Commit Graph

509 Commits

Author SHA1 Message Date
Sergey Zimin
b99fc9d29f [MTIA] Support loading Tensors on mtia:0 for pytorch code (#149327)
Summary: The diff includes updates to the PyTorch code to enable loading tensors to MTIA.

Reviewed By: PatriceVignola

Differential Revision: D71176848

Pull Request resolved: https://github.com/pytorch/pytorch/pull/149327
Approved by: https://github.com/ezyang
2025-03-20 18:05:15 +00:00
cyy
b61a556427 Turn onnx functions into static (#147598)
To avoid exposing ONNX symbols.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/147598
Approved by: https://github.com/justinchuby
2025-02-21 07:40:28 +00:00
cyy
8f291e8c00 Fix clang-tidy warnings in torch/jit (#146963)
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/146963
Approved by: https://github.com/davidberard98
2025-02-15 03:36:59 +00:00
cyy
379bbef23c Enable more C++ warnings (#143355)
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/143355
Approved by: https://github.com/albanD
2024-12-27 05:46:57 +00:00
PyTorch MergeBot
9255ffc841 Revert "Enable more C++ warnings (#143355)"
This reverts commit daa3ffe0eb.

Reverted https://github.com/pytorch/pytorch/pull/143355 on behalf of https://github.com/malfet due to It fails internal build system as it kind of breaks separation between native and native/cpu ([comment](https://github.com/pytorch/pytorch/pull/143355#issuecomment-2562961546))
2024-12-26 17:13:10 +00:00
cyy
daa3ffe0eb Enable more C++ warnings (#143355)
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/143355
Approved by: https://github.com/albanD
2024-12-21 09:19:02 +00:00
Richard Barnes
82ce888273 c10::string_view -> std::string_view in more places (#142517)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/142517
Approved by: https://github.com/malfet
2024-12-12 19:45:59 +00:00
cyy
da1e120dfd [2/N] Replace c10::sv with std::sv (#139456)
Follows  #139453

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139456
Approved by: https://github.com/ezyang, https://github.com/malfet
2024-11-08 01:48:00 +00:00
cyy
bf1b8adee6 Turn static inline into static function (#139843)
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139843
Approved by: https://github.com/ezyang
2024-11-07 23:58:18 +00:00
PyTorch MergeBot
68f1b52d8a Revert "Turn static inline into static function (#139843)"
This reverts commit 72d3f5b26d.

Reverted https://github.com/pytorch/pytorch/pull/139843 on behalf of https://github.com/ZainRizvi due to Sorry but this is causing tests to fail on trunk. See [GH job link](https://github.com/pytorch/pytorch/actions/runs/11729669425/job/32675829894) [HUD commit link](72d3f5b26d) ([comment](https://github.com/pytorch/pytorch/pull/139843#issuecomment-2463354131))
2024-11-07 22:29:45 +00:00
cyy
72d3f5b26d Turn static inline into static function (#139843)
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139843
Approved by: https://github.com/ezyang
2024-11-07 19:08:41 +00:00
PyTorch MergeBot
7e02386303 Revert "[2/N] Replace c10::sv with std::sv (#139456)"
This reverts commit 028c5d3426.

Reverted https://github.com/pytorch/pytorch/pull/139456 on behalf of https://github.com/ZainRizvi due to Sorry but this breaks internally. @ezyang can you please help get this landed? See D65546398 for more details ([comment](https://github.com/pytorch/pytorch/pull/139456#issuecomment-2462768891))
2024-11-07 17:00:59 +00:00
Nikita Shulga
c0c6bf4ef2 Don't use deprecated type properties in UpsampleKernel (#139399)
By replacing `at::CPU(dtype)` pattern with `at::device(kCPU).dtype(dtype)` pattern

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139399
Approved by: https://github.com/Skylion007
ghstack dependencies: #139353
2024-11-06 13:34:45 +00:00
PyTorch MergeBot
53299b8a38 Revert "Don't use deprecated type properties in UpsampleKernel (#139399)"
This reverts commit 0058f71002.

Reverted https://github.com/pytorch/pytorch/pull/139399 on behalf of https://github.com/malfet due to And it was backed out again due to the internal usages of deprecated API ([comment](https://github.com/pytorch/pytorch/pull/139358#issuecomment-2459740090))
2024-11-06 13:23:43 +00:00
cyy
028c5d3426 [2/N] Replace c10::sv with std::sv (#139456)
Follows  #139453

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139456
Approved by: https://github.com/ezyang
2024-11-06 01:50:38 +00:00
Nikita Shulga
0058f71002 Don't use deprecated type properties in UpsampleKernel (#139399)
By replacing `at::CPU(dtype)` pattern with `at::device(kCPU).dtype(dtype)` pattern

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139399
Approved by: https://github.com/Skylion007
ghstack dependencies: #139353, #139358
2024-11-05 00:29:58 +00:00
PyTorch MergeBot
4a3ee96427 Revert "Don't use deprecated type properties in UpsampleKernel (#139399)"
This reverts commit 9d096e4d9f.

Reverted https://github.com/pytorch/pytorch/pull/139399 on behalf of https://github.com/ZainRizvi due to Change reverted internally due to broken builds. See D65378845 ([comment](https://github.com/pytorch/pytorch/pull/139358#issuecomment-2455959040))
2024-11-05 00:13:48 +00:00
cyy
419a7e197d [6/N] Fix Wextra-semi warning (#139605)
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139605
Approved by: https://github.com/ezyang
2024-11-04 13:43:16 +00:00
cyy
4a2da52137 [1/N] Replace c10::sv with std::sv (#139453)
Picks some safe replacements.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139453
Approved by: https://github.com/Skylion007
2024-11-01 05:39:37 +00:00
Nikita Shulga
9d096e4d9f Don't use deprecated type properties in UpsampleKernel (#139399)
By replacing `at::CPU(dtype)` pattern with `at::device(kCPU).dtype(dtype)` pattern

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139399
Approved by: https://github.com/Skylion007
ghstack dependencies: #139353, #139358
2024-10-31 17:32:19 +00:00
Richard Barnes
8f62832189 c10::nullopt -> std::nullopt (#138701)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/138701
Approved by: https://github.com/Skylion007, https://github.com/malfet
2024-10-24 15:03:32 +00:00
Richard Barnes
dbf0fa811a Remove C10_HOST_CONSTEXPR_EXCEPT_WIN_CUDA and CONSTEXPR_EXCEPT_WIN_CUDA (#138479)
BC linter suppressed due to removal of `tools/linter/adapters/constexpr_linter.py`
Pull Request resolved: https://github.com/pytorch/pytorch/pull/138479
Approved by: https://github.com/eqy, https://github.com/malfet
2024-10-24 07:51:05 +00:00
FFFrog
af0bc75460 Remove deprecated alias macro(1/3) (#137556)
**Detailed Descriptions:**
- Remove AT_ERROR Macro

Pull Request resolved: https://github.com/pytorch/pytorch/pull/137556
Approved by: https://github.com/ezyang
2024-10-21 17:32:32 +00:00
Richard Barnes
fddabc6e0b C10_UNUSED to [[maybe_unused]] (#6357) (#138364)
Summary: Pull Request resolved: https://github.com/pytorch/executorch/pull/6357

Pull Request resolved: https://github.com/pytorch/pytorch/pull/138364
Approved by: https://github.com/Skylion007, https://github.com/eqy
2024-10-19 13:17:43 +00:00
Richard Barnes
542f7c8383 Eliminate C10_NODISCARD (#138336)
Test Plan: Sandcastle

Reviewed By: swolchok

Pull Request resolved: https://github.com/pytorch/pytorch/pull/138336
Approved by: https://github.com/Skylion007
2024-10-19 02:54:06 +00:00
PyTorch MergeBot
375dcb960f Revert "Avoid some dangling reference warnings (#132535)"
This reverts commit f3d7a02716.

Reverted https://github.com/pytorch/pytorch/pull/132535 on behalf of https://github.com/clee2000 due to broke some internal builds D64479234 ([comment](https://github.com/pytorch/pytorch/pull/132535#issuecomment-2419983509))
2024-10-17 16:23:36 +00:00
Isuru Fernando
f3d7a02716 Avoid some dangling reference warnings (#132535)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/132535
Approved by: https://github.com/aaronenyeshi
2024-10-16 13:41:12 +00:00
Richard Barnes
b7f798caa4 Use C10_UNUSED instead of (void)X (#137239)
Summary:
Auto-generated with
```
buck run //scripts/rbarnes/regex_multiline_replacer:regex_multiline_replacer -- --find '^(\s*for\s*\()(const.*\n)\s*\(void\)[A-Za-z]+;\s*//\s*Suppress.*\s*\n(.*)'  --replace '\1C10_UNUSED \2\3' `find caffe2/ -regex ".*\.\(cpp\|h\)"`
```

Differential Revision: D33432600

Pull Request resolved: https://github.com/pytorch/pytorch/pull/137239
Approved by: https://github.com/Skylion007
2024-10-15 14:32:59 +00:00
cyy
7bbdf87517 [22/N] Fix clang-tidy warnings in jit (#134829)
Follows  #134537

Pull Request resolved: https://github.com/pytorch/pytorch/pull/134829
Approved by: https://github.com/ezyang
2024-09-19 19:24:42 +00:00
cyy
ec3f52dd27 [21/N] Fix clang-tidy warnings in jit (#134537)
Follows #133399

Pull Request resolved: https://github.com/pytorch/pytorch/pull/134537
Approved by: https://github.com/Skylion007
2024-08-28 03:22:01 +00:00
Zhengxu Chen
59b3f5911d [sigmoid] Support custom obj deserialization. (#133463)
Summary:
It seems we have multiple places deserializing torchbind objects. Moving the code around so that every load essentially share the same implementation.

Also added a test case "package_reader_testing" which load back the archive file in Python and eagerly validate the numerical result.

Test Plan: buck test mode/opt sigmoid/inference/test:e2e_test_cpu

Reviewed By: SherlockNoMad

Differential Revision: D61235770

Pull Request resolved: https://github.com/pytorch/pytorch/pull/133463
Approved by: https://github.com/ydwu4
2024-08-15 17:58:44 +00:00
Daniil Kutz
62cd065de2 Validate that node TK_ASSIGN have field initialized (#127878)
Fixes segmentation fault during model load via C++ API.

An `Assign` statement (`TK_ASSIGN` type) have 3 fields: `lhs`, `rhs` and `type`. Field `type` is of type `Maybe`, which means it could be not presented. During model load in `import_source.cpp` field `type` is dereferenced without validation.

It is similar error that have been fixed in #106041.

Fixes #127877

Pull Request resolved: https://github.com/pytorch/pytorch/pull/127878
Approved by: https://github.com/malfet
2024-08-14 15:27:58 +00:00
cyy
5b3b2b9cc7 [7/N] Fix clang-tidy warnings in jit (#131996)
Follows #131986

Pull Request resolved: https://github.com/pytorch/pytorch/pull/131996
Approved by: https://github.com/ezyang
2024-07-29 01:21:18 +00:00
Nikita Shulga
62e566b345 [BE] Remove suppression of inconsistent missing overrides (#131524)
This should prevent regressions like the ones fixed by https://github.com/pytorch/pytorch/pull/131204

- Remove global `-Wno-error=inconsistent-missing-override`
- Wrap offending includes (protobuf and asmjit) with `C10_DIAGNOSTIC_PUSH_AND_IGNORE` and `C10_DIAGNOSTIC_POP_AND_IGNORED`
- Add `override` keyword to `at::namespace::tunable::StreamTimer` and `LLVMCodeGenImpl`

Pull Request resolved: https://github.com/pytorch/pytorch/pull/131524
Approved by: https://github.com/atalman
2024-07-24 10:07:36 +00:00
cyy
f4dcf2ae93 [1/N] Change #include <c10/util/Optional.h> to #include <optional> (#128301)
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/128301
Approved by: https://github.com/ezyang, https://github.com/r-barnes
2024-07-08 07:03:53 +00:00
cyy
efb73eda51 [2/N] Fix some violations of unused-function and unused-variable checks in torch_cpu (#129878)
Follows #128670

Pull Request resolved: https://github.com/pytorch/pytorch/pull/129878
Approved by: https://github.com/ezyang
2024-07-04 00:39:28 +00:00
cyy
26de2c2487 [3/N] Enable clang-tidy on torch/csrc/jit/serialization/* (#129850)
Follows #129300.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/129850
Approved by: https://github.com/ezyang
2024-07-02 20:08:48 +00:00
cyy
eb1583dbc1 [2/N] Fix clang-tidy warnings in torch/csrc/jit/serialization (#129300)
Follows #129055
Pull Request resolved: https://github.com/pytorch/pytorch/pull/129300
Approved by: https://github.com/ezyang
2024-07-01 01:09:00 +00:00
cyy
2c7c286fa4 [1/N] Fix clang-tidy warnings in torch/csrc/jit/serialization (#129055)
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/129055
Approved by: https://github.com/r-barnes
2024-06-21 14:56:31 +00:00
PyTorch MergeBot
846bb30e13 Revert "[1/N] Change #include <c10/util/Optional.h> to #include <optional> (#128301)"
This reverts commit bd72e28314.

Reverted https://github.com/pytorch/pytorch/pull/128301 on behalf of https://github.com/huydhn due to Sorry for reverting your change but it fails XLA build bd72e28314. Please rebase your PR before relanding because I think the failure is hidden by an unrelated broken trunk XLA failure from your current base commit ([comment](https://github.com/pytorch/pytorch/pull/128301#issuecomment-2169035822))
2024-06-15 01:58:20 +00:00
cyy
bd72e28314 [1/N] Change #include <c10/util/Optional.h> to #include <optional> (#128301)
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/128301
Approved by: https://github.com/ezyang
2024-06-14 23:21:01 +00:00
cyy
67d52d7fcb [caffe2] Remove import_legacy.cpp (#126149)
I think they are for Caffe2 and should be deleted.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/126149
Approved by: https://github.com/r-barnes
2024-05-24 19:47:32 +00:00
Gustav Larsson
44e47d5bd0 [onnx.export] Avoid linear loop over symbol_dim_map (#123029)
This PR is part of an effort to speed up torch.onnx.export (#121422).

- Doing a reverse look-up in `symbol_dim_map` incurs a linear cost in number of symbols. This happens for each node, so incurs a quadratic cost to the whole export.
- Add a reverse look-up `dim_symbol_map` that is kept in parallel of `symbol_dim_map`. This avoids a linear time look-up, which creates a quadratic export time complexity.
- This is a highly pragmatic solution. If someone more familiar with the code base has a better solution, I'm interested to hear about it.
- Resolves (9) in #121422.

(partial fix of #121422)

Pull Request resolved: https://github.com/pytorch/pytorch/pull/123029
Approved by: https://github.com/justinchuby
2024-05-15 17:22:30 +00:00
Richard Barnes
ed327876f5 [codemod] c10:optional -> std::optional (#126135)
Generated by running the following from PyTorch root:
```
find . -regex ".*\.\(cpp\|h\|cu\|hpp\|cc\|cxx\)$" | grep -v "build/" | xargs -n 50 -P 4 perl -pi -e 's/c10::optional/std::optional/'
```

`c10::optional` is just an alias for `std::optional`. This removes usages of that alias in preparation for eliminating it entirely.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/126135
Approved by: https://github.com/Skylion007, https://github.com/malfet, https://github.com/albanD, https://github.com/aaronenyeshi
2024-05-14 19:35:51 +00:00
Richard Barnes
b9e7b35912 Remove caffe2 from more build files (#125898)
Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/125898
Approved by: https://github.com/Skylion007
2024-05-13 18:37:59 +00:00
Zhengxu Chen
1115a25c36 Add obc counter for TS migration. (#125986)
Summary: Since table caffe2_pytorch_usage_stats only has 1 day retention which renders it useless for TS migration purposes, we want to build a lightweight counter mechanism to collect usage data about torch jit APIs which can monitor the usage decline in the long term.

Test Plan: CI

Reviewed By: SherlockNoMad

Differential Revision: D57216847

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125986
Approved by: https://github.com/gmagogsfm
2024-05-11 05:14:02 +00:00
PyTorch MergeBot
ea347fa6ce Revert "Fix & optimze open device registration test. (#124712)"
This reverts commit f03cf9d4dc.

Reverted https://github.com/pytorch/pytorch/pull/124712 on behalf of https://github.com/kit1980 due to breaking internal builds ([comment](https://github.com/pytorch/pytorch/pull/124712#issuecomment-2086971499))
2024-04-30 20:00:37 +00:00
FFFrog
f03cf9d4dc Fix & optimze open device registration test. (#124712)
Fixes #100152

1. Fix the wrong tests about lazy init for PrivateUse1 named foo
2. Fix wrong backend meta registry mechanism when compiling with clang++( compiling with g++ work well)(introduced by static variable in inline function)
3. Refactor the tests and make it more flexible
4. Disable the two tests temporarily
    - test_open_device_storage_pin_memory
    - test_compile_autograd_function_aliasing
Pull Request resolved: https://github.com/pytorch/pytorch/pull/124712
Approved by: https://github.com/albanD, https://github.com/malfet
2024-04-29 18:55:38 +00:00
liqunfu
bbe846f430 Add symbolic_opset19.py and symbolic_opset20.py to support opset 19/20, extend opset 18 support (#118828)
Start to fix https://github.com/pytorch/pytorch/issues/114801

Co-authored-by: Thiago Crepaldi <thiagofc@microsoft.com>
Co-authored-by: Justin Chu <justinchuby@users.noreply.github.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118828
Approved by: https://github.com/thiagocrepaldi
2024-03-22 18:01:33 +00:00
cyy
5f9b432494 [2/N] Replace std::tie with structural binding (#119879)
This PR follows #119774, Python generated code was changed to use structural binding.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/119879
Approved by: https://github.com/albanD
2024-02-15 02:56:34 +00:00