aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Mantis#2860. Thank you kindly, Thomax for a patch that:Charles Krinke2008-12-211-1/+1
| | | | | | | Improves map image representation.
* * Re-implement packet tracking in IClientAPI so we can see what's going on ↵Teravus Ovares2008-12-215-1/+28
| | | | in the client network in the simstats manager. This makes packets in per second, packets out per second, and unacked bytes work again in the simulator stats section.
* Slowing things down even more on TPs, to see if that helps the European ↵diva2008-12-202-3/+7
| | | | folks connected to osgrid.
* Mantis#2796. Thank you kindly, Gerhard for a patch that addresses:Charles Krinke2008-12-206-24/+139
| | | | | | | | | | | | | | On a call of llVolumeDetect(1) (or any other number !=0) volume detection is enabled. Together with VD, the phantom flag is set to the GUI. On a call of llVolumeDetect(0), vd detection is switched of again, also the phantom state is removed. On a call to llSetState(STATE_PHANTOM, false) while VD is active, also VD is switched off. The same is true for unchecking the phantom flag via GUI. This allows to take back VD without the need to script just by removing the phantom flag. Things missing in this patch: persistance of the volume-detection flag. This needs more discussion and will be included in another patch soon.
* This reduces DNS lookups in TPs by a factor of 4. Let's see if this fixes ↵diva2008-12-203-13/+36
| | | | some of the crashes experienced by ppl in Europe connected to OSGrid. If DNS lookup does not resolve, they may experience a substantial increase in "Region not available", but they shouldn't crash.
* * Add two more missing defaults that use GetBoolean without a default in ↵Teravus Ovares2008-12-201-2/+2
| | | | standalone. (not sure if this will fix your issue daTwitch.. it's probably not even a bug)
* Fix Mantis#2878 for now by not resetting physical values on rez (only on ↵Homer Horwitz2008-12-201-1/+1
| | | | | | | switch). We still might have to split the rotational velocities into a physical and a non-physical part...
* Slowing things down a little bit in TPs, with a couple of extra Thread.Sleep ↵diva2008-12-202-2/+10
| | | | to give enough time for the destination to set up CAPs for the incoming agent before sending EAC to the client.
* * Adds defaults for some GetBoolean config values that previously did not ↵Teravus Ovares2008-12-201-3/+3
| | | | have them.
* The previous fix for the EQ might have been too radical, and some early ↵diva2008-12-191-15/+4
| | | | events now may be getting dropped because the EQ is not there yet. So, let's add an EQ immediately upon registering caps.
* let you specify a hex number as the color in the vector renderingSean Dague2008-12-191-1/+11
| | | | | module. This expands the colors you can use.
* - Set acceleration and rotational velocity to 0, tooHomer Horwitz2008-12-191-7/+10
| | | | | | - Pull it up a bit, so the physics is notified of it, too, before being switched off.
* * refactor: move sound trigger from scene into sepearate SoundModuleJustin Clarke Casey2008-12-193-19/+114
|
* Stop prims that get non-physical. Fixes Mantis#2859.Homer Horwitz2008-12-191-0/+7
|
* Remove forgotten debug message from my last commit.Homer Horwitz2008-12-191-1/+0
|
* Fix for more child agent mismanagement (like #2846). On enqueueing events, ↵diva2008-12-191-5/+44
| | | | we should not be creating new queues; if the queue doesn't exist -- too bad, event is dropped. That tends to happen just after the client is closed, and the EQ also closed. If someone places an event in that queue after that, that event should be dropped instead of creating a new, erroneous queue.
* * minor: Just some method commentary before turning to another subjectJustin Clarke Casey2008-12-191-2/+10
|
* minor: remove mono compiler warningJustin Clarke Casey2008-12-191-1/+0
|
* * refactor: move saved named prims to xml2 method out into the serialization ↵Justin Clarke Casey2008-12-195-26/+36
| | | | module
* minor: Remove some serialization module scene wrappersJustin Clarke Casey2008-12-193-35/+26
|
* * refactor: remove xml serialization wrappers from SceneJustin Clarke Casey2008-12-192-18/+6
|
* * refactor: Rename IRegionSerialiser and move into interfaces namespaceJustin Clarke Casey2008-12-198-21/+19
|
* Reset Expire date for tempOnRez objects on rez. Fixes Mantis#2848Homer Horwitz2008-12-191-4/+13
|
* * minor: move IGesturesModule into the correct namespaceJustin Clarke Casey2008-12-191-1/+1
|
* * refactor: Remove archiver module scene wrappersJustin Clarke Casey2008-12-194-34/+15
|
* * Add missing file from last commitJustin Clarke Casey2008-12-191-0/+56
|
* * refactor: Move tree code out into a separate moduleJustin Clarke Casey2008-12-195-57/+215
|
* * Commit patch from cmickeyb. #2871. Optimized float array for the terrain ↵Teravus Ovares2008-12-191-5/+13
| | | | | | | heightfield to reduce cpu usage on new client significantly. Thanks cmickeyb!
* Make llSetAlpha and llSetTexture properly queue full updates, so changesMelanie Thielker2008-12-181-1/+1
| | | | | | are visible right away
* Removing a check in Teleports. This check should be there, I think, but it ↵diva2008-12-181-20/+22
| | | | may be causing problems right now.
* Fix and streamline the RegionHandshake stuff for estate managersMelanie Thielker2008-12-181-8/+3
|
* * refactor: move gestures code out from Scene into its own moduleJustin Clarke Casey2008-12-183-45/+155
|
* This may fix mantis #2855. There was a race condition on the ↵diva2008-12-181-9/+49
| | | | TextureDownloadModule upon clients (ScenePresences) being closed. If there were still textures to send, the UserTextureServices was created again, but pointing to the old IClient that had just been closed, which made things not work upon that user returning to that region.
* Enabled complex meshing for simple box prims with non-zero shearDahlia Trimble2008-12-181-2/+3
|
* Stop hacked godmode from being able to actually move other people'sMelanie Thielker2008-12-181-0/+4
| | | | | | attachments
* * Added "show queues" command that shows throttling queues for all clients.lbsa712008-12-184-5/+53
| | | | | *** This only works for LLCLientView at the moment ***
* * Catch a WebException in the LoadImageURLModuleTeravus Ovares2008-12-181-33/+40
|
* This innocuous change in the TextureDownloadModule suddenly made the texture ↵diva2008-12-181-1/+4
| | | | problems (#2855) go away for me. I'm committing it, so that others can test and refute, to make sure I'm not hallucinating.
* Fix attachment persistence which got broken with some inventoryMelanie Thielker2008-12-171-1/+6
| | | | | | changes a few revs back.
* Reapply the packet optimization patch, with changesMelanie Thielker2008-12-173-57/+42
|
* * minor: Comment out "Informing neighbours" debug message that pops up ↵Justin Clarke Casey2008-12-171-1/+1
| | | | | | | | whenever some aspect of the avatar changes. * with Diva's blessing ;)
* Revert 7754 and 7755 pending a reworkMelanie Thielker2008-12-173-47/+59
|
* Stop packets that may be resent from being returned to the packet pool.Melanie Thielker2008-12-171-1/+8
| | | | | | FWIW
* * revert r7724 so that PresenceChildStatus() starts throwing NRE's again Justin Clarke Casey2008-12-172-3/+8
| | | | | | * apparantly logout code relies on this happening in certain circumstances. Really, the root issue needs to be investigated.
* Change the PacketQueue to use the stored size as wellMelanie Thielker2008-12-171-23/+16
|
* Marry AckData to LLQueItem, and store packet data and length there forMelanie Thielker2008-12-172-36/+31
| | | | | | use everywhere. Each packet gets serialized only once now in PacketHandler
* * remove mono compiler warningsJustin Clarke Casey2008-12-176-27/+27
| | | | | | * should work - the last compile failure looks like a random glitch...
* * minor: Comment out high volume 'Getting MapItems' messagesJustin Clarke Casey2008-12-171-15/+15
| | | | | | * make worldmap module log tags consistent
* Kicking child agents off on the first login after a client crash. The child ↵diva2008-12-172-1/+25
| | | | | | | | agents kickoff was missing from when this is run in grid mode. But, worse, turns out that this whole feature of kicking off the user after a crash was entirely missing from standalone mode.
* * Only serialize packets once for the length calculation for throttle queues ↵Justin Clarke Casey2008-12-171-16/+25
| | | | rather than twice
generated by cgit v1.1 at 2024-11-02 18:38:40 +0000