Merge branch 'master' into 3d_noc

.github/workflows/nightly_test.yml

@@ -0,0 +1,110 @@
+name: NightlyTest
+
+on:
+  # We want to run the CI when anything is pushed to master.
+  # Since master is a protected branch this only happens when a PR is merged.
+  # This is a double check in case the PR was stale and had some issues.
+  push:
+    branches:
+      - master
+    paths-ignore: # Prevents from running if only docs are updated
+      - 'doc/**'
+      - '**/*README*'
+      - '**.md'
+      - '**.rst'
+  pull_request:
+    paths-ignore: # Prevents from running if only docs are updated
+      - 'doc/**'
+      - '**/*README*'
+      - '**.md'
+      - '**.rst'
+  workflow_dispatch:
+  schedule:
+  - cron: '0 0 * * *' # daily
+
+# We want to cancel previous runs for a given PR or branch / ref if another CI
+# run is requested.
+# See: https://docs.github.com/en/actions/using-jobs/using-concurrency
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  cancel-in-progress: true
+
+env:
+  # default compiler for all non-compatibility tests
+  MATRIX_EVAL: "CC=gcc-11 && CXX=g++-11"
+
+jobs:
+  Run-tests:
+    # Prevents from running on forks where no custom runners are available
+    if: ${{ github.repository_owner == 'verilog-to-routing' }}
+
+    timeout-minutes: 420
+
+    container: ubuntu:jammy
+
+    runs-on: [self-hosted, Linux, X64]
+
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - {test: "vtr_reg_nightly_test1",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
+          - {test: "vtr_reg_nightly_test1_odin", cores: "16", options: "",          cmake: "-DWITH_ODIN=ON",                                               extra_pkgs: ""}
+          - {test: "vtr_reg_nightly_test2",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
+          - {test: "vtr_reg_nightly_test2_odin", cores: "16", options: "",          cmake: "-DWITH_ODIN=ON",                                               extra_pkgs: ""}
+          - {test: "vtr_reg_nightly_test3",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
+          - {test: "vtr_reg_nightly_test3_odin", cores: "16", options: "",          cmake: "-DWITH_ODIN=ON",                                               extra_pkgs: ""}
+          - {test: "vtr_reg_nightly_test4",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
+          - {test: "vtr_reg_nightly_test4_odin", cores: "16", options: "",          cmake: "-DWITH_ODIN=ON",                                               extra_pkgs: ""}
+          - {test: "vtr_reg_nightly_test5",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
+          - {test: "vtr_reg_nightly_test6",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
+          - {test: "vtr_reg_nightly_test7",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
+          - {test: "vtr_reg_strong",             cores: "16", options: "",          cmake: "-DVTR_ASSERT_LEVEL=3",                                         extra_pkgs: "libeigen3-dev"}
+          - {test: "vtr_reg_strong_odin",        cores: "16", options: "",          cmake: "-DVTR_ASSERT_LEVEL=3 -DWITH_ODIN=ON",                          extra_pkgs: "libeigen3-dev"}
+          - {test: "vtr_reg_strong_odin",        cores: "16", options: "-skip_qor", cmake: "-DVTR_ASSERT_LEVEL=3 -DVTR_ENABLE_SANITIZE=ON -DWITH_ODIN=ON", extra_pkgs: "libeigen3-dev"}
+          - {test: "vtr_reg_system_verilog",     cores: "16", options: "",          cmake: "-DYOSYS_F4PGA_PLUGINS=ON",                                     extra_pkgs: ""}
+          - {test: "odin_reg_strong",            cores: "16", options: "",          cmake: "-DWITH_ODIN=ON",                                               extra_pkgs: ""}
+          - {test: "parmys_reg_strong",          cores: "16", options: "",          cmake: "-DYOSYS_F4PGA_PLUGINS=ON",                                     extra_pkgs: ""}
+
+    env:
+      DEBIAN_FRONTEND: "noninteractive"
+
+    steps:
+
+      # TODO: This runnner is running on a self-hosted CPU. In order to upgrade
+      #       to v4, need to upgrade the machine to support node20.
+    - uses: actions/checkout@v3
+      with:
+        submodules: 'true'
+
+    - name: Setup
+      run: stdbuf -i0 -i0 -e0 ./.github/scripts/hostsetup.sh
+
+    - name: Install external libraries
+      run: apt install -y ${{ matrix.extra_pkgs }}
+      if: ${{ matrix.extra_pkgs }}
+
+    - name: Execute test script
+      run: stdbuf -i0 -o0 -e0 ./.github/scripts/run-vtr.sh
+      env:
+        VPR_NUM_WORKERS: 4
+        VTR_TEST: ${{ matrix.test }}
+        VTR_TEST_OPTIONS: ${{ matrix.options }}
+        VTR_CMAKE_PARAMS: ${{ matrix.cmake }}
+        NUM_CORES: ${{ matrix.cores }}
+
+    - name: Upload test results
+      # If the job was not cancelled, we want to save the result (this includes
+      # when the job fails). See warning here:
+      # https://docs.github.com/en/actions/learn-github-actions/expressions#always
+      if: ${{ !cancelled() }}
+      # TODO: This runnner is running on a self-hosted CPU. In order to upgrade
+      #       to v4, need to upgrade the machine to support node20.
+      uses: actions/upload-artifact@v3
+      with:
+        name: ${{matrix.test}}_test_results
+        path: |
+          **/results*.gz
+          **/plot_*.svg
+          **/qor_results*.tar.gz
+

.github/workflows/test.yml

@@ -34,83 +34,6 @@ env:
   MATRIX_EVAL: "CC=gcc-11 && CXX=g++-11"
 
 jobs:
-
-
-  Run-tests:
-    # Prevents from running on forks where no custom runners are available
-    if: ${{ github.repository_owner == 'verilog-to-routing' }}
-
-    timeout-minutes: 420
-
-    container: ubuntu:jammy
-
-    runs-on: [self-hosted, Linux, X64]
-
-    strategy:
-      fail-fast: false
-      matrix:
-        include:
-          - {test: "vtr_reg_nightly_test1",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
-          - {test: "vtr_reg_nightly_test1_odin", cores: "16", options: "",          cmake: "-DWITH_ODIN=ON",                                               extra_pkgs: ""}
-          - {test: "vtr_reg_nightly_test2",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
-          - {test: "vtr_reg_nightly_test2_odin", cores: "16", options: "",          cmake: "-DWITH_ODIN=ON",                                               extra_pkgs: ""}
-          - {test: "vtr_reg_nightly_test3",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
-          - {test: "vtr_reg_nightly_test3_odin", cores: "16", options: "",          cmake: "-DWITH_ODIN=ON",                                               extra_pkgs: ""}
-          - {test: "vtr_reg_nightly_test4",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
-          - {test: "vtr_reg_nightly_test4_odin", cores: "16", options: "",          cmake: "-DWITH_ODIN=ON",                                               extra_pkgs: ""}
-          - {test: "vtr_reg_nightly_test5",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
-          - {test: "vtr_reg_nightly_test6",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
-          - {test: "vtr_reg_nightly_test7",      cores: "16", options: "",          cmake: "",                                                             extra_pkgs: ""}
-          - {test: "vtr_reg_strong",             cores: "16", options: "",          cmake: "-DVTR_ASSERT_LEVEL=3",                                         extra_pkgs: "libeigen3-dev"}
-          - {test: "vtr_reg_strong_odin",        cores: "16", options: "",          cmake: "-DVTR_ASSERT_LEVEL=3 -DWITH_ODIN=ON",                          extra_pkgs: "libeigen3-dev"}
-          - {test: "vtr_reg_strong_odin",        cores: "16", options: "-skip_qor", cmake: "-DVTR_ASSERT_LEVEL=3 -DVTR_ENABLE_SANITIZE=ON -DWITH_ODIN=ON", extra_pkgs: "libeigen3-dev"}
-          - {test: "vtr_reg_system_verilog",     cores: "16", options: "",          cmake: "-DYOSYS_F4PGA_PLUGINS=ON",                                     extra_pkgs: ""}
-          - {test: "odin_reg_strong",            cores: "16", options: "",          cmake: "-DWITH_ODIN=ON",                                               extra_pkgs: ""}
-          - {test: "parmys_reg_strong",          cores: "16", options: "",          cmake: "-DYOSYS_F4PGA_PLUGINS=ON",                                     extra_pkgs: ""}
-
-    env:
-      DEBIAN_FRONTEND: "noninteractive"
-
-    steps:
-
-      # TODO: This runnner is running on a self-hosted CPU. In order to upgrade
-      #       to v4, need to upgrade the machine to support node20.
-    - uses: actions/checkout@v3
-      with:
-        submodules: 'true'
-
-    - name: Setup
-      run: stdbuf -i0 -i0 -e0 ./.github/scripts/hostsetup.sh
-
-    - name: Install external libraries
-      run: apt install -y ${{ matrix.extra_pkgs }}
-      if: ${{ matrix.extra_pkgs }}
-
-    - name: Execute test script
-      run: stdbuf -i0 -o0 -e0 ./.github/scripts/run-vtr.sh
-      env:
-        VPR_NUM_WORKERS: 4
-        VTR_TEST: ${{ matrix.test }}
-        VTR_TEST_OPTIONS: ${{ matrix.options }}
-        VTR_CMAKE_PARAMS: ${{ matrix.cmake }}
-        NUM_CORES: ${{ matrix.cores }}
-
-    - name: Upload test results
-      # If the job was not cancelled, we want to save the result (this includes
-      # when the job fails). See warning here:
-      # https://docs.github.com/en/actions/learn-github-actions/expressions#always
-      if: ${{ !cancelled() }}
-      # TODO: This runnner is running on a self-hosted CPU. In order to upgrade
-      #       to v4, need to upgrade the machine to support node20.
-      uses: actions/upload-artifact@v3
-      with:
-        name: ${{matrix.test}}_test_results
-        path: |
-          **/results*.gz
-          **/plot_*.svg
-          **/qor_results*.tar.gz
-
-
   Build:
     name: 'B: Building VtR'
     runs-on: ubuntu-22.04

BUILDING.md

@@ -46,7 +46,7 @@ If you download a different version of those tools, then those versions may not
 
 To verfiy that VTR has been installed correctly run::
 
-    ./vtr_flow/scripts/run_vtr_task.py regression_tests/vtr_reg_basic/basic_timing
+    ./vtr_flow/scripts/run_vtr_task.py ./vtr_flow/tasks/regression_tests/vtr_reg_basic/basic_timing
 
 The expected output is::
 

doc/src/api/vpr/grid.rst

@@ -0,0 +1,10 @@
+========
+Grid
+========
+
+DeviceGrid
+-------
+
+.. doxygenclass:: DeviceGrid
+   :project: vpr
+   :members:

doc/src/api/vpr/index.rst

@@ -7,6 +7,7 @@ VPR API
    :maxdepth: 1
 
    contexts
+   grid
    mapping
    netlist
    route_tree

doc/src/api/vpr/rr_graph.rst

@@ -35,3 +35,9 @@ RRSpatialLookup
    :project: librrgraph
    :members:
 
+rr_graph_utils
+---------------
+
+.. doxygenfile:: rr_graph_utils.h
+   :project: librrgraph
+   :sections: briefdescription detaileddescription func prototype user-defined public-func

doc/src/api/vprinternals/router_lookahead.rst

@@ -0,0 +1,35 @@
+==============
+Router Lookahead
+==============
+
+e_router_lookahead
+----------
+.. doxygenenum:: e_router_lookahead
+   :project: vpr
+
+
+RouterLookahead
+----------
+.. doxygenclass:: RouterLookahead
+   :project: vpr
+   :members:
+
+MapLookahead
+----------
+.. doxygenclass:: MapLookahead
+   :project: vpr
+
+make_router_lookahead
+----------
+.. doxygenfunction:: make_router_lookahead
+   :project: vpr
+
+get_cached_router_lookahead
+----------
+.. doxygenfunction:: get_cached_router_lookahead
+   :project: vpr
+
+invalidate_router_lookahead_cache
+----------
+.. doxygenfunction:: invalidate_router_lookahead_cache
+   :project: vpr

doc/src/api/vprinternals/vpr_router.rst

@@ -7,4 +7,5 @@ VPR Router
 .. toctree::
    :maxdepth: 1
 
-   router_heap
+   router_heap
+   router_lookahead

doc/src/vpr/command_line_usage.rst

@@ -1306,6 +1306,15 @@ The following options are only valid when the router is in timing-driven mode (t
 
     **Default:** ``1.2``
 
+.. option:: --astar_offset <float>
+
+    Sets how aggressive the directed search used by the timing-driven router is.
+    It is a subtractive adjustment to the lookahead heuristic.
+
+    Values between 0 and 1e-9 are resonable; higher values may increase quality at the expense of run-time.
+
+    **Default:** ``0.0``
+
 .. option:: --router_profiler_astar_fac <float>
 
     Controls the directedness of the timing-driven router's exploration when doing router delay profiling of an architecture.

libs/libarchfpga/src/device_grid.h

@@ -6,6 +6,7 @@
 #include <cmath>
 #include "vtr_ndmatrix.h"
 #include "physical_types.h"
+#include "vtr_geometry.h"
 
 /**
  * @brief s_grid_tile is the minimum tile of the fpga
@@ -73,6 +74,18 @@ class DeviceGrid {
         return grid_[tile_loc.layer_num][tile_loc.x][tile_loc.y].height_offset;
     }
 
+    ///@brief Returns a rectangle which represents the bounding box of the tile at the given location.
+    inline vtr::Rect<int> get_tile_bb(const t_physical_tile_loc& tile_loc) const {
+        t_physical_tile_type_ptr tile_type = get_physical_type(tile_loc);
+
+        int tile_xlow = tile_loc.x - get_width_offset(tile_loc);
+        int tile_ylow = tile_loc.y - get_height_offset(tile_loc);
+        int tile_xhigh = tile_xlow + tile_type->width - 1;
+        int tile_yhigh = tile_ylow + tile_type->height - 1;
+
+        return {{tile_xlow, tile_ylow}, {tile_xhigh, tile_yhigh}};
+    }
+
     ///@brief Return the metadata of the tile at the specified location
     inline const t_metadata_dict* get_metadata(const t_physical_tile_loc& tile_loc) const {
         return grid_[tile_loc.layer_num][tile_loc.x][tile_loc.y].meta;

libs/libarchfpga/src/physical_types.h

@@ -925,10 +925,10 @@ struct t_logical_block_type {
     std::vector<t_physical_tile_type_ptr> equivalent_tiles; ///>List of physical tiles at which one could
                                                             ///>place this type of netlist block.
 
-    std::unordered_map<int, t_pb_graph_pin*> pin_logical_num_to_pb_pin_mapping;                   /* pin_logical_num_to_pb_pin_mapping[pin logical number] -> pb_graph_pin ptr} */
-    std::unordered_map<const t_pb_graph_pin*, int> primitive_pb_pin_to_logical_class_num_mapping; /* primitive_pb_pin_to_logical_class_num_mapping[pb_graph_pin ptr] -> class logical number */
-    std::vector<t_class> primitive_logical_class_inf;                                             /* primitive_logical_class_inf[class_logical_number] -> class */
-    std::unordered_map<const t_pb_graph_node*, t_class_range> pb_graph_node_class_range;
+    std::unordered_map<int, t_pb_graph_pin*> pin_logical_num_to_pb_pin_mapping;                    /* pin_logical_num_to_pb_pin_mapping[pin logical number] -> pb_graph_pin ptr} */
+    std::unordered_map<const t_pb_graph_pin*, int> primitive_pb_pin_to_logical_class_num_mapping;  /* primitive_pb_pin_to_logical_class_num_mapping[pb_graph_pin ptr] -> class logical number */
+    std::vector<t_class> primitive_logical_class_inf;                                              /* primitive_logical_class_inf[class_logical_number] -> class */
+    std::unordered_map<const t_pb_graph_node*, t_class_range> primitive_pb_graph_node_class_range; /* primitive_pb_graph_node_class_range[primitive_pb_graph_node ptr] -> class range for that primitive*/
 
     // Is this t_logical_block_type empty?
     bool is_empty() const;
@@ -1240,6 +1240,12 @@ class t_pb_graph_node {
 
     int placement_index;
 
+    /*
+     * There is a root-level pb_graph_node assigned to each logical type. Each logical type can contain multiple primitives.
+     * If this pb_graph_node is associated with a primitive, a unique number is assigned to it within the logical block level.
+     */
+    int primitive_num = OPEN;
+
     /* Contains a collection of mode indices that cannot be used as they produce conflicts during VPR packing stage
      *
      * Illegal modes do arise when children of a graph_node do have inconsistent `edge_modes` with respect to

libs/libarchfpga/src/physical_types_util.cpp

@@ -965,7 +965,7 @@ t_class_range get_pb_graph_node_class_physical_range(t_physical_tile_type_ptr /*
                                                      const t_pb_graph_node* pb_graph_node) {
     VTR_ASSERT(pb_graph_node->is_primitive());
 
-    t_class_range class_range = logical_block->pb_graph_node_class_range.at(pb_graph_node);
+    t_class_range class_range = logical_block->primitive_pb_graph_node_class_range.at(pb_graph_node);
     int logical_block_class_offset = sub_tile->primitive_class_range[sub_tile_relative_cap].at(logical_block).low;
 
     class_range.low += logical_block_class_offset;

libs/librrgraph/src/base/check_rr_graph.cpp

@@ -9,6 +9,7 @@
 #include "physical_types_util.h"
 
 #include "describe_rr_node.h"
+#include "rr_graph_utils.h"
 
 /*********************** Subroutines local to this module *******************/
 
@@ -389,16 +390,29 @@ void check_rr_node(const RRGraphView& rr_graph,
 
     switch (rr_type) {
         case SOURCE:
-        case SINK:
             if (type == nullptr) {
                 VPR_FATAL_ERROR(VPR_ERROR_ROUTE,
                                 "in check_rr_node: node %d (type %d) is at an illegal clb location (%d, %d).\n", inode, rr_type, xlow, ylow);
             }
+
             if (xlow != (xhigh - type->width + 1) || ylow != (yhigh - type->height + 1)) {
                 VPR_FATAL_ERROR(VPR_ERROR_ROUTE,
                                 "in check_rr_node: node %d (type %d) has endpoints (%d,%d) and (%d,%d)\n", inode, rr_type, xlow, ylow, xhigh, yhigh);
             }
             break;
+        case SINK: {
+            if (type == nullptr) {
+                VPR_FATAL_ERROR(VPR_ERROR_ROUTE,
+                                "in check_rr_node: node %d (type %d) is at an illegal clb location (%d, %d).\n", inode, rr_type, xlow, ylow);
+            }
+
+            vtr::Rect<int> tile_bb = grid.get_tile_bb({xlow, ylow, layer_num});
+            if (xlow < tile_bb.xmin() || ylow < tile_bb.ymin() || xhigh > tile_bb.xmax() || yhigh > tile_bb.ymax()) {
+                VPR_FATAL_ERROR(VPR_ERROR_ROUTE,
+                                "in check_rr_node: node %d (type %d) has endpoints (%d,%d) and (%d,%d), which is outside the bounds of the grid tile containing it.\n", inode, rr_type, xlow, ylow, xhigh, yhigh);
+            }
+            break;
+        }
         case IPIN:
         case OPIN:
             if (type == nullptr) {