Refactor and Optimize trace_link.py for Enhanced Readability and Performance #90
Conversation
TaekyungHeo
requested review from
kingchc,
louisfeng,
sunghlin,
wfu-fb,
shengbao-zheng and
briancoutinho
as code owners
|
Hi @TaekyungHeo! Thank you for your pull request. We require contributors to sign our Contributor License Agreement, and yours needs attention. You currently have a record in our system, but the CLA is no longer valid, and will need to be resubmitted. ProcessIn order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA. Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with If you have received this in error or have any questions, please contact us at [email protected]. Thanks! |
TaekyungHeo
changed the title
TaekyungHeo
force-pushed
the
refactor
branch
2 times, most recently
from
8596483 to
6c6524c
Compare
facebook-github-bot
added
the
CLA Signed
TaekyungHeo
force-pushed
the
refactor
branch
from
f8ec036 to
e709784
Compare
|
@briancoutinho has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@TaekyungHeo I'm not sure why but import failed :( One request can you try to rebase on master |
|
@TaekyungHeo has updated the pull request. You must reimport the pull request before landing. |
|
Updated, @briancoutinho . |
|
@TaekyungHeo has updated the pull request. You must reimport the pull request before landing. |
The `handle_kineto_segmentation` function is intended to support kineto traces cross multiple iterations by splitting a trace into several segments according to the provided annotations. Unfortunately, this function is not operating as expected, leading to errors. It is advisable to remove it.
The multi-iteration support feature for PyTorch execution traces is designed to facilitate the handling of traces over multiple iterations. Unfortunately, this feature is not functioning as expected and is leading to errors. It is advisable to remove it.
This commit introduces support for inter-thread dependencies within the Chakra framework. By examining Kineto traces via chrome://tracing, one can observe multiple CPU threads and their implicit dependencies. This update explicitly encodes these dependencies in the output trace, enabling accurate handling by subsequent tools.
This commit adds stream ID encoding to GPU operators. This ensures that all operators within the same stream are executed in the correct order, supporting intra-stream dependencies.
Introduced exclusive duration calculation for Kineto operators in the TraceLinker class. This update differentiates between inclusive and exclusive durations, providing a clearer distinction in the profiling data. Exclusive durations are now calculated to identify the actual time spent in individual operations, excluding overlaps with child operators.
|
@TaekyungHeo has updated the pull request. You must reimport the pull request before landing. |
|
@TaekyungHeo has updated the pull request. You must reimport the pull request before landing. |
|
Move trace_link.py to mlcommons/chakra based on the discussion between Meta and NVIDIA. (#100) |
Summary
This pull request introduces significant improvements to
trace_link.py. The changes aim to enhance the readability, maintainability, and performance of the codebase, addressing several issues identified in the previous implementation.Key highlights of this pull request include:
trace_link.pyto improve code readability and maintainability. The refactoring effort focused on restructuring the code for better understanding and future development, accompanied by comprehensive comments and logging messages to facilitate easier debugging and maintenance.