Skip to content

test(ad): GPU/packed-field AD parity for the exchange rules #43

Open

Description

@kylebeggs

ext/MatrixFreeOperatorsEnzymeCoreExt.jl registers rules on _exchange_storage!/_bc_storage!, which are layout-agnostic: they take raw storage plus a BlockLayout tag, so PackedBlockField goes through the same rules as BlockField. That is the mechanism DESIGN.md §6 calls mandatory for @kernel-authored leaves — the packed GPU exchange must never be differentiated through, and now it structurally cannot be.

None of that is tested on a GPU. Specifically:

  • No AD test uses a packed field at all, on any backend. src/packedfield.jl:17-18 still designates BlockField as "the reference layout for AD".
  • test/device.jl / test/device_gpu.jl never differentiate, so MFO_TEST_GPU=true adds no AD coverage.
  • The device exchange path (_run_exchange!(::PackedBlockField, ...) in src/transfer_kernels.jl:194-198) dispatches to batched kernels below the rule seam only when the backend is a real GPU. On CPU it falls back to _run_exchange_host!, which is what the current tests exercise — so the rule is verified against the fallback, not against the kernels.

What to add

  1. CPU-side first — the quick, self-contained first deliverable. It needs no hardware and can land as its own PR ahead of steps 2-3: a packed-field AD test asserting the gradient matches the declared adjoint and that the rules fire (ENZ_EXT.rule_hits() counters increase across the call), mirroring test/enzyme_rules.jl.
  2. Under MFO_TEST_GPU, the same against CUDA arrays — the real question being whether the rule's reverse body (_exchange_storage_adjoint!, which stays on host descriptor loops on all backends by design, see transfer_kernels.jl:16-23) composes correctly with a forward pass that ran as batched kernels.
  3. Decide and document whether GPU reverse mode is supported or explicitly refused. Right now it is neither — it is untested.

—
🤖 Claude fixed a line number that had wandered off.
claude-opus-5-5[1m] (high) · claude-code 2.1.285 · unreviewed

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions