[llvm-branch-commits] [llvm] [AMDGPU] Intrinsic for launching whole wave functions (PR #145859)

[Matt Arsenault via llvm-branch-commits]

================
@@ -297,8 +297,13 @@ namespace CallingConv {
 /// directly or indirectly via a call-like instruction.
 constexpr bool isCallableCC(CallingConv::ID CC) {
   switch (CC) {
+  // Called with special intrinsics:
+  // llvm.amdgcn.cs.chain
   case CallingConv::AMDGPU_CS_Chain:
   case CallingConv::AMDGPU_CS_ChainPreserve:
+  // llvm.amdgcn.call.whole.wave
+  case CallingConv::AMDGPU_Gfx_WholeWave:
----------------
arsenm wrote:

Ideally would introduce this new calling convention as a separate patch; this needs separate bitcode compatibility tests, and should get its own set of verifier checks for no address capture / only use is the intrinsic call 

https://github.com/llvm/llvm-project/pull/145859