Disable warpReduce under the LTS driver#1337
Merged
alexbaden merged 1 commit intollvm-targetfrom Jun 12, 2024
Merged
Conversation
Contributor
|
What's the issue to fix it on LTS? |
Contributor
Author
|
#1336 linked above |
Contributor
Thanks, I thought that issue is for this PR, and not for the future investigation. |
whitneywhtsang
approved these changes
Jun 12, 2024
Comment on lines
+90
to
+93
| const bool isLTS = | ||
| op->getParentOfType<ModuleOp>()->hasAttr("triton_gpu.is_lts"); | ||
| if (isLTS) | ||
| return false; |
Contributor
There was a problem hiding this comment.
[optional] IMO triton_gpu.is_lts is clear, no need to create a variable.
Suggested change
| const bool isLTS = | |
| op->getParentOfType<ModuleOp>()->hasAttr("triton_gpu.is_lts"); | |
| if (isLTS) | |
| return false; | |
| if (op->getParentOfType<ModuleOp>()->hasAttr("triton_gpu.is_lts")) | |
| return false; |
Contributor
Author
|
Still fails after the fox127 upgrade so I'm going to merge this. |
Merged
etiotto
pushed a commit
that referenced
this pull request
Jul 10, 2024
The workarounds were added in #1275 and #1337. All huggingface training float32 models pass with the LTS workaround removed: https://github.com/intel/intel-xpu-backend-for-triton/actions/runs/9865658421 Signed-off-by: Whitney Tsang <whitney.tsang@intel.com>
wdziurdz
pushed a commit
that referenced
this pull request
Apr 7, 2026
Fixes #1179 Signed-off-by: Gregory Shimansky <gregory.shimansky@intel.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Allows all HF Float32 Training accuracy models to pass under LTS driver and upstream PyTorch.
cc #1336