[all-commits] [llvm/llvm-project] a910a0: [mlir][x86] Fix - Replace `load` with `transfer_re...

Arun Thangamani via All-commits all-commits at lists.llvm.org
Tue Apr 28 01:33:25 PDT 2026


  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a910a0b4b5252627c0150c26ba94cc8479ad6e1a
      https://github.com/llvm/llvm-project/commit/a910a0b4b5252627c0150c26ba94cc8479ad6e1a
  Author: Arun Thangamani <arun.thangamani at intel.com>
  Date:   2026-04-28 (Tue, 28 Apr 2026)

  Changed paths:
    M mlir/lib/Dialect/X86/Transforms/VectorContractToPackedTypeDotProduct.cpp
    M mlir/test/Dialect/X86/vector-contract-to-packed-type-dotproduct.mlir

  Log Message:
  -----------
  [mlir][x86] Fix - Replace `load` with `transfer_read` to support on tensor type. (#194543)

This `patch` replaces `vector.load` operation with
`vector.transfer_read` op, such that the re-write lowers
`vector.contract` ops to `bf16_avx512_dp`.



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


More information about the All-commits mailing list