[ROCm] slow torch.sum optimization by increasing max_values_per_thread in reduce config #135624
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
…d in reduce config (#135397) Fixes #132964 This change is to optimize torch.sum() performance by increasing max_values_per_thread in setReduceConfig() for ROCm platform. By increasing this parameter, it uses fewer threadblocks and improved the performance. Test: Tested on MI300x and H100, and now the MI300x perf improved to 3205GByte/s from ~1690GByte/s for the test case and is slightly better than H100 (3136GByte/s). Also tested with other different sizes of tensors and also see perf improvement. ```python import torch from triton.testing import do_bench x = torch.randn(2**30, device='cuda') ms = do_bench(lambda: x.sum(dim=-1)) bandwidth_gbyte = x.numel() * x.dtype.itemsize / (10**9) time_s = ms / 1000 bw_per_second = bandwidth_gbyte / time_s print(bw_per_second) ``` Co-author: @carlobertolli Pull Request resolved: #135397 Approved by: https://github.com/eqy, https://github.com/malfet (cherry picked from commit eb38ee2)
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/135624
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 3 Unrelated FailuresAs of commit 70e2277 with merge base b7eb725 ( NEW FAILURE - The following job has failed:
BROKEN TRUNK - The following jobs failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
pytorch-bot
bot
added
ciflow/rocm
eqy
approved these changes
Sep 11, 2024
pruthvistony
approved these changes
Sep 12, 2024
Collaborator
|
@atalman , |
hongxiayang
approved these changes
Sep 12, 2024
Contributor
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
Collaborator
|
closing the stale PR and also this change cant be pushed release/2.5 since its late. |
github-actions
bot
deleted the
cherry-pick-135397-by-pytorch_bot_bot_
branch
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.
Fixes #132964
This change is to optimize torch.sum() performance by increasing max_values_per_thread in setReduceConfig() for ROCm platform.
By increasing this parameter, it uses fewer threadblocks and improved the performance for large tensors.
Test:
Tested on MI300x and H100, and now the MI300x perf improved to 3205GByte/s from ~1690GByte/s for the test case and is slightly better than H100 (3136GByte/s).
Co-author: @carlobertolli @hongxiayang
cc @jeffdaily @sunway513 @jithunnair-amd @pruthvistony @ROCmSupport @dllehr-amd @jataylo @hongxiayang @naromero77amd