman Maurer
b737d631f1 [maven-release-plugin] prepare for next development iteration 2014-06-12 16:20:52 +02:00
Norman Maurer
1709113a1f [maven-release-plugin] prepare release netty-4.0.20.Final 2014-06-12 16:14:48 +02:00
Norman Maurer
76043bc8c8 Make use of an array to store FastThreadLocals and so allow to also use it in PooledByteBufAllocator that is instanced by users.
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
2014-06-12 15:43:20 +02:00
belliottsmith
1ac2ff8d7b Introduce FastThreadLocal which uses an EnumMap and a predefined fixed set of possible thread locals
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
2014-06-12 15:43:20 +02:00