Implement WU architecture support

hw/rtl/VX_config.vh

@@ -88,15 +88,48 @@
 `endif
 
 `ifndef NUM_CORES
-`define NUM_CORES 8
+`define NUM_CORES 1
 `endif
 
 `ifndef NUM_WARPS
-`define NUM_WARPS 8
+`define NUM_WARPS 4
+`endif
+
+`ifndef NUM_TENSOR_WARPS
+`define NUM_TENSOR_WARPS 2
+`endif
+
+`define NUM_SCALAR_WARPS (`NUM_WARPS - `NUM_TENSOR_WARPS)
+
+`define WU_CONFIG_STATIC_ASSERTS \
+    generate \
+        if (!(`NUM_WARPS > 0)) begin : g_wu_num_warps_gt_zero \
+            invalid_NUM_WARPS_must_be_greater_than_zero __wu_config_error(); \
+        end \
+        if (!(`NUM_TENSOR_WARPS > 0)) begin : g_wu_num_tensor_warps_gt_zero \
+            invalid_NUM_TENSOR_WARPS_must_be_greater_than_zero __wu_config_error(); \
+        end \
+        if (!(`NUM_TENSOR_WARPS < `NUM_WARPS)) begin : g_wu_num_tensor_warps_lt_num_warps \
+            invalid_NUM_TENSOR_WARPS_must_be_smaller_than_NUM_WARPS __wu_config_error(); \
+        end \
+        if (!(`NUM_SCALAR_WARPS > 0)) begin : g_wu_num_scalar_warps_gt_zero \
+            invalid_NUM_SCALAR_WARPS_must_be_greater_than_zero __wu_config_error(); \
+        end \
+    endgenerate
+
+`define IS_SCALAR_WARP(wid) ((wid) < `NUM_SCALAR_WARPS)
+`define IS_TENSOR_WARP(wid) ((wid) >= `NUM_SCALAR_WARPS)
+
+`ifndef TENSOR_NUM_GPRS
+`define TENSOR_NUM_GPRS 8
+`endif
+
+`ifndef TENSOR_NUM_FPRS
+`define TENSOR_NUM_FPRS 8
 `endif
 
 `ifndef NUM_THREADS
-`define NUM_THREADS 8
+`define NUM_THREADS 4
 `endif
 
 `ifndef NUM_BARRIERS