[all-commits] [llvm/llvm-project] 8f4f26: [MLGO] ML Regalloc Priority Advisor

[Eric Wang via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 8f4f26ba5bd04f7b335836021e5e63b4236c0305
      https://github.com/llvm/llvm-project/commit/8f4f26ba5bd04f7b335836021e5e63b4236c0305
  Author: Eric Wang <flpha0830 at gmail.com>
  Date:   2022-09-29 (Thu, 29 Sep 2022)

  Changed paths:
    A llvm/lib/Analysis/models/gen-regalloc-priority-test-model.py
    M llvm/lib/CodeGen/CMakeLists.txt
    A llvm/lib/CodeGen/MLRegallocPriorityAdvisor.cpp
    M llvm/lib/CodeGen/RegAllocPriorityAdvisor.cpp
    A llvm/test/CodeGen/MLRegalloc/Inputs/reference-prio-log-noml.txt
    A llvm/test/CodeGen/MLRegalloc/dev-mode-prio-logging.ll

  Log Message:
  -----------
  [MLGO] ML Regalloc Priority Advisor

The bulk of the implementation is common between 'release' mode (==AOT-ed
model) and 'development' mode (for training), the main difference is
that in development mode, we may also log features (for training logs),
inject scoring information and then produce the log file.

Differential Revision: https://reviews.llvm.org/D133616