Patch from 2010-10-20/22. Didn't apply cleanly so I redid it
manually and tested/checked it. There was only one difference
with imprudence: for some reason Singularity tries to see
if the "teleport history" floater is visible while initializing
the menu's, something that imprudence doesn't do. For the rest
the patch is basically the same. Nevertheless, I left all code
exactly as it was in Singularity (the only thing that this
commit changes is the intend of the original patch: in debug
mode, fail with an assertion when LindenUserDir is called while
it is not initialized; and then fix everything to make it work
again.
Original commit message:
The LindenUserDir (~/.imprudence/first_last/) cannot be initialized
before the user logged in. However, several singletons (that only can be
initialized once) depend on this directory for initialization. Therefore
we have to take care not to instantiate those singletons until after the
user logged in.
With regard to webit, this fixes the browser_profile (cache and cookies)
directory that the builtin browser uses.
LL has since removed high-res snapshots using render targets. I find this change to be favorable, as rendertargets were very finicky to get working with anti-aliasing. Also, deferred rendering uses many rendertargets that depend on screen resolution. Changing resolution every screenshot is.. not very awesome.
There is some deviation from LL's viewer. I've kept the old tiled glow pass enabled for non-deferred, as it fixes issues with tile edges truncating glow propogation. However, this does no work with deferred yet. I need to pin down why. I assume using binding one of the RenderTargets is required for deferred.
Additionally, the usage of a RenderTarget for snapshots is what prevented me from fully enabling my supersampling settings. Now that that hurdle is gone, I consider including this new setting by default to be safe enough.
And a note: Do not remove the 'tiling' variable when merging with v2 changes, as singu's glow pass is 'special'. There are other fixes absent from LL's viewer that require knowing if the render is tiled or not. (water reflections/cloud freezing)
Need to test:
localassetbrowser
preview related floaters
hgfloatertexteditor
maps
media textures! Currently very hacky
web browser
alpha masks on avatars
bumpmaps
Are all sky components appearing?
LLViewerDynamicTexture (texture baking, browser, animated textures, anim previews, etc)
Snapshot related features
Customize avatar
vfs floater
UI textures in general
Texture priority issues
Deferred fastalpha behaves a bit better.
Pulled over patch for https://jira.secondlife.com/browse/STORM-336 and https://jira.secondlife.com/browse/STORM-1011 from linden repo
Sky rendered using new LL method. Assuming this fixes issues on AMD cards(works on cat 11.2)
Added a few things missed related to spatial-parition updating.
Added 'SkipReflectOcclusionUpdates' setting that prevents occlusion updates for reflection pass. Less taxing on CPU.
Enabled bulk glReadPixels for snapshots if target-resolution matches render-resolution
Snapshot thumbnail generated from downscaled render image, instead of requring an extra pass.
-const'd all readonly LLCached[COA]Controls to prevent altering by mistake.
-added get() accessor to LLCached[COA]Controls to return cached type explicitly without manual casts.
-added missing operators to LLCachedCOAControl
-Converted a few more settings to LLCached[COA]Control
-Fixed LLColor4U -> LLSD -> LLColor4 problems in LLCachedControl
Ascent functionality fixes:
-Tag coloring pulled out of getClientInfo(Redundant. idleUpdateNameTag already handled it)
--Tag and color substituted at render-time. Simple boolean logic. Neglible perf penalty.
-Fixed llinfos spam if AscentShowSelfTagColor = false
-Client tag updated less superfluously
Added LLObjectList::getAvatar(LLUUID) for type-safe and quicker lookup
Added F32 overload to LLSD because using F64 needlessly is silly.
(And its lack of F32 overload caused issues with LLCachedControl<F32>)
