| Age | Commit message (Collapse) | Author |
|
|
|
First test returns quarter to a half the throughput, do two tests.
May be caused by driver, may be some 'energy saving', but not important
enough to spend time investingating. It was working the same way prior
to ExtraFPS, but viewer was running an extra CPU test that 'preheated'
the system.
Also increasing minimum throughput as numerous new features, like
mirrors and pbr were added and requirements are now higher.
|
|
This commit contains performance optimisations in the the pipeline,
framebuffer, vertexbuffer, reflection probes, shadows. It also fixes
many opengl errors, modifies the opengl debugging, and adds
a visuals effects panel.
|
|
|
|
|
|
Make `LLGLSLShader::finishProfile()` accept a string pathname instead of a
bool and, in addition to logging statistics to the viewer log, output
statistics to that file as JSON. The calls that used to pass
`emit_report=false` now pass `report_name=std::string()`.
Make llviewerdisplay.cpp's `display()` function synthesize a profile filename
in the viewer's logs directory, and pass that filename to
`LLGLSLShader::finishProfile()`.
(cherry picked from commit d5712689d36a1ee1af32242706901fde7229b08d)
|
|
|
|
* #2434 Intel HD4000 compatibility pass
* Fixes crash on old drivers when loading shaders
* Fixes 30 second hang on startup
* Fixes occasional dev build crash in LLWearableList
---------
Co-authored-by: Brad Linden <brad@lindenlab.com>
|
|
- Use GL_NVX_gpu_memory_info when available
- Disable transparent water on Mid+ and lower
- Adjust GPU benchmark to better tell the truth
- Texture bias tune up
- viewer-private/#277 - Report foreground_time in viewer stats
|
|
and llunittype.h for now
|
|
project/gltf_development
|
|
* Disable unloading of objects in background.
* Add unlit GLTF shader variant
|
|
"possible loss of precision" warnings
|
|
|
|
# Conflicts:
# autobuild.xml
# indra/cmake/CMakeLists.txt
# indra/cmake/GoogleMock.cmake
# indra/llaudio/llaudioengine_fmodstudio.cpp
# indra/llaudio/llaudioengine_fmodstudio.h
# indra/llaudio/lllistener_fmodstudio.cpp
# indra/llaudio/lllistener_fmodstudio.h
# indra/llaudio/llstreamingaudio_fmodstudio.cpp
# indra/llaudio/llstreamingaudio_fmodstudio.h
# indra/llcharacter/llmultigesture.cpp
# indra/llcharacter/llmultigesture.h
# indra/llimage/llimage.cpp
# indra/llimage/llimagepng.cpp
# indra/llimage/llimageworker.cpp
# indra/llimage/tests/llimageworker_test.cpp
# indra/llmessage/tests/llmockhttpclient.h
# indra/llprimitive/llgltfmaterial.h
# indra/llrender/llfontfreetype.cpp
# indra/llui/llcombobox.cpp
# indra/llui/llfolderview.cpp
# indra/llui/llfolderviewmodel.h
# indra/llui/lllineeditor.cpp
# indra/llui/lllineeditor.h
# indra/llui/lltextbase.cpp
# indra/llui/lltextbase.h
# indra/llui/lltexteditor.cpp
# indra/llui/lltextvalidate.cpp
# indra/llui/lltextvalidate.h
# indra/llui/lluictrl.h
# indra/llui/llview.cpp
# indra/llwindow/llwindowmacosx.cpp
# indra/newview/app_settings/settings.xml
# indra/newview/llappearancemgr.cpp
# indra/newview/llappearancemgr.h
# indra/newview/llavatarpropertiesprocessor.cpp
# indra/newview/llavatarpropertiesprocessor.h
# indra/newview/llbreadcrumbview.cpp
# indra/newview/llbreadcrumbview.h
# indra/newview/llbreastmotion.cpp
# indra/newview/llbreastmotion.h
# indra/newview/llconversationmodel.h
# indra/newview/lldensityctrl.cpp
# indra/newview/lldensityctrl.h
# indra/newview/llface.inl
# indra/newview/llfloatereditsky.cpp
# indra/newview/llfloatereditwater.cpp
# indra/newview/llfloateremojipicker.h
# indra/newview/llfloaterimsessiontab.cpp
# indra/newview/llfloaterprofiletexture.cpp
# indra/newview/llfloaterprofiletexture.h
# indra/newview/llgesturemgr.cpp
# indra/newview/llgesturemgr.h
# indra/newview/llimpanel.cpp
# indra/newview/llimpanel.h
# indra/newview/llinventorybridge.cpp
# indra/newview/llinventorybridge.h
# indra/newview/llinventoryclipboard.cpp
# indra/newview/llinventoryclipboard.h
# indra/newview/llinventoryfunctions.cpp
# indra/newview/llinventoryfunctions.h
# indra/newview/llinventorygallery.cpp
# indra/newview/lllistbrowser.cpp
# indra/newview/lllistbrowser.h
# indra/newview/llpanelobjectinventory.cpp
# indra/newview/llpanelprofile.cpp
# indra/newview/llpanelprofile.h
# indra/newview/llpreviewgesture.cpp
# indra/newview/llsavedsettingsglue.cpp
# indra/newview/llsavedsettingsglue.h
# indra/newview/lltooldraganddrop.cpp
# indra/newview/llurllineeditorctrl.cpp
# indra/newview/llvectorperfoptions.cpp
# indra/newview/llvectorperfoptions.h
# indra/newview/llviewerparceloverlay.cpp
# indra/newview/llviewertexlayer.cpp
# indra/newview/llviewertexturelist.cpp
# indra/newview/macmain.h
# indra/test/test.cpp
|
|
LF, and trim trailing whitespaces as needed
|
|
|
|
|
|
|
|
|
|
|
|
|
|
blending. (#63)
* SL-19203 WIP -- Integrate SSR with reflection probes. Decruft LLRenderTarget.
* SL-19203 WIP -- Re-integrate SSR. Incidental decruft.
* SL-19203 WIP -- SSR frame delta correction (still broken for Z)
* SL-19203 WIP -- SSR frame delta Z fix
* SL-19203 WIP -- Make SSR toggleable again and disable SSR in cube snapshots.
* SL-19203 WIP -- Soften sphere probe transitions and fix reflections on void water (make fallback probe a simple terrain+water+sky probe). Remove parallax correction for automatic probes to reduce artifacts.
* SL-19203 Tune probe blending.
* SL-19203 Cleanup.
|
|
SL-18869, SL-18772 Overhaul VBO management, restore occlusion culling, intel compatibility pass, etc
|
|
|
|
|
|
|
|
|
|
bandwidth levels and bias settings based on CPU speed.
|
|
(decruftify settings, compatibility pass).
|
|
textures
|
|
|
|
optimization bugs.
|
|
|
|
|
|
|
|
|
|
|
|
buffer during benchmark"
|
|
|
|
including safety checks and some refactoring
|
|
|
|
|
|
|
|
Introduce helper classes to manage paired initProfile() / finishProfile()
calls and gBenchmarkProgram.bind() / unbind() calls.
Make TextureHolder a class instead of a struct.
Per Henri Beauchamp, since gpu_benchmark() takes a very early exit if
(!gGLManager.mHasTimerQuery), subsequent tests of mHasTimerQuery are
redundant. Remove.
One of those tests controls the busted_finish bool, which can never become
true. Remove that and all tests on it.
|
|
A classic-C array doesn't destroy its individual elements, but a std::vector
does. Use a std::vector<LLRenderTarget> for dest, so each LLRenderTarget will
be destroyed. ~LLRenderTarget() calls its release() method.
|
|
Ruslan tracked the observed crash to assignments (to create a dummy triangle)
through an LLStrider<LLVector3> obtained from getVertexStrider(). When
getVertexStrider() returns false, produce a warning and just skip the rest of
the benchmark test.
The one bit of explicit cleanup apparently required by that early exit is a
call to LLImageGL::deleteTextures() to match the preceding generateTextures()
call. Wrap both in a new TextureHolder class whose destructor takes care of
cleanup. The only other references to the corresponding U32 array are a couple
calls to LLTexUnit::bindManual(); add a bind() method to support that.
Also fix apparent bug in the LL_DARWIN special case for "improbably high and
likely incorrect": the code assigned -1.f (the "couldn't compute" value) to
gbps, overlooking the fact that gbps is unconditionally recomputed below. In
the "likely incorrect" stanza, simply return -1.f instead.
|
|
In case of buff->getVertexStrider(v) return false it mean that glMapBufferRange() return NULL
The next three lines can be the reason of this crash.
|
|
|
|
|