Commits · d9847cde4841140a95404ea7b7d3a57f8bfbf976 · Roger Ferrer / llvm-epi · GitLab

Jul 31, 2023
- [GlobalISel] convergent intrinsics · d9847cde
  Sameer Sahasrabuddhe authored Jul 31, 2023
  d9847cde
Jul 29, 2023
- Revert "[MachineScheduler] Track physical register dependencies per-regunit" · e2e3f068
  Jay Foad authored Jul 29, 2023
  e2e3f068
- [MachineScheduler] Track physical register dependencies per-regunit · 1a54671d
  Jay Foad authored Jul 28, 2023
  1a54671d
- [MachineScheduler] Test case for physical register dependencies · 5a64c89c
  Jay Foad authored Jul 28, 2023
  5a64c89c
Jul 28, 2023
- AMDGPU/GlobalISel: Set dead on scc on manually selected instructions · 3240ae70
  Matt Arsenault authored Jul 28, 2023
  3240ae70
- [AMDGPU] Allow 8,16 bit sources in calculateSrcByte · 391249d1
  Jeffrey Byrnes authored Jun 29, 2023
  391249d1
Jul 27, 2023
- AMDGPU: Always custom lower extract_subvector · 95e5a461
  Matt Arsenault authored Jul 27, 2023
  95e5a461
- Revert "[CodeGen]Allow targets to use target specific COPY instructions for live range splitting" · a496c8be
  Vitaly Buka authored Jul 26, 2023
  a496c8be
- [AMDGPU] Propagate constants for llvm.amdgcn.wave.reduce.umin/umax · 14620536
  Pravin Jagtap authored Jul 26, 2023
  14620536
Jul 26, 2023
- [AMDGPU] Fix PromoteAlloca Subvector Stores for Single Elements · a8aabba5
  pvanhout authored Jul 26, 2023
  a8aabba5
- [AMDGPU] Precommit tests for D156308 · 6a767fbc
  pvanhout authored Jul 26, 2023
  6a767fbc
Jul 25, 2023
- [AMDGPU] Add dynamic stack bit info to kernel-resource-usage Rpass output · 7a4968b5
  Corbin Robeck authored Jul 25, 2023
  7a4968b5
- [FPEnv][AMDGPU] Correct strictfp tests. · 76c22b18
  Kevin P. Neal authored Jul 25, 2023
  76c22b18
- AMDGPU: Correctly expand f64 sqrt intrinsic · e3fd8f83
  Matt Arsenault authored Nov 20, 2022
  e3fd8f83
- AMDGPU: Add more sqrt f64 lowering tests · 47b3ada4
  Matt Arsenault authored Jun 20, 2023
  47b3ada4
- [AMDGPU] Allow vector access types in PromoteAllocaToVector · 3cd4afce
  pvanhout authored Jul 19, 2023
  3cd4afce
- [AMDGPU] Use SSAUpdater in PromoteAlloca · 3890a3b1
  pvanhout authored Jun 28, 2023
  3890a3b1
Jul 24, 2023
- RegisterCoaleser: Fix empty subrange verifier error · 0d797b71
  Matt Arsenault authored Jun 27, 2023
  0d797b71
- RegisterCoalescer: Fix verifier error on redef of subregister for live out implicit_defs · 2a53b6c0
  Matt Arsenault authored Jun 21, 2023
  2a53b6c0
- AMDGPU: Implement combineRepeatedFPDivisors · e561e7cb
  Matt Arsenault authored Jul 01, 2023
  
  e561e7cb
- [AMDGPU] Fix llvm.amdgcn.wave.reduce.umax/umin MIR tests · d163b76c
  Pravin Jagtap authored Jul 24, 2023
  d163b76c
- [AMDGPU] Add llvm.amdgcn.wave.reduce.umin/umax Intrinsic. · c48ed93c
  Pravin Jagtap authored Jul 24, 2023
  c48ed93c
Jul 22, 2023
- AMDGPU: Implement new 2ulp fdiv lowering · 8406c356
  Matt Arsenault authored Jul 16, 2023
  8406c356
- AMDGPU: Refactor AMDGPUCodeGenPrepare fdiv handling · 6699c370
  Matt Arsenault authored Jul 19, 2023
  6699c370
Jul 21, 2023
- AMDGPU: Overhaul and improve rcp and rsq f32 formation · 8287f3af
  Matt Arsenault authored Jul 03, 2023
  8287f3af
- AMDGPU: Add baseline test for fdiv combine · 37512d76
  Matt Arsenault authored Jul 01, 2023
  
  37512d76
- [AMDGPU][RFC] Update isLegalAddressingMode for GFX9 SMEM signed offsets · e45a0c29
  Jay Foad authored Jul 18, 2023
  e45a0c29
- [AMDGPU] Add tests for SMEM addressing modes in CodeGenPrepare · 787bef0b
  Jay Foad authored Jul 20, 2023
  787bef0b
- AMDGPU: Add flag to disable fdiv processing in IR pass · d33ab054
  Matt Arsenault authored Jul 19, 2023
  d33ab054
- AMDGPU: Expand rsq testing to cover contract flag · b2d58b59
  Matt Arsenault authored Jul 17, 2023
  b2d58b59
- AMDGPU: Fold fsub [+-0] into fneg when folding source modifiers · fb54afd1
  Matt Arsenault authored Jul 18, 2023
  fb54afd1
- AMDGPU: Regenerate test checks · 881e9f29
  Matt Arsenault authored Jul 20, 2023
  881e9f29
- AMDGPU: Add baseline test for folding fsub into fneg modifiers · ca34f1bd
  Matt Arsenault authored Jul 18, 2023
  
  ca34f1bd
- AMDGPU: Filter out contract flags when lowering exp · 02955132
  Matt Arsenault authored Jul 20, 2023
  02955132
- AMDGPU: Add some new baseline tests for exp lowering · 076bc374
  Matt Arsenault authored Jul 20, 2023
  
  076bc374
Jul 20, 2023
- Revert "[MachineLICM] Handle Subloops" · 351b4c17
  Jingu Kang authored Jul 20, 2023
  351b4c17
- [MachineLICM] Handle Subloops · 50dd383d
  Jingu Kang authored Jul 19, 2023
  50dd383d
Jul 19, 2023
- [AMDGPUAttributor][FIX] No endless recursion for recursive initializers · d015018c
  Johannes Doerfert authored Jul 19, 2023
  d015018c
- [AMDGPU] Combine the SDAG and GISel versions of the fmed3.ll test. · 1b324272
  Ivan Kosarev authored Jul 19, 2023
  1b324272
- [AMDGPU] Insert s_nop before s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) · 7fa7a08f
  Jay Foad authored Jul 19, 2023
  7fa7a08f