[compiler toolkit] Port joint_ac_pass from simplefsdp #2051
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
meta-cla
bot
added
the
CLA Signed
yiming0416
force-pushed
the
yiming/add_reshard_after_forward_ac_pass
branch
from
0d70d9e to
88c79b6
Compare
ruisizhang123
approved these changes
Nov 18, 2025
| return gm | ||
|
|
||
|
|
||
| # Apply activation checkpointing on joint graph before partitioner |
Contributor
There was a problem hiding this comment.
We use the "joint_ac_pass" here to implement FSDP's reshard_after_forward. Maybe better to rename it to fsdp_reshard_after_fwd_pass?
| ) | ||
|
|
||
| joint_custom_passes = [] | ||
| joint_custom_passes.append(validate_flex_attn_annotation_pass) |
Contributor
There was a problem hiding this comment.
just out of curiosity: do we care about the order of passes applied in joint_custom_passes?
Contributor
Author
There was a problem hiding this comment.
we probably care the order of passes when later we have more joint_custom_passes.
Right now the validate_flex_attn_annotation_pass is just validating if flex attention nodes are annotated and will error out if it fails. So I put it at first.
yiming0416
force-pushed
the
yiming/add_reshard_after_forward_ac_pass
branch
from
88c79b6 to
528b319
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR integrates the changes in #1970 to compiler toolkit (applying
joint_ac_passon the joint graph graph to tag nodes based onreshard_after_forwardflag)Also did some refactor for applying graph passes in compiler toolkit experiments. We will have two kinds of passes
joint_custom_passes: these are passes to be applied on the captured joint graph before partitioner. By default we
validate_flex_attn_annotation_passandfsdp_reshard_after_fwd_passcompiler_passes: there are passes to be applied on partitioned fwd and bwd graphs as backend optimizations. By default there is none. We can indicate
autobucketing_reordering_passandregional_inductor_passusing configs.