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

[GPU] scatter element update fix and dynamic impl #29087

Merged
merged 3 commits into from
Feb 21, 2025
Merged

[GPU] scatter element update fix and dynamic impl #29087

merged 3 commits into from
Feb 21, 2025
+49 −9

Conversation

michal-miotk
Copy link
Contributor

Details:

  • added input idx calculation because output idx was different due to output padding
  • part of accuracy fix of model FasterRCNN_ResNet50_FPN_V2.onnx

Tickets:

  • small part of 101294

@michal-miotk michal-miotk requested review from a team as code owners February 20, 2025 10:38
@github-actions github-actions bot added the category: GPU OpenVINO GPU plugin label Feb 20, 2025
@vladimir-paramuzov vladimir-paramuzov added this to the 2025.1 milestone Feb 21, 2025
@vladimir-paramuzov vladimir-paramuzov added this pull request to the merge queue Feb 21, 2025
@sshlyapn sshlyapn removed this pull request from the merge queue due to a manual request Feb 21, 2025
sshlyapn
sshlyapn reviewed Feb 21, 2025
View reviewed changes
.../intel_gpu/src/kernel_selector/kernels/scatter_update/scatter_elements_update_kernel_ref.cpp
auto dispatchData = SetDefault(prim_params, i == 1);
kd.kernels[i].params.workGroups.global = dispatchData.gws;
kd.kernels[i].params.workGroups.local = dispatchData.lws;
kd.kernels[i].skip_execution = KernelData::SkipKernelExecution(prim_params);
Copy link
Contributor

@sshlyapn sshlyapn Feb 21, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

I think similar logic as bellow should also be applied here for proper paddings handling:

openvino/src/plugins/intel_gpu/src/kernel_selector/kernels/scatter_update/scatter_nd_update_kernel_ref.cpp

Lines 166 to 168 in 285a73a

// Do not skip copy stage if output buffer is not empty or requires modification
if (i == 0 && prim_params.outputs[0].LogicalSize() != 0 && prim_params.outputs[0] != prim_params.inputs[0])
kd.kernels[i].skip_execution = false;

@p-durandin p-durandin added this pull request to the merge queue Feb 21, 2025
Merged via the queue into openvinotoolkit:master with commit b6fd0f9 Feb 21, 2025
168 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@sshlyapn sshlyapn sshlyapn left review comments

@vladimir-paramuzov vladimir-paramuzov vladimir-paramuzov approved these changes

Assignees
No one assigned
Labels
category: GPU OpenVINO GPU plugin
Projects
None yet
Milestone
2025.1
Development

Successfully merging this pull request may close these issues.
4 participants
@michal-miotk @sshlyapn @vladimir-paramuzov @p-durandin