summaryrefslogtreecommitdiff
path: root/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
AgeCommit message (Expand)Author
2025-06-26[AMDGPU] Enable vectorization of i8 values. (#134934)Gheorghe-Teodor Bercea
2025-06-21[CostModel] Add a DstTy to getShuffleCost (#141634)David Green
2025-06-18AMDGPU: Report special input intrinsics as free (#141948)Matt Arsenault
2025-06-18AMDGPU: Cost model for minimumnum/maximumnum (#141946)Matt Arsenault
2025-06-18AMDGPU: Move fpenvIEEEMode into TTI (#141945)Matt Arsenault
2025-06-18AMDGPU: Reduce cost of f64 copysign (#141944)Matt Arsenault
2025-06-18AMDGPU: Fix cost model for 16-bit operations on gfx8 (#141943)Matt Arsenault
2025-06-03[ValueTracking] Make Depth last default arg (NFC) (#142384)Ramkumar Ramachandra
2025-05-20[AMDGPU] Add make.buffer.rsrc to InferAddressSpaces (#140770)Krzysztof Drewniak
2025-05-19[AMDGPU] Add a new amdgcn.load.to.lds intrinsic (#137425)Krzysztof Drewniak
2025-05-01[CostModel] Make Op0 and Op1 const in getVectorInstrCost. NFC (#137631)David Green
2025-04-24[AMDGPU] Use variadic isa<>. NFC. (#137016)Jay Foad
2025-04-23[CostModel] Remove optional from InstructionCost::getValue() (#135596)David Green
2025-04-22[TTI] Fix discrepancies in prototypes between interface and implementations (...Sergei Barannikov
2025-04-22[TTI] Make all interface methods const (NFCI) (#136598)Sergei Barannikov
2025-04-21[TTI] Constify BasicTTIImplBase::thisT() (NFCI) (#136575)Sergei Barannikov
2025-02-04[AMDGPU] Remove special cases in TTI::getMemcpyLoop(Residual)LoweringType (#1...Fabian Ritter
2025-01-29[KernelInfo] Implement new LLVM IR pass for GPU code analysis (#102944)Joel E. Denny
2024-10-28[AMDGPU] Use wider loop lowering type for LowerMemIntrinsics (#112332)Fabian Ritter
2024-10-11[TTI][AMDGPU] Allow targets to adjust `LastCallToStaticBonus` via `getInlinin...Shilei Tian
2024-10-11[NFC] Rename `Intrinsic::getDeclaration` to `getOrInsertDeclaration` (#111752)Rahul Joshi
2024-10-11[AMDGPU] Enable unaligned scratch accesses (#110219)Fabian Ritter
2024-10-09[TTI] NFC: Port TLI.shouldSinkOperands to TTI (#110564)Jeffrey Byrnes
2024-10-09AMDGPU: Remove flat/global fmin/fmax intrinsics (#105642)Matt Arsenault
2024-10-03[AMDGPU] Qualify auto. NFC. (#110878)Jay Foad
2024-08-26[NFC] Remove dead codeLuke Drummond
2024-08-22AMDGPU: Remove global/flat atomic fadd intrinics (#97051)Matt Arsenault
2024-08-09AMDGPU: Correct costs of saturating add/sub intrinsics (#100808)Matt Arsenault
2024-08-09TTI: Check legalization cost of abs nodes (#100523)Matt Arsenault
2024-08-01AMDGPU: Improve cost handling of canonicalize (#101479)Matt Arsenault
2024-07-30AMDGPU: Enable vectorization of v2f16 copysign (#100799)Matt Arsenault
2024-07-30AMDGPU: Improve cost handling of fma/fmuladd (#100798)Matt Arsenault
2024-07-29[LowerMemIntrinsics][NFC] Use Align in TTI::getMemcpyLoopLoweringType (#100984)Fabian Ritter
2024-06-28[IR] Add getDataLayout() helpers to Function and GlobalValue (#96919)Nikita Popov
2024-06-27[IR] Add getDataLayout() helpers to BasicBlock and Instruction (#96902)Nikita Popov
2024-06-27AMDGPU: Remove ds_fmin/ds_fmax intrinsics (#96739)Matt Arsenault
2024-06-23AMDGPU: Remove ds atomic fadd intrinsics (#95396)Matt Arsenault
2024-05-21[AMDGPU] Vectorize more 16 bit shuffles (#90648)Jeffrey Byrnes
2024-04-09[AArch64] Add costs for ST3 and ST4 instructions, modelled as store(shuffle)....David Green
2024-02-12[TTI]Fallback to SingleSrcPermute shuffle kind, if no direct estimation forAlexey Bataev
2023-12-19GFX12: Add LoopDataPrefetchPass (#75625)Mariusz Sikora
2023-12-15[AMDGPU] - Add address space for strided buffers (#74471)Jessica Del
2023-12-15[AMDGPU] CodeGen for GFX12 VFLAT, VSCRATCH and VGLOBAL instructions (#75493)Mirko BrkuĊĦanin
2023-12-13[AMDGPU] Update IEEE and DX10_CLAMP for GFX12 (#75030)Piotr Sobczak
2023-12-11[llvm] Use StringRef::{starts,ends}_with (NFC) (#74956)Kazu Hirata
2023-11-22[llvm][TypeSize] Fix addition/subtraction in TypeSize. (#72979)Sander de Smalen
2023-10-27Replace TypeSize::{getFixed,getScalable} with canonical TypeSize::{Fixed,Scal...Fangrui Song
2023-10-15[AMDGPU] Fix cost of fast unsafe f32 fdiv (#68988)Jay Foad
2023-08-18[TTI]Add InsertSubvector pattern in improveShuffleKindFromMask().Alexey Bataev
2023-07-11AMDGPU: Partially fix not respecting dynamic denormal modeMatt Arsenault
generated by cgit v1.2.3 (git 2.25.1) at 2026-10-23 19:18:38 +0000