[all-commits] [llvm/llvm-project] b00288: [AMDGPU] Extending wave reduction intrinsics for `...

[Aaditya via All-commits]

  Branch: refs/heads/users/easyonaadit/amdgpu/wave-reduce-intrinsics-arithmetic
  Home:   https://github.com/llvm/llvm-project
  Commit: b00288d734d7310cb708ea9336de1fc4af5bb66c
      https://github.com/llvm/llvm-project/commit/b00288d734d7310cb708ea9336de1fc4af5bb66c
  Author: Aaditya <Aaditya.AlokDeshpande at amd.com>
  Date:   2025-09-10 (Wed, 10 Sep 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.max.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.min.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.umax.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.umin.ll

  Log Message:
  -----------
  [AMDGPU] Extending wave reduction intrinsics for `i64` types - 1

Supporting Min/Max Operations: `min`, `max`, `umin`, `umax`


  Commit: b5ac2215c9c9eb2b1f1c304f4bc3e71becc10dbc
      https://github.com/llvm/llvm-project/commit/b5ac2215c9c9eb2b1f1c304f4bc3e71becc10dbc
  Author: Aaditya <Aaditya.AlokDeshpande at amd.com>
  Date:   2025-09-10 (Wed, 10 Sep 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp

  Log Message:
  -----------
  Addressing Review Comments


  Commit: 5d5f775756c80b1a35789e377a7614ed2ba8e07d
      https://github.com/llvm/llvm-project/commit/5d5f775756c80b1a35789e377a7614ed2ba8e07d
  Author: Aaditya <Aaditya.AlokDeshpande at amd.com>
  Date:   2025-09-10 (Wed, 10 Sep 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.max.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.min.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.umax.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.umin.ll

  Log Message:
  -----------
  Using `S_MOV_B64_IMM_PSEUDO` instead of dealing with legality concerns.


  Commit: cffbb21c9c964b6ac3af4acbd79bd2c9d8f45642
      https://github.com/llvm/llvm-project/commit/cffbb21c9c964b6ac3af4acbd79bd2c9d8f45642
  Author: Aaditya <Aaditya.AlokDeshpande at amd.com>
  Date:   2025-09-10 (Wed, 10 Sep 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.add.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.sub.ll

  Log Message:
  -----------
  [AMDGPU] Extending wave reduction intrinsics for `i64` types - 2

Supporting Arithemtic Operations: `add`, `sub`


  Commit: afb2d771c3d677ce530bd0a56832ff668c05eff5
      https://github.com/llvm/llvm-project/commit/afb2d771c3d677ce530bd0a56832ff668c05eff5
  Author: Aaditya <Aaditya.AlokDeshpande at amd.com>
  Date:   2025-09-10 (Wed, 10 Sep 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.add.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.sub.ll

  Log Message:
  -----------
  Marking dead scc


  Commit: eb9b409144edacf5283359db19a9c730f9c71688
      https://github.com/llvm/llvm-project/commit/eb9b409144edacf5283359db19a9c730f9c71688
  Author: Aaditya <Aaditya.AlokDeshpande at amd.com>
  Date:   2025-09-10 (Wed, 10 Sep 2025)

  Changed paths:
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.add.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.sub.ll

  Log Message:
  -----------
  Checking for targets with native 64-bit `add`/`sub` support


  Commit: 6497e4f1e2882d583404586d43d553138ecf2968
      https://github.com/llvm/llvm-project/commit/6497e4f1e2882d583404586d43d553138ecf2968
  Author: Aaditya <Aaditya.AlokDeshpande at amd.com>
  Date:   2025-09-10 (Wed, 10 Sep 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.max.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.min.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.umax.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.reduce.umin.ll

  Log Message:
  -----------
  removing unused variable


Compare: https://github.com/llvm/llvm-project/compare/b00288d734d7%5E...6497e4f1e288

To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications