linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Damien Le Moal <damien.lemoal@opensource.wdc.com>
To: Johannes Thumshirn <Johannes.Thumshirn@wdc.com>,
	"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>
Subject: Re: [PATCH 6/8] zonefs: Add active seq file accounting
Date: Tue, 19 Apr 2022 20:26:41 +0900	[thread overview]
Message-ID: <38720285-3f18-dcee-fcee-f2c223b1d52a@opensource.wdc.com> (raw)
In-Reply-To: <PH0PR04MB741645F790D210401E2CD2EB9BF29@PH0PR04MB7416.namprd04.prod.outlook.com>

On 4/19/22 20:06, Johannes Thumshirn wrote:
> On 19/04/2022 13:03, Damien Le Moal wrote:
>> On 4/19/22 19:59, Johannes Thumshirn wrote:
>>> On 18/04/2022 03:12, Damien Le Moal wrote:
>>>> +/*
>>>> + * Manage the active zone count. Called with zi->i_truncate_mutex held.
>>>> + */
>>>> +static void zonefs_account_active(struct inode *inode)
>>>> +{
>>>> +	struct zonefs_sb_info *sbi = ZONEFS_SB(inode->i_sb);
>>>> +	struct zonefs_inode_info *zi = ZONEFS_I(inode);
>>>> +
>>>
>>> Nit:	lockdep_assert_held(&zi->i_truncate_mutex);
>>
>> If I add that, lockdep screams during mount as the inodes mutex is not
>> held when the zone inodes are initialized and zonefs_account_active()
>> called. We could add a wrapper function for this, but I did not feel it
>> was necessary.
> 
> OK, but then the 'Called with zi->i_truncate_mutex held.' comment is invalid
> and should be removed.

Not invalid. When mounting, nobody but the mount process can touch inodes,
so not holding the mutex is OK.

Will see how to clean this up.


-- 
Damien Le Moal
Western Digital Research

  reply	other threads:[~2022-04-19 11:26 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-18  1:11 [PATCH 0/8] zonefs improvements Damien Le Moal
2022-04-18  1:12 ` [PATCH 1/8] zonefs: Clear inode information flags on inode creation Damien Le Moal
2022-04-19  9:27   ` Johannes Thumshirn
2022-04-18  1:12 ` [PATCH 2/8] zonefs: Fix management of open zones Damien Le Moal
2022-04-19  9:38   ` Johannes Thumshirn
2022-04-18  1:12 ` [PATCH 3/8] zonefs: Rename super block information fields Damien Le Moal
2022-04-19  9:29   ` Johannes Thumshirn
2022-04-18  1:12 ` [PATCH 4/8] zonefs: Always do seq file write open accounting Damien Le Moal
2022-04-19  9:55   ` Johannes Thumshirn
2022-04-18  1:12 ` [PATCH 5/8] zonefs: Export open zone resource information through sysfs Damien Le Moal
2022-04-19 10:06   ` Johannes Thumshirn
2022-04-18  1:12 ` [PATCH 6/8] zonefs: Add active seq file accounting Damien Le Moal
2022-04-19 10:59   ` Johannes Thumshirn
2022-04-19 11:02     ` Damien Le Moal
2022-04-19 11:06       ` Johannes Thumshirn
2022-04-19 11:26         ` Damien Le Moal [this message]
2022-04-18  1:12 ` [PATCH 7/8] documentation: zonefs: Cleanup the mount options section Damien Le Moal
2022-04-19 10:14   ` Johannes Thumshirn
2022-04-18  1:12 ` [PATCH 8/8] documentation: zonefs: Document sysfs attributes Damien Le Moal
2022-04-19 10:14   ` Johannes Thumshirn

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=38720285-3f18-dcee-fcee-f2c223b1d52a@opensource.wdc.com \
    --to=damien.lemoal@opensource.wdc.com \
    --cc=Johannes.Thumshirn@wdc.com \
    --cc=linux-fsdevel@vger.kernel.org \
    /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).