Searched refs:device (Results 1 – 12 of 12) sorted by relevance
| /libCEED/backends/sycl/ |
| H A D | ceed-sycl-common.sycl.cpp | 32 auto sycl_devices = sycl::device::get_devices(device_type); in CeedInit_Sycl() 44 sycl::device sycl_device{sycl_devices[device_id]}; in CeedInit_Sycl()
|
| H A D | ceed-sycl-common.hpp | 36 sycl::device sycl_device;
|
| H A D | ceed-sycl-compile.sycl.cpp | 71 static inline int CeedJitCompileSource_Sycl(Ceed ceed, const sycl::device &sycl_device, const std::… in CeedJitCompileSource_Sycl() 86 static int CeedLoadModule_Sycl(Ceed ceed, const sycl::context &sycl_context, const sycl::device &sy… in CeedLoadModule_Sycl()
|
| H A D | online_compiler.hpp | 101 online_compiler(const sycl::device &) in online_compiler() argument
|
| /libCEED/examples/python/ |
| H A D | tutorial-1-vector.ipynb | 208 …device memory in your `libceed.Vector`s. In the following example, we create a `libceed.Vector` wi…
|
| /libCEED/doc/sphinx/source/ |
| H A D | libCEEDapi.md | 11 …cription that enables efficient evaluation on a variety of computational device types (selectable … 178 …asses of numerical algorithms -- parallel (multi-device) linear algebra for $\bm{P}$, sparse (on-d… 183 The idea is that every MPI rank can use any logical device it is assigned at runtime. 191 …tion to the original operator on **L-vector** level (i.e. independently on each device / MPI task). 243 creates a logical device `ceed` on the specified *resource*, which could also be a coprocessor such… 245 …hich will have discretion over the implementations of all objects created with this logical device. 316 If the user has arrays available on a device, they can be provided using `CEED_MEM_DEVICE`.
|
| H A D | intro.md | 21 …n-time selection of implementations tuned for a variety of computational device types, including C…
|
| H A D | releasenotes.md | 15 …- Add `CEED_RUNNING_JIT_PASS` compiler definition for wrapping header files that device JiT compil… 108 - Fix bug in setting device id for GPU backends. 268 …ext` object to manage user QFunction context data and reduce copies between device and host memory. 595 but also add corresponding device (e.g., GPU) pointers to the data. Coherency is handled
|
| /libCEED/backends/sycl-ref/ |
| H A D | ceed-sycl-ref-basis.sycl.cpp | 45 const sycl::device &sycl_device = sycl_queue.get_device(); in CeedBasisApplyInterp_Sycl() 149 const sycl::device &sycl_device = sycl_queue.get_device(); in CeedBasisApplyGrad_Sycl()
|
| /libCEED/doc/papers/joss/ |
| H A D | paper.md | 111 …a `CeedVector`, which typically encapsulates zero-copy access to host or device memory provided by… 133 …exposed as `libceed.Vector` in Python, and supports no-copy host and GPU device interperability wi…
|
| /libCEED/ |
| H A D | README.md | 20 … that enable efficient operator evaluation on a variety of computational device types (CPUs, GPUs,… 221 Users can specify a device for all CUDA, HIP, and MAGMA backends through adding `:device_id=#` afte…
|
| /libCEED/backends/sycl-gen/ |
| H A D | ceed-sycl-gen-operator-build.sycl.cpp | 227 …std::vector allowed_sg_sizes = sycl_data->sycl_device.get_info<sycl::info::device::sub_group_size… in CeedOperatorBuildKernel_Sycl_gen()
|