[Mlir-commits] [clang] [mlir] [CIR] Implement "uniform_work_group_size" func attribute. (PR #214708)

[Konstantinos Parasyris]

================
@@ -421,8 +421,15 @@ void CIRGenModule::constructAttributeList(
       }
     }
 
-    // TODO(cir): Quite a few CUDA and OpenCL attributes are added here, like
-    // uniform-work-group-size.
+    // TODO(cir): Quite a few CUDA and OpenCL attributes are added here.
+
+    // OpenCL v2.0 Work groups may be whether uniform or not.
+    // '-cl-uniform-work-group-size' compile option gets a hint
+    // to the compiler that the global work-size be a multiple of
+    // the work-group size specified to clEnqueueNDRangeKernel
+    // (i.e. work groups are uniform).
----------------
koparasy wrote:

NIT: 
```
// -cl-uniform-work-group-size / -foffload-uniform-block: work groups are
// uniform (global work-size is a multiple of work-group size).
```
The OG text has typo's. 

https://github.com/llvm/llvm-project/pull/214708