[Navi3x] Fix Gridwise_multiple_d operation (#649)

* Add CMake Option "USE_OPT_NAVI3X"

* fix bug

example/30_grouped_conv_fwd_multiple_d/run_grouped_conv_fwd_bias_relu_add_wmma_example.inc

@@ -74,8 +74,8 @@ using DeviceConvFwdInstance =
         8,           // BBlockTransferSrcScalarPerVector
         8,           // BBlockTransferDstScalarPerVector_BK1
         true,        // BBlockLdsExtraN
-        1,
-        1,
+        4,
+        2,
         S<1, 32, 1, 8>,
         8>;
 

include/ck/tensor_operation/gpu/grid/gridwise_gemm_multiple_d_wmma_cshuffle.hpp

@@ -431,6 +431,9 @@ struct GridwiseGemmMultipleD_k0mk1_k0nk1_mn_wmma_cshuffle
         constexpr auto b_block_desc_k0perblock_nperblock_k1 =
             GetBBlockDescriptor_K0PerBlock_NPerBlock_K1();
 
+        constexpr auto cshuffle_block_desc_mshrepeat_mpershrepeat_nshrepeat_npershrepeat =
+            GetCShuffleBlockDescriptor_MShRepeat_MPerShRepeat_NShRepeat_NPerShRepeat();
+
         constexpr auto max_lds_align = K1;
 
         constexpr auto a_block_space_size_aligned = math::integer_least_multiple(
@@ -439,8 +442,13 @@ struct GridwiseGemmMultipleD_k0mk1_k0nk1_mn_wmma_cshuffle
         constexpr auto b_block_space_size_aligned = math::integer_least_multiple(
             b_block_desc_k0perblock_nperblock_k1.GetElementSpaceSize(), max_lds_align);
 
-        return (a_block_space_size_aligned * sizeof(ADataType) +
-                b_block_space_size_aligned * sizeof(BDataType));
+        constexpr auto c_block_space_size_aligned = math::integer_least_multiple(
+            cshuffle_block_desc_mshrepeat_mpershrepeat_nshrepeat_npershrepeat.GetElementSpaceSize(),
+            max_lds_align);
+
+        return math::max((a_block_space_size_aligned * sizeof(ADataType) +
+                          b_block_space_size_aligned * sizeof(BDataType)),
+                         c_block_space_size_aligned * sizeof(CShuffleDataType));
     }
 
     // block_id to matrix tile idx (m0, n0) mapping are controlled by {M01, N01}