[Mlir-commits] [mlir] [MLIR][Arith] Partially fold and / or / xor const tripples (PR #195588)

Max Graey llvmlistbot at llvm.org
Mon May 4 02:59:15 PDT 2026


MaxGraey wrote:

In LLVM, I think Reassoc or OptimizeAndOrXor passes handles this.  But that might be a bit of overkill for simple canonicalization in MLIR. Right now, it tries fold until fix-point is reached with limit to 10 iterations.

https://github.com/llvm/llvm-project/pull/195588


More information about the Mlir-commits mailing list