Give broadcast_coalesced tensors different version counters #13594
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
colesbury
approved these changes
Nov 7, 2018
torch/csrc/cuda/comm.cpp
Outdated
| // See NOTE [ Version Counter in comm.*_coalesced ] | ||
| AT_ASSERT(t.is_variable()); | ||
| Variable var = t; | ||
| device_outputs.push_back(std::move(make_variable(var.data(), false))); |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
torch/csrc/cuda/comm.cpp
Outdated
| // See NOTE [ Version Counter in comm.*_coalesced ] | ||
| AT_ASSERT(t.is_variable()); | ||
| Variable var = t; | ||
| device_outputs.push_back(std::move(make_variable(var.data(), false))); |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
Contributor
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@ssnl is landing this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
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Outdated suggestions cannot be applied.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.
In
broadcast_coalesced, since multiple variables can be "views" of a big flattened tensor, they can share the same version counter. However, this base flat tensor is not exposed and they don't share any memory locations, so this is not necessary. Furthermore, it can cause problems, e.g., when two buffers are broadcast together inDataParalleland one of them is modified in-place duringforwardbut the other is needed in backward, autograd engine will complain.Fixing the bug discovered at #13350 (comment)
edit: This is a very real problem. E.g., consider using Spectral Norm + Batch Norm together.