Brister/always tiled reduction #144008

blaine-rister · 2024-12-31T00:08:52Z

Fixes #ISSUE_NUMBER

Test the CI with tiled reductions always on. This might catch some bugs.

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

…to brister/prefer_tiling

pytorch-bot · 2024-12-31T00:08:56Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/144008

📄 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.

❌ 18 New Failures

As of commit 7b39569 with merge base d88a8c4 ():

NEW FAILURES - The following jobs have failed:

Check Labels / Check labels (gh)
RuntimeError: Error checking labels: PR does not have required labels
inductor / cuda12.4-py3.10-gcc9-sm86 / test (inductor_huggingface, 1, 1, linux.g5.4xlarge.nvidia.gpu) (gh)
YituTechConvBert
inductor / cuda12.4-py3.10-gcc9-sm86 / test (inductor_timm, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
eca_halonext26ts
inductor / cuda12.4-py3.10-gcc9-sm86 / test (inductor_timm, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
swin_base_patch4_window7_224
inductor / cuda12.4-py3.10-gcc9-sm86 / test (inductor_torchbench, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
hf_T5_generate
inductor / cuda12.4-py3.10-gcc9-sm86 / test (inductor_torchbench, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
pyhpc_isoneutral_mixing
inductor / unit-test / cuda12.4-py3.10-gcc9-sm86 / test (inductor_cpp_wrapper, 1, 1, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor.py::GPUTests::test_mutations_loop_fusion_cuda
inductor / unit-test / cuda12.4-py3.10-gcc9-sm86 / test (inductor_distributed, 1, 1, linux.g5.12xlarge.nvidia.gpu) (gh)
distributed/_composable/fsdp/test_fully_shard_compile.py::TestFullyShardCompile::test_compiled_autograd_ctx
inductor / unit-test / cuda12.4-py3.10-gcc9-sm86 / test (inductor, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleGpu::test_aoti_debug_printer_sym_inputs_cuda
inductor / unit-test / cuda12.4-py3.10-gcc9-sm86 / test (inductor, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor.py::GPUTests::test_bucketize_add_autotune_cuda
inductor / unit-test / cuda12.4-py3.12-gcc9-sm86 / test (inductor, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleGpu::test_aoti_debug_printer_sym_inputs_cuda
inductor / unit-test / cuda12.4-py3.12-gcc9-sm86 / test (inductor, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor.py::GPUTests::test_aoti_eager_support_out_cuda
inductor / unit-test / cuda12.4-py3.13-gcc9-sm86 / test (inductor, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleGpu::test_aoti_debug_printer_sym_inputs_cuda
inductor / unit-test / cuda12.4-py3.13-gcc9-sm86 / test (inductor, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor.py::GPUTests::test_aoti_eager_support_out_cuda
inductor / unit-test / linux-jammy-cpu-py3.12-gcc11-inductor-halide / test (inductor-halide, 1, 1, linux.12xlarge) (gh)
inductor/test_halide.py::CpuHalideTests::test_index_propagation_nested_indirect_indexing_cpu
inductor / unit-test / linux-jammy-cpu-py3.12-gcc11-inductor-triton-cpu / test (inductor-triton-cpu, 1, 1, linux.12xlarge) (gh)
inductor/test_triton_cpu_backend.py::CpuTritonTests::test_as_strided_scatter_cpu
inductor-rocm / rocm6.2-py3.10-inductor / test (inductor, 1, 2, linux.rocm.gpu.2) (gh)
inductor/test_aot_inductor.py::AOTInductorTestABICompatibleGpu::test_aoti_debug_printer_sym_inputs_cuda
inductor-rocm / rocm6.2-py3.10-inductor / test (inductor, 2, 2, linux.rocm.gpu.2) (gh)
inductor/test_torchinductor_opinfo.py::TestInductorOpInfoCUDA::test_comprehensive__unsafe_masked_index_put_accumulate_cuda_float32

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

github-actions · 2024-12-31T00:10:36Z

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.

…ction

github-actions · 2025-03-02T07:34:04Z

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.

blaine-rister and others added 30 commits July 26, 2024 00:22

draft: prefer tiling to higher dimensions

b88c192

fix none dense stride

6fa6c69

add unit test; try to fix for none order path

7de6dff

fix passing [] rather than None

1fbf80b

fix bug for test_index_put_reinplace. return x for actual_strides too

1b68e78

draft: prefer tiling to higher dimensions

462c87a

bug fix

cfc0d15

Merge branch 'brister/prefer_tiling' of github.com:pytorch/pytorch in…

11420e4

…to brister/prefer_tiling

fix 3D leading dims

8183766

remove successful tests

8362008

Merge remote-tracking branch 'origin/main' into brister/prefer_tiling

5f0c245

update tests for ND pointwise. Tiled reduction is out of scope

29bddfa

pass type checking, parametrize a few more tests

762edd2

fix test

bac7d50

Add a test for 5D tensors

6e502de

fix type annotations for python 3.8

4e1b398

sort tilings in descending order

8543226

remove unexpected success; fix bug for channel_last

a313b78

skip non-dense+reinterpretview; add list convert

607a229

remove unexpected success

6d0afc9

add a new API require_exact_strides

6419105

for view operation, it's ok the stride is not equal.

1ab9ccd

add conversion for symint and symmul

1b5cb3f

Merge branch 'brister/prefer_tiling' into brister/custom_strides_pass

4a77ba9

fix none dense stride

bf01a24

add unit test; try to fix for none order path

273aa7c

fix passing [] rather than None

350a88e

fix bug for test_index_put_reinplace. return x for actual_strides too

d8381d7

remove successful tests

0a003d3

remove unexpected success; fix bug for channel_last

60be746

blaine-rister added 16 commits December 13, 2024 18:03

Merge branch 'main' into brister/tiled_reduction

d533470

single inheritance

fbf346e

inline helper functions which are only used once

6da703c

ordered dict to avoid ND

29495a9

revert CantSplit change

5510059

simplify 3d tiling

b885fd1

revert is_compatible_with_numels

76ed913

revert mask changes

16d1589

revert format

0dbc5ea

restore type hint

98740ac

revert format

931ecb9

generalize index expression tiling search

04b7a65

add fusion test

4756597

add another multi reduction test

a3b1bff

Merge branch 'main' into brister/tiled_reduction

a46235d

turn tiled reductions always on

8223cab

pytorch-bot bot added ciflow/inductor module: inductor labels Dec 31, 2024

blaine-rister mentioned this pull request Dec 31, 2024

[Inductor] Support tiling reduction dimensions #137243

Closed

blaine-rister added 5 commits December 31, 2024 00:40

prototype using global rnumel. Doesn't work yet

0def433

use reduction_numel to complete partial tiling

a960fb1

Merge branch 'brister/reduction_numel' into brister/always_tiled_redu…

c324ed0

…ction

fix tiling completion

305c4db

Merge branch 'brister/reduction_numel' into brister/always_tiled_redu…

7b39569

…ction

github-actions bot added the Stale label Mar 2, 2025

blaine-rister closed this Mar 25, 2025

github-actions bot deleted the brister/always_tiled_reduction branch April 27, 2025 02:18

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Brister/always tiled reduction #144008

Brister/always tiled reduction #144008

Uh oh!

blaine-rister commented Dec 31, 2024 •

edited by pytorch-bot bot

Loading

Uh oh!

pytorch-bot bot commented Dec 31, 2024 •

edited

Loading

Uh oh!

github-actions bot commented Dec 31, 2024

Uh oh!

github-actions bot commented Mar 2, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Brister/always tiled reduction #144008

Brister/always tiled reduction #144008

Uh oh!

Conversation

blaine-rister commented Dec 31, 2024 • edited by pytorch-bot bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented Dec 31, 2024 • 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/144008

❌ 18 New Failures

Uh oh!

github-actions bot commented Dec 31, 2024

This PR needs a release notes: label

Uh oh!

github-actions bot commented Mar 2, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

blaine-rister commented Dec 31, 2024 •

edited by pytorch-bot bot

Loading

pytorch-bot bot commented Dec 31, 2024 •

edited

Loading

This PR needs a `release notes:` label