Refactors CustomSchedule.py for scale out of schedules#2574
Merged
Conversation
Contributor
Author
|
Note: this is only the changes for the file |
1 task
- Functions are a scalable way to organize custom schedule implementations. Takes all existing schedule implementations and puts each in its own function. Over time, we can factor out the functions to a separate module, if desired.
- use helpers for _get_schedule... functions to make a consistent calling convention.
talumbau
force-pushed
the
users/talumbau/cms_functions
branch
from
1044690 to
5694bbf
Compare
msujon-AMD
approved these changes
Nov 10, 2025
msujon-AMD
changed the base branch from
users/common/cvs_dev
to
hipblaslt_common_cms_dev
perfci run on commit 93c031e |
talumbau
added a commit
that referenced
this pull request
Nov 12, 2025
We are about to land a significant number of custom schedules for various tile sizes and transpose orientations. By using the convention "a schedule is described in a function" we can reasonably scale the code out, instead of a chain of if/elif blocks that spans hundreds or thousands of lines. Initially we keep the existing schedules in functions inside CustomSchedule.py, but we can refactor as needed. TENSILELITE_CLIENT_ARGS="--build-type Debug --gpu-targets gfx950 --clean" tox -e py39 -- Tensile/Tests -k custom_mainloop_scheduling.yaml <!-- Briefly summarize test outcomes. -->
talumbau
added a commit
that referenced
this pull request
Nov 12, 2025
We are about to land a significant number of custom schedules for various tile sizes and transpose orientations. By using the convention "a schedule is described in a function" we can reasonably scale the code out, instead of a chain of if/elif blocks that spans hundreds or thousands of lines. Initially we keep the existing schedules in functions inside CustomSchedule.py, but we can refactor as needed. TENSILELITE_CLIENT_ARGS="--build-type Debug --gpu-targets gfx950 --clean" tox -e py39 -- Tensile/Tests -k custom_mainloop_scheduling.yaml <!-- Briefly summarize test outcomes. -->
sebvince
pushed a commit
to sebvince/rocm-libraries
that referenced
this pull request
Nov 12, 2025
## Motivation We are about to land a significant number of custom schedules for various tile sizes and transpose orientations. By using the convention "a schedule is described in a function" we can reasonably scale the code out, instead of a chain of if/elif blocks that spans hundreds or thousands of lines. Initially we keep the existing schedules in functions inside CustomSchedule.py, but we can refactor as needed. ## Technical Details ## Test Plan TENSILELITE_CLIENT_ARGS="--build-type Debug --gpu-targets gfx950 --clean" tox -e py39 -- Tensile/Tests -k custom_mainloop_scheduling.yaml ## Test Result <!-- Briefly summarize test outcomes. -->
ammallya
pushed a commit
that referenced
this pull request
Feb 3, 2026
Re-enable group mode instances for the Pytorch receipt and resolve linker errors for torch SDPA [ROCm/composable_kernel commit: 7fe50dc]
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.
Motivation
We are about to land a significant number of custom schedules for various tile sizes and transpose orientations. By using the convention "a schedule is described in a function" we can reasonably scale the code out, instead of a chain of if/elif blocks that spans hundreds or thousands of lines. Initially we keep the existing schedules in functions inside CustomSchedule.py, but we can refactor as needed.
Technical Details
Test Plan
TENSILELITE_CLIENT_ARGS="--build-type Debug --gpu-targets gfx950 --clean" tox -e py39 -- Tensile/Tests -k custom_mainloop_scheduling.yaml
Test Result