From: "Javier González" <javier@javigon.com>
To: Damien Le Moal <Damien.LeMoal@wdc.com>
Cc: "axboe@kernel.dk" <axboe@kernel.dk>,
SelvaKumar S <selvakuma.s1@samsung.com>,
"sagi@grimberg.me" <sagi@grimberg.me>,
Kanchan Joshi <joshi.k@samsung.com>,
"linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>,
Nitesh Shetty <nj.shetty@samsung.com>,
"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
"kbusch@kernel.org" <kbusch@kernel.org>,
"mb@lightnvm.io" <mb@lightnvm.io>, "hch@lst.de" <hch@lst.de>
Subject: Re: [PATCH 1/4] block: Add zone flags to queue zone prop.
Date: Thu, 2 Jul 2020 10:34:30 +0200 [thread overview]
Message-ID: <20200702083430.miax2cd44mkhc5fb@mpHalley.local> (raw)
In-Reply-To: <CY4PR04MB37511008EEBF1A77DB4F423BE76D0@CY4PR04MB3751.namprd04.prod.outlook.com>
On 02.07.2020 07:54, Damien Le Moal wrote:
>On 2020/07/02 15:55, Javier González wrote:
>> From: Javier González <javier.gonz@samsung.com>
>>
>> As the zoned block device will have to deal with features that are
>> optional for the backend device, add a flag field to inform the block
>> layer about supported features. This builds on top of
>> blk_zone_report_flags and extendes to the zone report code.
>>
>> Signed-off-by: Javier González <javier.gonz@samsung.com>
>> Signed-off-by: SelvaKumar S <selvakuma.s1@samsung.com>
>> Signed-off-by: Kanchan Joshi <joshi.k@samsung.com>
>> Signed-off-by: Nitesh Shetty <nj.shetty@samsung.com>
>> ---
>> block/blk-zoned.c | 3 ++-
>> drivers/block/null_blk_zoned.c | 2 ++
>> drivers/nvme/host/zns.c | 1 +
>> drivers/scsi/sd.c | 2 ++
>> include/linux/blkdev.h | 3 +++
>> 5 files changed, 10 insertions(+), 1 deletion(-)
>>
>> diff --git a/block/blk-zoned.c b/block/blk-zoned.c
>> index 81152a260354..0f156e96e48f 100644
>> --- a/block/blk-zoned.c
>> +++ b/block/blk-zoned.c
>> @@ -312,7 +312,8 @@ int blkdev_report_zones_ioctl(struct block_device *bdev, fmode_t mode,
>> return ret;
>>
>> rep.nr_zones = ret;
>> - rep.flags = BLK_ZONE_REP_CAPACITY;
>> + rep.flags = q->zone_flags;
>
>zone_flags seem to be a fairly generic flags field while rep.flags is only about
>the reported descriptors structure. So you may want to define a
>BLK_ZONE_REP_FLAGS_MASK and have:
>
> rep.flags = q->zone_flags & BLK_ZONE_REP_FLAGS_MASK;
>
>to avoid flags meaningless for the user being set.
>
>In any case, since *all* zoned block devices now report capacity, I do not
>really see the point to add BLK_ZONE_REP_FLAGS_MASK to q->zone_flags, especially
>considering that you set the flag for all zoned device types, including scsi
>which does not have zone capacity. This makes q->zone_flags rather confusing
>instead of clearly defining the device features as you mentioned in the commit
>message.
>
>I think it may be better to just drop this patch, and if needed, introduce the
>zone_flags field where it may be needed (e.g. OFFLINE zone ioctl support).
I am using this as a way to pass the OFFLINE support flag to the block
layer. I used this too for the attributes. Are you thinking of a
different way to send this?
I believe this fits too for capacity, but we can just pass it in
all report as before if you prefer.
>
>> +
>> if (copy_to_user(argp, &rep, sizeof(struct blk_zone_report)))
>> return -EFAULT;
>> return 0;
>> diff --git a/drivers/block/null_blk_zoned.c b/drivers/block/null_blk_zoned.c
>> index b05832eb21b2..957c2103f240 100644
>> --- a/drivers/block/null_blk_zoned.c
>> +++ b/drivers/block/null_blk_zoned.c
>> @@ -78,6 +78,8 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q)
>> }
>>
>> q->limits.zoned = BLK_ZONED_HM;
>> + q->zone_flags = BLK_ZONE_REP_CAPACITY;
>> +
>> blk_queue_flag_set(QUEUE_FLAG_ZONE_RESETALL, q);
>> blk_queue_required_elevator_features(q, ELEVATOR_F_ZBD_SEQ_WRITE);
>>
>> diff --git a/drivers/nvme/host/zns.c b/drivers/nvme/host/zns.c
>> index 0642d3c54e8f..888264261ba3 100644
>> --- a/drivers/nvme/host/zns.c
>> +++ b/drivers/nvme/host/zns.c
>> @@ -81,6 +81,7 @@ int nvme_update_zone_info(struct gendisk *disk, struct nvme_ns *ns,
>> }
>>
>> q->limits.zoned = BLK_ZONED_HM;
>> + q->zone_flags = BLK_ZONE_REP_CAPACITY;
>> blk_queue_flag_set(QUEUE_FLAG_ZONE_RESETALL, q);
>> free_data:
>> kfree(id);
>> diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
>> index d90fefffe31b..b9c920bace28 100644
>> --- a/drivers/scsi/sd.c
>> +++ b/drivers/scsi/sd.c
>> @@ -2967,6 +2967,7 @@ static void sd_read_block_characteristics(struct scsi_disk *sdkp)
>> if (sdkp->device->type == TYPE_ZBC) {
>> /* Host-managed */
>> q->limits.zoned = BLK_ZONED_HM;
>> + q->zone_flags = BLK_ZONE_REP_CAPACITY;
>> } else {
>> sdkp->zoned = (buffer[8] >> 4) & 3;
>> if (sdkp->zoned == 1 && !disk_has_partitions(sdkp->disk)) {
>> @@ -2983,6 +2984,7 @@ static void sd_read_block_characteristics(struct scsi_disk *sdkp)
>> "Drive-managed SMR disk\n");
>> }
>> }
>> +
>> if (blk_queue_is_zoned(q) && sdkp->first_scan)
>> sd_printk(KERN_NOTICE, sdkp, "Host-%s zoned block device\n",
>> q->limits.zoned == BLK_ZONED_HM ? "managed" : "aware");
>> diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
>> index 8fd900998b4e..3f2e3425fa53 100644
>> --- a/include/linux/blkdev.h
>> +++ b/include/linux/blkdev.h
>> @@ -512,12 +512,15 @@ struct request_queue {
>> * Stacking drivers (device mappers) may or may not initialize
>> * these fields.
>> *
>> + * Flags represent features as described by blk_zone_report_flags in blkzoned.h
>> + *
>> * Reads of this information must be protected with blk_queue_enter() /
>> * blk_queue_exit(). Modifying this information is only allowed while
>> * no requests are being processed. See also blk_mq_freeze_queue() and
>> * blk_mq_unfreeze_queue().
>> */
>> unsigned int nr_zones;
>> + unsigned int zone_flags;
>> unsigned long *conv_zones_bitmap;
>> unsigned long *seq_zones_wlock;
>> #endif /* CONFIG_BLK_DEV_ZONED */
>>
>
>And you are missing device-mapper support. DM target devices have a request
>queue that would need to set the zone_flags too.
Ok. I looked at it and I thought that this would be inherited by the
underlying device. I will add it in V3.
Javier
_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
next prev parent reply other threads:[~2020-07-02 8:34 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-02 6:54 [V2 PATCH 0/4] ZNS: Extra features for current patche Javier González
2020-07-02 6:54 ` [PATCH 1/4] block: Add zone flags to queue zone prop Javier González
2020-07-02 7:54 ` Damien Le Moal
2020-07-02 8:34 ` Javier González [this message]
2020-07-02 8:49 ` Damien Le Moal
2020-07-02 10:27 ` Javier González
2020-07-03 5:20 ` Damien Le Moal
2020-07-03 6:28 ` Javier González
2020-07-03 8:35 ` Damien Le Moal
2020-07-02 6:54 ` [PATCH 2/4] block: add support for zone offline transition Javier González
2020-07-02 8:10 ` Damien Le Moal
2020-07-02 8:39 ` Javier González
2020-07-02 8:52 ` Damien Le Moal
2020-07-02 6:54 ` [PATCH 3/4] nvme: Add consistency check for zone count Javier González
2020-07-02 8:16 ` Damien Le Moal
2020-07-02 8:19 ` Johannes Thumshirn
2020-07-02 8:27 ` Javier González
2020-07-02 6:54 ` [PATCH 4/4] block: add attributes to zone report Javier González
2020-07-02 8:27 ` Damien Le Moal
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200702083430.miax2cd44mkhc5fb@mpHalley.local \
--to=javier@javigon.com \
--cc=Damien.LeMoal@wdc.com \
--cc=axboe@kernel.dk \
--cc=hch@lst.de \
--cc=joshi.k@samsung.com \
--cc=kbusch@kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=mb@lightnvm.io \
--cc=nj.shetty@samsung.com \
--cc=sagi@grimberg.me \
--cc=selvakuma.s1@samsung.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).