shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/llvm/lib/Target/AMDGPU
History
Austin Kerbow 89503bda38
[AMDGPU] Add structural stall heuristic to scheduling strategies (#169617) 
Implements a structural stall heuristic that considers both resource
hazards and latency constraints when selecting instructions. In coexec,
this changes the pending queue from a binary “not ready to issue”
distinction into part of a unified candidate comparison. Pending
instructions still identify structural stalls in the current cycle, but
they are now evaluated directly against available instructions by stall
cost, making the heuristics both more intuitive and more expressive.

- Add getStructuralStallCycles() to GCNSchedStrategy that computes the
number of cycles an instruction must wait due to:
  - Resource conflicts on unbuffered resources (from the SchedModel)
  - Sequence-dependent hazards (from GCNHazardRecognizer)

- Add getHazardWaitStates() to GCNHazardRecognizer that returns the
number
of wait states until all hazards for an instruction are resolved,
providing cycle-accurate hazard information for scheduling heuristics.
2026-03-23 11:33:43 -07:00
..
AsmParser
[AMDGPU] Allow bank conflicts on src0 for V_DUAL_MOV_B32 for gfx1170 (#186100)
2026-03-12 13:49:52 +01:00
Disassembler
[AMDGPU] Add GFX11_7Insts feature, eliminate isGFX1170 helpers. NFC (#185878)
2026-03-11 17:05:18 +01:00
MCA
Reapply "[MC][TableGen] Expand Opcode field of MCInstrDesc" (#180321) (#180954)
2026-02-12 09:17:02 +01:00
MCTargetDesc
[AMDGPU] Remove unused forward declaration of GCNSubtarget (#187695)
2026-03-20 13:26:57 +00:00
TargetInfo
[llvm] annotate interfaces in llvm/Target for DLL export (#143615)
2025-06-17 13:28:45 -07:00
Utils
[AMDGPU] Introduce ASYNC_CNT on GFX1250 (#185810)
2026-03-20 09:52:05 +00:00
AMDGPU.h
[AMDGPU] Cgscc amdgpu attributor boilerplate NFC (#179719)
2026-03-09 15:14:03 -04:00
AMDGPU.td
[AMDGPU] Update features for gfx1170 (#186107)
2026-03-20 17:04:17 +01:00
AMDGPUAliasAnalysis.cpp
…
AMDGPUAliasAnalysis.h
…
AMDGPUAlwaysInlinePass.cpp
[AMDGPU] Make use of getFunction and getMF. NFC. (#167872)
2025-11-14 11:00:57 +00:00
AMDGPUAnnotateUniformValues.cpp
[CodeGen] Drop uses of BranchInst (#186391)
2026-03-13 21:51:38 +00:00
AMDGPUArgumentUsageInfo.cpp
[AMDGPU] Remove AMDGPUArgumentUsageInfo pass (#182490)
2026-02-23 18:47:01 +05:30
AMDGPUArgumentUsageInfo.h
[AMDGPU] Remove AMDGPUArgumentUsageInfo pass (#182490)
2026-02-23 18:47:01 +05:30
AMDGPUAsanInstrumentation.cpp
[llvm] Use ConstantInt::getAllOnesValue()
2025-12-09 12:02:39 +01:00
AMDGPUAsanInstrumentation.h
…
AMDGPUAsmPrinter.cpp
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
AMDGPUAsmPrinter.h
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
AMDGPUAtomicOptimizer.cpp
[CodeGen] Drop uses of BranchInst (#186391)
2026-03-13 21:51:38 +00:00
AMDGPUAttributes.def
[AMDGPU][Attributor] Add AAAMDGPUClusterDims (#158076)
2025-09-15 15:04:33 -04:00
AMDGPUAttributor.cpp
[AMDGPU] Cgscc amdgpu attributor boilerplate NFC (#179719)
2026-03-09 15:14:03 -04:00
AMDGPUBarrierLatency.cpp
[AMDGPU] Pre-GFX10 does not need added latency for workgroup fences (#177157)
2026-01-27 10:24:05 +09:00
AMDGPUBarrierLatency.h
[AMDGPU] Ignore wavefront barrier latency during scheduling DAG mutation (#168500)
2025-11-19 17:49:14 +09:00
AMDGPUCallingConv.td
AMDGPU: Change ABI of 16-bit scalar values for gfx6/gfx7 (#175795)
2026-01-22 18:34:06 +00:00
AMDGPUCallLowering.cpp
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
AMDGPUCallLowering.h
[AMDGPU] Remove AMDGPUCallLowering dependency on AMDGPUTargetLowering. NFC. (#187008)
2026-03-17 12:38:25 +00:00
AMDGPUCodeGenPrepare.cpp
[AMDGPU] Standardize on using AMDGPU::getNullPointerValue. NFC. (#187037)
2026-03-17 17:08:16 +00:00
AMDGPUCoExecSchedStrategy.cpp
[AMDGPU] Add structural stall heuristic to scheduling strategies (#169617)
2026-03-23 11:33:43 -07:00
AMDGPUCoExecSchedStrategy.h
[AMDGPU] Add structural stall heuristic to scheduling strategies (#169617)
2026-03-23 11:33:43 -07:00
AMDGPUCombine.td
[AMDGPU][GlobalISel] Combine (or s64, zext(s32)) (#151519)
2025-10-24 17:25:00 +02:00
AMDGPUCombinerHelper.cpp
[AMDGPU] Remove NoSignedZerosFPMath uses (#178343)
2026-01-30 09:18:40 +08:00
AMDGPUCombinerHelper.h
[AMDGPU][GlobalISel] Combine for breaking s64 and/or into two s32 insts (#151731)
2025-08-20 17:32:29 +02:00
AMDGPUCtorDtorLowering.cpp
[AMDGPUCtorDtorLowering] Remove unnecessary pointer arithmetic (#182869)
2026-02-23 17:04:49 +01:00
AMDGPUCtorDtorLowering.h
…
AMDGPUExportClustering.cpp
[AMDGPU] Remove unused includes (NFC) (#141376)
2025-05-24 14:48:46 -07:00
AMDGPUExportClustering.h
…
AMDGPUExportKernelRuntimeHandles.cpp
…
AMDGPUExportKernelRuntimeHandles.h
…
AMDGPUFeatures.td
[AMDGPU] Use subtarget feature for flat offset bit width instead of arch checks (#183742)
2026-03-05 10:58:42 +01:00
AMDGPUFrameLowering.cpp
…
AMDGPUFrameLowering.h
…
AMDGPUGenRegisterBankInfo.def
…
AMDGPUGISel.td
AMDGPU: Improve codegen for VOP2 v_dot2c_f32_f16/bf16 (#179225)
2026-03-23 16:55:53 +01:00
AMDGPUGlobalISelDivergenceLowering.cpp
[AMDGPU] Use AMDGPULaneMaskUtils in SILowerI1Copies (#186170)
2026-03-17 10:19:15 +01:00
AMDGPUGlobalISelUtils.cpp
AMDGPU/GlobalISel: Add waterfall lowering in regbanklegalize (#145912)
2025-07-23 16:47:00 +02:00
AMDGPUGlobalISelUtils.h
AMDGPU/GlobalISel: Add waterfall lowering in regbanklegalize (#145912)
2025-07-23 16:47:00 +02:00
AMDGPUHazardLatency.cpp
[AMDGPU] Add scheduling DAG mutation for hazard latencies (#170075)
2026-02-03 11:10:28 +09:00
AMDGPUHazardLatency.h
[AMDGPU] Add scheduling DAG mutation for hazard latencies (#170075)
2026-02-03 11:10:28 +09:00
AMDGPUHSAMetadataStreamer.cpp
[AMDGPU] Make uniform-work-group-size a valueless attribute (#183925)
2026-03-01 21:29:55 +00:00
AMDGPUHSAMetadataStreamer.h
[Target] Add "override" where appropriate (NFC) (#165083)
2025-10-25 06:23:43 -07:00
AMDGPUIGroupLP.cpp
[AMDGPU] Use empty() instead of size() comparisons. NFC. (#187424)
2026-03-19 16:08:56 +00:00
AMDGPUIGroupLP.h
Remove unused <vector> inclusion (#166997)
2025-11-08 20:42:00 +00:00
AMDGPUImageIntrinsicOptimizer.cpp
…
AMDGPUInsertDelayAlu.cpp
[AMDGPU] Treat F64 TRANS instructions as VALU for S_DELAY_ALU insertion (#180940)
2026-02-11 19:14:54 +00:00
AMDGPUInstCombineIntrinsic.cpp
[AMDGPU] fold a call to implictarg.ptr to a poison with no-implicitarg-ptr (#186925)
2026-03-17 16:37:56 -04:00
AMDGPUInstrInfo.cpp
AMDGPU: Fix treating unknown mem operands as uniform (#170309)
2025-12-02 16:19:46 +00:00
AMDGPUInstrInfo.h
[AMDGPU] Remove unused forward declaration
2026-03-19 13:15:18 +00:00
AMDGPUInstrInfo.td
[AMDGPU][GFX12.5] Reimplement monitor load as an atomic operation (#177343)
2026-02-09 09:57:27 +01:00
AMDGPUInstructions.td
[AMDGPU] Fix and simplify patterns selecting fsub to v_fma_mix_f32 (#180169)
2026-02-06 14:39:13 +00:00
AMDGPUInstructionSelector.cpp
AMDGPU: Improve codegen for VOP2 v_dot2c_f32_f16/bf16 (#179225)
2026-03-23 16:55:53 +01:00
AMDGPUInstructionSelector.h
AMDGPU: Improve codegen for VOP2 v_dot2c_f32_f16/bf16 (#179225)
2026-03-23 16:55:53 +01:00
AMDGPUISelDAGToDAG.cpp
AMDGPU: Improve codegen for VOP2 v_dot2c_f32_f16/bf16 (#179225)
2026-03-23 16:55:53 +01:00
AMDGPUISelDAGToDAG.h
AMDGPU: Improve codegen for VOP2 v_dot2c_f32_f16/bf16 (#179225)
2026-03-23 16:55:53 +01:00
AMDGPUISelLowering.cpp
[AMDGPU] Saturate at i16 for f16 to i1/i8 conversion (#187467)
2026-03-20 13:15:40 +00:00
AMDGPUISelLowering.h
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
AMDGPULaneMaskUtils.h
[AMDGPU] Remove unused CmpLGOp instruction (#180195)
2026-02-27 10:37:44 +00:00
AMDGPULateCodeGenPrepare.cpp
[AMDGPU] LiveRegOptimizer: consider i8/i16 binops on SDWA (#155800)
2025-12-15 12:04:33 -05:00
AMDGPULegalizerInfo.cpp
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
AMDGPULegalizerInfo.h
AMDGPU/GlobalISel: Lower G_EXTRACT/INSERT in legalizer (#181036)
2026-03-12 10:03:44 -07:00
AMDGPULibCalls.cpp
AMDGPU: Fix selection failure on fast vector rootn (#185035)
2026-03-09 10:22:47 +00:00
AMDGPULibFunc.cpp
AMDGPU: Try to fix leak in AMDGPULibFunc (#182583)
2026-02-20 15:07:19 -08:00
AMDGPULibFunc.h
AMDGPU: Perform libcall recognition to replace fast OpenCL pow (#182135)
2026-02-19 11:49:32 +01:00
AMDGPULowerBufferFatPointers.cpp
[atomicrmw] fminimumnum/fmaximumnum support (#187030)
2026-03-18 09:35:49 +01:00
AMDGPULowerExecSync.cpp
[NFCI] replace getValueType with new getGlobalSize query (#177186)
2026-01-22 13:55:53 -05:00
AMDGPULowerIntrinsics.cpp
[AMDGPU] Add s_cluster_barrier on gfx1250 (#159175)
2025-09-16 14:49:48 -07:00
AMDGPULowerKernelArguments.cpp
Reapply "[AMDGPU] Propagate alias information in AMDGPULowerKernelArguments." (#174977)
2026-01-20 16:39:08 +00:00
AMDGPULowerKernelAttributes.cpp
AMDGPU: Annotate grid_dims ABI load with range metadata (#185610)
2026-03-16 23:31:32 +01:00
AMDGPULowerModuleLDSPass.cpp
[AMDGPU] Use empty() instead of size() comparisons. NFC. (#187424)
2026-03-19 16:08:56 +00:00
AMDGPULowerVGPREncoding.cpp
[AMDGPU] Simplify loop in AMDGPULowerVGPREncoding::handleCoissue. NFC. (#187511)
2026-03-19 15:36:37 +00:00
AMDGPULowerVGPREncoding.h
[AMDGPU] Register amdgpu-lower-vgpr-encoding pass in npm (#156971)
2025-09-05 00:07:47 -07:00
AMDGPUMachineFunctionInfo.cpp
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
AMDGPUMachineFunctionInfo.h
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
AMDGPUMachineModuleInfo.cpp
[AMDGPU][gfx1250] Support "cluster" syncscope (#157641)
2025-09-10 11:41:43 +02:00
AMDGPUMachineModuleInfo.h
[AMDGPU][gfx1250] Support "cluster" syncscope (#157641)
2025-09-10 11:41:43 +02:00
AMDGPUMacroFusion.cpp
…
AMDGPUMacroFusion.h
…
AMDGPUMarkLastScratchLoad.cpp
Revert "[NFC][LiveStacks] Use vectors instead of map and unordred_map" (#180421)
2026-02-08 16:54:51 +00:00
AMDGPUMCInstLower.cpp
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
AMDGPUMCInstLower.h
[AMDGPU] Standardize on using AMDGPU::getNullPointerValue. NFC. (#187037)
2026-03-17 17:08:16 +00:00
AMDGPUMCResourceInfo.cpp
[MC] Rename PrivateGlobalPrefix to InternalSymbolPrefix. NFC (#185164)
2026-03-10 01:03:27 -07:00
AMDGPUMCResourceInfo.h
[AMDGPU] upstream barrier count reporting part1 (#154409)
2025-08-19 16:42:31 -07:00
AMDGPUMemoryUtils.cpp
[NFCI] replace getValueType with new getGlobalSize query (#177186)
2026-01-22 13:55:53 -05:00
AMDGPUMemoryUtils.h
[AMDGPU] Remove isKernelLDS, add isKernel(const Function &). NFC. (#167300)
2025-11-25 15:43:18 +00:00
AMDGPUMIRFormatter.cpp
[AMDGPU] Introduce custom MIR formatting for s_wait_alu (#176316)
2026-01-31 10:46:59 -08:00
AMDGPUMIRFormatter.h
[AMDGPU] Introduce custom MIR formatting for s_wait_alu (#176316)
2026-01-31 10:46:59 -08:00
AMDGPUPassRegistry.def
[AMDGPU] Cgscc amdgpu attributor boilerplate NFC (#179719)
2026-03-09 15:14:03 -04:00
AMDGPUPerfHintAnalysis.cpp
[llvm] Remove unused includes of SmallSet.h (NFC) (#154893)
2025-08-22 10:33:46 -07:00
AMDGPUPerfHintAnalysis.h
[llvm] Use "= default" (NFC) (#166088)
2025-11-02 17:16:47 -08:00
AMDGPUPostLegalizerCombiner.cpp
Fix typos and spelling errors across codebase (#156270)
2026-01-13 11:52:46 -05:00
AMDGPUPredicateControl.td
…
AMDGPUPreLegalizerCombiner.cpp
[AMDGPU] Make use of getFunction and getMF. NFC. (#167872)
2025-11-14 11:00:57 +00:00
AMDGPUPreloadKernArgProlog.cpp
[AMDGPU] Enable kernarg preload on gfx1250 (#153686)
2025-08-14 16:29:53 -07:00
AMDGPUPreloadKernArgProlog.h
…
AMDGPUPreloadKernelArguments.cpp
[AMDGPU] Make use of getFunction and getMF. NFC. (#167872)
2025-11-14 11:00:57 +00:00
AMDGPUPrepareAGPRAlloc.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
AMDGPUPrepareAGPRAlloc.h
AMDGPU: Add pass to replace constant materialize with AV pseudos (#149292)
2025-07-18 17:15:38 +09:00
AMDGPUPrintfRuntimeBinding.cpp
AMDGPU: Stop checking for r600 in printf pass (#183536)
2026-02-26 17:18:36 +01:00
AMDGPUPromoteAlloca.cpp
[AMDGPU] Assert non-array alloca does have a size (#183834)
2026-02-28 10:32:36 -05:00
AMDGPUPromoteKernelArguments.cpp
…
AMDGPUPTNote.h
…
AMDGPURegBankCombiner.cpp
[AMDGPU] Fix legacy index in fmed3 optimization (#177426)
2026-01-28 10:22:22 +00:00
AMDGPURegBankLegalize.cpp
AMDGPU/GlobalISel: Regbanklegalize rules for G_PHI (#179735)
2026-03-06 14:01:50 +01:00
AMDGPURegBankLegalizeHelper.cpp
AMDGPU/GlobalISel: RegBankLegalize rules for s_sleep_var, s_prefetch (#187382)
2026-03-19 11:12:29 -07:00
AMDGPURegBankLegalizeHelper.h
AMDGPU/GlobalISel: RegBankLegalize rules for ctlz/cttz_zero_undef (#186546)
2026-03-16 10:21:47 -07:00
AMDGPURegBankLegalizeRules.cpp
AMDGPU/GlobalISel: RegBankLegalize rules for live_mask (#187833)
2026-03-23 11:01:46 -07:00
AMDGPURegBankLegalizeRules.h
AMDGPU/GlobalISel: RegBankLegalize rules for s_sleep_var, s_prefetch (#187382)
2026-03-19 11:12:29 -07:00
AMDGPURegBankSelect.cpp
[NFC][CodeGen] Adopt MachineFunctionProperties convenience accessors (#141101)
2025-05-23 08:30:29 -07:00
AMDGPURegisterBankInfo.cpp
[AMDGPU] Codegen for min/max instructions for gfx1170 (#185625)
2026-03-12 12:32:56 +01:00
AMDGPURegisterBankInfo.h
…
AMDGPURegisterBanks.td
AMDGPU: Start using RegClassByHwMode for wavesize operands
2025-11-11 15:07:59 -08:00
AMDGPURemoveIncompatibleFunctions.cpp
[AMDGPU] Add target features to guard DPP controls (#182391)
2026-02-20 07:59:10 -07:00
AMDGPURemoveIncompatibleFunctions.h
…
AMDGPUReserveWWMRegs.cpp
[AMDGPU] Remove unused includes (NFC) (#141376)
2025-05-24 14:48:46 -07:00
AMDGPUReserveWWMRegs.h
…
AMDGPUResourceUsageAnalysis.cpp
AMDGPU: Avoid crashing on statepoint-like pseudoinstructions (#170657)
2025-12-29 19:08:08 +01:00
AMDGPUResourceUsageAnalysis.h
[AMDGPU] upstream barrier count reporting part1 (#154409)
2025-08-19 16:42:31 -07:00
AMDGPURewriteAGPRCopyMFMA.cpp
Revert "[NFC][LiveStacks] Use vectors instead of map and unordred_map" (#180421)
2026-02-08 16:54:51 +00:00
AMDGPURewriteOutArguments.cpp
[AMDGPU] Make use of getFunction and getMF. NFC. (#167872)
2025-11-14 11:00:57 +00:00
AMDGPURewriteUndefForPHI.cpp
…
AMDGPUSearchableTables.td
AMDGPU: llvm.amdgcn.ds.add/sub.gs.reg.rtn are sources of divergence (#186883)
2026-03-17 08:30:27 -07:00
AMDGPUSelectionDAGInfo.cpp
[AMDGPU] TableGen-erate SDNode descriptions (#168248)
2025-11-17 00:10:07 +00:00
AMDGPUSelectionDAGInfo.h
[AMDGPU] TableGen-erate SDNode descriptions (#168248)
2025-11-17 00:10:07 +00:00
AMDGPUSetWavePriority.cpp
[llvm] Replace SmallSet with SmallPtrSet (NFC) (#154068)
2025-08-18 07:01:29 -07:00
AMDGPUSplitModule.cpp
[AMDGPU] Fix alias handling in module splitting functionality (#187295)
2026-03-19 10:51:39 -05:00
AMDGPUSplitModule.h
…
AMDGPUSubtarget.cpp
[AMDGPU] Add dereferenceable retAttr to a call to llvm.amdgcn.implicitarg.ptr (#182206)
2026-03-03 21:58:12 +01:00
AMDGPUSubtarget.h
[AMDGPU] Use subtarget feature for flat offset bit width instead of arch checks (#183742)
2026-03-05 10:58:42 +01:00
AMDGPUSwLowerLDS.cpp
[AMDGPU] Avoid unnecessary zero-index GEPs
2026-02-13 15:10:42 +01:00
AMDGPUTargetMachine.cpp
[AMDGPU] Add ML-oriented coexec scheduler selection and queue handling (#169616)
2026-03-23 09:30:01 -07:00
AMDGPUTargetMachine.h
[AMDGPU] Add ML-oriented coexec scheduler selection and queue handling (#169616)
2026-03-23 09:30:01 -07:00
AMDGPUTargetObjectFile.cpp
…
AMDGPUTargetObjectFile.h
…
AMDGPUTargetTransformInfo.cpp
[AMDGPU] Implement LSR cost model for GFX9+ (#184138)
2026-03-23 12:18:11 +01:00
AMDGPUTargetTransformInfo.h
[AMDGPU] Implement LSR cost model for GFX9+ (#184138)
2026-03-23 12:18:11 +01:00
AMDGPUUniformIntrinsicCombine.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
AMDGPUUnifyDivergentExitNodes.cpp
[CodeGen] Drop uses of BranchInst (#186391)
2026-03-13 21:51:38 +00:00
AMDGPUUnifyDivergentExitNodes.h
…
AMDGPUWaitSGPRHazards.cpp
[AMDGPU] Introduce custom MIR formatting for s_wait_alu (#176316)
2026-01-31 10:46:59 -08:00
AMDGPUWaitSGPRHazards.h
…
AMDKernelCodeT.h
…
BUFInstructions.td
[AMDGPU] Asynchronous loads from global/buffer to LDS on pre-GFX12 (#180466)
2026-02-11 05:26:58 +00:00
CaymanInstructions.td
…
CMakeLists.txt
[AMDGPU] Add ML-oriented coexec scheduler selection and queue handling (#169616)
2026-03-23 09:30:01 -07:00
DSDIRInstructions.td
[AMDGPU] Add VDSDIR encoding to gfx13 (#181620)
2026-02-17 09:25:14 +01:00
DSInstructions.td
Add VDS encoding for gfx13 (#187693)
2026-03-20 20:31:29 +01:00
EvergreenInstructions.td
[AMDGPU] Make rotr illegal (#166558)
2025-12-05 12:39:50 +00:00
EXPInstructions.td
[AMDGPU] Add VEXPORT encoding for gfx13 (#181788)
2026-02-19 08:46:50 +01:00
FLATInstructions.td
[AMDGPU] Add VFLAT, VGLOBAL, VSCRATCH to gfx13 (#182811)
2026-03-05 14:24:44 +01:00
GCNCreateVOPD.cpp
[AMDGPU] Fix VOPD checks for commuting OpX and OpY (#178772)
2026-01-30 15:13:33 +00:00
GCNDPPCombine.cpp
[NFC][MI] Tidy Up RegState enum use (2/2) (#177090)
2026-01-23 00:19:03 -08:00
GCNDPPCombine.h
[NFC][CodeGen] Adopt MachineFunctionProperties convenience accessors (#141101)
2025-05-23 08:30:29 -07:00
GCNHazardRecognizer.cpp
[AMDGPU] Add structural stall heuristic to scheduling strategies (#169617)
2026-03-23 11:33:43 -07:00
GCNHazardRecognizer.h
[AMDGPU] Add structural stall heuristic to scheduling strategies (#169617)
2026-03-23 11:33:43 -07:00
GCNILPSched.cpp
…
GCNIterativeScheduler.cpp
[AMDGPU] Iterative scheduling must behave the same with/without debug (#178460)
2026-02-02 09:15:48 -06:00
GCNIterativeScheduler.h
[AMDGPU] Teach iterative schedulers about IGLP (#134953)
2025-04-11 15:34:49 -07:00
GCNMinRegStrategy.cpp
…
GCNNSAReassign.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
GCNNSAReassign.h
…
GCNPreRALongBranchReg.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
GCNPreRALongBranchReg.h
…
GCNPreRAOptimizations.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
GCNPreRAOptimizations.h
…
GCNProcessors.td
[AMDGPU] Add gfx12-5-generic subtarget (#183381)
2026-02-25 13:34:48 -08:00
GCNRegPressure.cpp
[AMDGPU][Scheduler] Add GCNRegPressure-based methods to GCNRPTarget (#182853)
2026-02-27 15:21:45 +01:00
GCNRegPressure.h
[AMDGPU][Scheduler] Add GCNRegPressure-based methods to GCNRPTarget (#182853)
2026-02-27 15:21:45 +01:00
GCNRewritePartialRegUses.cpp
[NFC][CodeGen] Add helper function to check SubReg validity (#181489)
2026-02-19 12:39:11 -08:00
GCNRewritePartialRegUses.h
…
GCNSchedStrategy.cpp
[AMDGPU] Add structural stall heuristic to scheduling strategies (#169617)
2026-03-23 11:33:43 -07:00
GCNSchedStrategy.h
[AMDGPU] Add structural stall heuristic to scheduling strategies (#169617)
2026-03-23 11:33:43 -07:00
GCNSubtarget.cpp
[AMDGPU] Add ML-oriented coexec scheduler selection and queue handling (#169616)
2026-03-23 09:30:01 -07:00
GCNSubtarget.h
[AMDGPU] Update features for gfx1170 (#186107)
2026-03-20 17:04:17 +01:00
GCNVOPDUtils.cpp
[AMDGPU] Allow bank conflicts on src0 for V_DUAL_MOV_B32 for gfx1170 (#186100)
2026-03-12 13:49:52 +01:00
GCNVOPDUtils.h
[AMDGPU] VOPD/VOPD3 changes for gfx1250 (#147602)
2025-07-10 14:15:01 -07:00
MIMGInstructions.td
[AMDGPU] Add suffix _d4 to tensor load/store with 4 groups D#, NFC (#184176)
2026-03-03 14:10:38 -08:00
R600.h
…
R600.td
CodeGen: Make all targets override pseudos with pointers (#159881)
2025-11-26 14:57:14 +00:00
R600AsmPrinter.cpp
…
R600AsmPrinter.h
…
R600ClauseMergePass.cpp
…
R600ControlFlowFinalizer.cpp
[NFCI][AMDGPU] Use GET_SUBTARGETINFO_MACRO in GCNSubtarget.h and R600Subtarget.h (#177402)
2026-01-25 09:38:42 -05:00
R600Defines.h
…
R600EmitClauseMarkers.cpp
…
R600ExpandSpecialInstrs.cpp
[CodeGen] Remove FinalizeMachineBundles pass (#149806)
2025-07-23 11:36:49 +01:00
R600FrameLowering.cpp
…
R600FrameLowering.h
…
R600InstrFormats.td
…
R600InstrInfo.cpp
[AMDGPU] Make use of getFunction and getMF. NFC. (#167872)
2025-11-14 11:00:57 +00:00
R600InstrInfo.h
[TableGen] Return int32_t from InstrMapping table lookup functions. NFC. (#182079)
2026-02-18 18:49:58 +00:00
R600InstrInfo.td
…
R600Instructions.td
[AMDGPU] Use FPImmLeaf for float constants, fix build_vector patterns (#178018)
2026-01-28 10:51:42 +01:00
R600ISelDAGToDAG.cpp
[AMDGPU] Remove unused includes (NFC) (#141376)
2025-05-24 14:48:46 -07:00
R600ISelLowering.cpp
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
R600ISelLowering.h
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
R600MachineCFGStructurizer.cpp
[AMDGPU] Use empty() instead of size() comparisons. NFC. (#187424)
2026-03-19 16:08:56 +00:00
R600MachineFunctionInfo.cpp
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
R600MachineFunctionInfo.h
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
R600MachineScheduler.cpp
…
R600MachineScheduler.h
…
R600MCInstLower.cpp
[AMDGPU] Standardize on using AMDGPU::getNullPointerValue. NFC. (#187037)
2026-03-17 17:08:16 +00:00
R600OpenCLImageTypeLoweringPass.cpp
…
R600OptimizeVectorRegisters.cpp
[perf] Replace copy-assign by move-assign in llvm/lib/Target/AMDGPU/ (#179460)
2026-02-03 14:24:31 +00:00
R600Packetizer.cpp
[TableGen] Add mapping from processor ID to resource index for packetizer (#158182)
2025-09-17 09:12:37 +08:00
R600Processors.td
[NFCI][AMDGPU] Use GET_SUBTARGETINFO_MACRO in GCNSubtarget.h and R600Subtarget.h (#177402)
2026-01-25 09:38:42 -05:00
R600RegisterInfo.cpp
…
R600RegisterInfo.h
…
R600RegisterInfo.td
…
R600Schedule.td
…
R600Subtarget.cpp
[AMDGPU] Remove FeaturePromoteAlloca (#177636)
2026-02-23 15:24:57 -05:00
R600Subtarget.h
Reapply "AMDGPU: Annotate group size ABI loads with range metadata (#185420)" (#185588)
2026-03-10 07:41:26 +00:00
R600TargetMachine.cpp
[CodeGen][NewPM] Adjust pipeline for AsmPrinter
2026-02-23 17:23:25 -08:00
R600TargetMachine.h
[CodeGen][NewPM] Plumb MCContext through buildCodeGenPipeline
2026-02-23 17:21:32 -08:00
R600TargetTransformInfo.cpp
[IR] Split Br into UncondBr and CondBr (#184027)
2026-03-11 12:31:10 +00:00
R600TargetTransformInfo.h
[AMDGPU] Remove R600TargetTransformInfo dependency on AMDGPUTargetLowering. NFC. (#187014)
2026-03-17 13:40:49 +00:00
R700Instructions.td
…
SIAnnotateControlFlow.cpp
[CodeGen] Drop uses of BranchInst (#186391)
2026-03-13 21:51:38 +00:00
SIDefines.h
[AMDGPU][MC] Update old and add new min/max instructions for gfx1170 (#184601)
2026-03-09 18:38:45 +01:00
SIFixSGPRCopies.cpp
[MIR][MachineInstr] Update MachineInstr::eraseFromParent() to return an iterator (#179787)
2026-03-13 13:39:13 -07:00
SIFixSGPRCopies.h
…
SIFixVGPRCopies.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
SIFixVGPRCopies.h
…
SIFoldOperands.cpp
[AMDGPU] Clear no convergence flag on operand folding. NFCI (#179438)
2026-02-03 10:46:26 -08:00
SIFoldOperands.h
[NFC][CodeGen] Adopt MachineFunctionProperties convenience accessors (#141101)
2025-05-23 08:30:29 -07:00
SIFormMemoryClauses.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
SIFormMemoryClauses.h
…
SIFrameLowering.cpp
[AMDGPU] Make chain functions receive a stack pointer (#184616)
2026-03-06 11:01:42 +01:00
SIFrameLowering.h
[CodeGen] Replace (Min,Max)CSFrameIndex with flag on frame object [NFCI] (#170905)
2025-12-06 14:00:55 -08:00
SIInsertHardClauses.cpp
[AMDGPU] Adjust hard clause rules for gfx1250 (#152592)
2025-08-07 14:59:31 -07:00
SIInsertWaitcnts.cpp
[AMDGPU][SIInsertWaitcnts][NFC] SGPRInfo: Move score selection logic closer (#186518)
2026-03-20 16:30:52 -07:00
SIInstrFormats.td
CodeGen: Add RegisterClass by HwMode (#158269)
2025-09-19 20:08:51 +09:00
SIInstrInfo.cpp
[AMDGPU] Add basic verification for source modifiers (#186733)
2026-03-20 20:32:30 +01:00
SIInstrInfo.h
[AMDGPU] Handle S_WAIT_XCNT in SIInstrInfo::isWaitcnt (#187726)
2026-03-23 10:19:43 +00:00
SIInstrInfo.td
AMDGPU: Improve codegen for VOP2 v_dot2c_f32_f16/bf16 (#179225)
2026-03-23 16:55:53 +01:00
SIInstructions.td
[AMDGPU] DPP implementations for Wave Reduction (#185814)
2026-03-18 13:11:53 +05:30
SIISelLowering.cpp
[AMDGPU] DPP implementations for Wave Reduction (#187214)
2026-03-23 11:41:12 +05:30
SIISelLowering.h
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
SILateBranchLowering.cpp
[AMDGPU] Teach SILateBranchLowering pass to preserve MachineLoopInfo (#178276)
2026-01-29 22:23:00 +05:30
SILoadStoreOptimizer.cpp
[AMDGPU] Disable negative imm offset for async load/store instructions (#185078)
2026-03-06 21:55:39 +00:00
SILoadStoreOptimizer.h
[NFC][CodeGen] Adopt MachineFunctionProperties convenience accessors (#141101)
2025-05-23 08:30:29 -07:00
SILowerControlFlow.cpp
[AMDGPU] Fix DomTree preservation in SILowerControlFlow when nodes are removed (#176691)
2026-01-25 09:45:44 +05:30
SILowerControlFlow.h
…
SILowerI1Copies.cpp
[AMDGPU] Use AMDGPULaneMaskUtils in SILowerI1Copies (#186170)
2026-03-17 10:19:15 +01:00
SILowerI1Copies.h
[AMDGPU] Use AMDGPULaneMaskUtils in SILowerI1Copies (#186170)
2026-03-17 10:19:15 +01:00
SILowerSGPRSpills.cpp
[AMDGPU] Fix typos "SPGR" / "VPGR" in comments
2026-03-02 11:48:38 +00:00
SILowerSGPRSpills.h
[NFC][CodeGen] Adopt MachineFunctionProperties convenience accessors (#141101)
2025-05-23 08:30:29 -07:00
SILowerWWMCopies.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
SILowerWWMCopies.h
…
SIMachineFunctionInfo.cpp
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
SIMachineFunctionInfo.h
[AMDGPU] Rename AMDGPUMachineFunction to AMDGPUMachineFunctionInfo. NFC. (#187276)
2026-03-18 20:29:47 +00:00
SIMachineScheduler.cpp
[AMDGPU] Use empty() instead of size() comparisons. NFC. (#187424)
2026-03-19 16:08:56 +00:00
SIMachineScheduler.h
[CodeGen] Use VirtRegOrUnit where appropriate (NFCI) (#167730)
2025-11-13 10:26:58 +00:00
SIMemoryLegalizer.cpp
[MIR][MachineInstr] Update MachineInstr::eraseFromParent() to return an iterator (#179787)
2026-03-13 13:39:13 -07:00
SIModeRegister.cpp
…
SIModeRegisterDefaults.cpp
[AMDGPU] Remove DX10_CLAMP and IEEE bits from gfx1170 (#182107)
2026-03-04 12:16:41 +01:00
SIModeRegisterDefaults.h
…
SIOptimizeExecMasking.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
SIOptimizeExecMasking.h
…
SIOptimizeExecMaskingPreRA.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
SIOptimizeExecMaskingPreRA.h
…
SIOptimizeVGPRLiveRange.cpp
[LLVM][CodeGen] Remove pass initialization calls from pass constructors (#173061)
2026-01-21 08:44:51 -08:00
SIOptimizeVGPRLiveRange.h
[NFC][CodeGen] Adopt MachineFunctionProperties convenience accessors (#141101)
2025-05-23 08:30:29 -07:00
SIPeepholeSDWA.cpp
Reland "[NFC][MI] Tidy Up RegState enum use (1/2)" (#176277)
2026-01-16 13:05:06 -08:00
SIPeepholeSDWA.h
…
SIPostRABundler.cpp
[NFC][AMDGPU] Remove pass initialization from pass constructors (#180149)
2026-02-06 09:05:14 -08:00
SIPostRABundler.h
…
SIPreAllocateWWMRegs.cpp
[regalloc][LiveRegMatrix][AMDGPU] Fix LiveInterval dangling pointers in LiveRegMatrix. (#168556)
2026-01-27 16:08:58 +01:00
SIPreAllocateWWMRegs.h
…
SIPreEmitPeephole.cpp
[CodeGen] Use separate MBB number for analyses (#187086)
2026-03-18 07:35:36 +01:00
SIProgramInfo.cpp
[AMDGPU] Remove DX10_CLAMP and IEEE bits from gfx1170 (#182107)
2026-03-04 12:16:41 +01:00
SIProgramInfo.h
[AMDGPU] upstream barrier count reporting part1 (#154409)
2025-08-19 16:42:31 -07:00
SIRegisterInfo.cpp
[AMDGPU] Remove alignment constraint from spill pseudos (#177317)
2026-03-09 09:33:23 +05:30
SIRegisterInfo.h
[RegAlloc] Change the computation of CSRCost (#177226)
2026-02-04 10:08:57 -08:00
SIRegisterInfo.td
[AMDGPU] Remove unused regclass SGPR_HI16. NFCI. (#186431)
2026-03-13 21:38:10 +00:00
SISchedule.td
[AMDGPU] Update gfx1250 sched model to latest table (#182923)
2026-02-23 13:25:07 -08:00
SIShrinkInstructions.cpp
Revert "[AMDGPU] Generate more swaps" (#187723)
2026-03-20 12:03:20 -05:00
SIShrinkInstructions.h
…
SIWholeQuadMode.cpp
[AMDGPU] Fix crash in SIWholeQuadMode with debug instructions. (#178282)
2026-01-30 02:11:07 +01:00
SIWholeQuadMode.h
[NFC][CodeGen] Adopt MachineFunctionProperties convenience accessors (#141101)
2025-05-23 08:30:29 -07:00
SMInstructions.td
[AMDGPU] Add FeatureGFX13 and SMEM encoding for gfx13 (#177567)
2026-01-26 14:16:36 +01:00
SOPInstructions.td
[AMDGPU] Use s_cvt_i32/u32_f32 instructions for saturated uniform conversions (#187711)
2026-03-23 10:56:44 +00:00
VIInstrFormats.td
…
VINTERPInstructions.td
[AMDGPU] Add VINTERP encoding to gfx13 (#182481)
2026-03-02 10:50:08 +01:00
VOP1Instructions.td
[AMDGPU] Remove _e32 patterns for i32 saturated conversions (#187715)
2026-03-20 15:41:13 +00:00
VOP2Instructions.td
AMDGPU: Improve codegen for VOP2 v_dot2c_f32_f16/bf16 (#179225)
2026-03-23 16:55:53 +01:00
VOP3Instructions.td
[AMDGPU] Add GFX11_7Insts feature, eliminate isGFX1170 helpers. NFC (#185878)
2026-03-11 17:05:18 +01:00
VOP3PInstructions.td
AMDGPU: Fix src2_modifiers for v_dot2_f32_f16/bf16 (#179224)
2026-03-23 15:44:09 +01:00
VOPCInstructions.td
[AMDGPU] Mark VOPC_e64 instructions rematerializable (#186195)
2026-03-13 21:16:18 +00:00
VOPDInstructions.td
[AMDGPU][MC] Update old and add new min/max instructions for gfx1170 (#184601)
2026-03-09 18:38:45 +01:00
VOPInstructions.td
AMDGPU: Fix src2_modifiers for v_dot2_f32_f16/bf16 (#179224)
2026-03-23 15:44:09 +01:00