Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Binary Shift operators #10034

Closed
Tracked by #9702 ...
VirdhatchaniKN opened this issue Jul 9, 2024 · 6 comments
Closed
Tracked by #9702 ...

Binary Shift operators #10034

VirdhatchaniKN opened this issue Jul 9, 2024 · 6 comments
Assignees
Labels
op_cat: eltwise Op Generalization Generalization and relaxations of requirements in Ops P1

Comments

@VirdhatchaniKN
Copy link
Contributor

VirdhatchaniKN commented Jul 9, 2024

Master issue : #9702

Op Requirement :

  • ElemwiseBinary/ELEMWISE_BINARY_LEFT_SHIFT
  • ElemwiseBinary/ELEMWISE_BINARY_RIGHT_SHIFT

Issues :

Initial Implementation done in #8636
This implementation requires the involvement of a lot of ops, which may lead to performance issues.

@KalaivaniMCW
Copy link
Contributor

Hi @rtawfik01 @rdjogoTT

Is it possible to support shift operation for two tensors (like it is available in bitwise operation)?
At present, the second operand for shift operation takes only a single integer parameter
Image

Bitwise ops:
Image

rdjogoTT added a commit that referenced this issue Dec 12, 2024
rdjogoTT added a commit that referenced this issue Dec 12, 2024
rdjogoTT added a commit that referenced this issue Dec 12, 2024
rdjogoTT added a commit that referenced this issue Dec 12, 2024
rdjogoTT added a commit that referenced this issue Dec 12, 2024
rdjogoTT added a commit that referenced this issue Dec 12, 2024
### Ticket
[Link to Github
Issue](#10034)

### What's changed
Implemented binary left/right shift as binary SFPU OPs.

### Checklist
- [x] Post commit CI passes:
https://github.com/tenstorrent/tt-metal/actions/runs/12287866881
- [x] Blackhole Post commit:
https://github.com/tenstorrent/tt-metal/actions/runs/12287867702
- [x] New/Existing tests provide coverage for changes - will be added in
future PR
@rdjogoTT
Copy link
Contributor

LLK added here: #15926

@rdjogoTT rdjogoTT removed their assignment Dec 13, 2024
@rdjogoTT rdjogoTT removed the LLK label Dec 13, 2024
@cmaryanTT
Copy link

@eyonland - please assign someone for implementation of this. The LLK should be present, per Radomir's comment. This should be high priority because it could be used to address some customer requests.

@eyonland
Copy link
Contributor

Pending merge of
#16055

KalaivaniMCW added a commit that referenced this issue Dec 17, 2024
KalaivaniMCW added a commit that referenced this issue Dec 18, 2024
KalaivaniMCW added a commit that referenced this issue Dec 18, 2024
KalaivaniMCW added a commit that referenced this issue Dec 18, 2024
KalaivaniMCW added a commit that referenced this issue Dec 18, 2024
KalaivaniMCW added a commit that referenced this issue Dec 18, 2024
### Ticket
Link to Github Issue #10034 

### Problem description
Binary left/right shift operators

### What's changed
add ttnn support for sfpu binary bitwise shift operators

### Checklist
- [x] Post commit CI passes
https://github.com/tenstorrent/tt-metal/actions/runs/12372840148
https://github.com/tenstorrent/tt-metal/actions/runs/12394449160
- [ ] Blackhole Post commit (if applicable)
- [ ] Model regression CI testing passes (if applicable)
- [ ] Device performance regression CI testing passes (if applicable)
- [ ] **(For models and ops writers)** Full [new
models](https://github.com/tenstorrent/tt-metal/actions/workflows/full-new-models-suite.yaml)
tests passes
- [x] New/Existing tests provide coverage for changes
@eyonland
Copy link
Contributor

@VirdhatchaniKN , would you confirm that this can be closed now?

@VirdhatchaniKN
Copy link
Contributor Author

Closing ticket as PR #16055 has been merged to main

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
op_cat: eltwise Op Generalization Generalization and relaxations of requirements in Ops P1
Projects
None yet
Development

No branches or pull requests

9 participants