Skip to content

[API Compatibility] change prefix matcher for 4 distributed apis -part#879

Merged
zhwesky2010 merged 7 commits into
PaddlePaddle:masterfrom
Manfredss:ApiEnhance_all_gather_object
Jul 15, 2026
Merged

[API Compatibility] change prefix matcher for 4 distributed apis -part#879
zhwesky2010 merged 7 commits into
PaddlePaddle:masterfrom
Manfredss:ApiEnhance_all_gather_object

Conversation

@Manfredss

@Manfredss Manfredss commented May 5, 2026

Copy link
Copy Markdown
Collaborator

PR Docs

torch.distributed.all_gather_object
torch.distributed.group.WORLD
torch.distributed.init_process_group
torch.distributed.ProcessGroup

PR APIs

torch.distributed.all_gather_object
torch.distributed.group.WORLD
torch.distributed.init_process_group
torch.distributed.ProcessGroup

related PR:

@paddle-bot

paddle-bot Bot commented May 5, 2026

Copy link
Copy Markdown

Thanks for your contribution!

@paddle-bot paddle-bot Bot added the contributor External developers label May 5, 2026
@Manfredss Manfredss changed the title [API Compatibility] change prefix matcher for distributed.all_gather_object -part [API Compatibility] change prefix matcher for 4 distributed apis -part May 6, 2026
@Manfredss Manfredss self-assigned this Jul 8, 2026
@Manfredss
Manfredss requested a review from zhwesky2010 July 8, 2026 03:47
Comment thread tests/test_distributed_group_WORLD.py Outdated
g = torch.distributed.group.WORLD
"""
)
expect = textwrap.dedent(

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

注意检查下AI的代码,单测是否真正运行了?是否符合规范?

distributed系列单测统一放到tests/distributed目录下

@Manfredss
Manfredss requested a review from zhwesky2010 July 14, 2026 11:23
Comment thread tests/distributed/group_WORLD.py Outdated
rank = dist.get_rank()
torch.cuda.set_device(rank)

result = dist.get_rank(dist.group.WORLD)

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

result = dist.group.WORLD

Comment thread tests/distributed/ProgressGroup.py Outdated
rank = dist.get_rank()
torch.cuda.set_device(rank)

print(dist.ProcessGroup)

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

  1. 这个文件名不对
  2. 这个正常做一下调用

Comment thread tests/test_distributed_group_WORLD.py Outdated
obj = APIBase("torch.distributed.group.WORLD")


def test_case_1():

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

这几个测试可以删除了,分布式统一在distributed中进行测试

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

好的

rank = dist.get_rank()
torch.cuda.set_device(rank)

pg = dist.group.WORLD

@zhwesky2010 zhwesky2010 Jul 15, 2026

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

这个测试没有调用dist.ProcessGroup?先合入了,后面顺带优化下吧。

@zhwesky2010
zhwesky2010 merged commit cebd649 into PaddlePaddle:master Jul 15, 2026
8 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

contributor External developers

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants