On Sat Feb 8, 2025 at 2:58 AM JST, Zhi Wang wrote:
> Introduce a kernel doc to describe the GSP message handling policy.
>
> Signed-off-by: Zhi Wang <zhiw@xxxxxxxxxx>
> ---
> Documentation/gpu/nouveau.rst | 3 +++
> .../gpu/drm/nouveau/include/nvkm/subdev/gsp.h | 17 +++++++++++++++++
> 2 files changed, 20 insertions(+)
>
> diff --git a/Documentation/gpu/nouveau.rst b/Documentation/gpu/nouveau.rst
> index 0f34131ccc27..b8c801e0068c 100644
> --- a/Documentation/gpu/nouveau.rst
> +++ b/Documentation/gpu/nouveau.rst
> @@ -27,3 +27,6 @@ GSP Support
>
> .. kernel-doc:: drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c
> :doc: GSP message queue element
> +
> +.. kernel-doc:: drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h
> + :doc: GSP message handling policy
> diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h
> index bc16510261b8..2d0b80a733d7 100644
> --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h
> +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h
> @@ -31,6 +31,23 @@ typedef int (*nvkm_gsp_msg_ntfy_func)(void *priv, u32 fn, void *repv, u32 repc);
> struct nvkm_gsp_event;
> typedef void (*nvkm_gsp_event_func)(struct nvkm_gsp_event *, void *repv, u32 repc);
>
> +/**
> + * DOC: GSP message handling policy
> + *
> + * When sending a GSP RPC command, there can be multiple cases of handling
> + * the GSP RPC messages, which are the reply of GSP RPC commands, according
> + * to the requirement of the callers and the nature of the GSP RPC commands.
> + *
> + * If none is specified, the policy is the callers don't care. Immediately
> + * return to the caller after the GSP RPC command is issued.
> + *
> + * NVKM_GSP_RPC_REPLY_RECV - If specified, wait and receive the entire GSP
> + * RPC message after the GSP RPC command is issued.
> + *
> + * NVKM_GSP_RPC_REPLY_POLL - If specified, wait for the specific reply and
> + * discard the reply before returning to the caller.
> + *
> + */
> enum {
> NVKM_GSP_RPC_REPLY_RECV = 1,
> NVKM_GSP_RPC_REPLY_POLL,
It would IMHO look more natural if the documentation was introduced at
the same time as the variants themselves, so they are not left
undocumented for some range of the git history.
