================
@@ -240,12 +251,49 @@ void SpillPlacement::run(MachineFunction &mf, EdgeBundles 
*Bundles,
   TodoList.clear();
   TodoList.setUniverse(bundles->getNumBundles());
 
+  // Check if we should flatten frequencies for AMDGPU to avoid PGO-related
----------------
yxsamliu wrote:

Agreed. Refactored this so the uniformity gating logic is no longer implemented 
inline in SpillPlacement.

- PGO-use now annotates instrumented IR blocks with `offload-block-uniformity` 
metadata.
- Added a target-agnostic `OffloadBlockUniformityAnalysis` that propagates that 
info into MachineFunction.
- SpillPlacement now only queries the analysis (no attribute parsing / modulo 
mapping).

Updated in the latest PR head.

https://github.com/llvm/llvm-project/pull/177665
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to