[Mlir-commits] [mlir] [mlir][LLVM] Use int32_t to indirectly construct GEPArg (PR #79562)

[Andrei Golubev]

================
@@ -524,9 +525,9 @@ static void splitIntegerStore(const DataLayout &dataLayout, Location loc,
 
     // We create an `i8` indexed GEP here as that is the easiest (offset is
     // already known). Other patterns turn this into a type-consistent GEP.
-    auto gepOp =
-        rewriter.create<GEPOp>(loc, address.getType(), rewriter.getI8Type(),
-                               address, ArrayRef<GEPArg>{currentOffset});
+    auto gepOp = rewriter.create<GEPOp>(
+        loc, address.getType(), rewriter.getI8Type(), address,
+        ArrayRef<GEPArg>{static_cast<int32_t>(currentOffset)});
----------------
andrey-golubev wrote:

it participates in other expressions involving `unsigned` values (e.g. as "accumulator" below) so figured the cast here is cleaner.

https://github.com/llvm/llvm-project/pull/79562