Fix bitvector test by precompiling the kernel in library #1929
Conversation
This file contains hidden or 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
yhmtsai
added
1:ST:ready-for-review
ginkgo-bot
added
reg:build
1 task
yhmtsai
added
1:ST:ready-to-merge
yhmtsai
force-pushed
the
fix_bitvector
branch
from
9d80394 to
fc31474
Compare
yhmtsai
added
the
1:ST:no-changelog-entry
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.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.
thrust has the issue when the kernel are compiled in different shared library.
We have used
THRUST_CUB_WRAPPED_NAMESPACEto avoid the issue between Ginkgo and the other library.However, this issue still bothers us between test and ginkgo library.
We need bitvector kernel implementation in header because of #1832
To work around this issue, I decide to pre-compile the kernels for tests and avoid any thrust kernel compilation from test.
Another workaround is to skip this test when building shared library with CUDA < 12.4
CCCL waives this
THRUST_CUB_WRAPPED_NAMESPACEin 2.3.0 and CUDA 12.4 use CCCL 2.3.1 already.related issue: NVIDIA/cccl#655
From the issue, this issue happens after cub 2.0.0.
It is raised from #1841 cuda 12.2 test