From: Bart Van Assche <bvanassche@acm.org>
To: Damien Le Moal <dlemoal@kernel.org>,
linux-block@vger.kernel.org, Jens Axboe <axboe@kernel.dk>,
linux-scsi@vger.kernel.org,
"Martin K . Petersen" <martin.petersen@oracle.com>,
dm-devel@lists.linux.dev, Mike Snitzer <snitzer@redhat.com>,
linux-nvme@lists.infradead.org, Keith Busch <kbusch@kernel.org>,
Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH v3 08/30] block: Introduce zone write plugging
Date: Thu, 28 Mar 2024 15:20:14 -0700 [thread overview]
Message-ID: <688607d9-6dfe-4c37-81e8-4fea91ec8da8@acm.org> (raw)
In-Reply-To: <20240328004409.594888-9-dlemoal@kernel.org>
On 3/27/24 5:43 PM, Damien Le Moal wrote:
> + /*
> + * Initialize the zoen write plug with an extra reference so that
> + * it is not freed when the zone write plug becomes idle without
> + * the zone being full.
> + */
zoen -> zone
> +static void disk_zone_wplugs_work(struct work_struct *work)
> +{
> + struct gendisk *disk =
> + container_of(work, struct gendisk, zone_wplugs_work);
> + struct blk_zone_wplug *zwplug;
> + unsigned long flags;
> +
> + spin_lock_irqsave(&disk->zone_wplugs_lock, flags);
> +
> + while (!list_empty(&disk->zone_wplugs_err_list)) {
> + zwplug = list_first_entry(&disk->zone_wplugs_err_list,
> + struct blk_zone_wplug, link);
> + list_del_init(&zwplug->link);
> + blk_get_zone_wplug(zwplug);
> + spin_unlock_irqrestore(&disk->zone_wplugs_lock, flags);
> +
> + disk_zone_wplug_handle_error(disk, zwplug);
> + disk_put_zone_wplug(zwplug);
> +
> + spin_lock_irqsave(&disk->zone_wplugs_lock, flags);
> + }
> +
> + spin_unlock_irqrestore(&disk->zone_wplugs_lock, flags);
> +}
What is the maximum number of iterations the above while loop can
perform? I'm wondering whether a cond_resched() call should be added.
> +
> + /* Wait for the zone write plugs to be RCU-freed. */
> + rcu_barrier();
> +}
It is not clear to me why the above rcu_barrier() call is necessary? I'm
not aware of any other kernel code where kfree_rcu() is followed by an
rcu_barrier() call.
> +static int disk_alloc_zone_resources(struct gendisk *disk,
> + unsigned int max_nr_zwplugs)
> +{
> + unsigned int i;
> +
> + disk->zone_wplugs_hash_bits =
> + min(ilog2(max_nr_zwplugs) + 1, BLK_ZONE_MAX_WPLUG_HASH_BITS);
If max_nr_zwplugs is a power of two, the above formula will result in a
hash table with a size that is twice the size of max_nr_zwplugs.
Shouldn't ilog2(max_nr_zwplugs) + 1 be changed into
ilog2(roundup_pow_of_two(max_nr_zwplugs))?
Otherwise this patch looks fine to me.
Thanks,
Bart.
next prev parent reply other threads:[~2024-03-28 22:20 UTC|newest]
Thread overview: 109+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-03-28 0:43 [PATCH v3 00/30] Zone write plugging Damien Le Moal
2024-03-28 0:43 ` [PATCH v3 01/30] block: Do not force full zone append completion in req_bio_endio() Damien Le Moal
2024-03-28 4:10 ` Christoph Hellwig
2024-03-28 18:14 ` Bart Van Assche
2024-03-28 22:43 ` Damien Le Moal
2024-03-28 23:03 ` Jens Axboe
2024-03-28 0:43 ` [PATCH v3 02/30] block: Restore sector of flush requests Damien Le Moal
2024-03-28 0:43 ` [PATCH v3 03/30] block: Remove req_bio_endio() Damien Le Moal
2024-03-28 4:13 ` Christoph Hellwig
2024-03-28 21:28 ` Bart Van Assche
2024-03-28 22:42 ` Damien Le Moal
2024-03-28 0:43 ` [PATCH v3 04/30] block: Introduce blk_zone_update_request_bio() Damien Le Moal
2024-03-28 4:14 ` Christoph Hellwig
2024-03-28 5:20 ` Damien Le Moal
2024-03-28 5:42 ` Christoph Hellwig
2024-03-28 5:54 ` Damien Le Moal
2024-03-28 21:31 ` Bart Van Assche
2024-03-28 0:43 ` [PATCH v3 05/30] block: Introduce bio_straddles_zones() and bio_offset_from_zone_start() Damien Le Moal
2024-03-28 21:32 ` Bart Van Assche
2024-03-28 0:43 ` [PATCH v3 06/30] block: Allow using bio_attempt_back_merge() internally Damien Le Moal
2024-03-28 0:43 ` [PATCH v3 07/30] block: Remember zone capacity when revalidating zones Damien Le Moal
2024-03-28 21:38 ` Bart Van Assche
2024-03-28 22:40 ` Damien Le Moal
2024-03-28 0:43 ` [PATCH v3 08/30] block: Introduce zone write plugging Damien Le Moal
2024-03-28 4:48 ` Christoph Hellwig
2024-03-28 22:20 ` Bart Van Assche [this message]
2024-03-28 22:38 ` Damien Le Moal
2024-03-29 18:20 ` Bart Van Assche
2024-03-28 0:43 ` [PATCH v3 09/30] block: Pre-allocate zone write plugs Damien Le Moal
2024-03-28 4:30 ` Christoph Hellwig
2024-03-28 5:28 ` Damien Le Moal
2024-03-28 5:46 ` Christoph Hellwig
2024-03-28 6:02 ` Damien Le Moal
2024-03-28 6:03 ` Christoph Hellwig
2024-03-28 6:18 ` Damien Le Moal
2024-03-28 6:22 ` Christoph Hellwig
2024-03-28 6:33 ` Damien Le Moal
2024-03-28 6:38 ` Christoph Hellwig
2024-03-28 6:51 ` Damien Le Moal
2024-03-28 6:52 ` Christoph Hellwig
2024-03-28 6:53 ` Damien Le Moal
2024-03-28 22:25 ` Bart Van Assche
2024-03-28 22:29 ` Bart Van Assche
2024-03-28 22:33 ` Damien Le Moal
2024-03-28 0:43 ` [PATCH v3 10/30] block: Fake max open zones limit when there is no limit Damien Le Moal
2024-03-28 4:49 ` Christoph Hellwig
2024-03-29 20:37 ` Bart Van Assche
2024-03-28 0:43 ` [PATCH v3 11/30] block: Allow zero value of max_zone_append_sectors queue limit Damien Le Moal
2024-03-28 4:49 ` Christoph Hellwig
2024-03-29 20:50 ` Bart Van Assche
2024-03-28 0:43 ` [PATCH v3 12/30] block: Implement zone append emulation Damien Le Moal
2024-03-28 4:50 ` Christoph Hellwig
2024-03-29 21:22 ` Bart Van Assche
2024-03-29 21:26 ` Bart Van Assche
2024-03-28 0:43 ` [PATCH v3 13/30] block: Allow BIO-based drivers to use blk_revalidate_disk_zones() Damien Le Moal
2024-03-28 0:43 ` [PATCH v3 14/30] dm: Use the block layer zone append emulation Damien Le Moal
2024-03-28 0:43 ` [PATCH v3 15/30] scsi: sd: " Damien Le Moal
2024-03-28 4:50 ` Christoph Hellwig
2024-03-28 10:49 ` Johannes Thumshirn
2024-03-29 21:27 ` Bart Van Assche
2024-03-28 0:43 ` [PATCH v3 16/30] ublk_drv: Do not request ELEVATOR_F_ZBD_SEQ_WRITE elevator feature Damien Le Moal
2024-03-28 4:50 ` Christoph Hellwig
2024-03-29 21:28 ` Bart Van Assche
2024-03-28 0:43 ` [PATCH v3 17/30] null_blk: " Damien Le Moal
2024-03-28 4:51 ` Christoph Hellwig
2024-03-29 21:29 ` Bart Van Assche
2024-04-02 6:43 ` Chaitanya Kulkarni
2024-03-28 0:43 ` [PATCH v3 18/30] null_blk: Introduce zone_append_max_sectors attribute Damien Le Moal
2024-03-28 4:51 ` Christoph Hellwig
2024-03-29 21:35 ` Bart Van Assche
2024-03-30 0:33 ` Damien Le Moal
2024-04-02 6:44 ` Chaitanya Kulkarni
2024-03-28 0:43 ` [PATCH v3 19/30] null_blk: Introduce fua attribute Damien Le Moal
2024-03-28 4:52 ` Christoph Hellwig
2024-03-29 21:36 ` Bart Van Assche
2024-04-02 6:42 ` Chaitanya Kulkarni
2024-03-28 0:43 ` [PATCH v3 20/30] nvmet: zns: Do not reference the gendisk conv_zones_bitmap Damien Le Moal
2024-04-02 6:45 ` Chaitanya Kulkarni
2024-03-28 0:44 ` [PATCH v3 21/30] block: Remove BLK_STS_ZONE_RESOURCE Damien Le Moal
2024-03-29 21:37 ` Bart Van Assche
2024-03-28 0:44 ` [PATCH v3 22/30] block: Simplify blk_revalidate_disk_zones() interface Damien Le Moal
2024-03-29 21:41 ` Bart Van Assche
2024-03-28 0:44 ` [PATCH v3 23/30] block: mq-deadline: Remove support for zone write locking Damien Le Moal
2024-03-28 4:52 ` Christoph Hellwig
2024-03-29 21:43 ` Bart Van Assche
2024-03-28 0:44 ` [PATCH v3 24/30] block: Remove elevator required features Damien Le Moal
2024-03-29 21:44 ` Bart Van Assche
2024-03-28 0:44 ` [PATCH v3 25/30] block: Do not check zone type in blk_check_zone_append() Damien Le Moal
2024-03-29 21:45 ` Bart Van Assche
2024-03-28 0:44 ` [PATCH v3 26/30] block: Move zone related debugfs attribute to blk-zoned.c Damien Le Moal
2024-03-28 4:52 ` Christoph Hellwig
2024-03-29 19:00 ` Bart Van Assche
2024-03-28 0:44 ` [PATCH v3 27/30] block: Replace zone_wlock debugfs entry with zone_wplugs entry Damien Le Moal
2024-03-28 4:53 ` Christoph Hellwig
2024-03-29 18:54 ` Bart Van Assche
2024-03-28 0:44 ` [PATCH v3 28/30] block: Remove zone write locking Damien Le Moal
2024-03-29 18:57 ` Bart Van Assche
2024-03-28 0:44 ` [PATCH v3 29/30] block: Do not force select mq-deadline with CONFIG_BLK_DEV_ZONED Damien Le Moal
2024-03-28 4:53 ` Christoph Hellwig
2024-03-28 0:44 ` [PATCH v3 30/30] block: Do not special-case plugging of zone write operations Damien Le Moal
2024-03-28 4:54 ` Christoph Hellwig
2024-03-28 6:43 ` Damien Le Moal
2024-03-28 6:51 ` Christoph Hellwig
2024-03-28 6:54 ` Damien Le Moal
2024-03-29 18:58 ` Bart Van Assche
2024-03-28 23:05 ` (subset) [PATCH v3 00/30] Zone write plugging Jens Axboe
2024-03-28 23:13 ` Damien Le Moal
2024-03-28 23:27 ` Jens Axboe
2024-03-28 23:33 ` 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=688607d9-6dfe-4c37-81e8-4fea91ec8da8@acm.org \
--to=bvanassche@acm.org \
--cc=axboe@kernel.dk \
--cc=dlemoal@kernel.org \
--cc=dm-devel@lists.linux.dev \
--cc=hch@lst.de \
--cc=kbusch@kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=snitzer@redhat.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 read-only IMAP folder(s) and NNTP newsgroup(s).