[OpenMP] Increase max teams in OMPGridValues

[kevinsala]

Summary

Increasing the maximum number of teams in OMPGridValues to 64K for both NVIDIA and AMDGPU.

[jdoerfert]

LG

llvm/include/llvm/Frontend/OpenMP/OMPGridValues.h

constexpr unsigned maxWarpNumber() const {
     return GV_Max_WG_Size / GV_Warp_Size;
   }
 };
 
 /// For AMDGPU GPUs
 static constexpr GV AMDGPUGridValues64 = {
     256,  // GV_Slot_Size
     64,   // GV_Warp_Size
-    128,  // GV_Max_Teams
+    (1 << 16),  // GV_Max_Teams
     896,  // GV_SimpleBufferSize
     1024, // GV_Max_WG_Size,
     256,  // GV_Default_WG_Size
 };
 
 static constexpr GV AMDGPUGridValues32 = {
     256,  // GV_Slot_Size
     32,   // GV_Warp_Size
-    128,  // GV_Max_Teams
+    (1 << 16),  // GV_Max_Teams
     896,  // GV_SimpleBufferSize
     1024, // GV_Max_WG_Size,
     256,  // GV_Default_WG_Size
 };
 
 template <unsigned wavesize> constexpr const GV &getAMDGPUGridValues() {
   static_assert(wavesize == 32 || wavesize == 64, "Unexpected wavesize");
   return wavesize == 32 ? AMDGPUGridValues32 : AMDGPUGridValues64;
 }
 
 /// For Nvidia GPUs
 static constexpr GV NVPTXGridValues = {
     256,  // GV_Slot_Size
     32,   // GV_Warp_Size
-    1024, // GV_Max_Teams
+    (1 << 16), // GV_Max_Teams
     896,  // GV_SimpleBufferSize
     1024, // GV_Max_WG_Size
     128,  // GV_Default_WG_Size
 };
 
 } // namespace omp
 } // namespace llvm
 
 #endif // LLVM_FRONTEND_OPENMP_OMPGRIDVALUES_H