linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Dave Kleikamp <dave.kleikamp@oracle.com>
To: Oleg Kanatov <okanatov@gmail.com>
Cc: Oleg Kanatov <okanatov@astralinux.ru>,
	jfs-discussion@lists.sourceforge.net,
	linux-kernel@vger.kernel.org, lvc-project@linuxtesting.org
Subject: Re: [PATCH] jfs: makes diUnmount/diMount in jfs_mount_rw atomic
Date: Thu, 10 Nov 2022 15:47:10 -0600	[thread overview]
Message-ID: <85b67cd9-2d00-f2f2-1335-efab7240d65d@oracle.com> (raw)
In-Reply-To: <20221028122254.21081-1-okanatov@astralinux.ru>

Applied with trivial changes. I cleaned up the commit message a little 
and removed a couple blank lines.

Thanks!

On 10/28/22 7:22AM, Oleg Kanatov wrote:
> jfs_mount_rw can call diUnmount and then diMount. These calls
> change the imap pointer.
> In case JFS filesystem is mounted on root (/), between these two
> calls there may be calls of function jfs_lookup().
> The jfs_lookup() function calls jfs_iget(), which, in its turn,
> calls diRead(). The latter references the imap pointer.
> That may cause diRead() to refer to a pointer "freed" in
> diUnmount().
> This commit makes calls diUnmount()/diMount() be atomic so that
> nothing will read the imap pointer until the whole remount is
> completed.
> 
> Signed-off-by: Oleg Kanatov <okanatov@gmail.com>
> ---
>   fs/jfs/jfs_imap.c  | 2 +-
>   fs/jfs/jfs_mount.c | 6 ++++++
>   2 files changed, 7 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c
> index 799d3837e7c2..390cbfce391f 100644
> --- a/fs/jfs/jfs_imap.c
> +++ b/fs/jfs/jfs_imap.c
> @@ -310,8 +310,8 @@ int diRead(struct inode *ip)
>   	iagno = INOTOIAG(ip->i_ino);
>   
>   	/* read the iag */
> -	imap = JFS_IP(ipimap)->i_imap;
>   	IREAD_LOCK(ipimap, RDWRLOCK_IMAP);
> +	imap = JFS_IP(ipimap)->i_imap;
>   	rc = diIAGRead(imap, iagno, &mp);
>   	IREAD_UNLOCK(ipimap);
>   	if (rc) {
> diff --git a/fs/jfs/jfs_mount.c b/fs/jfs/jfs_mount.c
> index 48d1f70f786c..972b9ff18723 100644
> --- a/fs/jfs/jfs_mount.c
> +++ b/fs/jfs/jfs_mount.c
> @@ -234,12 +234,18 @@ int jfs_mount_rw(struct super_block *sb, int remount)
>   
>   		truncate_inode_pages(sbi->ipimap->i_mapping, 0);
>   		truncate_inode_pages(sbi->ipbmap->i_mapping, 0);
> +
> +		IWRITE_LOCK(sbi->ipimap, RDWRLOCK_IMAP);
> +
>   		diUnmount(sbi->ipimap, 1);
>   		if ((rc = diMount(sbi->ipimap))) {
> +			IWRITE_UNLOCK(sbi->ipimap);
>   			jfs_err("jfs_mount_rw: diMount failed!");
>   			return rc;
>   		}
>   
> +		IWRITE_UNLOCK(sbi->ipimap);
> +
>   		dbUnmount(sbi->ipbmap, 1);
>   		if ((rc = dbMount(sbi->ipbmap))) {
>   			jfs_err("jfs_mount_rw: dbMount failed!");

      reply	other threads:[~2022-11-10 21:47 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-28 12:22 [PATCH] jfs: makes diUnmount/diMount in jfs_mount_rw atomic Oleg Kanatov
2022-11-10 21:47 ` Dave Kleikamp [this message]

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=85b67cd9-2d00-f2f2-1335-efab7240d65d@oracle.com \
    --to=dave.kleikamp@oracle.com \
    --cc=jfs-discussion@lists.sourceforge.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lvc-project@linuxtesting.org \
    --cc=okanatov@astralinux.ru \
    --cc=okanatov@gmail.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).