Make run_dtensor_rng_op compatible with compile_on_one_rank#177447
Closed
aorenste wants to merge 9 commits intogh/aorenste/220/basefrom
Closed
Make run_dtensor_rng_op compatible with compile_on_one_rank#177447aorenste wants to merge 9 commits intogh/aorenste/220/basefrom
aorenste wants to merge 9 commits intogh/aorenste/220/basefrom
Conversation
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/177447
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 1 Pending, 2 Unrelated FailuresAs of commit 1c8fbdc with merge base 417a890 ( NEW FAILURE - The following job has failed:
BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Mar 14, 2026
aorenste
added a commit
that referenced
this pull request
Mar 14, 2026
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. ghstack-source-id: 71bd32b Pull Request resolved: #177447
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. [ghstack-poisoned]
aorenste
added a commit
that referenced
this pull request
Mar 16, 2026
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. ghstack-source-id: 7a7bbae Pull Request resolved: #177447
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. [ghstack-poisoned]
aorenste
added a commit
that referenced
this pull request
Mar 16, 2026
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. ghstack-source-id: 0f5f81d Pull Request resolved: #177447
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. [ghstack-poisoned]
aorenste
added a commit
that referenced
this pull request
Mar 17, 2026
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. ghstack-source-id: 328f827 Pull Request resolved: #177447
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. [ghstack-poisoned]
aorenste
added a commit
that referenced
this pull request
Mar 18, 2026
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. ghstack-source-id: 991e325 Pull Request resolved: #177447
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. [ghstack-poisoned]
aorenste
added a commit
that referenced
this pull request
Mar 18, 2026
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. ghstack-source-id: 52b8581 Pull Request resolved: #177447
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. [ghstack-poisoned]
aorenste
added a commit
that referenced
this pull request
Mar 18, 2026
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. ghstack-source-id: a6c27a8 Pull Request resolved: #177447
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. [ghstack-poisoned]
aorenste
added a commit
that referenced
this pull request
Mar 19, 2026
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. ghstack-source-id: 8d87e47 Pull Request resolved: #177447
yiming0416
approved these changes
Mar 19, 2026
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. [ghstack-poisoned]
aorenste
added a commit
that referenced
this pull request
Mar 19, 2026
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. ghstack-source-id: af73b79 Pull Request resolved: #177447
Contributor
Author
|
test/distributed/tensor/test_random_ops.py::DistTensorRandomOpTest::test_pipeline_parallel_manual_seed is a pre-existing failure in trunk |
Collaborator
Merge startedYour change will be merged while ignoring the following 3 checks: inductor / inductor-cpu-test / test (cpu_inductor_torchbench, 1, 2, linux.2xlarge.amx, unstable), inductor / inductor-test / test (inductor_torchbench, 2, 2, linux.g5.4xlarge.nvidia.gpu), dtensor / dtensor-test / test (dtensor, 1, 1, lf.linux.g5.12xlarge.nvidia.gpu) Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
AaronWang04
pushed a commit
to AaronWang04/pytorch
that referenced
this pull request
Mar 31, 2026
…177447) Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. Pull Request resolved: pytorch#177447 Approved by: https://github.com/yiming0416 ghstack dependencies: pytorch#177446
nklshy-aws
pushed a commit
to nklshy-aws/pytorch
that referenced
this pull request
Apr 7, 2026
…177447) Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim) when compile_on_one_rank is active. Previously, mesh.get_coordinate() returned concrete rank-specific integers that got baked into the compiled graph, producing different graphs on different ranks. Also refactors test_compile_on_one_rank.py to extract graph-comparison helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph) and adds a test for DTensor random op graph consistency. Authored with Claude. Pull Request resolved: pytorch#177447 Approved by: https://github.com/yiming0416 ghstack dependencies: pytorch#177446
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
3 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.
Stack from ghstack (oldest at bottom):
Use mesh._sym_get_coordinate() in _compute_rng_offsets so that RNG
offset values become symbolic SymInts (via _runtime_compute_coordinate_on_dim)
when compile_on_one_rank is active. Previously, mesh.get_coordinate()
returned concrete rank-specific integers that got baked into the compiled
graph, producing different graphs on different ranks.
Also refactors test_compile_on_one_rank.py to extract graph-comparison
helpers (_assert_graphs_identical_across_ranks, _compile_and_capture_graph)
and adds a test for DTensor random op graph consistency.
Authored with Claude.