Merging in remote changes

hpvm/docs/components/index.rst

@@ -7,7 +7,7 @@ HPVM consists of a few relatively independent key components.
 * HPVM code generator: a few ``opt`` passes that lowers HPVM IR to LLVM IR,
   which is then compiled into object code and binary.
 
-:doc:`Compilation process of HPVM </references/hpvm-specification>`
+:doc:`Compilation process of HPVM </specifications/hpvm-spec>`
 shows how these 2 components work together.
 In addition, there are:
 

hpvm/docs/conf.py

@@ -98,7 +98,7 @@ html_theme_options = {
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,
 # so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = []
+html_static_path = ["_static"]
 
 # If not '', a 'Last updated on:' timestamp is inserted at every page bottom,
 # using the given strftime format.

hpvm/docs/developerdocs/index.rst

@@ -6,4 +6,6 @@ Developer Documents
 
    approximation-implementation
    cnn-models
+   compilation-process
    configuration-format
+   port-to-hpvm-c

hpvm/docs/developerdocs/port-to-hpvm-c.rst

+Porting a Program from C to HPVM-C
+==================================
+
+The following represents the required steps to port a regular C program into an HPVM program with HPVM-C. These steps are described at a high level; for more detail, please see `hpvm-cava </hpvm/test/benchmarks/hpvm-cava>`_ provided in `benchmarks </hpvm/test/benchmarks>`_.
+
+* Separate the computation that will become a kernel into its own (leaf node) function and add the attributes and target hint.
+* Create a level 1 wrapper node function that will describe the thread-level parallelism (for the GPU). The node will:
+
+  * Use the ``createNode[ND]()`` method to create a kernel node and specify how many threads will execute it.
+  * Bind its arguments to the kernel arguments.
+
+* If desired, create a level 2 wrapper node function which will describe the threadblock-level parallalism (for the GPU). This node will:
+
+  * Use the ``createNode[ND]()`` method to create a level 1 wrapper node and specify how many threadblocks will execute it.
+  * Bind its arguments to its child node's arguments.
+
+* A root node function that creates all the top-level wrapper nodes, binds their arguments, and connects their edges.
+
+  * Each root node represents a DFG.
+
+* All the above node functions have the combined arguments of all the kernels that are nested at each level. 
+* The host code will have to include the following:
+
+  * Initialize the HPVM runtime using the ``init()`` method.
+  * Create an argument struct for each DFG and assign its member variables.
+  * Add all the memory that is required by the kernel into the memory tracker.
+  * Launch the DFG by calling the ``launch()`` method on the root node function, and passing the corresponding argument struct.
+  * Wait for the DFG to complete execution.
+  * Read out any generated memory using the ``request_mem()`` method.
+  * Remove all the tracked memory from the memory tracker.

hpvm/docs/gallery.rst

+Gallery
+=======
+
+This gallery contains example tradeoff curves for the 10 DNN benchmarks in HPVM.
+The performances shown are measured on Jetson TX2;
+these numbers are close to what is shown in `ApproxTuner <https://dl.acm.org/doi/10.1145/3437801.3446108>`_,
+with small differences due to variation in autotuning and profiling.
+
+.. list-table:: Tradeoff curves of 9 benchmark DNN networks.
+   :widths: 30 30
+   :header-rows: 0
+
+   * - .. image:: _static/lenet_mnist.png
+          :target: _static/lenet_mnist.png
+     - .. image:: _static/alexnet2_cifar10.png
+          :target: _static/alexnet2_cifar10.png
+   * - .. image:: _static/alexnet_cifar10.png
+          :target: _static/alexnet_cifar10.png
+     - .. image:: _static/alexnet_imagenet.png
+          :target: _static/alexnet_imagenet.png
+   * - .. image:: _static/vgg16_cifar10.png
+          :target: _static/vgg16_cifar10.png
+     - .. image:: _static/vgg16_cifar100.png
+          :target: _static/vgg16_cifar100.png
+   * - .. image:: _static/resnet18_cifar10.png
+          :target: _static/resnet18_cifar10.png
+     - .. image:: _static/resnet50_imagenet.png
+          :target: _static/resnet50_imagenet.png
+   * - .. image:: _static/mobilenet_cifar10.png
+          :target: _static/mobilenet_cifar10.png
+     -

hpvm/docs/getting-started.rst

@@ -222,7 +222,7 @@ while ``configs_profiled.png`` shows the final performance-accuracy tradeoff cur
 
 An example of ``configs_profiled.png`` looks like this (proportion of your image may be different):
 
-.. image:: tradeoff-curves/alexnet2_cifar10.png
+.. image:: _static/alexnet2_cifar10.png
 
 -----------------------
 

hpvm/docs/index.rst

@@ -52,8 +52,9 @@ Documentation
    getting-started
    tests
    components/index
-   references/index
+   specifications/index
    developerdocs/index
+   gallery
 
 Indices and tables
 ------------------

hpvm/docs/install.rst

@@ -19,12 +19,19 @@ The following components are required to be installed on your machine to build H
 
 Python must be strictly 3.6 (any subversion from 3.6.0 to 3.6.13).
 Alternatively, if you use Anaconda for package management,
-we provide a conda environment file that covers all Python and package requirements:
+we provide a conda environment file that covers all Python and package requirements
+(``hpvm/env.yaml`` can be found in the repository):
 
 .. code-block:: bash
 
    conda env create -n hpvm -f hpvm/env.yaml
 
+This creates the conda environment ``hpvm``.
+If you use this method, remember to activate the environment each time you enter a bash shell:
+
+.. code-block:: bash
+
+   conda activate hpvm
 
 Supported Architectures
 -----------------------

hpvm/docs/references/index.rst

-References
-============
-
-Below are some technical details of HPVM system and the HPVM-C language.
-
-.. toctree::
-   :maxdepth: 1
-
-   hpvm-c
-   hpvm-specification
-   compilation-process

hpvm/docs/references/hpvm-c.rst → hpvm/docs/specifications/hpvm-c-spec.rst

@@ -117,35 +117,3 @@ Atomically computes the bitwise XOR of ``v`` and the value stored at memory loca
 
 ``void __hpvm__barrier()``:raw-html-m2r:`<br>`
 Local synchronization barrier across dynamic instances of current leaf node.
-
-Porting a Program from C to HPVM-C
-==================================
-
-The following represents the required steps to port a regular C program into an HPVM program with HPVM-C. These steps are described at a high level; for more detail, please see `hpvm-cava </hpvm/test/benchmarks/hpvm-cava>`_ provided in `benchmarks </hpvm/test/benchmarks>`_.
-
-
-* Separate the computation that will become a kernel into its own (leaf node) function and add the attributes and target hint.
-* Create a level 1 wrapper node function that will describe the thread-level parallelism (for the GPU). The node will:
-
-  * Use the ``createNode[ND]()`` method to create a kernel node and specify how many threads will execute it.
-  * Bind its arguments to the kernel arguments.
-
-* If desired, create a level 2 wrapper node function which will describe the threadblock-level parallalism (for the GPU). This node will:
-
-  * Use the ``createNode[ND]()`` method to create a level 1 wrapper node and specify how many threadblocks will execute it.
-  * Bind its arguments to its child node's arguments.
-
-* A root node function that creates all the top-level wrapper nodes, binds their arguments, and connects their edges.
-
-  * Each root node represents a DFG.
-
-* All the above node functions have the combined arguments of all the kernels that are nested at each level. 
-* The host code will have to include the following:
-
-  * Initialize the HPVM runtime using the ``init()`` method.
-  * Create an argument struct for each DFG and assign its member variables.
-  * Add all the memory that is required by the kernel into the memory tracker.
-  * Launch the DFG by calling the ``launch()`` method on the root node function, and passing the corresponding argument struct.
-  * Wait for the DFG to complete execution.
-  * Read out any generated memory using the ``request_mem()`` method.
-  * Remove all the tracked memory from the memory tracker.