[all-commits] [llvm/llvm-project] 9744b4: [CIR][AMDGPU] Add support for AMDGCN rsq and rsq_c...

[Ayokunle Amodu via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 9744b475e28e180599fd1a42d956f74b0a1aaed8
      https://github.com/llvm/llvm-project/commit/9744b475e28e180599fd1a42d956f74b0a1aaed8
  Author: Ayokunle Amodu <ayokunle321 at gmail.com>
  Date:   2026-07-15 (Wed, 15 Jul 2026)

  Changed paths:
    M clang/lib/CIR/CodeGen/CIRGenBuiltinAMDGPU.cpp
    M clang/test/CIR/CodeGenHIP/builtins-amdgcn-gfx1250.hip
    M clang/test/CIR/CodeGenHIP/builtins-amdgcn-vi-f16.hip
    M clang/test/CIR/CodeGenHIP/builtins-amdgcn.hip

  Log Message:
  -----------
  [CIR][AMDGPU] Add support for AMDGCN rsq and rsq_clamp builtins (#197349)

Adds codegen for the following AMDGCN reciprocal square root builtins:

- __builtin_amdgcn_rsq (double)
- __builtin_amdgcn_rsqf (float)
- __builtin_amdgcn_rsqh (half)
- __builtin_amdgcn_rsq_bf16 (bfloat16)
- __builtin_amdgcn_rsq_clamp (double)
- __builtin_amdgcn_rsq_clampf (float)

These are lowered to the corresponding `llvm.amdgcn.rsq` and
`llvm.amdgcn.rsq.clamp` intrinsic calls.



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