[Mlir-commits] [flang] [mlir] [MLIR] Add ComplexTOROCDL pass (PR #144926)

[Krzysztof Drewniak]

================
@@ -4105,8 +4106,10 @@ class FIRToLLVMLowering
     // GPU library calls, the rest can be converted to LLVM intrinsics, which
     // is handled in the mathToLLVM conversion. The lowering to libm calls is
     // not needed since all math operations are handled this way.
-    if (isAMDGCN)
+    if (isAMDGCN) {
       mathConvertionPM.addPass(mlir::createConvertMathToROCDL());
+      mathConvertionPM.addPass(mlir::createConvertComplexToROCDL());
----------------
krzysz00 wrote:

AI's got a point on this one

https://github.com/llvm/llvm-project/pull/144926