Reduce elementwise extension size #1976
Merged
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
…function Doing so reduces the binary size of elementwise operations extension Before: ``` (dev_dpctl) opavlyk@mtl-world:~/repos/dpctl$ ls -l dpctl/tensor/_tensor_elementwise_impl.cpython-312-x86_64-linux-gnu.so -rw-r--r-- 1 opavlyk opavlyk 38659896 Jan 19 20:58 dpctl/tensor/_tensor_elementwise_impl.cpython-312-x86_64-linux-gnu.so ``` After: ``` dev_dpctl) opavlyk@mtl-world:~/repos/dpctl$ ls -l dpctl/tensor/_tensor_elementwise_impl.cpython-312-x86_64-linux-gnu.so -rw-r--r-- 1 opavlyk opavlyk 37176600 Jan 21 06:36 dpctl/tensor/_tensor_elementwise_impl.cpython-312-x86_64-linux-gnu.so ``` Added static assertions to offset_utils to ensure that indexers are device copyable.
oleksandr-pavlyk
force-pushed
the
reduce-elementwise-extension-size
branch
from
2a526d2 to
3f90e9b
Compare
|
Deleted rendered PR docs from intelpython.github.com/dpctl, latest should be updated shortly. 🤞 |
|
Array API standard conformance tests for dpctl=0.19.0dev0=py310h93fe807_490 ran successfully. |
ndgrigorian
reviewed
Jan 21, 2025
|
Array API standard conformance tests for dpctl=0.19.0dev0=py310h93fe807_491 ran successfully. |
ndgrigorian
approved these changes
Jan 22, 2025
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Outdated suggestions cannot be applied.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.
This PR factors out inline submit to populate padded vector in specializations for binary operations on matrix and a vector.
Doing so allows to generate fewer of such kernels, resulting in the binary size decrease.
Before:
After:
Also this PR adds
static_assertinoffset_utils.hppto verify that indexers are device copyable.It also sneaks in changes of defining local typename for the functor being submitted in
cgh.parallel_forto simplify the invocation.