v0.14.4

This is hot-fix for 0.14.3 release.

Added

• Added dpctl.tensor.less_equal, dpctl.tensor.greater, dpctl.tensor.greater_equal: #1239

Changed

• Optimized in-place arithmetic operations for updating matrix with rows/columns via broadcasting: #1244

Fixed

• Fixed handling of 0d arrays in dpctl.tensor.sum: #1238

v0.14.3

Added

• Added support of axis=None in dpctl.tensor.concat #1125
• Added caching for dpctl.SyclDevice.filter_string property #1127
• Added dpctl.tensor.isdtype from array API #1133
• Added dpctl.tensor.unstack, dpctl.tensor.moveaxis, dpctl.tensor.swapaxes #1137, #1174
• Allow for mutation of dpctl.tensor.usm_ndarray.flags.writable #1141
• Added dpctl.tensor.where from array API #1147
• Include libtensor headers in dpctl installation layout #1185
• Added new properties of dpctl.tensor.usm_ndarray object #1199
• Added a list of unary and binary elementwise functions from array API:
  • #1203: dpctl.tensor.add, dpctl.tensor.divide, dpctl.tensor.isnan, dpctl.tensor.isinf, dpctl.tensor.isfinite, dpctl.tensor.cos, dpctl.tensor.abs, dpctl.tensor.equal
  • #1205: dpctl.tensor.sqrt
  • #1209: implements out keyword argument
  • #1211: dpctl.tensor.multiply, dpctl.tensor.subtract
  • #1214: dpctl.tensor.not_equal
  • #1216: dpctl.tensor.exp, dpctl.tensor.sin
  • #1217: dpctl.tensor.real, dpctl.tensor.imag, dpctl.tensor.proj
  • #1218: dpctl.tensor.log, dpctl.tensor.log1p, dpctl.tensor.expm1
  • #1221: dpctl.tensor.floor_divide
  • #1235: dpctl.tensor.less
  • #1237: in-place support for addition, multiplication and subtraction
• Added dpctl.tensor.all and dpctl.tensor.any #1204
• Added dpctl.tensor.sum #1210

Changed

• Updated examples of native Python extensions built using dpctl #1108
• Used security flags to compile and link native extensions of dpctl #1109
• Changed types of dpctl.tensor.finfo and dpctl.tensor.iinfo output structure per array API spec #1110
• Consolidated multiple USM temporaries life-time management host_tasks to improve test suite stability #1111
• MAINT: Improved cmake target dependency tracking #1112
• MAINT: Improved docstrings for existing dpctl.tensor functions #1123
• Changed default value of mode keyword in dpctl.tensor.take and dpctl.take.put from clip to wrap #1132
• Added support for (nested) sequence of dpctl.tensor.usm_ndarray objects in dpctl.tensor.asarray #1139
• Improved exception handling in dpctl.tensor.usm_ndarray.__setitem__ special method #1146
• Simplified implementation of copy-and-cast kernels and removed special casing for 2D arrays to conserve binary size #1165
• Improved speed of dpctl.tensor.usm_ndarray printing functionality #1187
• Require DPC++ RT 2023.1 to build and run dpctl #1195
• Compile offloading native extensions with -fno-sycl-id-queries-fit-in-int fixing gh-1184, #1200
• Transition to conda-forge ecosystem #1213

Fixed

• Fix to add empty values check for dpctl.tensor.place #1105, #1106
• Fixed gh-1089 by improving dpctl.tensor.asarray handling of NumPy arrays viewing into host-accessible USM allocation objects.
• MAINT: Fixed build break with newer GCC and SYCLOS #1118
• Fixed a bug in basic indexing of dpctl.tensor.usm_ndarray #1136

v0.14.2

Added

• Added dpctl.SyclDevice.partition_max_sub_devices property #1005
• Added dpctl.program.SyclKernel.max_sub_group_size property #1028
• Implemented printing of usm_ndarray #1013, #1043, #1060
• Implemented support for advanced indexing for dpctl.tensor.usm_ndarray #1095, #1097, #1099, #1101
• Implemented support for platform listing in dpctl.__main__ script #1014
• Improved performance of dpctl.tensor.asnumpy #1026
• Added UsmNDArray_Make* C-API for constructing dpctl.tensor.usm_ndarray from native allocations #1050, #1067
• Added support for dpctl.SyclDevice.native_vector_width_* device descriptors #1075
• Added dpctl::tensor::usm_ndarray::get_shape_vector and dpctl::tensor::usm_ndarray::get_strides_vector methods #1090

Changed

• Removed dpctl.select_host_device, dpctl.has_host_device, dpctl.SyclDevice.is_host, and dpctl.SyclDevice.has_aspect_host since support for host device has been removed in DPC++ 2023 and from SYCL 2020 spec #1028

• usm_ndarrayis made writable by default #1012, and writable flag is now checked by __setitem__.

• Added convenience signature for C++ utility function in "dpctl4pybind11.hpp" #1016

• Improved error reported when attempting to submit kernel that uses a data-type unsupported by target device #1018, #1040

• Updated C++ code to require DPC++ 2023.0.0 or newer #1028, #1066

• The dpctl.tensor.Device class supports print_device_info method #1029, equality comparison, and hashing #1048

• Updated version of pybind11 used to 2.10.2 #1031

• Improved internal utility responsible for reduction of iteration space dimensionality #1044, #1054

• Changed return type of DCPCTLUSM_GetPointerType function in SyclInterface library #1061, #1065

• Updated supported version of DLPack to 0.8 #1073

• Implemented queue cache per context/device pair and deployed it in dpctl.memory, dpctl.tensor.from_dlpack and dpctl.tensor array creation functions #1076, #1079

• Maintainance, CI work: #1001, #1009, #1011, #1024, #1030, #1032, #1035, #1037, #1039, #1041, #1045, #1047, #1055, #1057, #1059, #1068, #1070, #1074,#1077, #1078, #1081, #1084, #1085, #1088, #1086, #1092, #1093

Fixed

• Fixed error gh-998 in forming Python exception, #999.
• A small memory leak fixed, #1000
• Improved dtype support in dpctl.tensor.full, PR #1002
• Added missing header file #1008 fixing gh-1007
• Fixed a typo in device-specific dtype mapping #1015
• Fixed default device integer type to align with NumPy's behavior on Windows #1017
• Fixed unexpected overflow in dpctl.tensor.linspace when one of the parameters is the largest floating point value #1034
• Constructors dpctl.tensor.empty, dpctl.tensor.zeros, and usm_ndarray constructor itself no longer allow to create array with data-types not supported by targeted device #1042
• Fixed parameter validation in dpctl.SyclQueue constructor #1052
• Fixed usm_type of the resulting array in dpctl.tensor.tril and dpctl.tensor.triu functions #1062
• Used DPC++ configuration files to ensure correct use of conda compiler toolchain on Linux #1072
• Fixed issue with empty argument of dpctl.tensor.meshgrid function #1080
• Fixed linking problem on Windows enabling dpctl to be functional on Windows for devices not supporting some data types #1083

Full Changelog: 0.14.0...0.14.2

v0.14.0

[0.14.0] - 11/18/2022

Added

• Implemented dpctl.tensor.linspace function from array-API #875.
• Implemented dpctl.tensor.eye function from array-API #896.
• Implemented dpctl.tensor.tril and dpctl.tensor.triu functions from array-API #910.
• Added data type objects to dpctl.tensor namespace, finfo, iinfo, can_cast, and result_type functions #913.
• Implemented dpctl.tensor.meshgrid creation function from array-API #920.
• Implemented convenience class to represent output of dpctl.tensor.usm_ndarray.flags property #921.
• Added new device attributes and kernel's device-specific attributes #894.
• Added dpctl.utils.onetrace_enabled context manager for targeted trace collection #903.
• Added support for stream keyword in __dlpack__ method, enabling support for sending usm_ndarray using mpi4py #906.
• dpctl.tensor.asarray can now transition data between incompatible devices, #951.
• Introduced "syclinterface/dpctl_sycl_types_casters.hpp" header file with declaration of conversion routines between SYCL type pointers and SyclInterface library opaque pointers #960.
• Added C-API to dpctl.program.SyclKernel and dpctl.program.SyclProgram. Added type casters for new types to "dpctl4pybind11" and added an example demonstrating its use #970.
• Introduced "dpctl/sycl.pxd" Cython declaration file to streamline use of SYCL functions from Cython, and added an example demonstrating its use #981.
• Added experimental support for sharing data allocated on sub-devices via dlpack #984.
• Added dpctl.SyclDevice.sub_group_sizes property to retrieve supported sizes of sub-group by the device #985.

Changed

• Improved queue compatibility testing in dpctl.tensor's implementation module #900.
• Added automatic measurement of array-API conformance test suite in CI #901.
• Improved performance of array metadata transfer from host to device #912.
• Used os.add_dll_directory on Windows to ensure that DPCTLSyclInterface library can be found #918.
• Refactored dpctl.tensor's implementation module #941 to streamline adding new functionality. Streamlined dpctl::tensor::usm_ndarray class implementation.
• Added debugging messaging in case when DPCTLDynamicLib::getSymbol encounters errors #956.
• Updated code base according to changes in DPC++ compiler #952, #957, #958.
• Changed dpctl to use pybind11 2.10.1 #967.
• Extended dpctl.tensor.full to accept 0d and higher dimensional arrays for fill-value parameter #982 and #995.

Fixed

• Improved SyclDevice constructor error message #893.
• Fixed issue gh-890 about dpctl.tensor.reshape function #915.
• Fixed unexpected UnboundLocalError exception in #922.
• Fixed bugs in dpctl.tensor.arange in #945.
• Fixed issue with type inferencing in dpctl.tensor.asarray in #949.
• Added missing docstrings for dpctl.SyclDevice properties #964.

v0.13.0

Added

• Implemented and deployed dedicated kernels for copying with casting #781, used in __setitem__, implementaion of asarray, dpctl.tensor.copy functions.
• Implemented dedicated copying kernel for dpctl.tensor.reshape function #810, added support for copy keyword #807.
• Implemented dedicated kernel to copy with casting from numpy.ndarray into dpctl.tensor.usm_ndarray #817.
• Implemented dpctl.tensor.permute_dims function from array-API #787.
• Implemented dpctl.tensor.expand_dims function from array-API #788.
• Implemented dpctl.tensor.squeeze function from array-API #790.
• Implemented dpctl.tensor.broadcast_to function from array-API #791.
• Implemented dpctl.tensor.broadcast_arrays function from array-API #798.
• Implemented dpctl.tensor.flip function from array-API #801.
• Implemented dpctl.tensor.usm_ndarray.mT property per array-API #805.
• Implemented dpctl.tensor.roll function from array-API #809.
• Implemented dpctl.tensor.arange function from array-API #814.
• Implemented dpctl.tensor.zeros function from array-API #816.
• Implemented dpctl.tensor.zeros function from array-API #816.
• Implemented dpctl.tensor.ones, dpctl.tensor.full, dpctl.tensor.empty_like, dpctl.tensor.zeros_like, dpctl.tensor.ones_like, dpctl.tensor.full_like functions from array-API #822.
• Implemented DPCTLQueue_Memset function in SyclInterface library #812, and exposed it for dpctl.memory.MemoryUSM* classes #815.
• Implemented dpctl.utils.get_coerced_usm_type to deduced usm type of the output array from types of input arrays in compute-follows-data execution model #797.
• Added dpctl.SyclDevice.profiling_timer_resolution property #825.
• Added dpctl.SyclDevice.platform and dpctl.SyclPlatform.default_context properties #827.
• Provided pybind11 example for functions working on dpctl.tensor.usm_ndarray container applying oneMKL functions #780, #793, #819. The example was expanded to demonstrate implementing iterative linear solvers (Chebyshev solver, and Conjugate-Gradient solver) by asynchronously submitting individual SYCL kernels from Python #821, #833, #838.
• Wrote manual page about working with dpctl.SyclQueue #829.
• Added cmake scripts to dpctl package layout and a way to query the location #853.
• Implemented dpctl.tensor.concat function from array-API #867.
• Implemented dpctl.tensor.stack function from array-API #872.

Changed

• Enhanced coverage collection for SyclInterface library by also collecting it during pytest run and combining traces with those collected during C-test run #818. This change also allows to not rebuild SyclInterface library when building C-test executable.
• Exported keep_args_alive utility in dpctl4pybind11.hpp header #820. The utility uses sycl::handler::host_task to keep given Python arguments alive until eac sycl::event from the given vector of events is complete. The host task is scheduled on the SYCL queue provided as the first argument.
• Changed the size of struct underlying dpctl.SyclEvent to avoid storing Python object previously used to keep kernel arguments scheduled with dpctl.SyclQueue.submit #823.
• Fixed docstring for dpctl.SyclTimer #824.
• Changed type of exceptions raised on failure to create dpctl.SyclDevice from ValueError to dpctl.SyclDeviceCreationError #826.
• Improved performance of pybind11 type casters #837.
• Changed implementation of dpctl.SyclProgram from using deprecated sycl::program to sycl::kernel_bundle #845.
• Removed deprecated device aspects, added new supported aspects #844.
• Updated vendored dlpack.h to version 0.7 #847.

Fixed

• Fixed dpctl.lsplatform() to work correctly when used from within Jupyter notebook #800.
• Fixed script to drive debug build #835 and fixed code to compile in debug mode #836.
• Fixed filter selector string produced in outputs of dpctl.lsplatform(verbosity=2) and dpctl.SyclDevice.print_device_info #866.
• Fixed issue with slicing reported in gh-870 in #871.

New contributor: @npolina4 contributed #867, #872 and reported #870

v0.12.0

What's changed in 0.12.0

Added

• Properties added to MemoryUSM* objects. #647
• Added dpctl.tensor.asarray #646
• Implemented DLPack support for usm_ndarray #682
• Exported dpctl.tensor.Device class #708 #718
• Added testing of examples in CI #722
• Added user manuals to dpctl documentation #712 #773

Changed

• Folder dpctl-capi/ renamed to libsyclinterface/ in sources and documentation. #666
  #768
• Added workflow to publish rendered documentation on PRs #673 #753 #726
• Synchronization functions and USM allocation functions release GIL #736 #766
• dpctl.SyclEvent destructor is made non-blocking #751

Fixed

• Fixed for issue in code of dpctl.tensor.usm_ndarray.T #653
• Fixed issue with dpctl.tensor.reshape's affect on contiguity flags of usm_ndarray #695
• Fixed handling of empty list by dpctl.tensor.asarray #694
• Fixed type inference with array of empty arrays in dpctl.tensor.asarray #697
• Fixed issue gh-698 with dpctl.tensr.asarray #709
• Fixed performance of item assignment from numpy array #724
• DPCTLDeviceMgr_GetNumDevices should not operate on rejected devices #737
• Fixed issue gh-729 for dpctl.tensor.reshape applied to 0-element usm_ndarray #756
• Fixed issue gh-728 with dpctl.tensor.astype #757
• Fixed type in memory overlapping test #770
• Fixed issue with operator.pos for dpctl.tensor.usm_ndarray #783
• Only call PyThread_Ensure from host_task if the main-thread interpreter is initialized and not finalizing #776 #778 #721

Full Changelog: 0.11.4...0.12.0

0.11.4

What's Changed

• Fix tests for nested context factories expecting for integration environment by @PokhodenkoSA in #705

Full Changelog: 0.11.3...0.11.4

0.11.3

Fixed

• Set the last byte in allocated char array to zero [cherry picked from #650] (#699)

Full Changelog: 0.11.2...0.11.3

0.11.2

Added

• Extending dpctl.device_context with nested contexts (#678)

Fixed

• Fixed issue #649 about incorrect behavior of .T method on sliced arrays (#653)

Full Changelog: 0.11.1...0.11.2

0.11.1

Changed

• Replaced uses of clang compiler with icx executable (#665)