gtests: graph: unit: add gtests for verifying multi-consumers in pm

Author: ElaineBao

tests/gtests/graph/unit/utils/test_pattern_matcher_cpu.cpp

@@ -1266,6 +1266,78 @@ TEST(test_utils_pattern_matcher, ComplexRepetition) {
     ASSERT_EQ(fusion_ops.size(), 3U);
 }
 
+TEST(test_utils_pattern_matcher, SharedInput) {
+    /* Pattern that captures shared input to two MatMuls
+              |
+             / \
+        MatMul  MatMul
+             \ /
+           Multiply
+              |
+    */
+    auto graphp = std::make_shared<pb_graph_t>();
+    auto pmm1 = graphp->append_op(MatMul);
+    auto pmm2 = graphp->append_op(MatMul);
+    graphp->create_input_port(0, pmm1, 0);
+    graphp->create_input_port(0, pmm2, 0);
+    auto pmul = graphp->append_op(
+            Multiply, {in_edge(0, pmm1, 0), in_edge(1, pmm2, 0)});
+    UNUSED(pmul);
+
+    // test with a graph that has the shared input
+    graph_t agraph;
+    op_t matmul1 {0, MatMul, "matmul1"};
+    op_t matmul2 {1, MatMul, "matmul2"};
+    op_t multiply {2, Multiply, "multiply"};
+
+    std::vector<logical_tensor_t> lt_vec = create_logical_tensors(6);
+    matmul1.add_input(lt_vec[0]);
+    matmul1.add_input(lt_vec[1]);
+    matmul1.add_output(lt_vec[2]);
+    matmul2.add_input(lt_vec[0]);
+    matmul2.add_input(lt_vec[3]);
+    matmul2.add_output(lt_vec[4]);
+    multiply.add_input(lt_vec[2]);
+    multiply.add_input(lt_vec[4]);
+    multiply.add_output(lt_vec[5]);
+
+    ASSERT_EQ(agraph.add_op(&matmul1), status::success);
+    ASSERT_EQ(agraph.add_op(&matmul2), status::success);
+    ASSERT_EQ(agraph.add_op(&multiply), status::success);
+    agraph.finalize();
+
+    std::vector<op_t *> fusion_ops;
+    EXPECT_TRUE(match_pattern(agraph.get_ops()[0].get(), graphp, fusion_ops));
+    ASSERT_EQ(fusion_ops.size(), 3U);
+
+    // test with a graph that does not have the shared input
+    graph_t agraph2;
+    op_t matmul3 {0, MatMul, "matmul1"};
+    op_t matmul4 {1, MatMul, "matmul2"};
+    op_t multiply2 {2, Multiply, "multiply"};
+
+    std::vector<logical_tensor_t> lt_vec2 = create_logical_tensors(7);
+    matmul3.add_input(lt_vec2[0]);
+    matmul3.add_input(lt_vec2[1]);
+    matmul3.add_output(lt_vec2[2]);
+    matmul4.add_input(lt_vec2[3]);
+    matmul4.add_input(lt_vec2[4]);
+    matmul4.add_output(lt_vec2[5]);
+    multiply2.add_input(lt_vec2[2]);
+    multiply2.add_input(lt_vec2[5]);
+    multiply2.add_output(lt_vec2[6]);
+
+    ASSERT_EQ(agraph2.add_op(&matmul3), status::success);
+    ASSERT_EQ(agraph2.add_op(&matmul4), status::success);
+    ASSERT_EQ(agraph2.add_op(&multiply2), status::success);
+    agraph2.finalize();
+
+    std::vector<op_t *> fusion_ops2;
+    EXPECT_FALSE(
+            match_pattern(agraph2.get_ops()[0].get(), graphp, fusion_ops2));
+    ASSERT_EQ(fusion_ops2.size(), 0U);
+}
+
 TEST(test_utils_pattern_matcher, ParallelMatmul) {
     auto graphp = std::make_shared<pb_graph_t>();
     // Pattern that captures shared input to three MatMuls