Re: [RFC PATCH] block: Make request_queue lockdep splats show up earlier

[Bart Van Assche <bvanassche@xxxxxxx>]

On 3/17/25 10:11 AM, Thomas Hellström wrote:
> diff --git a/block/blk-core.c b/block/blk-core.c
> index d6c4fa3943b5..4aa439309406 100644
> --- a/block/blk-core.c
> +++ b/block/blk-core.c
> @@ -455,6 +455,12 @@ struct request_queue *blk_alloc_queue(struct queue_limits *lim, int node_id)
>   	lockdep_init_map(&q->q_lockdep_map, "&q->q_usage_counter(queue)",
>   			 &q->q_lock_cls_key, 0);
>   
> +	/* Prime io_lockdep_map as reclaim tainted */
> +	fs_reclaim_acquire(GFP_KERNEL);
> +	rwsem_acquire_read(&q->io_lockdep_map, 0, 0, _RET_IP_);
> +	rwsem_release(&q->io_lockdep_map, _RET_IP_);
> +	fs_reclaim_release(GFP_KERNEL);
> +
>   	q->nr_requests = BLKDEV_DEFAULT_RQ;
>   
>   	return q;

Hmm ... my understanding is that it is fine if FS code calls block layer
code but also that block layer code never should call FS code.

Thanks,

Bart.