Lightning-AI / Lightning-AI/lightning-thunder

Propagate tag information throughout a trace lifetime

Open
#1,199 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

tracing architecture tracing time
Dominant language
Python
Stars
1.5k
Forks
121
PR merge metrics
No merged PRs in 30d

Description

## 🐛 Bug

PR #1048 introduced Proxy object tagging. However, when a trace is manipulated the tag information is lost, how can we make sure that the tags are not discarded?

As of now there are many ways to create and manipulate traces, does it make sense here to start a simplification of the UX and declare a default way to create and manipulate a trace so we have more control over it and we can enforce this kind of constraints(keeping the same tag)?

### Expected behavior

When setting a tag, that tag should always accompany its proxy even after transformations.

cc @apaz-cli, @t-vi

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading PR #1048 to understand Proxy object tagging, then trace the documented ways traces are created and manipulated. Determine where tag information is discarded and verify that a tag remains attached to its proxy after each transformation; the issue does not name specific files or tests.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
compilers
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.