Skip to content

[GPU] Improve max-allocation error guidance and enable_large_allocations docs#34894

Open
Sahilbhatane wants to merge 6 commits intoopenvinotoolkit:masterfrom
Sahilbhatane:34856-gpu-max-alloc-error-guidance
Open

[GPU] Improve max-allocation error guidance and enable_large_allocations docs#34894
Sahilbhatane wants to merge 6 commits intoopenvinotoolkit:masterfrom
Sahilbhatane:34856-gpu-max-alloc-error-guidance

Conversation

@Sahilbhatane
Copy link

Details:

  • When a single allocation exceeds the device max allocation size, engine::check_allocatable() fails with an assertion. The message had a small formatting issue (missing space before "Please") and didn’t point users to the GPU_ENABLE_LARGE_ALLOCATIONS / ov::intel_gpu::hint::enable_large_allocations workaround.
  • Updated the error message in src/plugins/intel_gpu/src/runtime/engine.cpp to fix the spacing and include a clear suggestion to enable GPU_ENABLE_LARGE_ALLOCATIONS when applicable.
  • Aligned the internal option description in src/plugins/intel_gpu/include/intel_gpu/runtime/options.inl with the actual behavior. The check is against the device max allocation size, not just “>4GB”.
  • Extended the public property documentation in src/inference/include/openvino/runtime/intel_gpu/properties.hpp to clarify that enabling large allocations also bypasses the max-allocation-size check, which helps when a single buffer exceeds CL_DEVICE_MAX_MEM_ALLOC_SIZE.
  • These changes are limited to user-facing messages and documentation. Validation was done by reviewing the updated text across all affected files.

Tickets:

AI Assistance:

  • AI assistance used: yes/partially
  • AI was used for issue triage, tracing the config/property flow through the codebase, and helping draft this PR description. All code and documentation changes were applied and reviewed manually in the repository.

@github-actions github-actions bot added category: inference OpenVINO Runtime library - Inference category: GPU OpenVINO GPU plugin category: CPP API OpenVINO CPP API bindings labels Mar 24, 2026
@sys-openvino-ci sys-openvino-ci added the ExternalPR External contributor label Mar 24, 2026
@Sahilbhatane Sahilbhatane marked this pull request as ready for review March 24, 2026 14:47
@Sahilbhatane Sahilbhatane requested review from a team as code owners March 24, 2026 14:47
Copilot AI review requested due to automatic review settings March 24, 2026 14:47
Copy link
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

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

Pull request overview

Improves Intel GPU plugin user-facing guidance around failures caused by a single allocation exceeding the device’s maximum allocatable buffer size, and clarifies the behavior of the enable_large_allocations option/property in internal and public docs.

Changes:

  • Update the max-allocation assertion message to fix formatting and point users to GPU_ENABLE_LARGE_ALLOCATIONS.
  • Align the internal option description for ov::intel_gpu::hint::enable_large_allocations with the real check (device max allocation size rather than “>4GB”).
  • Clarify the public property documentation to mention bypassing the max-allocation-size check and its usefulness when exceeding CL_DEVICE_MAX_MEM_ALLOC_SIZE.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File Description
src/plugins/intel_gpu/src/runtime/engine.cpp Improves the assertion text for “single allocation exceeds device max alloc size” and adds a workaround hint.
src/plugins/intel_gpu/include/intel_gpu/runtime/options.inl Updates the option help text to describe the behavior more accurately than the previous “>4GB” wording.
src/inference/include/openvino/runtime/intel_gpu/properties.hpp Extends the public property doxygen to document bypassing the max-alloc-size check and relates it to CL_DEVICE_MAX_MEM_ALLOC_SIZE.

@p-durandin
Copy link
Contributor

build_jenkins

@Sahilbhatane
Copy link
Author

@p-durandin Re-requesting build. Updated the branch, all checks passed on previous build.

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

Labels

category: CPP API OpenVINO CPP API bindings category: GPU OpenVINO GPU plugin category: inference OpenVINO Runtime library - Inference ExternalPR External contributor

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants