oneMKL

oneMKL copied to clipboard

[CUDA] Add support for cuSparse

4

oneMKL provides support for sparse BLAS operations, and NVIDIA provides an interface to sparse BLAS operations optimized for their hardware in the [cuSparse](https://docs.nvidia.com/cuda/cusparse/index.html) library. Much as oneMKL provides support for...

jasonsewall-intel

# Description This PR adds support to use the cuRAND backend with hipSYCL. The approach is similar though simpler than #144: Since hipSYCL does not support `host_task` but instead implements...

nilsfriess

error while loading shared libraries: libonemkl.so.0: cannot open shared object file: No such file or directory

2

Hi, I executed SYCL application on NVIDIA GPU which is having oneMKL library. I built the open source MKL library for NVIDIA GPU's. The code is generating an output, i...

ManjulaChalla

Missing gemm_batch data types

5

# Summary I believe there are some missing gemm_batch implementations, looking at the oneMKL docs it seems this should support. A `gemm_batch` with, two half matrices as input, a float...

AidanBeltonS

CMake with "RelWithDebInfo" build type links to debug libraries

2

# Summary The MKLConfig.cmake file included with MKL causes debug libraries to be linked when the "RelWithDebInfo" build type is used. This results in linker errors. This happens due to...

nickanthony-dgl

const scaling paramters not supported for gemm_batch

1

# Summary gemm_batch and possibly other batch methods do not allow `const Ts*` types for the `alpha` and `beta` scaling parameters. I believe this should be the case as it...

AidanBeltonS

[CMake][cuBLAS]Unable to build cublas backend with a default CMake command

3

# Summary I am not able to build the cuBLAS backend without working around some CMake logic and adding CMake flags. # Version I am using the latest commit on...

Rbiessy

Illegal instruction error with oneMKL mt2203 instruction

16

# Summary MersenneTwisterGP11213 Sample migrated from CUDA fails on NVIDIA GPU with opensource oneMKL library - _Illegal instruction (core dumped)_ ``` checkCudaErrors(DPCT_CHECK_ERROR(prngGPU = dpct::rng::create_host_rng( dpct::rng::random_engine_type::mt2203))); ``` # Version onemkl version...

ghost

[CMake] New users can struggle with the CBLAS testing dependency when building oneMKL - what to do?

1

**The problem** Both in the issues ([here](https://github.com/oneapi-src/oneMKL/issues/436)), and privately, I've supported new users struggling to build the oneMKL library. Their aim was to build and install oneMKL. In both cases,...

hjabird

`gemm` throws exception on PVC

16

# Summary I'm trying to use `gemm` on PVC, but it keeps throwing an exception. Please let me know where I'm going wrong. I am attempting to use `gemm` and...

BenBrock