Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Promotion from AMD internal branch for 2024.Q2.1 #2899

Closed
wants to merge 3 commits into from
Closed

Conversation

qiaojbao
Copy link
Collaborator

@qiaojbao qiaojbao commented May 14, 2024

No description provided.

@qiaojbao qiaojbao requested a review from a team as a code owner May 14, 2024 09:39
[Continuations] Add helper commonMaterializable()
[Continuations] Add one extra SROA
[Continuations] Add PostProcessing test.
[Continuations] Check that resume functions report no stack usage
[Continuations] Cleanup includes.
[Continuations] Fix corruption seen in RaytracingReflections
[Continuations] Fix first payload argument dword calculation
[Continuations] Fix outgoing register count for kernelentry
[Continuations] Get DispatchSystemData type from GetLocalRootIndex.
[Continuations] Handle gpurt GetSetting
[Continuations] Ignore life time markers when doing store-to-load forwarding
[Continuations] Keep _AmdAcceptHitAttributes calls
[Continuations] Lower _GetResumePoint correctly
[Continuations] Move stackptr lowering
[Continuations] Promote _AmdAwait arguments correctly
[Continuations] Refactor RT shader stage enum
[Continuations] Rematerialize single source shuffle
[Continuations] Remove old unused helper
[Continuations] Remove use of [[no_unique_address]]
[Continuations] Set up proper register count for Traversal
[Continuations] Spill in PayloadCopyHelper
[Continuations] Support TriangleVertexPositions in continuations
[Continuations] Update test to reduce diff
[Continuations] Use int as name for gpurt settings
[Continuations] Use llvmraytracing in more places
[RT] Set the entry name as modulename
Add 'override' to SPIRVBaseVariable::validate()
Add AmdExtD3DShaderIntrinsics_ShaderMarker
Add DbgInfo api to ModuleBunch
Add detecting more patterns which can be lowered into fmul_legacy and fma_legacy
Add dynamicVgprBlockSize option
Add GpurtVersionFlags
Add Phoenix2 support
Add ShaderModuleUsage::usePrimitiveId
Apply WQM to result of WWM subgroup operations
Clean GEP formation for NGG atomic add
Cleanup the old PAL metadata code path
CrossModuleInliner: add check for consistent target module
CrossModuleInliner: make struct layout independent of NDEBUG
DataLayout index sizes for SPIRV address spaces
Detect read-only buffers and mark them as invariant
Expand GEP compatibility in tests
Handle new return type for some IRBuilder methods
Handle new return type from DIBuilder
lgc: add strided buffer load
lgc: correct stage detection for subgroup builder
lgc: missing code for load.strided.buffer.desc
lgc: Add LoadBufferAddrOp
lgc: Allow hoisting common instructions
lgc: Disassemble archive-of-ELFs
lgc: move CreateDebugBreak to BuilderCommon
lgc: Set flat workgroup size for cs_chain
LowerRayTracingPipeline: Avoid redundant GEPs
Move replaceAllPointerUses into CompilerUtils
Pack mesh shader outputs in LDS space
Reduce SW XFB LDS usage
Refine advanced blend func with value-type arguments
Refine dump info for shader module id
Refine pipeline dump when shader module identifier is used
Remove checks for gfxip >= 10
Remove LDS granularity rounding
Remove ShaderModuleUsage::useGenericBuiltIn
Remove the accidentally-added file for *.pipe.swp
Report new RequireFullPipeline error
Return frozen poison value in _AmdGetUninitialized
Return the state whether the texture is accessed by texelFetch function
Set MaxOutgoingVgprCount into SGPR and pass it
Simplify LDS GEP formation
Some renaming in mesh shader
Support uint64 AmdTraceRaySampleGpuTimer()
TargetFeatures for amdgpu_cs_chain_preserve funcs
Update llvm-dialects submodule
Use 32-bit GEPs consistently
Use CmpInst::Create instead of new to create instruction
Use convertUsersOfConstantsToInstruction throughout
VertexFetch: vertex offset is added twice in vertex offset mode
Fix check_llpc test regression
Fix color export offline compilation
Fix compiler warnings
Fix error on unused private field
Fix ShaderDbg issues
Fix the check for reserving a dummy vertex attribute
Fix unusual case imported by new game:STARFIELD
Fix updating threshold issue when merge with pipeline cache which cause PAL assert
@amdvlk-admin
Copy link
Collaborator

Test summary for commit fba3592

CTS tests (Failed: 0/148738)
  • Built with version 1.3.5.2.RC
  • Ubuntu navi3x, Srdcvk
    • Passed: 36314/74367 (48.8%)
    • Failed: 0/74367 (0.0%)
    • Not Supported: 38053/74367 (51.2%)
    • Warnings: 0/74367 (0.0%)
    Ubuntu navi2x, Srdcvk
    • Passed: 36343/74371 (48.9%)
    • Failed: 0/74371 (0.0%)
    • Not Supported: 38028/74371 (51.1%)
    • Warnings: 0/74371 (0.0%)

@amdvlk-admin
Copy link
Collaborator

Test summary for commit 3f33e94

CTS tests (Failed: 0/148738)
  • Built with version 1.3.5.2.RC
  • Ubuntu navi3x, Srdcvk
    • Passed: 36314/74367 (48.8%)
    • Failed: 0/74367 (0.0%)
    • Not Supported: 38053/74367 (51.2%)
    • Warnings: 0/74367 (0.0%)
    Ubuntu navi2x, Srdcvk
    • Passed: 36343/74371 (48.9%)
    • Failed: 0/74371 (0.0%)
    • Not Supported: 38028/74371 (51.1%)
    • Warnings: 0/74371 (0.0%)

@qiaojbao qiaojbao closed this May 15, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants