Commit Graph

177 Commits

Author SHA1 Message Date
Melanie Thielker
85a9c305a2 Remove AgentID and GroupOD from the signature of SOG.Copy(). They were never
used, but made for a very mispleading read of the code in the callers.
2010-06-28 01:43:26 +01:00
Melanie Thielker
4cdda2cae9 Fix Copy on Ray, Drag Copy and other little things. Removed the wrong and
nonworking ownership assignment in SOG, which messed things up before.
No longer trust the client to send the ID of the person something is copied
as, since it allows to run a script with someone else's permissions. Properly
adjust inventory ownership and perms.
2010-06-28 01:36:40 +01:00
Melanie Thielker
660a560d3b Make drag copy and copy-on-ray handle friends list perms properly 2010-06-27 20:59:51 +01:00
Justin Clark-Casey (justincc)
52b711af82 stop KeyNotFoundException() being thrown in RemoveScenePresence if the agent isn't present in the presence dictionary
the code to do this was there but was being circumvented by newmap[agentID] before the check actually took place
2010-06-25 18:55:20 +01:00
Dan Lake
ca2abc43ad Refactor SendCoarseLocations for better performance. Instead of computing list of all locations fresh for every scene presence on every frame, we will instead compute the list once every 50 frames and send to all connected presences at that time. Also, we only add 60 items to the list when there are more than 60 presences in the scene. For 1000 users, this change yields a 99.8% reduction in list processing and a 98% reduction in network bandwidth for coarse locations. 2010-06-08 16:44:18 -07:00
Justin Clark-Casey (justincc)
b6076d7b33 Reduce number of full updates sent on region crossing for attachments/huds to 1 from 3
This is one step towards reducing hud glitches on region crossing, since the viewer fails to display prims if it receives child full updates before the root prim full update
This commit also introduces a mechanism in LLClientView to stop child attachment updates ever going out before the root one
This is a very temporary mechanism and will be commented out when the next step of the fix (to give root prims higher udpate priority) is committed
This code is a foreport from the equivalent changes in 0.6.9-post-fixes
2010-06-08 15:07:57 +01:00
Melanie Thielker
e515467c5e Fix create selection getting overwritten by multiple updates for the same prim. 2010-05-31 17:55:56 +01:00
Justin Clark-Casey (justincc)
5287c5f7fb rearrange SceneGraph.AddSceneObject() to return earlier if an object with that uuid is already in the scene
this means that we don't perform pointless work
2010-05-21 21:57:56 +01:00
Justin Clark-Casey (justincc)
f83acf533b Fix a problem where SceneGraph.AddSceneObject() would return false on successfully adding an object rather than true, in defiance of its method documentation
This meant that the returns were inconsistent - false would be returned both for various scene object failure conditions (e.g. root part was null) and if the object was successfully added.
2010-05-21 21:22:53 +01:00
Dan Lake
36bcab5f07 Refactor scene presence list for lockless iteration. Lock contention will now only be for simultaneous add/removes of scene presences from the scene. 2010-05-15 02:20:23 -07:00
Melanie
22b3217113 Fix link security issue 2010-04-30 11:46:50 +01:00
Melanie
21cad5d3ac All scripts are now created suspended and are only unsuspended when the object
is fully rezzed and all scripts in it are instantiated. This ensures that link
messages will not be lost on rez/region crossing and makes heavily scripted
objects reliable.
2010-04-19 06:29:26 +01:00
Justin Clark-Casey (justincc)
57c293d604 refactor: crudely move DetachSingleAttachmentToGround() over to the AttachmentsModule 2010-04-16 23:15:13 +01:00
Dan Lake
62e0b53ca4 Renamed TryGetAvatar to TryGetScenePresence on SceneManager, SceneBase, Scene and SceneGraph. This was the only change in this patch to keep it isolated from other recent changes to the same set of files. 2010-03-19 15:16:44 -07:00
Dan Lake
859bc717a4 Cleaned up access to scenepresences in scenegraph. GetScenePresences and GetAvatars have been removed to consolidate locking and iteration within SceneGraph. All callers which used these to then iterate over presences have been refactored to instead pass their delegates to Scene.ForEachScenePresence(Action<ScenePresence>). 2010-03-19 15:16:35 -07:00
Dan Lake
73e9b0be72 Inconsistent locking of ScenePresence array in SceneGraph. Fixed by eliminating option to return the actual list. Callers can now either request a copy of the array as a new List or ask the SceneGraph to call a delegate function on every ScenePresence. Iteration and locking of the ScenePresences now takes place only within the SceneGraph class.
This patch also applies a fix to Combat/CombatModule.cs which had unlocked iteration of the ScenePresences and inconsistent try/catch around the use of those ScenePresences.
2010-03-17 11:21:27 -07:00
Justin Clark-Casey (justincc)
582375509c refactor: move RezSingleAttachmentFromInventory() from SceneGraph to AttachmentsModule 2010-03-12 22:48:49 +00:00
Justin Clark-Casey (justincc)
b9f5cd75bc refactor: move client invoked AttachObject from SceneGraph to AttachmentsModule 2010-03-12 22:39:15 +00:00
Justin Clark-Casey (justincc)
395f343498 refactor: Move DetachSingleAttachmentToInv to region module
need to rationalize method names later
2010-03-06 00:07:47 +00:00
Justin Clark-Casey (justincc)
60553e62a3 refactor: begin to move attachments code into a region module 2010-03-05 23:18:47 +00:00
Justin Clark-Casey (justincc)
edb176447b Fix bug where approximately half the time, attachments would rez only their root prim until right clicked (or otherwise updated).
The root cause of this problem was that multiple ObjectUpdates were being sent on attachment which differed enough to confuse the client.
Sometimes these would eliminate each other and sometimes not, depending on whether the scheduler looked at the queued updates.
The solution here is to only schedule the ObjectUpdate once the attachment code has done all it needs to do.
2010-03-03 22:14:06 +00:00
Justin Clark-Casey (justincc)
660ebe52cf Actually make EventManager.OnAttach() fire when an object is attached. Previously, only detach was firing! 2010-03-03 18:40:33 +00:00
Melanie
1f45e6c649 Merge branch 'master' into presence-refactor
This brings presence-refactor up to master again
2010-02-22 03:00:59 +00:00
Jeff Ames
af265e001d Formatting cleanup. 2010-02-15 19:21:56 +09:00
Melanie
c033223c63 Merge branch 'master' into presence-refactor 2010-02-15 00:20:48 +00:00
Revolution
9821c4f566 Revolution is on the roll again! :)
Fixes: Undo, T-pose of others on login, modifiedBulletX works again, feet now stand on the ground instead of in the ground, adds checks to CombatModule. Adds: Redo, Land Undo, checks to agentUpdate (so one can not fall off of a region), more vehicle parts. Finishes almost all of LSL (1 function left, 2 events).

Direct flames and kudos to Revolution, please

Signed-off-by: Melanie <melanie@t-data.com>
2010-02-14 22:18:46 +00:00
Melanie
baaf660511 Merge branch 'master' into presence-refactor
This was a large, heavily conflicted merge and things MAY have got broken.
Please check!
2010-02-08 15:53:38 +00:00
Melanie
2953bee1f4 Revert "change position of OnAttach event firing so that this also happens when a user teleports into a region"
The behavior introduced here is not compatible with SL

This reverts commit b6bee4999c.
2010-02-04 17:11:06 +00:00
Justin Clark-Casey (justincc)
b6bee4999c change position of OnAttach event firing so that this also happens when a user teleports into a region 2010-02-03 15:35:41 +00:00
Justin Clark-Casey (justincc)
4c1740f7d8 Actually make EventManager.OnAttach() fire when an object is attached. Previously, only detach was firing! 2010-02-01 20:15:36 +00:00
Justin Clark-Casey (justincc)
53a01dc422 Don't try to start attachment scripts or send updates if the attachment itself has been unsuccessful 2010-02-01 20:06:43 +00:00
Diva Canto
42f978a478 * HGScene is no more.
* Moved a few key inventory access methods from Scene.Inventory to an IInventoryAccessModule module
2010-01-30 09:23:07 -08:00
Melanie
9453c42631 Experimental: Remove explicit property sends 2009-12-06 02:54:19 +00:00
Melanie
1a695875a7 Skip single prims in unlink selections 2009-12-06 02:54:09 +00:00
Melanie
f71025aeed Change locking to not mix explicit Monitor.* calls with lock() 2009-12-06 02:54:00 +00:00
Melanie
46d5add175 Lock updates out while linking and unlinking 2009-12-06 02:53:42 +00:00
Melanie
e3a22ff37d Eliminate multiple updates on link/unlink 2009-12-06 02:53:34 +00:00
John Hurliman
59eb378d16 Small performance tweaks to code called by the heartbeat loop 2009-10-28 23:10:16 -07:00
John Hurliman
588361e2a2 Experimental change to use an immutable array for iterating ScenePresences, avoiding locking and copying the list each time it is accessed 2009-10-23 01:02:36 -07:00
Melanie
c4969d47d9 Merge branch 'master' into vehicles 2009-10-22 07:12:10 +01:00
John Hurliman
142008121e * Change Util.FireAndForget to use ThreadPool.UnsafeQueueUserWorkItem(). This avoids .NET remoting and a managed->unmanaged->managed jump. Overall, a night and day performance difference
* Initialize the LLClientView prim full update queue to the number of prims in the scene for a big performance boost
* Reordered some comparisons on hot code paths for a minor speed boost
* Removed an unnecessary call to the expensive DateTime.Now function (if you *have* to get the current time as opposed to Environment.TickCount, always use DateTime.UtcNow)
* Don't fire the queue empty callback for the Resend category
* Run the outgoing packet handler thread loop for each client synchronously. It seems like more time was being spent doing the execution asynchronously, and it made deadlocks very difficult to track down
* Rewrote some expensive math in LandObject.cs
* Optimized EntityManager to only lock on operations that need locking, and use TryGetValue() where possible
* Only update the attachment database when an object is attached or detached
* Other small misc. performance improvements
2009-10-19 15:19:09 -07:00
John Hurliman
fdb2a75ad3 Committing the second part of Jim Greensky @ Intel Lab's patch, re-prioritizing updates 2009-10-17 18:01:22 -07:00
John Hurliman
4b75353cbf Object update prioritization by Jim Greensky of Intel Labs, part one. This implements a simple distance prioritizer based on initial agent positions. Re-prioritizing and more advanced priority algorithms will follow soon 2009-10-15 16:35:27 -07:00
John Hurliman
d44b50ee46 * Removed some of the redundant broadcast functions in Scene and SceneGraph so it is clear who/what the broadcast is going to each time
* Removed two redundant parameters from SceneObjectPart
* Changed some code in terse update sending that was meant to work with references to work with value types (since Vector3 and Quaternion are structs)
* Committing a preview of a new method for sending object updates efficiently (all commented out for now)
2009-10-15 15:25:02 -07:00
Melanie
6deef7d0f3 Merge branch 'master' into vehicles 2009-10-15 21:14:13 +01:00
Dan Lake
5976ac16b0 Optimized heartbeat by calling Update() only on updated objects.
During the heartbeat loop, Update() is called on every SceneObjectGroup which in turn checks if any SceneObjectPart has changed. For large regions (> 100k prims) this work consumes 20-30% of a CPU even though there are only a few objects updating each frame.

There is only one other reason to check every object on every frame, and that is the case where a script has registered the object with an "at target" listener. We can easily track when an object is registered or unregistered with an AtTarget, so this is not a reason to check every object every heartbeat.

In the attached patch, I have added a dictionary to the scene which tracks the objects which have At Targets. Each heartbeat, the AtTarget() function will be called on every object registered with a listener for that event. Also, I added a dictionary to SceneGraph which stores references to objects which have been queued for updates during the heartbeat. At each heartbeat, Update() is called only on the objects which have generated updates during that beat.
2009-10-13 19:32:59 -07:00
Melanie
89d23a1fa2 Revert "Rewrote parts of the code that were double-locking different objects. This is about half of the code base reviewed."
This reverts commit e992ca0255.
2009-10-07 01:45:49 +01:00
Diva Canto
e992ca0255 Rewrote parts of the code that were double-locking different objects. This is about half of the code base reviewed. 2009-10-06 15:39:53 -07:00
Melanie
41ff39414b Merge branch 'master' into vehicles 2009-10-01 14:08:15 +01:00
opensim
827b0fb199 Commit initial version of KittoFlora's vehicle changes 2009-09-30 18:51:02 +02:00