Re: [PATCH v3 04/10] lib/igt_drm_netlink: add get_error_counter support

[Jani Nikula <jani.nikula@intel.com>]

On Mon, 24 Aug 2026, Ravi Kishore Koppuravuri <ravi.kishore.koppuravuri@intel.com> wrote:
> Add netlink request/response handling for DRM_RAS_CMD_GET_ERROR_COUNTER.
>
> Signed-off-by: Ravi Kishore Koppuravuri <ravi.kishore.koppuravuri@intel.com>
> ---
> v2:Handled ACK/FINISH/ERR nl responses using separate callbacks
>    Added a check whether error_value updated by valid callback or not
>    Added a check whether response required for the command or not
> ---
> ---
>  lib/igt_drm_netlink.c | 264 +++++++++++++++++++++++++++++++++++++++++-
>  lib/igt_drm_netlink.h |  14 +++
>  2 files changed, 276 insertions(+), 2 deletions(-)
>
> diff --git a/lib/igt_drm_netlink.c b/lib/igt_drm_netlink.c
> index fec1639b4..76223f3de 100644
> --- a/lib/igt_drm_netlink.c
> +++ b/lib/igt_drm_netlink.c
> @@ -5,6 +5,7 @@
>  
>  #include <stdbool.h>
>  #include <stdint.h>
> +#include <errno.h>
>  #include <stdio.h>
>  #include <stdlib.h>
>  #include <string.h>
> @@ -17,6 +18,225 @@
>  #include "igt_core.h"
>  #include "igt_drm_netlink.h"
>  
> +static int ras_command_cb(struct nl_msg *msg, void *arg)
> +{
> +	struct app_context *ctx = arg;
> +	struct nlmsghdr *nlh;
> +	struct genlmsghdr *gnlh;
> +	int ret;
> +
> +	nlh = nlmsg_hdr(msg);
> +	gnlh = nlmsg_data(nlh);
> +
> +	switch (gnlh->cmd) {
> +	case DRM_RAS_CMD_GET_ERROR_COUNTER: {
> +		struct nlattr *attrs[DRM_RAS_A_ERROR_COUNTER_ATTRS_MAX + 1];
> +
> +		ret = genlmsg_parse(nlh, 0, attrs,
> +				    DRM_RAS_A_ERROR_COUNTER_ATTRS_MAX, NULL);
> +		if (ret < 0)
> +			return NL_SKIP;
> +
> +		if (!attrs[DRM_RAS_A_ERROR_COUNTER_ATTRS_ERROR_VALUE])
> +			return NL_SKIP;
> +
> +		ctx->error_value = nla_get_u32(attrs[DRM_RAS_A_ERROR_COUNTER_ATTRS_ERROR_VALUE]);
> +		break;
> +	}
> +	default:
> +		return NL_SKIP;
> +	}
> +
> +	ctx->reply_received = true;
> +	if (ctx->ack_received)
> +		ctx->cmd_done = true;
> +
> +	return NL_OK;
> +}
> +
> +static int ras_error_cb(struct sockaddr_nl *nla, struct nlmsgerr *err, void *arg)
> +{
> +	struct app_context *ctx = arg;
> +
> +	ctx->last_nl_error = err->error;
> +	ctx->cmd_done = true;
> +
> +	return NL_STOP;
> +}
> +
> +static int ras_ack_cb(struct nl_msg *msg, void *arg)
> +{
> +	struct app_context *ctx = arg;
> +
> +	ctx->ack_received = true;
> +	if (!ctx->response_required || ctx->reply_received)
> +		ctx->cmd_done = true;
> +
> +	return NL_STOP;
> +}
> +
> +static int ras_finish_cb(struct nl_msg *msg, void *arg)
> +{
> +	struct app_context *ctx = arg;
> +
> +	ctx->cmd_done = true;
> +
> +	return NL_STOP;
> +}
> +
> +static int register_callbacks(struct nl_cb *cb, struct app_context *ctx)
> +{
> +	int ret;
> +
> +	ret = nl_cb_set(cb, NL_CB_VALID, NL_CB_CUSTOM, ras_command_cb, ctx);
> +	if (ret < 0)
> +		return ret;
> +
> +	ret = nl_cb_set(cb, NL_CB_ACK, NL_CB_CUSTOM, ras_ack_cb, ctx);
> +	if (ret < 0)
> +		return ret;
> +
> +	ret = nl_cb_set(cb, NL_CB_FINISH, NL_CB_CUSTOM, ras_finish_cb, ctx);
> +	if (ret < 0)
> +		return ret;
> +
> +	ret = nl_cb_err(cb, NL_CB_CUSTOM, ras_error_cb, ctx);
> +	if (ret < 0)
> +		return ret;
> +
> +	return 0;
> +}
> +
> +static int send_and_recv_nl_msg(struct app_context *ctx,
> +				struct nl_cb *cb,
> +				struct nl_msg *msg)
> +{
> +	int ret;
> +
> +	ret = nl_send_auto(ctx->sock, msg);
> +	nlmsg_free(msg);
> +	if (ret < 0) {
> +		nl_cb_put(cb);
> +		return ret;
> +	}
> +
> +	/* NL Reply and ACK arrive as separate responses */
> +	while (!ctx->cmd_done) {
> +		ret = nl_recvmsgs(ctx->sock, cb);
> +		if (ret < 0)
> +			break;
> +	}
> +	nl_cb_put(cb);
> +
> +	return ret < 0 ? ret : 0;
> +}
> +
> +static int send_command(struct app_context *ctx, uint8_t cmd)
> +{
> +	struct nl_cb *cb;
> +	struct nl_msg *msg;
> +	void *msg_head;
> +	int ret;
> +
> +	msg = nlmsg_alloc();
> +	if (!msg)
> +		return -ENOMEM;
> +
> +	ctx->last_nl_error = 0;
> +	ctx->cmd_done = false;
> +	ctx->reply_received = false;
> +	ctx->ack_received = false;
> +	ctx->response_required = false;
> +
> +	msg_head = genlmsg_put(msg,
> +			       NL_AUTO_PORT,
> +			       NL_AUTO_SEQ,
> +			       ctx->family_id,
> +			       0,
> +			       NLM_F_REQUEST | NLM_F_ACK,
> +			       cmd,
> +			       DRM_RAS_FAMILY_VERSION);
> +	if (!msg_head) {
> +		nlmsg_free(msg);
> +		return -ENOMEM;
> +	}
> +
> +	switch (cmd) {
> +	case DRM_RAS_CMD_GET_ERROR_COUNTER:
> +		ctx->response_required = true;
> +		ret = nla_put_u32(msg,
> +				  DRM_RAS_A_ERROR_COUNTER_ATTRS_NODE_ID,
> +				  ctx->node_id);
> +		if (ret < 0) {
> +			nlmsg_free(msg);
> +			return ret;
> +		}
> +
> +		ret = nla_put_u32(msg,
> +				  DRM_RAS_A_ERROR_COUNTER_ATTRS_ERROR_ID,
> +				  ctx->error_id);
> +		if (ret < 0) {
> +			nlmsg_free(msg);
> +			return ret;
> +		}
> +		break;
> +	default:
> +		nlmsg_free(msg);
> +		return -EOPNOTSUPP;
> +	}
> +
> +	cb = nl_cb_alloc(NL_CB_DEFAULT);
> +	if (!cb) {
> +		nlmsg_free(msg);
> +		return -ENOMEM;
> +	}
> +
> +	ret = register_callbacks(cb, ctx);
> +	if (ret < 0) {
> +		nl_cb_put(cb);
> +		nlmsg_free(msg);
> +		return ret;
> +	}
> +
> +	return send_and_recv_nl_msg(ctx, cb, msg);
> +}
> +
> +static int init_app_context(struct app_context *ctx)
> +{
> +	if (!ctx)
> +		return -EINVAL;
> +
> +	ctx->sock = NULL;
> +	ctx->node_id = UINT32_MAX;
> +	ctx->error_id = UINT32_MAX;
> +	ctx->error_value = 0;
> +	ctx->last_nl_error = 0;
> +	ctx->cmd_done = false;
> +	ctx->reply_received = false;
> +	ctx->ack_received = false;
> +	ctx->response_required = false;
> +	ctx->family_id = -1;
> +
> +	return 0;
> +}
> +
> +static void cleanup_app_context(struct app_context *ctx)
> +{
> +	if (!ctx)
> +		return;
> +
> +	ctx->sock = NULL;
> +	ctx->node_id = UINT32_MAX;
> +	ctx->error_id = UINT32_MAX;
> +	ctx->error_value = 0;
> +	ctx->last_nl_error = 0;
> +	ctx->cmd_done = false;
> +	ctx->reply_received = false;
> +	ctx->ack_received = false;
> +	ctx->response_required = false;
> +	ctx->family_id = -1;
> +}
> +
>  void igt_cleanup_nl_socket(struct app_context *ctx)
>  {
>  	if (!ctx || !ctx->sock)
> @@ -24,14 +244,20 @@ void igt_cleanup_nl_socket(struct app_context *ctx)
>  
>  	nl_close(ctx->sock);
>  	nl_socket_free(ctx->sock);
> -	ctx->sock = NULL;
> -	ctx->family_id = -1;
> +
> +	cleanup_app_context(ctx);
>  
>  	igt_debug("Cleaned up netlink socket.\n");
>  }
>  
>  int igt_init_nl_socket(struct app_context *ctx)
>  {
> +	int ret;
> +
> +	ret = init_app_context(ctx);
> +	if (ret < 0)
> +		return ret;
> +
>  	ctx->sock = nl_socket_alloc();
>  	if (!ctx->sock)
>  		return -1;
> @@ -58,3 +284,37 @@ int igt_init_nl_socket(struct app_context *ctx)
>  		  DRM_RAS_FAMILY_NAME, ctx->family_id);
>  	return 0;
>  }
> +
> +int igt_get_error_counter(struct app_context *ctx)
> +{
> +	int ret;
> +
> +	if (!ctx || !ctx->sock || ctx->family_id < 0)
> +		return -EINVAL;
> +
> +	if (ctx->node_id == UINT32_MAX ||
> +	    ctx->error_id == UINT32_MAX ||
> +	    ctx->error_id == 0) {
> +		igt_warn("Invalid node_id (%u) or error_id (%u) provided. "
> +			 "node_id should be >= 0 and error_id should be >= 1.\n",
> +			 ctx->node_id, ctx->error_id);
> +		return -EINVAL;
> +	}
> +
> +	ctx->error_value = UINT32_MAX;
> +
> +	ret = send_command(ctx, DRM_RAS_CMD_GET_ERROR_COUNTER);
> +	if (ret < 0)
> +		return ret;
> +
> +	if (ctx->error_value == UINT32_MAX) {
> +		igt_warn("No valid error counter reply: node_id=%u error_id=%u kernel_error=%d\n",
> +			 ctx->node_id, ctx->error_id, ctx->last_nl_error);
> +		return ctx->last_nl_error ? ctx->last_nl_error : -ENODATA;
> +	}
> +
> +	igt_debug("Retrieved error counter: node_id=%u error_id=%u value=%u\n",
> +		  ctx->node_id, ctx->error_id, ctx->error_value);
> +
> +	return 0;
> +}
> diff --git a/lib/igt_drm_netlink.h b/lib/igt_drm_netlink.h
> index f681a3a81..92ac665e2 100644
> --- a/lib/igt_drm_netlink.h
> +++ b/lib/igt_drm_netlink.h
> @@ -9,17 +9,31 @@
>  #include <stdbool.h>
>  #include <stdint.h>

Okay so this change should add them I guess.

>  
> +#include <linux/genetlink.h>
> +
> +#include <netlink/attr.h>
> +#include <netlink/handlers.h>
> +#include <netlink/msg.h>

None of these are used here though.

>  #include <netlink/netlink.h>
>  
>  #include <drm-uapi/drm_ras.h>
>  
>  struct app_context {
>  	struct nl_sock *sock;
> +	uint32_t node_id;
> +	uint32_t error_id;
> +	uint32_t error_value;
> +	int last_nl_error;
> +	bool cmd_done;
> +	bool reply_received;
> +	bool ack_received;
> +	bool response_required;
>  	int family_id;
>  };
>  
>  void igt_cleanup_nl_socket(struct app_context *ctx);
>  int igt_init_nl_socket(struct app_context *ctx);
> +int igt_get_error_counter(struct app_context *ctx);

This is really inconsistent naming, though.

>  
>  #endif /* IGT_DRM_NETLINK_H */

-- 
Jani Nikula, Intel