CUDA Plugin EP: NHWC Cleanup & Hardening

[tianleiwu]

Summary

Unifies the NHWC-eligible op allowlist between the bundled CUDA EP and the CUDA plugin EP into a single shared header, adds kernel-miss diagnostics, and expands NHWC test coverage from 4 ops to 11.

Motivation

The bundled EP (cuda_execution_provider.cc) and the plugin EP (plugin/cuda_ep.cc) independently maintained their own copies of the NHWC allowlist. This created a maintenance hazard where ops could be added to one but not the other, leading to silent divergence. Additionally, there was no runtime diagnostic when the framework rewrote a node to the NHWC domain but the plugin EP lacked a matching kernel — failures were silent fallbacks to CPU.

Key Changes

Shared NHWC Allowlist (cuda_nhwc_ops.h)

Item	Detail
New file	onnxruntime/core/providers/cuda/cuda_nhwc_ops.h
Contents	IsNhwcEligibleOnnxOp(), IsNhwcEligibleMsOp(), IsNhwcEligible() inline functions
Ops covered	AveragePool, BatchNormalization, Conv, ConvTranspose, DepthToSpace, GlobalAveragePool, GlobalMaxPool, GridSample, LRN, MaxPool, SpaceToDepth (+ MS-domain GridSample)

Bundled EP Refactor (cuda_execution_provider.cc)

• Removed the static std::unordered_set<std::string_view> cuda_nhwc_onnx_ops and the inline domain check logic.
• Replaced with a single call to cuda::IsNhwcEligible(node_domain, node_op_type).

Plugin EP Refactor & Diagnostics (plugin/cuda_ep.cc)

• ShouldConvertDataLayoutForOpImpl: Replaced ~20 lines of static set + domain checks with a single cuda::IsNhwcEligible() call.
• GetCapabilityImpl: Added a WARNING-level diagnostic in the else branch (kernel not found). When a node in the com.ms.internal.nhwc domain has no registered kernel, the log emits the op type, domain, version, and node name — making future NHWC registration gaps immediately visible at session creation.

Expanded NHWC Test Coverage (test_cuda_plugin_ep.py)

• Added _assert_nhwc_domain_assigned() helper that verifies NHWC layout transformation occurred by checking for framework-inserted Transpose nodes in the EP's assignment info.
• Added _run_nhwc_model_test() helper combining domain assertion + numerical validation.
• Updated 4 existing NHWC tests (Conv, BatchNormalization, MaxPool, AveragePool) to include structural assertions.
• Added 7 new NHWC test methods:
  • test_nhwc_conv_transpose
  • test_nhwc_global_max_pool
  • test_nhwc_global_average_pool
  • test_nhwc_depth_to_space
  • test_nhwc_space_to_depth
  • test_nhwc_lrn
  • test_nhwc_grid_sample

Testing Notes

Run the full CUDA plugin EP test suite with NHWC enabled:

bash .env/cuda13_plugin.sh --build --install --test_plugin

Or run only the NHWC tests directly:

cd onnxruntime/test/python/transformers
ORT_TEST_CUDA_PLUGIN_EP=1 python -m unittest \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_conv \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_batch_normalization \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_maxpool \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_avgpool \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_conv_transpose \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_global_max_pool \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_global_average_pool \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_depth_to_space \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_space_to_depth \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_lrn \
  test_cuda_plugin_ep.TestCudaPluginEP.test_nhwc_grid_sample

All 86 tests in the suite pass (11 NHWC + 75 existing), with no regressions.

[Copilot]

Pull request overview

This PR consolidates the CUDA NHWC layout-conversion allowlist into a shared header used by both the in-tree CUDA EP and the CUDA plugin EP, adds a plugin-side warning diagnostic for NHWC-domain kernel misses, and expands CUDA plugin EP NHWC test coverage with new structural assertions plus additional operator tests.

Changes:

• Introduce a shared NHWC eligibility helper (cuda_nhwc_ops.h) and replace duplicated allowlists in both CUDA EP implementations with cuda::IsNhwcEligible(...).
• Add a warning diagnostic in the CUDA plugin EP when an internal NHWC-domain node has no matching registered kernel.
• Expand and harden NHWC-focused plugin EP tests, adding helpers and new op coverage.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 4 comments.

File	Description
onnxruntime/test/python/transformers/test_cuda_plugin_ep.py	Adds NHWC structural assertions and expands NHWC operator coverage in CUDA plugin EP tests.
onnxruntime/core/providers/cuda/plugin/cuda_ep.cc	Switches layout-conversion logic to shared helper and adds NHWC kernel-miss warning diagnostics.
onnxruntime/core/providers/cuda/cuda_nhwc_ops.h	New shared header providing unified NHWC eligibility allowlist helpers.
onnxruntime/core/providers/cuda/cuda_execution_provider.cc	Refactors NHWC eligibility check to use the shared helper.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

You can commit the suggested changes from lintrunner.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated no new comments.

[yuslepukhin]

Approving with comments.