Re: [PATCH v1 1/2] drm/xe/xe_ras: Add structures and commands for RAS GPU health indicator

[Rodrigo Vivi <rodrigo.vivi@intel.com>]

On Thu, Apr 16, 2026 at 03:06:09PM +0530, Soham Purkait wrote:
> Add the sysctrl commands and response structures for GPU health
> indicator supported by RAS.
> 

As Andy pointed out, please improve the commit message.

> Signed-off-by: Soham Purkait <soham.purkait@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_ras_types.h             | 65 +++++++++++++++++++
>  drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h | 15 +++++
>  2 files changed, 80 insertions(+)
>  create mode 100644 drivers/gpu/drm/xe/xe_ras_types.h
> 
> diff --git a/drivers/gpu/drm/xe/xe_ras_types.h b/drivers/gpu/drm/xe/xe_ras_types.h
> new file mode 100644
> index 000000000000..a05cfd971acd
> --- /dev/null
> +++ b/drivers/gpu/drm/xe/xe_ras_types.h
> @@ -0,0 +1,65 @@
> +/* SPDX-License-Identifier: MIT */
> +/*
> + * Copyright © 2026 Intel Corporation
> + */
> +
> +#ifndef _XE_RAS_TYPES_H_
> +#define _XE_RAS_TYPES_H_
> +
> +#include <linux/types.h>
> +
> +/**
> + * typedef xe_ras_health_status_t - Device health status values (8-bit)
> + *
> + * Three-state health indicator value reported by system controller commands.
> + *
> + * The expected values are:
> + * - 0: ok  - The device is healthy and operating within normal parameters.
> + * - 1: warning  - The device is experiencing minor issues but is still operational.
> + * - 2: critical  - The device is in a critical state and may not be operational.
> + */
> +typedef u8 xe_ras_health_status_t;

Please use enum instead of this...

> +
> +/**
> + * struct xe_ras_health_get_input - Input for XE_SYSCTRL_CMD_GET_HEALTH
> + */
> +struct xe_ras_health_get_input {
> +	/** @reserved: Reserved for future use, must be 0 */
> +	u32 reserved[2];
> +} __packed;
> +
> +/**
> + * struct xe_ras_health_get_response - Response for XE_SYSCTRL_CMD_GET_HEALTH
> + */
> +struct xe_ras_health_get_response {
> +	/** @current_health: Current health status (OK/WARNING/CRITICAL) */
> +	xe_ras_health_status_t current_health;
> +	/** @reserved: Reserved for alignment */
> +	u8 reserved[3];
> +} __packed;
> +
> +/**
> + * struct xe_ras_health_set_input - Input for XE_SYSCTRL_CMD_SET_HEALTH
> + */
> +struct xe_ras_health_set_input {
> +	/** @new_health: New health status to set */
> +	xe_ras_health_status_t new_health;
> +	/** @reserved: Reserved for alignment */
> +	u8 reserved[3];
> +} __packed;
> +
> +/**
> + * struct xe_ras_health_set_response - Response for XE_SYSCTRL_CMD_SET_HEALTH
> + */
> +struct xe_ras_health_set_response {
> +	/** @operation_status: Status of set operation (RAS_STATUS_*) */
> +	u32 operation_status;
> +	/** @current_health: Health status after this change */
> +	xe_ras_health_status_t current_health;
> +	/** @reserved: Reserved for alignment */
> +	u8 reserved[3];
> +	/** @reserved_2: Reserved for future expansion */
> +	u32 reserved_2[2];
> +} __packed;
> +
> +#endif /* _XE_RAS_TYPES_H_ */
> diff --git a/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h b/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h
> index 89456aec6097..fc73e02c3202 100644
> --- a/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h
> +++ b/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h
> @@ -10,6 +10,21 @@
>  
>  #include "abi/xe_sysctrl_abi.h"
>  
> +/**
> + * enum xe_sysctrl_mailbox_command_id - RAS Command ID's for GFSP group
> + *
> + * @XE_SYSCTRL_CMD_GET_HEALTH: Get current health status
> + * @XE_SYSCTRL_CMD_SET_HEALTH: Set new health status
> + */
> +enum xe_sysctrl_mailbox_command_id {
> +	XE_SYSCTRL_CMD_GET_HEALTH = 0x0B,
> +	XE_SYSCTRL_CMD_SET_HEALTH = 0x0C
> +};
> +
> +enum xe_sysctrl_group {
> +	XE_SYSCTRL_GROUP_GFSP = 1
> +};
> +
>  /**
>   * struct xe_sysctrl_mailbox_command - System Controller mailbox command
>   */
> -- 
> 2.34.1
>