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.
This reduces base memory churn of every client connection, improving the garbage collection situation.
The effect is a significant portion of base load (an avatar standing still on a completely blank island)
but will probably still be swallowed up by other memory use on active regions.
Tests have shown no noticeable impact on speed of processing incoming packets, though setting remains
in case a switch back is needed.
Setting this to false will block all restart requests from the viewer even if they are otherwise legitimate.
One use is to block region restarts if necessary whilst restart functionality remains buggy or triggers bugs in modules,
though these should be fixed as soon as practicable.
Default is true, as has been the case historically.
Simple attempt to make avatars better shaped.
Replace parameter 'avatarCapsuleRadius' with 'avatarCapsuleWidth'
and 'avatarCapsuleDepth'.
More tweeking to avatar height calculation. A little better but
short avatar's feet are above the terrain and tall avatar's feet
are a little below the ground.
nant_0.91~alpha2+dfsg-3_all.deb in Ubuntu 12.04 and earlier actually ignored these due to a bug
However, nant 0.92~rc1+dfsg-2 in Ubuntu 12.10 fixes this bug (possibly https://github.com/nant/nant/pull/39).
Which makes nant time-consumingly copy these files when the aren't actually used.
Tested removal of <copy> on both nant 0.91 and nant 0.92
Will be submitting this patch to prebuild project for comment though I suspect there's nobody there to pay attention.
This allows a closed grid to delete asset types other than maptile remotely.
Only operational if AllowRemoteDelete = true also.
Defaults to false - do not enable if anybody other than you can make asset service requests.
This is because macosx mono is 32-bit and this can't p/invoke 64-bit binaries.
However, the reverse is also true.
If OpenSimulator stops working for you then please complain! Long term alternative is probably to build a fat binary with both architectures.
This retrieves and caches information from the PresenceService to only send messages to online users.
This is reported to much improve performance for large groups where most users are offline.
Cache is 20 seconds to balance requests against users not receiving messages until cache updates.
This is an alternative to an approach where login/logout notification is sent directly from simulator to groups service.
However, I'm not convinced that this PresenceService approach is actually better. Needs more thought.
Even when an avatar is standing still, it's sending in a constant stream of AgentUpdate packets that the client creates new UDPPacketBuffer objects to handle.
This option pools those objects. This reduces memory churn.
Currently off by default. Works but the scope can be expanded.
We can provide modular ini for connectors...
look for our configuration in the following places...
1) in the default ini/-inifile
2) in the named file (ConfigName) located in the configured directory
(see Robust[.HG].ini [Start] section for ConfigDirectory)
3) in the repository named in the connector (ConfigURL)
In this case, the file will be written into the configured
directory with the specified
See example connector/service @ https://github.com/BlueWall/SlipStream
for testing.
Rearrangement and cleanup of shape collection code. Much more readable.
Enabling and use of collision filters and masks.
Addition of ID to body creation BulletSimAPI calls so always set in
shape for collision reporting.
Change default of ShouldSplitSimulationIslands and ShouldRandomizeSolverOrder
from 'false' to 'true'. When 'false', this suppresses NO_CONTACT_RESPONSE
which makes volume detect fail.
This controls how many undo steps the simulator will store for each prim.
Default is now 20 rather than 5 as it briefly was.
The default number could be increased through this is a memory tradeoff which will scale with the number of prims in the sim and level of activity.
This resends appearance uuids to avatars in the scene once a minute.
I have seen this help in the past resolve grey appearance problems where viewers have for unknown reasons sometimes ignored the packet.
The overhead is very small since only the UUIDs are sent - the viewer then requests the texture only if it does not have it cached.
This setting will not help with cloudy avatars which are usually due to the viewer not uploading baked texture data or uploading something that isn't valid JPEG2000
This is because libopenjpeg 1.5 appears to require a minimum of glibc 2.14, whereas at least one fairly recent distro (openSUSE 11.4 from 2011-03-10) only has glibc
Further investigation pending.
Among other things this allows pCampbot to work under Windows since libopenmateverse now ships the same log4net.dll (publicly signed) as OpenSimulator
This also updates the libopenmetaverse embedded libopenjpeg from 1.3 to 1.5.
DLL naming and mapping for non-Windows libopenjpeg changes to remove version number to make future udpates easier and bring it into line with names of other shipped DLLs.
libopenjpeg updates have been made for OSX, Windows (32 and 64 bit) and Linux (32 and 64 bit). Please report any issues.
Fix inconsistencies between configuration parameter names and their description names. Changing the configuration parameters for non physical prim size min-max from Nonphys* to NonPhys*.
Please update your OpenSim.ini and Regions.ini to reflect these changes.
This is to resolve a problem where an asset marked as local but not temporary but still used in the scene would be removed.
The timed expiry scan no longer tries to refetch assets from the scene that are not currently in the cache - this is not helpful since it just drags a lot of data into the cache that may never be referenced.
This removes the DeepScanBeforePurge option since setting this to false will introduce the above problem. This previously had a default of true.
Viewer LL 3.3.4 and before sometimes fail to properly redisplay dynamic textures that have a small data length compared to pixel size when pulled from cache.
This appears to happen when the data length is smaller than the estimate discard level 2 size the viewer uses when making this GetTexture request.
This commit works around this by always regenerating dynamic textures that fall below this threshold rather than reusing them if ReuseDynamicTextures = true
This can be controlled by the [Textures] ReuseDynamicLowDataTextures config setting which defaults to false.
If true, this setting reuses dynamically generated textures (i.e. created through osSetDynamicTextureData() and similar OSSL functions) where possible rather than always regenerating them.
This results in much quicker updates viewer-side but may bloat the asset cache (though this is fixable).
Also, sometimes issue have been seen where dynamic textures do not transfer to the viewer properly (permanently blurry).
If this happens and that flag is set then they are not regenerated, the viewer has to clear cache or wait for 24 hours before all cached uuids are invalidated.
CUrrently experimental. Default is false, as before.
This is necessary because commit 8131a24 (Tue Mar 27 10:08:13 2012) started passing the config section name rather than hardcoding "AssetService"
This meant that the HG external-facing asset service tried to read ConnectionString from [HGAssetService] rather than [AssetService].
On SQLite, not finding this meant that it fell back to [DatabaseService], which is set for OpenSim.db rather than Asset.db.
Therefore, all external asset requests returned null.
Solution taken here is to create an [HGAssetService] section with the same ConnectionString as [AssetService].
This bug does not affect normal MySQL/MSSQL config since they use the [DatabaseService] connection string anyway.
Addresses http://opensimulator.org/mantis/view.php?id=6200, many thanks to DanBanner for identifying the exact problem commit which was very helpful.
This was a regression from OpenSimulator 0.7.3.1 which did not contain this bug.
This also fixes computation of avatar mass.
Added parameter MaxPersistantManifoldPoolSize.
Fixed a parameter setting bug which caused crashes of there were
more than 400 or so physical objects. I tested up to 5000.
Updated BulletSim DLLs and SOs.
Add description and default of false (as before) to OpenSimDefaults.ini
If set to true, this config switch will resend avatar appearance information (a small amount of UUID data, not the baked textures themselves) to other avatars in the sim every 60 seconds.
For me, this has helped with situations where avatars appear persistently grey - the LL viewer sometimes did not appear to request assets the first time the appearance data was sent.
However, this switch will not help with other appearance failure situations (e.g. failure to bake assets).
This setting is experimental but will not have any significant impact on the simulator if turned to true.
This is to try and resolve a strange Initialization exception for the AppearanceManager type that only occurs on Windows 64 bit (not Windows 32 bit or Linux!)
Built from https://github.com/openmetaversefoundation/libopenmetaverse/commits/master
This update incorporates a fix by Melanie in libomv where textures would sometimes rotate slightly when other properties were changed.
Hence addresses http://opensimulator.org/mantis/view.php?id=6100
This update also incorporates libomv bot code updates that significantly reduce the warning spam for pCampbot.
The change rate of libomv is low and many changes refer to bot or code not used by OpenSimulator, so I think it is sufficiently stable to use evne though it's not a release.
and into its own class. The BulletSim data structures track
individual prims as linksets of 1 so most of the prim code is not
different between a linked and unlinked object.
Only files ending in .ini and .xml are actually loaded.
Also improves the help at the top of Regions.ini.example to mention that OpenSimulator would generate Regions.ini anyway on first startup.
Also changes separator from comma to bar to match existing [LoginService] config features.
Divergence of config names for identical facilities in different places makes for an unnecessarily confusing user experience.
This is to provide an indication of what's happening now that the default isn't to report every single script start.
Changes XEngine logging level in OpenSim.exe.config from WARN to INFO.
This is to reduce log spam from script loading, which is especially spammy for avatar movements with scripted attachments.
All important messages are at warn or above.
If you still want/need to see these messages, set <level value="DEBUG"/> in the <logger name="OpenSim.Region.ScriptEngine.XEngine"> section of OpenSim.exe.config.
This affects no other package logs, which still output at the root configured level (currently DEBUG by default).
The packet was actually being handled but not acted on.
This change extends the default timeout for paused clients to 5 minutes
and makes both the paused and non-paused timeout periods configurable.
If active, the physics module can return arbitrary stat counters that can be seen via the MonitoringModule
(http://opensimulator.org/wiki/Monitoring_Module)
This is only active in OdeScene if collect_stats = true in [ODEPhysicsSettings].
This patch allows OdeScene to collect elapsed time information for calls to the ODE native collision methods to assess what proportion of time this takes compared to total physics processing.
This data is returned as ODENativeCollisionFrameMS in the monitoring module, updated every 3 seconds.
The performance effect of collecting stats is probably extremely minor, dwarfed by the rest of the physics code.
Without this, LL 3.3.1 continually pushes LLInventoryModelFetchDescendentsResponder::error 499 to its log.
This cap will be ignored by older viewers - UDP inventory will work normally.
BlueWall