[llvm] [Offload] Remove handling for COV4 binaries from offload/ (PR #131033)

Matt Arsenault via llvm-commits llvm-commits at lists.llvm.org
Wed Mar 12 20:51:57 PDT 2025


================
@@ -3386,20 +3385,16 @@ Error AMDGPUKernelTy::launchImpl(GenericDeviceTy &GenericDevice,
   if (auto Err = AMDGPUDevice.getStream(AsyncInfoWrapper, Stream))
     return Err;
 
-  // Only COV5 implicitargs needs to be set. COV4 implicitargs are not used.
-  if (ImplArgs &&
-      getImplicitArgsSize() == sizeof(hsa_utils::AMDGPUImplicitArgsTy)) {
-    ImplArgs->BlockCountX = NumBlocks[0];
-    ImplArgs->BlockCountY = NumBlocks[1];
-    ImplArgs->BlockCountZ = NumBlocks[2];
-    ImplArgs->GroupSizeX = NumThreads[0];
-    ImplArgs->GroupSizeY = NumThreads[1];
-    ImplArgs->GroupSizeZ = NumThreads[2];
-    ImplArgs->GridDims = NumBlocks[2] * NumThreads[2] > 1
-                             ? 3
-                             : 1 + (NumBlocks[1] * NumThreads[1] != 1);
-    ImplArgs->DynamicLdsSize = KernelArgs.DynCGroupMem;
-  }
+  ImplArgs->BlockCountX = NumBlocks[0];
----------------
arsenm wrote:

Maybe leave a comment on the version 

https://github.com/llvm/llvm-project/pull/131033


More information about the llvm-commits mailing list