linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <djwong@kernel.org>
To: Leah Rumancik <leah.rumancik@gmail.com>
Cc: linux-ext4@vger.kernel.org, tytso@mit.edu
Subject: Re: [PATCH 1/2] ext4: wipe filename upon file deletion
Date: Thu, 25 Mar 2021 17:05:36 -0700	[thread overview]
Message-ID: <20210326000536.GA22091@magnolia> (raw)
In-Reply-To: <20210325181220.1118705-1-leah.rumancik@gmail.com>

On Thu, Mar 25, 2021 at 06:12:19PM +0000, Leah Rumancik wrote:
> Zero out filename field when file is deleted. Also, add mount option
> nowipe_filename to disable this filename wipeout if desired.
> 
> Signed-off-by: Leah Rumancik <leah.rumancik@gmail.com>

I'm totally cribbing off of the previous decade's dirname wipe patch[1].

[1] https://lore.kernel.org/linux-ext4/1309468923-5677-1-git-send-email-achender@linux.vnet.ibm.com/T/#ma0442145ca50bb6e62f8e3502d607c758dd24418

> ---
>  fs/ext4/ext4.h  |  1 +
>  fs/ext4/namei.c |  4 ++++
>  fs/ext4/super.c | 11 ++++++++++-
>  3 files changed, 15 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
> index 826a56e3bbd2..8011418176bc 100644
> --- a/fs/ext4/ext4.h
> +++ b/fs/ext4/ext4.h
> @@ -1247,6 +1247,7 @@ struct ext4_inode_info {
>  #define EXT4_MOUNT2_JOURNAL_FAST_COMMIT	0x00000010 /* Journal fast commit */
>  #define EXT4_MOUNT2_DAX_NEVER		0x00000020 /* Do not allow Direct Access */
>  #define EXT4_MOUNT2_DAX_INODE		0x00000040 /* For printing options only */
> +#define EXT4_MOUNT2_WIPE_FILENAME       0x00000080 /* Wipe filename on del entry */
>  
>  
>  #define clear_opt(sb, opt)		EXT4_SB(sb)->s_mount_opt &= \
> diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
> index 883e2a7cd4ab..ae6ecabd4d97 100644
> --- a/fs/ext4/namei.c
> +++ b/fs/ext4/namei.c
> @@ -2492,6 +2492,10 @@ int ext4_generic_delete_entry(struct inode *dir,
>  			else
>  				de->inode = 0;
>  			inode_inc_iversion(dir);
> +
> +			if (test_opt2(dir->i_sb, WIPE_FILENAME))
> +				memset(de_del->name, 0, de_del->name_len);

You probably ought to erase the file type information too.

> +
>  			return 0;
>  		}
>  		i += ext4_rec_len_from_disk(de->rec_len, blocksize);
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index b9693680463a..5e8737b3f171 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -1688,7 +1688,7 @@ enum {
>  	Opt_dioread_nolock, Opt_dioread_lock,
>  	Opt_discard, Opt_nodiscard, Opt_init_itable, Opt_noinit_itable,
>  	Opt_max_dir_size_kb, Opt_nojournal_checksum, Opt_nombcache,
> -	Opt_prefetch_block_bitmaps,
> +	Opt_prefetch_block_bitmaps, Opt_nowipe_filename,
>  #ifdef CONFIG_EXT4_DEBUG
>  	Opt_fc_debug_max_replay, Opt_fc_debug_force
>  #endif
> @@ -1787,6 +1787,7 @@ static const match_table_t tokens = {
>  	{Opt_test_dummy_encryption, "test_dummy_encryption"},
>  	{Opt_inlinecrypt, "inlinecrypt"},
>  	{Opt_nombcache, "nombcache"},
> +	{Opt_nowipe_filename, "nowipe_filename"},
>  	{Opt_nombcache, "no_mbcache"},	/* for backward compatibility */
>  	{Opt_prefetch_block_bitmaps, "prefetch_block_bitmaps"},
>  	{Opt_removed, "check=none"},	/* mount option from ext2/3 */
> @@ -2007,6 +2008,8 @@ static const struct mount_opts {
>  	{Opt_max_dir_size_kb, 0, MOPT_GTE0},
>  	{Opt_test_dummy_encryption, 0, MOPT_STRING},
>  	{Opt_nombcache, EXT4_MOUNT_NO_MBCACHE, MOPT_SET},
> +	{Opt_nowipe_filename, EXT4_MOUNT2_WIPE_FILENAME, MOPT_CLEAR | MOPT_2 |
> +		MOPT_EXT4_ONLY},
>  	{Opt_prefetch_block_bitmaps, EXT4_MOUNT_PREFETCH_BLOCK_BITMAPS,
>  	 MOPT_SET},
>  #ifdef CONFIG_EXT4_DEBUG
> @@ -2621,6 +2624,10 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
>  	} else if (test_opt2(sb, DAX_INODE)) {
>  		SEQ_OPTS_PUTS("dax=inode");
>  	}
> +
> +	if (!test_opt2(sb, WIPE_FILENAME))
> +		SEQ_OPTS_PUTS("nowipe_filename");

Interesting how this time around it's a mount option...

At the risk of dredging up bad old memories, any chance you'd want to
select this if the file being unlinked has EXT4_SECRM_FL set?

Also, uh, I guess this is a change in default behavior?  Now you have to
opt-out of wiping filenames?  I suppose it's not a big deal
performance-wise since we're logging and writing buffers to disk anyway,
but I bet there's at last a few people who have recovered accidental
deltree invocations this way...

--D

> +
>  	ext4_show_quota_options(seq, sb);
>  	return 0;
>  }
> @@ -4161,6 +4168,8 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
>  	if (def_mount_opts & EXT4_DEFM_DISCARD)
>  		set_opt(sb, DISCARD);
>  
> +	set_opt2(sb, WIPE_FILENAME);
> +
>  	sbi->s_resuid = make_kuid(&init_user_ns, le16_to_cpu(es->s_def_resuid));
>  	sbi->s_resgid = make_kgid(&init_user_ns, le16_to_cpu(es->s_def_resgid));
>  	sbi->s_commit_interval = JBD2_DEFAULT_MAX_COMMIT_AGE * HZ;
> -- 
> 2.31.0.291.g576ba9dcdaf-goog
> 

  parent reply	other threads:[~2021-03-26  0:06 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-25 18:12 [PATCH 1/2] ext4: wipe filename upon file deletion Leah Rumancik
2021-03-25 18:12 ` [PATCH 2/2] ext4: add ioctl EXT4_FLUSH_JOURNAL Leah Rumancik
2021-03-26  1:21   ` Darrick J. Wong
2021-03-29 15:06     ` Leah Rumancik
2021-03-29 22:10       ` harshad shirwadkar
2021-03-30 16:32         ` Darrick J. Wong
2021-03-30 17:15           ` Theodore Ts'o
2021-03-30 17:18             ` Darrick J. Wong
2021-03-30 17:17   ` Darrick J. Wong
2021-03-30 18:14     ` Leah Rumancik
2021-03-26  0:05 ` Darrick J. Wong [this message]
2021-03-26 23:05 ` [PATCH 1/2] ext4: wipe filename upon file deletion Andreas Dilger
2021-03-27  1:43   ` harshad shirwadkar
2021-03-27  2:08     ` Darrick J. Wong
2021-03-29 16:06       ` Leah Rumancik
2021-03-29 19:05         ` Andreas Dilger
2021-03-30  1:07           ` Theodore Ts'o

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=20210326000536.GA22091@magnolia \
    --to=djwong@kernel.org \
    --cc=leah.rumancik@gmail.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=tytso@mit.edu \
    /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).