Fix indexing of large tensors #20920
Closed
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
pytorchbot
added
module: cuda
Member
No, because the "32-bit" code path uses "fast integer division" ( I slightly prefer my patch (#20919). The 32-bit + split is slightly faster than 64-bit indexing (23.5 ms vs. 33.3 ms for the repro you wrote for #20888) and has fewer specializations. None of this really matters because we're not really optimizing for the giant Tensor case and the extra template specializations are unlikely to have a significant impact on overall compilation time. |
Collaborator
Author
|
Closed in favor of #20919 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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Outdated suggestions cannot be applied.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.
Uses
int64_tOffsetCalculator when necessary, fix for #20888cc @colesbury.
@colesbury, should it be
uint32_therepytorch/aten/src/ATen/native/TensorIterator.cpp
Line 628 in 4704322
uint32_t, notint32_t.