* Register added nodes into EditorOnly expression collection to be saved/displayed correctly
* Fix conversion invalidation when a node is deleted
* Change Substrate version meaning:
* -1 no conversion has run
* 0 the conversion fonction has run and no conversion was done. Or the material has been changed/touched since its conversion
* >=0 the conversion function has run
#rb sebastien.hillaire
[CL 36760583 by charles derousiers in 5.5 branch]
Refactor CookByTheBookFinishedInternal to move all optional writes into the per-platform forloop so we can share data calculated for each platform between ReferencedSet.txt and the AssetRegistry.
Remove the no-longer-used SCOPED_BOOT_TIMING macros; we profile cook times using UE_SCOPED_HIERARCHICAL_COOKTIMER now.
#rnx
#rb Zousar.Shaker
[CL 36759205 by matt peters in 5.5 branch]
Fix lingering hard reference to deleted landscapes being held by landscape info objects.
#rb jonathan.bard
[CL 36758879 by roey borsteinas in 5.5 branch]
Force SM6 for preview VK SM5, Mac SM5
Enable Lumen Support for Preview VK SM5, Mac SM5
Modifying FeatureLevelSwitch to respect the parent ShaderPlatform FeatureLevel and not the one set by preview in order to have consistent material logic with the Shader Platform we are trying to preview
#rb Jack.Porter, jeannoe.morissette
[CL 36758806 by florin pascu in 5.5 branch]
- Various legacy game thread code assumes the render thread will never be more than one frame behind. The original change in 36468180 switched from syncing the GT with the RT, to syncing the GT with the RHIT. That left the render thread "floating" in the center of the pipeline, and led to cases where resources are deleted too soon.
- New approach is to always sync with the GT with the N-1 RT frame, so the GT is never too far ahead of the RT. This maintains compatibility with the legacy GT code paths. In addition to the GT->RT sync, we also sync with the RHIT to prevent the engine running ahead, which was the original bug that 36468180 was fixing.
- "r.GTSyncType" mode 0 now allows for 1 frame of GT->RT overlap, and 2 frames of GT->RHIT overlap.
- For debugging purposes, "r.GTSyncType" can also be made negative, which increases the number of GT->RHIT overlap frames, e.g. "r.GTSyncType -3" gives 5 frames of GT->RHIT overlap. While this is not overly useful in a shipped title, it can be used to prove the correctness of the rendering pipeline.
- Merged the FDeferredCleanupInterface / FPendingCleanupObjects processing into the FFrameEndSync code path, plus made FFrameEndSync a static singleton. This allows us to manage the N frames of overlap between the GT and RHIT in a central place, and correctly cleanup deferred resources when the RT fences have passed.
In future, we will need to revisit this as part of the frame pacing initiative. Game thread code should be written to not require render thread fences for correctness / threadsafety.
#jira UE-223692
#rb zach.bethel
[CL 36758377 by luke thatcher in 5.5 branch]
#rnx
[FYI] Charles.deRousiers
#tests Add a test for CPU lightmass with Substrate (opaque, unlit, translucent, emissive, legacy and slabs with colored transmittance)
[CL 36758335 by sebastien hillaire in 5.5 branch]
* Allows paths copied like this to be pasted into Perforce directly, in order to expand the Depot or Workspace trees
#rb Dominic.Couture, JeanFrancois.Dube
[CL 36757611 by sebastien lussier in 5.5 branch]
* Also switch to SComboBox<FName> instead of SComboBox<TSharedPtr<FName>>
#jira UE-217994
#rb Thomas.Sarkanen
[CL 36757409 by lucas dower in 5.5 branch]
Also fix double-entry in the context menu coming from parent/child classes
#jira UE-219402
#rb Jurre.deBaare
[RN] Fix crash in 'call function from anim graph' node BP menu population
[CL 36757252 by thomas sarkanen in 5.5 branch]
Adds support for "indirect" references, addressing the missing referencing properties.
There are still cases in which we cannot access properties, or their details. In that case, this CL adds a simple node stating that currently that info cannot be retrieved.
Note: there are plans to address some limitations in 5.6, also targeting the support for specific types, e.g. showing Material Parameters, Functions and Texture Samplers name instead of their "raw" C++ property name.
#jira UE-222011
#rb aditya.ravichandran
[CL 36757234 by dario mazzanti in 5.5 branch]
* This fixes an issue where copy-pasting a shared transition rule will not display the graph in the 'My Blueprint' tab and will result in operations such as Find References not taking into account all graphs.
#jira UE-215302
#rb Thomas.Sarkanen
[CL 36757223 by lucas dower in 5.5 branch]
Replaced with SetAnimInstanceClass. Fixed up deprecation warnings and places thast were callling both (!)
PR #11100: Deprecatead the 'SetAnimClass' function. Redundant call to 'SetAnimIn?
#jira UE-200548
#rb Nicholas.Frechette
[RN] Deprecated USkeletalMeshComponent::SetAnimClass function
[CL 36757163 by TheLumbee in 5.5 branch]
#rb Matt.Peters
#jira UE-208080
#rnx
- When FBulkDataRegistryImpl::Teardown is run we were cancelling FUpdatePayloadWorker by signalling to them that they should abandon their work, but we did not make sure that all of the jobs completed.
- In most cases this was fine but FUpdatePayloadWorker::DoWork will attempt to load editor bulkdata payloads first, before checking for cancellation and even if we did check first, it would be possible for currently running jobs to be mid loading of the payload when Teardown is run. In either of these scenarios we could potentially end up attempting to access the asset registry which would be cleaned up shortly after Teardown is called and end up asserting.
- To fix this we now check if the work should be abandoned before we start trying to read the bulkdata payload, then record how many jobs are currently reading payloads.
- In ::Teardown we can then check how many jobs are currently reading payloads and wait until that value reaches zero before returning.
- Note that this issue will only really show up on machines with very low worker core limits, when the editor is closed very shortly after loading and the project being loaded has manypackages in older pre 5.0 formats.
[CL 36757097 by paul chipchase in 5.5 branch]
