[llvm] [NVPTX] Make i16x2 a native type and add supported vec instructions (PR #65432)

Artem Belevich via llvm-commits llvm-commits at lists.llvm.org
Wed Sep 6 11:52:26 PDT 2023


================
@@ -1678,7 +1681,8 @@ bool NVPTXDAGToDAGISel::tryStore(SDNode *N) {
   MVT ScalarVT = SimpleVT.getScalarType();
   unsigned toTypeWidth = ScalarVT.getSizeInBits();
   if (SimpleVT.isVector()) {
-    assert((StoreVT == MVT::v2f16 || StoreVT == MVT::v2bf16) &&
+    assert((StoreVT == MVT::v2f16 || StoreVT == MVT::v2bf16 ||
+            StoreVT == MVT::v2i16) &&
----------------
Artem-B wrote:

We may want to extract it into an `isv2x16vt` helper function, so we don't clutter code with this boilerplate comparison in many places.

https://github.com/llvm/llvm-project/pull/65432


More information about the llvm-commits mailing list