Now when initializing a ScriptModule during the torch.jit.load()
process, there is already a cpp module backing the thing. That means
that setting training will overwrite whatever the initialized
ScriptModule had.

This PR splits apart the common "set up internal state" part of the
Module __init__ and calls that from ScriptModule.__init__ and
Module.__init__, leaving the "nn.Module-specific" part (setting
`self.training`) for the nn.Module __init__

ghstack-source-id: 9b2ba8a15c43cf230363e4cd10ba4ad3ac4931f7
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23680

* Slightly improve dataloader docs on when auto-batching is disabled

* fix typo

Summary:
Changelog:
- Use narrow instead of narrow_copy while returning
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23591

Test Plan:
- All tests should pass to ensure that the change is correct

Fixes https://github.com/pytorch/pytorch/issues/23580

Differential Revision: D16581174

Pulled By: ezyang

fbshipit-source-id: 1b6bf7d338ddd138ea4c6aa6901834dd202ec79c

Summary:
accidently calls clone, but what we want is creating an empty tensor and set storage.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/23452
ghstack-source-id: 87438096

Differential Revision: D16442756

fbshipit-source-id: 6d5663f82c9bd4e9de8fc846c52992477843af6a

This sets up the docs build in dry-run mode. If everything looks okay I
will enable it.

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/23556

Test Plan:
- Run ci

Imported from OSS

Differential Revision: D16563747

Pulled By: zou3519

fbshipit-source-id: 104371b3712c00b073a82e5145090e7bd6fd2d53

Summary:
Changelog:
- Rename `gels` to `lstsq`
- Fix all callsites
- Rename all tests
- Create a tentative alias for `lstsq` under the name `gels` and add a deprecation warning to not promote usage.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23460

Test Plan: - All tests should pass to confirm that the patch is correct

Differential Revision: D16547834

Pulled By: colesbury

fbshipit-source-id: b3bdb8f4c5d14c7716c3d9528e40324cc544e496

Summary:
After https://github.com/pytorch/pytorch/issues/23455, there is no need of this preprocessing in Python scripts.
They will be automatically processed in CMake (plus CPPFLAGS here
probably meant to be CXXFLAGS).

Reference:

- https://cmake.org/cmake/help/v3.15/envvar/CFLAGS.html
- https://cmake.org/cmake/help/v3.15/envvar/CXXFLAGS.html
- https://cmake.org/cmake/help/v3.15/envvar/LDFLAGS.html
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23528

Differential Revision: D16561561

Pulled By: ezyang

fbshipit-source-id: 962a27a2b0a18db0f95477ad067a2611e4128187

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/22917

Differential Revision: D16521429

Pulled By: pbelevich

fbshipit-source-id: 80ae583c6486d6948431b79e1452902bdf2cfbc3

Summary:
Only check for cmake dependencies we directly depend on (e.g., hipsparse but not rocsparse)

Use cmake targets for ROCm where possible.

While there, update the docker CI build infrastructure to only pull in packages by name we directly depend on (anticipating the demise of, e.g., miopengemm). I do not anticipate a docker rebuild to be necessary at this stage as the changes are somewhat cosmetic.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23527

Differential Revision: D16561010

Pulled By: ezyang

fbshipit-source-id: 87cd9d8a15a74caf9baca85a3e840e9d19ad5d9f

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/23229

Test Plan: Imported from OSS

Differential Revision: D16494413

Pulled By: zou3519

fbshipit-source-id: 4acb85e5a4ad09bf5f7cbb84cc8d4ceac0cd9967

Summary:
Syncing worker requirement mismatches to improve remote build time.

Created actions:
LARGE: 66
MEDIUM: 649
XLARGE: 1

Updated actions:
From LARGE to MEDIUM: 18
From LARGE to XLARGE: 2
From MEDIUM to LARGE: 20
From XLARGE to LARGE: 1

Differential Revision: D16559356

fbshipit-source-id: a51ef034265649314661ab0e283089a069a20437

Summary:
Fixes https://github.com/pytorch/pytorch/issues/21406
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23433

Differential Revision: D16524135

Pulled By: ailzhang

fbshipit-source-id: e7684fec60c9b9db9a09f8ac157b13c8dde1bdd2

Summary:
When a user tries to change metadata of a tensor created from `.data` or `.detach()`, we currently shows an error message "<function_name> is not allowed on Tensor created from .data or .detach()". However, this error message doesn't suggest what the right fix should look like. This PR improves the error message.

Closes https://github.com/pytorch/pytorch/issues/23393.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23504

Differential Revision: D16547415

Pulled By: yf225

fbshipit-source-id: 37f4a0385442e2b0966386fb14d3d938ecf4230c

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/23524

Differential Revision: D16549562

fbshipit-source-id: 58351fc2858d495b135023626116f6f565c8e9b1

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23270
ghstack-source-id: 87389530

Differential Revision: D16448942

fbshipit-source-id: e6b578f0e97776112259d7ea38e143e4716ec273

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/23542

Test Plan: Imported from OSS

Differential Revision: D16557122

Pulled By: suo

fbshipit-source-id: c86578aa2c55f44ed5d573d33874a82244df3d09

Differential Revision:
D16526027

Original commit changeset: 109f2968430d

fbshipit-source-id: c27252540ec6b7da60739eb7dcc8b1650672c226

Differential Revision:
D16554694

Original commit changeset: 0fae4458f18c

fbshipit-source-id: 08aa0c292fa5b2dbdd0d1f0e59f531416edef760

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/23535

Test Plan: Imported from OSS

Differential Revision: D16554694

Pulled By: suo

fbshipit-source-id: 0fae4458f18c06ffbd484905ad7836dce9ce69cc

Summary:
Previously these were left out which would lead to confusing messages,
now it looks something like:

```
torch.jit.frontend.UnsupportedNodeError: import statements aren't
supported
:
at ../test.py:13:9
    def bad_fn(self):
        import pdb
        ~~~~~~ <--- HERE
'__torch__.X' is being compiled since it was called from 'fn'
at ../test.py:16:12
def fn(x):
    return X(10)
           ~~~~ <--- HERE
```

Fixes #23453
](https://our.intern.facebook.com/intern/diff/16526027/)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23454

Pulled By: driazati

Differential Revision: D16526027

fbshipit-source-id: 109f2968430dbf51ee91b1b3409badfd557d19a4

Summary:
Use the recursive script API in the existing docs

TODO:
* Migration guide for 1.1 -> 1.2
Pull Request resolved: https://github.com/pytorch/pytorch/pull/21612

Pulled By: driazati

Differential Revision: D16553734

fbshipit-source-id: fb6be81a950224390bd5d19b9b3de2d97b3dc515

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23521

non-fbgemm path should have the same arguments as fbgemm path.

Reviewed By: jianyuh

Differential Revision: D16547637

fbshipit-source-id: bb00d725fb968cbee32defb8facd2799a7e79bb4

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/23154

Test Plan: Imported from OSS

Differential Revision: D16441913

Pulled By: suo

fbshipit-source-id: a79f2c3e06a33cbd79b2e3333f16c069f356f451

Summary:
This resolves two issues in one shot:

- sub shouldn't be available for bool type.
- When sub is applied to an unsupported type, the current error messages
  shows "add_cpu/add_cuda is not implemented for [type]". They should be
  "sub_cpu/sub_cuda" instead.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23519

Differential Revision: D16548770

Pulled By: izdeby

fbshipit-source-id: fe404a2a97b8d11bd180ec41364bf8e68414fb15

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23276

This diff introduces a new feature to simplify benchmarking the backward path of ops. Here is an example:

```
...
self.input_one = torch.rand(M, N, K, requires_grad=self.auto_set())
self.input_two = torch.rand(M, N, K, requires_grad=self.auto_set())
...
```

In this way, the benchmark will generate three different test cases.
1. input_one requires grad
2. input_two requires grad
3. both inputs require grad

Here is a sample output:
```
# Benchmarking PyTorch: add
# Mode: Eager
# Name: add_M1_N8_K8_bwdall
# Input: M: 1, N: 8, K: 8
Backward Execution Time (us) : 863.744

# Benchmarking PyTorch: add
# Mode: Eager
# Name: add_M1_N8_K8_bwd1
# Input: M: 1, N: 8, K: 8
Backward Execution Time (us) : 727.915

# Benchmarking PyTorch: add
# Mode: Eager
# Name: add_M1_N8_K8_bwd2
# Input: M: 1, N: 8, K: 8
Backward Execution Time (us) : 687.626
```

Reviewed By: zheng-xq

Differential Revision: D16450355

fbshipit-source-id: 50ae0916e81c3ff9f0c482ed6d386319eb15b305

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/23417

Test Plan:
cd docs; make html

Imported from OSS

Differential Revision: D16523781

Pulled By: ilia-cher

fbshipit-source-id: d6c09e8a85d39e6185bbdc4b312fea44fcdfff06

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/23249

Test Plan: Imported from OSS

Differential Revision: D16466587

Pulled By: wanchaol

fbshipit-source-id: a721da01b2da0ef90cac80b77f1285102e3b1118

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/23248

Test Plan: Imported from OSS

Differential Revision: D16466588

Pulled By: wanchaol

fbshipit-source-id: 3c3d5dec2cea2f9cb080eadaef457cc62ac3fbe0

Summary:
No real change on the CI since currently the default latest is 0.4.0. houseroad bddppq
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23517

Differential Revision: D16550375

Pulled By: bddppq

fbshipit-source-id: a669b8af678c79c4d6909300b28458fe6b7cd30c

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/23247

Test Plan: Imported from OSS

Differential Revision: D16466590

Pulled By: wanchaol

fbshipit-source-id: cf52721eacd177d9040564790382db13a9fcc2fe

Summary:
Install clang-tidy (from LLVM 8) for the `clang_tidy` job.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23518

Differential Revision: D16549621

Pulled By: ezyang

fbshipit-source-id: b1d20641380cdfdb0589249770b98163528fa69f

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/22926

Differential Revision: D16546369

Pulled By: colesbury

fbshipit-source-id: 56f7ef4476e586dee19366fdb720085d1c2f2027

Summary:
avoid Include the same header file twice
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23418

Differential Revision: D16546422

Pulled By: colesbury

fbshipit-source-id: 5cd868cce73d9199ced9b6f2f6f57bf42e5a5d5b

Summary:
There is an internal fbcode assert that fails if i do not add these checks.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23511

Differential Revision: D16545606

Pulled By: eellison

fbshipit-source-id: cd3a799850bae8f052f9d81c1e4a2678fda19317

Summary:
PyTorch test sets a policy() method to assertLeaksNoCudaTensors.
Whenever a test is run, assertLeaksNoCudaTensors is called,
which in turn calls CudaMemoryLeakCheck, which in turn calls
initialize_cuda_context_rng, where it executes torch.randn
on each device, where a kernel is launched on each device.

Since the kernel may not finish on device 1, the assertion
self.assertTrue(s1.query()) fails.

The fix is to insert

        torch.cuda.synchronize(d0)
        torch.cuda.synchronize(d1)

at the beginning of the test so that previously launched kernels finish before the real
test begins.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23520

Differential Revision: D16547701

Pulled By: soumith

fbshipit-source-id: 42ad369f909d534e15555493d08e9bb99dd64b6a