update the examples for SDK version 1.3.0

README.rst

@@ -57,20 +57,20 @@ The sample applications available are:
   (GEMM) using the ``collectives`` library.
 * ``residual``: Computes the norm of the residual of a matrix-vector
   multiplication. Builds on the ``gemv-checkerboard-pattern`` example.
-* ``stencil-v2``: A 3D 25-point stencil finite difference code for solving a
+* ``25-pt-stencil``: A 3D 25-point stencil finite difference code for solving a
   wave equation with a source perturbation.
-* ``bandwidthTest``: Benchmarks the bandwidth of data transfers between host
+* ``bandwidth-test``: Benchmarks the bandwidth of data transfers between host
   and device using the ``memcpy`` framework and the ``SdkRuntime`` host API.
 * ``spmv-hypersparse``: Computes a sparse matrix-vector product using a
   hypersparse matrix.
-* ``stencil-3d-7pts``: Computes a sparse matrix-vector product using a matrix
-  generated by a 7-point stencil.
-* ``powerMethod``: Implements the Power method to compute the eigenvector
+* ``7pt-stencil-spmv``: Computes a sparse matrix-vector product using a matrix
+  generated by a 3D 7-point stencil.
+* ``power-method``: Implements the Power method to compute the eigenvector
   of the largest eigenvalue of a matrix generated by a 7-point stencil.
-* ``conjugateGradient``: Implements the Conjugate Gradient (CG) method to 
+* ``conjugate-gradient``: Implements the Conjugate Gradient (CG) method to 
   approximate the solution to a system of linear equations ``A*x = b``,
   where ``A`` is a matrix generated by a 7-point stencil.
-* ``preconditionedConjugateGradient``: Implements the Preconditioned Conjugate
+* ``preconditioned-conjugate-gradient``: Implements the Preconditioned Conjugate
   Gradient method (PCG) to approximate the solution to a system of linear
   equations ``A*x = b``, where ``A`` is a matrix generated by a 7-point
   stencil.
@@ -89,15 +89,20 @@ The sample applications available are:
 * ``FFT``: Implements 1D and 2D Discrete Fourier Transforms (DFT).
 * ``single-tile-matvec``: Implements highly optimized ``N x N`` matrix-vector
   products, in which each PE performs the same matrix-vector computation.
+* ``row-col-broadcast``: Benchmarks the bandwidth of data transfers between
+  host and device, where data is broadcast across a row or column of PEs,
+  using ``memcpy_h2d_colbcast`` and ``memcpy_h2d_rowbcast``.
+* ``game-of-life``: Implements Conway's Game of Life, where each PE is treated
+  as a single cell.
 
 Branches
 --------
 
 For each release of the SDK, there is a corresponding release tag in this
 repository which contains a version of the CSL examples which are compatible
-with that SDK release. For example, the tag ``rel-sdk-1.2.0`` in this
+with that SDK release. For example, the tag ``rel-sdk-1.3.0`` in this
 repository contains a version of the CSL examples which will work (compile and
-simulate) with the SDK 1.2.0 release. The ``master`` branch is identical to the
+simulate) with the SDK 1.3.0 release. The ``master`` branch is identical to the
 newest release.
 
 Full backward compatibility of the SDK is not guaranteed.

RELEASE-NOTES.rst

@@ -4,6 +4,19 @@ Release Notes
 The following are the release notes for the CSL Examples repository,
 ``csl-examples``.
 
+Version 1.3.0
+-------------
+
+- The examples are improved and updated to comply with the SDK version 1.3.0.
+
+- A new example program ``row-col-broadcast`` has been introduced which
+  benchmarks the bandwidth of data transfers between host and device,
+  where data is broadcast across a row or column of PEs,
+  using the new ``memcpy_h2d_colbcast`` and ``memcpy_h2d_rowbcast`` APIs.
+
+- A new example program ``game-of-life`` has been introduced which implements
+  Conway's Game of Life, where each PE is treated as a single cell.
+
 Version 1.2.0
 -------------
 

benchmarks/stencil-3d-7pts/README.rst → benchmarks/7pt-stencil-spmv/README.rst

@@ -1,5 +1,5 @@
-stencil-3d-7pts
-===============
+3D 7-Point Stencil SpMV
+=======================
 
 This example evaluates the performance of 7-point stencil. The kernel records
 the ``start`` and ``end`` of ``spmv`` by tsc counter. In addition the tsc

benchmarks/stencil-3d-7pts/cmd_parser.py → benchmarks/7pt-stencil-spmv/cmd_parser.py

@@ -47,6 +47,8 @@ def parse_args():
       "-n",
       default=1, type=int,
       help="number of columns")
+  parser.add_argument("--simulator", action="store_true",
+                      help="Runs on simulator")
   parser.add_argument(
       "-k",
       default=1, type=int,
@@ -74,10 +76,9 @@ def parse_args():
   parser.add_argument(
       "--run-only",
       help="Run only", action="store_true")
-  # arch = wse1 or wse2
   parser.add_argument(
       "--arch",
-      help="wse1 or wse2. Default is wse1 when not supplied.")
+      help="wse2 or wse3. Default is wse2 when not supplied.")
   parser.add_argument(
       "--width-west-buf",
       default=0, type=int,
@@ -108,4 +109,7 @@ def parse_args():
     print(f"create {logs_dir} to store log files")
     os.mkdir(logs_dir)
 
+  if args.cmaddr is None:
+      args.simulator = False
+
   return args, logs_dir

benchmarks/stencil-3d-7pts/commands.sh → benchmarks/7pt-stencil-spmv/commands_wse2.sh

@@ -2,7 +2,7 @@
 
 set -e
 
-cslc ./layout.csl --arch wse2 --fabric-dims=12,7 --fabric-offsets=4,1 \
+cslc ./src/layout.csl --arch wse2 --fabric-dims=12,7 --fabric-offsets=4,1 \
 --params=width:5,height:5,MAX_ZDIM:5 --params=BLOCK_SIZE:2 --params=C0_ID:0 \
 --params=C1_ID:1 --params=C2_ID:2 --params=C3_ID:3 --params=C4_ID:4 --params=C5_ID:5 \
 --params=C6_ID:6 --params=C7_ID:7 --params=C8_ID:8 -o=out \

benchmarks/7pt-stencil-spmv/commands_wse3.sh

@@ -0,0 +1,11 @@
+#!/usr/bin/env bash
+
+set -e
+
+cslc ./src/layout.csl --arch wse3 --fabric-dims=12,7 --fabric-offsets=4,1 \
+--params=width:5,height:5,MAX_ZDIM:5 --params=BLOCK_SIZE:2 --params=C0_ID:0 \
+--params=C1_ID:1 --params=C2_ID:2 --params=C3_ID:3 --params=C4_ID:4 --params=C5_ID:5 \
+--params=C6_ID:6 --params=C7_ID:7 --params=C8_ID:8 -o=out \
+--memcpy --channels=1 --width-west-buf=0 --width-east-buf=0
+cs_python ./run.py -m=5 -n=5 -k=5 --latestlink out --channels=1 \
+--width-west-buf=0 --width-east-buf=0 --zDim=5 --run-only