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.

Sign up for GitHub

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

Jump to bottom

Move CCL ops to ttnn #9486

Closed
13 of 17 tasks
Tracked by #9322
ayerofieiev-tt opened this issue Jun 17, 2024 · 7 comments
Closed
13 of 17 tasks
Tracked by #9322

Move CCL ops to ttnn #9486

ayerofieiev-tt opened this issue Jun 17, 2024 · 7 comments
Assignees
@SeanNijjar @Aswinmcw
Labels
op_cat: ccl ttnn-merge

Comments

@ayerofieiev-tt
Copy link
Member

ayerofieiev-tt commented Jun 17, 2024 •
edited by Aswinmcw
Loading

Consolidate Tensor Manipulation ops in ttnn

Plan

all_gather

line_all_gather

  • replace c++ usage (not used)
  • Add ttnn op
  • replace python usage
  • remove python binding
  • consolidate into proper structure

reduce_scatter

  • replace c++ usage (not used)
  • Add ttnn op
  • replace python usage
  • remove python binding
  • consolidate into proper structure

CCL common (test and source) after above are done

  • remaining standalone and ccl common files under tt_eager/tt_dnn/op_library/ccl
    • ccl_common, README, ccl_host_datastructures, and contents of tt_eager/tt_dnn/op_library/ccl/shared_with_host/
  • Unit tests under tests/tt_eager/ops/ccl

Work done so far:

  • All 3 CCL ops are merged to main

Testing

For the best coverage, I recommend to run these workflows. If some of them fails, check if it is the same fail as on main:

  • Post commit CI
  • [post-commit] models tests
  • Device perf regressions and output report
  • Model perf regressions and output report
  • Nightly fast dispatch model regression CI
  • T3000 frequent tests
  • T3000 demo tests
@ayerofieiev-tt ayerofieiev-tt mentioned this issue Jun 17, 2024
Closed
31 tasks
@ayerofieiev-tt ayerofieiev-tt changed the title CCL - Sean Nijjar Move CCL to ttnn Jun 17, 2024
@ayerofieiev-tt ayerofieiev-tt changed the title Move CCL to ttnn Move CCL ops to ttnn Jun 17, 2024
ayerofieiev-tt added a commit that referenced this issue Jun 18, 2024
@ayerofieiev-tt
#9486: change ttl all_gather call with ttnn
4d5cbce
@ayerofieiev-tt ayerofieiev-tt mentioned this issue Jun 18, 2024
Merged
4 tasks
ayerofieiev-tt added a commit that referenced this issue Jun 18, 2024
@ayerofieiev-tt
#9486: Replace ttl.tensor.all_gather with ttnn.all_gather (#9517)
8c96c4f 
* #9486: change ttl all_gather call with ttnn

* #0: correct arguments
ayerofieiev-tt added a commit that referenced this issue Jun 18, 2024
@ayerofieiev-tt
#9486: remove tt_lib binding for all_gather
fd14d74
@ayerofieiev-tt ayerofieiev-tt mentioned this issue Jun 18, 2024
Merged
1 task
ayerofieiev-tt added a commit that referenced this issue Jun 18, 2024
@ayerofieiev-tt
#9486: Remove all_gather binding from tt_lib (#9525)
b88cf09 
#9486: remove tt_lib binding for all_gather
ayerofieiev-tt added a commit that referenced this issue Jun 19, 2024
@ayerofieiev-tt
#9486: ttnn.all_gather allows tensor or a list of tensors
eac3a74
@ayerofieiev-tt ayerofieiev-tt mentioned this issue Jun 19, 2024
Closed
2 tasks
ayerofieiev-tt added a commit that referenced this issue Jun 21, 2024
@ayerofieiev-tt
#9486: Revert removal of all_gather bindings from tt_lib
ef22a4a 
This reverts commit b88cf09.
ayerofieiev-tt added a commit that referenced this issue Jun 21, 2024
@ayerofieiev-tt
#9486: revert from ttnn to tt_lib all_gather in llama2_70 t3k
4b52249
ayerofieiev-tt added a commit that referenced this issue Jun 24, 2024
@ayerofieiev-tt
#9568: Revert t3k demo llama70 to use tt_lib all_gather (#9612)
358ebae 
* #9486: Revert removal of all_gather bindings from tt_lib

This reverts commit b88cf09.

* #9486: revert from ttnn to tt_lib all_gather in llama2_70 t3k
SeanNijjar added a commit that referenced this issue Jun 26, 2024
@SeanNijjar
#9486: use mesh tensor for allgather test and re-enable tests
21f5054
SeanNijjar added a commit that referenced this issue Jun 26, 2024
@SeanNijjar
#9486: use mesh tensor for allgather test and re-enable tests
415526e
SeanNijjar added a commit that referenced this issue Jun 26, 2024
@SeanNijjar
#9486: use mesh tensor for allgather test and re-enable tests
88e8d99
SeanNijjar added a commit that referenced this issue Jun 26, 2024
@SeanNijjar
#9486: use mesh tensor for allgather test and re-enable tests
18582e4
SeanNijjar added a commit that referenced this issue Jun 26, 2024
@SeanNijjar
#9486: use mesh tensor for allgather test and re-enable tests
943d660
@Aswinmcw Aswinmcw self-assigned this Jul 2, 2024
Aswinmcw added a commit that referenced this issue Jul 3, 2024
@Aswinmcw
#9486: Merge line_all_gather to TTNN
decb82e
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move CCL common to TTNN
1d3a3b2
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: re-organize namespace
4c444c8
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move kernel files into kernels directory
1781f85
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Merge CCL reduce_scatter to TTNN
e6e5353
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move pytests to TTNN
650dbde
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Replace ttdnn op with ttnn
bb615e5
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Merge line_all_gather to TTNN
149367e
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move CCL kernel files to TTNN
026ba42
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move CCL common to TTNN
77eca4e
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: re-organize namespace
a4ac4e7
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move kernel files into kernels directory
2a3b558
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Merge CCL reduce_scatter to TTNN
65d48cb
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move pytests to TTNN
3c559e0
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Replace ttdnn op with ttnn
21309a6
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Merge line_all_gather to TTNN
c40e47d
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move CCL kernel files to TTNN
20a9371
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move CCL common to TTNN
52ace56
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: re-organize namespace
4d9777d
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move kernel files into kernels directory
75121b8
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Merge CCL reduce_scatter to TTNN
6c3e9c2
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Move pytests to TTNN
b9eb5e0
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Replace ttdnn op with ttnn
de1d373
Aswinmcw added a commit that referenced this issue Jul 22, 2024
@Aswinmcw
#9486: Merge CCL reduce_scatter to TTNN (#9979)
01c7e81 
* #9486: Merge line_all_gather to TTNN

* #9486: Move CCL kernel files to TTNN

* #9486: Move CCL common to TTNN

* #9486: re-organize namespace

* #9486: Move kernel files into kernels directory

* #9486: Merge CCL reduce_scatter to TTNN

* #9486: Move pytests to TTNN

* #0: Fix issues

* #0: Rebased

#0: Rebased

* #9486: Replace ttdnn op with ttnn
@Aswinmcw
Copy link
Contributor

All 3 CCL ops are now merged to main
cc. @SeanNijjar @ayerofieiev-tt

@SeanNijjar
Copy link
Contributor

Last remaining items are the (gtest) unit tests.

@SeanNijjar
Copy link
Contributor

All 3 CCL ops are now merged to main cc. @SeanNijjar @ayerofieiev-tt

Thanks @Aswinmcw!

@ayerofieiev-tt
Copy link
Member Author

Last remaining items are the (gtest) unit tests.

Is it on @Aswinmcw ?

@Aswinmcw
Copy link
Contributor

If moving gtests are in priority, I guess only few modifications are needed, if needed, let me know, I will try and raise one PR in couple of days

@eyonland
Copy link
Contributor

This has been migrated.

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

@SeanNijjar SeanNijjar

@Aswinmcw Aswinmcw

Labels
op_cat: ccl ttnn-merge
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
4 participants
@SeanNijjar @eyonland @Aswinmcw @ayerofieiev-tt