About doca-rdmi
An agent skill for external developers using DOCA RDMI to initiate one-sided RDMA operations from the DPA datapath. It guides connection lifecycles, DPA handle handoff, version and experimental-symbol checks, and practical workflows while distinguishing RDMI from general-purpose RDMA.
Key features
- 01RDMI versus RDMA selection guidance
- 02Connection and poster lifecycle guidance
- 03DPA completion and handle handoff
- 04Experimental symbol and version checks
- 05Configure, build, run, and debug workflows
Use cases
- 01Posting RDMA writes from a DPA kernel to a remote responder
- 02Setting up an RDMI connection with DPA completions
- 03Checking symbol status in an installed DOCA version
- 04Diagnosing errors from RDMI calls
Install
Copy the command and run it in a terminal.
npx skills add https:/ / github.com/ nvidia/ skills --skill doca-rdmiAfter installing, type/doca-rdmiin Claude Code to confirm it loaded.
Bundled files
References and scripts installed along with SKILL.md.
BENCHMARK.mdCAPABILITIES.mdskill-card.mdskill.oms.sigSKILLCARD.yamlTASKS.mdevals/evals.jsonDetails
From the project READMEDOCA RDMA Initiator
Examples
- "DPA kernel needs to post RDMA writes to a remote responder."
— this is in scope. Stay on
doca_rdmi_*; do not rewrite it as host-CPUdoca-rdma. - "Function not found on doca_rdmi_ at link time."* — audit whether the symbol is EXPERIMENTAL on this DOCA before inventing a fallback API.
Instructions
Follow Loading order below. Open the tasks companion for configure / build / debug. Open the capabilities companion for RDMI vs RDMA and the experimental-symbol rule. Do not open every companion file.
Where to start: This skill assumes DOCA is already installed and
the user is doing hands-on RDMI work on a host or BlueField with
the DOCA package set that ships the doca-rdmi library. Open
the tasks companion if the user wants to do something
(install / configure / build / modify / run / test / debug / use);
open the capabilities companion when the question is
what can RDMI express on this version — the object model, the
DPA-side handle types, the relationship to doca-rdma, the
EXPERIMENTAL-tag policy, and the safety overlay.
End-to-end "walk me through doca-rdmi" questions are answerable entirely from this skill. Go straight to
end-to-end (quickref),
which carries the self-contained install-check → device/cap
discovery → sample → pkg-config build → run → debug walkthrough
with the exact commands. You do not need to open doca-setup or
doca-programming-guide to answer an RDMI build/run/debug question.
Route to doca-setup/SKILL.md when the required
DOCA prerequisites are absent, partial, or version-mismatched.
Example questions this skill answers well
The CLASSES of RDMI questions this skill is built to answer, each with one worked example. The agent should treat the class as the load-bearing piece — the worked example is a single instance.
- "Should I use
doca-rdmiordoca-rdmafor this case?" — worked example: "I have a DPA kernel that needs to fire 1 MB RDMA writes at a remote responder; which library?". Answered by the initiator-side vs general-purpose selection rule inCapabilities and modes
surface-selection table + the routing back todoca-rdma/SKILL.mdwhen the use case is two-sided or host-CPU initiated. - "How do I bring up an RDMI connection on the DPA datapath?" —
worked example: "create a
doca_rdmi_connection, attach a DPA completion context, hand the DPA-side handle to my kernel". Answered by the connection-object lifecycle inCapabilities and modes
- the configure walk in
configure.
- "How do connection and poster relate — when do I need both?"
— worked example: "my application receives work requests AND
posts RDMA writes; do I need a
doca_rdmi_connectionplus adoca_rdmi_poster, or one of them?". Answered by the two-object model inCapabilities and modes
- the modify-from-sample slot table in
modify.
- "How do I drive completions on the DPA side?" — worked
example: "hook the connection to a
doca_dpa_completionso my kernel polls completions directly". Answered by the DPA-side completion-attach pattern inCapabilities and modes
- the run-side wiring in
run, cross-linked into
doca-dpa/SKILL.mdfor the DPA programming surface itself. - "Is the symbol I want available — and is it stable enough to
ship?" — worked example: "is
doca_rdmi_poster_postGA on my installed DOCA, or still EXPERIMENTAL?". Answered by the EXPERIMENTAL-tag policy inVersion compatibility
- the version-discovery rule
(
pkg-config --modversion doca-rdmi) pinned in
configure.
- the version-discovery rule
(
- "What does this
DOCA_ERROR_*from adoca_rdmi_*call mean?" — worked example: "DOCA_ERROR_BAD_STATEfromdoca_rdmi_connection_dpa_completion_attach". Answered by the RDMI overlay on the cross-library taxonomy inError taxonomy
- the layered ladder in ## debug
that escalates to
doca-debug/SKILL.md.
- the layered ladder in ## debug
that escalates to
Audience
This skill serves external developers building DPA-resident DOCA
applications that need to initiate one-sided RDMA operations
against a remote responder — i.e., users whose accelerator-side
code wants to post sends, writes, or reads directly from the
accelerator without round-tripping through the host CPU. The
canonical caller is a DPA kernel that has been compiled with
doca-dpacc-compiler and runs on the BlueField DPA datapath; a
GPU-side caller that drives the DPU's RDMA queues is the
sister case routed to doca-gpi/SKILL.md. This
skill is not for NVIDIA developers contributing to DOCA RDMI
itself, and it is not the right surface for general host-CPU
two-sided RDMA — that belongs to
doca-rdma/SKILL.md.
Language scope
DOCA RDMI ships as a C library with the pkg-config module name
doca-rdmi. The library's host-side surface
(doca_rdmi_connection_*, doca_rdmi_poster_*) is C; the DPA-side
surface that the accelerator kernel uses is also C, compiled against
the DOCA DPA toolchain documented in
doca-dpa/SKILL.md. Other-language consumers (Rust,
Go, Python, …) consume the host-side *.so through FFI; the skill's
contribution in that case is to keep the connection / poster
lifecycle, the EXPERIMENTAL-tag policy, the DPA-side handoff rules,
and the safety overlay language-neutral, and to route the agent to
the public C ABI as the authoritative surface that any wrapper will
eventually call. The DPA-side surface is not wrappable in another
language — it is compiled and linked into the DPA binary itself.
When to load this skill
Load this skill when the user is doing hands-on DOCA RDMI work on a host or BlueField with DOCA installed. Concretely:
- Deciding between
doca-rdmianddoca-rdmafor a new one-sided RDMA workload that runs from the DPA datapath. - Creating a
doca_rdmi_connectionordoca_rdmi_poster, attaching adoca_dpa_completionor adoca_verbs_cq, and starting the context on the DPA datapath. - Wiring the DPA-side handle returned by
doca_rdmi_connection_get_dpa_handle/doca_rdmi_poster_get_dpa_handleinto a DPA kernel that calls the matching device-side header (doca_rdmi_dev_connection.h,doca_rdmi_dev_poster.h,doca_rdmi_dev_cqe.h). - Acknowledging receive completions on the host side via
doca_rdmi_connection_recv_ackafter the DPA kernel consumed them. - Auditing which RDMI symbols are EXPERIMENTAL on the installed DOCA version, before declaring an RDMI-using component production-stable.
- Debugging a
DOCA_ERROR_*returned by adoca_rdmi_*call and deciding whether the cause is a configuration mistake, a lifecycle ordering bug, an unsupported capability on this device, or a layer below DOCA.
Do not load this skill for general DOCA orientation, install of
DOCA itself, or two-sided host-CPU RDMA questions. For those, use
doca-public-knowledge-map/SKILL.md,
doca-setup/SKILL.md, and
doca-rdma/SKILL.md respectively.
What this skill provides
This is a thin loader. The body keeps only the orientation needed to pick the right next file. The substantive RDMI-specific material lives in two companion files:
- the capabilities companion — what RDMI can express on this version: the
doca_rdmi_connectionanddoca_rdmi_posterobject model, the DPA-side completion-attach pattern, the relationship todoca-verbs(RDMI builds on adoca_verbs_context) and todoca-dpa(the accelerator-side datapath), the EXPERIMENTAL-tag rule for version handling, the RDMI overlay on the cross-libraryDOCA_ERROR_*taxonomy, the observability surface (completion events on the PE / DPA-side completions), and the safety policy that gates posting work from an accelerator kernel into a remote responder's memory. - the tasks companion — step-by-step workflows for the eight in-scope verbs:
install,configure,build,modify,run,test,debug,use. Plus a## rollbackoverlay (RDMI-specific five-step teardown for the verbs / connection / DPA-attach / MR stack) and the 5-phase universal debug-loop instantiation appended to## debug. Plus aDeferred task verbsblock that points out-of-scope questions at the right next skill.
The skill assumes a host or BlueField where DOCA is already
installed at the standard location and the user has the privileges
their public install profile expects. It does not cover installing
DOCA itself — that path goes through
doca-setup/SKILL.md.
What this skill deliberately does not ship
This skill is agent guidance, not a samples or templates bundle. To keep the boundary clean, it deliberately does not contain — and pull requests should not add:
- Pre-written DOCA RDMI application source code, in any
language. The agent's job is to route the user to verified
reference code on the user's installed DOCA and to prescribe a
minimum-diff modification via the universal modify-a-sample
workflow in
doca-programming-guide/SKILL.md, layered with the RDMI-specific overrides inmodify. Because every RDMI
symbol is EXPERIMENTAL at the time of writing, the skill refuses to author RDMI source code from documentation prose — the API can change between releases and the resulting code may not even compile. - Standalone build manifests (
meson.build,CMakeLists.txt,Cargo.toml, …) parked inside the skill. The agent constructs the build manifest in the user's project directory against the user's installed DOCA, wherepkg-config --modversion doca-rdmiis the source of truth. - DPA-side kernel templates. The DPA-side surface is owned by
doca-dpa/SKILL.md; RDMI's DPA-side headers are consumed by the DPA programming model documented there. This skill names the RDMI-specific handoff (the DPA handle type, the completion-attach call) but does not author DPA kernels. - A
samples/,bindings/, orreference/subtree of any kind. A mock or incomplete artifact in this skill's tree, even one labeled "reference", is misleading: users will read it as buildable.
Loading order
- Read this skill first to confirm the user's question is in scope.
- For the RDMI object model, the DPA-side handoff pattern, the EXPERIMENTAL-tag policy, the error taxonomy, observability, and safety policy, see the capabilities companion.
- For step-by-step workflows — install, configure, build, modify, run, test, debug, use — see the tasks companion.
Both companion files cross-link to each other and to
doca-public-knowledge-map/SKILL.md
whenever the right answer is "look it up in the public docs or the
installed package layout" rather than "RDMI-specific guidance".
Related skills
doca-rdma/SKILL.md— the higher-level RDMA library covering two-sided and host-CPU-initiated RDMA. RDMI is the focused initiator-side surface; doca-rdma is the right answer for the majority of RDMA use cases. The selection table in ## Capabilities and modes is the load-bearing decision aid.doca-dpa/SKILL.md— the DOCA DPA programming surface. RDMI returns DPA-side handles (doca_dpa_dev_rdmi_connection_t,doca_rdmi_dev_poster_t) that the DPA kernel uses through the device-side headers (doca_rdmi_dev_connection.h,doca_rdmi_dev_poster.h,doca_rdmi_dev_cqe.h); the DPA toolchain, kernel build, and execution model are owned by that skill.doca-gpi/SKILL.md— the GPU-side sister of this skill. GPI is the channel/queue surface a CUDA kernel uses to initiate RDMA; RDMI is the DPA-side surface. Both layer on the same DOCA RDMA / verbs substrate; either may apply depending on whether the initiator is on the DPA or on the GPU.doca-public-knowledge-map/SKILL.md— the routing table for every public DOCA documentation source and the on-disk layout of an installed DOCA package.doca-setup/SKILL.md— env preparation, install verification, and the I have no install yet path with the public NGC DOCA container.doca-programming-guide/SKILL.md— general DOCA programming patterns shared by every library: the canonicalpkg-config+ meson build pattern, the universal modify-a-shipped-sample first-app workflow, the universal Core-context lifecycle, the cross-libraryDOCA_ERROR_*taxonomy. This skill layers RDMI specifics on top.doca-debug/SKILL.md— the cross-cutting debug ladder (install / version / build / link / runtime / program / driver). RDMI-specific debug overlays on top of it.doca-hardware-safety/SKILL.md— the bundle-wide hardware-safety meta-policy. The## Safety policyoverlay in the capabilities companion cross-links it.doca-version/SKILL.md— the version detection / four-way match rule every per-artifact## Version compatibilityanchor builds on. This skill quotes the RDMI-specific overlay only.doca-structured-tools-contract/SKILL.md— the JSON-schema contracts for the agent-preferred structured helpers (env probe, capability snapshot, version-matrix lookup); the## Command appendixin the tasks companion defers to them before falling back to the manual chain.