add more instances to layernorm

example/42_groupnorm/groupnorm_sigmoid_fp16.cpp

@@ -59,7 +59,7 @@ using DeviceInstance =
                                                       1,   // ClusterM
                                                       1024,  // ClusterK
                                                       1,   // SliceM
-                                                      8,   // SliceK
+                                                      32,   // SliceK
                                                       1,   // SrcVecDim (0=M, 1=K)
                                                       2,   // SrcScalarPerVector
                                                       1,   // GammaVecDim (0=M, 1=K)

library/src/tensor_operation_instance/gpu/normalization/device_layernorm_f16_instance.cpp

@@ -31,7 +31,9 @@ using device_layernorm_f16_instances = std::tuple<
         DeviceLayernormImpl<F16, F16, F16, F32, F16, OutElementwise, Rank, Reduce, 256, 2, 128, 1, 32, 1, 8, 1, 8, 1, 8, 8>,
         DeviceLayernormImpl<F16, F16, F16, F32, F16, OutElementwise, Rank, Reduce, 256, 1, 256, 1, 8, 1, 8, 1, 8, 1, 8, 8>,
         DeviceLayernormImpl<F16, F16, F16, F32, F16, OutElementwise, Rank, Reduce, 256, 1, 256, 1, 16, 1, 8, 1, 8, 1, 8, 8>,
-        DeviceLayernormImpl<F16, F16, F16, F32, F16, OutElementwise, Rank, Reduce, 256, 1, 256, 1, 32, 1, 8, 1, 8, 1, 8, 8>
+        DeviceLayernormImpl<F16, F16, F16, F32, F16, OutElementwise, Rank, Reduce, 256, 1, 256, 1, 32, 1, 8, 1, 8, 1, 8, 8>,
+        DeviceLayernormImpl<F16, F16, F16, F32, F16, OutElementwise, Rank, Reduce, 1024, 1, 1024, 1, 32, 1, 8, 1, 8, 1, 8, 8>,
+        DeviceLayernormImpl<F16, F16, F16, F32, F16, OutElementwise, Rank, Reduce, 1024, 1, 1024, 1, 8, 1, 2, 1, 2, 1, 2, 2>
     // clang-format on
     >;