From: Jan Kara <jack@suse.cz>
To: Chengguang Xu <cgxu519@mykernel.net>
Cc: miklos@szeredi.hu, jack@suse.cz, amir73il@gmail.com,
linux-unionfs@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-kernel@vger.kernel.org,
Chengguang Xu <charliecgxu@tencent.com>
Subject: Re: [RFC PATCH V6 4/7] ovl: set 'DONTCACHE' flag for overlayfs inode
Date: Fri, 26 Nov 2021 10:20:23 +0100 [thread overview]
Message-ID: <20211126092023.GD13004@quack2.suse.cz> (raw)
In-Reply-To: <20211122030038.1938875-5-cgxu519@mykernel.net>
On Mon 22-11-21 11:00:35, Chengguang Xu wrote:
> From: Chengguang Xu <charliecgxu@tencent.com>
>
> Set 'DONTCACHE' flag to overlayfs inode so that
> upper inode to be always synced before eviction.
>
> Signed-off-by: Chengguang Xu <charliecgxu@tencent.com>
> ---
> fs/overlayfs/inode.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/fs/overlayfs/inode.c b/fs/overlayfs/inode.c
> index 027ffc0a2539..c4472299d5df 100644
> --- a/fs/overlayfs/inode.c
> +++ b/fs/overlayfs/inode.c
> @@ -791,6 +791,7 @@ void ovl_inode_init(struct inode *inode, struct ovl_inode_params *oip,
> ovl_copyattr(realinode, inode);
> ovl_copyflags(realinode, inode);
> ovl_map_ino(inode, ino, fsid);
> + d_mark_dontcache(inode);
> }
Doesn't this effectively disable dcache for overlayfs dentries? I mean e.g.
whenever overlayfs file is closed, we will drop its dentry & inode from the
cache. Upper and lower inodes / dentries stay in cache so no disk access
should be needed to reconstruct overlayfs dentry & inode but still it may
be a bit costly? I guess others more familiar with overlayfs have to judge.
Honza
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
next prev parent reply other threads:[~2021-11-26 9:22 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-22 3:00 [RFC PATCH V6 0/7] implement containerized syncfs for overlayfs Chengguang Xu
2021-11-22 3:00 ` [RFC PATCH V6 1/7] ovl: setup overlayfs' private bdi Chengguang Xu
2021-11-26 8:51 ` Jan Kara
2021-11-22 3:00 ` [RFC PATCH V6 2/7] ovl: mark overlayfs inode dirty when it has upper Chengguang Xu
2021-11-26 9:10 ` Jan Kara
2021-11-26 13:06 ` Chengguang Xu
2021-11-26 14:32 ` Jan Kara
2021-11-22 3:00 ` [RFC PATCH V6 3/7] ovl: implement overlayfs' own ->write_inode operation Chengguang Xu
2021-11-26 9:14 ` Jan Kara
2021-11-26 13:09 ` Chengguang Xu
2021-11-22 3:00 ` [RFC PATCH V6 4/7] ovl: set 'DONTCACHE' flag for overlayfs inode Chengguang Xu
2021-11-26 9:20 ` Jan Kara [this message]
2021-11-22 3:00 ` [RFC PATCH V6 5/7] fs: export wait_sb_inodes() Chengguang Xu
2021-11-26 9:20 ` Jan Kara
2021-11-22 3:00 ` [RFC PATCH V6 6/7] ovl: introduce ovl_sync_upper_blockdev() Chengguang Xu
2021-11-26 9:21 ` Jan Kara
2021-11-22 3:00 ` [RFC PATCH V6 7/7] ovl: implement containerized syncfs for overlayfs Chengguang Xu
2021-11-22 7:40 ` Amir Goldstein
2021-11-26 5:03 ` Chengguang Xu
2021-11-26 9:25 ` Jan Kara
2021-11-27 9:26 ` 回复:[RFC PATCH V6 0/7] " Chengguang Xu
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=20211126092023.GD13004@quack2.suse.cz \
--to=jack@suse.cz \
--cc=amir73il@gmail.com \
--cc=cgxu519@mykernel.net \
--cc=charliecgxu@tencent.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-unionfs@vger.kernel.org \
--cc=miklos@szeredi.hu \
/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).