pass file to nvbench (still has free error)

benchmarks/CMakeLists.txt

@@ -8,7 +8,7 @@ foreach(SOURCE IN LISTS BENCHMARK_SOURCES)
   add_executable(${BENCHMARK_NAME} ${SOURCE})
   target_link_libraries(${BENCHMARK_NAME} 
     PRIVATE essentials
-    PRIVATE nvbench::main
+    PRIVATE nvbench::nvbench
   )
   get_target_property(ESSENTIALS_ARCHITECTURES 
     essentials CUDA_ARCHITECTURES

benchmarks/bench.cu

@@ -3,64 +3,80 @@
 #include <gunrock/formats/formats.hxx>
 #include <gunrock/cuda/cuda.hxx>
 #include <gunrock/framework/operators/for/for.hxx>
-#include <gunrock/io/sample_large.hxx>
 #include <nvbench/nvbench.cuh>
 #include <iostream>
 #include <gunrock/algorithms/algorithms.hxx>
 #include <gunrock/algorithms/mst.hxx>
 #include <gunrock/algorithms/bfs.hxx>
+#include <cxxopts.hpp>
 
-namespace gunrock {
-namespace benchmark {
+using namespace gunrock;
+using namespace memory;
 
-void mst_bench(nvbench::state& state) {
-  // Build a graph using a sample csr.
-  auto csr = io::sample_large::csr();
-  auto G =
-      graph::build::from_csr<memory_space_t::device, graph::view_t::csr>(csr);
+using csr_t = format::csr_t<memory_space_t::device, vertex_t, edge_t, weight_t>;
+csr_t csr;
 
-  // Initialize the context.
-  cuda::device_id_t device = 0;
-  cuda::multi_context_t context(device);
+void mst_bench(nvbench::state& state) {
+  state.collect_dram_throughput();
+  state.collect_l1_hit_rates();
+  state.collect_l2_hit_rates();
+  state.collect_loads_efficiency();
+  state.collect_stores_efficiency();
 
-  // --
-  // Params and memory allocation
-  thrust::device_vector<weight_t> mst_weight(1);
+  thrust::device_vector<vertex_t> row_indices(csr.number_of_nonzeros);
+  thrust::device_vector<vertex_t> column_indices(csr.number_of_nonzeros);
+  thrust::device_vector<edge_t> column_offsets(csr.number_of_columns + 1);
 
   // --
-  // GPU Run
-  state.exec(nvbench::exec_tag::sync, [&](nvbench::launch& launch) {
-    gunrock::mst::run(G, mst_weight.data().get());
-  });
-}
+  // Build graph + metadata
 
-void bfs_bench(nvbench::state& state) {
-  // Build a graph using a sample csr.
-  auto csr = io::sample_large::csr();
   auto G =
-      graph::build::from_csr<memory_space_t::device, graph::view_t::csr>(csr);
+      graph::build::from_csr<memory_space_t::device,
+                             graph::view_t::csr /* | graph::view_t::csc */>(
+          csr.number_of_rows,               // rows
+          csr.number_of_columns,            // columns
+          csr.number_of_nonzeros,           // nonzeros
+          csr.row_offsets.data().get(),     // row_offsets
+          csr.column_indices.data().get(),  // column_indices
+          csr.nonzero_values.data().get(),  // values
+          row_indices.data().get(),         // row_indices
+          column_offsets.data().get()       // column_offsets
+      );
 
   // Initialize the context.
   cuda::device_id_t device = 0;
   cuda::multi_context_t context(device);
 
   // --
   // Params and memory allocation
-  vertex_t single_source = 0;
-  vertex_t n_vertices = G.get_number_of_vertices();
-  thrust::device_vector<vertex_t> distances(n_vertices);
-  thrust::device_vector<vertex_t> predecessors(n_vertices);
+  thrust::device_vector<weight_t> mst_weight(1);
 
   // --
   // GPU Run
   state.exec(nvbench::exec_tag::sync, [&](nvbench::launch& launch) {
-    gunrock::bfs::run(G, single_source, distances.data().get(),
-                      predecessors.data().get());
+    gunrock::mst::run(G, mst_weight.data().get());
   });
 }
 
-NVBENCH_BENCH(mst_bench);
-NVBENCH_BENCH(bfs_bench);
+int main(int argc, char** argv) {
+  std::string filename = argv[1];
+
+  if (util::is_market(filename)) {
+    io::matrix_market_t<vertex_t, edge_t, weight_t> mm;
+    csr.from_coo(mm.load(filename));
+  } else if (util::is_binary_csr(filename)) {
+    csr.read_binary(filename);
+  } else {
+    std::cerr << "Unknown file format: " << filename << std::endl;
+    exit(1);
+  }
+
+  char* args[argc - 1];
+  args[0] = argv[0];
+  for (int i = 1; i < argc + 1; i++) {
+    args[i] = argv[i + 1];
+  }
 
-}  // namespace benchmark
-}  // namespace gunrock
+  NVBENCH_BENCH(mst_bench);
+  NVBENCH_MAIN_BODY(argc - 1, args);
+}