[all-commits] [llvm/llvm-project] e52494: [AMDGPU] Price scalar integer to fp casts by sourc...

[Dmitry Sidorov via All-commits]

  Branch: refs/heads/users/mrsidims/fix-amdgpu-scalar-int-fp-cast-cost
  Home:   https://github.com/llvm/llvm-project
  Commit: e52494828f8f261114c71014e68b2e21a8930f99
      https://github.com/llvm/llvm-project/commit/e52494828f8f261114c71014e68b2e21a8930f99
  Author: Dmitry Sidorov <Dmitry.Sidorov at amd.com>
  Date:   2026-10-02 (Fri, 02 Oct 2026)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
    M llvm/test/Analysis/CostModel/AMDGPU/narrow-int-to-bfloat.ll
    M llvm/test/Analysis/CostModel/AMDGPU/narrow-int-to-fp.ll

  Log Message:
  -----------
  [AMDGPU] Price scalar integer to fp casts by source width and sign

Scalar sources between a byte and 31 bits fell to the default cost of one
while the matching vector lanes were already priced, which skewed the
difference SLP weighs a bundle against. Such a source is extended before
the conversion, and what the extension takes depends on the width, on the
sign and on whether the subtarget has SDWA and 16 bit instructions.
Sources narrower than a byte are left alone, because their vector form is
not priced either.


  Commit: 293fc582c7d6feb98ea0daf4028c5c89e97ccd77
      https://github.com/llvm/llvm-project/commit/293fc582c7d6feb98ea0daf4028c5c89e97ccd77
  Author: Dmitry Sidorov <Dmitry.Sidorov at amd.com>
  Date:   2026-10-02 (Fri, 02 Oct 2026)

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

  Log Message:
  -----------
  format


  Commit: f161776d203ce7918a3123bbd0a17f23b2d88801
      https://github.com/llvm/llvm-project/commit/f161776d203ce7918a3123bbd0a17f23b2d88801
  Author: Dmitry Sidorov <Dmitry.Sidorov at amd.com>
  Date:   2026-10-02 (Fri, 02 Oct 2026)

  Changed paths:
    M llvm/test/CodeGen/AMDGPU/slp-int-to-fp.ll

  Log Message:
  -----------
  update slp-int-to-fp.ll


  Commit: 6cbd60ef5fc42bb6c1e3f91530f7889e91607a1b
      https://github.com/llvm/llvm-project/commit/6cbd60ef5fc42bb6c1e3f91530f7889e91607a1b
  Author: Dmitry Sidorov <Dmitry.Sidorov at amd.com>
  Date:   2026-10-02 (Fri, 02 Oct 2026)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
    M llvm/test/Analysis/CostModel/AMDGPU/narrow-int-to-bfloat.ll
    M llvm/test/Analysis/CostModel/AMDGPU/narrow-int-to-fp.ll

  Log Message:
  -----------
  few fixes


  Commit: ca7e60362682a6699cadd6287d1ca301d6a26077
      https://github.com/llvm/llvm-project/commit/ca7e60362682a6699cadd6287d1ca301d6a26077
  Author: Dmitry Sidorov <Dmitry.Sidorov at amd.com>
  Date:   2026-10-02 (Fri, 02 Oct 2026)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
    M llvm/test/Analysis/CostModel/AMDGPU/narrow-int-to-bfloat.ll
    M llvm/test/Analysis/CostModel/AMDGPU/narrow-int-to-fp.ll

  Log Message:
  -----------
  [AMDGPU] Do not price the extension of a loaded i40, i48 or i56 in int to fp casts

A widened constant or invariant load still pays it.


  Commit: d05a0a5f83ec58c33dfa6867670376bd56b02e4e
      https://github.com/llvm/llvm-project/commit/d05a0a5f83ec58c33dfa6867670376bd56b02e4e
  Author: Dmitry Sidorov <Dmitry.Sidorov at amd.com>
  Date:   2026-10-02 (Fri, 02 Oct 2026)

  Changed paths:
    M llvm/test/CodeGen/AMDGPU/slp-int-to-fp.ll

  Log Message:
  -----------
  update slp-int-to-fp.ll


Compare: https://github.com/llvm/llvm-project/compare/69c89de2c9d6...d05a0a5f83ec

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