Age | Commit message (Collapse) | Author | |
---|---|---|---|
2021-12-09 | SL-16461 disable occlusion-based reflection update | Dave Houlton | |
2021-12-09 | SL-16471 Removed unneeded MeshOpt generation method | Andrey Kleshchev | |
2021-12-08 | SL-16468 Fix for crash when enabling highlight transparent (add rigged mesh ↵ | Runitai Linden | |
support to highlight transparent). | |||
2021-12-08 | SL-16469 Fix for reflection/refraction passes breaking occlusion queries. | Runitai Linden | |
2021-12-07 | SL-16462 Fix for broken water clip plane during water reflection/refraction ↵ | Runitai Linden | |
map render. | |||
2021-12-07 | SL-16454 FIXED Performance viewer login screen is black after a Release ↵ | Mnikolenko Productengine | |
viewer session. | |||
2021-12-06 | SL-16447 Fix for broken benchmark. | Runitai Linden | |
2021-12-06 | Merge branch 'DRTVWR-546' of ssh://bitbucket.org/lindenlab/viewer into ↵ | Runitai Linden | |
DRTVWR-546 | |||
2021-12-06 | SL-16451 Fix for rigged meshes disappearing when rendered as static. | Runitai Linden | |
2021-12-06 | SL-16282 FIXED Friend thumbnails are flickering | Mnikolenko Productengine | |
2021-12-06 | SL-16202 Put Multi-threaded GL behind a feature flag and update featuretable ↵ | Dave Parks | |
(decruftify settings, compatibility pass). | |||
2021-12-03 | SL-16436 and SL-16327 Fix for RenderDebugGL test failures and fix for grey ↵ | Dave Parks | |
textures | |||
2021-12-02 | SL-16405 Remove “pass == <magic number>” anti-pattern from various ↵ | Dave Parks | |
drawpools. | |||
2021-12-02 | SL-14403 Remove unused shader binds | Andrey Kleshchev | |
remains from glod | |||
2021-12-02 | SL-16420 Upload's physics LODs are broken in preview #2 | Andrey Kleshchev | |
2021-12-02 | mac build fix | Andrey Kleshchev | |
remove unused variable | |||
2021-12-01 | SL-16420 Upload's physics LODs are broken in preview | Andrey Kleshchev | |
2021-12-01 | Merged in euclid-pre-16386 (pull request #793) | Euclid Linden | |
Purge LLGLSLShader::sNoFixedFunction, and all that flows from it. No functional changes. Approved-by: Michael Pohoreski Approved-by: Dave Parks | |||
2021-12-01 | SL-16386 skip void return | Dave Houlton | |
2021-11-30 | SL-16386 consolidate all VertexShader capability checks to a single ↵ | Dave Houlton | |
LLPipeline::shadersLoaded() | |||
2021-11-30 | SL-16386 remove references to (const true) LLPipeline::mVertexShadersEnabled | Dave Houlton | |
2021-11-30 | SL-16386 replace gutted LLRender::setAlphaRejectSettings() with explicit flush() | Dave Houlton | |
2021-11-30 | SL-16386 purge write-only state member LLCubeMap::mTextureCoordStage | Dave Houlton | |
2021-11-30 | SL-16386 purge no-op fxn LLTexUnit::setTextureAlphaBlend() | Dave Houlton | |
2021-11-30 | SL-16386 purge no-op fxn LLTexUnit::setTextureColorBlend() | Dave Houlton | |
2021-11-30 | SL-16386 purge no-op fxn LLTexUnit::setTextureBlendType() | Dave Houlton | |
2021-11-30 | SL-16386 remove references to (const true) LLGLSLShader::sNoFixedFunction | Dave Houlton | |
2021-11-30 | SL-16421: Destroy the "General" ThreadPool as soon as cleanup starts. | Nat Goodspeed | |
Introduce LLAppViewer::onCleanup(), a method that accepts a nullary callable to execute once viewer shutdown begins. Fire the collected callables in LLAppViewer::cleanup(). In llstartup.cpp, instead of declaring a static unique_ptr and relying on static object destruction to clean up the "General" ThreadPool, bind the pointer to the new ThreadPool into an onCleanup() lambda that will delete it when called. ~ThreadPool() takes care of orderly shutdown. | |||
2021-11-30 | SL-16421: Unify LLAppViewer::cleanup() indentation. | Nat Goodspeed | |
Use hard tabs because most of the existing function uses those. | |||
2021-11-30 | SL-16421: Unify LLAppViewer class declaration alignment. | Nat Goodspeed | |
Use hard tabs as most of the class declaration already uses those. | |||
2021-11-30 | SL-16369 Remove Hardware Skinning setting | Mnikolenko Productengine | |
2021-11-29 | DRTVWR-542 merge, purge GLOD copyrights from xml | Dave Houlton | |
2021-11-29 | Merge branch 'DRTVWR-542-meshopt' into DRTVWR-546 | Andrey Kleshchev | |
2021-11-29 | SL-16094: Merge branch 'DRTVWR-546' into glthread | Nat Goodspeed | |
2021-11-29 | SL-16408 fixed regression - crash on exit | Mnikolenko Productengine | |
2021-11-25 | SL-16408 Fix for crashing on disconnect | Mnikolenko Productengine | |
2021-11-24 | SL-16094: Clean up a bit more merge cruft. | Nat Goodspeed | |
2021-11-24 | SL-16094: Fix WorkQueue test for correct behavior of runFor(). | Nat Goodspeed | |
Turns out that one of our WorkQueue integration tests was relying on the incorrect runFor() behavior that we just fixed, so the test broke. Now that runFor() doesn't wait around for work to be posted, use an explicit wait loop instead. To support this, add LLCond::get(functor), where functor must accept a const reference to the stored data. This new get() returns whatever the functor returns, allowing a caller to peek at the stored data. Also use universal references for all remaining LLCond functor arguments. | |||
2021-11-24 | DRTVWR-546, SL-16220, SL-16094: Undo previous glthread branch revert. | Nat Goodspeed | |
Reverting a merge is sticky: it tells git you never want to see that branch again. Merging the DRTVWR-546 branch, which contained the revert, into the glthread branch undid much of the development work on that branch. To restore it we must revert the revert. This reverts commit 029b41c0419e975bbb28454538b46dc69ce5d2ba. | |||
2021-11-24 | SL-16400: Make WorkQueue::runFor() and runUntil() stop when done. | Nat Goodspeed | |
runFor(interval) and runUntil(timestamp) are intended, and documented, to run *no longer than* the specified time. Instead, the initial implementation always waited the full specified time, hoping for work to arrive. Fix that: once we clear work that's already pending, return right away. | |||
2021-11-24 | SL-16094: Fix merge glitches from previous revert. | Nat Goodspeed | |
2021-11-23 | SL-16094, SL-16400: Merge branch 'DRTVWR-546' into glthread | Nat Goodspeed | |
2021-11-23 | SL-16400: Address a couple shutdown crashes. | Nat Goodspeed | |
It can happen that we try to post() work for LLWindowWin32's window thread after the thread's WorkQueue has been closed. Also, instead of giving the "General" ThreadPool static lifespan, put it on the heap, anchored with a static unique_ptr. | |||
2021-11-23 | SL-16400: Add ThreadPool::start() method, and call it. | Nat Goodspeed | |
It's sometimes important to finish other initialization before launching the threads in the ThreadPool, so make that an explicit step. In particular, we were launching the LLImageGL texture thread before initializing the GL context, resulting in all gray textures. | |||
2021-11-23 | SL-9436 Fix for glow disappearing when 100% transparent. | Dave Parks | |
2021-11-23 | SL-16368 Fix for crashing on right-clicking mesh avatar on 32bit viewer | Mnikolenko Productengine | |
2021-11-23 | SL-16239 Fix for slowdown on AMD GPUs (disable core profile and remove ↵ | Dave Parks | |
volatile members from LLVertexBuffer) | |||
2021-11-23 | Merge branch 'glthread' of ssh://bitbucket.org/lindenlab/viewer into glthreadx | Nat Goodspeed | |
2021-11-23 | SL-16094: Fix second startup hang. | Nat Goodspeed | |
Add LLWindowWin32Thread::Post(), like post() except it uses PostMessage() to send the work item to the window thread. Support this in mainWindowProc(). Move LLWindowWin32::recreateWindow()'s destroy_window_handler() call onto the window thread. Delaying destruction of the old HWND ensures that we can use PostMessage() and GetMessage() with that HWND to pass the lambda work item. Moreover, it's likely to be less buggy to call DestroyWindow() on the same thread that created the window. Make recreateWindow()'s window thread lambda bind the window class parameters by value, rather than binding 'this' and back-referencing LLWindowWin32 members. Make recreateWindow() construct the window thread lambda and then decide whether to pass it to the window thread using post() or Post(), depending on whether we have a current HWND -- therefore whether the window thread is blocked on GetMessage(). That means we can eliminate the kickWindowThread() call. Make destroy_window_handler() accept HWND by value rather than by non-const reference. Since it doesn't attempt to modify the caller's value, this is a better match for the function's semantics anyway -- but importantly, it lets us pass a const HWND. | |||
2021-11-23 | SL-16094: Support ll_convert<std::string>(const char*) | Nat Goodspeed | |
and correspondingly, ll_convert<std::wstring>(const wchar_t*). Now that we're using ll_convert() for single-argument stringize(arg), make sure it can efficiently handle the simple case of constructing a string from a const char pointer. |