pt: avoid torch.tensor(constant) during forward #3421
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
torch.tensor(constant) copies memory from the CPU to GPU, so it is host blocking and should be avoided in the `forward` method. Signed-off-by: Jinzhe Zeng <[email protected]>
Signed-off-by: Jinzhe Zeng <[email protected]>
Signed-off-by: Jinzhe Zeng <[email protected]>
Signed-off-by: Jinzhe Zeng <[email protected]>
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## devel #3421 +/- ##
=======================================
Coverage 76.98% 76.98%
=======================================
Files 427 427
Lines 36753 36753
Branches 1621 1621
=======================================
Hits 28295 28295
Misses 7590 7590
Partials 868 868 ☔ View full report in Codecov by Sentry. |
for more information, see https://pre-commit.ci
anyangml
approved these changes
Mar 6, 2024
wanghan-iapcm
approved these changes
Mar 6, 2024
iProzd
approved these changes
Mar 8, 2024
Signed-off-by: Jinzhe Zeng <[email protected]>
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Outdated suggestions cannot be applied.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.
torch.tensor(constant)copies memory from the CPU to the GPU, so it is host blocking and should be avoided in theforwardmethod.Before, the CPU waited for the GPU using
cudaStreamSynchronize, blocking the CPU from doing the following things, where the CPU memory needs to be copied to the GPU, a.k.a. host-to-device (H2D).After this PR, all ops in the energy loss are asynchronous, as no H2D happens.
