Add scaling arguments to bsr_dense_addmm #136104
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
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/136104
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 825a557 with merge base 357b7fb ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
As in the title. Tackles https://github.com/pytorch/ao/pull/821/files#r1759821413 [ghstack-poisoned]
As in the title. Tackles https://github.com/pytorch/ao/pull/821/files#r1759821413 The PR assumes that the existing tuning parameters are good also when using scaling arguments. This needs to be verified as a follow-up task. [ghstack-poisoned]
As in the title. Tackles https://github.com/pytorch/ao/pull/821/files#r1759821413 The PR assumes that the existing tuning parameters are good also when using scaling arguments. This needs to be verified as a follow-up task. [ghstack-poisoned]
As in the title. Tackles https://github.com/pytorch/ao/pull/821/files#r1759821413 The PR assumes that the existing tuning parameters are good also when using scaling arguments. This needs to be verified as a follow-up task. Also, this PR redefines triton-contiguous tensors: the tensor must have strides not larger than 1. This will now allow zero strides that previously triggered `contiguous` call although the underlying memory buffer was contiguous. Re: "a considerable slow-down occurs because tensor data is copied element-wise rather than chunk-wise" - this note should refer to a code (torch or triton?) that implements the element/chunk-wise copy so that we could verify that allowing zero strides indeed would not trigger element-wise copies. Atm, the performance increase in ViT-H benchmarks (that involve using 0 strides) is an evidence that allowing zero strides does not lead to slow-downs. [ghstack-poisoned]
cpuhrsch
approved these changes
Sep 16, 2024
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Chao1Han
pushed a commit
to Chao1Han/pytorch
that referenced
this pull request
Sep 20, 2024
As in the title. Tackles https://github.com/pytorch/ao/pull/821/files#r1759821413 The PR assumes that the existing tuning parameters are good also when using scaling arguments. This needs to be verified as a follow-up task. Also, this PR redefines triton-contiguous tensors: the tensor must have strides not larger than 1. This will now allow zero strides that previously triggered `contiguous` call although the underlying memory buffer was contiguous. Re: "a considerable slow-down occurs because tensor data is copied element-wise rather than chunk-wise" - this note should refer to a code (torch or triton?) that implements the element/chunk-wise copy so that we could verify that allowing zero strides indeed would not trigger element-wise copies. Atm, the performance increase in ViT-H benchmarks (that involve using 0 strides) is an evidence that allowing zero strides does not lead to slow-downs. Pull Request resolved: pytorch#136104 Approved by: https://github.com/cpuhrsch
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
As in the title.
Tackles https://github.com/pytorch/ao/pull/821/files#r1759821413
The PR assumes that the existing tuning parameters are good also when using scaling arguments. This needs to be verified as a follow-up task.
Also, this PR redefines triton-contiguous tensors: the tensor must have strides not larger than 1. This will now allow zero strides that previously triggered
contiguouscall although the underlying memory buffer was contiguous.Re: "a considerable slow-down occurs because tensor data is copied element-wise rather than chunk-wise" - this note should refer to a code (torch or triton?) that implements the element/chunk-wise copy so that we could verify that allowing zero strides indeed would not trigger element-wise copies. Atm, the performance increase in ViT-H benchmarks (that involve using 0 strides) is an evidence that allowing zero strides does not lead to slow-downs.
Stack from ghstack (oldest at bottom):