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.

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

[Auto Parallel] Upgrade fluid comm operators to be compatible with new comm library #56088

Merged
merged 44 commits into from
Sep 8, 2023

Conversation

GhostScreaming
Copy link
Contributor

@GhostScreaming GhostScreaming commented Aug 8, 2023

PR types

Others

PR changes

Others

Description

Pcard-70448

Based on compatible new communication library upgrade PR #56604. Old communication library is used in default, export FLAGS_dynamic_static_unified_comm=1 to use new communication library.

Following operators are upgraded compatibly now: alltoall, barrier, c_concat, c_allreduce, c_sync_stream_op, c_wait_comm_op, c_wait_compute_op, recv_v2_op, send_v2_op, c_reduce_op, c_allgather_op, c_scatter_op, c_reducescatter_op.

wentaoyu and others added 5 commits August 8, 2023 20:12
@paddle-bot
Copy link

paddle-bot bot commented Aug 8, 2023

你的PR提交成功,感谢你对开源项目的贡献!
请关注后续CI自动化测试结果,详情请参考Paddle-CI手册
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

1. remove useless loggings.
2. Fix conditional compilation for HIP.
3. Fix problems of test_pass_generation_pipeline.py. It calls paddle.distributed.init_parallel_env() at first,
then auto.Engine calls _init_comm(), which will calls process_group.instantiate(). However, init_parallel_env() will call
paddle.distributed.barrier(), it will call CreateNCCLEnvCache and create corresponding NCCLCommContext. But dev_id is not
set, as a result, NCCLCommContext's dev_ctx is not initialized.
@GhostScreaming GhostScreaming changed the title [WIP] compatible new comm library upgrade [Auto Parallel] Upgrade fluid communication operators to be compatible with new comm library Aug 24, 2023
@GhostScreaming GhostScreaming changed the title [Auto Parallel] Upgrade fluid communication operators to be compatible with new comm library [Auto Parallel] Upgrade fluid comm operators to be compatible with new comm library Aug 24, 2023
Copy link

@hitywt hitywt left a comment

Choose a reason for hiding this comment

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

LGTM

zhiqiu
zhiqiu previously approved these changes Sep 6, 2023
Copy link
Contributor

@zhiqiu zhiqiu left a comment

Choose a reason for hiding this comment

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

LGTM

Copy link
Contributor

@luotao1 luotao1 left a comment

Choose a reason for hiding this comment

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

LGTM for flags

Copy link
Contributor

@XieYunshen XieYunshen left a comment

Choose a reason for hiding this comment

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

LGTM

Copy link
Contributor

@zhiqiu zhiqiu left a comment

Choose a reason for hiding this comment

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

LGTM

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants