[all-commits] [llvm/llvm-project] c22876: [MLIR] Add extra locking during cubin generation.

[Stephan Herhut via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: c22876b550f5905b1cb7a4a0f329e76fac3e0232
      https://github.com/llvm/llvm-project/commit/c22876b550f5905b1cb7a4a0f329e76fac3e0232
  Author: Stephan Herhut <herhut at google.com>
  Date:   2020-04-22 (Wed, 22 Apr 2020)

  Changed paths:
    M mlir/lib/Conversion/GPUToCUDA/ConvertKernelFuncToCubin.cpp

  Log Message:
  -----------
  [MLIR] Add extra locking during cubin generation.

We also need to lock the LLVMDialect mutex when initializing
LLVM targets or destroying llvm modules concurrently. Added another
scoped lock to that effect.

Differential Revision: https://reviews.llvm.org/D78580