Merge branch 'sparse/csr/linalg_solve' of github.com:krshrimali/pytor…

…ch into sparse/csr/linalg_solve

[CUDA Pinned Memory] [Retry] Alternative implementation of pinned mem…

…ory allocator focusing on multi-threaded scalability (pytorch#69299)

Summary:
Pull Request resolved: pytorch#69299

pytorch#68906 + pytorch#68749 plugged one correctness hole (non-blocking copies of offset pinned memory tensors) while introducing another (non-blocking copies of pinned memory tensors with a non-standard DataPtr context).

In this revision, we use both the tensor data pointer and context to attempt to identify the originating block in the pinned memory allocator.

Test Plan: New unit tests added to cover the missing case previously.

Reviewed By: yinghai

Differential Revision: D32787087

fbshipit-source-id: 7fbc8c4fb74fcc6b47618587d6a599ef51b9fac0

Merge branch 'sparse/csr/linalg_solve' of github.com:krshrimali/pytor…

…ch into sparse/csr/linalg_solve

Merge branch 'sparse/csr/linalg_solve' of github.com:krshrimali/pytor…

…ch into sparse/csr/linalg_solve

Update on "Add OpInfo test to check that floating point inputs in OpI…

…nfos have requires_grad set to True"

This test detected a number of sampling methods that were not generating
the samples as expected, e.g. `index_put`, `cosine_embedding`, `stft`, but
perhaps most notably the generator for `BinOps`.

It also detected that `reminder` and `fmod` did not have implemented the
backward formula for the second input. I added this in the previous PR.

[ghstack-poisoned]

Update on "Add OpInfo test to check that floating point inputs in OpI…

…nfos have requires_grad set to True"

This test detected a number of sampling methods that were not generating
the samples as expected, e.g. `index_put`, `cosine_embedding`, `stft`, but
perhaps most notably the generator for `BinOps`.

It also detected that `reminder` and `fmod` did not have implemented the
backward formula for the second input. I added this in the previous PR.

[ghstack-poisoned]

Update on "no longer coalesce sparse COO tensors before comparison"

cc nikitaved pearu cpuhrsch @IvanYashchuk

[ghstack-poisoned]

Update on "use `torch.testing.assert_equal` in `TestCase.assertEqual`"


Supersedes pytorch#58981.

cc mruberry

Differential Revision: [D33542994](https://our.internmc.facebook.com/intern/diff/D33542994)

[ghstack-poisoned]

Update on ".github: Enable windows binary builds"

Signed-off-by: Eli Uriegas <eliuriegasfb.com>

[ghstack-poisoned]

Update on "ci: Enable tests for iOS on GHA"

These were left out of the intial migration for some reason so this just
transfers over those tests

Signed-off-by: Eli Uriegas <eliuriegasfb.com>

[ghstack-poisoned]