[mlir] Fix typo s/opreations/operations (#163544)

This commit is contained in:
peledins-zimperium 2025-12-26 14:03:25 +02:00 committed by GitHub
parent 7c2bbfae3e
commit 64496be8e0
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 3 additions and 3 deletions

View file

@ -505,7 +505,7 @@ private:
/// A dense backward dataflow analysis propagating lattices after and before the
/// execution of every operation across the IR by implementing transfer
/// functions for opreations.
/// functions for operations.
///
/// `LatticeT` is expected to be a subclass of `AbstractDenseLattice`.
template <typename LatticeT>

View file

@ -700,7 +700,7 @@ private:
/// - `throughValue` is the payload value the handle to which is consumed,
/// when it is the case, null when the operation handle is consumed
/// directly.
/// Looks at the payload opreations associated with `otherHandle` and if any
/// Looks at the payload operations associated with `otherHandle` and if any
/// of these operations has an ancestor (or is itself) listed in
/// `potentialAncestors`, records the error message describing the use of the
/// invalidated handle. Does nothing if `otherHandle` already has a reporter

View file

@ -275,7 +275,7 @@ module attributes { transform.with_named_sequence } {
// Vectorize the remaining non-unit dimensions in structured operations.
// This essentially rewrites operations on `tensor<5x64xf32>` into
// opreations on `vector<5x64xf32>`. Further lowering in MLIR and LLVM will
// operations on `vector<5x64xf32>`. Further lowering in MLIR and LLVM will
// decompose this into a sequence of operations on single-dimensional
// vectors of the platform-relevant size, e.g., `vector<16xf32>` for AVX512.
// High-level vector primitives, such as `vector.transpose` and