aboutsummaryrefslogtreecommitdiff
path: root/llvm/lib/Target/AMDGPU/BUFInstructions.td
AgeCommit message (Expand)AuthorFilesLines
6 days[AMDGPU] Support f64 atomics on gfx1250 (#151172)Changpeng Fang1-1/+11
14 days[AMDGPU] Remove some duplicated lines. NFC. (#128029)Jay Foad1-1/+0
2025-07-18[AMDGPU] Support nv memory instructions modifier on gfx1250 (#149582)Stanislav Mekhanoshin1-0/+1
2025-06-24AMDGPU: Remove Formatted MUBUF instructions from gfx1250 support (#145590)Changpeng Fang1-8/+18
2025-06-24AMDGPU: Remove MTBUF instructions from gfx1250 support (#145563)Changpeng Fang1-11/+16
2025-04-28[Targets] Migrate from atomic_load_8/16/32/64 to atomic_load_nonext_8/16/32/6...Craig Topper1-3/+3
2025-04-25[SelectionDAG][Targets] Replace atomic_load_8/atomic_load_16 with atomic_load...Craig Topper1-3/+4
2025-03-25[AMDGPU][GFX11] buffer_load_lds_{size} instructions do not exist (#132916)Juan Manuel Martinez Caamaño1-44/+0
2025-01-08[AMDGPU] Disallow null for more resource operands (#121941)Jay Foad1-1/+1
2025-01-03[AMDGPU][MC] Allow null where 128b or larger dst reg is expected (#115200)Jun Wang1-3/+3
2024-12-12[TableGen] Replace WantRoot/WantParent SDNode properties with flags (#119599)Sergei Barannikov1-2/+4
2024-11-25AMDGPU: Support buffer_atomic_pk_add_bf16 for gfx950 (#117599)Matt Arsenault1-0/+2
2024-11-20[AMDGPU] Remove hasPostISelHook for atomics. NFC. (#116791)Jay Foad1-1/+0
2024-11-18AMDGPU: Handle gfx950 96/128-bit buffer_load_lds (#116681)Matt Arsenault1-8/+16
2024-11-01[AMDGPU] Simplify GFX12 VBUFFER definitions. NFC. (#114403)Jay Foad1-5/+3
2024-10-31AMDGPU/GlobalISel: Handle atomic sextload and zextload (#111721)Matt Arsenault1-0/+5
2024-09-23[AMDGPU][MC] Instructions not to be supported in GFX940 (#109225)Jun Wang1-2/+3
2024-09-11[AMDGPU] Make use of multiclass inheritance. NFC.Jay Foad1-3/+2
2024-07-25[LLVM][AMDGPU] Copy isConvergent from Pseudo to Real instructions (#99658)Acim Maravic1-0/+3
2024-07-10AMDGPU: Add subtarget feature for memory atomic fadd f64 (#96444)Matt Arsenault1-4/+6
2024-06-25AMDGPU: Handle legal v2bf16 atomicrmw fadd for gfx12 (#95930)Matt Arsenault1-1/+6
2024-06-23AMDGPU: Start selecting flat/global atomicrmw fmin/fmax. (#95592)Matt Arsenault1-6/+20
2024-06-21AMDGPU: Start fixing inconsistencies in usage of SubtargetPredicate (#96337)Matt Arsenault1-37/+29
2024-06-21AMDGPU: Fix overriding SubtargetPredicate in MUBUF_Real_gfx90a (#96351)Matt Arsenault1-3/+3
2024-06-18AMDGPU: Create pseudo to real mapping for flat/buffer atomic fmin/fmax (#95591)Matt Arsenault1-50/+53
2024-06-17AMDGPU: Cleanup selection patterns for buffer loads (#95378)Matt Arsenault1-42/+30
2024-06-17AMDGPU: Remove .v2bf16 buffer atomic fadd intrinsics (#95783)Matt Arsenault1-1/+1
2024-06-14[AMDGPU] Change getLdStRegisterOperand to !cond for better diagnostic (#95475)Joe Nash1-6/+4
2024-06-13AMDGPU: Fix buffer intrinsic handling for various 16-bit elements. (#95376)Matt Arsenault1-0/+8
2024-06-13DAG: Replace bitwidth with type in suffix in atomic tablegen ops (#94845)Matt Arsenault1-2/+2
2024-06-13AMDGPU: Fix using wrong memory type for non-image resource intrinsics (#94911)Matt Arsenault1-0/+4
2024-06-13[AMDGPU][GFX12] Support disassembling MUBUF instructions with arbitrary FORMA...Ivan Kosarev1-2/+8
2024-06-12AMDGPU: Handle buffer load/store for 64-bit element typesMatt Arsenault1-0/+9
2024-05-31[AMDGPU] Copy Defs and Uses from Pseudo to Real Instructions (#93004)Fabian Ritter1-0/+6
2024-05-27[AMDGPU] Legalize and select raw/struct_buffer_load with tfe (#93310)Mirko Brkušanin1-0/+9
2024-05-09[AMDGPU] Create AMDGPUMnemonicAlias tablegen class (#89288)Joe Nash1-3/+9
2024-04-17[AMDGPU] Fix predicates for BUFFER_ATOMIC_FMIN/FMAX patterns (#89066)Jay Foad1-1/+1
2024-03-25[AMDGPU] Extend zero initialization of return values for TFE (#85759)David Stuttard1-2/+3
2024-03-13[AMDGPU] Remove unneeded MnemonicAlias. NFC.Jay Foad1-1/+0
2024-03-12[AMDGPU] Add missing GFX10 buffer format d16 hi instructions (#84809)Jay Foad1-3/+2
2024-03-08[AMDGPU] Simplify and refactor VBUFFER_Real class definitions. NFC. (#84521)Jay Foad1-17/+14
2024-03-08[AMDGPU] Use get_BUF_ps to default real_name of BUF instructions. NFC. (#84524)Jay Foad1-13/+18
2024-03-08[AMDGPU] Simplify renamed BUF instruction definitions. NFC. (#84503)Jay Foad1-125/+114
2024-03-07[AMDGPU] Make use of Mnem_gfx11_gfx12. NFC.Jay Foad1-1/+1
2024-03-05[AMDGPU] Remove unneeded BUF _impl multiclasses. NFC. (#84034)Jay Foad1-59/+45
2024-03-05[AMDGPU] Use BUF multiclasses to reduce repetition. NFC. (#84003)Jay Foad1-316/+289
2024-03-05[AMDGPU] Only use the BUF Base_ prefix for multiple architectures. NFC.Jay Foad1-6/+6
2024-03-05[AMDGPU] Remove Base_MUBUF_Real_Atomic_gfx11. NFC. (#83994)Jay Foad1-8/+3
2024-03-05[AMDGPU] Rename get_MUBUF_ps and use it for MTBUF too. NFC. (#83991)Jay Foad1-20/+17
2024-03-01[AMDGPU] Remove AtomicNoRet class and getAtomicNoRetOp table (#83593)Jay Foad1-59/+29
generated by cgit v1.1 at 2025-08-04 11:11:07 +0000