linux-kernel - Re: [PATCH v2] block: ublk: enable zoned storage support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <87wn4386dw.fsf@metaspace.dk>
Date:   Mon, 27 Feb 2023 16:36:15 +0100
From:   Andreas Hindborg <nmi@...aspace.dk>
To:     Minwoo Im <minwoo.im.dev@...il.com>
Cc:     linux-block@...r.kernel.org, Hans Holmberg <Hans.Holmberg@....com>,
        Matias Bjorling <Matias.Bjorling@....com>,
        Niklas Cassel <Niklas.Cassel@....com>,
        kernel test robot <lkp@...el.com>,
        Ming Lei <ming.lei@...hat.com>, Jens Axboe <axboe@...nel.dk>,
        open list <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] block: ublk: enable zoned storage support


Minwoo Im <minwoo.im.dev@...il.com> writes:

> On 23-02-24 21:05:01, Andreas Hindborg wrote:
>> +static int ublk_report_zones(struct gendisk *disk, sector_t sector,
>> +			     unsigned int nr_zones, report_zones_cb cb,
>> +			     void *data)
>> +{
>> +	struct ublk_device *ub;
>> +	unsigned int zone_size;
>> +	unsigned int first_zone;
>> +	int ret = 0;
>> +
>> +	ub = disk->private_data;
>> +
>> +	if (!(ub->dev_info.flags & UBLK_F_ZONED))
>> +		return -EINVAL;
>> +
>> +	zone_size = disk->queue->limits.chunk_sectors;
>> +	first_zone = sector >> ilog2(zone_size);
>> +	nr_zones = min(ub->ub_disk->nr_zones - first_zone, nr_zones);
>> +
>> +	for (unsigned int i = 0; i < nr_zones; i++) {
>> +		struct request *req;
>> +		blk_status_t status;
>> +		struct blk_zone info = {0};
>> +
>> +		req = blk_mq_alloc_request(disk->queue, REQ_OP_DRV_IN, 0);
>> +
>> +		if (IS_ERR(req)) {
>> +			ret = PTR_ERR(req);
>> +			goto out;
>> +		}
>
> Can we just return directly just like above (-EINVAL)?
>
> if (IS_ERR(req))
> 	return PTR_ERR(req);

Yes, I agree.

>
>> +
>> +		req->__sector = sector;
>> +
>> +		ret = blk_rq_map_kern(disk->queue, req, &info, sizeof(info),
>> +				      GFP_KERNEL);
>> +
>> +		if (ret)
>> +			goto out;
>
> If we really have to use goto here, then I think we have choices:
> put blk_mq_free_request(req); here or put it to the out: area.

Lets free here and drop the goto.

Thanks,
Andreas

>> +
>> +		status = blk_execute_rq(req, 0);
>> +		ret = blk_status_to_errno(status);
>> +		if (ret)
>> +			goto out;
>> +
>> +		blk_mq_free_request(req);
>> +
>> +		ret = cb(&info, i, data);
>> +		if (ret)
>> +			goto out;
>> +
>> +		/* A zero length zone means don't ask for more zones */
>> +		if (!info.len) {
>> +			nr_zones = i;
>> +			break;
>> +		}
>> +
>> +		sector += zone_size;
>> +	}
>> +	ret = nr_zones;
>> +
>> + out:
>> +	return ret;
>> +}