Execution spaces: support for memory backends and execution policies #543
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
cnpetra
changed the title
6 tasks
cameronrutherford
requested changes
Dec 13, 2022
There was a problem hiding this comment.
Perhaps the biggest PR I have ever reviewed. Amazing work @cnpetra, and while I made 43 comments, they are all relatively minor.
| void hiopVectorRajaPar::print(FILE* file/*=nullptr*/, const char* msg/*=nullptr*/, int max_elems/*=-1*/, int rank/*=-1*/) const | ||
| template<class MEM, class POL> | ||
| void hiopVectorRaja<MEM,POL>:: | ||
| print(FILE* file, const char* msg/*=NULL*/, int max_elems/*=-1*/, int rank/*=-1*/) const |
There was a problem hiding this comment.
These default values should be enforced either in header file or here... I commented on another function that had default values removed, so wanted to make sure that functionality wasn't lost.
cnpetra
force-pushed
the
hwbackends_memory_dev
branch
from
e59d2e0 to
df6136e
Compare
cnpetra
force-pushed
the
hwbackends_memory_dev
branch
from
339dbbb to
4794a18
Compare
This was referenced Dec 19, 2022
Closed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Abstraction needed to support multiple hardware backends. The goal is to allow multiple "backends" to run concurrently, e.g., Cuda/Hip linear algebra with Raja linear algebra. This proposed implementation differentiates between memory backend and execution policy to allow using Raja with Umpire memory, Raja with Cuda/Hip memory, or Raja with std memory.
Passes tests on Crusher (with Umpire memory backend and RAJA exec policies), in addition to all the CIs. Tested with CUDA memory backend and RAJA exec policies manually on Ascent.