[DTensor] expand_to_full_mesh_op_strategy filters mixed partials by wconstab · Pull Request #173614 · pytorch/pytorch

wconstab · 2026-01-28T03:49:51Z

Stack from ghstack (oldest at bottom):

-> [DTensor] expand_to_full_mesh_op_strategy filters mixed partials #173614

Per #172609, we do not allow
mixed partial placements in one DTensor spec. It is therefore pointless
to generate strategies that encourage mixed partial types.

Before this PR, any OP strategy that has a rule for more than one
partial type and uses the expand util would have combined them
inappropriately. After this PR, the expansion util filters out any such
combinations.

Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. [ghstack-poisoned]

pytorch-bot · 2026-01-28T03:49:55Z

This PR needs a `release notes:` label

If your changes are user facing and intended to be a part of release notes, please use a label starting with release notes:.

If not, please add the topic: not user facing label.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "topic: not user facing"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

pytorch-bot · 2026-01-28T03:49:57Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/173614

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki

Note: Links to docs will display an error until the docs builds have been completed.

❌ 2 Cancelled Jobs, 4 Unrelated Failures

As of commit 21537d4 with merge base 4c6817d ():

CANCELLED JOBS - The following jobs were cancelled. Please retry:

inductor / unit-test / inductor-cpu-core-build (3.12) / build (gh)
The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
inductor / unit-test / inductor-halide-build / build (gh)
The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.

FLAKY - The following job failed but was likely due to flakiness present on trunk:

inductor / unit-test / inductor-test / test (inductor, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (similar failure)
test/inductor/test_torchinductor_opinfo.py::TestInductorOpInfoCUDA::test_comprehensive_nn_functional_linear_cuda_float32

BROKEN TRUNK - The following jobs failed but was present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

inductor / inductor-test / test (inductor_torchbench, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (trunk failure)
pytorch_CycleGAN_and_pix2pix
inductor / unit-test / inductor-test / test (inductor, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (trunk failure)
test/inductor/test_torchinductor_opinfo.py::TestInductorOpInfoCUDA::test_comprehensive_pca_lowrank_cuda_float32

UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:

inductor / inductor-test-cuda13 / test (inductor_torchbench, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (#174930)
pytorch_CycleGAN_and_pix2pix

This comment was automatically generated by Dr. CI and updates every 15 minutes.

Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. ghstack-source-id: eb4a927 Pull Request resolved: #173614

…rtials" Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. [ghstack-poisoned]

Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. ghstack-source-id: 8cca0a6 Pull Request resolved: #173614

…rtials" Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. [ghstack-poisoned]

Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. ghstack-source-id: 2984751 Pull Request resolved: #173614

pianpwk · 2026-02-10T00:15:42Z

+        for spec in spec_list:
+            if spec is not None:
+                partial_reduce_ops = {
+                    p.reduce_op for p in spec.placements if isinstance(p, Partial)


maybe a weird case, but I would be wary of NormPartial() and P(sum) having the same reduce_op attribute

well i thought we were deleting NormPartial soon? but if not i should harden this.

@pianpwk i've updated to cover this.

…rtials" Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. [ghstack-poisoned]

Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. ghstack-source-id: 2483827 Pull Request resolved: #173614

…rtials" Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. [ghstack-poisoned]

Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. ghstack-source-id: e4407b4 Pull Request resolved: #173614

…rtials" Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. [ghstack-poisoned]

Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. ghstack-source-id: 38e8412 Pull Request resolved: #173614

wconstab · 2026-02-18T20:51:59Z

@pytorchbot merge -i

pytorchmergebot · 2026-02-18T20:54:23Z

Merge started

Your change will be merged while ignoring the following 4 checks: inductor / inductor-cpu-build / build, inductor / unit-test / inductor-halide-build / build, inductor / unit-test / inductor-cpu-core-build (3.12) / build, inductor / inductor-test-cuda13 / test (inductor_torchbench, 2, 2, linux.g5.4xlarge.nvidia.gpu)

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

…3614) Per #172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. Pull Request resolved: #173614 Approved by: https://github.com/pianpwk, https://github.com/zpcore

…orch#173614) Per pytorch#172609, we do not allow mixed partial placements in one DTensor spec. It is therefore pointless to generate strategies that encourage mixed partial types. Before this PR, any OP strategy that has a rule for more than one partial type and uses the expand util would have combined them inappropriately. After this PR, the expansion util filters out any such combinations. Pull Request resolved: pytorch#173614 Approved by: https://github.com/pianpwk, https://github.com/zpcore

This was referenced Jan 28, 2026

[DTensor] single-dim expander raises clear inplace error #173572

Closed

[DTensor] Update TP api to support single-dim strategies #173567

Closed

wconstab mentioned this pull request Jan 28, 2026

[DTensor] Complete single-dim pointwise rule #172278

Closed

pytorch-bot Bot added the ciflow/inductor label Jan 28, 2026

pytorch-bot Bot added the release notes: distributed (dtensor) release notes category label Jan 28, 2026

wconstab requested review from pianpwk and zpcore January 29, 2026 17:12

pianpwk approved these changes Feb 10, 2026

View reviewed changes

zpcore approved these changes Feb 10, 2026

View reviewed changes

pytorch-bot Bot added the ciflow/trunk Trigger trunk jobs on your pull request label Feb 18, 2026

pytorchmergebot added the merging label Feb 18, 2026

pytorchmergebot added the Merged label Feb 18, 2026

pytorchmergebot closed this in 372ae3c Feb 18, 2026

pytorchmergebot removed the merging label Feb 18, 2026

github-actions Bot deleted the gh/wconstab/510/head branch March 21, 2026 02:24

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[DTensor] expand_to_full_mesh_op_strategy filters mixed partials#173614

[DTensor] expand_to_full_mesh_op_strategy filters mixed partials#173614
wconstab wants to merge 6 commits intogh/wconstab/510/basefrom
gh/wconstab/510/head

wconstab commented Jan 28, 2026 •

edited

Loading

Uh oh!

pytorch-bot Bot commented Jan 28, 2026

Uh oh!

pytorch-bot Bot commented Jan 28, 2026 •

edited

Loading

Uh oh!

pianpwk Feb 10, 2026

Uh oh!

wconstab Feb 10, 2026

Uh oh!

wconstab Feb 18, 2026 •

edited

Loading

Uh oh!

wconstab commented Feb 18, 2026

Uh oh!

pytorchmergebot commented Feb 18, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

wconstab commented Jan 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot Bot commented Jan 28, 2026

This PR needs a release notes: label

Uh oh!

pytorch-bot Bot commented Jan 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/173614

❌ 2 Cancelled Jobs, 4 Unrelated Failures

Uh oh!

pianpwk Feb 10, 2026

Choose a reason for hiding this comment

Uh oh!

wconstab Feb 10, 2026

Choose a reason for hiding this comment

Uh oh!

wconstab Feb 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

wconstab commented Feb 18, 2026

Uh oh!

pytorchmergebot commented Feb 18, 2026

Merge started

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

wconstab commented Jan 28, 2026 •

edited

Loading

This PR needs a `release notes:` label

pytorch-bot Bot commented Jan 28, 2026 •

edited

Loading

wconstab Feb 18, 2026 •

edited

Loading