Commit message (Collapse) | Author | Files | Lines | ||
---|---|---|---|---|---|
2009-10-26 | * Switched from OpenJPEG to CSJ2K in Meshmerizer | John Hurliman | 1 | -3/+1 | |
* Tested the previous patch and found no regressions | |||||
2009-10-26 | Experimental change of PhysicsVector to Vector3. Untested | John Hurliman | 3 | -34/+56 | |
2009-10-26 | Added calls to GC.AddMemoryPressure() when unmanaged memory is allocated for ↵ | John Hurliman | 1 | -0/+6 | |
ODE (helps the GC make better scheduling choices), and a call to GC.Collect() right before logins are enabled for a region. Although this doesn't change actual memory usage, it improves the reported usage from OpenSim and the operating system | |||||
2009-10-25 | Changing Scene.ForEachClient to use the synchronous for loop instead of ↵ | John Hurliman | 1 | -1/+1 | |
Parallel. This is quite possibly the source of some deadlocking, and at the very least the synchronous version gives better stack traces * Lock the LLUDPClient RTO math * Add a helper function for backing off the RTO, and follow the optional advice in RFC 2988 to clear existing SRTT and RTTVAR values during a backoff * Removing the unused PrimitiveBaseShape.SculptImage parameter * Improved performance of SceneObjectPart instantiation * ZeroMesher now drops SculptData bytes like Meshmerizer, to allow the texture data to be GCed * Improved typecasting speed in MySQLLegacyRegionData.BuildShape() * Improved the instantiation of PrimitiveBaseShape | |||||
2009-10-19 | Formatting cleanup. | Jeff Ames | 1 | -3/+3 | |
2009-10-18 | Zero out PrimitiveBaseShape.SculptData after the JPEG2000 data has been ↵ | John Hurliman | 1 | -0/+3 | |
decoded to allow garbage collection on it | |||||
2009-10-13 | * Eliminated unnecessary parameters from LLUDPServer.SendPacketData() | John Hurliman | 1 | -1/+1 | |
* Changed PrimMesher's Quat.Identity to return <0,0,0,1> instead of <0,0,0,1.1> | |||||
2009-10-11 | sync PrimMesher with r46 | dahlia | 2 | -19/+782 | |
2009-10-11 | disambiguate System.Path (preparation for new PrimMesher) | dahlia | 1 | -1/+1 | |
2009-10-05 | Eliminate pinned Mesh data on managed heap by using IntPtrs to memory ↵ | Dan Lake | 1 | -54/+112 | |
allocated on the unmanaged heap. This prevents fragmentation of the managed heap and the resulting stress on GC. A region with ~150,000 prims using ODE and Meshmerizer saw memory remain flat around 1.2GB as opposed to 1.5GB and continually growing due to pinned memory. This patch complements the unique mesh dictionary patch applied to Meshmerizer but is independent. The net effect is a 60-75% reduction in memory for our largest regions. | |||||
2009-10-04 | only cache mesh if meshing was successful | dahlia | 1 | -8/+4 | |
re-enable sculpt mesh caching | |||||
2009-10-04 | flexible extrusion shape defaults to linear | dahlia | 1 | -1/+1 | |
2009-10-04 | Exclude sculpt proxies from mesh caching | dahlia | 1 | -6/+23 | |
2009-09-30 | Create j2k cache directory if it doesn't exist. The Meshmerizer needs it there. | Diva Canto | 1 | -0/+14 | |
2009-09-29 | Add copyright header. Formatting cleanup. | Jeff Ames | 2 | -2/+2 | |
2009-09-25 | corrections to viewerMode AddPos() | dahlia | 2 | -1/+40 | |
2009-09-24 | Meshmerizer stores dictionary of unique Meshes keyed on construction ↵ | Dan Lake | 2 | -64/+92 | |
parameters. CreateMesh() returns a Mesh from the dictionary or creates a new Mesh if it has not been created before. Meshes are never purged from the dictionary. The raw Mesh data is discarded once the memory is pinned for ODE use. All copies of the same prim/mesh use the same pinned memory. ONLY IMPLEMENTED AND TESTED WITH MESHMERIZER AND ODE Signed-off-by: dahlia <dahliaTrimble@gmailDotCom> | |||||
2009-09-02 | eliminate unnecessary storage of mesh objects which are no longer necessary ↵ | dahlia | 1 | -1/+0 | |
after recent startup optimizations | |||||
2009-09-03 | Add copyright headers. Formatting cleanup. | Jeff Ames | 1 | -2/+2 | |
2009-09-01 | Applying Mantis #4079. Thank you, dslake | Melanie | 2 | -109/+57 | |
2009-08-03 | starting to refactor path generation out of prim extrusion | dahlia | 1 | -2/+18 | |
2009-07-30 | Flag certain meshes to avoid vertex list compaction. This should help speed ↵ | Dahlia Trimble | 1 | -0/+2 | |
up meshing time at startup. | |||||
2009-07-14 | add a constructor overload and an AddPos method to SculptMesh | Dahlia Trimble | 1 | -0/+33 | |
2009-07-11 | Some improvements to prim face numbering - only affects viewers using PrimMesher | Dahlia Trimble | 1 | -11/+17 | |
2009-06-10 | Formatting cleanup. | Jeff Ames | 1 | -3/+1 | |
2009-06-06 | some code cleanup | Dahlia Trimble | 1 | -3/+0 | |
test commit with cygwin svn | |||||
2009-06-04 | Fix comparison of UUID with null. | Jeff Ames | 1 | -1/+1 | |
2009-06-01 | Minor: Change OpenSim to OpenSimulator in older copyright headers and ↵ | Jeff Ames | 5 | -5/+5 | |
LICENSE.txt. | |||||
2009-05-31 | reduce verbosity of sculpt map cache messages to reporting cache failures only | Dahlia Trimble | 1 | -2/+2 | |
2009-05-29 | some clean up of sculpt map caching code | Dahlia Trimble | 1 | -5/+3 | |
remove a redundant debug message | |||||
2009-05-29 | reinstate a hopefully more robust experimental decoded sculpt map caching scheme | Dahlia Trimble | 1 | -3/+5 | |
2009-05-29 | disable sculpt map caching until a better method of avoiding asset requests ↵ | Dahlia Trimble | 1 | -5/+8 | |
can be found | |||||
2009-05-29 | Experimental decoded sculpt map caching | Dahlia Trimble | 1 | -17/+48 | |
2009-05-22 | cleaning out warnings. | Dr Scofield | 1 | -37/+37 | |
NOTE: we currently have a gazillion warnings caused stuff flagged as "obsolete" (OGS1 stuff) --- what's up with that? | |||||
2009-05-19 | Sculpt mesher refactor | Dahlia Trimble | 1 | -58/+98 | |
adds some previously missing geometry to sculpties new LOD improves vertex accuracy fix torus mode mesh edge joining sync with primmesher r37 | |||||
2009-05-15 | some sculpted prim geometry accuracy and meshing speed improvements | Dahlia Trimble | 1 | -22/+7 | |
2009-05-08 | Added a Copy() method to PrimMesh and SculptMesh as suggested by dmiles. ↵ | Dahlia Trimble | 2 | -20/+75 | |
Sync PrimMesher.cs and SculptMesh.cs with PrimMesher.dll r36. | |||||
2009-05-03 | alter behavior of sculpted prim "Inside out" setting. Addresses Mantis #3514 | Dahlia Trimble | 1 | -1/+1 | |
2009-04-24 | some code cleanup | Dahlia Trimble | 1 | -18/+0 | |
sync with primmesher r35 | |||||
2009-04-24 | Limit hollow size of physics proxy to 95% | Dahlia Trimble | 2 | -2/+4 | |
2009-04-18 | Add some documentation. (note this is *not* a thinly veiled attempt to ↵ | Dahlia Trimble | 1 | -1/+3 | |
increase my commit frequency *wink*) | |||||
2009-04-11 | * Instead of referencing mesh stuff in the physics plugin.. change the IMesh ↵ | Teravus Ovares | 1 | -3/+13 | |
Interface. (blame prebuild) | |||||
2009-04-05 | unspecified sculpt stitching mode now defaults to plane instead of sphere. ↵ | Dahlia Trimble | 1 | -1/+3 | |
Addresses Mantis #3403 | |||||
2009-04-03 | fix an error in hollow cylinder face number calculation | Dahlia Trimble | 1 | -1/+2 | |
2009-04-02 | partial support for sculpted prim "inside out" setting | Dahlia Trimble | 1 | -11/+38 | |
2009-04-01 | Hard code scaled image resolution to 96 dpi prior to sculpt meshing to ↵ | Dahlia Trimble | 1 | -2/+2 | |
prevent a mono 2.4 failure. Thanks to cmickeyb for pointing out the failure. | |||||
2009-03-26 | add x-axis mirror capability to sculpted prim mesh - addresses Mantis #3342 | Dahlia Trimble | 2 | -2/+16 | |
2009-03-26 | make some arrays static to prevent excessive re-initialization - suggested ↵ | Dahlia Trimble | 1 | -5/+5 | |
by jhurliman | |||||
2009-03-15 | fixed propagation of normalized sculpt mesh vertex normals | Dahlia Trimble | 1 | -3/+3 | |
2009-02-24 | remove log4net dependency and from PrimMesher.cs | Dahlia Trimble | 2 | -7/+18 | |
sync PrimMesher.cs with PrimMesher.dll version 29 on forge |