linux-doc.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Yu Kuai <yukuai@kernel.org>
To: Li Nan <linan666@huaweicloud.com>,
	Yu Kuai <yukuai1@huaweicloud.com>,
	corbet@lwn.net, agk@redhat.com, snitzer@kernel.org,
	mpatocka@redhat.com, song@kernel.org, yukuai3@huawei.com,
	hare@suse.de
Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
	dm-devel@lists.linux.dev, linux-raid@vger.kernel.org,
	yi.zhang@huawei.com, yangerkun@huawei.com,
	johnny.chenyi@huawei.com
Subject: Re: [PATCH v3 06/11] md/md-bitmap: delay registration of bitmap_ops until creating bitmap
Date: Sat, 19 Jul 2025 17:45:06 +0800	[thread overview]
Message-ID: <a41300dd-22af-4619-9b68-a5cd6f8aa259@kernel.org> (raw)
In-Reply-To: <d1ee0f33-40e9-5586-36ec-88192747998f@huaweicloud.com>

Hi,

在 2025/7/19 14:49, Li Nan 写道:
>
>
> 在 2025/7/18 17:23, Yu Kuai 写道:
>> From: Yu Kuai <yukuai3@huawei.com>
>>
>> Currently bitmap_ops is registered while allocating mddev, this is fine
>> when there is only one bitmap_ops.
>>
>> Delay setting bitmap_ops until creating bitmap, so that user can choose
>> which bitmap to use before running the array.
>>
>> Signed-off-by: Yu Kuai <yukuai3@huawei.com>
>> ---
>>   Documentation/admin-guide/md.rst |  3 ++
>>   drivers/md/md.c                  | 82 +++++++++++++++++++++-----------
>>   2 files changed, 56 insertions(+), 29 deletions(-)
>>
>> diff --git a/Documentation/admin-guide/md.rst 
>> b/Documentation/admin-guide/md.rst
>> index 356d2a344f08..03a9f5025f99 100644
>> --- a/Documentation/admin-guide/md.rst
>> +++ b/Documentation/admin-guide/md.rst
>> @@ -388,6 +388,9 @@ All md devices contain:
>>        bitmap
>>            The default internal bitmap
>>   +If bitmap_type is not none, then additional bitmap attributes will be
>> +created after md device KOBJ_CHANGE event.
>> +
>>   If bitmap_type is bitmap, then the md device will also contain:
>>       bitmap/location
>> diff --git a/drivers/md/md.c b/drivers/md/md.c
>> index d8b0dfdb4bfc..639b0143cbb1 100644
>> --- a/drivers/md/md.c
>> +++ b/drivers/md/md.c
>> @@ -674,9 +674,11 @@ static void no_op(struct percpu_ref *r) {}
>>     static bool mddev_set_bitmap_ops(struct mddev *mddev)
>>   {
>> +    struct bitmap_operations *old = mddev->bitmap_ops;
>>       struct md_submodule_head *head;
>>   -    if (mddev->bitmap_id == ID_BITMAP_NONE)
>> +    if (mddev->bitmap_id == ID_BITMAP_NONE ||
>> +        (old && old->head.id == mddev->bitmap_id))
>>           return true;
>>         xa_lock(&md_submodule);
>> @@ -694,6 +696,18 @@ static bool mddev_set_bitmap_ops(struct mddev 
>> *mddev)
>>         mddev->bitmap_ops = (void *)head;
>>       xa_unlock(&md_submodule);
>> +
>> +    if (mddev->bitmap_ops->group) {
>> +        if (sysfs_create_group(&mddev->kobj, mddev->bitmap_ops->group))
>> +            pr_warn("md: cannot register extra bitmap attributes for 
>> %s\n",
>> +                mdname(mddev));
>> +        else
>> +            /*
>> +             * Inform user with KOBJ_CHANGE about new bitmap
>> +             * attributes.
>> +             */
>> +            kobject_uevent(&mddev->kobj, KOBJ_CHANGE);
>> +    }
>>       return true;
>>     err:
>> @@ -703,28 +717,25 @@ static bool mddev_set_bitmap_ops(struct mddev 
>> *mddev)
>>     static void mddev_clear_bitmap_ops(struct mddev *mddev)
>>   {
>> +    if (mddev->bitmap_ops && mddev->bitmap_ops->group)
>> +        sysfs_remove_group(&mddev->kobj, mddev->bitmap_ops->group);
>> +
>>       mddev->bitmap_ops = NULL;
>>   }
>>     int mddev_init(struct mddev *mddev)
>>   {
>> -    if (!IS_ENABLED(CONFIG_MD_BITMAP)) {
>> +    if (!IS_ENABLED(CONFIG_MD_BITMAP))
>>           mddev->bitmap_id = ID_BITMAP_NONE;
>> -    } else {
>> +    else
>>           mddev->bitmap_id = ID_BITMAP;
>
> 'bitmap_id' is set here.
>
>> -        if (!mddev_set_bitmap_ops(mddev))
>> -            return -EINVAL;
>> -    }
>>         if (percpu_ref_init(&mddev->active_io, active_io_release,
>> -                PERCPU_REF_ALLOW_REINIT, GFP_KERNEL)) {
>> -        mddev_clear_bitmap_ops(mddev);
>> +                PERCPU_REF_ALLOW_REINIT, GFP_KERNEL))
>>           return -ENOMEM;
>> -    }
>>         if (percpu_ref_init(&mddev->writes_pending, no_op,
>>                   PERCPU_REF_ALLOW_REINIT, GFP_KERNEL)) {
>> -        mddev_clear_bitmap_ops(mddev);
>>           percpu_ref_exit(&mddev->active_io);
>>           return -ENOMEM;
>>       }
>> @@ -752,6 +763,7 @@ int mddev_init(struct mddev *mddev)
>>       mddev->resync_min = 0;
>>       mddev->resync_max = MaxSector;
>>       mddev->level = LEVEL_NONE;
>> +    mddev->bitmap_id = ID_BITMAP;
>
> This change is wrong.

Yes, this line from v1 should be removed. Will change this in v3.

Thanks,
Kuai



  reply	other threads:[~2025-07-19  9:45 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-18  9:23 [PATCH v3 00/11] md/llbitmap: md/md-llbitmap: introduce a new lockless bitmap Yu Kuai
2025-07-18  9:23 ` [PATCH v3 01/11] md: add a new parameter 'offset' to md_super_write() Yu Kuai
2025-07-18  9:23 ` [PATCH v3 02/11] md: factor out a helper raid_is_456() Yu Kuai
2025-07-18  9:23 ` [PATCH v3 03/11] md/md-bitmap: support discard for bitmap ops Yu Kuai
2025-07-18  9:23 ` [PATCH v3 04/11] md: add a new mddev field 'bitmap_id' Yu Kuai
2025-07-18  9:23 ` [PATCH v3 05/11] md/md-bitmap: add a new sysfs api bitmap_type Yu Kuai
2025-07-18  9:23 ` [PATCH v3 06/11] md/md-bitmap: delay registration of bitmap_ops until creating bitmap Yu Kuai
2025-07-19  6:49   ` Li Nan
2025-07-19  9:45     ` Yu Kuai [this message]
2025-07-21  6:01   ` Hannes Reinecke
2025-07-21  6:10     ` Yu Kuai
2025-07-18  9:23 ` [PATCH v3 07/11] md/md-bitmap: add a new method skip_sync_blocks() in bitmap_operations Yu Kuai
2025-07-18  9:23 ` [PATCH v3 08/11] md/md-bitmap: add a new method blocks_synced() " Yu Kuai
2025-07-18  9:23 ` [PATCH v3 09/11] md: add a new recovery_flag MD_RECOVERY_LAZY_RECOVER Yu Kuai
2025-07-18  9:23 ` [PATCH v3 10/11] md/md-bitmap: make method bitmap_ops->daemon_work optional Yu Kuai
2025-07-18  9:23 ` [PATCH v3 11/11] md/md-llbitmap: introduce new lockless bitmap Yu Kuai
2025-07-21  6:20   ` Hannes Reinecke
2025-07-21  7:12     ` Yu Kuai
2025-07-21  7:20       ` Hannes Reinecke

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=a41300dd-22af-4619-9b68-a5cd6f8aa259@kernel.org \
    --to=yukuai@kernel.org \
    --cc=agk@redhat.com \
    --cc=corbet@lwn.net \
    --cc=dm-devel@lists.linux.dev \
    --cc=hare@suse.de \
    --cc=johnny.chenyi@huawei.com \
    --cc=linan666@huaweicloud.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-raid@vger.kernel.org \
    --cc=mpatocka@redhat.com \
    --cc=snitzer@kernel.org \
    --cc=song@kernel.org \
    --cc=yangerkun@huawei.com \
    --cc=yi.zhang@huawei.com \
    --cc=yukuai1@huaweicloud.com \
    --cc=yukuai3@huawei.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).