Re: [PATCH v2] block: don't show io_timeout if driver has no timeout handler

[Bart Van Assche <bvanassche@acm.org>]

On Tue, 2019-04-02 at 21:14 +0800, Weiping Zhang wrote:
> [ ... ]
> -static struct attribute *default_attrs[] = {
> +static struct attribute *queue_attrs[] = {
>  	&queue_requests_entry.attr,
>  	&queue_ra_entry.attr,
>  	&queue_max_hw_sectors_entry.attr,
> @@ -770,6 +770,25 @@ static struct attribute *default_attrs[] = {
>  	NULL,
>  };
> 
> [ ... ]
>
>  static ssize_t
> @@ -890,7 +909,6 @@ static const struct sysfs_ops queue_sysfs_ops = {
>  
>  struct kobj_type blk_queue_ktype = {
>  	.sysfs_ops	= &queue_sysfs_ops,
> -	.default_attrs	= default_attrs,
>  	.release	= blk_release_queue,
>  };
>  
> @@ -939,6 +957,14 @@ int blk_register_queue(struct gendisk *disk)
>  		goto unlock;
>  	}
>  
> +	ret = sysfs_create_group(&q->kobj, &queue_attr_group);
> +	if (ret) {
> +		blk_trace_remove_sysfs(dev);
> +		kobject_del(&q->kobj);
> +		kobject_put(&dev->kobj);
> +		goto unlock;
> +	}
> +
>  	if (queue_is_mq(q)) {
>  		__blk_mq_register_dev(dev, q);
>  		blk_mq_debugfs_register(q);

This kind of change introduces an unacceptable race against udev. I think we
should wait with making this change until the patch series "kobject: Add
default group support to kobj_type and replace subsystem uses" is upstream.

Bart.