Justin Clark-Casey (justincc)
f7b8c54c24
Add comment for experimental effect of removing the Thread.Sleep(20) in ODEPrim.changevelocity()
2011-11-04 23:43:17 +00:00
Justin Clark-Casey (justincc)
ccca6ba935
Stop llPushObject() from causing problems by adding force via a taint rather than directly.
...
This isn't a perfect solution since there can be a race between the taint processing and taint setting, as force needs to be reset after processing.
Needs careful locking in the future.
2011-11-04 23:12:01 +00:00
Justin Clark-Casey (justincc)
9fdd1753fa
Add taint target velocity for ODECharacters as is already done for ODECharacter position and position and velocity for ODEPrims.
...
This is to help stop surprises if the velocity is set in the middle of physics calculations, though this probably isn't a huge problem.
It's more for consistency and for the next step of removing some scene locks
2011-10-29 02:30:33 +01:00
Justin Clark-Casey (justincc)
ef8370fb8e
tidy up OdeCharacter so that we just use OpenMetaverse.Vector3 assignment directly where possible, instead of transferring X, Y and Z components separately
...
some of this is probably a hold over from using ODE.Vector3, which is still necessary in some places.
2011-10-29 02:07:28 +01:00
Justin Clark-Casey (justincc)
a5ea9f8830
Move position set from taint to logically better position at top of ODECharacter.ProcessTaints() though this makes no practical difference
2011-10-29 01:46:22 +01:00
Justin Clark-Casey (justincc)
5ae8de3c00
Stop setting _position as well as m_taint_position in ODECharacter.Position
...
setting position at the same time as taint appears to undermine the whole purpose of taint
testing doesn't reveal any obvious regressions in doing this
2011-10-29 01:39:48 +01:00
Justin Clark-Casey (justincc)
820242bc49
Fix a bug I introduced yesterday in ODE physics where prim scripts would only receive the very first collision.
2011-10-27 02:05:59 +01:00
Justin Clark-Casey (justincc)
6a74a4c12b
Clear OdeCharacter CollisionEventUpdate when we subscribe or unsubscribe from collision events
2011-10-25 22:46:42 +01:00
Justin Clark-Casey (justincc)
0f83f87233
Remove unused fields from CollisionEventUpdate
2011-10-25 22:39:08 +01:00
Justin Clark-Casey (justincc)
b9f106f484
When sending object collision updates, don't null out and recreate the CollisionEventUpdate() if the number of collisions falls to zero. Reuse the existing one instead.
2011-10-25 22:28:40 +01:00
Justin Clark-Casey (justincc)
5d37f0471e
For ScenePresence collision events, instead of creating a new CollisionEventsThisFrame every time we need to send some new ones, reuse the existing one instead.
...
This assumes that the listener is using the data synchronously, which is currently the case.
2011-10-25 22:19:17 +01:00
Justin Clark-Casey (justincc)
7b6b36cee9
Fix bug where collision event listeners were not removed once the listener had gone away.
...
This was causing continuous use of temporary memory even when all avatars had left the scene.
Memory does leak but it does cause more calls to the garbage collector, which would pause the scene thread for a very short while during collection.
2011-10-25 21:40:51 +01:00
Justin Clark-Casey (justincc)
1fbb379536
minor: rename a parameter in OdeScene.Simulate() from actor -> prim since it's an OdePrim
2011-10-25 21:15:37 +01:00
Justin Clark-Casey (justincc)
4241ee5dfa
very minor removal of old commented out line of code in OdeScene
2011-10-21 00:04:36 +01:00
Justin Clark-Casey (justincc)
30fe66d3ab
remove unnecessary null check on _collisionEventPrim
2011-10-20 20:58:29 +01:00
Justin Clark-Casey (justincc)
9c43020876
Get OdeScene to use passed in time step rather than hard-coded 0.089
...
However, I still don't recommend changing MinFrameTime from 0.089, high values do not work well and lower values don't seem to make much difference
2011-10-20 20:48:51 +01:00
Justin Clark-Casey (justincc)
b63ec987b0
For now, stop passing timeStep into methods where it's not actually used.
2011-10-20 17:54:32 +01:00
Justin Clark-Casey (justincc)
581885da75
Temporarily put in log lines to record time taken to set terrain in OdeScene.
2011-10-18 21:21:15 +01:00
Justin Clark-Casey (justincc)
03202ada29
Store scene identifier passed in to OdeScene for later debug messages
2011-10-18 21:11:13 +01:00
Justin Clark-Casey (justincc)
1d4cd76e8a
Don't bother taking OdeLock during OdeScene construction, since there can be no contention until the object is constructed.
2011-10-18 21:03:41 +01:00
Justin Clark-Casey (justincc)
f10a824e47
Remove unnecessary lock of OdeLock in OdePrim.changeadd()
...
This taint can only ever be processed from the OdeScene.Simulate() loop, which already locks OdeLock.
2011-10-18 20:58:59 +01:00
Justin Clark-Casey (justincc)
5515c45e3b
minor: method doc to explain a lock of OdeLock
2011-10-18 20:56:11 +01:00
Justin Clark-Casey (justincc)
71d221cdc0
Remove the unused CollisionLocker from ODE
...
Despite its name, this wasn't actually being used in any collision checking
2011-10-18 17:35:58 +01:00
Justin Clark-Casey (justincc)
0c041ce12f
Implement osNpcSit(). This is still in development so don't trust it
...
Format is osNpcSit(<npc-uuid>, <target-uuid>, OS_NPC_SIT_IMMEDIATE)
e.g. osNpcSit(npc, llGetKey(), OS_NPC_SIT_IMMEDIATE);
At the moment, sit only succeeds if the part has a sit target set.
NPC immediately sits on the target even if miles away - they do not walk up to it.
This method is in development - it may change so please don't trust it yet.
Standing will follow shortly since that's kind of important once you're sitting :)
2011-10-17 01:42:31 +01:00
Justin Clark-Casey (justincc)
978fb3d482
reduce access to ODECharacter methods to make code analysis easier. Eliminate redundant argument on ProcessTaints()
2011-10-15 01:41:39 +01:00
Justin Clark-Casey (justincc)
9cc2694776
restrict unnecessary access levels on ODEPrim fields/properties
2011-10-15 01:31:09 +01:00
Justin Clark-Casey (justincc)
f2132329a3
refactor: make methods that do not need to be public in ODE private or internal to aid code reading/analysis. Remove some unused method arguments
2011-10-15 01:20:40 +01:00
Justin Clark-Casey (justincc)
4bfc2f5cde
Change hardcoded ODE total frame time to match the default total frame time (0.09375 -> 0.089).
...
No apparant ill effects - because the default stepsize is 0.2, there are still 5 physics steps per physics frame.
This is a precursor to using the elapsed value passed in (and now changeable in config).
2011-10-14 21:07:57 +01:00
Justin Clark-Casey (justincc)
20da04fd0c
More method doc and formatting changes. Makes DestroyOdeStructures() private
2011-10-13 21:42:24 +01:00
Justin Clark-Casey (justincc)
227db07f2f
refactor: move 3x copy/pasted ode structure removal code in ODECharacter into a DestroyOdeStructures() method
...
also adds some method doc
2011-10-13 18:19:13 +01:00
Justin Clark-Casey (justincc)
9090039324
Remove unused local variable i from OdeScene.Simulate()
2011-10-05 22:15:28 +01:00
Justin Clark-Casey (justincc)
1e798136c3
adjust some whitespace to trigger another build, to check the last failure was just a glitch
2011-09-17 01:33:55 +01:00
Justin Clark-Casey (justincc)
6f542f73d4
Stop the avatar stalling on its first boarder cross when using the ODE plugin
...
When upgrading the previously child agent to a root, the code was setting the Size parameter on the ODECharacter PhysicsActor.
This in turn reset Velocity, which cause the border stall.
I'm fixing this by commenting out the Velocity = Vector3.Zero lines since they don't appear to play a useful purpose
2011-09-16 23:57:16 +01:00
Justin Clark-Casey (justincc)
951ffad81e
If SP.MoveToTarget has been called with a force walk, begin by landing the avatar.
...
There is a bug here - once an avatar has landed it glides to its new position instead of performing a walk animation
2011-08-11 00:23:54 +01:00
Justin Clark-Casey (justincc)
7f499ff3f3
Add a OS_NPC_LAND_AT_TARGET option to osMoveToTarget()
...
Default for this function is now not to automatically land.
This allows better control by scripts when an avatar is going to be landing on a prim rather than the ground.
Stopping the avatar involves faking a collision, to avoid the pid controller making it overshoot.
A better approach would be to gradually slow the avatar as we near the target
2011-08-10 23:56:19 +01:00
Justin Clark-Casey (justincc)
05e94ff27e
Move common gemo/agent map name code into CreateGeom()
...
Fix build break.
2011-08-01 07:04:13 +01:00
Justin Clark-Casey (justincc)
f9d6a91252
Instead of preserving old name in geom_name_map in change size or shape, use the Name property instead.
...
This is equivalent since the prim 'name' is never changed. In fact, this propery is never used for prims
2011-08-01 06:52:52 +01:00
Justin Clark-Casey (justincc)
40a78db182
comment out unused code in OdeScene.TriCallback()
2011-08-01 06:47:45 +01:00
Justin Clark-Casey (justincc)
ccb4b76242
On geom removal, remove the name from the OdeScene.geom_name_map too
2011-08-01 06:40:29 +01:00
Justin Clark-Casey (justincc)
f79df6f43f
remove the unused ODEPrim.prev_geom field
2011-08-01 06:35:59 +01:00
Justin Clark-Casey (justincc)
f32dbef647
When an ODE geom is removed (as when a non-phantom prim is deleted, resized or shape changed, also remove the OdeScene.actor_name_map entry pointing to the phys actor
...
This is to stop a small memory leak over time when prims are deleted or phantom-toggled
2011-08-01 06:32:30 +01:00
Justin Clark-Casey (justincc)
6618948ff9
refactor: centralize prim geom removal code from four places to one
2011-08-01 06:15:02 +01:00
Justin Clark-Casey (justincc)
509200d5cd
minor: add note to RemovePrimThreadLocked() to the effect that it contrary to the summary, it is being called from within Simulate() lock (OdeLock)
2011-08-01 05:48:27 +01:00
Justin Clark-Casey (justincc)
dfa2f7d715
If a prim changes size or shape, add actor to _parent_scene.actor_name_map with new prim_geom key, as the old one becomes invalid.
...
This resolves http://opensimulator.org/mantis/view.php?id=5603 where changing size or shape would stop collision_start being fired in a running script.
In both this and existing code we are not removing old actors from actor_name_map when the existing prim_geom is removed, which leads to a small memory leak over time.
This needs to be fixed.
2011-08-01 05:43:04 +01:00
Justin Clark-Casey (justincc)
2102964826
minor: indentation correction
2011-08-01 05:14:16 +01:00
Justin Clark-Casey (justincc)
d041276517
Remove _mesh field since the mesh data no longer needs to be stored after it's initially used.
...
This may improve memory usage for regions using mesh and sculpts, though I suspect that it doesn't address the current memory leak.
2011-07-31 01:03:52 +01:00
Justin Clark-Casey (justincc)
bd8f538f80
refactor: Remove argument to pass in an initial mesh to OdePrim since this is no longer required and it prevents removal of the _mesh field (which is only used temporarily)
...
If passing in a mesh becomes important again in the future then this can be reinstated.
2011-07-31 00:49:49 +01:00
Justin Clark-Casey (justincc)
196a774b24
minor: correct method doc for last commit
2011-07-31 00:45:20 +01:00
Justin Clark-Casey (justincc)
454312f5bc
refactor: rename CreateGeom _mesh argument to mesh, so as to not confuse this with the pre-existing _mesh field
2011-07-31 00:44:23 +01:00
Justin Clark-Casey (justincc)
f1ce17071d
minor: method doc
2011-07-31 00:43:02 +01:00
