Platforms that do support uniform buffer objects can now provide batched primitive data through UBO. There is a limit UBO range that can be accessed in shaders, so we group instances into batches that fit into this limit. Switch uses 64KB views, other platfroms16KB views. For each primitive we allocate 512Bytes and 256Bytes for instance. Mobile platforms that do not support UBO will use a desktop version of GPUScene.
There are a few things that still missing: Dynamic mesh passes, static lighting
#rb ola.olsson, benjamin.rouveyrol
[CL 26354848 by marc audy in ue5-main branch]
- A minor QoL improvement for captures.
#rb Rob.Srinivasiah, Erica.Stella
#jira none
#review @Robert.Srinivasiah, @Erica.Stella
#preflight 642f0eb6b306e98c632026f3
[CL 24965490 by Arciel Rekman in ue5-main branch]
- Fixed GPU scene resources not being available for the base pass. The external access queue is flushed only on the graphics pipe for the base pass and then later for both graphics | async compute. This is necessary to avoid
having to immediately transition back from Graphics | AsyncCompute when updating shadow GPU scene buffers, which introduces an immediate stall on any in-flight async compute work on certain platforms.
- Fixed issue where RHI transition validation log was being hashed including the stack breadcrumb, which was causing log spam.
- Merged reflection capture / atmosphere external access queues together with the main scene rendering one to reduce passes.
- Fixed incorrect RHI validation assert in hardware ray tracing passes.
- Fixed IES texture atlas assert due to not using external access mode on the texture.
#rb christopher.waters
#preflight 6387b1ba766a2aeed29463e0
#jira UE-170797
[CL 23339379 by zach bethel in ue5-main branch]
*** This change will incur a full shader invalidation across all platforms ***
Issues:
- Some platforms require async compute dispatch indirect arguments to not cross specific memory boundaries
- This places restrictions on the valid sizes for a dispatch indirect argument set. We were not conforming to these restrictions which could result in GPU crashes on these async passes
Fixes:
- FRHIDispatchIndirectParameters is padded out to meet per-platform memory boundary restrictions
- This is driven via new per-platform preprocessor define PLATFORM_DISPATCH_INDIRECT_ARGUMENT_BOUNDARY_SIZE
- Some platforms require FRHIDispatchIndirectParameters to align with their internal structure hence we cannot universally size to meet all platform's requirements
- Introduce new FRHIDispatchIndirectParametersNoPadding for uses when we explicitly do not want the padding and otherwise avoid the memory boundary restrictions
- Revise and expand indirect argument validation code to catch further such issues in the future
- Update shaders which write to dispatch indirect argument buffers to account for optional per-platform padding
- New utility function WriteDispatchIndirectArgs introduced to faciliate this
- platforms which require other than the default nonpadded dispatch indirect arguments must define DISPATCH_INDIRECT_UINT_COUNT and their own WriteDispatchIndirectArgs in their CommonPlatform.ush
- move creation of DispatchIndirectGraphicsCommandSignature command signature to be per-platform
- DispatchIndirectGraphicsCommandSignature and DispatchIndirectComputeCommandSignature stride changed to account for additional padding on impacted platforms
Testing:
- ran Lyra with and without async compute Lumen on impacted platforms as well as Win64
- ran FN replay on impacted platforms
#rb Krzysztof.Narkowicz, Ben.Woodhouse, Benjamin.Rouveyrol
#jira UE-167950
#preflight 6359563b2e6690262a11bc06
[CL 22862498 by eric mcdaniel in ue5-main branch]
- Specify source/destination offsets and number of bytes.
#rb christopher.waters
[FYI] zach.bethel
#preflight skip
[CL 22434285 by tiago costa in ue5-main branch]
This change includes significant refactor work performed in //UE5/Dev-ParallelRendering. A brief summary of the work is as follows:
Refactored RHI command lists
- Removal of the "immediate" async compute command list
- Introduced an "active pipe" on each command list, allowing RHICmdLists to record work for either graphics or async compute. Pipes can be selected using the SwitchPipeline() function, or the FRHICommandListScopedPipeline helper.
- New explicit command list submission RHI API (RHIFinalizeContext, RHISubmitCommandLists). The IRHICommandContextContainer type has been removed.
- Explicit GPU submission is automatically appended to the immediate command list when it is dispatched to the RHI thread.
Platform RHI implementations
- The new submission API has been implemented across all platforms. Some platforms required a significant refactor.
#rb Mihnea.Balta,Kenzo.Terelst
#jira UE-139550
#preflight 6332e3641003050806d802ef
[CL 22239063 by luke thatcher in ue5-main branch]
Changed iOS to use Memoryless flag to match other platforms when creating MSAA targets
#jira UE-162047
#rb Dmitriy.Dyomin, Jack.Porter
[FYI] Wei.Liu
#preflight 632c351c7b582f58ab2c8563
[CL 22144723 by carl lloyd in ue5-main branch]
Focus started on GenerateMips and AddClearUAVPass.
#jira none
#rb mihnea.balta
#preflight 62fe768d200ff87e07c5e1ad
[CL 21448320 by christopher waters in ue5-main branch]
Clean obsolete codes in mobile postprocess.
#jira UE-156188
#rb Dmitriy.Dyomin
#preflight 62b274f08e5ee14ea77b8d10
[CL 20769595 by Wei Liu in ue5-main branch]
- Adding support for bindless resources/samplers. The loose resources have to be wrapped in a macro that generates an index and proxy resource when bindless is enabled. Reflection has to parse these differently from other resources since they're parsed as uint parameters. The runtime also has to bind these differently since they end up needing to find their real resource in the binding data and then update a constant buffer.
- To assist in conversions, the shader compiler will detect if a shader is compiled with both bindless and non-bindless resources/samplers and will emit errors if that is the case.
- Adding support for bindless Uniform Buffer resources/samplers. Because Uniform Buffers aren't per-shader, they'll always update their constants with the bindless indices of their resources.
- Adding more flags to FShaderCodePackedResourceCounts required changing all the array initializations to individual parameters since not all the new flags are used everywhere.
- No shaders have been configured to support bindless resources/samplers. Yet.
#jira UE-139616
#rb zach.bethel, arciel.rekman
#preflight 6282b8ec44349a6581a21a39
[CL 20250348 by christopher waters in ue5-main branch]
- Added resource pool counters and events.
- Added AllocatePooledBuffer method and refactored pool to no longer take a command list.
- Refactored swap chain barrier logic to be a bit cleaner.
- Added helper methods to cast between views.
- Added power of two alignment option to buffer pool.
- Added GetTypeHash implementations for RDG SRV | UAV descriptors.
#preflight 62631046006fa20b683d130f
[CL 19873407 by zach bethel in ue5-main branch]
