swap ordering in magma backend names

add cuda and hip tags to MAGMA backends

Merge pull request #628 from CEED/jeremy/ceed-hash

Make - install hashing headers, used in MFEM

Make - install hashing headers, used in MFEM

Merge pull request #629 from CEED/jeremy/oneapi

Make - Intel oneAPI C/C++ compiler renamed to icx/icpx

Make - Intel oneAPI C/C++ compiler renamed to icx/icpx

add magma-det backend for HIP mode

Merge pull request #610 from CEED/jeremy/cuda-qf-pointer

QF - add interface to set device user function

QFunction - compile cuda/hip only functions separately

QF - add interface to set device user function

Merge branch 'main' into icl/hip-magma

Consolidate CUDA backends (#623)

* Travis - allow icc failure for now

* move cuda-reg basis init kernels to cuda-shared

* move cuda-reg restrictions to cuda-ref

* change delegate ceeds for

Consolidate CUDA backends (#623)

* Travis - allow icc failure for now

* move cuda-reg basis init kernels to cuda-shared

* move cuda-reg restrictions to cuda-ref

* change delegate ceeds for previous uses of cuda-reg

* remove cuda-reg backend

* update hip restrictions to match cuda

* update backends list in README

* make style

* update release notes for removal of cuda-reg

Co-authored-by: jeremylt <thompson.jeremy.luke@gmail.com>

show more ...

detect MAGMA build from library symbols

Add HIP support for MAGMA backend

update HIP libs (#619)

Merge branch 'main' into yohann/fix-cuda-gen

OCCA Backend Update (#305)

* [Docs] Update docs for new OCCA backend

* [Fortran] CeedVectorRestoreArray fix

* [Test] Updates t002-ceed test to support query params

* [Make] Adds tidy for cp

OCCA Backend Update (#305)

* [Docs] Update docs for new OCCA backend

* [Fortran] CeedVectorRestoreArray fix

* [Test] Updates t002-ceed test to support query params

* [Make] Adds tidy for cpp files

* [OCCA] Updates OCCA backend

* PR feedback: Update StrideType naming

* PR feedback: Fixed USER_STRIDES vs BACKEND_STRIDES usage

* [OCCA] Add comments to code generation

* [CI] Removes commit from OCCA build

* PR Feedback: Update README

* PR Feedback: Removed compiler warnings

* Fix restriction function changes

* occa: track AssembleLinear -> LinearAssemble

* [OCCA][Vector] Adds takeArray method

* [OCCA][Restriction] Fixes restriction strides

* [OCCA][Operator] Adds point block diagonal registration

* [OCCA][Operator] Fixes

* OCCA - update debug color for visability

* Travis - fix extra fi

* OCCA - adjust kernel, multi component derivatives are expected as [dim][comp][q]

* OCCA - adjust basis kernel args in operator kernel to agree with expected [dim][comp][q] ordeing

* OCCA - fix uninitalized memory in grad transpose 3d kernel

* OCCA - fix Elayout description

* OCCA - fix bad dimensions in basis kernel

* OCCA - fix TakeArray to sync before returning array pointer

* MFEM - print error when test fails

* OCCA - fix 2d grad kernels

* OCCA - flag digonal of composite operators unsupported

* OCCA - fix restoreArray logic

* OCCA - minor cleanup with GPU

* Travis - add 'make info' where able for debugging

* OCCA - explictily test OpenCL mode

* OCCA - drop restrict for ElemRestriction kernels, OpenCL doesn't like it

* OCCA - explicitly test cuda and hip versions of OCCA backend

* OCCA - explicitly test OpenMP mode in OCCA

* Tests - modify check for VLA support for OCCA to catch all OCCA modes

* WIP - test possible OCCA fix for PowerPC

* OCCA - separate CPU modes for testing as well

* Readme - update list of OCCA backend modes

* Makefile - fix unterminated addprefix

* OCCA - enable direct access to OCCA Serial mode

* OCCA - add comments to registration

* Makefile - remove extra )

* OCCA - remove pass by reference C++ syntax for OpenCL compatability

* OCCA - drop use of @restrict for OpenCL

* OCCA - remove OpenCL mode, not fully supported in OCCA (see OCCA issue #166)

* OCCA - fixing rebase issues

* OCCA - Fix implementation of QFunctionContext

* OCCA - move GetContextSize so ierr check actually works

* Travis - use libOCCA instead of jeremylt/occa

* Junit - update OCCA test skip list

* Make - simplify OCCA check for enabled modes

Co-authored-by: Jed Brown <jed@jedbrown.org>
Co-authored-by: Jeremy L. Thompson <jeremy.thompson@colorado.edu>

show more ...

QFunction Context Data Object (#596)

* Ctx - create context object for QFunction context data

* Context - rename UserContext -> QFunctionContext

* Ctx - add lcov markers

* Ctx - fix leak in

QFunction Context Data Object (#596)

* Ctx - create context object for QFunction context data

* Context - rename UserContext -> QFunctionContext

* Ctx - add lcov markers

* Ctx - fix leak in identity QFunctions

* Hip/Cuda - rename sync functions for vector/context

* Tests - lcov marker update

* QFunction - drop unused function

* Python - fix copy-paste errors

* Ctx - update notes for Fortran usage

* Fortran - drop unneeded cast

Co-authored-by: Jed Brown <jed@jedbrown.org>

* Interface - use void* for SetData interfaces

* Make - use call quiet for NVCC

* Interface - use void* for GetData interfaces

* Make - add quiet call option for examples

* Makefile - create common makefile to reduce duplication/complexity in example makefiles

Co-authored-by: Jed Brown <jed@jedbrown.org>

show more ...

Merge pull request #598 from CEED/jeremy/cuda-small-fixes

Small Cuda/Hip Fixes

Makefile - small HIP improvements

Fix CodeCov Reports (#597)

* Tests - use qfunction headers for Fortran tests to improve bypass gcov issue

* Gitlab - use latest gcc on Noether

* Cuda/Hip - add case in reciprocal for completen

Fix CodeCov Reports (#597)

* Tests - use qfunction headers for Fortran tests to improve bypass gcov issue

* Gitlab - use latest gcc on Noether

* Cuda/Hip - add case in reciprocal for completeness

* Cuda - remove duplicate case

* Makefile - exclude fortran test headers from make style

* Travis - update to Focal

* Cov - adjust style to be consistent and avoid false misses

* Travis - update comments and style

show more ...

Merge pull request #594 from CEED/jed/hip-lcov-excl

backends/hip: exclude testing of unimplemented code

Makefile - minor reordering

Device - put device kernels in separate 'kernels' folder in backends

Merge pull request #591 from CEED/natalie/hip-ref-v2

Add hip-ref backend