aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Use a thread abort safe version of OpenMetaverse.DoubleDictionary with the ↵Justin Clark-Casey (justincc)2012-12-072-1/+511
| | | | | | | | | | aim of avoiding OpenSimulator problems due to script thread aborts. When an object is removed, its scripts are stopped and then the thread running them is aborted if stop takes too long. However, it appears that aborting a thread at just the wrong moment when it is obtaining a ReaderWriterLockSlim lock can leave this lock in an inconsistent state. One symptom of this is that mono leaps to 100% cpu and a vm thread dump reveals lots of threads waiting for a ReaderWriterLockSlim lock without any thread actually holding it. This is probably the same problem as encountered originally in commit 12cebb12 This commit looks to plaster this problem by putting lock obtaining methods inside finally blocks which should be uninterruptible by thread aborts.
* minor: change method doc on GetTextureHandler.TryParseRange(), mainly to ↵Justin Clark-Casey (justincc)2012-12-061-2/+2
| | | | trigger another build
* Allow GetTexture calls with no second value in the range header (e.g. just ↵Justin Clark-Casey (justincc)2012-12-061-1/+34
| | | | | | | | 5333-) It looks like the latest Kokua is doing this. As per http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html, leaving off the second value is legal This indicates the caller wants the rest of the entity.
* Instead of printing script errors out to console, put to debug log so that ↵Justin Clark-Casey (justincc)2012-12-061-1/+11
| | | | | | | | we also get timestamps. This commit also adds script name, part name, uuid, etc. for later identification. This information has been sent to console since 2009 but may be turned down if it proves too noisy. However, I still currently need it to investigate some region problems probably triggered by scripting.
* minor: Fix more compiler warnings in CoreModules tests by properly ↵Justin Clark-Casey (justincc)2012-12-055-5/+16
| | | | overriding OpenSimTestCase.SetUp()
* Add engine-wide events queued and events processed numbers to output of ↵Justin Clark-Casey (justincc)2012-12-051-1/+14
| | | | "xengine status" console command. For debugging purposes.
* Remove some mono warnings in script tests, chiefly where SetUp() wasn't ↵Justin Clark-Casey (justincc)2012-12-055-7/+13
| | | | properly calling to OpenSimTestCase.SetUp()
* Add IScriptInstance.EventsQueued to match EventsProcessed instead of asking ↵Justin Clark-Casey (justincc)2012-12-053-4/+15
| | | | callers to lock and directly inspect the EventQueue
* For now, sort "show scripts" output in descending order sorted by events ↵Justin Clark-Casey (justincc)2012-12-051-3/+21
| | | | | | processed. For debug purposes - should later add options to allow different sorting or show only highest 10, etc.
* Add IScriptInstance.EventsProcessed stat so that we can record this ↵Justin Clark-Casey (justincc)2012-12-053-0/+10
| | | | information and display in "show scripts" for debug purposes
* Remove very probably unused PrimitiveBaseShape(bool) constructor to reduce ↵Justin Clark-Casey (justincc)2012-12-051-12/+6
| | | | code complexity
* Don't pointlessly set ExtraParams = byte[1] in PrimitiveBaseShape since this ↵Justin Clark-Casey (justincc)2012-12-051-3/+0
| | | | is ignored by the ExtraParams properly anyway
* minor: tidy up spacing at bottom of MundaneFrameworkTestsJustin Clark-Casey (justincc)2012-12-051-6/+2
|
* In BaseHttpServer.HandleRequest(), use Culture.SetCurrentCulture() rather ↵Justin Clark-Casey (justincc)2012-12-051-1/+1
| | | | than creating a new CultureInfo separately
* In XEngine.DoOnRezScript() use Culture.SetCurrentCulture() instead of ↵Justin Clark-Casey (justincc)2012-12-051-2/+1
| | | | constructing a new CultureInfo separately
* On XEngine.ProcessEventHandler(), instead of creating a new CultureInfo on ↵Justin Clark-Casey (justincc)2012-12-051-2/+1
| | | | | | | every call use the single one set by Culture.SetCurrentCulture() This is slightly different in that SetCurrentCulture() does not use overridden settings if the system culture matches en-US but some settings there have been changed. This is what we want - we do not want to use any system overriden settings.
* minor: also comment out the debug log message which reports searching for ↵Justin Clark-Casey (justincc)2012-12-041-2/+2
| | | | child agents in simulator scenes for now.
* minor: Comment out "Delivering IM to..." messages for now.Justin Clark-Casey (justincc)2012-12-041-3/+3
|
* minor: Put Scene.PhysicsRequestAsset() into standard C# xml format.Justin Clark-Casey (justincc)2012-12-041-4/+11
|
* Bug fix in OfflineMessageModule. Mantis #6446Diva Canto2012-12-041-1/+1
|
* Merge branch 'master' of ssh://opensimulator.org/var/git/opensimDiva Canto2012-12-0418-322/+724
|\
| * Merge branch 'master' of ssh://opensimulator.org/var/git/opensimBlueWall2012-12-0314-287/+578
| |\
| | * BulletSim: Reduce idle region physics overhead where there are MANYRobert Adams2012-12-037-80/+95
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | static objects by more restrictive selection of objects that collide with static objects. Rename collision mask fuctions from 'filter' to 'group' so it is clear what is being set. Rename BulletSimAPI.SetCollisionFilterMask() to SetCollisionGroupMask to match above. Restore passing of time step to linear and angular motion component routines. Use buffering vehicle physical parameter get/set routines consistantly. Make range enforcement clearer by using ClampInRange() function for parameter setting. Remove commented out experimental vehicle calculations.
| | * BulletSim: rework angular corrections to remove any hybrid code and compute ↵Robert Adams2012-12-034-111/+115
| | | | | | | | | | | | absolute collections.
| | * BulletSim: add stubs for generalization of preStep actions. Will eventually ↵Robert Adams2012-12-032-3/+22
| | | | | | | | | | | | replace the specialized vehicle processing with preStep event processing. Add TODO comments about this feature. Redo line endings in TODO file to be all Linux.
| | * BulletSim: begin tracking a TODO list. There just are so many things to ↵Robert Adams2012-12-031-0/+112
| | | | | | | | | | | | remember to do.
| | * BulletSim: revert angular vertical attraction from motor to code. The motor ↵Robert Adams2012-12-031-47/+19
| | | | | | | | | | | | code did not return the restoring difference but the current value. Remove unused commented out code.
| | * BulletSim: format vehicle detail logging messages so vehicle changs are ↵Robert Adams2012-12-032-17/+17
| | | | | | | | | | | | grouped better in the log output.
| | * BulletSim: localize vehicle property setting so the vehicle prim is only ↵Robert Adams2012-12-031-33/+128
| | | | | | | | | | | | updated at the end of the vehicle simulation step and the push of the physics property update event only happens if the properties are actually changed.
| | * BulletSim: Add DumpActivationInfo2 function. Change static objects from ↵Robert Adams2012-12-014-10/+15
| | | | | | | | | | | | DISABLE_SIMULATION to ISLAND_SLEEPING. Update DLLs and SOs to add DumpActivationInfo2 function.
| | * BulletSim: remove time scaling of computed vehicle absolute velocity since ↵Robert Adams2012-11-293-48/+73
| | | | | | | | | | | | Bullet will scale the movement by the time slice. Restore LIMIT_MOTOR_UP to definitition of BOAT simce some vehicle engines use it even for land vehicles. Push vehicle parameter updates through the regular property update to solve vehicles floating off when they should be stopped.
| | * BulletSim: add copyright header where it is missing. Remove some unnecessary ↵Robert Adams2012-11-292-3/+31
| | | | | | | | | | | | 'using' requirements so testing framework is less complicated.
| | * BulletSim: add expanded call to IMesher/Meshmerizer which enables/disables ↵Robert Adams2012-11-294-10/+26
| | | | | | | | | | | | mesh caching. Since BulletSim caches and tracks the unmanaged memory version of meshes, the Meshmerizer itself does not need to cache built meshes once BulletSim has made the physical proxy mesh.
| | * BulletSim: reverse direction of hover correction. Removes problem with ↵Robert Adams2012-11-291-8/+8
| | | | | | | | | | | | vehicles being orbited.
| * | XmlRpcGridRouterBlueWall2012-12-034-19/+77
| |/ | | | | | | | | | | | | | | | | Flesh out XmlRpcGridRouter to reap unused channels from gateway when scripts or objects are removed, or when the llCloseRemoteDataChannel is called. See: http://http://forge.opensimulator.org/gf/project/xmlrpcrouter/ or https://github.com/BlueWall/XmlRpcRouter for php gateway and test code.
| * BulletSim: fix boats floating low by removing LIMIT_MOTOR_UP flag from ↵Robert Adams2012-11-281-3/+15
| | | | | | | | TYPE_BOAT definition.
| * BulletSim: move GetWaterLevelAtXYZ from BSScene to BSPhysTerrain.Robert Adams2012-11-287-22/+63
| |
* | Improvement in HGInstantMessageService: account for the existence of an ↵Diva Canto2012-12-042-10/+140
|/ | | | offline IM service.
* Merge branch 'master' of ssh://opensimulator.org/var/git/opensimDiva Canto2012-11-273-4/+66
|\
| * Stop logging spurious asset data for {0} is zero length error for sculpts/meshJustin Clark-Casey (justincc)2012-11-282-1/+9
| | | | | | | | | | | | | | | | ODEPrim, for instance, always hits this code path twice at the moment Firstly before any sculpt data has been loaded (hence the spurious message) Secondly when any sculpt data has been loaded or failed to load (when the message would be valid). Hence comment this out and rely on the message in ODEPrim.MeshAssetReceived() instead (though this is not ideal since it requires all physics plugins to copy/paste similar code).
| * Merge branch 'master' of ssh://opensimulator.org/var/git/opensimJustin Clark-Casey (justincc)2012-11-284-205/+296
| |\
| * | Show many more primitive properties on console command "show part name/id/pos"Justin Clark-Casey (justincc)2012-11-281-3/+57
| | |
* | | Bug fix in Offline IM for inventory transfers. The items were being placed ↵Diva Canto2012-11-271-11/+15
| |/ |/| | | | | twice in the receiver's inventory.
* | Merge branch 'master' of ssh://opensimulator.org/var/git/opensimDiva Canto2012-11-27120-2727/+3739
|\ \
| * | BulletSim: reorganize linear movement routine into separate subroutines ↵Robert Adams2012-11-271-58/+79
| | | | | | | | | | | | enabling external calibration routines and unit tests.
| * | BulletSim: fix terrain mesh generation for problem with regions that have ↵Robert Adams2012-11-271-6/+16
| | | | | | | | | | | | unequal edge heights. Thanks UBit.
| * | BulletSim: reorganize angular movement routine into separate subroutines ↵Robert Adams2012-11-271-95/+114
| | | | | | | | | | | | enabling external calibration routines and unit testing.
| * | BulletSim: implementation of vertical attraction motor.Robert Adams2012-11-271-41/+62
| | |
| * | BulletSim: add 'infinite' timescale that does not reduce motor target or ↵Robert Adams2012-11-271-10/+27
| |/ | | | | | | friction.
| * Fix database service unit test failures by temporarily reverting ↵Justin Clark-Casey (justincc)2012-11-261-1/+7
| | | | | | | | | | | | | | | | | | BasicDataServiceTest extending OpenSimTestCase. Mono 2.4.3 doesn't like this when running nunit, with nunit throwing AssetTests`2 : System.MemberAccessException : Cannot create an instance of OpenSim.Data.Tests.AssetTests`2[TConn,TAssetData] because Type.ContainsGenericParameters is true. and similar. Mono 2.10.8.1 does not have this issue. So will wait until min version of mono bumps before restoring.
generated by cgit v1.1 at 2024-12-30 17:23:29 +0000