linux-ext4 - Re: [PATCH] ext4: fix error when itable blocks is greater than s_itb_per

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]

Date:   Wed, 3 Aug 2022 09:18:59 +0200
From:   Lukas Czerner <lczerner@...hat.com>
To:     Michael Wu <michael@...winnertech.com>
Cc:     tytso@....edu, adilger.kernel@...ger.ca,
        linux-ext4@...r.kernel.org, linux-kernel@...r.kernel.org,
        allwinner-opensource-support@...winnertech.com
Subject: Re: [PATCH] ext4: fix error when itable blocks is greater than
 s_itb_per_group

On Tue, Aug 02, 2022 at 10:10:29AM +0800, Michael Wu wrote:
> The following error occurs when mounting the ext4 image made by image
> making tool:
> "ext4_init_inode_table:1301:comm ext4lazyinit:Something is wrong with group
> 0: used itable blocks: 491; itable unused count: 0."
> 
> Currently all the inodes in block group0 and ext4 image is divided by
> s_inodes_per_group. That leads to a hazard: we can't ensure all
> s_inodes_per_group are divisible by s_inodes_per_block. For example, when
> the s_inodes_per_group (equals to 7851) is divided by s_inodes_per_block
> (which is 16), because 7851 is undivisible by 16, we get the wrong result
> 490, while 491 is expected.
> 
> So, we suggest that s_itb_per_group should equal to
> DIV_ROUND_UP(s_inodes_per_group, s_inodes_per_block) instead of directly
> getting the result from s_inodes_per_group/s_inodes_per_block.

Hi Michael,

mke2fs is making sure that we completely fill the inote table blocks.
This is a corrupted image and so AFAICT ext4 is doing the right thing
here. There does not seem to be a problem to fix, unless you can somehow
trick mke2fs to make a file system like this.

-Lukas

> 
> Signed-off-by: Michael Wu <michael@...winnertech.com>
> ---
>  fs/ext4/super.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index 845f2f8aee5f..76cbd638ea10 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -4796,8 +4796,8 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb)
>  			 sbi->s_inodes_per_group);
>  		goto failed_mount;
>  	}
> -	sbi->s_itb_per_group = sbi->s_inodes_per_group /
> -					sbi->s_inodes_per_block;
> +	sbi->s_itb_per_group = DIV_ROUND_UP(sbi->s_inodes_per_group,
> +					    sbi->s_inodes_per_block);
>  	sbi->s_desc_per_block = blocksize / EXT4_DESC_SIZE(sb);
>  	sbi->s_sbh = bh;
>  	sbi->s_mount_state = le16_to_cpu(es->s_state) & ~EXT4_FC_REPLAY;
> -- 
> 2.29.0
>