Rename dppl to dppy (#42)

Co-authored-by: etotmeni <elena.totmenina@intel.com>
Co-authored-by: Diptorup Deb <diptorup.deb@intel.com>

Author: 3 people

.gitignore

@@ -0,0 +1,27 @@
+*.pyc
+*.o
+*.so
+*.dylib
+*.pyd
+*.pdb
+*.egg-info
+*.sw[po]
+*.out
+*.ll
+.coverage
+.nfs*
+tags
+MANIFEST
+
+build/
+docs/_build/
+docs/gh-pages/
+dist/
+htmlcov/
+.idea/
+.vscode/
+.mypy_cache/
+.ipynb_checkpoints/
+__pycache__/
+
+docs/source/developer/autogen*

HowTo.rst

@@ -7,7 +7,7 @@ are listed below with the help of sample code snippets. In this release we have
 the implementation of the OAK approach described in MS138 in section 4.3.2. The
 new decorator is described below.
 
-To access the features driver module have to be imported from numba_dppy.dppl_driver
+To access the features driver module have to be imported from numba_dppy.dppy_driver
 
 New Decorator
 =============
@@ -61,7 +61,7 @@ Primitive types are passed by value to the kernel, currently supported are int,
 Math Kernels
 ============
 
-This release has support for math kernels. See numba_dppy/tests/dppl/test_math_functions.py
+This release has support for math kernels. See numba_dppy/tests/dppy/test_math_functions.py
 for more details.
 
 
@@ -170,6 +170,6 @@ Testing
 
 All examples can be found in numba_dppy/examples/
 
-All tests can be found in numba_dppy/tests/dppl and can be triggered by the following command:
+All tests can be found in numba_dppy/tests/dppy and can be triggered by the following command:
 
 ``python -m numba.runtests numba_dppy.tests``

numba_dppy/CHANGE_LOG

@@ -1,7 +1,7 @@
-NUMBA Version 0.48.0 + DPPL Version 0.3.0 (June 29, 2020)
+NUMBA Version 0.48.0 + DPPY Version 0.3.0 (June 29, 2020)
 --------------------------------------------------------
 
 This release includes:
 
-* Caching of dppl.kernels which will improve performance.
+* Caching of dppy.kernels which will improve performance.
 * Addition of support for Intel Advisor which will help in profiling applications.

numba_dppy/__init__.py

@@ -4,9 +4,9 @@
 
 
 Extensions to Numba for Intel GPUs introduce two new features into Numba:
-    a.  A new backend that has a new decorator called @dppl.kernel that
+    a.  A new backend that has a new decorator called @dppy.kernel that
         exposes an explicit kernel programming interface similar to the
-        existing Numba GPU code-generation backends. The @dppl.kernel
+        existing Numba GPU code-generation backends. The @dppy.kernel
         decorator currently implements a subset of OpenCL’s API through
         Numba’s intrinsic functions.
 
@@ -20,48 +20,48 @@
 Explicit Kernel Prgoramming with new Docorators:
 
 
-@dppl.kernel
+@dppy.kernel
 
-    The @dppl.kernel decorator can be used with or without extra arguments.
+    The @dppy.kernel decorator can be used with or without extra arguments.
     Optionally, users can pass the signature of the arguments to the
     decorator. When a signature is provided to the DK decorator the version
     of the OpenCL kernel generated gets specialized for that type signature.
 
     ---------------------------------------------------------------------------
-    @dppl.kernel
+    @dppy.kernel
     def data_parallel_sum(a, b, c):
-        i = dppl.get_global_id(0)
+        i = dppy.get_global_id(0)
         c[i] = a[i] + b[i]
     ---------------------------------------------------------------------------
 
     To invoke the above function users will need to provide a
     global size (OpenCL) which is the size of a (same as b and c) and a
-    local size (dppl.DEFAULT_LOCAL_SIZE if user don't want to specify).
+    local size (dppy.DEFAULT_LOCAL_SIZE if user don't want to specify).
     Example shown below:
 
     ---------------------------------------------------------------------------
-    data_parallel_sum[len(a), dppl.DEFAULT_LOCAL_SIZE](dA, dB, dC)
+    data_parallel_sum[len(a), dppy.DEFAULT_LOCAL_SIZE](dA, dB, dC)
     ---------------------------------------------------------------------------
 
 
-@dppl.func
+@dppy.func
 
-    The @dppl.func decorator is the other decorator provided in the explicit
+    The @dppy.func decorator is the other decorator provided in the explicit
     kernel programming model. This decorator allows users to write “device”
     functions that can be invoked from inside DK functions but cannot be invoked
     from the host. The decorator also supports type specialization as with the
-    DK decorator. Functions decorated with @dppl.func will also be JIT compiled
-    and inlined into the OpenCL Program containing the @dppl.kernel function
-    calling it. A @dppl.func will not be launched as an OpenCL kernel.
+    DK decorator. Functions decorated with @dppy.func will also be JIT compiled
+    and inlined into the OpenCL Program containing the @dppy.kernel function
+    calling it. A @dppy.func will not be launched as an OpenCL kernel.
 
     ---------------------------------------------------------------------------
-    @dppl.func
+    @dppy.func
     def bar(a):
         return a*a
 
-    @dppl.kernel
+    @dppy.kernel
     def foo(in, out):
-        i = dppl.get_global_id(0)
+        i = dppy.get_global_id(0)
         out[i] = bar(in[i])
     ---------------------------------------------------------------------------
 
@@ -71,13 +71,13 @@ def foo(in, out):
     The following table has the list of intrinsic functions that can be directly
     used inside a DK function. All the functions are equivalent to the similarly
     named OpenCL function. Wherever there is an implementation difference
-    between the Numba-PyDPPL version and the OpenCL version, the difference is
+    between the Numba-DPPY version and the OpenCL version, the difference is
     explained in table. Note that these functions cannot be used anywhere else
     outside of a DK function in a Numba application. Readers are referred to the
     OpenCL API specs to review the functionality of each function.
 
     +----------------------+----------------------------+----------------------+
-    | Numba-DPPL intrinsic | Equivalent OpenCL function |         Notes        |
+    | Numba-DPPY intrinsic | Equivalent OpenCL function |         Notes        |
     +----------------------+----------------------------+----------------------+
     | get_global_id        | get_global_id              |                      |
     +----------------------+----------------------------+----------------------+
@@ -121,7 +121,7 @@ def foo(in, out):
     |print             |print(varargs)                 |The print function is a  |
     |                  |                               |subset of the OpenCL     |
     |                  |                               |printf function. The     |
-    |                  |                               |Numba-DPPL version of    |
+    |                  |                               |Numba-DPPY version of    |
     |                  |                               |print supports only int, |
     |                  |                               |string, and float        |
     |                  |                               |arguments.               |
@@ -160,16 +160,16 @@ def foo(in, out):
 
 
 
-Complete Example using @dppl.kernel:
+Complete Example using @dppy.kernel:
 
     ---------------------------------------------------------------------------
     import numpy as np
-    import numba_dppy, numba_dppy as dppl
+    import numba_dppy, numba_dppy as dppy
     import dpctl
 
-    @dppl.kernel
+    @dppy.kernel
     def data_parallel_sum(a, b, c):
-        i = dppl.get_global_id(0)
+        i = dppy.get_global_id(0)
         c[i] = a[i] + b[i]
 
     def driver(device_env, a, b, c, global_size):
@@ -181,7 +181,7 @@ def driver(device_env, a, b, c, global_size):
         print("before : ", dA._ndarray)
         print("before : ", dB._ndarray)
         print("before : ", dC._ndarray)
-        data_parallel_sum[global_size, dppl.DEFAULT_LOCAL_SIZE](dA, dB, dC)
+        data_parallel_sum[global_size, dppy.DEFAULT_LOCAL_SIZE](dA, dB, dC)
         device_env.copy_array_from_device(dC)
         print("after : ", dC._ndarray)
 
@@ -509,11 +509,11 @@ def main():
 if dppy_present:
     from .device_init import *
 else:
-    raise ImportError("Importing dppl failed")
+    raise ImportError("Importing numba-dppy failed")
 
 def test(*args, **kwargs):
     if not dppy_present and not is_available():
-        dppl_error()
+        dppy_error()
 
     return numba.testing.test("numba_dppy.tests", *args, **kwargs)