Re: [PATCH v6 05/10] lib/igt_drm_ras_nl: add get_error_threshold command support

["Koppuravuri, Ravi Kishore" <ravi.kishore.koppuravuri@intel.com>]

Hi Soham,

On 30-10-2026 15:57, Purkait, Soham wrote:
> Hi Ravi,
>
> On 29-10-2026 14:29, Ravi Kishore Koppuravuri wrote:
>> Add support for GET_ERROR_THRESHOLD to fetch the current value of error
>> threshold
>>
>> Signed-off-by: Ravi Kishore Koppuravuri 
>> <ravi.kishore.koppuravuri@intel.com>
>> ---
>> v6:Removed normalize_nl_error() for standard errno in 
>> igt_drm_ras_nl_get_error_threshold()
>> v4:Renamed the library functions with prefix "igt_drm_netlink_*" (Nikula
>>     Jani & Kamil)
>> v2:Added a check whether error_threshold updated by valid callback or 
>> not
>> ---
>>   lib/igt_drm_ras_nl.c | 52 +++++++++++++++++++++++++++++++++++++++++---
>>   lib/igt_drm_ras_nl.h |  4 ++++
>>   2 files changed, 53 insertions(+), 3 deletions(-)
>>
>> diff --git a/lib/igt_drm_ras_nl.c b/lib/igt_drm_ras_nl.c
>> index 7e6a61f11..d930b6e33 100644
>> --- a/lib/igt_drm_ras_nl.c
>> +++ b/lib/igt_drm_ras_nl.c
>> @@ -75,6 +75,14 @@ static int ras_command_cb(struct nl_msg *msg, void 
>> *arg)
>>           ctx->error_value = 
>> nla_get_u32(attrs[DRM_RAS_A_ERROR_COUNTER_ATTRS_ERROR_VALUE]);
>>           break;
>>       }
>> +    case DRM_RAS_CMD_GET_ERROR_THRESHOLD: {
>> +        if (!attrs[DRM_RAS_A_ERROR_COUNTER_ATTRS_ERROR_THRESHOLD])
>> +            return NL_SKIP;
>> +
>> +        ctx->error_threshold =
>> + nla_get_u32(attrs[DRM_RAS_A_ERROR_COUNTER_ATTRS_ERROR_THRESHOLD]);
>> +        break;
>> +    }
>>       default:
>>           return NL_SKIP;
>>       }
>> @@ -230,6 +238,7 @@ static int send_command(struct 
>> igt_drm_ras_nl_context *ctx, uint8_t cmd)
>>       }
>>         switch (cmd) {
>> +    case DRM_RAS_CMD_GET_ERROR_THRESHOLD:
>>       case DRM_RAS_CMD_GET_ERROR_COUNTER:
>>           ctx->response_required = true;
>>           ret = nla_put_u32(msg,
>> @@ -280,6 +289,7 @@ static void reset_drm_ras_nl_context(struct 
>> igt_drm_ras_nl_context *ctx)
>>       ctx->reply_received = false;
>>       ctx->ack_received = false;
>>       ctx->response_required = false;
>> +    ctx->error_threshold = 0;
>>       ctx->family_id = -1;
>>   }
>>   @@ -326,10 +336,8 @@ int igt_drm_ras_nl_init(struct 
>> igt_drm_ras_nl_context *ctx)
>>       return 0;
>>   }
>>   -int igt_drm_ras_nl_get_error_counter(struct igt_drm_ras_nl_context 
>> *ctx)
>> +static int validate_inputs(struct igt_drm_ras_nl_context *ctx, 
>> uint8_t cmd)
>>   {
>> -    int ret;
>> -
>>       if (!ctx || !ctx->sock || ctx->family_id < 0) {
>>           igt_warn("Invalid DRM RAS netlink context\n");
>>           return -EINVAL;
>> @@ -343,6 +351,17 @@ int igt_drm_ras_nl_get_error_counter(struct 
>> igt_drm_ras_nl_context *ctx)
>>           return -EINVAL;
>>       }
>>   +    return 0;
>> +}
>> +
>> +int igt_drm_ras_nl_get_error_counter(struct igt_drm_ras_nl_context 
>> *ctx)
>> +{
>> +    int ret;
>> +
>> +    ret = validate_inputs(ctx, DRM_RAS_CMD_GET_ERROR_COUNTER);
>> +    if (ret < 0)
>> +        return ret;
>> +
>>       ctx->error_value = UINT32_MAX;
>>         ret = send_command(ctx, DRM_RAS_CMD_GET_ERROR_COUNTER);
>> @@ -361,3 +380,30 @@ int igt_drm_ras_nl_get_error_counter(struct 
>> igt_drm_ras_nl_context *ctx)
>>         return 0;
>>   }
>> +
>> +int igt_drm_ras_nl_get_error_threshold(struct igt_drm_ras_nl_context 
>> *ctx)
>> +{
>> +    int ret;
>> +
>> +    ret = validate_inputs(ctx, DRM_RAS_CMD_GET_ERROR_THRESHOLD);
>> +    if (ret < 0)
>> +        return ret;
>> +
>> +    ctx->error_threshold = UINT32_MAX;
> Could be initialized to 0 as Valid threshold are as 1..16

Agree. Although 0 is currently outside the valid threshold range, 
initializing error_threshold to UINT32_MAX before fetching makes it easy 
to distinguish a received threshold from a missing response. This 
remains reliable even if the valid range changes in the future.

Thanks,

Ravi Kishore K.

>
> Apart from this all LGTM
> Reviewed-by: Soham Purkait <soham.purkait@intel.com>
>
> Regards,
> Soham
>> +
>> +    ret = send_command(ctx, DRM_RAS_CMD_GET_ERROR_THRESHOLD);
>> +    if (ret < 0)
>> +        return ret;
>> +
>> +    if (ctx->error_threshold == UINT32_MAX) {
>> +        igt_warn("No valid error threshold 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 threshold: node_id=%u error_id=%u 
>> threshold=%u\n",
>> +          ctx->node_id, ctx->error_id, ctx->error_threshold);
>> +
>> +    return 0;
>> +}
>> diff --git a/lib/igt_drm_ras_nl.h b/lib/igt_drm_ras_nl.h
>> index f51784524..86cfd79d0 100644
>> --- a/lib/igt_drm_ras_nl.h
>> +++ b/lib/igt_drm_ras_nl.h
>> @@ -44,11 +44,15 @@ struct igt_drm_ras_nl_context {
>>         /** @family_id: Generic netlink family ID */
>>       int family_id;
>> +
>> +    /** @error_threshold: Error threshold value to query/set */
>> +    uint32_t error_threshold;
>>   };
>>     void igt_drm_ras_nl_cleanup(struct igt_drm_ras_nl_context *ctx);
>>   int igt_drm_ras_nl_init(struct igt_drm_ras_nl_context *ctx);
>>   int igt_drm_ras_nl_get_error_counter(struct igt_drm_ras_nl_context 
>> *ctx);
>> +int igt_drm_ras_nl_get_error_threshold(struct igt_drm_ras_nl_context 
>> *ctx);
>>     #endif /* IGT_DRM_RAS_NL_H */