Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Move frequency retrieval to the beginning and manual input when error #1483

Open
wants to merge 1 commit into
base: develop
Choose a base branch
from
Open

Move frequency retrieval to the beginning and manual input when error #1483

wants to merge 1 commit into from
+233 −73

Conversation

waynehuangntu
Copy link
Contributor

@waynehuangntu waynehuangntu commented Dec 24, 2024 •
edited
Loading

  1. Move the max frequency retrieval step to the beginning of the process.
  2. If an error occurs while retrieving or processing the maximum frequency, the program will pause and prompt the user to manually input the frequency before proceeding.
    image

jichangjichang
jichangjichang previously approved these changes Jan 6, 2025
View reviewed changes
KKyang
KKyang reviewed Jan 6, 2025
View reviewed changes
Copy link
Contributor

@KKyang KKyang left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we try to get the freq at the beginning of the process?

@vin-huang
Copy link
Collaborator

internal CI passed
http://math-ci.amd.com/job/main/job/precheckin/job/hipBLASLt/job/PR-1483/1/pipeline-graph/

KKyang
KKyang previously approved these changes Jan 13, 2025
View reviewed changes
@waynehuangntu waynehuangntu dismissed stale reviews from KKyang and jichangjichang via a05f7d9 January 14, 2025 06:18
@waynehuangntu waynehuangntu changed the title Add retry mechanism for frequency retrieval and manual input when error Move frequency retrieval to the beginning and manual input when error Jan 14, 2025
KKyang
KKyang previously approved these changes Jan 14, 2025
View reviewed changes
jichangjichang
jichangjichang previously approved these changes Jan 14, 2025
View reviewed changes
@jichangjichang
Copy link
Collaborator

please provide tenslitelite tox test result.

@waynehuangntu waynehuangntu dismissed stale reviews from jichangjichang and KKyang via c531281 January 16, 2025 07:15
@waynehuangntu
Copy link
Contributor Author

Fixed bugs when tuning grouped gemm.
Will provide tenslitelite tox test result soon.

@waynehuangntu
Copy link
Contributor Author

Main Update: For frequency retrieval, first attempt using hipDeviceGetAttribute. If it fails, fall back to rocm-smi -s. If that also fails, prompt the user to manually input the frequency.

Below is the result of tox:
image

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@jichangjichang jichangjichang jichangjichang approved these changes

@KKyang KKyang KKyang left review comments

@vin-huang vin-huang Awaiting requested review from vin-huang vin-huang is a code owner

@imcarsonliao imcarsonliao Awaiting requested review from imcarsonliao imcarsonliao is a code owner

@hcman2 hcman2 Awaiting requested review from hcman2 hcman2 is a code owner

@Serge45 Serge45 Awaiting requested review from Serge45 Serge45 is a code owner

@Jinp800125 Jinp800125 Awaiting requested review from Jinp800125 Jinp800125 is a code owner

@TonyYHsieh TonyYHsieh Awaiting requested review from TonyYHsieh TonyYHsieh is a code owner

@solaslin solaslin Awaiting requested review from solaslin solaslin is a code owner

At least 2 approving reviews are required to merge this pull request.

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@waynehuangntu @vin-huang @jichangjichang @KKyang