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

[NVIDIA] TensorIterator Body as CUDA Graph #755

Closed
wants to merge 14 commits into from
Closed

[NVIDIA] TensorIterator Body as CUDA Graph #755

wants to merge 14 commits into from

Conversation

apavliuk55
Copy link
Contributor

@apavliuk55 apavliuk55 commented Nov 6, 2023 •
edited
Loading

Details:

  • TensorIterator optimization to use CUDA graph as a body of the iterations loop.

Ticket:

  • 122556
  • 122561
  • 122558

Depends on #759

@github-actions github-actions bot added the category: NVIDIA plugin OpenVINO NVIDIA plugin label Nov 6, 2023
@apavliuk55 apavliuk55 force-pushed the feature/ti-body-as-cuda-graph branch 2 times, most recently from 7451673 to 2ee3dff Compare November 15, 2023 18:36
@apavliuk55 apavliuk55 changed the title [WIP][NVIDIA] TensorIterator Body as CUDA Graph [NVIDIA] TensorIterator Body as CUDA Graph Nov 15, 2023
@apavliuk55 apavliuk55 marked this pull request as ready for review November 15, 2023 19:30
@apavliuk55 apavliuk55 requested a review from a team as a code owner November 15, 2023 19:30
@apavliuk55 apavliuk55 force-pushed the feature/ti-body-as-cuda-graph branch from 2ee3dff to 95acbac Compare November 20, 2023 13:36
@apavliuk55 apavliuk55 force-pushed the feature/ti-body-as-cuda-graph branch 2 times, most recently from 7ee678f to 69104bd Compare December 8, 2023 12:55
Andrii Pavliuk added 14 commits December 28, 2023 00:51
[NVIDIA] Add operator==/!= to DevicePointer
a51a7de
[NVIDIA] Add CUDA::NodeParams, CUDA::TransferNode, CUDA::KernelNode
8be74b7
[NVIDIA] Add kernel args getters for Insert/Slice
bdcee9c
[NVIDIA] Add KernelNodeTest and TransferNodeTest
5145804
[NVIDIA] Fix review issues
6d5aed4
[NVIDIA] Add launchers to TI, refactor Execute()
3e553b1
[NVIDIA] Add TiCudaGraphInfo
b5e7404
[NVIDIA] Update TI to support CUDA graph as a body of iterations loop
4ce98f4
[NVIDIA] Add operator== for dim3, KernelNode and NodeParams
d7f3957
[NVIDIA] Update Run() of *TopologyRunners to take non-const context r…
f453633 
…eference
[NVIDIA] Remove TiCudaGraphInfo, add set_current_graph(), add_new_gra…
a267341 
…ph_info(), get_current_graph_info(), select_current_graph()
[NVIDIA] Change IsCudaGraphCompatible() interface to GetCudaGraphComp…
80dff6f 
…atibility() using enum
[NVIDIA] Add ExecuteGraph() to IOperationExec/OperationBase
dad499c
[NVIDIA] Remove paramsGraph_/resultsGraph_ from CudaGraphInfo
802ecf5
@apavliuk55 apavliuk55 force-pushed the feature/ti-body-as-cuda-graph branch from 69104bd to 802ecf5 Compare December 28, 2023 08:13
@apavliuk55 apavliuk55 mentioned this pull request Dec 28, 2023
Merged
@p-durandin p-durandin mentioned this pull request Jan 23, 2024
Merged
@p-durandin p-durandin closed this Jan 25, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
category: NVIDIA plugin OpenVINO NVIDIA plugin
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@apavliuk55 @p-durandin