[Deprecated][inductor] switch CppCodeCache to new cpp_builder. (take 2) #129144

xuhancn · 2024-06-20T15:36:12Z

The old one is damaged on Pre CI: #128303, this is new one.

This PR is the implemention of #124245 step 2.1

Switch CppCodeCache to new CPP builder.
Switch AotCodeCompiler to new CPP builder.
Switch HalideCodeCache to new CPP builder.

TODO:
Current still need to optimize code structure and circle import. I will do this work in further PR. I have did the exploration: #128330

cc @peterjc123 @mszhanyi @skyline75489 @nbcsm @vladimir-aubrecht @iremyux @Blackhex @cristianPanaite @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10 @ezyang @anijain2305 @chauhang @voznesenskym @penguinwu @EikanWang @Guobing-Chen @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire

pytorch-bot · 2024-06-20T15:36:16Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/129144

📄 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.

❌ 17 New Failures, 32 Unrelated Failures

As of commit d282bc0 with merge base 93a33bf ():

NEW FAILURES - The following jobs have failed:

inductor / cuda12.1-py3.10-gcc9-sm86 / test (inductor, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor.py::CpuTests::test_new_cpp_build_logical_cpu
inductor / cuda12.1-py3.12-gcc9-sm86 / test (inductor, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor.py::CpuTests::test_new_cpp_build_logical_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 4, 5, linux.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_multilayer_var_lowp_dynamic_shapes_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9-sm86 / test (default, 2, 5, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_multilayer_var_lowp_dynamic_shapes_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9-sm86 / test (default, 4, 5, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor.py::CpuTests::test_neg_index_cpu
trunk / linux-focal-cuda12.4-py3.10-gcc9-experimental-split-build-test / test (default, 1, 5, linux.4xlarge.nvidia.gpu) (gh)
inductor/test_cpu_repro.py::CPUReproTests::test_reduction_float_to_int64
trunk / linux-focal-cuda12.4-py3.10-gcc9-experimental-split-build-test / test (default, 2, 5, linux.4xlarge.nvidia.gpu) (gh)
inductor/test_cpu_repro.py::CPUReproTests::test_cpp_kernel_profile
trunk / linux-focal-cuda12.4-py3.10-gcc9-experimental-split-build-test / test (default, 4, 5, linux.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_multilayer_var_lowp_dynamic_shapes_cpu
trunk / linux-focal-cuda12.4-py3.10-gcc9-experimental-split-build-test / test (default, 5, 5, linux.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_tmp_not_defined_issue2_dynamic_shapes_cpu
trunk / linux-focal-cuda12.4-py3.10-gcc9-experimental-split-build-test / test (nogpu_AVX512, 1, 1, linux.2xlarge) (gh)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_neg_index_dynamic_shapes_cpu
trunk / linux-focal-cuda12.4-py3.10-gcc9-experimental-split-build-test / test (nogpu_NO_AVX2, 1, 1, linux.2xlarge) (gh)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_neg_index_dynamic_shapes_cpu
trunk / linux-focal-cuda12.4-py3.10-gcc9-sm86 / test (default, 2, 5, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_multilayer_var_lowp_dynamic_shapes_cpu
trunk / linux-focal-cuda12.4-py3.10-gcc9-sm86 / test (default, 3, 5, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_cpu_repro.py::CPUReproTests::test_cpp_kernel_profile
trunk / linux-focal-cuda12.4-py3.10-gcc9-sm86 / test (default, 4, 5, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor.py::CpuTests::test_neg_index_cpu
trunk / linux-focal-cuda12.4-py3.10-gcc9-sm86 / test (default, 5, 5, linux.g5.4xlarge.nvidia.gpu) (gh)
inductor/test_torchinductor_codegen_dynamic_shapes.py::DynamicShapesCodegenGPUTests::test_new_cpp_build_logical_dynamic_shapes_cuda
trunk / linux-focal-rocm6.1-py3.8 / test (default, 1, 2, linux.rocm.gpu) (gh)
inductor/test_torchinductor.py::CpuTests::test_AllenaiLongformerBase_repro_cpu
trunk / linux-focal-rocm6.1-py3.8 / test (default, 2, 2, linux.rocm.gpu) (gh)
inductor/test_torchinductor.py::CpuTests::test_neg_index_cpu

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

inductor / cuda12.1-py3.10-gcc9-sm86 / test (inductor_huggingface, 1, 1, linux.g5.4xlarge.nvidia.gpu) (gh) (matched linux rule in flaky-rules.json)
Error response from daemon: a prune operation is already running
inductor / cuda12.1-py3.12-gcc9-sm86 / test (inductor, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (matched linux rule in flaky-rules.json)
Error response from daemon: a prune operation is already running

UNSTABLE - The following jobs failed but were likely due to flakiness present on trunk and has been marked as unstable:

inductor / rocm6.1-py3.8-inductor / test (inductor, 1, 1, linux.rocm.gpu.2, unstable) (gh) (#128871)
'test/inductor/test_max_autotune.py::TestTuningProcess::test_tuning_pool_multiple_devices'
pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 1, 5, linux.4xlarge.nvidia.gpu) (gh) (related job)
inductor/test_torchinductor.py::CpuTests::test_new_cpp_build_logical_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 2, 5, linux.4xlarge.nvidia.gpu) (gh) (related job)
inductor/test_torchinductor.py::CpuTests::test_AllenaiLongformerBase_repro_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 3, 5, linux.4xlarge.nvidia.gpu) (gh) (related job)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_new_cpp_build_logical_dynamic_shapes_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 5, 5, linux.4xlarge.nvidia.gpu) (gh) (related job)
inductor/test_torchinductor_codegen_dynamic_shapes.py::DynamicShapesCodegenCpuTests::test_neg_index_dynamic_shapes_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9-experimental-split-build / test (default, 1, 5, linux.4xlarge.nvidia.gpu, unstable) (gh) (#129080)
inductor/test_torchinductor.py::CpuTests::test_new_cpp_build_logical_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9-experimental-split-build / test (default, 2, 5, linux.4xlarge.nvidia.gpu, unstable) (gh) (#129080)
inductor/test_torchinductor.py::CpuTests::test_AllenaiLongformerBase_repro_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9-experimental-split-build / test (default, 3, 5, linux.4xlarge.nvidia.gpu, unstable) (gh) (#129080)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_new_cpp_build_logical_dynamic_shapes_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9-experimental-split-build / test (default, 4, 5, linux.4xlarge.nvidia.gpu, unstable) (gh) (#129080)
inductor/test_torchinductor_codegen_dynamic_shapes.py::DynamicShapesCodegenCpuTests::test_new_cpp_build_logical_dynamic_shapes_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9-experimental-split-build / test (default, 5, 5, linux.4xlarge.nvidia.gpu, unstable) (gh) (#129080)
inductor/test_torchinductor_codegen_dynamic_shapes.py::DynamicShapesCodegenCpuTests::test_neg_index_dynamic_shapes_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9-sm86 / test (default, 3, 5, linux.g5.4xlarge.nvidia.gpu) (gh) (related job)
inductor/test_torchinductor.py::CpuTests::test_AllenaiLongformerBase_repro_cpu
pull / linux-focal-cuda12.1-py3.10-gcc9-sm86 / test (default, 5, 5, linux.g5.4xlarge.nvidia.gpu) (gh) (related job)
inductor/test_torchinductor_codegen_dynamic_shapes.py::DynamicShapesCodegenCpuTests::test_neg_index_dynamic_shapes_cpu
pull / linux-focal-py3.11-clang10 / test (default, 1, 3, linux.2xlarge) (gh) (related job)
inductor/test_extension_backend.py::ExtensionBackendTests::test_open_device_registration
pull / linux-focal-py3.11-clang10 / test (default, 2, 3, linux.2xlarge) (gh) (related job)
inductor/test_torchinductor.py::CpuTests::test_AllenaiLongformerBase_repro_cpu
pull / linux-focal-py3.11-clang10 / test (default, 3, 3, linux.2xlarge) (gh) (related job)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_multilayer_var_dynamic_shapes_cpu
pull / linux-focal-py3.12-clang10 / test (default, 1, 3, linux.2xlarge) (gh) (related job)
inductor/test_extension_backend.py::ExtensionBackendTests::test_open_device_registration
pull / linux-focal-py3.12-clang10 / test (default, 2, 3, linux.2xlarge) (gh) (related job)
inductor/test_torchinductor.py::CpuTests::test_AllenaiLongformerBase_repro_cpu
pull / linux-focal-py3.12-clang10 / test (default, 3, 3, linux.2xlarge) (gh) (related job)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_multilayer_var_dynamic_shapes_cpu
pull / linux-focal-py3.12-clang10-experimental-split-build / test (default, 1, 3, linux.2xlarge, unstable) (gh) (#129248)
inductor/test_extension_backend.py::ExtensionBackendTests::test_open_device_registration
pull / linux-focal-py3.12-clang10-experimental-split-build / test (default, 2, 3, linux.2xlarge, unstable) (gh) (#129248)
inductor/test_torchinductor.py::CpuTests::test_AllenaiLongformerBase_repro_cpu
pull / linux-focal-py3.12-clang10-experimental-split-build / test (default, 3, 3, linux.2xlarge, unstable) (gh) (#129248)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_multilayer_var_dynamic_shapes_cpu
pull / linux-focal-py3.8-clang10 / test (default, 1, 3, linux.2xlarge) (gh) (related job)
inductor/test_extension_backend.py::ExtensionBackendTests::test_open_device_registration
pull / linux-focal-py3.8-clang10 / test (default, 2, 3, linux.2xlarge) (gh) (related job)
inductor/test_torchinductor.py::CpuTests::test_AllenaiLongformerBase_repro_cpu
pull / linux-focal-py3.8-clang10 / test (default, 3, 3, linux.2xlarge) (gh) (related job)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_multilayer_var_dynamic_shapes_cpu
pull / linux-jammy-py3.10-clang15-asan / test (default, 2, 6, linux.4xlarge) (gh) (related job)
inductor/test_torchinductor_codegen_dynamic_shapes.py::DynamicShapesCodegenCpuTests::test_new_cpp_build_logical_dynamic_shapes_cpu
pull / linux-jammy-py3.10-clang15-asan / test (default, 3, 6, linux.4xlarge) (gh) (related job)
inductor/test_torchinductor.py::CpuTests::test_new_cpp_build_logical_cpu
pull / linux-jammy-py3.8-gcc11 / test (default, 1, 3, linux.2xlarge) (gh) (related job)
inductor/test_extension_backend.py::ExtensionBackendTests::test_open_device_registration
pull / linux-jammy-py3.8-gcc11 / test (default, 2, 3, linux.2xlarge) (gh) (related job)
inductor/test_torchinductor.py::CpuTests::test_AllenaiLongformerBase_repro_cpu
pull / linux-jammy-py3.8-gcc11 / test (default, 3, 3, linux.2xlarge) (gh) (related job)
inductor/test_torchinductor_dynamic_shapes.py::DynamicShapesCpuTests::test_multilayer_var_dynamic_shapes_cpu
trunk / linux-focal-cuda12.4-py3.10-gcc9-experimental-split-build-test / test (default, 3, 5, linux.4xlarge.nvidia.gpu) (gh) (related job)
inductor/test_torchinductor.py::CpuTests::test_AllenaiLongformerBase_repro_cpu

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

xuhancn · 2024-06-20T21:43:48Z

@pytorchbot rebase -b main

pytorchmergebot · 2024-06-20T21:45:08Z

@pytorchbot started a rebase job onto refs/remotes/origin/main. Check the current status here

pytorchmergebot · 2024-06-20T21:45:11Z

Successfully rebased xu_switch_to_new_cpp_builder_2 onto refs/remotes/origin/main, please pull locally before adding more changes (for example, via git checkout xu_switch_to_new_cpp_builder_2 && git pull --rebase)

xuhancn · 2024-06-20T22:16:48Z

@pytorchmergebot rebase

pytorchmergebot · 2024-06-20T22:18:08Z

@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here

pytorchmergebot · 2024-06-20T22:18:11Z

Tried to rebase and push PR #129144, but it was already up to date. Try rebasing against main by issuing:
@pytorchbot rebase -b main

xuhancn · 2024-06-21T07:36:43Z

Hi @desertfire could you please help on test it on fb_code environment? Thanks.

facebook-github-bot · 2024-06-21T14:01:13Z

@desertfire has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

xuhancn · 2024-06-22T18:47:37Z

@pytorchmergebot rebase -b main

pytorchmergebot · 2024-06-22T18:48:54Z

@pytorchbot started a rebase job onto refs/remotes/origin/main. Check the current status here

pytorchmergebot · 2024-06-22T18:48:55Z

Rebase failed due to Command git -C /home/runner/work/pytorch/pytorch rebase refs/remotes/origin/main pull/129144/head returned non-zero exit code 1

Rebasing (1/2)
Auto-merging torch/_inductor/codecache.py
CONFLICT (content): Merge conflict in torch/_inductor/codecache.py
error: could not apply d26c975560... switch to new cpp_builder.
hint: Resolve all conflicts manually, mark them as resolved with
hint: "git add/rm <conflicted_files>", then run "git rebase --continue".
hint: You can instead skip this commit: run "git rebase --skip".
hint: To abort and get back to the state before "git rebase", run "git rebase --abort".
hint: Disable this message with "git config advice.mergeConflict false"
Could not apply d26c975560... switch to new cpp_builder.

Raised by https://github.com/pytorch/pytorch/actions/runs/9627595117

revert unrelated changes.

…uhancn/pytorch into xu_switch_to_new_cpp_builder_2

xuhancn · 2024-06-22T19:02:52Z

@pytorchmergebot rebase -b main

pytorchmergebot · 2024-06-22T19:04:08Z

@pytorchbot started a rebase job onto refs/remotes/origin/main. Check the current status here

pytorchmergebot · 2024-06-22T19:04:10Z

Rebase failed due to Command git -C /home/runner/work/pytorch/pytorch rebase refs/remotes/origin/main pull/129144/head returned non-zero exit code 1

Rebasing (1/3)
Rebasing (2/3)
Auto-merging torch/_inductor/codecache.py
CONFLICT (content): Merge conflict in torch/_inductor/codecache.py
error: could not apply d26c975560... switch to new cpp_builder.
hint: Resolve all conflicts manually, mark them as resolved with
hint: "git add/rm <conflicted_files>", then run "git rebase --continue".
hint: You can instead skip this commit: run "git rebase --skip".
hint: To abort and get back to the state before "git rebase", run "git rebase --abort".
hint: Disable this message with "git config advice.mergeConflict false"
Could not apply d26c975560... switch to new cpp_builder.

Raised by https://github.com/pytorch/pytorch/actions/runs/9627657810

This reverts commit f659a14.

xuhancn · 2024-06-24T17:51:26Z

hi @desertfire , I tried to fix issue by this commit: 1dd0d10
Detailed analysis synced with you via slack, thanls.

facebook-github-bot · 2024-06-24T18:52:08Z

@desertfire has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

Original PRs is damaged by confilct and rebase: #128303, #129144 This PR just switch `HalideCodeCache` to new cpp_builder and it is not `fb_code` related. It can merge without `fb_code` test. Let's land this change firstly. Pull Request resolved: #129441 Approved by: https://github.com/jgong5, https://github.com/desertfire

xuhancn requested a review from desertfire June 20, 2024 15:36

pytorch-bot bot added ciflow/inductor module: inductor labels Jun 20, 2024

This was referenced Jun 20, 2024

[RFC] Add new CPP builder for inductor on pytorch Windows #124245

Open

[Deprecated][inductor] switch CppCodeCache to new cpp_builder. #128303

Closed

xuhancn requested a review from jgong5 June 20, 2024 15:38

xuhancn added module: windows Windows support for PyTorch ciflow/trunk Trigger trunk jobs on your pull request topic: not user facing topic category intel This tag is for PR from Intel labels Jun 20, 2024

pytorchbot added the open source label Jun 20, 2024

switch to new cpp_builder.

d26c975

pytorchmergebot force-pushed the xu_switch_to_new_cpp_builder_2 branch from 992de3d to d26c975 Compare June 20, 2024 21:45

revert unrelated changes.

b44fd39

drisspg added the oncall: pt2 label Jun 22, 2024

Merge branch 'main' into xu_switch_to_new_cpp_builder_2

7b1a3b2

xuhancn added 2 commits June 22, 2024 18:59

switch to new cpp_builder.

727cd31

revert unrelated changes.

Merge branch 'xu_switch_to_new_cpp_builder_2' of https://github.com/x…

961500f

…uhancn/pytorch into xu_switch_to_new_cpp_builder_2

xuhancn added 3 commits June 23, 2024 06:35

add debug message for fb_code.

0b84c44

fix new cpp_builder output path in fb_code.

f659a14

Revert "fix new cpp_builder output path in fb_code."

2dc10f7

This reverts commit f659a14.

janeyx99 added the triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module label Jun 24, 2024

xuhancn added 5 commits June 24, 2024 16:24

sync and fix test_new_cpp_build_logical

d8fffc5

fix mac os issue.

1b0536f

Merge branch 'main' into xu_switch_to_new_cpp_builder_2

6744f85

dispatch CppCodeCache fb_code to compile_file

1dd0d10

fix mac os issue.

c1d640c

fix output_path wrong.

799f706

xuhancn added 3 commits June 25, 2024 03:24

Merge branch 'pytorch:main' into xu_switch_to_new_cpp_builder_2

057582f

sync to latest main code and fix test_new_cpp_build_logical error.

01e1ca3

optimize code.

d282bc0

This was referenced Jun 25, 2024

[inductor] switch HalideCodeCache to new cpp_builder. #129441

Closed

[Out of Date] switch CppCodeCache to new cpp_builder. #129466

Closed

xuhancn changed the title ~~[inductor] switch CppCodeCache to new cpp_builder. (take 2)~~ [Deprecated][inductor] switch CppCodeCache to new cpp_builder. (take 2) Jun 30, 2024

xuhancn closed this Jul 4, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Deprecated][inductor] switch CppCodeCache to new cpp_builder. (take 2) #129144

[Deprecated][inductor] switch CppCodeCache to new cpp_builder. (take 2) #129144

xuhancn commented Jun 20, 2024 •

edited by pytorch-bot bot

Loading

pytorch-bot bot commented Jun 20, 2024 •

edited

Loading

xuhancn commented Jun 20, 2024

pytorchmergebot commented Jun 20, 2024

pytorchmergebot commented Jun 20, 2024

xuhancn commented Jun 20, 2024

pytorchmergebot commented Jun 20, 2024

pytorchmergebot commented Jun 20, 2024

xuhancn commented Jun 21, 2024

facebook-github-bot commented Jun 21, 2024

xuhancn commented Jun 22, 2024

pytorchmergebot commented Jun 22, 2024

pytorchmergebot commented Jun 22, 2024

xuhancn commented Jun 22, 2024

pytorchmergebot commented Jun 22, 2024

pytorchmergebot commented Jun 22, 2024

xuhancn commented Jun 24, 2024

facebook-github-bot commented Jun 24, 2024

[Deprecated][inductor] switch CppCodeCache to new cpp_builder. (take 2) #129144

[Deprecated][inductor] switch CppCodeCache to new cpp_builder. (take 2) #129144

Conversation

xuhancn commented Jun 20, 2024 • edited by pytorch-bot bot Loading

pytorch-bot bot commented Jun 20, 2024 • edited Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/129144

❌ 17 New Failures, 32 Unrelated Failures

xuhancn commented Jun 20, 2024

pytorchmergebot commented Jun 20, 2024

pytorchmergebot commented Jun 20, 2024

xuhancn commented Jun 20, 2024

pytorchmergebot commented Jun 20, 2024

pytorchmergebot commented Jun 20, 2024

xuhancn commented Jun 21, 2024

facebook-github-bot commented Jun 21, 2024

xuhancn commented Jun 22, 2024

pytorchmergebot commented Jun 22, 2024

pytorchmergebot commented Jun 22, 2024

xuhancn commented Jun 22, 2024

pytorchmergebot commented Jun 22, 2024

pytorchmergebot commented Jun 22, 2024

xuhancn commented Jun 24, 2024

facebook-github-bot commented Jun 24, 2024

xuhancn commented Jun 20, 2024 •

edited by pytorch-bot bot

Loading

pytorch-bot bot commented Jun 20, 2024 •

edited

Loading