[Refactor] Unify the shared pass prefix between vm and graph

src/relay/backend/build_module.cc

@@ -313,57 +313,7 @@ class RelayBuildModule : public runtime::ModuleNode {
       relay_module_ptr->Update(main_glb_var, new_main);
     }
 
-    Array<Pass> pass_seqs;
-    Array<runtime::String> entry_functions{"main"};
-    pass_seqs.push_back(transform::RemoveUnusedFunctions(entry_functions));
-    pass_seqs.push_back(transform::ToBasicBlockNormalForm());
-
-    // Run all dialect legalization passes.
-    pass_seqs.push_back(relay::qnn::transform::Legalize());
-
-    // Legalize pass is restricted to homogeneous execution for now.
-    if (targets.size() == 1) {
-      pass_seqs.push_back(transform::Legalize());
-    }
-
-    pass_seqs.push_back(transform::SimplifyInference());
-
-    // Convert Dynamic ops to static versions
-    pass_seqs.push_back(transform::DynamicToStatic());
-
-    PackedFunc fskip = PackedFunc([](TVMArgs args, TVMRetValue* rv) {
-      Expr expr = args[0];
-      *rv = false;
-      if (expr.as<CallNode>()) {
-        auto call_node = expr.as<CallNode>();
-        auto op_node = call_node->op.as<OpNode>();
-        if (op_node->name == "cast") {
-          auto attrs = call_node->attrs.as<CastAttrs>();
-          if (attrs->dtype == DataType::Int(32)) {
-            *rv = true;
-          }
-        }
-      }
-    });
-    pass_seqs.push_back(transform::EliminateCommonSubexpr(fskip));
-    pass_seqs.push_back(transform::SimplifyExpr());
-    pass_seqs.push_back(transform::CombineParallelConv2D(3));
-    pass_seqs.push_back(transform::CombineParallelDense(3));
-    pass_seqs.push_back(transform::CombineParallelBatchMatmul(3));
-    pass_seqs.push_back(transform::FoldConstant());
-    pass_seqs.push_back(transform::FoldScaleAxis());
-    pass_seqs.push_back(transform::CanonicalizeCast());
-    pass_seqs.push_back(transform::CanonicalizeOps());
-
-    // Alter layout transformation is only applied to homogeneous execution yet.
-    if (targets.size() == 1) {
-      pass_seqs.push_back(transform::InferType());
-      pass_seqs.push_back(transform::AlterOpLayout());
-    }
-
-    // Fast math optimizations.
-    pass_seqs.push_back(transform::FastMath());
-    pass_seqs.push_back(transform::FoldConstant());
+    Array<Pass> pass_seqs = GetPrefixOpts(targets, false);
 
     if (targets.size() == 1) {
       const auto& target = (*targets.begin()).second;

src/relay/backend/utils.cc

@@ -120,6 +120,71 @@ TVM_STATIC_IR_FUNCTOR(ReprPrinter, vtable)
                 << ",\n  relay_primfuncs=" << node->relay_primfuncs << ")";
     });
 
+Array<Pass> GetPrefixOpts(const Map<tvm::Integer, tvm::Target>& targets, bool is_vm) {
+  Array<Pass> pass_seqs;
+  Array<runtime::String> entry_functions{"main"};
+  pass_seqs.push_back(transform::RemoveUnusedFunctions(entry_functions));
+  pass_seqs.push_back(transform::ToBasicBlockNormalForm());
+  // Run all dialect legalization passes.
+  pass_seqs.push_back(relay::qnn::transform::Legalize());
+
+  // Legalize pass is restricted to homogeneous execution for now.
+  if (targets.size() == 1) {
+    pass_seqs.push_back(transform::Legalize());
+  }
+
+  pass_seqs.push_back(transform::SimplifyInference());
+
+  if (is_vm) {
+    // eta expand to support constructors in argument position
+    pass_seqs.push_back(transform::EtaExpand(
+        /* expand_constructor */ true, /* expand_global_var */ false));
+  } else {
+    // Convert Dynamic ops to static versions
+    pass_seqs.push_back(transform::DynamicToStatic());
+  }
+
+  PackedFunc fskip = PackedFunc([](TVMArgs args, TVMRetValue* rv) {
+    Expr expr = args[0];
+    if (expr.as<CallNode>()) {
+      auto call_node = expr.as<CallNode>();
+      auto op_node = call_node->op.as<OpNode>();
+      if (op_node->name == "cast") {
+        auto attrs = call_node->attrs.as<CastAttrs>();
+        if (attrs->dtype == DataType::Int(32)) {
+          *rv = true;
+        }
+      }
+    }
+    *rv = false;
+  });
+  pass_seqs.push_back(transform::EliminateCommonSubexpr(fskip));
+  pass_seqs.push_back(transform::SimplifyExpr());
+  if (is_vm) {
+    pass_seqs.push_back(transform::InlinePrimitives());
+  }
+  pass_seqs.push_back(transform::CombineParallelConv2D(3));
+  pass_seqs.push_back(transform::CombineParallelDense(3));
+  pass_seqs.push_back(transform::CombineParallelBatchMatmul(3));
+  pass_seqs.push_back(transform::FoldConstant());
+  pass_seqs.push_back(transform::FoldScaleAxis());
+  pass_seqs.push_back(transform::CanonicalizeCast());
+  pass_seqs.push_back(transform::CanonicalizeOps());
+
+  // Alter layout transformation is only applied to homogeneous execution yet.
+  if (targets.size() == 1) {
+    if (!is_vm) {
+      pass_seqs.push_back(transform::InferType());
+    }
+    pass_seqs.push_back(transform::AlterOpLayout());
+  }
+
+  // Fast math optimizations.
+  pass_seqs.push_back(transform::FastMath());
+  pass_seqs.push_back(transform::FoldConstant());
+  return pass_seqs;
+}
+
 }  // namespace backend
 }  // namespace relay
 }  // namespace tvm

src/relay/backend/utils.h

@@ -28,6 +28,7 @@
 #include <tvm/driver/driver_api.h>
 #include <tvm/relay/expr.h>
 #include <tvm/relay/expr_functor.h>
+#include <tvm/relay/qnn/transform.h>
 #include <tvm/relay/transform.h>
 #include <tvm/relay/type.h>
 #include <tvm/target/codegen.h>
@@ -44,7 +45,12 @@
 
 namespace tvm {
 namespace relay {
+namespace transform {
+Pass InlinePrimitives();
+}
+
 namespace backend {
+using Pass = tvm::transform::Pass;
 
 /*!
  * \brief The static storage information produced by memory planning.
@@ -410,6 +416,15 @@ inline bool IsCompileEngineCacheDisabled() {
       .value();
 }
 
+/*!
+ * \brief Get a shared optimization pass prefix between vm and graph executor.
+ *
+ * \param targets The device type to `Target` mapping.
+ * \param is_vm A boolean indicating if the passes are used for vm or graph executor.
+ * \return An array of passes.
+ */
+Array<Pass> GetPrefixOpts(const Map<tvm::Integer, tvm::Target>& targets, bool is_vm);
+
 }  // namespace backend
 }  // namespace relay
 }  // namespace tvm

src/relay/backend/vm/compiler.cc

@@ -1042,57 +1042,7 @@ IRModule VMCompiler::OptimizeModule(IRModule mod, const TargetsMap& targets_arg,
     mod->Add(gvar, f);
   }
 
-  Array<Pass> pass_seqs;
-  Array<runtime::String> entry_functions{"main"};
-  pass_seqs.push_back(transform::RemoveUnusedFunctions(entry_functions));
-  pass_seqs.push_back(transform::ToBasicBlockNormalForm());
-  // Run all dialect legalization passes.
-  pass_seqs.push_back(relay::qnn::transform::Legalize());
-
-  // Legalize pass is restricted to homogeneous execution for now.
-  if (targets.size() == 1) {
-    pass_seqs.push_back(transform::Legalize());
-  }
-
-  // eta expand to support constructors in argument position
-  pass_seqs.push_back(transform::EtaExpand(
-      /* expand_constructor */ true, /* expand_global_var */ false));
-
-  pass_seqs.push_back(transform::SimplifyInference());
-  PackedFunc fskip = PackedFunc([](TVMArgs args, TVMRetValue* rv) {
-    Expr expr = args[0];
-    if (expr.as<CallNode>()) {
-      auto call_node = expr.as<CallNode>();
-      auto op_node = call_node->op.as<OpNode>();
-      if (op_node->name == "cast") {
-        auto attrs = call_node->attrs.as<CastAttrs>();
-        if (attrs->dtype == DataType::Int(32)) {
-          *rv = true;
-        }
-      }
-    }
-    *rv = false;
-  });
-  pass_seqs.push_back(transform::EliminateCommonSubexpr(fskip));
-  pass_seqs.push_back(transform::SimplifyExpr());
-  pass_seqs.push_back(transform::InlinePrimitives());
-
-  pass_seqs.push_back(transform::CombineParallelConv2D(3));
-  pass_seqs.push_back(transform::CombineParallelDense(3));
-  pass_seqs.push_back(transform::CombineParallelBatchMatmul(3));
-  pass_seqs.push_back(transform::FoldConstant());
-  pass_seqs.push_back(transform::FoldScaleAxis());
-  pass_seqs.push_back(transform::CanonicalizeCast());
-  pass_seqs.push_back(transform::CanonicalizeOps());
-
-  // Alter layout transformation is only applied to homogeneous execution yet.
-  if (targets.size() == 1) {
-    pass_seqs.push_back(transform::AlterOpLayout());
-  }
-
-  // Fast math optimizations.
-  pass_seqs.push_back(transform::FastMath());
-  pass_seqs.push_back(transform::FoldConstant());
+  Array<Pass> pass_seqs = relay::backend::GetPrefixOpts(targets, true);
 
   if (targets_.size() > 1) {
     // Handle heterogeneous compilation.