[all-commits] [llvm/llvm-project] 427b22: [AMDGPU] Extend TargetParser VGPR queries for dyna...

[Chinmay Deshpande via All-commits]

  Branch: refs/heads/users/chinmaydd/amdgpu-targetparser-vgpr-apis
  Home:   https://github.com/llvm/llvm-project
  Commit: 427b224fcea36fe24323a3f6516adb900ceabed3
      https://github.com/llvm/llvm-project/commit/427b224fcea36fe24323a3f6516adb900ceabed3
  Author: Chinmay Deshpande <chdeshpa at amd.com>
  Date:   2026-09-27 (Sun, 27 Sep 2026)

  Changed paths:
    M llvm/include/llvm/TargetParser/AMDGPUTargetParser.h
    M llvm/lib/TargetParser/AMDGPUTargetParser.cpp
    M llvm/unittests/TargetParser/TargetParserTest.cpp

  Log Message:
  -----------
  [AMDGPU] Extend TargetParser VGPR queries for dynamic mode and encoding

Add a VGPR encoding granule query with explicit GPU kind and wave size.
Extend the allocation granule and addressable VGPR queries with an optional
dynamic block size, including the eight-block limit and gfx90a-family
exceptions already used by the backend.

Cover static and dynamic limits, encoding granules, wave sizes, and the
GPUKind and SubArch overloads in TargetParser unit tests. Existing callers
continue to select static mode by default.

Change-Id: I1d389875d25f02b07beb0ea788f16735220007fd



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