[all-commits] [llvm/llvm-project] e1a812: [AMDGPU] Support double type in atomic optimizer. ...

[Pravin Jagtap via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: e1a8120a63cdb6c9567b0f68d9a0390e4f5da184
      https://github.com/llvm/llvm-project/commit/e1a8120a63cdb6c9567b0f68d9a0390e4f5da184
  Author: Pravin Jagtap <prjagtap at amd.com>
  Date:   2024-03-22 (Fri, 22 Mar 2024)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUAtomicOptimizer.cpp
    M llvm/test/CodeGen/AMDGPU/GlobalISel/fp64-atomics-gfx90a.ll
    M llvm/test/CodeGen/AMDGPU/fp64-atomics-gfx90a.ll
    M llvm/test/CodeGen/AMDGPU/global_atomic_optimizer_fp_rtn.ll
    M llvm/test/CodeGen/AMDGPU/global_atomics_optimizer_fp_no_rtn.ll
    M llvm/test/CodeGen/AMDGPU/global_atomics_scan_fadd.ll
    M llvm/test/CodeGen/AMDGPU/global_atomics_scan_fmax.ll
    M llvm/test/CodeGen/AMDGPU/global_atomics_scan_fmin.ll
    M llvm/test/CodeGen/AMDGPU/global_atomics_scan_fsub.ll

  Log Message:
  -----------
  [AMDGPU] Support double type in atomic optimizer. (#84307)

Presently the atomic optimizer supports only 32-bit operations. Plan is
to extend the atomic optimizer for 64-bit operations for compute and
graphics. This patch extends support for double type for `uniform
values` only. Going forward, will extend the support for divergent
values. Adding support for divergent values requires
extending/legalizing readfirstlane, readlane, writelane, etc ops for
64-bit operations to avoid `bitcast` noise that we have currently.

---------

Authored-by: Pravin Jagtap <Pravin.Jagtap at amd.com>



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