aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region/ClientStack/Linden (follow)
Commit message (Collapse)AuthorAgeFilesLines
* enqueue also if m_nextPackets[category] is not null. This is really theUbitUmarov2014-08-271-4/+8
| | | | | top element of a category queue, equivalente to using a queue.peek() if avaiable
* dont append acks to a resend packetUbitUmarov2014-08-261-1/+1
|
* *test* back to no sog kills, little retouch on attachment updates,UbitUmarov2014-08-251-1/+1
| | | | NameValue AttachItemID is a root part thing only
* *test* cleanup the bugs nestUbitUmarov2014-08-251-6/+8
|
* change enconding of attachment updates NameValue and State fieldsUbitUmarov2014-08-251-0/+19
|
* clean sendKill a bit, remove a IsChild that is set too early in a callingUbitUmarov2014-08-241-22/+2
| | | | path
* on updates, send rotations using livomv Quaternion.toBytes() againUbitUmarov2014-08-231-4/+2
|
* Remove entities from updates queues on kill. Do it sync so enqueues afterUbitUmarov2014-08-231-0/+6
| | | | the kill work
* add a direct sendpartfullUpdate to send a full object update to a part,UbitUmarov2014-08-211-0/+26
| | | | optionally overriding its parentID. check what it does to attachments
* revert droping udp packet resends after 6 retries, keep resending.UbitUmarov2014-08-211-10/+0
|
* remove from use the UpdatesResend on resending udp packets. Just resendUbitUmarov2014-08-202-36/+54
| | | | | the UDP packet. Also just loose packets we tried to send 6 times already (ll says 3) A viewer may just beeing ignoring them, or then the link is just dead.
* send zero velocity again on avatar full update or its uglyUbitUmarov2014-08-191-1/+1
|
* fix the encoding of rotation in updates, not just using the next field toUbitUmarov2014-08-191-15/+12
| | | | override w bytes. ( specially having it commented )
* still another ...UbitUmarov2014-08-191-2/+1
|
* fix a missed blocking of sending updates the the new attach points aboveUbitUmarov2014-08-191-2/+1
| | | | hud indexes
* NextAnimationSequenceNumber be a udpserver variable with random startUbitUmarov2014-08-162-2/+22
|
* reduce ping cliping lower limitUbitUmarov2014-08-141-2/+2
|
* reduce ping filter time constantUbitUmarov2014-08-141-1/+1
|
* add a estimator of client ping time, and painfully make it visible in showUbitUmarov2014-08-143-5/+41
| | | | connections console command
* Merge branch 'ubitworkmaster'Melanie Thielker2014-08-132-63/+68
|\
| * remove HandleCompleteMovementIntoRegion delay hack from llUDPserver. If weUbitUmarov2014-08-132-11/+17
| | | | | | | | | | need a delay, we need to do it at end of HandleUseCircuitCode before feeding pending packets (including that one) into processing queue.
| * process AgentUpdates in order with rest of packets. Only give higher ↵UbitUmarov2014-08-131-2/+4
| | | | | | | | priority to chat
| * take agentUpdate checks out of llUDPserver (disabling useless debug) andUbitUmarov2014-08-122-43/+35
| | | | | | | | do it only where its supposed to be done..
| * do the significance test invalidation at MoveAgentToRegionUbitUmarov2014-08-121-4/+3
| |
| * no need to go from cos into squared sin when just abs of cos is as goodUbitUmarov2014-08-121-5/+7
| |
| * force AgentUpdate after CompleteAgentMovement to pass by significance testUbitUmarov2014-08-121-5/+9
| |
* | Revert "make HandlerRegionHandshakeReply processing async and delay it a ↵Melanie Thielker2014-08-131-6/+2
|/ | | | | | | | | | bit. This" This reverts commit 30f00bfb14cce582382bd37b1e22062af664ec64. Conflicts: OpenSim/Region/ClientStack/Linden/UDP/LLClientView.cs
* actually let delay match throttles as original designUbitUmarov2014-08-101-11/+9
|
* OutgoingPacketHandler can not be paced by hits on SendPacket()UbitUmarov2014-08-101-4/+8
|
* also on other caseUbitUmarov2014-08-091-0/+11
|
* return even zero costs ( prims on other regions )UbitUmarov2014-08-091-1/+1
|
* move from RegisterInventoryServiceHandlers toUbitUmarov2014-08-091-9/+14
| | | | RegisterRegionServiceHandlers what belongs there ( readabilitly only )
* Reduce delay time on regin handshake reply to speed up terrain sending.Melanie Thielker2014-08-071-1/+1
| | | | Terrain IS more important than avatar!
* minor clean, dont check for cache if we aren't using it..UbitUmarov2014-08-071-82/+22
|
* reply to cached checks with same serial numberUbitUmarov2014-08-071-3/+4
|
* DANGER... changed bakedtextures caching. Assuming grid baking is cacheUbitUmarov2014-08-072-135/+99
| | | | only, reduced number of accesses to it. TESTING
* bakemodule didnt like last changesUbitUmarov2014-08-062-13/+8
|
* some debug to remove laterUbitUmarov2014-08-062-1/+9
|
* try to make the baked textures cache workUbitUmarov2014-08-061-33/+40
|
* check...UbitUmarov2014-08-041-17/+17
|
* Revert "start sending terrain in scenePresence after well defined avatar. Minor"UbitUmarov2014-08-042-37/+25
| | | | This reverts commit 05a2feba5d780c57c252891a20071800fd9f2e3e.
* start sending terrain in scenePresence after well defined avatar. MinorUbitUmarov2014-08-042-25/+37
| | | | change on significante AgentUpdate check.
* comment out a debug msg. more on gods being GodsUbitUmarov2014-08-031-1/+1
|
* make HandlerRegionHandshakeReply processing async and delay it a bit. ThisUbitUmarov2014-07-301-2/+6
| | | | | | | delays things like terrain height map sending, giving chance to completemovent to send more priority information. POssible things on this should be on other event trigger when its more apropriate really done.
* Send new parcel permissions to activate viewer options - done right this timeMelanie Thielker2014-07-291-4/+6
|
* Add new fields to parcel updateMelanie Thielker2014-07-291-1/+11
|
* dont send packets twiceUbitUmarov2014-07-271-1/+0
|
* add queue transition code needs review/more testingUbitUmarov2014-07-251-79/+115
|
* remove old and dead code that its only making it hard to understandUbitUmarov2014-07-251-287/+18
|
* change previus code, send a NoEvents response back to http server, evenUbitUmarov2014-07-241-2/+1
| | | | | if we know there is no one there, so that the server finishes handling it.