Allow dynamic loading of TensorRT lib #3
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
…corresponding enums
…led for now due to deprecation
theHamsta
force-pushed
the
dynloading
branch
5 times, most recently
from
050258b to
7696256
Compare
Author
|
This would obviously also require building all feature permutations in CI. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
This is a draft of allowing dynamic loading of the TensorRT libs. Actual diff here cefe2c1
It is on top of the refactoring of #2 . I would wait whether #2 will be adopted or not and rebase accordingly. I would also wait with this PR on how this projects decides to select TensorRT versions (either via TensorRT present in dev enviornment or by checking in TensorRT headers)
The premise of this change is that both linking of the TensorRT lib and the ONNX parser link and dynamic loading should be Cargo features to which a user of the crate can opt-in or opt-out (no strong opinion what the default features of the crates should be).
Making dynload the default would enable vendoring the TRT headers and not require the TensorRT libs to be present in the dev environment. One could even vendor the headers of different TensorRT version and select the concrete version via Cargo feature or by analyzing the TensorRT library to be dynamically loaded.
Also the ONNX parser should be a Cargo feature. This way users that don't use ONNX don't need to generate unnecessary code.