linux-ext4 - Re: [PATCH RESEND v2 10/12] e2fsck: Add option to force encoded filename verification

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <877dpp2xqe.fsf@collabora.com>
Date:   Thu, 10 Dec 2020 17:48:57 -0300
From:   Gabriel Krisman Bertazi <krisman@...labora.com>
To:     Arnaud Ferraris <arnaud.ferraris@...labora.com>
Cc:     linux-ext4@...r.kernel.org, drosen@...gle.com, ebiggers@...nel.org,
        tytso@....edu
Subject: Re: [PATCH RESEND v2 10/12] e2fsck: Add option to force encoded
 filename verification

Arnaud Ferraris <arnaud.ferraris@...labora.com> writes:

> From: Gabriel Krisman Bertazi <krisman@...labora.com>
>
> This is interesting for !strict filesystems as part of the encoding
> update procedure. Once the filesystem is known to not have badly encoded
> filenames, the update is trivial, thanks to the stability of assigned
> code points in the unicode specification.

Ted,

I thought of this after your comment on LSFMM 2019 that strict mode
made the update of the unicode version trivial.  I think this patch is a
dependency for ext4 to support a newer unicode version, which I'd like
to support in the near feature, such that we don't lag too much behind
the Unicode specification.

>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@...labora.com>
> Signed-off-by: Arnaud Ferraris <arnaud.ferraris@...labora.com>
> ---
>  e2fsck/e2fsck.h | 1 +
>  e2fsck/pass2.c  | 5 +++--
>  e2fsck/unix.c   | 4 ++++
>  3 files changed, 8 insertions(+), 2 deletions(-)
>
> diff --git a/e2fsck/e2fsck.h b/e2fsck/e2fsck.h
> index dcaab0a1..f324e92c 100644
> --- a/e2fsck/e2fsck.h
> +++ b/e2fsck/e2fsck.h
> @@ -177,6 +177,7 @@ struct resource_track {
>  #define E2F_OPT_ICOUNT_FULLMAP	0x20000 /* use an array for inode counts */
>  #define E2F_OPT_UNSHARE_BLOCKS  0x40000
>  #define E2F_OPT_CLEAR_UNINIT	0x80000 /* Hack to clear the uninit bit */
> +#define E2F_OPT_CHECK_ENCODING  0x100000 /* Force verification of encoded filenames */
>  
>  /*
>   * E2fsck flags
> diff --git a/e2fsck/pass2.c b/e2fsck/pass2.c
> index b6514de8..f93e2b53 100644
> --- a/e2fsck/pass2.c
> +++ b/e2fsck/pass2.c
> @@ -1049,9 +1049,10 @@ static int check_dir_block(ext2_filsys fs,
>  	ctx = cd->ctx;
>  
>  	/* We only want filename encoding verification on strict
> -	 * mode. */
> +	 * mode or if explicitly requested by user. */
>  	if (ext2fs_test_inode_bitmap2(ctx->inode_casefold_map, ino) &&
> -	    (ctx->fs->super->s_encoding_flags & EXT4_ENC_STRICT_MODE_FL))
> +	    ((ctx->fs->super->s_encoding_flags & EXT4_ENC_STRICT_MODE_FL) ||
> +	     (ctx->options & E2F_OPT_CHECK_ENCODING)))
>  		cf_dir = 1;
>  
>  	if (ctx->flags & E2F_FLAG_RUN_RETURN)
> diff --git a/e2fsck/unix.c b/e2fsck/unix.c
> index 1cb51672..0a9012e5 100644
> --- a/e2fsck/unix.c
> +++ b/e2fsck/unix.c
> @@ -753,6 +753,9 @@ static void parse_extended_opts(e2fsck_t ctx, const char *opts)
>  			ctx->options |= E2F_OPT_UNSHARE_BLOCKS;
>  			ctx->options |= E2F_OPT_FORCE;
>  			continue;
> +		} else if (strcmp(token, "check_encoding") == 0) {
> +			ctx->options |= E2F_OPT_CHECK_ENCODING;
> +			continue;
>  #ifdef CONFIG_DEVELOPER_FEATURES
>  		} else if (strcmp(token, "clear_all_uninit_bits") == 0) {
>  			ctx->options |= E2F_OPT_CLEAR_UNINIT;
> @@ -784,6 +787,7 @@ static void parse_extended_opts(e2fsck_t ctx, const char *opts)
>  		fputs("\tbmap2extent\n", stderr);
>  		fputs("\tunshare_blocks\n", stderr);
>  		fputs("\tfixes_only\n", stderr);
> +		fputs("\tcheck_encoding\n", stderr);
>  		fputc('\n', stderr);
>  		exit(1);
>  	}

-- 
Gabriel Krisman Bertazi