Project

General

Profile

Feature #2254

GPU extensions for Google Tests

Added by Szilárd Páll about 3 years ago. Updated 16 days ago.

Status:
Resolved
Priority:
Normal
Assignee:
-
Category:
mdrun
Target version:
-
Difficulty:
uncategorized
Close

Description

As discussed in the past, we could make use of some compact abstraction layer for GPU framework glue code that wraps boilerplate code required to manage devices, initialization, teardown of contexts, queues, etc. Ideally, this would be a compact base-class which provides the API with CUDA and OpenCL flavors implemented.

As an example for such a class for OpenCL testing see:
https://github.com/clMathLibraries/clBLAS/blob/master/src/tests/BlasBase.cpp

Associated revisions

Revision 37969a6d (diff)
Added by Artem Zhmurov 16 days ago

Use new GPU infrastructure in MDLib tests

This make use of common device testing infrastructure in MDLib tests,
where both GPU and CPU implementations are tested. The GPU runners
will now be executed on all the detected devices, not only on the
default one. Also, this will allow to use the MDLib tests in OpenCL
and SYCL, where proper device context object is needed.

Closes #3317

Closes #2254

Related #2092

History

#1 Updated by Artem Zhmurov 16 days ago

  • Status changed from New to Resolved

Also available in: Atom PDF