Make require_contiguous require exact strides instead of stride order #148235

eellison · 2025-03-01T01:37:51Z

Stack from ghstack (oldest at bottom):

-> Make require_contiguous require exact strides instead of stride order #148235

cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov

[ghstack-poisoned]

pytorch-bot · 2025-03-01T01:37:55Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/148235

📄 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 or our office hours

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

❌ 24 New Failures, 1 Unrelated Failure

As of commit 3d8508c with merge base 4e160d5 ():

NEW FAILURES - The following jobs have failed:

inductor / unit-test / cuda12.4-py3.10-gcc9-sm86 / test (inductor, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleCpu::test_freezing_cpu
inductor / unit-test / cuda12.4-py3.10-gcc9-sm86 / test (inductor, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleCpu::test_linear_freezing_cpu
inductor / unit-test / cuda12.4-py3.12-gcc9-sm86 / test (inductor, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleCpu::test_freezing_cpu
inductor / unit-test / cuda12.4-py3.12-gcc9-sm86 / test (inductor, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleCpu::test_linear_freezing_cpu
inductor / unit-test / cuda12.4-py3.13-gcc9-sm86 / test (inductor, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleCpu::test_freezing_cpu
inductor / unit-test / cuda12.4-py3.13-gcc9-sm86 / test (inductor, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleCpu::test_linear_freezing_cpu
inductor-rocm / rocm6.3-py3.10-inductor / test (inductor, 1, 2, linux.rocm.gpu.2) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleCpu::test_linear_freezing_cpu
Lint / lintrunner-noclang / linux-job (gh)
>>> Lint for torch/_inductor/ir.py:
pull / linux-focal-cuda12.4-py3.10-gcc9 / test (default, 2, 5, linux.4xlarge.nvidia.gpu) (gh)
inductor/test_aot_inductor_arrayref.py::AOTInductorTestABICompatibleCpuWithStackAllocation::test_freezing_cpu_with_stack_allocation
pull / linux-focal-cuda12.4-py3.10-gcc9 / test (default, 4, 5, linux.4xlarge.nvidia.gpu) (gh)
inductor/test_mkldnn_pattern_matcher.py::TestPatternMatcher::test_linear_relu_dynamic_fp16
pull / linux-focal-cuda12.4-py3.10-gcc9 / test (default, 5, 5, linux.4xlarge.nvidia.gpu) (gh)
inductor/test_mkldnn_pattern_matcher.py::TestPatternMatcher::test_linear_dynamic_fp16
pull / linux-focal-cuda12.4-py3.10-gcc9-sm89 / test (default, 2, 5, linux.g6.4xlarge.experimental.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleCpu::test_freezing_cpu
pull / linux-focal-cuda12.4-py3.10-gcc9-sm89 / test (default, 3, 5, linux.g6.4xlarge.experimental.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleCpu::test_linear_freezing_cpu
pull / linux-focal-cuda12.4-py3.10-gcc9-sm89 / test (default, 4, 5, linux.g6.4xlarge.experimental.nvidia.gpu) (gh)
inductor/test_aot_inductor_arrayref.py::AOTInductorTestABICompatibleCpuWithStackAllocation::test_freezing_cpu_with_stack_allocation
pull / linux-focal-cuda12.4-py3.10-gcc9-sm89 / test (default, 5, 5, linux.g6.4xlarge.experimental.nvidia.gpu) (gh)
inductor/test_mkldnn_pattern_matcher.py::TestPatternMatcher::test_linear_relu_dynamic_fp16
pull / linux-focal-py3.13-clang10 / test (default, 3, 5, linux.4xlarge) (gh)
inductor/test_mkldnn_pattern_matcher.py::TestPatternMatcher::test_linear_relu_dynamic_fp16
pull / linux-focal-py3.13-clang10 / test (default, 4, 5, linux.4xlarge) (gh)
inductor/test_mkldnn_pattern_matcher.py::TestPatternMatcher::test_linear_add_bias
pull / linux-focal-py3.9-clang10 / test (default, 4, 5, linux.4xlarge) (gh)
inductor/test_aot_inductor_arrayref.py::AOTInductorTestABICompatibleCpuWithStackAllocation::test_freezing_cpu_with_stack_allocation
pull / linux-focal-py3.9-clang10 / test (default, 5, 5, linux.4xlarge) (gh)
inductor/test_mkldnn_pattern_matcher.py::TestPatternMatcher::test_linear_relu_dynamic_fp16
pull / linux-jammy-py3.10-clang15-asan / test (default, 3, 6, linux.4xlarge) (gh)
inductor/test_aot_inductor_arrayref.py::AOTInductorTestABICompatibleCpuWithStackAllocation::test_freezing_cpu_with_stack_allocation
pull / linux-jammy-py3.10-clang15-asan / test (default, 5, 6, linux.4xlarge) (gh)
inductor/test_mkldnn_pattern_matcher.py::TestPatternMatcher::test_linear_unary
pull / linux-jammy-py3.10-clang15-asan / test (default, 6, 6, linux.4xlarge) (gh)
inductor/test_mkldnn_pattern_matcher.py::TestPatternMatcher::test_linear_relu_dynamic_fp16
pull / linux-jammy-py3.9-gcc11 / test (default, 2, 5, linux.2xlarge) (gh)
inductor/test_mkldnn_pattern_matcher.py::TestPatternMatcher::test_linear_binary
pull / linux-jammy-py3.9-gcc11 / test (default, 4, 5, linux.2xlarge) (gh)
inductor/test_mkldnn_pattern_matcher.py::TestPatternMatcher::test_linear_add_bias

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

inductor / linux-jammy-cpu-py3.9-gcc11-inductor / test (inductor_torchbench_cpu_smoketest_perf, 1, 1, linux.24xl.spr-metal) (gh) (similar failure)
Process completed with exit code 1.

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

ghstack-source-id: 34a8ccc Pull Request resolved: #148235

[ghstack-poisoned]

ghstack-source-id: 7d1223a Pull Request resolved: #148235

eellison · 2025-03-03T17:06:01Z

cc @leslie-fang-intel I made this pr after our difficulties with ir.require_contiguous in #146746. now i'm seeing a number of mkldnn/cpu errors. would you mind debugging ?

leslie-fang-intel · 2025-03-04T03:00:25Z

I think it may relate to mkldnn tensor, which seems fail to get the data ptr. cc @CaoE will help on this.

ghstack-source-id: 7d1223a Pull Request resolved: pytorch#148235

CaoE · 2025-03-04T06:12:17Z

Hi @eellison This is because the mkldnn tensor returns an invalid stride and fail to get the data ptr. I added a check for mkldnn tensor to see if it can fix these failures #148424.

ghstack-source-id: 7d1223a Pull Request resolved: pytorch#148235

eellison · 2025-03-13T17:13:08Z

any updates @CaoE and @leslie-fang-intel ?

leslie-fang-intel · 2025-03-14T00:27:09Z

any updates @CaoE and @leslie-fang-intel ?

Hi @CaoE, I thought we already fix it. Doesn't we?

CaoE · 2025-03-14T01:56:13Z

Hi @eellison @leslie-fang-intel #148424 This test shows adding checks for mkldnn tensor could fix it.
How to place these checks still needs to be confirmed #148424 (comment).

ghstack-source-id: 7d1223a Pull Request resolved: #148235

github-actions · 2025-05-13T02:13:30Z

Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as Stale.
Feel free to remove the Stale label if you feel this was a mistake.
If you are unable to remove the Stale label please contact a maintainer in order to do so.
If you want the bot to never mark this PR stale again, add the no-stale label.
Stale pull requests will automatically be closed after 30 days of inactivity.

Update

43bc90d

[ghstack-poisoned]

pytorch-bot bot added ciflow/inductor module: inductor labels Mar 1, 2025

eellison added a commit that referenced this pull request Mar 1, 2025

Make require_contiguous require exact strides instead of stride order

a7687e3

ghstack-source-id: 34a8ccc Pull Request resolved: #148235

eellison added the topic: not user facing topic category label Mar 1, 2025

Update

3d8508c

[ghstack-poisoned]

eellison added a commit that referenced this pull request Mar 1, 2025

Make require_contiguous require exact strides instead of stride order

190182c

ghstack-source-id: 7d1223a Pull Request resolved: #148235

CaoE pushed a commit to CaoE/pytorch that referenced this pull request Mar 4, 2025

Make require_contiguous require exact strides instead of stride order

7124a0a

ghstack-source-id: 7d1223a Pull Request resolved: pytorch#148235

CaoE pushed a commit to CaoE/pytorch that referenced this pull request Mar 5, 2025

Make require_contiguous require exact strides instead of stride order

abd7683

ghstack-source-id: 7d1223a Pull Request resolved: pytorch#148235

pytorch-bot bot pushed a commit that referenced this pull request Mar 14, 2025

Make require_contiguous require exact strides instead of stride order

b577583

ghstack-source-id: 7d1223a Pull Request resolved: #148235

pytorch-bot bot pushed a commit that referenced this pull request May 5, 2025

Make require_contiguous require exact strides instead of stride order

6d0f10e

ghstack-source-id: 7d1223a Pull Request resolved: #148235

eellison mentioned this pull request May 6, 2025

Add torch._C.Tag.needs_contiguous_strides #152859

Closed

pytorch-bot bot pushed a commit that referenced this pull request May 12, 2025

Make require_contiguous require exact strides instead of stride order

4570cee

ghstack-source-id: 7d1223a Pull Request resolved: #148235

github-actions bot added the Stale label May 13, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Make require_contiguous require exact strides instead of stride order #148235

Make require_contiguous require exact strides instead of stride order #148235

Make require_contiguous require exact strides instead of stride order #148235

Are you sure you want to change the base?

Make require_contiguous require exact strides instead of stride order #148235

Conversation

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/148235

❌ 24 New Failures, 1 Unrelated Failure