Merge branch 'master' into impl-kaiser

Author: vtavana

.github/workflows/array-api-skips.txt

@@ -6,3 +6,9 @@ array_api_tests/test_operators_and_elementwise_functions.py::test_clip
 # unexpected result is returned - unmute when dpctl-1986 is resolved
 array_api_tests/test_operators_and_elementwise_functions.py::test_asin
 array_api_tests/test_operators_and_elementwise_functions.py::test_asinh
+
+# advanced indexing relating issues (waiting a fix from dpctl)
+array_api_tests/test_array_object.py::test_getitem_arrays_and_ints_1[1]
+array_api_tests/test_array_object.py::test_getitem_arrays_and_ints_1[None]
+array_api_tests/test_array_object.py::test_getitem_arrays_and_ints_2[1]
+array_api_tests/test_array_object.py::test_getitem_arrays_and_ints_2[None]

.github/workflows/check-mkl-interfaces.yaml

@@ -48,7 +48,7 @@ jobs:
 
       - name: Merge conda env files
         run: |
-          conda-merge ${{ env.build-with-oneapi-env }} ${{ env.dpctl-pkg-env }} ${{ env.oneapi-pkgs-env }} > ${{ env.environment-file }}
+          conda-merge ${{ env.dpctl-pkg-env }} ${{ env.oneapi-pkgs-env }} ${{ env.build-with-oneapi-env }} > ${{ env.environment-file }}
           cat ${{ env.environment-file }}
 
       - name: Upload artifact

.github/workflows/conda-package.yml

@@ -93,13 +93,13 @@ jobs:
         continue-on-error: true
         run: conda build --no-test --python ${{ matrix.python }} --numpy 2.0 ${{ env.channels-list }} conda-recipe
         env:
-          MAX_BUILD_CMPL_MKL_VERSION: '2025.1a0'
+          MAX_BUILD_CMPL_MKL_VERSION: '2025.2a0'
 
       - name: ReBuild conda package
         if: steps.build_conda_pkg.outcome == 'failure'
         run: conda build --no-test --python ${{ matrix.python }} --numpy 2.0 ${{ env.channels-list }} conda-recipe
         env:
-          MAX_BUILD_CMPL_MKL_VERSION: '2025.1a0'
+          MAX_BUILD_CMPL_MKL_VERSION: '2025.2a0'
 
       - name: Upload artifact
         uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
@@ -625,6 +625,7 @@ jobs:
           python -m pytest --json-report --json-report-file=${{ env.json-report-file }} --disable-deadline --skips-file ${{ env.array-api-skips-file }} array_api_tests || true
         env:
           ARRAY_API_TESTS_MODULE: 'dpnp'
+          ARRAY_API_TESTS_VERSION: '2024.12'
           SYCL_CACHE_PERSISTENT: 1
         working-directory: ${{ env.array-api-tests-path }}
 

.github/workflows/openssf-scorecard.yml

@@ -68,6 +68,6 @@ jobs:
 
       # Upload the results to GitHub's code scanning dashboard.
       - name: "Upload to code-scanning"
-        uses: github/codeql-action/upload-sarif@5f8171a638ada777af81d42b55959a643bb29017 # v3.28.12
+        uses: github/codeql-action/upload-sarif@1b549b9259bda1cb5ddde3b41741a82a2d15a841 # v3.28.13
         with:
           sarif_file: results.sarif

.github/workflows/pre-commit.yml

@@ -29,7 +29,7 @@ jobs:
         uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
 
       - name: Set up python
-        uses: actions/setup-python@42375524e23c412d93fb67b49958b491fce71c38 # v5.4.0
+        uses: actions/setup-python@8d9ed9ac5c53483de85588cdf95a591a75ab9f55 # v5.5.0
         with:
           python-version: '3.13'
 

.pre-commit-config.yaml

@@ -2,13 +2,13 @@
 # See https://pre-commit.com/hooks.html for more hooks
 repos:
 -   repo: https://github.com/PyCQA/bandit
-    rev: '1.7.9'
+    rev: '1.8.3'
     hooks:
     -   id: bandit
         pass_filenames: false
         args: ["-r", "dpnp", "-lll"]
 -   repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.6.0
+    rev: v5.0.0
     hooks:
     -   id: check-ast
     -   id: check-builtin-literals
@@ -44,17 +44,18 @@ repos:
     -   id: rst-inline-touching-normal
     -   id: text-unicode-replacement-char
 -   repo: https://github.com/codespell-project/codespell
-    rev: v2.3.0
+    rev: v2.4.1
     hooks:
     -   id: codespell
+        args: ["-L", "abd"]  # ignore "abd" used in einsum tests
         additional_dependencies:
             - tomli
 -   repo: https://github.com/psf/black
-    rev: 24.4.2
+    rev: 25.1.0
     hooks:
     -   id: black
 -   repo: https://github.com/pycqa/isort
-    rev: 5.13.2
+    rev: 6.0.1
     hooks:
     -   id: isort
         name: isort (python)
@@ -65,20 +66,20 @@ repos:
         name: isort (pyi)
         types: [pyi]
 -   repo: https://github.com/pycqa/flake8
-    rev: 7.1.0
+    rev: 7.1.2
     hooks:
     -   id: flake8
         args: ["--config=.flake8"]
         additional_dependencies:
             - flake8-docstrings==1.7.0
-            - flake8-bugbear==24.4.26
+            - flake8-bugbear==24.12.12
 -   repo: https://github.com/pocc/pre-commit-hooks
     rev: v1.3.5
     hooks:
     -   id: clang-format
         args: ["-i"]
 -   repo: https://github.com/gitleaks/gitleaks
-    rev: v8.18.4
+    rev: v8.24.0
     hooks:
     -   id: gitleaks
 -   repo: https://github.com/jumanjihouse/pre-commit-hooks
@@ -102,3 +103,8 @@ repos:
             "--disable=unused-wildcard-import"
             ]
         files: '^dpnp/(dpnp_iface.*|fft|linalg)'
+-   repo: https://github.com/macisamuele/language-formatters-pre-commit-hooks
+    rev: v2.14.0
+    hooks:
+    -   id: pretty-format-toml
+        args: [--autofix]

CHANGELOG.md

@@ -16,14 +16,16 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Changed
 
-* Extended `dpnp.fft.fftfreq` and `dpnp.fft.rfftfreq` functions to support `dtype` keyword per Python Array API spec 2024.12 [#2384](https://github.com/IntelPython/dpnp/pull/2384)
+* Improved performance of `dpnp.nansum`, `dpnp.nanprod`, `dpnp.nancumsum`, and `dpnp.nancumprod` by reusing `dpnp.nan_to_num` function in implementation of the functions [#2339](https://github.com/IntelPython/dpnp/pull/2339)
 * Allowed input array of `uint64` dtype in `dpnp.bincount` [#2361](https://github.com/IntelPython/dpnp/pull/2361)
 * The vector norms `ord={None, 1, 2, inf}` and the matrix norms `ord={None, 1, 2, inf, "fro", "nuc"}` now consistently return zero for empty arrays, which are arrays with at least one axis of size zero. This change affects `dpnp.linalg.norm`, `dpnp.linalg.vector_norm`, and `dpnp.linalg.matrix_norm`. Previously, dpnp would either raise errors or return zero depending on the parameters provided [#2371](https://github.com/IntelPython/dpnp/pull/2371)
-* Improved performance of `dpnp.nansum`, `dpnp.nanprod`, `dpnp.nancumsum`, and `dpnp.nancumprod` by reusing `dpnp.nan_to_num` function in implementation of the functions [#2339](https://github.com/IntelPython/dpnp/pull/2339)
+* Extended `dpnp.fft.fftfreq` and `dpnp.fft.rfftfreq` functions to support `dtype` keyword per Python Array API spec 2024.12 [#2384](https://github.com/IntelPython/dpnp/pull/2384)
+* Updated `dpnp.fix` to return output with the same data-type of input [#2392](https://github.com/IntelPython/dpnp/pull/2392)
 
 ### Fixed
 
 * Resolved an issue with an incorrect result returned due to missing dependency from the strided kernel on a copy event in `dpnp.erf` [#2378](https://github.com/IntelPython/dpnp/pull/2378)
+* Updated `conda create` commands build and install instructions of `Quick start guide` to avoid a compilation error [#2395](https://github.com/IntelPython/dpnp/pull/2395)
 
 
 ## [0.17.0] - 02/26/2025

doc/quick_start_guide.rst

@@ -88,14 +88,18 @@ On Linux:
 
 .. code-block:: bash
 
-    conda create -n build-env dpctl cython dpcpp_linux-64 mkl-devel-dpcpp tbb-devel onedpl-devel cmake scikit-build ninja pytest -c https://software.repos.intel.com/python/conda/ -c conda-forge
+    conda create -n build-env dpctl cython dpcpp_linux-64 mkl-devel-dpcpp tbb-devel \
+          onedpl-devel cmake scikit-build ninja pytest intel-gpu-ocl-icd-system     \
+          -c dppy/label/dev -c https://software.repos.intel.com/python/conda/ -c conda-forge
     conda activate build-env
 
 On Windows:
 
 .. code-block:: bash
 
-    conda create -n build-env dpctl cython dpcpp_win-64 mkl-devel-dpcpp tbb-devel onedpl-devel cmake scikit-build ninja pytest -c https://software.repos.intel.com/python/conda/ -c conda-forge
+    conda create -n build-env dpctl cython dpcpp_win-64 mkl-devel-dpcpp tbb-devel \
+          onedpl-devel cmake scikit-build ninja pytest intel-gpu-ocl-icd-system   \
+          -c dppy/label/dev -c https://software.repos.intel.com/python/conda/ -c conda-forge
     conda activate build-env
 
 To build and install the package on Linux OS, run:

dpnp/backend/extensions/blas/blas_py.cpp

@@ -62,8 +62,7 @@ PYBIND11_MODULE(_blas_impl, m)
     using event_vecT = std::vector<sycl::event>;
 
     {
-        dot_ns::init_dot_dispatch_vector<dot_impl_fn_ptr_t,
-                                         blas_ns::DotContigFactory>(
+        dot_ns::init_dot_dispatch_vector<blas_ns::DotContigFactory>(
             dot_dispatch_vector);
 
         auto dot_pyapi = [&](sycl::queue &exec_q, const arrayT &src1,
@@ -81,8 +80,7 @@ PYBIND11_MODULE(_blas_impl, m)
     }
 
     {
-        dot_ns::init_dot_dispatch_vector<dot_impl_fn_ptr_t,
-                                         blas_ns::DotcContigFactory>(
+        dot_ns::init_dot_dispatch_vector<blas_ns::DotcContigFactory>(
             dotc_dispatch_vector);
 
         auto dotc_pyapi = [&](sycl::queue &exec_q, const arrayT &src1,
@@ -101,8 +99,7 @@ PYBIND11_MODULE(_blas_impl, m)
     }
 
     {
-        dot_ns::init_dot_dispatch_vector<dot_impl_fn_ptr_t,
-                                         blas_ns::DotuContigFactory>(
+        dot_ns::init_dot_dispatch_vector<blas_ns::DotuContigFactory>(
             dotu_dispatch_vector);
 
         auto dotu_pyapi = [&](sycl::queue &exec_q, const arrayT &src1,

dpnp/backend/extensions/blas/dot_common.hpp

@@ -50,14 +50,13 @@ typedef sycl::event (*dot_impl_fn_ptr_t)(sycl::queue &,
 namespace dpctl_td_ns = dpctl::tensor::type_dispatch;
 namespace py = pybind11;
 
-template <typename dispatchT>
 std::pair<sycl::event, sycl::event>
     dot_func(sycl::queue &exec_q,
              const dpctl::tensor::usm_ndarray &vectorX,
              const dpctl::tensor::usm_ndarray &vectorY,
              const dpctl::tensor::usm_ndarray &result,
              const std::vector<sycl::event> &depends,
-             const dispatchT &dot_dispatch_vector)
+             const dot_impl_fn_ptr_t *dot_dispatch_vector)
 {
     const int vectorX_nd = vectorX.get_ndim();
     const int vectorY_nd = vectorY.get_ndim();
@@ -166,12 +165,10 @@ std::pair<sycl::event, sycl::event>
     return std::make_pair(args_ev, dot_ev);
 }
 
-template <typename dispatchT,
-          template <typename fnT, typename T>
-          typename factoryT>
-void init_dot_dispatch_vector(dispatchT dot_dispatch_vector[])
+template <template <typename fnT, typename T> typename factoryT>
+void init_dot_dispatch_vector(dot_impl_fn_ptr_t dot_dispatch_vector[])
 {
-    dpctl_td_ns::DispatchVectorBuilder<dispatchT, factoryT,
+    dpctl_td_ns::DispatchVectorBuilder<dot_impl_fn_ptr_t, factoryT,
                                        dpctl_td_ns::num_types>
         contig;
     contig.populate_dispatch_vector(dot_dispatch_vector);

dpnp/backend/extensions/window/common.hpp

@@ -67,12 +67,11 @@ sycl::event window_impl(sycl::queue &q,
     return window_ev;
 }
 
-template <typename dispatchT>
 std::pair<sycl::event, sycl::event>
     py_window(sycl::queue &exec_q,
               const dpctl::tensor::usm_ndarray &result,
               const std::vector<sycl::event> &depends,
-              const dispatchT &window_dispatch_vector)
+              const window_fn_ptr_t *window_dispatch_vector)
 {
     dpctl::tensor::validation::CheckWritable::throw_if_not_writable(result);
 

dpnp/dpnp_algo/dpnp_elementwise_common.py

@@ -41,6 +41,7 @@
     "DPNPI0",
     "DPNPAngle",
     "DPNPBinaryFunc",
+    "DPNPFix",
     "DPNPImag",
     "DPNPReal",
     "DPNPRound",
@@ -511,6 +512,48 @@ def __call__(self, x, deg=False, out=None, order="K"):
         return res
 
 
+class DPNPFix(DPNPUnaryFunc):
+    """Class that implements dpnp.fix unary element-wise functions."""
+
+    def __init__(
+        self,
+        name,
+        result_type_resolver_fn,
+        unary_dp_impl_fn,
+        docs,
+    ):
+        super().__init__(
+            name,
+            result_type_resolver_fn,
+            unary_dp_impl_fn,
+            docs,
+        )
+
+    def __call__(self, x, out=None, order="K"):
+        if not dpnp.is_supported_array_type(x):
+            pass  # pass to raise error in main implementation
+        elif dpnp.issubdtype(x.dtype, dpnp.inexact):
+            pass  # for inexact types, pass to calculate in the backend
+        elif out is not None and not dpnp.is_supported_array_type(out):
+            pass  # pass to raise error in main implementation
+        elif out is not None and out.dtype != x.dtype:
+            # passing will raise an error but with incorrect needed dtype
+            raise ValueError(
+                f"Output array of type {x.dtype} is needed, got {out.dtype}"
+            )
+        else:
+            # for exact types, return the input
+            if out is None:
+                return dpnp.copy(x, order=order)
+
+            if isinstance(out, dpt.usm_ndarray):
+                out = dpnp_array._create_from_usm_ndarray(out)
+            out[...] = x
+            return out
+
+        return super().__call__(x, out=out, order=order)
+
+
 class DPNPI0(DPNPUnaryFunc):
     """Class that implements dpnp.i0 unary element-wise functions."""
 

dpnp/dpnp_iface_linearalgebra.py

@@ -418,7 +418,7 @@ def einsum(
     >>> a = np.ones(64000).reshape(20, 40, 80)
 
     Basic `einsum`: 119 ms ± 26 ms per loop (evaluated on 12th
-    Gen Intel\u00AE Core\u2122 i7 processor)
+    Gen Intel\u00ae Core\u2122 i7 processor)
 
     >>> %timeit np.einsum("ijk,ilm,njm,nlk,abc->",a, a, a, a, a)
 

dpnp/dpnp_iface_mathematical.py

@@ -63,6 +63,7 @@
     DPNPI0,
     DPNPAngle,
     DPNPBinaryFunc,
+    DPNPFix,
     DPNPImag,
     DPNPReal,
     DPNPRound,
@@ -1728,14 +1729,14 @@ def ediff1d(ary, to_end=None, to_begin=None):
 Round to nearest integer towards zero.
 
 Round an array of floats element-wise to nearest integer towards zero.
-The rounded values are returned as floats.
+The rounded values have the same data-type as the input.
 
 For full documentation refer to :obj:`numpy.fix`.
 
 Parameters
 ----------
 x : {dpnp.ndarray, usm_ndarray}
-    An array of floats to be rounded.
+    Input array, expected to have a real-valued data type.
 out : {None, dpnp.ndarray, usm_ndarray}, optional
     Output array to populate.
     Array must have the correct shape and the expected data type.
@@ -1749,19 +1750,12 @@ def ediff1d(ary, to_end=None, to_begin=None):
 Returns
 -------
 out : dpnp.ndarray
-    An array with the rounded values and with the same dimensions as the input.
-    The returned array will have the default floating point data type for the
-    device where `a` is allocated.
-    If `out` is ``None`` then a float array is returned with the rounded values.
+    An array with the same dimensions and data-type as the input.
+    If `out` is ``None`` then a new array is returned
+    with the rounded values.
     Otherwise the result is stored there and the return value `out` is
     a reference to that array.
 
-Limitations
------------
-Parameters `where` and `subok` are supported with their default values.
-Keyword argument `kwargs` is currently unsupported.
-Otherwise ``NotImplementedError`` exception will be raised.
-
 See Also
 --------
 :obj:`dpnp.round` : Round to given number of decimals.
@@ -1782,7 +1776,7 @@ def ediff1d(ary, to_end=None, to_begin=None):
 array([ 2.,  2., -2., -2.])
 """
 
-fix = DPNPUnaryFunc(
+fix = DPNPFix(
     "fix",
     ufi._fix_result_type,
     ufi._fix,
@@ -1934,8 +1928,10 @@ def ediff1d(ary, to_end=None, to_begin=None):
 
 Notes
 -----
-Some spreadsheet programs calculate the "floor-towards-zero", in other words floor(-2.5) == -2.
-DPNP instead uses the definition of floor where floor(-2.5) == -3.
+Some spreadsheet programs calculate the "floor-towards-zero", where
+``floor(-2.5) == -2``. DPNP instead uses the definition of :obj:`dpnp.floor`
+where ``floor(-2.5) == -3``. The "floor-towards-zero" function is called
+:obj:`dpnp.fix` in DPNP.
 
 Examples
 --------

dpnp/dpnp_iface_statistics.py

@@ -797,8 +797,8 @@ def cov(
         Default: ``None``.
     dtype : {None, str, dtype object}, optional
         Data-type of the result. By default, the return data-type will have
-        at least floating point type based on the capabilities of the device on
-        which the input arrays reside.
+        the default floating point data-type of the device on which the input
+        arrays reside.
 
         Default: ``None``.