[llvm-branch-commits] [llvm] AMDGPU/GlobalISel: Temporal divergence lowering (non i1) (PR #124298)

Sameer Sahasrabuddhe via llvm-branch-commits llvm-branch-commits at lists.llvm.org
Wed Jan 29 22:18:39 PST 2025


================
@@ -342,6 +342,10 @@ template <typename ContextT> class GenericUniformityAnalysisImpl {
       typename SyncDependenceAnalysisT::DivergenceDescriptor;
   using BlockLabelMapT = typename SyncDependenceAnalysisT::BlockLabelMap;
 
+  // Use outside cycle with divergent exit
+  using UOCWDE =
----------------
ssahasra wrote:

Just a suggestion, I would consider giving the name "TemporalDivergenceList" to the entire type ``SmallVector<std::tuple<InstructionT* ...``, and then derive other names from the public member types exposed by the SmallVector. In particular, the function that returns this vector as a range can benefit from simply saying ``auto`` as its return types. This instance of "really long derived names" is justification enough for just saying ``auto`` instead. There is no value in repeating all these names.

https://github.com/llvm/llvm-project/pull/124298


More information about the llvm-branch-commits mailing list