aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region/ClientStack (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
| | * Putting this to exactly what it was yesterday around this time, so we can do ↵Diva Canto2009-10-092-182/+411
| | | | | | | | | | | | the 100-ppl load test in WP. We need to carefully play with this code in order to understand all the problems.
* | | * Minimized the number of times textures are pulled off the priority queueJohn Hurliman2009-10-147-67/+141
|/ / | | | | | | | | | | * OnQueueEmpty is still called async, but will not be called for a given category if the previous callback for that category is still running. This is the most balanced behavior I could find, and seems to work well * Added support for the old [ClientStack.LindenUDP] settings (including setting the receive buffer size) and added the new token bucket and global throttle settings * Added the AssetLoaderEnabled config variable to optionally disable loading assets from XML every startup. This gives a dramatic improvement in startup times for those who don't need the functionality every startup
* | * Split Task category into Task and StateJohn Hurliman2009-10-132-30/+45
| | | | | | | | * Crude prioritization hack
* | * Copied LocklessQueue.cs into OpenSim.Framework and added the .Count ↵John Hurliman2009-10-135-80/+103
| | | | | | | | | | | | | | | | | | property and .Clear() method * Changed the way the QueueEmpty callback is fired. It will be fired asynchronously as soon as an empty queue is detected (this can happen immediately following a dequeue), and will not be fired again until at least one packet is dequeued from that queue. This will give callbacks advanced notice of an empty queue and prevent callbacks from stacking up while the queue is empty * Added LLUDPClient.IsConnected checks in several places to prevent unwanted network activity after a client disconnects * Prevent LLClientView.Close() from being called twice every disconnect * Removed the packet resend limit and improved the client timeout check
* | * Consolidated adding / removing ClientManager IClientAPIs to two places in ↵John Hurliman2009-10-132-69/+5
| | | | | | | | | | | | | | | | Scene * Added some missing implementations of IClientAPI.RemoteEndPoint * Added a ClientManager.Remove(UUID) overload * Removed a reference to a missing project from prebuild.xml
* | * Fixed a bug where clients were being added to ClientManager twiceJohn Hurliman2009-10-133-48/+41
| | | | | | | | | | | | * Changed the ClientManager interface to reduce potential errors with duplicate or mismatched keys * Added IClientAPI.RemoteEndPoint, which can (hopefully) eventually replace IClientAPI.CircuitCode * Changed the order of operations during client shutdown
* | * Rewrote ClientManager to remove Lindenisms from OpenSim core, improve ↵John Hurliman2009-10-134-271/+107
| | | | | | | | | | | | | | | | performance by removing locks, and replace LLUDPClientCollection * Removed the confusing (and LL-specific) shutdowncircuit parameter from IClientAPI.Close() * Updated the LLUDP code to only use ClientManager instead of trying to synchronize ClientManager and m_clients * Remove clients asynchronously since it is a very slow operation (including a 2000ms sleep)
* | * Unregister event handlers in LLUDPServer when a client logs out and ↵John Hurliman2009-10-132-239/+221
| | | | | | | | | | | | | | | | disconnects * Move ViewerEffect handling to Scene.PacketHandlers * Removing the unused CloseAllAgents function * Trimming ClientManager down. This class needs to be reworked to keep LLUDP circuit codes from intruding into the abstract OpenSim core code
* | Avoid checking m_clients collection twice when a UseCircuitCode packet is ↵John Hurliman2009-10-132-28/+30
| | | | | | | | received
* | * Broke the circular reference between LLClientView and LLUDPClient. This ↵John Hurliman2009-10-132-59/+77
| | | | | | | | | | | | should speed up garbage collection on the large LLClientView objects, and also prevents handling packets for disconnected clients * Renamed local LLUDPClient variables to udpClient to avoid naming confusion between LLUDPClient and LLClientView
* | * Eliminated unnecessary parameters from LLUDPServer.SendPacketData()John Hurliman2009-10-131-5/+8
| | | | | | | | * Changed PrimMesher's Quat.Identity to return <0,0,0,1> instead of <0,0,0,1.1>
* | * Changed the "Packet exceeded buffer size" log line to debug and include ↵John Hurliman2009-10-092-8/+5
| | | | | | | | | | | | the packet type. This message is normal, but could be evidence of a message marked for zerocoding that probably shouldn't be * Changing OpenSimUDPBase back to high concurrency. High concurrency mode seems to make other problems happen faster, so it's helpful for working out other bugs and will probably
* | Slow down the packet receiving code again after new reports of thread storms.Melanie2009-10-091-4/+8
| |
* | Forgot to initialize m_writeLockJohn Hurliman2009-10-091-1/+1
| |
* | Added more debugging output to the "unrecognized source" warningJohn Hurliman2009-10-091-1/+2
| |
* | * Added a lock object for the write functions in LLUDPClientCollection ↵John Hurliman2009-10-093-210/+169
| | | | | | | | | | | | | | (immutable != concurrent write safety) * Allow the UDP server to bind to a user-specified port again * Updated to a newer version of OpenSimUDPBase that streamlines the code even more. This also reintroduces the highly concurrent packet handling which needs more testing
* | Simplified LLUDPClientCollection from three collections down to one. This ↵John Hurliman2009-10-083-95/+29
| | | | | | | | will prevent any potential problems from inconsistency between the internal collections
* | Fear the lockless LLUDP implementation!John Hurliman2009-10-081-192/+60
|/
* Putting the lock back in TryGetValue.Diva Canto2009-10-082-349/+349
|
* Fork UDPBase from libOMV into opensimMelanie2009-10-082-1/+341
|
* Added guards against the exception reported:Diva Canto2009-10-081-0/+9
| | | | | | | | | #2009-10-08 12:58:34,462 ERROR - OpenSim.Application [APPLICATION]: #APPLICATION EXCEPTION DETECTED: System.UnhandledExceptionEventArgs # #Exception: System.Exception: Object reference not set to an instance of an object. ---> System.NullReferenceException: Object reference not set to an instance of an object. # at OpenSim.Region.ClientStack.LindenUDP.J2KImage.SendFirstPacket(LLClientView client) in c:\Opensim\OpenSim\Region\ClientStack\LindenUDP\J2KImage.cs:line 230 # at OpenSim.Region.ClientStack.LindenUDP.J2KImage.SendPackets(LLClientView client, Int32 maxpack) in c:\Opensim\OpenSim\Region\ClientStack\LindenUDP\J2KImage.cs:line 84
* One last attempt at tunning the locking/no locking behaviour. The previous ↵Diva Canto2009-10-071-10/+16
| | | | one was too lax and made the important login packets fail.
* Another go at the deadlock: not locking TryGetValue, and just catch/ignore ↵Diva Canto2009-10-071-15/+31
| | | | exceptions.
* Changed the locking mechanism, because the existing one is causing deadlocks ↵Diva Canto2009-10-071-25/+100
| | | | to occur.
* replace LLClientView with proper versionMelanie2009-10-071-1898/+1332
|
* Replace LLClientView on the branch with master. UNCOMPILABLE!Melanie2009-10-071-1332/+1898
|
* Revert "Merging in diva's locking fixes"Melanie2009-10-071-69/+47
| | | | This reverts commit 832cc685138b2244529f10b54b373c34adb4a633.
* Commented noisy debugging about packet splittingJohn Hurliman2009-10-061-4/+4
|
* Merging in diva's locking fixesJohn Hurliman2009-10-061-47/+69
|\
| * Rewrote parts of the code that were double-locking different objects. This ↵Diva Canto2009-10-061-47/+69
| | | | | | | | is about half of the code base reviewed.
* | Applying the real fix (thank you for tracking that MSDN doc down Melanie)John Hurliman2009-10-061-5/+6
| |
* | Trying Melanie's fixJohn Hurliman2009-10-061-3/+4
| |
* | * Added a sanity check for Mono before trying to enumerate over an empty ↵John Hurliman2009-10-061-0/+3
| | | | | | | | | | | | SortedDictionary * Changed the order of a log line from DotNetEngine so you can tell whether or not it is actually loading
* | * Added missing references to prebuild.xml and commented out the LindenUDP ↵John Hurliman2009-10-067-195/+195
| | | | | | | | | | | | | | | | tests until a new test harness is written * Clients are no longer disconnected when a packet handler crashes. We'll see how this works out in practice * Added documentation and cleanup, getting ready for the first public push * Deleted an old LLUDP file
* | * Try/catch around EndInvoke() when Util.FireAndForget() returns to catch ↵John Hurliman2009-10-063-78/+112
| | | | | | | | | | | | | | exceptions thrown in the async method * Added packet stats handling to the new LLUDP implementation * Attempting to avoid a race condition when creating a new LLUDPClient
* | * Continued work on the new LLUDP implementation. Appears to be functioning, ↵John Hurliman2009-10-0613-2116/+1028
| | | | | | | | | | | | | | | | although not everything is reimplemented yet * Replaced logic in ThreadTracker with a call to System.Diagnostics that does the same thing * Added Util.StringToBytes256() and Util.StringToBytes1024() to clamp output at byte[256] and byte[1024], respectively * Fixed formatting for a MySQLAssetData error logging line
* | Merge branch 'master' of ssh://opensimulator.org/var/git/opensim into ↵John Hurliman2009-10-051-18/+24
|\ \ | |/ | | | | htb-throttle
| * Changed the lock object, to see if this helps at all. Also switched the ↵Diva Canto2009-10-051-11/+15
| | | | | | | | order of lock and try/catch in RemoveImageFromQueue.
| * Reduced locking.Diva Canto2009-10-041-8/+10
| |
* | Beginning work on the new LLUDP implementationJohn Hurliman2009-10-0512-2328/+1815
|/
* Guarding a line that is sometimes throwing a null pointer exception.Diva Canto2009-10-041-5/+6
|
* Trying to avoid the null pointer exception in J2KImage, RunUpdate.Diva Canto2009-10-041-1/+9
|
* Merge branch 'diva-textures-osgrid'Melanie2009-10-041-2/+2
|\
| * * Creates Util.UTF8 and switches some references of Encoding.UTF8 to ↵John Hurliman2009-10-021-2/+2
| | | | | | | | | | | | Util.UTF8 (not all references were switched since not all OpenSim libraries reference OpenSim.Framework) * Shrinks the largest in-memory object, the LLRAW.HeightmapLookupValue struct (only used for exporting to LLRAW terrain files), to the minimum possible size. This seems to have the odd side effect of cutting the size of the two double[256,256] terrain objects in half. Possibly an alignment optimization?
* | Merge branch 'diva-textures-osgrid'Melanie2009-10-031-126/+88
|\ \ | |/
| * * Removed code that is no longer used with Melanie's queue callback optimizationJohn Hurliman2009-10-021-126/+88
| | | | | | | | | | * Moved all priority queue access to helper functions to improve reliability and readability * New version of CSJ2K.dll (from libomv source tree) that fixes a JPEG2000 comment decoding bug
* | Merge branch 'diva-textures-osgrid'Melanie2009-10-022-182/+142
|\ \ | |/
| * * Removed even more unnecessary state variables from J2KImageJohn Hurliman2009-10-022-182/+142
| | | | | | | | | | * Fixed the incorrect "out of memory?" exception messages and replaced them with useful information * Cleaned up J2KImage and LLImageManager to conform to coding conventions
* | Merge branch 'diva-textures-osgrid' into diva-texturesMelanie2009-10-023-59/+19
|\ \ | |/
| * * Changed the flush logic to drop packets in non-transactional streams, and ↵John Hurliman2009-10-023-59/+19
| | | | | | | | | | | | to not fire any "put more data in the queues" callbacks * Minor tweaks to code formatting to make the callback chain for packet queuing easier to follow
generated by cgit v1.1 at 2024-11-05 11:31:00 +0000