[all-commits] [llvm/llvm-project] 305fb4: [AMDGPU] Add structural stall heuristic to schedul...

[Austin Kerbow via All-commits]

  Branch: refs/heads/users/kerbowa/structural-vs-latency-stall-heuristic
  Home:   https://github.com/llvm/llvm-project
  Commit: 305fb4fbfadccd26c39249a725e9e06bea8a9102
      https://github.com/llvm/llvm-project/commit/305fb4fbfadccd26c39249a725e9e06bea8a9102
  Author: Austin Kerbow <Austin.Kerbow at amd.com>
  Date:   2025-12-01 (Mon, 01 Dec 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUMLSchedStrategy.cpp
    M llvm/lib/Target/AMDGPU/AMDGPUMLSchedStrategy.h
    M llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp
    M llvm/lib/Target/AMDGPU/GCNHazardRecognizer.h
    M llvm/lib/Target/AMDGPU/GCNSchedStrategy.cpp
    M llvm/lib/Target/AMDGPU/GCNSchedStrategy.h
    M llvm/test/CodeGen/AMDGPU/ml-sched-effective-stall.mir

  Log Message:
  -----------
  [AMDGPU] Add structural stall heuristic to scheduling strategies

Implements a structural stall heuristic that considers both resource
hazards and latency constraints when selecting instructions from the
pending queue.

- Add getStructuralStallCycles() to GCNSchedStrategy that computes the
number of cycles an instruction must wait due to:
  - Resource conflicts on unbuffered resources (from the SchedModel)
  - Sequence-dependent hazards (from GCNHazardRecognizer)

- Add getHazardWaitStates() to GCNHazardRecognizer that returns the number
of wait states until all hazards for an instruction are resolved,
providing cycle-accurate hazard information for scheduling heuristics.



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