[graph_trainer][aot_fx_trace][graph_pp] Add dI/dW backward splitting#2727
Merged
Conversation
sanketpurandare
pushed a commit
that referenced
this pull request
Mar 27, 2026
stack-info: PR: #2727, branch: sanketpurandare/stack/6
sanketpurandare
force-pushed
the
sanketpurandare/stack/6
branch
from
b54d197 to
bed5c29
Compare
sanketpurandare
force-pushed
the
sanketpurandare/stack/5
branch
from
3e47d66 to
19150f9
Compare
This was referenced Mar 27, 2026
meta-cla
Bot
added
the
CLA Signed
sanketpurandare
requested review from
fegin,
tianyu-l,
wconstab and
wwwjn
as code owners
sanketpurandare
pushed a commit
that referenced
this pull request
Mar 27, 2026
stack-info: PR: #2727, branch: sanketpurandare/stack/6
sanketpurandare
force-pushed
the
sanketpurandare/stack/6
branch
from
bed5c29 to
e6c34d8
Compare
sanketpurandare
pushed a commit
that referenced
this pull request
Mar 27, 2026
stack-info: PR: #2727, branch: sanketpurandare/stack/6
sanketpurandare
force-pushed
the
sanketpurandare/stack/6
branch
from
e6c34d8 to
89371b2
Compare
sanketpurandare
pushed a commit
that referenced
this pull request
Apr 1, 2026
stack-info: PR: #2727, branch: sanketpurandare/stack/6
sanketpurandare
force-pushed
the
sanketpurandare/stack/6
branch
from
89371b2 to
baa2a42
Compare
sanketpurandare
force-pushed
the
sanketpurandare/stack/6
branch
from
baa2a42 to
0a6ed49
Compare
sanketpurandare
force-pushed
the
sanketpurandare/stack/6
branch
from
caeee4b to
37de310
Compare
sanketpurandare
force-pushed
the
sanketpurandare/stack/5
branch
from
74de9a6 to
398c43f
Compare
sanketpurandare
added a commit
that referenced
this pull request
May 8, 2026
stack-info: PR: #2727, branch: sanketpurandare/stack/6
sanketpurandare
force-pushed
the
sanketpurandare/stack/6
branch
from
37de310 to
47ebb7c
Compare
sanketpurandare
force-pushed
the
sanketpurandare/stack/6
branch
from
47ebb7c to
a6db1e3
Compare
sanketpurandare
requested review from
IvanKobzarev,
SherlockNoMad,
aditvenk and
xmfan
as code owners
aditvenk
approved these changes
Jun 23, 2026
aditvenk
left a comment
aditvenk
left a comment
Contributor
There was a problem hiding this comment.
LGTM -- some minor comments.
Add the GraphPP backward split pass for schedules that execute input-gradient and weight-gradient work as separate PP actions. The pass preserves the flat graph calling convention selected by the partitioner and returns explicit activation handoff metadata for the dW graph. Add unit coverage that reconstructs the full backward result from split dI and dW graphs and validates the no-op path when a graph has no input-gradient outputs. stack-info: PR: #2727, branch: sanketpurandare/stack/6
This was referenced Jun 30, 2026
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Stacked PRs:
[graph_trainer][aot_fx_trace][graph_pp] Add dI/dW backward splitting
Add the GraphPP backward split pass for schedules that execute input-gradient and weight-gradient work as separate PP actions. The pass preserves the flat graph calling convention selected by the partitioner and returns explicit activation handoff metadata for the dW graph.
Add unit coverage that reconstructs the full backward result from split dI and dW graphs and validates the no-op path when a graph has no input-gradient outputs.