[llvm] [AMDGPU] Optimize image sample followed by llvm.amdgcn.cvt.pkrtz into d16 variant (PR #145203)

[Georg Lehmann via llvm-commits]

================
@@ -247,6 +247,42 @@ simplifyAMDGCNImageIntrinsic(const GCNSubtarget *ST,
                                        ArgTys[0] = User->getType();
                                      });
         }
+
+        // Fold image.sample + cvt.pkrtz -> extractelement idx0 into a single
+        // d16 image sample.
----------------
DadSchoorse wrote:

> conversion of other input data formats uses round-to-nearest-even.

This is exactly the issue I mentioned. For fixed points formats, the data is not converted to fp32 and then truncated to fp16, it's directly rounded to fp16.

https://github.com/llvm/llvm-project/pull/145203