llvm-project

79225349 - [mlir][linalg] Add patterns to convert matmul to transposed variants (#89075)

[mlir][linalg] Add patterns to convert matmul to transposed variants (#89075) This adds patterns to convert from the Linalg matmul and batch_matmul ops to the transposed variants. By default the LHS matrix is transposed. Our work enabling a lowering path from linalg.matmul to ArmSME has revealed the current lowering results in non-contiguous memory accesses for the A matrix and very poor performance. These patterns provide a simple option to fix this.