aboutsummaryrefslogtreecommitdiff
path: root/llvm/lib/Target/AMDGPU/FLATInstructions.td
AgeCommit message (Expand)AuthorFilesLines
2024-08-22AMDGPU: Remove global/flat atomic fadd intrinics (#97051)Matt Arsenault1-13/+0
2024-08-21AMDGPU: Remove flat/global atomic fadd v2bf16 intrinsics (#97050)Matt Arsenault1-2/+0
2024-07-25[LLVM][AMDGPU] Copy isConvergent from Pseudo to Real instructions (#99658)Acim Maravic1-0/+2
2024-07-18AMDGPU: Loop over the types for global_load_tr16 pats (NFC) (#99551)Changpeng Fang1-6/+4
2024-07-18AMDGPU: Add back half and bfloat support for global_load_tr16 pats (#99540)Changpeng Fang1-0/+4
2024-07-10AMDGPU: Add subtarget feature for memory atomic fadd f64 (#96444)Matt Arsenault1-3/+3
2024-06-27AMDGPU: Fix no return atomicrmw fadd v2f16 selection for gfx908 (#96948)Matt Arsenault1-0/+1
2024-06-23AMDGPU: Start selecting flat/global atomicrmw fmin/fmax. (#95592)Matt Arsenault1-31/+45
2024-06-21[AMDGPU][GFX12] Add support for new block ls instructions (#96273)Mariusz Sikora1-0/+14
2024-06-18AMDGPU: Flat instructions do not have signed offsets gfx7-gfx11 (#95852)Matt Arsenault1-38/+22
2024-06-18AMDGPU: Create pseudo to real mapping for flat/buffer atomic fmin/fmax (#95591)Matt Arsenault1-54/+53
2024-06-17Reapply "AMDGPU: Handle legal v2f16/v2bf16 atomicrmw fadd for global/flat (#9...Matt Arsenault1-1/+5
2024-06-15Revert "AMDGPU: Handle legal v2f16/v2bf16 atomicrmw fadd for global/flat (#95...Nico Weber1-5/+1
2024-06-15AMDGPU: Handle legal v2f16/v2bf16 atomicrmw fadd for global/flat (#95394)Matt Arsenault1-1/+5
2024-06-13DAG: Replace bitwidth with type in suffix in atomic tablegen ops (#94845)Matt Arsenault1-6/+6
2024-06-05[AMDGPU] Move flat patterns instantiation down. NFC. (#94409)Stanislav Mekhanoshin1-112/+111
2024-06-04[AMDGPU] Do not override PseudoInstr in FLAT Pseudo definitions. NFC. (#94369)Jay Foad1-17/+4
2024-05-31[AMDGPU] Copy Defs and Uses from Pseudo to Real Instructions (#93004)Fabian Ritter1-0/+4
2024-05-09[AMDGPU] Create AMDGPUMnemonicAlias tablegen class (#89288)Joe Nash1-5/+9
2024-04-03[AMDGPU] Remove useless aliases for FLAT instructions. NFC. (#87462)Jay Foad1-2/+2
2024-03-25AMDGPU: Rename intrinsics and remove f16/bf16 versions for load transpose (#8...Changpeng Fang1-8/+4
2024-03-20[AMDGPU] Simplify definition of GLOBAL_LOAD_TR Real instructionsJay Foad1-4/+8
2024-03-20[AMDGPU] Remove FLAT_Real_AllAddr_gfx11 in favor of GLOBAL_Real_AllAddr_gfx11Jay Foad1-38/+28
2024-03-20[AMDGPU] Simplify GFX11/GFX12 FLAT instruction definitions. NFC. (#85819)Jay Foad1-368/+363
2024-03-18[AMDGPU] Remove unneeded addr mode predicates on FLAT Real instructions (#85641)Jay Foad1-15/+5
2024-03-17[AMDGPU] Fix predicates on FLAT scratch ST/SVS mode Pseudos (#85442)Jay Foad1-77/+69
2024-03-15[AMDGPU] Simplify definition of FLAT segment bits. NFC.Jay Foad1-4/+2
2024-03-15[AMDGPU] Clean up GFX10 FLAT saddr field definitionJay Foad1-2/+3
2024-03-15[AMDGPU] Simplify GFX11 and GFX12 FLAT saddr field definitionJay Foad1-15/+6
2024-03-15[AMDGPU] Simplify GFX10+ FLAT saddr field definitionJay Foad1-2/+2
2024-03-15[AMDGPU] Reduce duplication in FLAT atomic definitions (#85383)Jay Foad1-70/+73
2024-03-01AMDGPU: Simplify instruction definitions for global_load_tr_b64(b128) (#83601)Changpeng Fang1-2/+2
2024-03-01[AMDGPU] Remove AtomicNoRet class and getAtomicNoRetOp table (#83593)Jay Foad1-12/+6
2024-02-28AMDGPU: Copy a few Predicates from Pseudo to Real (#83365)Changpeng Fang1-0/+1
2024-01-31AMDGPU/NFC: Add predicate for supporting buffer/flat/global f64 atomics (#80209)Konstantin Zhuravlyov1-3/+3
2024-01-19[AMDGPU] Misc formatting fixes. NFC.Jay Foad1-4/+4
2024-01-18[AMDGPU] Add global_load_tr for GFX12 (#77772)Piotr Sobczak1-0/+35
2024-01-18[AMDGPU][GFX12] Add 16 bit atomic fadd instructions (#75917)Mariusz Sikora1-0/+4
2024-01-17[AMDGPU][GFX12] Add Atomic cond_sub_u32 (#76224)Mariusz Sikora1-6/+44
2024-01-16[AMDGPU] Remove VT helpers isFloatType, isPackedType, simplify isIntType (#77...Stanislav Mekhanoshin1-13/+8
2024-01-15[AMDGPU] Simplify GFX12 FLAT Real instruction definitions. NFC. (#78147)Jay Foad1-37/+42
2024-01-09[AMDGPU] Flip the default value of maybeAtomic. NFCI. (#75220)Jay Foad1-7/+0
2024-01-08[AMDGPU] Add new cache flushing instructions for GFX12 (#76944)Mirko Brkušanin1-1/+38
2024-01-02[AMDGPU] GFX12 global_atomic_ordered_add_b64 instruction and intrinsic (#76149)Jay Foad1-2/+9
2023-12-15[AMDGPU] CodeGen for GFX12 VFLAT, VSCRATCH and VGLOBAL instructions (#75493)Mirko Brkušanin1-11/+46
2023-12-14[AMDGPU][MC] Add GFX12 VFLAT, VSCRATCH and VGLOBAL encodings (#75193)Mirko Brkušanin1-2/+257
2023-10-25[AMDGPU][NFCI] Decouple actual register encodings from HWEncoding values. (#6...Ivan Kosarev1-4/+4
2023-10-23[AMDGPU] Remove unnecessary conditionality on atomic CSUB pseudo-ops (#69914)Stephen Thomas1-5/+1
2023-10-11[AMDGPU] Add encoding/decoding support for non-result-returning ATOMIC_CSUB i...Stephen Thomas1-5/+13
2023-08-31SelectionDAG: Swap operands of atomic_storeMatt Arsenault1-36/+12
generated by cgit v1.1 at 2024-09-17 17:14:01 +0000