[all-commits] [llvm/llvm-project] 6d18c2: [Matrix] Update load/store intrinsics.

[Florian Hahn via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 6d18c2067ef1f2450078c115966ebb2699b0558c
      https://github.com/llvm/llvm-project/commit/6d18c2067ef1f2450078c115966ebb2699b0558c
  Author: Florian Hahn <flo at fhahn.com>
  Date:   2020-06-18 (Thu, 18 Jun 2020)

  Changed paths:
    M llvm/docs/LangRef.rst
    M llvm/include/llvm/IR/Intrinsics.td
    M llvm/include/llvm/IR/MatrixBuilder.h
    M llvm/lib/IR/Verifier.cpp
    M llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
    M llvm/test/Transforms/LowerMatrixIntrinsics/bigger-expressions-double.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/const-gep.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/multiply-add-sub-double-row-major.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/propagate-backward.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/propagate-forward.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/propagate-mixed-users.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/propagate-multiple-iterations.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/remarks-inlining.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/remarks-shared-subtrees.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/remarks.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-double.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-float.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-i32.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-double.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-float.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-i32.ll
    M llvm/test/Verifier/matrix-intrinsics.ll
    M mlir/include/mlir/Dialect/LLVMIR/LLVMOps.td
    M mlir/test/Target/llvmir-intrinsics.mlir

  Log Message:
  -----------
  [Matrix] Update load/store intrinsics.

This patch adjust the load/store matrix intrinsics, formerly known as
llvm.matrix.columnwise.load/store, to improve the naming and allow
passing of extra information (volatile).

The patch performs the following changes:
 * Rename columnwise.load/store to column.major.load/store. This is more
   expressive and also more in line with the naming in Clang.
 * Changes the stride arguments from i32 to i64. The stride can be
   larger than i32 and this makes things more uniform with the way
   things are handled in Clang.
 * A new boolean argument is added to indicate whether the load/store
   is volatile. The lowering respects that when emitting vector
   load/store instructions
 * MatrixBuilder is updated to require both Alignment and IsVolatile
   arguments, which are passed through to the generated intrinsic. The
   alignment is set using the `align` attribute.

The changes are grouped together in a single patch, to have a single
commit that breaks the compatibility. We probably should be fine with
updating the intrinsics, as we did not yet officially support them in
the last stable release. If there are any concerns, we can add
auto-upgrade rules for the columnwise intrinsics though.

Reviewers: anemet, Gerolf, hfinkel, andrew.w.kaylor, LuoYuanke, nicolasvasilache, rjmccall, ftynse

Reviewed By: anemet, nicolasvasilache

Differential Revision: https://reviews.llvm.org/D81472