[PATCH] D157004: [mlir][ArmSME] Add vector to tile intrinsics

[Cullen Rhodes via Phabricator via llvm-commits]

c-rhodes marked an inline comment as done.
c-rhodes added a comment.

rebased and addressed comments before landing, @dcaballe @awarzynski thanks for review!



================
Comment at: mlir/test/Target/LLVMIR/arm-sme.mlir:319-328
+// Verify shape of predicate and vector must match
+llvm.func @arm_sme_vector_to_tile_invalid_types(%tileslice : i32,
+                                                %nxv4i1 : vector<[4]xi1>,
+                                                %nxv16i8 : vector<[16]xi8>) {
+  %tile = llvm.mlir.constant(0 : index) : i32
+  // expected-error @+1 {{failed to verify that all of {pg, vector} have same shape}}
+  "arm_sme.intr.write.horiz"(%tile, %tileslice, %nxv4i1, %nxv16i8) :
----------------
awarzynski wrote:
> Move to invalid.mlir
> Move to invalid.mlir

Moved to `arm-sme-invalid.mlir`


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D157004/new/

https://reviews.llvm.org/D157004