[occa] qdata offset fix in qfunction before restore in op apply

Nek: Remove a redundant echo statement

Nek: Add a Makefile for Nek5000 examples

Merge branch 'magma-dev' of https://github.com/CEED/libCEED into magma-dev

one more file ...following discussion and agreement, remove for now /cpu/magma as support is provided through /gpu/magma

mollowing discussion and agreement, remove for now /cpu/magma as support is provided through /gpu/magma

Additional tweaks for building using IBM XL Fortran.

In Makefile, use CUDA headers/libs/compiler from the same location.
Tweak the MAGMA link flags.

Merge branch 'deep-copy-focca' [PR #68]

* deep-copy-focca:
When creating a Q-function, create a deep copy of the 'focca' string input. This resolves an issue with the OCCA backend when called from

Merge branch 'deep-copy-focca' [PR #68]

* deep-copy-focca:
When creating a Q-function, create a deep copy of the 'focca' string input. This resolves an issue with the OCCA backend when called from Fortran: the 'focca' pointer becomes invalid after the Q-function create call, at least with the Intel Fortran compiler. This lead to incorrect content of the 'qFunctionName' field in the structure CeedQFunction_Occa which points inside 'focca'.

show more ...

When creating a Q-function, create a deep copy of the 'focca' string
input. This resolves an issue with the OCCA backend when called from
Fortran: the 'focca' pointer becomes invalid after the Q-func

When creating a Q-function, create a deep copy of the 'focca' string
input. This resolves an issue with the OCCA backend when called from
Fortran: the 'focca' pointer becomes invalid after the Q-function
create call, at least with the Intel Fortran compiler. This lead to
incorrect content of the 'qFunctionName' field in the structure
CeedQFunction_Occa which points inside 'focca'.

show more ...

change of the generator to reflect bug fix and improved kernel (from Azzam; Thanks)

add missing restores; fix t30 test - it was assuming allocation initializes values to 0; bug fixes thanks to Azzam; control of templates to be done on CPU or GPU

some defines that control what goes through GPU code

add macro for debugging CUDA kernels

add atomics.cuh in the backened. Before was in magma and this was creating problem for spack

backends: add missing CeedVectorRestoreArray calls

libCEED 0.2

Merge branch 'master' of github:dmed256/libceed [PR #62]

* 'master' of github:dmed256/libceed:
[occa] Updated verbose flag

t30-operator.c: add missing CeedVectorRestoreArrayRead

Merge branch 'jed/fix-occa-mem-leaks' [PR #61]

* jed/fix-occa-mem-leaks:
occa: fix memory leaks

occa: fix memory leaks

Valgrind still says there is memory lost under some of these (plus a lot
from Occa static constructors), but this at least makes an attempt to
free from our end.

[occa] Updated verbose flag

make: handle opt/debug in consistent way

I'm not wild about this "make NDEBUG=" interface for specifying an
optimization preference, but don't want to change it right now for the
release.

Merge pull request #60 from thilinarmtb/nek_readme_updates

Nek: Update README.md

Update README.md