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, 01 Mar 2017 09:29:13 +0000
From:   David Woodhouse <dwmw2@...radead.org>
To:     yangshukui <yangshukui@...wei.com>, linux-mtd@...ts.infradead.org,
        linux-kernel@...r.kernel.org
Subject: Re: Bug fix for uid/gid in jffs2

On Wed, 2017-03-01 at 16:44 +0800, yangshukui wrote:
> 
> @@ -146,8 +146,8 @@ struct jffs2_raw_inode
>       jint32_t ino;        /* Inode number.  */
>       jint32_t version;    /* Version number.  */
>       jmode_t mode;       /* The file's type or mode.  */
> -    jint16_t uid;        /* The file's owner.  */
> -    jint16_t gid;        /* The file's group.  */
> +    jint32_t uid;        /* The file's owner.  */
> +    jint32_t gid;        /* The file's group.  */

This is changing the binary data structures on the flash, and breaking
compatibility with all existing file systems. Your modified kernel
won't be able to mount existing JFFS2 images.

The way to do this, if we really do need 32-bit ugid support in JFFS2,
is to define a *new* node (JFFS2_NODE_INCOMPAT) type, and then support
reading both. You could even contrive a way to make it
JFFS2_NODE_ROCOMPAT; perhaps a new node type which contains *only* the
(32-bit) ownership information and overrides the 16-bit ones in the
original data nodes? 
Download attachment "smime.p7s" of type "application/x-pkcs7-signature" (4938 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ