Skip to content

Conversation

@xuhdev
Copy link
Collaborator

@xuhdev xuhdev commented Jun 27, 2019

Stack from ghstack:

Related to #20024, #22246, #22262

Differential Revision: D16183578

@pytorchbot pytorchbot added module: cuda Related to torch.cuda, and CUDA support in general module: operators labels Jun 27, 2019
xuhdev added 2 commits June 27, 2019 15:33
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
@xuhdev xuhdev requested a review from izdeby June 27, 2019 22:36
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
@xuhdev xuhdev requested a review from colesbury June 27, 2019 22:44
xuhdev added 2 commits June 27, 2019 16:05
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
@li-roy li-roy added the triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module label Jun 28, 2019
@xuhdev xuhdev requested a review from ngimel June 30, 2019 08:21
xuhdev added 2 commits June 30, 2019 09:23
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
Copy link
Member

@colesbury colesbury left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This doesn't look like it will work on non-contiguous Tensors. Use TensorIterator here, like you do for the CPU kernel. I think the only difference from the CPU version will be that you should call gpu_kernel instead of cpu_kernel

…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
@pytorchbot pytorchbot added the module: cpu CPU specific problem (e.g., perf, algorithm) label Jul 1, 2019
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
@xuhdev xuhdev requested a review from colesbury July 1, 2019 22:24
xuhdev added 2 commits July 1, 2019 15:27
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
xuhdev added 4 commits July 2, 2019 11:56
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
@xuhdev xuhdev requested a review from colesbury July 2, 2019 19:08
xuhdev added 4 commits July 2, 2019 20:23
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
- func: bitwise_not(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
dispatch:
CPU: _bitwise_not_out_cpu
CPU: bitwise_not_out
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No need to list them if the dispatch to the same function.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

There are other places like this, e.g.

CPU: dense_to_sparse
CUDA: dense_to_sparse
where CPU and CUDA dispatches go to the same functions - I assumed this is so that sparse tensors error out with a nice message if they call this function? If that's the case, makes sense to leave it here too.

…CUDA)."

Add a bitwise NOT operator for integer and Boolean types (CUDA).

Related to #20024, #22246, #22262

gh-metadata: pytorch pytorch 22320 gh/xuhdev/7/head
@zou3519 zou3519 deleted the gh/xuhdev/7/head branch July 10, 2019 19:20
zdevito pushed a commit to zdevito/ATen that referenced this pull request Jul 10, 2019
Summary: Pull Request resolved: pytorch/pytorch#22320

Test Plan: Imported from OSS

Differential Revision: D16183578

Pulled By: colesbury

fbshipit-source-id: 2f72cce5e10fd637be1ac87e1bbfe0937a661034
@facebook-github-bot
Copy link
Contributor

@colesbury merged this pull request in 574e808.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Merged module: cpu CPU specific problem (e.g., perf, algorithm) module: cuda Related to torch.cuda, and CUDA support in general open source triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module

Projects

None yet

Development

Successfully merging this pull request may close these issues.

10 participants